Sam's Laser FAQ, Copyright © 1994-2009, Samuel M. Goldwasser, All Rights Reserved.
I may be contacted via the
Sci.Electronics.Repair FAQ Email Links Page.

  • Back to Sam's Laser FAQ Table of Contents.

    Ar/Kr Ion Laser Testing, Maintenance, Repair

    Sub-Table of Contents



  • Back to Ar/Kr Ion Laser Testing, Maintenance, Repair Sub-Table of Contents.

    Introduction, Related Information

    Ion Laser Tender Loving Care

    Unlike the ubiquitous helium-neon laser, argon/krypton ion lasers will likely require some amount of attention during their life, even if there are no actual failures. This is due to the high current and power levels at which the tubes operate, and for external mirror lasers, the need to clean and align the mirrors and other optics. Where the current or/and output power is adjustable, the life of the tube and to some extent, the amount of maintenance as well, will depend on the operating point.

    This chapter deals with ion laser system and ion tube issues; ion laser power supply repair is discussed in the chapter: Ar/Kr Ion Laser Power Supplies. Some model specific repair and related issues may also be found in the chapter: Complete Ar/Kr Ion Laser Power Supply Schematics.

    Safety Issues When Troubleshooting Ion Lasers

    WARNING: Ion lasers are usually not isolated from the power line, even for those having a buck/boost transformer since it may be an autotransformer. Therefore, in almost all cases, either the power supply (preferred) or the test equipment will have to be isolated to even be able to make measurements. This is because the bridge rectifier(s) in the front end of the power supply makes it impossible to connect the ground lead of the scope (for example) to any useful reference without shorting it to ground through the scope. If connected to the power supply case (which should be at Earth Ground potential), there will be an AC component on the displayed waveform that will likely be so much larger than the desired signal as to make the measurement useless.

    The key here is that neither output of a bridge rectifier (which is most likely at the front end of the power supply) is at Earth Ground potential. They both have a large AC component with respect to Earth Ground. Consider:

    
                         D1
            H o-----+----|>|-------+---------+-----o DC+
                   ~|    D2        |+        |
          In from   +----|<|----+  |       +_|_
          AC line        D3     |  |      C ___
                    +----|>|----|--+       - |
                    |    D4     |            |
     ---+-o N o-----+----|<|----+------------+-----o DC-
        |          ~   Bridge       -
        +-o G o------------------------o Earth Ground (also connected to scope)
       _|_
      ///
    
    

    Hot (H) and Neutral (N) are tied together at the electrical service panel. Now think about what would happen if the scope test probe ground lead was connected to DC- without an isolation transformer. This would basically short out D4 and put D2 directly across the line (among other things). Not good. With an isolation transformer for the power supply, there will be no fireworks. However, an isolation transformer for the scope will not help unless it has an isolated ground, or the scope ground is disconnected.

    Take extreme care when making measurements. The laser itself should be powered via an isolation transformer if possible. This is much preferred to isolating the test equipment (or - gasp - cutting their grounding pins or using 3 to 2 pin grounding adapters without attaching the third wire). The reason is that isolating the power supply prevents the situation where accidental contact between the internal circuitry of the power supply or laser head, and earth ground, will be a shocking experience. If only the test equipment is isolated, the power supply and laser are still live with respect to earth ground. This will permit measurements to be made, but is still extremely dangerous.

    For tests of the control circuits or even the igniter, a modest size isolation transformer will be adequate. These are readily available new or surplus, or can be constructed inexpensively. It's only when the tube starts that the high current capability will be required. See the document: Troubleshooting and Repair of Consumer Electronic Equipment for more info on isolation transformers.

    Note that most isolation transformers DO NOT isolate the earth ground (third prong). Thus, if using one on an oscilloscope where signal ground and earth ground are tied together, there will still be fireworks, and the scope cabinet will still be a potential shock hazard (for the return from the live laser).

    Ultimately, safe troubleshooting of line-powered and/or high voltage systems comes down to good work habits - one hand in a pocket, elimination of Earth Ground points to touch, not working in hip-deep salt water, etc.

    Also see the document: Safety Guidelines for High Voltage and/or Line Powered Equipment.

    Web-Site with Related Information

    Evergreen Laser Corporation has a Web site which includes a significant amount of information on ion laser tube and power supply adjustment, alignment, failure modes, troubleshooting, and repair.



  • Back to Ar/Kr Ion Laser Testing, Maintenance, Repair Sub-Table of Contents.

    Tube Life, Effects of Abuse, Failure Modes, Rebuilding

    Argon/Krypton Ion Laser Tube Life

    At the amazingly high power that these tubes use, it isn't surprising that their expected lifetime is a strong function of tube current. While a tube may be rated for '10 A max' - and this is where you get the expected optical power output - the lifetime when running at that power level may be 1/10th that of the same tube idling at 5 A.

    For example, the following table shows the expected MTBF hours for an ALC-60X/Omni-532 compatible tube (specific model unknown) as a function of beam current:

       Plasma   -------- Laser Output Power (mW) --------
        Tube     Multi- ------ Gaussian TEM00 Mode ------
       Current    mode           ------- Pure Line ------
       (Amps)   -- All Lines --   457 nm  488 nm  514 nm   Lifetime (MTBF) Hours
      ---------------------------------------------------------------------------
          4        20      10       1.0      7.0     0.0      15,000 - 25,000
          6        50      30       2.0     17.6     7.5       8,000 - 15,000
          8       110      70       5.0     27.0    23.0       4,000 -  6,000
         10       220     130      10.0     44.0    42.0       1,500 -  2,000
         12       325     200      15.0     60.0    68.0       1,000 -  1,500
         14       430     280      22.0     81.0    98.0         500 -  1,000
    
    Thus, it makes a lot of sense to run at the lowest power that will be adequate for the application whenever possible to maximize tube life (not to mention keep your utility bills within reason!). For a multiline tube, reducing current may still get you the particular color(s) you need with adequate intensity and/or acceptable color balance. However, each model/age tube will have a different balance of lines at a given current (i.e., of the 10 for argon and 11 for krypton). Also see the section: Laser System Life.

    Note: The power output numbers, above, look like they apply to a very 'hot' or new tube - at least compared to what you'll typically find in a surplus 60X laser. The actual power (at a given current) for one of these will likely be somewhat lower. Keep this in mind if you see an advertizement for a 400 mW 60X! Also see the section: Expected Output Power from Surplus or Previously Owned Ion Lasers.

    (From: Steve Roberts (osteven@akrobiz.com).)

    10 A is absolute max on a average 60X tube. I once talked to a engineer who was on the 60X project, and he prefers 9 A as the upper limit. Keep in mind that's a MTBF for a new tube, not a used one. Also note that the factory sells lasers from 5 mW and up, therefore you don't know what you have in the cavity.

    The American PSU can in theory source 14 amps for short periods of time. However this is only on a modified unit. 10 Amps is about what they can do. 11 amps runs the pass-bank way hotter then its 100% duty cycle temp without readjustment, in my opinion.

    However I've never seen a manual for it and am going from what I've learned from trial and error. A few newer compact FET based gold boxes made by Marlin can only do 9 amps. Above that they go into meltdown, so one has to be careful.

    As for maximizing life of a small to mid-size tubes (up to 2 or 3 W) where the laser is to be used intermittently (between shows, for example - say 10 minutes on, 20 minutes off), it is better to drop it back to idle (the lowest current which will maintain the discharge) when the beam is not needed rather than shutting down completely. (However, large frame ion lasers should be turned off entirely between shows - they like to run at one particular current.)

    Maximizing Ion Tube Lifetime

    For a small air-cooled ion laser, tube life is more or less inversely related to tube current. See the section: Argon/Krypton Ion Laser Tube Life. However, at very low currents, I have seen plasma instability causing the power supply to drop out and attempt to restart - hard on both tubes and power supplies. This occurs with my high mileage Cyonics/Uniphase 2214-30SL if run at a current of less than about 4 A. The tube voltage climbs until the PSU can't maintain the discharge. During this time, it is likely that bad things like greatly increased heat dissipation are happening at the cathode. :(

    (From: Steve Roberts (osteven@akrobiz.com).)

    For larger water-cooled ion lasers reducing current increases tube life only up to a point. However, running consistently at low idle is not good for the tube. On short frame lasers there is a certain reference current. You can go about 6 to 7 amps below it and an amp or two above it and still have decent lifetime. This point is determined from the factory data sheet for a given tube and by users' experience.

    Running above this window leads to quickly shortened tube life. Running much below it can result in plasma instability and PSU damage, as well as eventual tube damage. The rebuilder I used to work with stressed running consistently for long periods at the reference current as the best way to avoid warranty repairs, and I'd tend to believe him.

    Also, depending on your PSU design, it is often recommended that you start at the high end of the current range. Otherwise, damage to the PSU and tube cathode may result from repetitive start pulses when the tube doesn't catch but just flashes. Some power supplies will start at a high current regardless of the control settings, and then drop back to the selected current or power.

    On white light lasers, or pure krypton, there usually is a 'knee' curve in the plot of light output versus current where the krypton lines really take off. I'd find the knee point for my tube and see where it is versus the factory recommended current and run right around that region or a little above it. What I mean is if you look at the Kr gain curve, you'll see its a very gentle slope from laser start to usually a point around the mid current range, where the line in the plot starts to shoot nearly straight up. If the tube is set up right, that breakpoint should be way below the recommended operating current. The knee point will shift higher as the laser ages, and Kr adsorption rates are much higher then Ar rates. 647/568 nm balance is a good indicator of tube pressure, lower pressures favor yellow, however red has much more gain then yellow, so you'll always see some red.

    Owning a good power meter and logging line power on a weekly basis with a prism is a good idea. One of the problems of owning a high power argon or mixed gas is that laser ownership at a point stops becoming science and turns into somewhat of a art.

    What Happens if the Tube is Run with Excessive Current (>10 A)?

    There are at least three issues:

    OK, here are the gory details:

    (From: Steve Roberts (osteven@akrobiz.com).)

    Lifetime decreases exponentially with current, and you can get to the high speed end easily enough. (See the chart in the section: Argon/Krypton Ion Laser Tube Life.) At currents not much over 10 amps, assuming you have a power supply with enough spare output compliance to not burn up, first thing that happens is it outgases from the cathode and bore, so that's the LAST time it lights. Then she starts sputtering and flashing, at this point some will just wink out from overpressure. If the pressure was low enough to begin with, now the anode starts glowing red very bright, and if it doesn't crack off, and if the cathode feed-throughs hold, and the bore doesn't crack from shock, you'd see maybe a watt or so for a few seconds until the cathode goes dead from meltdown or the return bores could no longer handle the flow.

    If you did reach this region, getting the cathode material off the stems and replacing the etched insides of the Brewster windows is hard. You can see the particulates accelerate inside the tube from the intracavity light and they go right to the windows at HIGH speeds, glowing like little fireflys and acting like fast little cruise missiles and stick when they hit the window. If they bounce, they go into the returns where the charge is more neutral.

    By this time the gas returns will be filled with crud, and when it cools off and you try to relight it will probably find the returns mucho more interesting then the bore.

    Interestly enough, new tubes are the hottest until they season in. It's the Brewster windows, cathode, and pressure that determine tube life.

    Hint: Got Brewster windows that are clean both inside and outside with good tube voltage, a positive delta-T, and good magnet?, and yet no matter what you do you get no power, put in two high reflectors for a while, ramp up the current and get rid of your color centers in the windows. This is a last resort sort of thing for bigger lasers.

    The former Soviet Union and a few U.S. researchers developed some neat tricks to avoid these things and get to 500 to 1,000 watts, but I m not talking. :-) The tricks are not all that easy to do outside of a lab anyways and they fry the coatings off the optics fast.

    HGM Medical, the surgical arm of American Laser, made some glycol cooled things roughly the size of a 60X. A water-to-air heat exchanger is built in. Bore sizes are similar, but the changes are in the gas returns, anodes, stems, tube pressure, PSU etc. They do 3.2 watts new at the OC, for a few seconds at a time at a VERY VERY low duty cycle. The limits are the anode and pressure, the anode is air cooled. If you hold the treat pedal down for more then 20 seconds, the power drops off like a falling stone as the pressure rises until the overtemp switch cuts out to protect the tube. Then you need to set for about 5 minutes and it comes back. These were for eye surgery, and the eye surgeon rarely needs the full 3 watts, 200 to 500 mW delivered power to the eye is more typical. Also the OC transmission and lasing threshold is a lot different on a power on demand tube. You may not get enough transmission on a normal X optic. They usually have PC in the part number and a diode aiming beam, since the tube doesn't lase until needed, call it fire on demand, once the tube gets stable, then the shutter opens.

    Factory manuals and practical experience are wonderful. ;-)

    Here are some more of the gory details, if you can stand them!

    The following take place in NO particular order: They are heat related and plasma/pressure related fast failures!!!

    1. Sheer overpower - it can't take the heat and the weakest link dies!

    2. Cathode gets stressed, the thin electron emitting surface can't keep up and boils off. Cruds up on windows and buries the lasant gas. Cathode lead seal failure due to I2R overtemp (very common in overdriven Lexels!). Remember that the cathode carries its own current plus the bore current. (A typical cathode is 70 to 90 watts of heater).

    3. Anode cracks. Anodes take the whole discharge and the plasma bombards them, heating them to beyond their design point which causes them to outgas and sputter, cruds up windows, and cracks glass to metal or ceramic to metal seals.

    4. Cooling failure due to hot spots/localized boiling cracks the tube. These things run close to critical at full power, BeO is 5 times more conductive to heat then aluminum, but there is a limit to what it can transfer. Many designs, especially high powered air-cooleds, run close to the border on heat capacity. A 5 watt graphite tube for example, has to radiate 150 watts of infrared per centimeter through its quartz jacket into the water at max PSU power delivery, or it's meltdown time! (Gotta find the citation for that, I think its the Lexel sales brochure or the CR52 manual.) The graphite disks reach nearly 1,000 °C. The limit is the fact that graphite disks have to be much smaller than the quartz tube for expansion reasons and therefore cooling becomes radiation based as IR through the quartz adsorbed by the water.

    5. Bore erosion, a Lexel-88 for example runs at something like 500 to 700 amps per square centimeter current density down a 2.08 mm bore. (e.g., 470 A/cm2 at 16 A, 520 A/cm2 at 18 A, 600 A/cm2 at 20 A) As you pump up the current, the magnet can't confine the plasma as well and it hits the bore walls and blooms out around the anode, eroding bore material which hits the windows, clogs the gas vents, etc. Get much above 2,000 amps per square cm and NOTHING makes a good bore.

      The maximum limits are 700 A/cm2 for graphite and 850 A/cm2 for BeO. It's somewhat higher if there are no gas returns being a tradeoff between mechanical strength and radial thermal conductivity to cooling water. However, the tube will suffer from cataphoresis (gas migration) if it there are no returns. Gas returns internal to the BeO weaken it and raise thermal resistance. (This information from: Kesik and Siejca, SPIE vol. 859, Laser Techonolgy II.)

    6. Gas return blockage, crud from any process that releases metal causes it to go down the gas return and deposit, leading to tubes that find it easier to light down the returns rather then the bore. VERY common on ALC/OMNI, especially old 643 whitelights.

    7. Gas depletion, you drive it harder, it buries itself into neutral objects and the bore harder. The pressure gets lower, the number of energetic particles go up and they attack things, leading to more of #1, #4, #5 and power falls off as there are more electrons then atoms to ionize. Hence you make a atom smasher internally and that electron beam etches things.

    8. Plasma going into wrong places. Look at a 168 tube or an Ionics tube, see the metal shroud around the cathode, It's there to keep the plasma off the thin glass cathode bell. Cranking up the current generates more plasma that can leak around this barrier as well as erode the barrier. I've seen overdriven tubes with holes in that metal shell, and that metal sputtering onto cold surfaces buries gas.

    9. Any crud released by any process listed above heads for the Brewster windows!!

    10. Most of these effects scale exponentially to the current, A.K.A. the laws of thermodynamics state that you will always loose gain, although sometimes you can just about break even. Many small inexpensive 90 V tubes are dying from minute one, they are actually overfilled to get some lifetime. Some effects crosstalk, especially low pressure at end of life. Low pressure is the "lifetime accelerator".

    11. Cooling systems that have crud in the water dump crud into tube walls and magnet spaces faster when they are hotter: blocked cooling = hot spot on tube walls = melting or cracking gas integrity failure. Just installed a new tube in one of those cases. One week before, that laser was basically brand new, construction of a new water main in the area resulted in years of pipeline crud filling the buildings water supply, a flow switch failed on, the tube boiled itself to death. The filter was intact, this was all mineral salts deposited out.

    There's another 10 or so nasties, but I'm outta time, chao!

    Buy a bigger tube then you need, turn it down to like less then 1/3 to 1/2 of its design power, cool it well, it will live nearly forever.

    Filament and Discharge Voltages for Some Typical Air-Cooled Ion Lasers

    All are assumed single line unless noted - 514 nm also available with slightly different output power.
                             Filament     Discharge 
        Manufacturer/Model     VRMS      VDC at IDC        Typical Output Power
     ------------------------------------------------------------------------------
        American Laser 60X      3.0    107-108 at 10 A  20-40 mW, higher multiline
      Cyonics/Uniphase 2301     3.0      105   at  8 A       5-20 mW 488 nm
           NEC GLG-3030         2.6       85   at 10 A        20 mW 488 nm
      Spectra-Physics 091-93    2.9       80   at 10 A        25 mW 488 nm
      Spectra-Physics 093-91    2.9      110   at 10 A  Greater than 25 mW, 488 nm
    

    Failure Modes of Ar/Kr Ion Tubes

    Also see the section: Hard-to-Start Ar/Kr Ion Tubes - Outgassing and Keeping Your Laser Healthy.

    Aside from obvious physical damage, there are a variety of conditions that can prevent a tube from lasing or result in a weak beam or one with rings or other artifacts:

    For a possible way of testing a tube without requiring a complete high current power supply, see the section: Pulsed Operation of an Ar/Kr Ion Tube.

    Effects of Contamination on Ion Tube Life and Performance

    (From: Steve Roberts (osteven@akrobiz.com).)

    Hydrogen and helium slip through glass and ceramic, and can stay hidden in certain metals used in tubes. They are also used in processing the porous cathode for various reasons. The heavy argon and krypton ions drive the lightweight hydrogen deep into the bore walls and metal portions of the tube, when a tube sets on the shelf, hydrogen can in some cases seep in from the walls and welds. This happens even in quartz or hard Pyrex, and more slowly into ceramics. Most of the hydrogen originates from welding of the seals and fusing of the ceramic to metal joints in a hydrogen furnace during processing. In glass tubes it can come from the cathode and metal structures.

    Hydrogen doesn't ionize easily. Thus it raises the tube pressure while increasing the resistance of the discharge. This can lead to a condition where the tube will not lase. Argon has a wide lasing pressure range but as the pressure climbs too high, the blue and violet lines trail off, the 514 nm and 528 nm (green) lines get a little brighter, then the green falls off and lasing ceases, even though in some cases the discharge will still start. Any gas in the tube other then argon or krypton diverts energy from the lasing process - some more then others - and can cause cathode poisoning.

    Tubes are carefully filled and designed to center the positive column and the negative dark space and negative glow in specific areas of the tube. That's one reason why cathodes are powered by center tapped AC - to move the arc around. When this alignment deteriorates, you get bore wall erosion, localized heating, and other undesirable effects leading to tube damage.

    Thus, a tube with large amounts of hydrogen that the getters can't handle needs reprocessing soon as the hydrogen will change the plasma physics in the tube by moving the cathode dark space and other regions of the discharge which promote erosion to places in the tube not designed for erosion. The main damage will be to the cathode, causing a hot spot to form ultimately leading to cathode sag or total failure (filament breakage). In fact, it is this small hot spot, not gravity, which is the most likely cause of cathode sag. The hot spot may also cause the electron emission agents mixed into the cathode matrix to boil off and dump more impurities into the tube. Low tube pressure will also move the discharge back toward the cathode, heating it excessively.

    Other impurities have similar damaging effects, including some that are much more reactive with the tube materials. Carbon dust, for example, will be ionized and accelerated toward the windows, damaging them.

    Getters in Ion Laser Tubes

    On a 60X tube, you should measure low ohms or a short from the cathode leads to the cathode endbell. This is the getter assembly on most air-cooleds. Having the endbell short to ground and passing current through it will release a cloud of Ba, Ti, Sr and some other gook, probably ruining the tube.

    You normally leave the end-bells floating. The only time you would connect to a end-bell is when you want to heat the getter. I don't know where the getter is on Cyonics tubes. Often they have a 3rd pin for it, if not, it's in the cathode end-bell. The getter is a corrigated or fan fold ribbon with little tabs sticking out of it located toward the front of the end-bell so it's not directly bombarded by plasma. It is an active device using the floating ions and electrons for activation. Current is passed through it at the factory to convert carbonates to pure metals and to outgas it.

    It is not safe to fire the getter on a 60X, I (Steve) have tried it a few imes and a hard start tube becomes a no start tube real fast! For some of their older tubes, Coherent has a getter procedure for a worse case situation where hydrogen has outgassed when the tube sets on a shelf and is thus at high pressure. Otherwise, no manufacturer would ever suggest lighting off the getter. It takes 7 to 10 A to heat the X getter, a ribbon secured to the cathode at one end. So, it really gets heated a bit during normal use. Further heating would just evolve more gas or burn off more material causing it to bury argon and/or liberate whatever it has gettered. The 60X getter is unique in that its coating is on both the end bell wall and the ribbon after its heated.

    On other tubes, getters are designed differently. On Lexels, for example, there is a getter in the ballast tank and the whole inside of the tank is coated. A electrode sticks out of the tank to activate the getter at the factory.

    The getter ts there mainly to collect hydrogen and water vapor. Having loose H2 (which diffuses out from the welds and brazes which are done in a hydrogen furnace) raises tube voltage and lowers power. That H2 has a tendency to make water vapor which is really hard on the cathode.

    Notes on Gas Pressure and Effects on Ion Tube Performance

    An ion tube must be run within a certain range of gas pressures for optimum performance as well as to prevent damage and maximize tube life (or the time between refurbs). This section deals mostly with large-frame, expensive, argon, krypton, and especially mixed gas ion laser. It is often cost effective to rebuild their tubes when it is no longer possible to obtain adequate power on the lines you want or performance suffers in other ways. These are also the sort of beasts which have built-in gas fill stems and other bells and whistles not found on the small air-cooled ion tubes most of us know and love. However, the basic information on effects on tube voltage and lasing applies to all sizes of ion tubes even if you don't have control of pressure and whether rebuilding is an option or not.

    (From: Steve Roberts (osteven@akrobiz.com).)

    If your tube is getting old and the gas pressure is starting to drop, refill the tube before you start damaging the cathode. Consult a professional to find out what your tube voltage drop should be for a given current. A lower voltage across the tube equals lower gas pressure. Don't wait too long or it won't be possible to rebuild it.

    The following comments were stimulated by being told by another laser tech why my customer's laser that had excess argon added by mistake should not be repumped. The tube isn't at quite high enough pressure to do major damage and the excess argon will burn off eventually. Hydrogen, on the other hand, tends to reappear for some reason, often without warning. I think the answer to that would only appear if we were to torture a ticklish Spectra-Physics or Coherent engineer for hours with a feather :-).

    You see, I had thought all along that argon was what needed burying by use, and some of it does need to be buried, but what argon that actually outgases gets reburied in the first 15 to 20 minutes, unless a fill valve from a reservoir is leaking or is opened too much by mistake on a big argon. It's not the argon pressure that prevents ignition, but the H2! Hydrogen seeps into the metal during manufacture of the tube, and can in some cases either leak out slowly over time or suddenly appear, most likely from thermal shock.

    Hydrogen causes a large increase in tube voltage, makes starting harder and depletes the population inversion, thus reducing power. When hydrogen or nitrogen is in the plasma, it generates large amounts of heat, thus damaging the tube, hydrogen also reacts with the cathode materials, reducing output and damaging the cathode. If a tube goes high in hydrogen and the getters wont clean it up over time, it's time for a rapid repump, to recover the tube from quick failure.

    I learned this the hard way today on a customer's 10 watt laser. The previous tech, not knowing he had a PSU problem, assumed it was low tube voltage (discharge), when in fact the tube voltage meter on the front panel was 25 volts out of calibration from a fried resistor. Each fill brought the laser tube up about 3 to 5 >volts on this model, and he did 5 or 6 fills. Thus, the nominal 354 V tube >drop was now actually 374 V and the PSU could still ignite it. But, the blue >and weaker green lines were suppressed by the high pressure. However, the >528 nm line really blossoms when the pressure is high. This is why a new tube >is so hot in the green but then goes blue as it is used and pressure goes down. >The overfilling resulted in a 2.5 watt drop in lasing power.

    The net result was at low currents the tube would not lase at all, and it pushed the lasing threshold (current) through the roof. And, getting the unit to its rated power required a current of 34 A instead of 31 A! We were lucky enough to have a factory test sheet on this one with all the detailed measurements. A high pressure tube manifests itself in two ways, it won't lase or lases at greatly reduced power, and the voltage across the tube increases. I have spent many hours on some units thinking I had a alignment problem, when in fact it was real high pressure.

    Once I fixed the ignite boost circuit - which was the original problem - the tube did start up but getting it aligned well enough to get any power was a factor of 10 harder then what it would have been at normal pressure.

    The gas fill on this unit is not by solenoid, but by two manual valves with a small space between them, you open one, fill the space, close it then open the other to add the small known quantity of gas. The previous tech thought it was necessary to close the fragile refill valves with Vice-Grips(tm), thus ruining the needle valves, he didn't realize the tube takes time to stabilize as the fresh gas has to be pumped to one end of the tube, heated, and then the new pressure would stabilize. This actually takes 15 to 30 minutes per fill.

    Some 500 hours of operation from now, it will be where it should be. All this was heard when I hauled the unit to the refurb shop for a more thorough evaluation. However if this were hydrogen outgassing, which raises the firing voltage and damages the cathode, a repump would be required. A repump would have also been required if the pressure was so high that lasing would not be possible, in this case about another 10 volts. When in doubt, call in a expert!

    Tube voltage drop is directly related to pressure in a Ion laser, charting the tube voltage is a easy way to monitor your laser, gradual increases in pressure over time suggest a leaky fill valve. Sudden increases in tube pressure suggest a tube that has possibly overheated and been damaged, hydrogen out gassing, or a recently activated autofill system. An autofill system getting out of control could rapidly overfill a tube. If you hear a sudden burst of solenoid valve clicks, something is going on. Most modern fill systems have a double solenoid valve with a small chamber between them, and a larger high pressure argon reservoir. Opening valve number one fills the chamber with a known amount of gas, then valve number one closes and valve number two opens the chamber to the tube. It then takes 5 to 20 minutes before you see a real stable change in tube voltage. Although you will see a sudden jump when the valve opens, it takes awhile for the gas to go against the discharge and flow down the tube. It is wise to let a laser run for 20 to 30 minutes before enabling fill, as well as checking across the tube with a multimeter once in a while to check the panel meter calibration.

    Keeping an eye on where the autofill enable switch is set is a good idea. On some older lasers it's a third key-switch position, and the operator is responsible for just pulsing the key momentarily to open the fill system solenoid - until the fill system buzzer stops beeping or the light goes out. That little "fill" button is a tempting thing for a visitor to push on some lasers, especially if the fill card has been rewired or bypassed by a tech tired of having his morning coffee ruined by a loud beeping while his recalcitrant laser stabilized. If you do anything to the fill circuit, carefully recalibrate it to the factory specs or add a extra key-switch to prevent it from firing.

    Some older mixed gas and krypton lasers have a bizarre version of the fill system called a "pressure pump". Since some of the the Kr red, yellow and green lines are fussy about lasing pressure, the pressure pump has a third larger chamber besides the fill chamber to pump the tube down. You increase the current, which raises the plasma temp and thus the gas pressure thanks to the laws of Boyle and Charles (remember P = VT?). A series of valves then open to the larger third chamber, which fills with hot gas acting to reduce the bore pressure. Later when you wish to raise the pressure, the third chamber is slowly dumped into the tube. These lasers tend to have a series of special markings on the tube voltage meter, to guide the operator in choosing the best pressure for a given line. The 647 nm red and 568 nm yellow lines share the same upper state. Whether you get red, or yellow, or a mix of both is determined by the gas pressure which favors a certain lower state for the transition. The 647 nm line likes high pressure, while the 568 nm line likes a much lower pressure. Running one of these requires some operator skill, patience, and a good reading of the manual.

    Remember, when working with fill systems, patience and care is needed, and time must be given for the tube to stabilize.

    Incidentally, if you do much work on lasers, it is wise to buy a cheap DMM, as when working on the laser you can often forget it's still across the tube, then when the tube winks out and restarts, the ignite pulse fries the meter into oblivion. (Or always put the meter before the igniter if you have access to the power supply and account for any wiring/other voltage drops.)

    And for those poor souls who would wish to repump a CR52 or other ancient coherent graphite with the vacuum access valves mounted on the reservoir, the valve actuators to hook onto the tube are available from CPC-Cryolab. Now I can't reccommend equipping modern ion laser tubes with O-ring sealed anything, having bad experience with various valves etc. Orings cannot be depended on at the low levels of vacuum used in a ion laser. Graphite tubes have such horrible gas cleanup and release cycles that something was needed to recover the tube if it got over or under pressured. On more modern tubes, a oring or most types of bellows valve is NOT going to stay sealed for more then a couple of days of operation. Nothing beats a copper pinch off or a molten glass seal off. However for a DIY sealed CO2 laser, where air leakage is not going to do much harm, these things are great!

    (From: Dan Glassburn (dan@niteliteproducts.com).)

    Concerning tube pressures and discharges....

    While under normal pressure range, the voltage at a particular current will be related to tube pressure - i.e., lower voltage equates to less pressure and higher voltage equates to higher pressure. However this is only within a particular range of pressure. Many tubes when running out of the normal operating range will reduce voltage then at some point start to increase tube voltage required to keep them ionized. This is more prevalent with solid bore tubes. The disk tubes just won't sustain ionization at lower pressures and higher currents. Many of the solid bore tubes will go into a "chirp" or oscillation which usually will cause havoc with the pass-bank. In the case of an air cooled, higher currents to deliver greater outputs will cause the tube to outgas more, especially if you don't have enough cooling. An easy way to see if the tube is low pressure in a 60X is to drive the current to it's maximum from and idle and see if the tube winks out. Many of the systems on the market do exhibit this symptom, although this is usually not how the customer normally uses the system. To get additional life from a low pressure air cooled, ramp the current up slowly. This allows the gas pressure in the tube to increase slowly and help sustain higher currents.

    Bad plasma oscillations usually occur if the tube pressure has decreased to the inversion point, where the impedance transforms from slightly positive to negative. Passing through zero is the trouble point. Oscillations again start occurring if the pressure gets almost too high to light or stay lit. For example, if you have a leak on the pumping station while lit. I recently have had 2 I90s on station and one of them had burned so low that it wasn't lighting. Adding a few puffs of gas made it unstable and only able to stabilize at high current. Puffing in about two volts of gas made it appear stable for a while, but to get ignition in the next few weeks operating at low currents I had to puff in more gas, even though the tube seemed to be at the correct voltage. (For these tubes, the inversion point seems to be a plateau in the curve.) So I finally over puffed it by about 3 volts, and now she always fires on the first pulse. So the worst plasma chirp seems to occur at very low pressures.

    More on Ion Laser Tube Cathode Characteristics

    (From: Steve Roberts (osteven@akrobiz.com).)

    Why does a healthy laser cathode climb in power (positive delta-T) when the plasma is lit? Rough explanation: Electron emission is an endothermic cooling process for the cathode. Sure you have to heat it to get it to emit, but it looses electrons, which by definition are a packet of energy. This results in cooling and a lower resistance of the cathode wire so there is an increase in current till the system reaches equilibrium. The cooling is perhaps roughly proportional to the cathode work function times the emission current.

    In a healthy ion laser tube with good gas pressure, you have many more positive ions then electrons floating around the tube. Thus, the tube voltage climbs as the beam current increases. There is plenty of gas to ionize, electrons are rapidly adsorbed and used, they have a short mean free path from atom to atom, but lots of atoms to ionize. That short mean free path and higher probability of collision means they have a longer path down the tube, hence the higher voltage.

    In a mediocre tube, there are less gas atoms due to lower pressure from plasma driven adsorption, and contaminates floating around the tube to adsorb gas and poison the cathode. But there is still healthy cathode emission, so you have a more of a break even condition - less gain, less ionization occurring, but still the tube is happy, just its failure mechanisms are starting to accelerate because of the lowered pressure. This is observable as less tube voltage climb with current and is the "Inversion Point"

    In a low pressure tube, there is a big electron cloud with just a few ions. The ions remaining are energetically accelerated and plasma etch tube structures. The electrons are also highly energetic due to a lack of collisions and thus do even more damage as they pass through the tube. By this time the cathodes emission is probably shot from contamination and plasma etching. Impurities don't get buried if there is no gas to accelerate them into the walls. The plasma is unstable and probably moving around by this point. As you increase tube current, you have a excess of electrons floating around and so the tube voltage actually goes lowers.

    So the first thing you do when you get a new laser, is chart the tube voltage versus current and check for a delta-T on the cathode. Delta-T is best observed by using a clamp-on AC ammeter around a cathode lead. Note that a few more modern high tech lasers have constant current transformers or autoswitching PSUs for their cathode and this may not apply to them. Also, if your cathode transformer is not correctly tapped or the buck boost on a older laser is misadjusted, you may not see a delta-T of either sort.

    Lowering the cathode current moves the first dark space in the discharge back toward the cathode and can reduce the amount of ions generated, resulting in etching of the cathode's active emitting surface, as well as running the risk of cooling the cathode into the 600 to 700 °C region where cathode sag occurs. Excessively raising the cathode voltage (say .5-.7 volts) results in the outer emitting layer "boiling off" and emission falls off fast, resulting in short tube life. The remaining emitting layer will tend to "spot", i.e., most of the emission will come from the remaining emissive material at a cool spot and the cathode will eventually melt or burn at that region. Its is however better to err on the warm side then then the cold.

    Some modern tubes with cathodes that are NOT tungsten-rhenium with a borate matrix or NOT a "dispenser" emitter run at about half the current of older cathodes, i.e., many Coherents now have the recommended cathode amperage stamped on the gas reservoir, and I have received reports from credible people that some Lexels are only consuming 15 to 17 amps instead of the normal 25 to 35 amps. This would be characteristic of a more modern Nitride ceramic cathode. So if in doubt, call the factory with the tube serial number and manufacturing date in hand. I have not yet heard of air cooleds using the expensive lower current cathodes, only the large and medium frames. However the ceramic cathode has a much longer life and greater electron emission.

    Thanks to Kim, Dan, and Dale for their contributions on this subject, responsibility for errors or incorrectness is my own. --- Steve)

    Comments on Filament Temperature and Thermal Cycling

    (From: Steve Roberts (osteven@akrobiz.com).)

    One contributing cause to cathode sag is something called the "Miller-Larson effect". It's a momentary plastic state in the filament caused by a change in the tungsten grain structure as its temperature passes through the range 600 to 675 °C. This results in a lengthening of the filament, and each time the filament goes through that range on warmup, it gets a bit thinner and a little longer. So, it might make sense to keep your filament on (or at reduced power but above 700 °C) during standby rather than shutting it off completely.

    Start plasma can suppress emission from the cathode, and if you have a flakey cathode transformer or one that's borderline, the filament can still be cold. One symptom of a cold cathode is the tube catches and starts for a few seconds but winks out as a stable hot spot won't develop on the cathode until its heated more and the barium kicks in. There is a point as you heat the cathode hotter that there is no gain in emission. So you want to be above the Larson-Miller region and yet not too hot. I just found a excellent book on the lifetime and operation of large radio frequency vacuum tubes in broadcast duty, that's where this is coming from. In fact, the broadcast folks don't shut their cathodes down if the downtime is going to be less then 4 hours, they run them at what's called "Black Heat" at about 800 DegC to avoid the stresses of startup.

    Unsagging a Sagging Filament

    A sagging filament may block or partially block the bore, making lasing rather difficult. However, the filament is distorted but not blocking the bore, it's probably best to leave well enough alone. Perhaps rotate the tube by 180 degrees so it sags the other way but don't do anything else as long as it works.

    (From: Steve Roberts (osteven@akrobiz.com).)

    If you just try to run at higher than normal temperature, it will be more likely to destroy the cathode permanently by blowing the emissive barium layer off, depleting it possibly in seconds!

    Part of cathode initial processing and reprocessing involves heating the cathode to drive the barium out of the porous tungsten matrix at hotter then operational temperatures, but it must be done in a vacuum at 10-7 Torr. It DOES NOT work in the inert gas.

    Thats what the famed car battery trick tries to do, blow the dirty non emitting regions off as dust into the tube and expose a little fresh barium. Thats bad too, its much better to have the tube reprocessed at that point then fill it with dust that will erode the bore and optics.

    The more correct thing to try is undercurrent the cathode and invert it. Pure Tungsten has a known sagging temperature region where it is butter soft at 600 to 700 °C. However, the cathode is not pure tungsten - it has rhenium, barium, aluminum and calcium - so the exact softening point is known only to the cathode manufacturers. Normally, the cathode runs at 1,080 to 1,120 °C by design. At 800 at °C it actually starts getting tougher in shear and you're not going to do that well because of the additives. You'll end up cooking and outgassing if not actually melting the nickel or molyB section that joins the tungsten to the kovar in the lead throughs.

    So if you want to try inverting it, get it to barely a warm soft red glow visible only in the dark and stay and watch it.

    About the High Cost of Refills and Refurbs

    Gases don't generally leak in or leak out of an ion tube at significant rates unless the seals were defective from the time of manufacture or stressed due to extreme overheating. However, after many hours of operation, the argon and/or krypton gas in an ion tube will become depleted - probably by being buried under sputtered cathode material on the tube walls. The pressure goes down, output power at a given current decreases, and eventually lasing ceases entirely. Usually, the gas atoms are trapped to stay. Large-frame ion lasers have gas reservoirs and fill valves to permit new gas to be added as the tube ages. Since these are not usually present on smaller tubes (the type you're likely to own), the only way to replenish the gas is to put the tube on a vacuum/gas system, break the seal, and go inside.

    Operation at high currents especially, with improper power supply regulation (resulting in plasma oscillations), overheating, or just long hours of use, can lead to failure of the filament structure, excessive sputtering with degradation of the Brewster windows or mirrors, and other irreversible damage. And, accidents can happen as ion tube are relatively fragile structures. "You dropped the wrench where?" :( Any of these would require a more extensive amount rebuilding commonly called a refurb.

    Whether a gas refill or total refurb is involved significantly affects the expenses involved. Depending on how much needs to be done and the size of the tube, the cost could be anywhere from a few hundred to a few thousand dollars. Where it is closer to the low end of this range, either of these is still likely to be a bargain compared to the prices of new ion tubes. For example, for a typical air-cooled laser like the Spectra-Physics 161B, typical list prices for replacement tubes are: 15 to 18 mW - $1,900; 20 to 25 mW - $2,750. Just think of what the biggies cost! OK, I'll give you a hint: You can buy a nice new automobile for the same amount of money!

    However, On a small air-cooled ion tube (say less then 350 mW), the cost in materials and labor to do a complete refurb exceeds the cost of buying an entire replacement system surplus (about $1,000 if it's done properly with a new cathode, decent welds, etc.).

    For just the refill, there will still probably be many hours of actual work over several days involved to gain access, pump it down, backfill, bake, pump it down again, backfill, bake, etc., and then seal it, with an extended burn-in using an ion laser power supply. During a major portion of this time, the tube will be the sole occupant of the expensive equipment required to perform the refill. If all they do is just put in some new gas and pinch the exhaust tube shut, only half a job was done and it probably won't last long. However, you could get lucky.

    The amount of gas used is much greater than just what is needed to finally fill the tube prior to seal-off since each backfill/pump down cycle wastes a lot. However, even where krypton is involved (which is more expensive than argon), the cost of the actual gas is still probably not a major part of the total expense.

    In addition to labor and gas that is used, you're paying for overhead in the way of the cost of the vacuum equipment, gas manifolds, inventory of gases and gas bottle rental, power supplies needed to test the tubes, rent, etc.

    Refurbs include much more work on top of the gas refill possibly including: cutting out the old cathode (filament and possibly a getter) assembly, cleaning the bore, replacing glass or ceramic to metal seals, installing a new cathode assembly, replacing Brewster windows or mirrors, etc. These steps are even more labor intensive and the cost of the replacement parts in this case IS significant.

    How much do you think an aspirin tablet costs to manufacture versus what you pay?? :)

    Also see the section: Ion Tube Rebuilding in Your Basement?.

    Now, for slightly more details:

    (Portions from: L. Michael Roberts (NewsMail@laserfx.com).)

    The procedure is actually quite complex. First the laser must be disassembled to remove the tube from the head - this process can take 3 to 6 hours depending on the type of laser head. If the cathode, glass, Brewster windows, etc. are in good shape then the tube can just be re-gassed. If not, these and other parts may need to be replaced - which would require disassembly of the tube. This involves glass blowing and/or replacement of the glass to metal seals (it isn't like you just remove a few screws and replace an O-ring!).

    The actual regassing requires that the tube be hooked to a vacuum system with an airtight seal. Then, typically the tube is pumped down in a temperature controlled oven for around 24 hours. The heat helps to get the last remnants of the old gas out of the tube. The tube is then typically flushed with a new gas fill and the procedure repeated. This 'rinse and repeat' is used to remove any trace of contaminants in the tube. After the tube is pumped down and baked for the second time, it is refilled with gas and the seal pinched off. Even the pinching off and sealing of the tube takes great precision and special equipment. The tube is then tested and reassembled into the laser head where it typically undergoes a 24 to 98 hour burn-in to assure that all is well.

    The procedure can consume 2 to 3 working days as well as many kilowatt-hours of electricity and involves a major investment in specialized and precision equipment - as well as a few dollars worth of gas. You also have to take into account that it is not always successful the first time and may have to be repeated. It is also possible that the tube can be damaged in the process and additional repairs will be needed - all of this factors into the cost. The refurbishing market is quite comparative and if prices could be any lower, they would!

    (From: Steve Roberts (osteven@akrobiz.com).)

    Hum, as one who does this from time-to-time, I can perhaps enlighten you a little more. Most lasers are sealed with a pinched off copper fill stem It looks like copper refrigeration tubing, but is actually oxygen free ultra high purity copper, about $60 a foot, and this is pinched off by a $3,500 tool that pinches off the copper and cold-welds it with about 30,000 PSI of force, using carbide dies. Here's the rub, the tube is filled with argon at a few Torr - a very small fraction of atmospheric pressure with argon. Should that pinch operation leak even slowly with a microscopic leak, the tube is shot and the rebuilder will be supplying the laser owner with a new tube, or having to completely rebuild the old one again. One speck of hard dust or a void in the copper is all it takes to ruin that pinch. Why not stick a valve on there you ask? Well, even the best valves leak or worse, will be turned by an overly inquisitive customer, a technician grasping at straws for some unrelated laser problem, or even by vibration of the cooling water. When a laser does have fill valves, they are attached to a another sealed low pressure chamber of argon - they don't exit into air.

    When you have tricky steps like that in the process, you have to bill extra to pad your losses when you have parts of the process that are really beyond your control.

    The header with lead through seals that holds the cathode has a series of glass to metal seals, each one of which is about $200 in quantity one, with a $250 cathode assembly mounted on it. All of these have to be installed with either a moly-mag or gold-indium braze because the use of any other welding process introduces flux into the tube which poisons the cathode and shortens life. The tube looks like iron, but it's actually nickel or something else exotic and it's a pain in the neck to weld. You have to do a weld with no pits or holes in it, even ones that don't go through the weld, as these trap pockets of gas or flux materials that can come back to haunt you. Indium and gold are two of the most expensive metals on the planet, and a typical large tube braze can contain enough of them in one or two joints to make a very respectable wedding ring. This whole thing has to be baked to 700 °F or more under vacuum, leading to possible cracks from stress where the metal bonds the ceramic. Then, the cathode has to be processed so it activates its electron emitting materials, and you risk burning it out at that point. It's also a risk that you may say, repair a leak, try to repump and then find out the cathode is shot. And then finding yourself cutting into the other end of the tube you just repaired. This can turn into a couple of days of work. Keep in mind that some or all of this must be done in a glove box and clean room to avoid contamination.

    The whole vacuum pump and fill valve assembly on the work station must be baked out to drive out water and other organics, using turbo pumps or ion pumps, and if you're doing it right, you also have a $20,000 helium leak tester to spot those tiny leaks that could stop the tube from functioning weeks or months down the road, as well as an equally expensive residual gas analyc-zer to make sure you have a clean vacuum.

    Even at $1,500 a repump and $4,000 to $5,000 a rebuild, it takes a lot of rebuilds and repumps to pay for that kind of equipment as well as feed your family and put money in the bank for hard times. This is why many rebuilders offer services beyond fixing lasers or also work on much less fragile YAGs or CO2s to help pay the bills.

    If you're in the USA, figure at least about $5000 to $10,000 a year per employee in taxes and health care costs as well. And, the process is tricky enough without having to do your own sales, shipping and answering the phone, so you need some help. When you have lasers needing 30 kW and 6 gallons of water per minute, your facility costs are not cheap. It's about 25$ an hour to run a large frame - much more if you figure in the initial costs and rebuild costs and PSU parts. Heck, some of the fuses in a big one are $40 each!

    Don't even get me started on the precision glasswork and polishing it takes to make a hard sealed Brewster window. Now if your making thousands of them, the price does come down a little, but one-offs are time consuming and not cheap.

    As for gas prices, 250 liters of moderate purity krypton was quoted to me at $269 (plus hazmat plus the cost of the bottle), with research grade about $60 more. I could only find one place in the US that would sell it in small quantities, as it is now a commodity item for semiconductor manufacture. Everybody else wanted me to take out a contract and the price changed daily as the stock market went up and down. As far as I'm concerned that's not cheap compared to argon at $10 a fill for 2,500 liters.

    I used to moan and groan about rebuild costs too until I saw what a cheap 'chop and pump' quickie rebuild does to tube life. Figure a large tube is 8,000 to 30,000 dollars from the factory without installation. That can easily justify the cost for rebuilding a commercial or scientific laser.

    (From: Dean Glassburn (Dean@niteliteproducts.com).)

    As one of those people that supposedly fleece others on refilling. Along with spending my entire weekend refilling a white light Lexel that just will not stabilize in pressure, I can attest to the difficulties involved.

    I always have a problem with people saying that it's just too easy. This is why tube rebuilders and reworkers have such a hard time. Until you do this for a living and find out all the shortcomings and have to explain to your customers why your reprocess only lasted 2 months or why you completely ruined their tube, one should be a little less flippant about saying that there is nothing to it and all the parts can be found cheaply. It must be nice to have a cushy job.

    And, if properly done, the equipment alone will cost more than a new 1999 Mustang GT.

    Ion Tube Rebuilding in Your Basement?

    Well, perhaps slightly more than the average basement will be needed. :-)

    For additional comments on ion laser rebuilding, see the sections: About the High Cost of Refills and Refurbs, Caveat Emptor, and Britt Pulsed Argon Ion Lasers. To get an idea of what is involved in getting started before actually cracking the seal on the tube, see the chapter: Amateur Laser Construction since much of the equipment and techniques that are discussed there will be required

    (From: Brian Bohan (camlaser@cambridgelasers.com).)

    If you think Laser repair prices are high, you should have seen the price of plasma tubes before reprocessing companies existed!! Over the years we have talked a few customers though basic regassing, if you are seriously interested in trying it yourself, I would be more happy to give you some pointers.

    (From: Steve Roberts (osteven@akrobiz.com).)

    If you're trying to pump one at home, you need to get the pressure to at least 10-6 Torr during cathode outgassing or they die about 2 minutes after you first light them. Lifetime is proportional to how well you get that cathode matrix sucked out at pump, we try for at least 10-7 Torr or better. Most laser companies and rebuilders use a big ion pump roughed by a cryopump with cryotraps. Death is by loss of cathode emission from surface poisoning and then you get a tube that wont restart. The list of things that will poison a active matrix cathode include nickel, carbon, copper, sulfer, iron, kovar, any gas other then hydrogen and the rare gases, most metals and just about any organic or nitrogen compound. welding and brazing fluxes do it quicker then anything else. everything needs to be TIG welded without flux or else hydrogen furnace brazed. Outgassing the cathode properly takes about 48 hours of slowly turning up the variac, waiting for the vacuum to go back down each time. Cathodes are incredibly spongy - they hold enormous amounts of gas.

    The gas is five nines or better purity. Ordinary tank argon may be used if you trap out the water with a getter. It depends on your gas companies refill practices, but often tank industrial grade argon is quite good. The difference between good industrial and pure grades in the US is whether the tank has been vacuum cleaned and heated in a oven before refill and then certified on a chromatograph or residual gas analyzer of some sort. Otherwise the tank is just roughly purged then filled. For experimenting its fine. The right grade of single isotope argon might pick up some more gain, but who could afford it?.

    On the krypton however, I would not skimp, its very sensitive to impurities.

    One would also like large amounts of decent helium or argon or xenon on one's station to clean the tube, but beware of using pure xenon in a narrow tube, it can and will explode. 20% AR 80% Xe is a good stable mix. One needs a lot of gas to "hot purge" the tube, i.e., fill it, run it on station, then suck it down quickly to pull out impurities. One must reach and maintain 1x10-6 Torr or better during cathode processing when you outgas huge amounts of gas. If you have a Coherent tube and are going in through the reservoir you may not need to reprocess the cathode if the tube is in decent health.

    Start at high pressure and work your way down. Very low pressure fills can get really conductive (extreme negative resistance or thyratron effect) and blow the pass-bank. (Just did that one last week! ouch, about $140 worth of fuses and diodes.) This is difficult if you can't run awhile on station.

    Also be aware of your tube's need for gas - is it a slow flow tube like a BeO bore or a fast flow tube such as a tungsten disk tube where the return bore may be 40 times the area of the arc bore. It's not fun to light off a fast flow tube and watch the plasma race down the bore and then wink out in a second or two from lack of gas.

    You also need really good gauging, at least a well calibrated thermocouple gauge or preferably a zero to one Torr capacitive manometer gauge (expensive).

    Your best weapon in getting a clean vacuum is a little heat! Heat tapes are wonderful, extruder band heaters are better yet and are not that expensive, and they clamp around standard pipe sizes. Heat the tube too. Crud then flows to the coolest part of the vacuum system.

    Change the pump oil OFTEN if you're pumping crud.

    Also, it might be wise to have some pure or inert waste gas flowing out of the connector when you go down onto the tube refill. That way it doesn't see air or water vapor, which makes getting a clean vacuum easier.

    I can't afford to do this at home, but backfilling the whole vacuum system with argon when you're not using it makes getting back down to high vacuum a lot easier.

    You'll need a good cathode transformer on a Variac. preferably a variac that can exceed normal line voltage by 30% or so.

    I hope you have a turbopump or ion pump or titanium sublimation pump as a diffusion won't do it due to oil contamination. Oil gets to the coolest part of the tube fast an sucks in gas, only to release it when disturbed.

    (From: Anonymous (localnet1@yahoo.com).)

    I have actually SEEN a medical I-90 ion laser pumped down by a Welch Duo-Seal rotary pump. It put out right around 5 W after the guy pumped it down, then the power rapidly fell of after a few hundred hours. That might not sound like a lot of life time, but to an experimenter, a few hundred hours is a few hundred nights of fiddling with the laser for an hour at a time. :) If you ever want to go that route, I guess the key is to pump down, and seal fast. Don't give hydrocarbons time to diffuse into the tube. Or, alternately, you could set up a argon purge system so that there is continual gas flow through the tube to the pump, thereby helping to prevent back-streaming. You can pump out through the tip off (exhaust tube) if you have a tip off sealing tool and put some high grade argon in the fill port. Just a guess though. Don't hold me liable if this doesn't work in practice. One more hint: If you go this way, use diffusion pump oil in your mechanical pump (but first make sure the viscosity is what your pump calls for). Diffusion pump oil is more expensive but back-streams a lot less. If you have access to some cheap Fomblin (and it normally isn't cheap!!!) use that, for next to no back-streaming at all.

    Removing Contamination on the Inside of Brewster Windows

    This is perhaps one of the most frustrating types of problems: Stuff (technical term!) stuck to the inaccessible inner surfaces of optics of laser tubes. While there probably is little hope of removing them, there are a variety of techniques that can be attempted before giving up. However, note that in some cases this is not just random debris but a coating caused by a tube run with excessive current or near end-of-life.

    (From: Phil Bergeron.)

    You can bet I WAS nervous that the RF might cause the power supply current to arc to ground through the Tesla Coil(!) when I did this to the Control Laser 10 watt argon laser with the tube lit at 550 V at 35 A. I am here today so it did not do that. These small Tesla coils have like a 1/2 spark gap inside and I guess even with the Rf there was just not enough voltage thank god. I was sweating for sure and imagined I was about to become black and crunchy like on the Saturday morning cartoons. Hey. I did say there was some risk didn't I? :)

    And it did work. The output went from 3 or 4 watts to over 8 watts in a minute. Alas the tube was so dirty the window just got coated again (cathode puke syndrome) and again until I got sick of it and declared the tube was a lost cause.



  • Back to Ar/Kr Ion Laser Testing, Maintenance, Repair Sub-Table of Contents.

    Cyonics Tube, Testing, Hard-to-Start Tubes, Tube and Igniter Problems

    Cyonics Argon Ion Tube

    This is typical of a well-built small sealed internal mirror argon ion tube. Uniphase Corporation acquired Cyonics in the late 1990s.

    For a detailed description and diagrams, refer to U.S. Patent #4,625,317: Internal Mirror Laser (Cyonics).

    Depending on model, these tubes should do 5 to 50 mW or more of 488 blue, 514 nm green, multiline, or some other combination at 9 A.

    It's designed for a 104 volt drop, so you can get it running with just a 7 to 10 ohm adjustable resistor and a bunch of filter caps.

    The tube wants to see about 2.8 to 3.0 V for the filament at 15 to 25 A, magnetron transformers with the HV winding whacked off are a good start. Adjust for 2.7 to 3.0 V across the filament after a 30 second warmup. Make sure it does not exceed 3.2 V after it is running.

    Tube current is 9 A maximum, 4.5 minimum. A simple 'heater' based power supply (115 VAC feeding 400 V, 20 A bridge, a few thousand uF of filter capacitance, and a fan cooled heating element for a ballast resistor) will do for a few hours until you buy or build a more sophisticated unit. But you must include an accurate means of monitoring tube current and will have to watch it like a hawk.

    The recommended fan is 225 cfm minimum, a bit less then for an ALC-60X. The patriot P2B3s are $22.95 from Marlin P. Jones and Associates quantity 1. Make sure flow is sucking air out of the vee shaped aluminum shroud around the tube.

    I (Sam) built the laser head for a Cyonics tube out of the aluminum box formerly from a defunct audio amplifier project, a fan ripped from a DEC BA11K expansion box power supply, and an igniter based on a flyback core. The entire power supply was also constructed from scrap parts. In fact, the only component that had any association with lasers in its former life was the tube itself. See the sections starting with: Sam's Even Simpler Ar/Kr Ion Laser Test Power Supply (SG-IX1/SG-IY1) for details including the purpose of the bicycle tire inner tube and the successful revival of my originally non-lasing Cyonics tube. :)

    Cyonics/Uniphase Mirror Alignment

    After many hours of use and thermal cycles or abuse (e.g., running at high current or loss of cooling), alignment may change enough to significantly reduce output power, possibly to 0 mW. Even if there is no output, all that may be needed to restore full power is a very slight adjustment of mirror alignment.

    These tubes have mirror mounts designed very much like those of modern HeNe lasers, with restricted regions that permit adjustment by careful bending. However, this isn't something to do with Vise-Grips(tm). Aside from being electrically live - usually with no isolation between the tube and the AC line, this is guaranteed to ruin your entire day. For minor alignment to peak power where gentle pressing side-ways on the mirror mount is sufficient, adjusters can be installed on these tubes to do the gentle pressing in a totally reversible manner. Sometimes, these tubes will be shipped with three-screw locking collars installed on the mirror mounts. They look similar to those found on many HeNe laser tubes. See Three-Screw Locking Collars Adjusters on Melles Griot HeNe Laser Tubes. Where locking collars are present, the use of properly insulated hex wrenches will permit correction of slight alignment errors. Where locking collars are not present, it may be possible to obtain a set from Uniphase or from a dead tube. Otherwise, it should be a relatively simple matter to fabricate a set. I have found that Melles Griot HeNe laser full size three-screw locking collars actually do fit the Uniphase argon ion tube. These adjusters were present on virtually all Melles Griot (non-barcode scanner) HeNe laser tubes (as shown above) until relatively recently and there are plenty of dead HeNe laser tubes floating around. (Melles Griot has eliminated locking collars on most new tubes.) The fit isn't perfect on on the tube in my 2214-30SLB - the adjuster's center hole is too wide - but they do work. In fact, there are already four access holes covered with plastic plugs at the correct longitudinal position around the cylindrical head. (Maybe these were designed for a locking collar with 4 holes!) It isn't possible to get three holes to line up quite perfectly but it is possible to orient the adjuster so a well insulated hex wrench with a ball (swivel?) tip can get to all three screws at not so terribly steep angles through three of the holes. Where your head doesn't have access holes, it would be very desirable to drill some.

    Slightly more forceful use of the hex wrench will permit permanent adjustment alignment such that the locking collar isn't needed, or is only needed for the fine tuning. Determine which screw(s) increase output power when tightened and then adjust them in very small increments, backing off afterwards to see if the power stays at an increased level. The idea is to go just far enough and no more. However, it's easy to go too far and mess up lasing entirely. Use with care!

    I had found that gently pressing on the HR-end mirror mount (with an insulated tool!) of my 2214-30SLB it was possible to increase power from around 12 mW to nearly 30 mW or more at 10 amps (according to the power monitor assuming 10 mW/V sensitivity, which tracks quite closely with a Newport laser power meter). By adding a Melles Griot three-screw locking collar to the OC (cathode-end) mirror mount, it's now possible to get 25 to 30 mW out of this old high mileage tube.

    I also installed a similar locking collar loosely on the mirror mount at the front of the head just in case I do any fine tuning. This required removing the light pickup and running time meter PCB and photodetector assembly as well as widening the hole in the plastic insulating plate. The trickiest part was that the 2 wires to the photodiode need to be carefully unplugged from sockets in the PCB before removal. Attempting to tweak this collar may require a super insulated wrench as the front mirror mount is probably the anode and would have multiple kV pulses on it when the tube is started.

    I recently had to realign a Uniphase 2101 laser which had no output. There were no results from gentle pressing. A HeNe alignment laser was needed to identify the OC mirror as being the one at fault. Then, Sam's Special Mirror Tweaker was used to adjust it as best as possible un-powered. When the laser was turned on, the alignment was close enough that there was some output at just above idle current. Both mirrors could then be adjusted live (with several layers of electrical tape insulating the metal parts of the tweaker!). There was only one shower of sparks when one of the filament studs must have shorted to something before the tube lit. :) (Probably only the low voltage filament power, not the high voltage!)

    It is not known how this particular laser ended up in a totally non-lasing condition. There is normally no stress on these mirror mounts and it doesn't seem likely that even the most determined shipping gorillas would be able to mess up mirror alignment so dramatically. I suppose it's possible that someone before me attempted alignment to lost it. After tweaking, it seems perfectly healthy - a 2101-40ML doing 6 mW at idle (4 A) and over 120 mW at max current. The tube voltage is 97 V at idle, 108 V at max current. In light mode, it goes from 6 mW to 90 mW using the power supply knob.

    Testing a Small Ar/Kr Ion Tube

    This was originally written for the Cyonics/Uniphase 22XX style internal mirror argon ion tubes but portions apply to others like the ALC 60X or Omnichrome 532 with minor and obvious changes. Comments on these and some other types are also noted below.

    The non-lasing tests can be performed even before constructing any sort of real power supply for the tube - to determine if it is worth proceeding. The lasing tests can be done with either a simple pulsed power supply (See the section: Pulsed Operation of an Ar/Kr Ion Tube or a proper DC ion laser power supply).

    Preliminary Tests

    Even before attempting to get a tube to lase, there are some very basic tests that can be performed to determine if it is even worth building or acquiring a power supply:

    If these tests all pass, it is probably worth proceeding with the construction or acquisition of a proper power supply for the tube.

    Initial Power Tests

    For initial tests with a home-built or commercial (DC) ion laser power supply, the tube can be run at a very low duty cycle without the cover/fan in place. Power up for say 4 to 5 seconds and power down, just don't rush, unlike HeNe tubes, these do not stablize instantly. This will determine that there is nothing catastrophically wrong with the tube or power supply.

    (Note that some of suggestions below only apply to the Cyonics/Uniphase tube - some others may not survive!)

    You then want to fire it up and run it, say 1 minute on or so. When all other causes of failure are accounted for, the igniter pulse becomes the biggest factor in the life of the tube. It (a) buries gas, (b) blows chunks out of the cathode until the cathode spot forms and (c) deposits junk on the internal optics. I'd say don't worry about it, as the Cyonics/Uniphase tube is really overbuilt. It has the same cathode and anode as a 5 W tube, and has so much metal that you need some time to heat it up. Unlike an ALC-60X tube which is critically cooled, this one will take some punishment, and no amount of energy you can dump into it with a 115 VAC line will rupture or implode or melt it. I have one sitting in the garage that contained all the fragments when the tungsten disks and cathodes melted, it held the vacuum as well.

    You could run the Cyonics/Uniphase tube at 14 amps for short periods of say 1 minute and do no damage at all, provided you have proper cooling and the cathode is no higher then 2.8 V. It's when you log hours and hours of operating with an improper supply that you will damage it, not during short term testing. In fact tubes are sometimes ran without cooling to recondition them a little (but I probably wouldn't recommend doing this at home unless the next stop would be the dumpster!)

    It's a good idea to power up and see what is going on then switch off, but it may take 5 to 7 seconds after ignition for the cathode to come up to final temp as pressure rises from the discharge heat. Hook a voltmeter across a current shunt in series with the tube, if it goes way out of range, just switch off. Don't panic. This thing has a much larger thermal mass then a HeNe tube, and the metal will desorb the pure gas over time, unlike the HeNe glass which destabilizes the discharge when it heats up enough for ion migration and the bore conducts/outgases CO2 and N2.

    I (Steve) do my initial checks for 30 to 60 seconds with the lid and fan off, often taking my time to stick the voltmeter leads across the tube and then the cathode, and my DVM is slow. Then I simply slide on the fan and cover and run it at 4 to 5 A for a while or shut down for 5 mins, which is about a 1% duty cycle. :-)

    You will pop a weak ceramic-to-metal seal long before you ever risk a catastrophic failure. The main safety hazard is reaching in and touching line side. If it fails the discharge will go out when the cathode opens up.

    There is a BeO (beryllium oxide) warning on the tube for those who would actually grind or smash a tube. Recent discussions with a shop that laser machines BeO have revealed that the hazard is only for a certain particle size. They cut it with CO2 lasers every day and set it up so the particles are outside a certain range without wearing space suits and using only a simple HEPA filter to recover the BeO dust for remelting as it's an expensive material with resale value.

    The Ultimate in a Bare Bones Test Supply

    This almost qualifies for inclusion in the laser humor department. Though, it's not as simple as the one using a set of 18 car batteries in series for a "portable" system. :)

    (From: Bob.)

    I remember a fellow who used an Oudin coil and two slice toaster to demonstrate 60X heads that were pulled from photocopiers. The Oudin was used to strike the tube and the toaster for current limiting. (I think he used a battery for the filament but am not sure as this was some time ago). He had a rectifier but no filter capacitor, and when he struck the tube by allowing a spark to jump form the oudin coil to the tube lead, the TV and radio got a ton of white noise interference. Current control was digitally determined by selecting one slice or two. :)

    (From: Sam.)

    Well, I would suggest enhancing this kludge with a filter capacitor but such an approach would certainly work for really quick and dirty testing. No, I am not recommending it! :) The minimum to be used for anything serious would be something along the lines of SG-IT1 or SG-IX1. These aren't really that much more complex electrically but would be a lot safer for you and your tube. See the chapter: Complete Ar/Kr Ion Laser Power Supply Schematics for details.

    Carl's Notes on Testing an Ion Laser Tube for the First Time

    (From: Carl Hannigan (carl@ppsfx.com).)

    You can use a HeNe supply between the anode and filament. There should be a weak ionization flash, and perhaps a sustained discharge glow if your HeNe supply has a strong enough start pulse to break down the resistance between the two electrodes. If the voltages produced by the HeNe supply are too small, you'll see nothing if the tube is good or bad.

    If the tube is glass, and you can see the filament and "getter" area inside it, a sure indication of a dead tube is a whitish or oxidized appearance on the normally silver metallic flash evaporated off of the getter when the tube is first started. If this area has a shiny metallic mirror-like appearance, things are good. If it looks like white powder on the inside of the tube, then it is probably up to air pressure and is dead. If the tube casing is opaque ceramic, you won't be able to see inside to use this test.

    You can positively test for gas integrity with a small Tesla coil, such as the Model BD-10A made by Electro-Technic Products (Go to "Leak and Pinhole Detection"). When it is turned on, and the tip is brought close to one of the clear glass sections of the laser tube, the high frequency radiation from this unit will ionize the gas inside the tube, which will produce a dim bluish glow. If you can see this, then the tube is probably okay. If you can't see any ionization glow, then the tube is probably up to air pressure and is dead.

    If you are attempting to test the tube with a Tesla coil or similar device, it is important for you to stay away from the windows, the filament, or any other glass-to-glass or glass-to-metal seals on the laser tube. A spark from the Tesla coil can penetrate and weaken the seals, and possibly cause a leak. Try this test only on sections of the tube which are clear glass, and preferably 100 mm (or four inches) away from the windows or any glass-metal seals.

    However, even if the tube is gas intact, it could still be at high pressure and be impossible to operate with a good power supply. You'll have to try it to find out for sure.

    High Pressure Tube Versus Power Supply Problems

    One of the common symptoms found when testing a (possibly newly configured) system for the first time is that the igniter fires, the tube flashes, but the discharge won't stay lit. (If the mirrors are properly aligned, there will also be flashes of laser output but mirror alignment may be another unknown so these won't always be present.) There may be a relatively easy test to determine whether the tube or power supply is at fault. The following assumes a small air-cooled ion tube running at around 100 V but obviously can be modified for other lasers.

    If you have an oscilloscope and a HV probe or are willing to risk the normal probe monitoring the voltage on the power supply side of the igniter transformer or blocking diode, see what happens at the instant the igniter fires. I haven't tried this but would expect that if there were a power supply problem, the voltage should dip to well below 140 V - down to the tube sustaining voltage at the peak current of the igniter pulse, perhaps 110 or 120 V, then drop lower, before cutting out. This would also occur if the light preamp were full on due to a circuit problem since in this case the power supply is also turning off the current. However, if the tube is high pressure, the voltage would not get down this low, perhaps not even to below 140 V, then climb up to the point where the power supply can't maintain adequate current through the tube.

    Hard-to-Start Ar/Kr Ion Tubes - Outgassing and Keeping Your Laser Healthy

    If you don't run an ion laser periodically, gas is released from the tube walls and then it can build to such a level that the tube won't start. So, you then have to spark them in the right place with an Oudin coil of the neon sign shop/vacuum lab variety till they start and bake them at full current for a few hours to drive the pressure down by burying gas into the metal parts of the tube. This is easy when you have a plasma core that is hotter then the surface of the sun and has 170 nm extreme ultraviolet photons bouncing around from every ionized atom in the tube!

    An Oudin coil is a sort of hand-held Tesla coil that is adjustable in output and runs off AC line. It may also be called a Tesla coil or spark coil. A new one costs about $160 from neon sign equipment suppliers but these are often found in high school physics labs and other places working with vacuum systems. You may be able to borrow one for for an afternoon (or longer). A home-built alternative is also possible. See: Home-Built Substitute for Oudin Coil.

    The Oudin coil is fun as sometimes you get all the energy stored in the PSU caps to flow through the air in a inch or two spark, pzzzzzap-flash-bang! (But see WARNING below if using a regulated power supply that might not be happy having its output shorted.) Using it around plastic bags of pure noble gases also is fun. You get 6" long sparks or glows. The only glitch is if you turn it up too high around your ion laser tube, it can puncture the Brewster stems which would be bad news indeed. The high voltage low current RF field doesn't damage the PSU or support electronics, even when you see corona off the leads of solid state components (though I wouldn't press my luck with MOSFETs or CMOS devices and especially keep clear of the light sense amp in the laser head!), yet it will light a nearby fluorescent lamp to the point you can read by it.)

    WARNING: The idea is to get the ion tube to start conducting. If an arc develops outside the tube to the HV return, it may damage expensive power supply components not as a result of the RF but due to the high current discharge pulse that follows. The PSU expects to see the normal ion tube voltage drop - an arc in air could be much less than this.

    Plug the coil into the wall and set the knob on the coil for a 2 to 4 cm spark to metal. Then hold it at the cathode-end of the tube where it can hit the air riser (metal box below the fan) and the bare ceramic of the tube with the power supply on (and trying to start). This will totally ionize the whole tube volume and may allow enough current to flow for the discharge to start.

    WARNING: Beware of turning up the spark coil so high you punch holes in the glass to metal seals on the Brewster stems.

    Note: When using a power supply like the Omni-150R with light feedback, a failure of the preamp in the head or component in the control loop can also result in the laser refusing to start with the dreaded 'tick-tick-tick' sound. If anything in the light control chain is pegged to the supply rail (or even if the control is turned down too low), the power supply will reduce current below the minimum required to maintain the discharge as soon as the tube starts thus effectively aborting. The light preamp can be damaged if you mishandled the sensor - it has piezoelectric properties and can blow the op-amp input if banged or dropped.

    If it succeeds, running the laser for a few hours will eventually bury enough gas that the igniter will handle starting on its own the next time.

    This problem results when you don't run the laser for a long time and it outgases, I shudder to think how many perfectly good tubes have been thrown away that just needed a little assist.

    A cure for this is to run these lasers at least 30 minutes every three weeks, like HeNe lasers, they like to run, it takes a couple of months of setting there for the pressure to really build. On some tubes this doesn't happen but it is a common problem even with the large frame water-cooled units and is worse with a krypton fill.

    The Krypton red and yellow lines share the same upper state, and what determines where they fall to ground state is the gas pressure. If your tube is on the low end, you get a lot of yellow; if it is high you get a lot of red. Large krypton lasers have a pressure control pump built into the tube. Two solenoid valves and a gas reservoir form the pump.

    Lower tube pressures cause more gas to be buried, leading to a runaway cycle. An argon tube with clean multi-line optics installed that only emits the 488 blue line is a dead giveaway for low pressure and end of life. However, it could just be dirty optics as the 488 is the highest gain line, it will lase with a 90% reflector, so try proper cleaning before throwing in the towel.

    Also see the section: Tips for Maximizing Ion Laser Tube Life.

    Notes on NEC Argon Ion Tubes

    The comments below were prompted when the NEC-3030 tube in the laser head Ben was using with his version of the home-built SG-IL1 power supply described in the section: Ben's Linear Ar/Kr Laser Power Supply (BJ/SG-IL1) refused to start for no apparent reason.

    (From: Steve Roberts (osteven@akrobiz.com).)

    9 Amps is flat out max for an NEC tube. They go on forever at 6 to 7 amps. I've seen over 40,000 hours when most others only last about 6,000.

    They only do 25 mW or so when new, and measuring blue light accurately without a thermocouple meter or very high quality silicon meter is difficult. If you can power up the light card with a regulated +/-12 or +/-15 VDC supply and the laser has a calibration sticker on the light jacks, it should be fairly accurate, +/-10% or better. Warning, light card and ignite circuit are on the same board - watch for ground loops. NEC brings out the light signal to their PSU for display on a front panel meter, so your unit should have been calibrated somehow.

    NEC tubes want to have a cooler filament than tubes like the ALC-60X - about 2.6 VRMS (compared to 3.0 VRMS for the others). They are really fussy about this - overtemping the filament causes it to outgas which can lead to hard starting.

    NECs are also particularly susceptible to ripple and overcurrent. They have some bizarre start up circuitry in their PSUs to insure ignition without damaging the tube by actually current limiting their start pulse.

    Sometimes for seemingly no reason at all, NEC tubes go south (probably they outgas some hydrogen - not good for the discharge) and need a little starting help. However, first check all the zeners and SCRs on the ignite board as well as for shorted/leaky caps. Since the NEC laser head generates its own boost voltage, something may be wrong on the ignite board. Make sure the filament volts/amps are adequate - cold filaments won't ignite!

    If this doesn't turn up anything, just getting some Tesla type HV near the glass bell may help. Leaking a spark across to the metal fins on the body may do it as well. Set the laser on a wood board along with a battery powered Tesla coil. Snake a single conductor lead over to the laser with the top cover off, then let the wood act as the return to the HV, if conditions are right, you might get some faint sparks to the fins to pre-ionize the gas.

    As soon as the tube starts, you have about 30 seconds to get the hood on (to provide cooling), so don't panic. It needs to run to get the pressure down, DO NOT interrupt the current flow for any reason. (Well, at least not anything less than life or limb threatening!)

    Assuming none of this works, you can consider the following two procedures but they ONLY apply to air-cooled NECs due to a unique structure in the tube. DO NOT attempt this with other lasers:

    Comments on Ion Tube Starting

    The ringing RF waveform from a proper starting coil aids in starting the tube softly, while insuring formation of the cathode spot using the first sharp pulse of the ringing wave. The pulse heats a small random portion of the cathode surface to start arc electron emission levels. The RF pulse also travels down the bore, ensuring a proper plasma formation by providing a ionized path to form the arc. Typically the waveform is at 100 to 500 kHz and the ignitor often has a small HV cap across the secondary coil to form a resonant circuit.

    At low tube pressures near end-of-life, that pulse shape really starts to matter as you start boiling off cathode material with the pulses as the tube really ages.

    On air cooled tubes, a hand-held Oudin coil such as a BD-10 can aid recalcitrant tubes that haven't ran in a while and are high pressure, or tubes that are low pressure near the the end-of-life and need a little help starting. The Oudin coil is a 500 kHz to 1 MHz sustained RF source at about 30 kV. Its "soft" RF with sharp ringing waves and high peak but low average power really can aid in starting if held near the cathode fins. Be careful not to puncture a glass to metal or ceramic to metal seal with it, and keep it away from head electronics. Its energy is looking for a fast way to ground and rarely will make it down a PSU umbilical.

    The Oudin coil has an obvious use to gently check if un-installed tubes are gas intact by setting it for a soft discharge and gently holding it near the Brewster windows in a darkened room and looking for a glow.

    More on Oudin Coils

    (From: Steve Roberts (osteven@akrobiz.com).)

    There is a lot more to a Oudin coil then a simple interrupter driven ignition coil circuit (though this is what it may appear to be from the outside). They have some unique windings for enhancing resonance and some caps in critical places. Mine is from the 1930s, insulated by wax. A few years ago I had to rewax it - paraffin doesn't work, only pure beeswax for some reason (so I had to rewax it again). I wish I had drawn out the details, but that is a true resonating Tesla coil in every sense, not just 60 hz pulsed AC.

    I've never got a start boost off anything but a pure RF spark, not a sharp pulse! - hence the confounded wood block setup I suggested.

    Oudin coils source so little current they generally do not harm the start boards, even the UJTs. It's not uncommon to stick one on a laser for 30 to 45 seconds with no harm to the electronics, its extremely high peak voltage, but NO current, but its broad spectrum RF that seems to contain (and perhaps adapt) to the natural frequency of whatever it's trying to excite, gas-wise.

    When we hit a tube with a true Tesla discharge (not just a few sparks here and there), it ionizes a large percentage of the atom in the tube, as the broad spectrum RF finds and crawls down every possible waveguide mode in the bore to the anode and beyond. This provides a path for the main start pulse through the hydrogen. On some tubes, the getters (if the tube has them) will start reacting and bite down on some hydrogen when they heat up and hot energetic ions are flying around to react with them.

    <begin humor>

    {
    Having your bore transported through wormhole space by a scalar RF field of course helps. With any sort of luck, the hydrogen doesn't get beamed back from the parallel universe. Oops, that was meant for alt.conspiracy... Sorry, wrong forum. :)
    }
    <end humor>

    The basic BD-10 Oudin coil, Tesla coil, spark coil - whatever it's called - is adequate. One supplier is Electro-Technic Products. Another one is EGL - Your Neon source (click on "accessories", "spark coil"). They both sell 5 or 6 versions. The cheapest is perfectly adequate.

    These also turn up on eBay quite frequently more likely to be called "Hand-held Tesla coils" or something similar. Sometimes, the listings associate them with quack medicine of the early 20th century but that's fine as long as they do what you want. :)

    BD-10 Anatomy and Repair

    The BD-10 is the classic hand-held Oudin, Tesla, or spark coil that runs off the AC line using a set of magnetically activated contacts as an interrupter. There are no active electronic components inside. The BD-10 consists of two sections. For lack of a better terminology, I'll call them the "rear" and "front". The rear consists of a heavy primary coil and thin secondary coil in what appears to be an autotransformer type of configuration. The primary is in series with the AC line and the interrupter contacts. One of the contacts is attached to a magnetic piece which is attracted to the core of the transformer when current flows through it. The other contact's position is adjustable via the knob on the bottom which thus determines how much pressure is applied to close the contacts - the more pressure, the longer the field builds up in the transformer core before current is interrupted by the contacts opening. Thus, when the adjustment knob is fully clockwise, the highest voltage is generated. Spacer washers can be added under the knob to limit the maximum voltage if desired.

    The rear and front sections are attached via two heavy wires. The front section is fully potted so I don't know what's inside but at its end is the output high voltage contact.

    The only user serviceable parts inside a BD-10 are the power cord and interrupter contacts. The transformer in the bottom section could conceivably be rewound if needed but it doesn't appear as though that would be much fun. Only Steve knows what's inside the top section but he's not telling (see the previous section).

    I acquired a vintage BD-10 (original General Electric) off of eBay for $30 that certainly looks and feels exactly like the one we used to test for vacuuum leaks in the cyclotron at our high school. Oh, and also for chasing unwanted visitors. :) (See the sections starting with: The Central High School Cyclotron.) See Electro-Technic Products for a sample photo (next to "Pinhole and Leak Detection").

    It worked fine generating sparks over an inch and a half long, but had a frayed and rather stiff cord, which I decided to replace. Everything inside appears to be in pristine condition, so maybe it's not really that old. However, there may be some minor differences in construction on really modern ones. Here's a step-by-step disassembly procedure for reference:

    1. Pull off the pointed metal electrode (or the "therapeutic applicator" if you have one of these originally intended for quack medicine!).

    2. Pull off the cylindrical piece behind it that holds the electrode - this may take a bit more force.

    3. At the bottom of the resulting recess is a 10-32 nut. To remove the nut may require a super thin-wall 5/16" nut driver or suitably improvised tool. The nut unscrews in the normal direction - counterclockwise. (Unscrewing the front plastic housing may loosen the nut along with it if it's not too tight. That might be risky though. See below.)

    4. Remove the knob by turning it fully counterclockwise and then some. One or more fiber washers may be present to limit how far the knob can be tightened, providing a settable limit on the maximum voltage. Don't lose them. The knob just screws onto the brass shaft until it's tight with nothing locking it in place.

    5. There are a pair of holes on the rear of the rear housing that may be filled with sealer or putty if no one has been inside before you. Gouge out the material with a pointed tool to reveal screws underneath, and remove them.

    6. Unscrew the front plastic housing (normal counterclockwise direction). Once loose, this should go easily as the internal section remains stationary. If there is any resistance, jiggle things to free it up. You don't want to rip the wires off that go between the front and rear sections (though that would be difficult). (If you're using this to also unscrew the 10-32 nut, make sure it turns with the housing.)

    7. Carefully pull the entire assembly from the rear housing.

    8. Inspect the contacts for evidence of wear or pitting, and proper positioning. They should not be anywhere near touching with the brass shaft on which the knob attaches fully counterclockwise (about 1/8th inch spacing) since this is how the device is turned off. Repair or adjust if needed.

    9. If removing the old cord, make a note of the wire locations and unsolder them. Take care not to overheat the lug for the black (Hot) wire excessively as it attaches to the very fine secondary wire and is held in place with a nylon screw. (The screw may need to be tightened after soldering.) Loosen the strain relief and remove the old cord. Any 3 wire grounded cord that fits will be acceptable as a replacement.

    Reassemble in reverse order.

    Home-Built Substitute for Oudin Coil

    Oudin coils or hand-held Tesla coils aren't the sort of thing to turn up at your typical garage sale (although you might find one at a high-tech flea market or hamfest). However, it should be possible to construct something suitable for dealing with hard-to-start Ar/Kr ion tubes at minimal cost.

    The typical hand-held Oudin coil (may also be called a Tesla coil or spark coil) from external appearances would seem to be just a line powered high voltage transformer driven like an ignition coil with a buzzer type of primary interrupter. However, from the section: More on Oudin Coils this is not the case (at least for some types) and produces significant radio frequency (RF) energy - not just a stream of pulses at 60 or 120 Hz.

    A high voltage high frequency source can be constructed from a TV or monitor flyback transformer using a transistor or two for drive. Alternatively, an interrupter (relay/buzzer) or low pulse rate transistor driven circuit using a flyback or ignition coil can be enhanced with suitable resonant components (additional capacitors) to promote the generation of RF energy. Note that if a flyback is used, it must NOT have an internal rectifier as that would simply result in voltage building up on the stray capacitance of the wiring, tube, etc., until a discharge took place inside the tube or an arc to something occurred outside - which isn't what we want.

    Some of these home-built devices won't have as great a maximum voltage as the genuine article but should still be adequate for our purposes. This may actually be an advantage as there is less likelihood of damaging the tube or seals with lower voltages.

    Igniter Problems and Troubleshooting

    Very often, when a tube won't start, it is the igniter rather than the tube or power supply that is at fault. The list of possible problems and tests below assumes an igniter design like the one described in the section: Ar/Kr Ion Tube Pulse Type Igniter (which is similar to that of the ALC-60X/Omni-532 and several other laser heads described in the chapter: Complete Ar/Kr Ion Laser Power Supply Schematics. This uses a thyristor (SCR) to discharge a small energy storage capacitor through the primary of a high voltage pulse transformer (similar in many ways to a xenon flash trigger circuit). The SCR is triggered either from a relaxation oscillator or manual switch. A boost supply (higher than the normal DC+/DC- of 150 V max or so) is used to power the igniter.

    WARNING: Ion lasers are usually not isolated from the power line. Take extreme care when making measurements. The laser itself must be powered via an isolation transformer (preferred) or the test equipment must be isolated (dangerous). For tests of a dead igniter, a modest size isolation transformer can be used as long as you don't actually start the tube (you won't succeed). Note that most isolation transformers DO NOT isolate the ground (third prong). Thus, if using one on a scope where signal ground and earth ground are tied together, there will still be fireworks.

    The easiest way to narrow down the fault to the laser head or power supply is to swap in a known good unit. Of course, not everyone has this luxury. :) Some simple tests:

    It may also be possible to test the output of the igniter by disconnecting it from the tube (with power off and everything discharged!) and then powering up and checking that it will arc 1/3" to 1/2" to a high value high voltage rated resistor connected to the HV return and/or with the main DC+ disconnected.

    WARNING: DO NOT attempt to arc directly to the HV return or ground as it is quite possible that the main supply will then discharge through the air in a spectacular flash-bang which will also blow pass-bank (MOSFETs or bipolar transistors) and other expensive parts!

    Note: Not all igniters are this energetic. Some types may be functioning properly and producing adequate start voltage for your tube even if they cannot pass the spark test. For example, Cyonics tubes start easily and can usually be fitted with a less powerful igniter. I haven't found such tests particularly useful in any case - if the SCR is charging to a reasonable voltage and triggering, the igniter output pulse is probably fine.

    Assuming that none of this helps, there are a variety of faults that can result in a smaller than normal ignite pulse or no pulse at all:

    More on Diagnosing "Tick-Tick-Tick" Problems

    Here's what to do if you were swapping heads or PSUs and now your laser just gives a "tick-tick-tick" (or even possibly nothing at all).

    (From: Steve Roberts (osteven@akrobiz.com).)

    Replace the light card op-amp and transistor in the 532 head, if its "light" signal is defective or oscillating, in some cases the PSU won't start.

    They fry if you sneeze. Mainly it's from being close to all that HV. One can't remote the card however as stretching the wires leads to noise pickup and oscillations. Short plasma tubes can break into damaging 100 kHz oscillations if they don't have a light loop. That's why there is always some light feedback in a OMNI system, even when wired for current mode. There is a pot labeled NOISE that adjusts the level of AC coupled anti-noise feedback, independent of what the light level pot is set.

    Also check all the ignite caps in the head. Build a little voltage doubler with a series resistor to charge each cap (unsoldered from the card). They should charge up to 400 volts or so and hold their charge a few seconds or more.

    It's also a good idea to check the head blocking diode with a HV supply. They can look just fine with a ohm meter on diode test, but then break down under HV. Also check the insulation around the blocking diode, if you have the older one with mylar washers, sometimes these tear, and you get a breakdown.

    Make sure you actually have a start boost voltage (400 to 600 V, not just the 150 VDC or so tube voltage). If you have no boost, the igniter will just faintly "tick" with no plasma glow. Often, the head ignite caps pop if there is a temporary over voltage spike from the supply, and just a "ticking" sound means you might have lost the boost voltage or the cap. Also, a "no anode voltage" condition will cause "ticking", so check your cables and head diode.

    test the igniter by attempting to arc it to the HV return if the main DC+ is connected! This will blow your MOSFETs immediately as the main discharge follows the igniter arc into something close to a short circuit. The MOSFETs need to see a plasma breakdown from the tube to adsorb that igniter energy. This isn't a beefy high energy arc lamp supply - it has just the minimum components to do its job on 115 VAC, let alone when rewired for 230 VAC. Omni makes a good product, but you have to obey the rules, set it up correctly, and leave it alone. There is little tolerance for tweaking or playing, it has to be set up right and left alone.

    Keep in mind that supplies generally have to be "matched" to a given head following a procedure in the manual. It's not a good idea to run a 106 to 109 volt tube from a supply set up for a 160 to 170 V tube like a 543/643.

    Matching is described in the PSU manual. It involves some jumper settings and resetting the loop gain and noise pots while monitoring with an ISOLATED floating oscilloscope. Also the filament tap on the transformer must be adjusted for the right voltage, which will be different on a 532/543 as 543 has a larger cathode coil. I only have manuals up to 155, I don't have docs on the 160/170 series. Matching is a non-trivial exercise as you can blow MOSFETs in the process if you don't follow the procedure exactly. What happens is this, if a supply is misadjusted then the MOSFETs have energy flowing through them when they are only biased halfway on or off, resulting in a blown MOSFET. A MOSFET is just a switch, it cannot really dissipate very energy by itself. That MOSFET has to turn off and on in 100 to 200 ns or so. If the turn-on or turn-off time isn't fast enough, then the MOSFET will heat up and blow. When the MOSFETs goes they takes little things with them, like the fast diode and some small electrolytics around that area. The failure can also damage the MOSFET gate drive circuits.

    Quite often, these sorts of problems start with a cabling or misconnection fault and it gets perplexing because minor damage results in both the PSU and head.

    Mike's Omni-543 Revival Saga

    (From: Mike Harrison (mike@whitewing.co.uk).)

    WARNING: The following contains material that some laser-lovers may find uncomfortable. However, please be assured there is a happy ending!

    A while ago I 'rescued' an Omnichrome 543 head at an auction of mostly junk, for the princely sum of UKP55 (including a car full of other rubbish that came in the same lot). After some playing, and replacing a dead igniter SCR, I got it running, and was seriously pleased to now have a nice powerful multiline argon ion laser in my collection, my only other "Argon experience" having been a Cyonics 2201. I spent much time building a nice power supply for it, but a few days later, it began refusing to start.

    After my first cry for help on the USENET newsgroup alt.lasers, Phil Fostini wrote me this very helpful message:

    "Omni tubes tend to like to strike down the gas return path of the tube. This may cause some outgassing and later make the tube harder to start. Boosting the prestart voltage up may help. Also, if it does restart keep it running for a while (5 hours) at a 6 amp current should help. Short "ON" times on tubes that have been sitting may also cause outgassing to happen. When allowed to run for some time it will clean up and you should see a change in the tube voltage. If all else fails there is the car battery trick but I strongly advise against that."

    I tried increasing the boost voltage, but no joy. It appeared that the boost cap was being discharged, and removal of the optics (so it couldn't lase) and looking down the bore seemed to confirm this - a bright-ish flash but no proper discharge. The igniter strike did seem to veer off to one side, instead of being a nice bright spot in the centre of the bore.

    So I now knew that if I could get it to strike just once, it could probably be recovered. So how to get it to strike?!

    I had heard of highly risky last-ditch tactics involving car batteries from Phil and others, but decided that this was somewhat extreme, and I was not (quite) that desperate yet.

    To summarise, below is the list of things I did to attempt to revive it. I don't know how many of these things helped (indeed some may well have made things worse!) but the bottom line is that it is has now struck properly, and will be running for the next day or so until I get the courage to turn it off.

    I'm sure I'm not the only one here to have experienced that feeling of sheer elation when a presumed-dead laser suddenly bursts into life. I don't even care that it will probably take me a couple of hours to realign the optics when I put them back on! I'm truly a happy bunny!

    Many thanks to Phil Fostini and Chris Leubner for helpful advice.

    Comments on Ion Laser Filament Testing

    (From: Gernot Stoffel (Beamchief@gmx.de).)

    I lately had the problem to proof function of a Spectra-Physics type 092-S metal tube's filament (from eBay!), that can't be seen directly, if you don't have X-ray eyes. It just behaved as a solid copper-block short, even on my 4-1/2 digit precision multimeter; no way to get another result with it. Thus, I was afraid this filament could be molten down in some way, shortening its length, so that proper tube function was impossible. I don't have any matching ion laser power supply yet so had to find another way of testing.

    So, sending 3.00 Amps from a proper stabilized DC power supply through that filament, I got 58.5 millivolts of voltage drop - and a pretty hot power supply's heat sink, too! ;-) So the filament resistance was 19.5 milliohms. This was little more than 1/7th of the specified resistance (1/8th Ohms) that could get calculated from the specified operation voltage and current.

    Tungsten has a *strong* temperature shift of resistance, I know (another reason, beside temperature bearing, to make that material ideal for light bulbs). So I tested a cold light bulb's resistance, finding that it was just 1/15th of the hot operation resistance (which can be calculated easily from specified bulb power and mains voltage, R=V2/P).

    Thus, you can think that my filament result just means: RELAX!

    Summary: Cold filament resistance seems to be about of 1/7th of operation filament resistance; for measurement just take a stabilized current of some amps and a precise voltage multimeter. Calculate V(drop)/I(stab)*7*I(spec) and you must get something like V(spec).

    (OK, here comes the 'think' section, too: Melting-down a filament of that kind would need at least triple the temperature of normal operation (about 20 times of cold resistance, for light bulbs stay stable at 15 times value). This means 80 times (!) higher power dissipation (goes by the forth power with temperature); and THAT would need about 15 times of the specified voltage (calculate yourself if you like)! Not even the heaviest power supply crash would produce this; so the whole problem might be *very* academic. A filament can be broken by shock or maybe erosion; but really can't melt down under any realistic conditions!)

    Any Hope for an Ar/Kr Ion Tube with a Broken or Damaged Filament?

    For a filament that is actually broken - continuity shows open or there is a visible gap - the short answer is no. Unless you are into amateur laser construction, do whatever your favorite religion calls for and give it a proper burial (actually, see below). However, such tubes (intact) or parts like Brewster stems/windows and mirrors may still be useful if you are into home-built lasers.

    The problem is that in order to sustain the discharge, the cathode/filament must be at the proper operating temperature. Unlike a fluorescent lamp, the arc can't sustain this. If filament power is lost during operation, the discharge will drop out in a few seconds on most power supplies - and this is hard on the tube. As it cools, the emissivity goes down and the voltage drop sustained at the filament will rise with proportionally increased power dissipation - several hundred watts more than it is designed to support when properly heated. If the power supply has enough compliance range to sustain the discharge with the increased voltage, bad things will happen to the filament pretty quickly.

    You may still be able to use the tube in short pulse mode for experimentation but life will be limited since each pulse will take a chunk out of what's left of the filament. At best you can get two or three pulses per second out of a dead tube. See the section: Pulsed Ion Tube Test Circuits.

    However they will not run cold cathode unless ran at very low pressures in a pulsed regime, or taken up to very high pressures with addition of a helium fill in a hollow cathode discharge or short large diameter arc. Then you only get one or two blue lines. The pressure would be around 40 to 50 Torr or so of helium and 350 millitorr or so of argon.

    If you aren't into pulsed operation and/or trying other gas fills, it is probably best to cut the Brewster stems or mirror mounts off for possible use in your home-built laser projects. But DON'T grind, file, or chip away at the ceramic - beryllia, BeO), wrap the tube in a few layers of plastic for environmental reasons, and dispose of it as solid waste, but not as scrap steel. It's not steel actually, but is usually a nickel alloy, as iron poisons cathodes.

    All of this assumes the filament is actually open. If a couple of its coils are shorted together and the bore is clear, you can try running the filament on reduced voltage (resulting in about the normal color) and run the laser at reduced tube current (maximum being proportional to the remaining hot part of the filament). This may or may not work and for an undetermined amount of time but could be better than nothing.

    Note that if the filament just looks bad with coils jumbled together or drooping, as long as the bore is clear, your tube may still work for a long time.

    Running a High Power Ion Tube in Your Basement

    (From: Steve Roberts (osteven@akrobiz.com).)

    You actually will have an easier time getting a big tube going then a small tube. It's real easy to overdrive a small one.

    First check that you can maintain water flow and MONITOR it. My water source is a well with a submerged pump powered by 220 VAC 136 feet down. Loosing drive to the pump would be an obvious problem, unless you have city water. Flow switch won't click out all the time on any model ion laser. They do have some flow induced hysteresis. You need to measure flow when doing this at home with larger tubes. A washing machine was switched on once on me and I nearly lost a MRA tube which was just switched off. I could hear the water boiling! When it was ALMOST too late, the water got hot enough from residual heat to melt the plastic hose fitting and blow out. That tube is just fine now, but required a day to dry out in its frame and needed a severe optics cleaning from the "rain". A Proteus flow switch with an analog readout added across its integrator cap is now mandatory when using that setup. Anyhow, here goes:

    1. Build/buy a clone of a Meditech or American laser switcher. On a good day with a tailwind I have easily gotten 25 amps down a 257 volt I90MRA, and that's about 12 volts too high in pressure. A corrected tube then ran at 27 to 32 A, but the air cooled meditech starts to oscillate at 32 to 33 A. Going to a water cooled cold plate cured that. You really need to put the filament transformer on a Variac or preferably, a constant voltage source at higher amps. When the line sags, the emission goes down and the tube falls off as you try to ram higher currents down it, hitting the upper limit in many cases. READ 4.5 WATTS IN THE BASEMENT FOR A FEW MINUTES AT A TIME. Same drop thing for the filament applies to the magnet and without a half decent magnet, most tubes won't light or even stay lit. Adding 2 to 3 times the regular number of caps improves things a lot, but the inrush surge can be a bummer. It needs 12 to 14 big darlington switches, but the inductors are small. Switching to MOSFETs requires tricky inductors. Fuse each transistor, or else!

    2. Hot water heater elements. Two to three of the big ones in parallel with a monster cap bank works well provided you use the hot coherent igniter with the pulse peaking and rectification network. Coherent igniters have a pulse forming network on them and that ignite pulse will take a small tube like an I60 or I70 almost to lasing on its own! Of course you need to add a few transistors across one of the resistors for light regulation and ripple killing. But using a FW Bell current sensor makes that easy. The standard hot water heater element threads into 1" pipe.

    3. Under test: Massive MOSFETs but cheap ($6 each) with no buck/boost transformer in linear mode, on a custom designed high flow heatsink. If that heatsink cant dissipate 1,000 watts in one second at 1 gpm, don't bother. if you don't monitor the outlet temp and inlet temp, don't built this. :-) Key here is the MOSFETs are almost isothermal on that heatsink, so their source resistances track, so if one heats up and goes higher Rs, the others will track with it. If they are spread far apart on the heatsink or the temps don't track well, boom! simple three op-amp circuit controls it, but current mode only. A Bruce Rodgers concept, and it works well. I'm can't talk about the heatsink design, that's proprietary. But not toooo difficult. However, you have to watch out for pressure induced warp (hint).

      To get real high power, you have to solve the cathode/magnet dropout issues, and thats not easy. but 3 watts all day out of a tungsten disk bore argon isn't that hard. Most small frame whitelights or kryptons have a tube voltage that is 20-25 volts down from a argon, and we get .4-.7 watts out depending on the tube, I've gotten 400 mW running kryptons from Lexel-88 supplies, and more than 700 mW off a single line red I90K.

      This is not for most amateurs, you have to really watch your cathode conditions and delta T. cathode monitoring circuits are a must if your going to do this for light shows. coherent takes this so seriously that they have been including a cathode current interlock on many of their lasers for years. You need to back off on the upper limit when running for long periods of time, so you have enough magnet and filament.

    Replacements for NEC GLG Argon Ion Laser Tubes

    (From: Steve Roberts.)

    National, Evergreen and LASOS, as well as JDS Uniphase all make drop in packages that replace a GLG, right down to the connectors. So did Spectra-Physics. as long as you can match the beam height, wavelength(s) and polarization (either polarized or random) the laser will work. You're most likely to get a unit from National, followed by Evergreen. I've known a lot of people who have had no problems with either evergreen or national. You'd be migrating your customer to a ceramic tube with easily double or triple the life and much better mode stability. As far as I know there hasn't been a bulk import of the NEC/SHOWA head into the USA in 15 years. There were tonnes of dying ones dumped on eBay in the mid 90s, and that was it. As it costs about $1,500 to rebuild a glass tube, vs. $800 for a ceramic one, they went out of favor fast.



  • Back to Ar/Kr Ion Laser Testing, Maintenance, Repair Sub-Table of Contents.

    Ar/Kr Ion Laser Lasing Problems

    What Could Prevent an Internal Mirror Tube from Lasing?

    If it has good pure argon at the right gas pressure, a working fan, the proper stable current down the bore and reasonably aligned clean mirrors it will lase. The 488 nm line will lase from .01 torr to 2-4 torr even with dirty mirrors, so it's almost guarenteed that you will get something. I doubt there is dirt or scratches in the optics as those would not leave the factory.

    Where these conditions are met, the most likely explanation is very slight mirror misalignment.

    It would not surprise me that just applying pressure in the right direction to one of the mirror mounts by slipping a steel tube (WARNING: High voltage - use something well insulated to grab it!!) over the mirror mount without loosening the set-screws (if present) will cause the beam to pop on. It should operate once the mirrors are somewhat aligned. Tubes like the Cyonics/Uniphase have a curved output mirror and a flat high reflector. Even if slighlty misaligned, this configuration is set up to tend to correct itself optically from small error but peaking may still be beneficial.

    Where there are locking collars with set-screws similar to those on many Melles Griot HeNe laser tubes (see Cyonics/Uniphase Argon Ion Tube Cathode-End), a somewhat less drastic procedure can be used to tweak alignment. These adjusters/locking collars are what the manufacturer actually uses to make final adjustments in alignment at the factory. But, with many thermal cycles, they can loosen resulting in a degradation output power. WARNING: Both ends are attached to the high voltage - take extreme care to insulate your hex wrench! CAUTION: Only very small adjustments are needed and these don't use fine thread screws. Only make adjustments with the tube lasing - if you lose the beam, it may be necessary to go through the entire alignment procedure from scratch!

    Mirror alignment can be tested and corrected using techniques similar to those described for sealed HeNe tubes. See the section: Checking and Correcting Mirror Alignment of Internal Mirror Laser Tubes in the chapter: HeNe Laser Testing, Maintenance, Repair as well as the section in this chapter: External Mirror Laser Cleaning and Alignment Techniques.

    Laser Beam Varies in Intensity

    A beam that comes and goes, flickers in intensity, oscillates (particularly at some multiple of the power line frequency), or starts out strong and gradually dies out with warmup, may be due to several causes. However, the most likely is that the tube current is set very close to the lasing threshold. On unregulated power supplies, even slight ripple on the main filter capacitors will result in significant current variation down the bore of the tube. As components heat up (like the current limiting resistors), the current may change slightly as well. And, as the tube heats up, there will be some change in its lasing characteristics. Running at a slightly higher current will probably make the symptoms disappear as far as the appearance of the laser output is concerned though there will still be variations detectable with a laser power meter. I've also found that on my Cyonics/Uniphase tubes at least, switching to a higher current for awhile and then back to the low setting will result in a higher output for a few seconds. I don't know why this should be the case - whether it is a tube or power supply related phenomenon but the current (on low) doesn't seem to be affected.

    Of course, first check to make sure the cooling system is fully functional. For air-cooled systems, this means the fan is sucking or blowing (as spec'd) at full speed, cooling fins free of obstructions, etc. For water-cooled systems, this means the flow rate is correct and the water temperature is low enough.

    Where the current is definitely well above threshold or increases as the tube warms up with light mode attempting and possibly failing to maintain selected output power, first check on mirror alignment. It may be on the hairy edge or may be changing due to thermal stress, particularly with internal mirror tubes. This has been found to happen with a National 60X replacement which went from lasing at full spec to almost non-lasing even at well above its rated current due to some shift in one of the mirrors.

    Erratic beam intensity may also be an indication of plasma oscillations. See the section: Plasma Oscillations and Other Instabilities.

    Note that this sort of behavior in a system with light feedback could indicate a failure of the sensor or control circuits or need for adjustment.

    Laser Beam Shifts Position Erratically

    The symptoms may be that the position of the beam jumps by a milliradian (e.g., 1 mm at a distance of a meter) or so over a period of a few seconds or minutes. It may do this in a cycle or with no apparent pattern.

    This could be due to mode cycling or mode hopping and is probably much more likely on a short tube. There are multiple paths down the bore of such a tube that can lase and very slight changes in mirror alignment due to thermal expansion can cause the active mode to jump between those that are possible for the gain curve. I expect that the longer the bore, the less likelihood of this happening. For example, on a HeNe laser with its high ratio of bore length to bore diameter, the effect is a very detectable change in output power but minimal if any shift in beam position.

    A simple test to confirm mode hopping as the cause would be to very gently press on each mirrors mount (with a well insulated tool!) to see if the behavior can be triggered by a slight change in mirror alignment. Just touching the mount may actually be enough. Assuming that the tube and/or resonator is properly designed, adjusting mirror alignment with the locking collar(s) or mirror mount screws may be all that is needed to remedy this annoying behavior.

    Of course, mechanical instability elsewhere could be affecting the beam direction - make sure the problem is actually in the laser!



  • Back to Ar/Kr Ion Laser Testing, Maintenance, Repair Sub-Table of Contents.

    Tips for Potential Ar/Kr Ion Lasers Enthusiasts

    Caveat Emptor

    A local 17 year old paid $700 for a NEC 3030 with 20,000 hours on it, it was missing the hour meter when he got it, it was rated for 40 mW and doing 10 when he got it, it took all of two months of intermittent use for it to die, and that was half a summers's spare wages for him, that could have been avoided by putting a voltmeter across the tube and measuring its drop, which is a function of pressure etc. WARNING: Lethal voltages and currents around the tube.

    A spark coil became a permanent resident at his house so he could somewhat prolong its life, by assisting it in starting. We got the 20,000 hour figure from a second hour meter buried on the start board. The higher the tube drop voltage the better, you measure it at the high and low current ranges and compare against a known good unit.

    There is a certain popular laser surplus company that buys these units, chops the seal off, fills them with a fresh argon fill at a lower pressure for more power, but doesn't bother to bake them out or replace the cathodes or Brewster windows or drill the bore out so it's clear of metal migration. They then sell them for $3,800 after setting the PSU upper limit to 11 amps or so. So you get a 175 milliwatt laser that lasts 3 months and dies from a dirty tube and bad cathode that was not reprocessed right in order to save time. While I can't print the name in a FAQ for fear of being sued, I can say: Ask your seller how your tube was processed. He should say: Oven bake out under vacuum for 24 hours, flashed the getter assembly, reactivated the cathode, lased it while on the pumping station or after reprocessing to check it, and replaced the Brewster angle windows and possibly the cathode. IF THESE STEPS WERE NOT DONE, think twice or get a good warranty and make sure they will likely still be in business when should you need to use it!

    With a complex laser - especially where the resonator mirrors are external (not inside the sealed tube) - it is extremely important to get the needed equipment and a manual so you can maintain and clean the laser. The air in a home environment is not really as dust-free as in a typical lab - and they still have problems with optics needing cleaning periodically even if not being used. However, don't overboard cleaning everyday just because you think the power has decreased - every cleaning is an opportunity to accidentally scratch the optics - and that will mean permanently lower performance!

    Above all, understand the safety implications of a higher power laser of this type. Furthermore, when first started, an ion laser may operate at or above its maximum rated power for a short amount of time - regardless of the control settings. Therefore, additional precautions are a good idea. A set of eye protection goggles for the range of wavelengths that your laser produces is highly recommended. Can you place a price on each of your eyeballs?

    Argon Laser Anonymous

    A big warning here is these are addictive, suddenly you find yourself trying holograms and illuminating low clouds and looking at the ramen scattering spectra of beer.



  • Back to Ar/Kr Ion Laser Testing, Maintenance, Repair Sub-Table of Contents.

    Cleaning of Laser Optics

    What's the BIG Deal About Cleaning?

    ALL of the optics (mirrors, Brewster windows, Littrow prisms, etalons, etc.) used inside a laser cavity are extremely delicate and easily damaged by contact with their surface or improper or excessive cleaning. Ultra-fine scratches that you wouldn't notice in a million years of eyeglass cleaning will result in a degradation of your laser's beam quality and output power. This is particularly true of the coated dielectric mirrors but to a lesser extent also affects the optical glass components as well. Cleaning, even when done properly, invariably results in some degradation of the optics surface.

    The reason this is so critical for external mirror lasers is that photons inside the cavity bounce back and forth dozens or hundreds or more times and every little scratch, speck of dust, smudge, or other blemish in the as close to perfect as possible surface can deflect, absorb, or otherwise corrupt the beam before any of it makes it way out the end of the laser! Even though your smile may appear to be reflected without degradation, in reality, the mirror may be useless for its intended purpose.

    The sections below provide information on how to get the maximum life out of your optics.

    Note: While these techniques may be overkill for optics outside the cavity (e.g., the external surfaces of the OC mirror on internal mirror HeNe laser tubes), they certainly won't hurt. And, getting into the habit of using the higher grade chemicals and taking proper care will prepare you for dealing with the more finicky optics inside the laser cavity. However, if all you are interested in at the moment is cleaning your little HeNe or other internal mirror tube, see the section: Internal Mirror HeNe Tube Optics.

    And, some options for cleaning aluminized bounce mirrors (not dielectric laser mirrors!) include:

    However, one must still take care not to scratch the surface with contaminated or abrasive wipes. And it won't hurt to use the same procedures as for high quality delicate laser mirrors!

    Note that some experts caution against cleaning aluminum or protected aluminum with isopropyl alcohol as it may degrade its surface.

    Make Sure Your Mirrors Aren't Going to Dissolve!

    The recommended chemicals and procedures below assume hard-coated dielectric mirrors for HeNe or ion lasers on a substrate like optical glass. Those of you with CO2 and excimer lasers with salt optics or group III-IV semiconductor optics will need other materials.

    Some older lasers may have soft-coated optics using dielectric materials that may be water or alcohol soluble before. These were manufactured in the days before they could reach the higher temperatures needed to do nice things with materials like hafnium oxide and titanium oxide. I don't know what is used to clean a soft-coated optic without damaging it terminally. :-( Probably ethyl acetate or something equally exotic. It is likely that water, alcohol (isopropyl or methyl), and acetone, will degrade the surface almost instantly, rendering the mirror useless. If the active area is clean, gently blow off the dust but otherwise leave well enough alone!

    How to tell if you have soft-coated optics? Carefully test the very edge of the coated area in a location that doesn't matter. If it is something you don't want to deal with, a bit of pure water, alcohol, or acetone (see the cleaning procedure below; drop and drag, don't rub!), will result in mottling, fine striations, or an otherwise ugly appearance to the surface after drying. The beam of a HeNe laser reflected off such an abused mirror onto a white card will show distinct interference patterns due to diffraction from the damaged surface.

    Consult your laser supplier (HaHa!) or optics manufacturer for more information (but you may need to dig their expert up from the grave for 30 year equipment!).

    In addition to the mirror surfaces, take care in attempting to clean the Brewster windows or mirror mounts of soft-sealed HeNe or ion laser tubes with alcohol or other solvents as the result may be immediate air leakage and a dead tube. The failure mechanism for this isn't clear - after all, it can take weeks to loosen up these optics by soaking when trying to salvage them for some other use. However, there is anecdotal evidence to suggest that instant tube death may result from such cleaning attempts. So, to be safe, avoid getting the area of the sealing adhesive wet with solvent.

    Chemicals and Supplies for Optics Cleaning

    For hard-coated (or uncoated) glass optics, you will need the following:

    1. Ultra high purity methanol - gas chromatograph grade or spectroscopic grade.

    2. Ultra high purity acetone - gas chromatograph grade or spectroscopic grade.

    3. For really bad fingerprints, hydrogen peroxide (3%) and high purity laboratory distilled water.

    4. Sterile Throat Swabs in individual packages on wood sticks without glue. Puritan(tm) and Qtips(tm) brands don't seem to have glue when purchased from a pharmacist in sterile form. Unwind the cotton carefully to look for glue. (Have your friendly Pharmacist order you a case - it's cheaper!)

    5. Lens tissue from a lab supplier in sealed envelopes. Using generic photographic tissue is not recommended. When I did a survey of laser refurb techs and light show techs, most recommended the Kodak tissue as it is sold in sealed packs. Many inexpensive tissues have impurities that can scratch soft laser optics.

    Alcohol or acetone sold over the counter in the United States may contain denaturing agents or tracking chemicals to prevent unintended uses such as drug making. Content (other than the alcohol and water) can vary greatly. Some 70% percent rubbing alcohols are quite pure while some 91% medicinals left a very noticeable hard-to-remove residue. A simple test of any solvent is to put a drop on a piece of glass and let it dry. If there is no visible residue, it's probably good enough for general electronic/mechanical cleaning and maybe cleaning of external optics. However, without actual testing, I would not recommend either of these for use inside a laser cavity. Additives/impurities will show up as a white film on the optics. In any case, it's best to get these chemicals from a lab supplier who can attest to the purity. (Kodak or Omnisolve is preferred). Keep the lids on the bottles so as not to pick up moisture from the air, and always pour out what you will need into a smaller container to avoid back contamination of your primary source from swabs or lens tissue. I flow warm filtered argon gas into my bottles before resealing. This aids in shelf life of these expensive fluids. DO NOT attempt to use common isopropyl (rubbing) alcohol for optics cleaning as it will leave a thin polymer film on your optic that is hard to see but really decreases lasing power. You can tell if you have water in your fluids as they will 'ball up' into droplets that hang around after the film of cleaning fluid evaporates.

    CAUTION: Hard-coated optics are only hard in a relative sort of way. No matter how careful you are, any contact with the surface degrades it slightly or worse. Therefore, don't be a cleaning fanatic - only do it when absolutely needed and use the proper materials and procedures. And make sure you have hard-coated optics *before* it is too late!

    Optics Cleaning Procedure

    Here are two cleaning procedures. The "drop and drag" method is potentially gentler but can't be used on all optics.

    The "wet and wipe" method should be used for mirrors and surfaces inside the laser, small mirrors you have a hard time holding in your hand, or mirrors where the holder prevents limits access to the sides of the optic, cavity optics for low power lasers (i.e., HeNe or air-cooled ion, HeCd, etc.), optics that need a real scrubbing, light pickoffs, tilted optics surfaces, or mirrors smaller then 10 mm.

    It is hard to precision swab a 10 mm mirror without a lot of practice. It usually requires holding the mirror sideways in your fingers to swab it, and you're likely to drop it. You usually leave contaminants at one edge of the mirror with a swab. This is why most large ion lasers offset the mirror surface from the face of the holder, to enable drop and drag cleaning. I strongly urge klutzes like myself (Steve) or nervous nellies to make a holder for cleaning their small optics. I'm getting older and I drop them more and more. (I've been known to drop $600 mirrors on the lab floor. --- Sam)

    For larger optics that can be removed from the laser with the entire surface is exposed, use the "drop and drag" method if possible.

    Wet and Wipe Method

    These two steps will be repeated for each optic surface. Note that quartz (often used for Brewster windows) is fairly robust. However, even modern hard dielectric coatings and common optical glass are much less forgiving. AVOID attempting to clean a dry surface - that is just asking for scratches. Use the technniques described below.

    1. Take a swab, wet it with acetone, let it set a moment, then flush it with more acetone. Wipe it from the top of the optic downward with a rolling motion. You want to just scrub the optic. This takes some pressure, but be gentle. Do exactly one pass with each swab and then discard the swab. Break it in half so you do not accidentally reuse it. Do this 2 to 3 times and then let the optic dry and examine it with a bright light for residual dust and films. Use lens tissue if very dirty, otherwise use swabs. Let gravity work to your advantage to get the contaminants to flow downward.

    2. Repeat the above with methanol till clean. Acetone kills grease, methanol cleans the surface.

    DO NOT clean any optics mount rubber O-rings as these will contaminate the optic with byproducts after the laser heats up if cleaned with other then distilled water.

    Here is a another similar method found in the instruction manual for the Spectra-Physics model 120 helium-neon laser:

    1. Using dry nitrogen, blow away any dust or lint on the optical surface. This step is extremely important as any dust or lint left on the optical surface can result in a scratch that will damage it permanently. If dry nitrogen is not available, an air bulb can be used to generate low pressure air for the same purpose.

    2. Wash your hands thoroughly with liquid detergent. This step is important for the reason that body oil and contaminants on the fingers can be transferred to the optical surfaces during the cleaning process resulting in re-contamination.

    3. Draw some (spectroscopic grade) acetone into an eye-dropper and squeeze out one drop (or two if necessary) to cover the optic surface. Then take a piece of lens tissue, place it on the wetted surface, and gently draw it across the optic to remove the contaminants that have dissolved or floated to the cleaning solvent surface. Use a lens tissue only once.
    (Note: This instruction manual dates from 1970. I do not know whether the mirrors in this laser were hard-coated or soft-coated. Acetone is safe for hard-coated optics but may damage at least some types of soft-coated optics.)

    Drop and Drag Method

    This requires practice on a round piece scrap glass before doing it on a optic. Lay a piece of high grade optics tissue lightly over the optic. Handle the tissues only by the edge. Grip the optic holder in your weak hand (i.e., your left hand if you are right-handed) and the tissue in the strong hand. Balance the tissue on the optic. With your strong hand drop methanol or acetone (if badly dirty or greasy) onto the tissue over the optic so it flows through the tissue and saturates on the optic. Raise your weak hand up a little and tilt the the optic and tissue toward your strong hand so it's about a 15 degree angle from vertical. You'll note with practice that when a certain amount of methanol has evaporated, that if you drag the tissue slowly and carefully across the optic, a white line of dry tissue appears, yet you still have some wetness holding down the side of the tissue moving with the tissue movement off the optic. In other words, a diminishing region of tissue has enough surface tension to hold the dry tissue down and you both swab the optic and 95% dry it in the same pass. You visually watch the evaporation as the tissue drys. Surface tension forces drive most of the dust and crud into the moving fluid, and the optic is literally pulled clean. You do not want to wait until the whole drop starts evaporating evenly across the optic, hence the 15 degree tilt that will favor one side to start evaporating. It's not easy, and it requires a good pure solvent (e.g., spectroscopic grade as described above). It won't work with water, only polar solvents that have a high vapor pressure.

    Links to Other Similar Methods

    In addition to the info above, there are a couple of Web sites that should be of interest.

    Polymer Based Optics Cleaning

    Photonics Cleaning Technologies (alternate Web site: Photonic Cleaning Technologies) had a product called "Opticlean" which consists of a special polymer in a fast drying solvent that is brushed onto the surface to be cleaned. After letting it dry completely (2 to 10 minutes), the resulting film is peeled off taking with it dust, dirt, grime, oil, and other contamination. Opticlean is suitable for bare glass surfaces as well as those with hard-coatings like most laser mirrors and intracavity optics. I have used it on laser mirrors with mixed results. At times it is truly impressive. After deliberately placing a very messy fingerprint on a HeNe laser HR mirror (as an experiment - really!), one application of Opticlean left the surface in spotless condition. On another mirror used as an OC for my one-Brewster laser tube test rig, it produced a surface as clean or cleaner than using any other technique I've ever tried. However, at other times - for no apparent reason - the resulting surface was no cleaner, and possibly even worse. Multiple applications might help. I would guess that the performance depends on the type of contamination but I don't yet have a good feel for predicting what happens, or what pretreatment would improve the situation. But the good news is that I have not seen any evidence of damage to glass optics or hard-coated mirrors from repeated applications of Opticlean, even if the surface doesn't end up spotless.

    They now have a supposedly new product called "First Contact" which I have not tested. It apparanetly has replaced Opticlean. The only obvious change is that it is red instead of clear.

    However, these products are not suitable for cleaning soft-coated or plastic optics as they may be damaged by the solvent.

    I have no affiliation with the company but it is a product I will be using. My only complaint is that the plastic bottles they put the extra solution in are apparently not solvent tight as it has partially evaporated from mine. But, admittedly, it's probably somewhat past the expiration date being a couple years old!

    Vapor Phase Optics Cleaning

    (From: Steve Hardy (hardy@sweng.stortek.com).)

    Another way of cleaning optics and other delicate objects is to construct a vapour phase degreaser.

    Get a metal can (e.g., a soft drink can with the top cut off) and make a loose fitting lid. Wrap copper pipe around the top of the can for cooling water.

    Suspend the object to be cleaned, face down, from the lid. Put methanol or some other suitable solvent in the can and heat over a hot plate (not an open flame!).

    When the solvent heats sufficiently, its vapour will rise to the top of the can. Pure solvent condenses on the object and drips back to the bottom, taking dirt with it. The cooling water creates an appropriate temperature gradient as well as economizing on solvent.

    Removing Dielectric Coatings?

    OK, so maybe this shouldn't be in the "Optics Cleaning" department, as it means cleaning off the coatings entirely. Exactly the opposite of the usual objectives, above.

    Here's the situation. You have some dielectric coated mirrors that have been seriously damaged, perhaps by improper cleaning, and thus useless for their intended purpose - say inside a laser resonator. Is there any way to remove the coating so they could possibly be used, say as a beam sampler or weak lens (if curved)? With old soft-coated optics, this would be no problem as almost any solvent will dissolve the coatings. But with modern hard-coated optics, the coating itself may be harder than the substrate and impervious to almost anything that won't also damage the substrate. The use of any abrasive would require repolishing of the substrate at the very least. A strong enough acid to dissolve the coating would probably also etch glass.

    The only way I know of to do this that even comes close to removing the coating without totally destroying the substrate is with ultra-fine steel wool, possibly helped slightly be soaking for a few weeks in water or a non-acidic solvent. By now, the professional optics types are probably rolling on the floor ready to hit the "reply flame" key on their PC. :) (Send me email if you like.)

    But I have done this on among other things, a couple of SP-125 OC mirrors that looked like someone had already tried to clean with sand paper. There were major scratches where the coating had been entirely scraped off all across the surface. These wouldn't even work half decently as external turning mirrors, let alone inside a laser cavity. However, the same "someone" had apparently not attempted to clean the AR coated outer surface, which was pristne. Since I wanted to preserve this, I put a generous amount of OptiClean (see above) on it and let dry to protect the AR coating. And this might also assure that it was really clean when removed!

    Gently rubbing with 0000 steel wool gradually wore away at the coating. It took a very long time to actually remove it entirely, perhaps equivalent to a half hour of constant rubbing. Considering how easy it is to damage these coatings by careless cleaning, this seems a bit amazing. And the layer directly on the substrate must be special, because that required even more time to remove, and it was impossible to entirely remove a fine ring where the mask had been during the coating process. But when done, the resulting surfaces were remarkably free of major defects. Only a few very fine scratches were evident under a bright light, possibly caused not by the steel wool but by the original abrasion to the mirror, when it was still a mirror. Sending a 10 mW green laser beam through these reprocessed optics produced virtually no detectable scatter or distortion in either the transmitted or reflected beam.

    The results using this technique won't be a pristine optic and it does take awhile. But it may not be as bad as one might think. And, at the very least, the amount of subsequent polishing that would be needed should be smaller than if an actual abrasive were used. Of course, your mileage may vary since there are all kinds of materials used for both hard coatings and optical glass (or other) substrates. In the case of my mirrors, the coating must have been softer than the steel wool which was softer than the glass. That may not be true in general.

    P.S. You didn't hear anything about this madness from me. :)

    (From: Salmon Egg (salmonegg@sbcglobal.net).)

    There may be some way to remove a dielectric coating without damaging the substrate, but I don't know what it could be. Hard coatings usually contain refractory oxides and fluorides. Aside from grinding and polishing, acids are typically used. In particular, acid of MF2 will generate HF which will etch silicates including glass. If the soak is carefully monitored, you might be able to catch the process before much surface damage occurs.

    Coatings containing silicon such as silica or silicon monoxide will require HF to remove. The only hope is that you might be able to undermine coatings in there that are not as tenacious.

    Also remember, most optical glasses are not as resistant to chemicals as lab ware. The best bet would be to use a borosilicate crown glass that resembles Pyrex brand chemical glass.



  • Back to Ar/Kr Ion Laser Testing, Maintenance, Repair Sub-Table of Contents.

    External Mirror Laser Cleaning and Alignment Techniques

    Lasers for Which These Procedures Apply

    The following was developed for a typical 100 mW external mirror argon ion laser (henceforth referred to as 'Argon'). It will, of course, also be suitable for a krypton ion or mixed gas laser. See the section: Maintenance, Alignment, and Modifications of the ALC-60X Laser Head for specific information pertaining to that laser (and the Omni-532) AFTER reading this set of instructions all the way through.

    This technique also works for copper vapor, CO2, 2 mirror YAG, and many other types of external mirror lasers including all similar type home-built lasers. It will also be suitable with obvious modifications and simplifications for lasers with an internal HR and external OC.

    (Note that what follows includes a more sophisticated version of the general procedure described in the section: Major Problems with Mirror Alignment for sealed internal mirror HeNe laser tubes. However, the basic principles are similar.)

    If you already have a working argon laser or *green* HeNe laser, you can do red HeNe lasers. These are aligned at the factory using the 488 nm or 457 nm lines of an argon laser. Hold up an unpowered (red) HeNe tube so a light shines down its bore and you will usually see a deep blue light transmitted through the mirrors. This is what we're going to use to our advantage. This technique was developed the hard way after discovering the techniques described by those who write laser books are not exactly tested in the lab and often written by someone who has graduate students to do it for them. The 'Cards with Crosses' technique only works on lasers that are nearly aligned. The approach described below works on anything including newly installed tubes that are not yet centered in their cradles as well as for very short lasers.

    My (Steve Roberts) thanks to Dale Harder at H&H Laser Refurb for teaching me this neat little trick for initial mirror alignment. Dale is the ultimate prefectionist. His lasers exceed their specs.

    It's also the only technique for low gain lasers short of an autocollimator or factory computerized search mode alignment jig.

    This technique works on the fact that the optimized dielectric mirrors used in laser cavities are often largely transparent or only partially reflective at wavelengths at least 100 nm away from the design wavelength.

    Mirrors within a milliradian (mR) of parallel is a hard angle to achieve, and that's what you are shooting for.

    Moral: Once it's aligned and tweaked, don't mess with it!!

    THE AUTHOR OF THE FOLLOWING ASSUMES NO RESPONSIBILITY FOR YOUR ACTIONS OR SAFETY. LASERS ARE FRAGILE, DANGEROUS, EXPENSIVE DEVICES. PROCEED AT OWN RISK!!!

    IF YOU ARE NOT COMFORTABLE WITH ATTEMPTING THIS OR THINK YOU ARE DOING SOMETHING WRONG, GET SOMEONE TO DO IT FOR YOU, THE GENERAL FEE FOR THIS AVERAGES $100 to $200 AN HOUR PLUS MATERIALS!!

    SAFETY: NEVER look down the bore of the laser you are aligning if it is under power. While extremely unlikely, all the wrong circumstances could converge to result in it lasing when you don't expect it! (There is also some UV from the discharge which isn't good for you.) When power is off, there is no danger except from the alignment laser beam passing all the way through the bore. Electrically, there is no risk of shock with all covers in place. But if adjustments need to be made inside, there could be exposed high voltage terminals. It is essential to unplug the power supply from the wall outlet and confirm that its main filter capacitors are fully discharged and/or disconnect the umbilical cable before touching anything inside the laser head. The ignite/start card capacitors can also hold a charge which isn't dangerous but it wouldn't hurt to check them just the same. The last thing you need is to be startled in the midst of delicate optics! Also see the general info in the section: Laser Safety and for ion lasers in particular: Argon/Krypton Ion Laser Safety.

    CAUTION: Some of the following steps require the removal of at least one mirror either for cleaning or during the alignment procedure. While most large-frame lasers use tubes with external mirrors, some do not. Obviously, it could ruin your whole day to remove a mirror and bring the tube up to air. If the mirror doesn't want to come off after removing expected screws, don't force anything!!! It's often possible to do the alignment with both mirrors in place by simply misadjusting the mirror closest to the alignment laser so that its reflections are off to one side. And, there is no need to clean the inside surface of permanently sealed mirrors!

    Required Equipment and Other Stuff

    1. Alignment jig or optical table (described below).
    2. A 5 to 7 mW HeNe laser with clean beam (henceforth referred to as the 'HeNe').
    3. A fluorescent orange (preferred) or yellow sticker (get these from office supply companies).
    4. A long thin sewing needle.
    5. A lab jack or custom made HeNe mount to set the beam height of the HeNe (see below).
    If you will be cleaning the optics (which is probably a good idea while you have the laser partially disassembled and assumed below, see the section: Chemicals and Supplies.

    The Alignment Jig

    I can't afford optical benches on my budget. I needed a long bench to work on my lasers. The solution was to go to the local aluminum company and see what they had in the computer as leftovers from a larger sheet that was cut to order for a customer. A 1/4" thick 3.5 foot long 14" wide piece of T6061 polished on one side was $40. A 16 foot piece of 1" x 1" finished square stock was 25 dollars.

    Initial Bore Alignment

    Watch out for stray beams that come off the Brewster windows!!!!!

    DANGER: Class III and Class IV power levels, use laser safety precautions appropriate for your laser. When your Argon is lasing, terminate its beam on a proper beam stop where you can't see the blocked beam. A cumulative exposure to a medium power reflection can cause eye damage.

    DANGER: DO NOT stare at diffuse or specular reflections!!!!

    You are now ready to deal with the Argon:

    I am assuming you have the output end of the Argon facing the HeNe.

    Cleaning the Optics

    The following applies to lasers with external mirrors. For internal mirror HeNe and Ar/Kr ion tubes, only the outer surface of the OC needs to be cleaned and this is a lot more forgiving than the optics inside the cavity.

    Old lasers often have a lot of dirt in them despite the O-ring seals, and getting rid of this dirt is a major concern as you don't wish to burn it into your optics. (Recall that there will be light flux with from 10 to 100 times the output power of your laser bouncing back and forth between its mirrors depending on its size and optics (e.g., 5 to 200 WATTs of circulating flux for 100 mW to 20 W ion lasers). It's a judgement call as to when this is needed. I (Steve) could not figure out how to provide general guidelines on this.

    See the section: Cleaning of Laser Optics for additional information including the required chemicals and supplies, WARNING about soft-coated optics, and other Web sites with optics cleaning procedures.

    These two steps will be repeated for each optic surface (starting with the Brewster windows):

    1. Take a swab, wet it with acetone, let it set a moment, then flush it with more acetone. Wipe it from the top of the Brewster Window down-word with a rolling motion. You want to just scrub the optic. This takes some pressure, but be gentle. Do exactly one pass with each swab and then discard the swab. Break it in half so you do not accidentally reuse it. Do this 2 to 3 times and then let it dry and examine the window with a bright light for residual dust and films. Use lens tissue if very dirty, otherwise use swabs. Use gravity to your advantage to get the contaminants to flow downward.

    2. Repeat the above with Methanol till clean. Acetone kills grease, methanol cleans the surface.
    Now repeat the same procedure on the high reflector mirror until clean and let dry. However, the coated mirrors are MUCH less robust than the quartz Brewster windows - be gentle! Clean the back side of the optic as well as the face, as contaminants migrate.

    DO NOT clean any optics mount rubber O-rings as these will contaminate the optic with byproducts after the laser heats up if cleaned with other then distilled water.

    Aligning the Rear Mirror

    Install the high reflector (HR, rear mirror) into the far end of the Argon making sure the coatings face inward.

    Slowly adjust the optics mount screws so that the weak REFLECTED HeNe beam from the front coated surface of the Argon HR mirror is visible on the face of the HeNe, then carefully walk the HeNe beam so it is right back into where it came from. If this is really good, the HeNe will flicker from you canceling out lasing with a third mirror, but this is rare. Note that each mirror will reflect 2 beams, 1 from the front and one from the back of the optic. You want the one off the coating. Note also the Brewster Windows may generate more reflections as well. You will end up with a bunch of dots dancing on the HeNe, keep track of the one you want. Keep working until you have a stable tight mirror mount with the beam centered in the hene beam. then back off some slack to leave room for adjustments of the screws, and recenter the beam. You're WORKING with FRESNEL reflections, or about 2 to 3 PERCENT of the HeNe beam - a few hundred microwatts at best - so turn off the room lights to see the weak beams.

    HeNe laser with reflected dot shows the HeNe alignment laser on its 3-point adjustable mount. The reflections of the HeNe beam from the mirror being aligned on the argon laser (off of the lower right corner of the photo, not shown), can be seen on the fluorescent sticker. In this case, the mirror still needs some more work!

    Aligning the Front Mirror

    Clean the FRONT Brewster window using the two step procedure described in the section: Cleaning the Optics, above.

    Carefully install the FRONT mirror (nearest the HeNe) after cleaning both sides using the same procedure.

    Carefully align it using the rear mirror procedure, above.

    Powering up the Argon - Final Alignment

    Making sure you have proper cooling for the argon laser, leave the HeNe on and switch the Argon ON and turn the tube current up to the upper limit, then back it off a little. This should be About 9 amps if you have a 10 Amp maximum laser. Let the laser warm up. DO NOT EXPECT IT TO LASE AT THIS POINT. If it does laser you are very lucky or you have a large frame (e.g., 1 meter long) laser.

    VERY SLIGHTLY loosen the rear mirror mount, NOT THE MIRROR ITSELF!! and slowly press it against the Mount holder or REAR PLATE of the laser, rock it back and forth slightly while doing this, you will see a small flash of laser light on the HeNe face.

    WARNING: THIS PROCEDURE IS FOR LASERS LESS THEN 250 MILLIWATTS ONLY. For bigger lasers, see the vertical search procedure in the laser manual or use the fine adjust screws or search bar!!! This flash will tell you which way you have to move the mirror mount screws, usually opposite of the way you have to hold or tilt the mirror.

    Once you have it steadly lasing, see the section: Walking the Mirrors in External Mirror Lasers to tune the resonator for maximum power and best beam quality.

    If you don't get a flash, repeat the alignment procedure until you get it!. This take patience and time. Commercial laser optics have a small amount of wedge to avoid creating ghost beams which interfere with the lasing process. This wedge may be what is messing you up as you may be trying to align the incorrect reflection!

    Walking the Mirrors in External Mirror Lasers

    The following applies to external mirror lasers where the mirror adjustments at both ends are accessible and permit small, precise, repeatable changes in alignment to be made easily.

    For a laser tube without screw adjusters but with compliant mirror mounts, see the section: Walking the Mirrors in Internal Mirror Laser Tubes for a modified (painful and risky) procedure that applies to common HeNe and argon ion internal mirror lasers.

    If your laser produces any sort of beam and the alignment of both mirrors independently is optimal (either after testing and/or after correcting it as described in the sections starting with: External Mirror Laser Cleaning and Alignment Techniques then its time to optimize the output power and beam quality by adjustments to both mirrors.

    Indications for the need of further alignment include:

    See Effects of Walking the Mirrors for an exaggerated (hopefully!) illustration of why this happens. As can be seen, although the mirrors may be perfectly parallel to each other and there is still some output, by not being aligned with the bore/capillary, portions of the beam are cut off, less than the full amount of gain medium participates in the lasing process, and there can be reflections from the walls and other structures in the tube to create artifacts.

    For external mirror lasers with fine adjustment screws on the mirror mounts, the "Walking the Mirrors" procedure isn't really at all difficult and can usually be performed quickly and painlessly without much risk of losing the beam entirely.

    For all measurements of output power, a laser power meter is highly desirable. It doesn't need to be fancy since maximizing power is what's important, not an accurate value. Anything that will convert photons to a meter reading will be fine including the absolutely trivial ones described in the sections starting with: Sam's Super Cheap and Dirty Laser Power Meter. It's just that your basic allotment of eyeballs isn't very good at detecting small changes in intensity! :) Note that mode cycling of your HeNe tube will result in small variations in output power - these can be annoying but need to be mentally discounted in determining the maximum power output readings.

    That's it! Now, if you aren't totally obsessive-compulsive, you will lock down the mirror adjusters and get on with your life. :)

    Peaking a Multi-Line Argon Ion Laser

    This specifically refers to the ALC-60X but should also apply to other models as long as the color/wavelength of the weakest line is known. Note that the power balance of the output lines also depends on tube current do this should probably be performed at the tube current that is to be used.

    As with any alignment, NEVER turn any bolt so far that the beam disappaers without going back to a lasing position!

    (From: L. Michael Roberts (NewsMail@LaserFX.com).)

    You can use a "quick and dirty" method to see if you can peak up the laser output power. Get a diffraction grating and bounce the laser off it so that you can see the coloured spots on the wall. You will want to be well back from the wall so that the lines (coloured spots) are separated and do not overlap.

    While looking at the deep violet spot, carefully adjust the X bolt on the back of the resonator. Careful adjustment means slowly turning 1/16 or 1/8 of a turn in one direction. If the laser becomes dimmer and/or the violet line disappears, STOP turning and return to the former position. Then try carefully turning in the other direction.

    What you are looking for is the brightest possible violet line output. Once you have obtained this using the X adjustment bolt, try doing the same procedure on the Y adjustment bolt. If either bolt had to be turned at all from it's starting position to increase violet output, then you need to go back and re-peak the other axis (in other words, adjust X for max violet, adjust Y for max violet, re-adjust X for max and then re-adjust Y for max). NEVER under any circumstances adjust the bolt that forms the right angle corner of the triangle described by the three adjustment bolts.

    DIsclaimer: While I have performed this procedure many times and it is relatively simple to do, you can dis-align the laser and loose all of your output if it is not done cautiously and correctly. The author, his company, his family, his heirs and assigns is/are not responsible in any way for any problems or damage to the laser you may cause by following this procedure.

    (From: Steve Roberts (osteven@akrobiz.com).)

    While LMR's quick and dirty method is the first thing I'd do when I got a unit with some hours on it, keep in mind that plugging a analog meter with a wide scale (you need that needle to see trends and digital doesn't sample fast enough) into the side light jacks will help with peaking and watching thermal drift over time. You can then log your light versus power curve with a digital meter. Forget about calibration, the light jacks are never accurate and will never match the .02 V/mW calibration claimed on a multiline laser. Make sure you use a isolated meter not connected to the power line! You'll need to push in the button on the side of the head to get a reading. The light jacks float about 70 volts above ground so be careful!

    But when it comes down to it, if you determine your cavity is majorly off alignment, then do the adjustments as Xrear Xfront, then Yrear Yfront, and so on, otherwise you'll find that doing random adjustments to X and Y with out a power monitor will eventually walk you off peak.

    Do you have a much fuzzier/larger diameter beam then when you got it? These are the principal signs of dirty optics.

    To really track your tube's condition, you need to monitor the tube voltage over time, and chart it versus amps, it falls off as tube pressure drops. Did the original tech send you a checkout slip with a tested and specified tube voltage when the laser was shipped?

    How Laser Manufacturers Align Lasers

    With the exception of fiber pigtailing which is easily and accurately automated, most factories use the visible beam technique - basically variations on the various methods described in this chapter - for HeNe, ion, CO2, solid state, and other common laser types. Solid state guys who can afford it may use an autocollimator.

    An autocollimator is a telescope with a light source and beamsplitter cube built in, with a reticle of some form to measure the misalignment , if its a quality one, it looks like a survayers instrument with the graduated mounts. Cheapies are $700 or so new. Obviously not worth it for doing a few lasers.

    However, while a good autocollimater spaced far away would get you a factor of 10 to 100 improvement over the visible beam technique, the final tweaking is still a human task. Some of the more expensive ion lasers using "Beamlok" or "Beamtrack" have stepper motors on the high reflector and can find lasing in two to ten minutes, but the initial bore alignment is still done manually. If your bore is very small, don't invest in the autocollimator as it won't help.

    Depending on the autocollimator, you can get every thing from cross hairs to units that read out error in degrees of arc, but its going to be megadollars and doesn't compensate for optical distortion when the rod heats up.

    Most HeNe lasers are aligned at the factory with a human operator, a 3 to 5 mW 488 nm argon ion alignment laser and a simple optical test bench for setting the tube concentric and parallel. (I've been told by an insider that for mass produced HeNe laser tubes, it's even simpler).

    Caution when Adjusting Large-Frame Lasers

    On some medium and large frame lasers, it is best to loosen the Brewster stem covers before proceeding with any large scale angle adjustments of the wobble plates as it may overstress or break the stems on some units. While this is not needed when using the normal adjusting screws, many times in cases of extremely bad misalignment you have to physically grab the plate and twist it, especially if you have just installed a home made optics mount that isn't exactly parallel, are prying on the mount with a screwdriver or other object or are doing a extended vertical search, etc. In these cases, loosening or partial removal of the stem covers may be required.

    For example, when aligning a Lexel-88 or other Lexel laser for the first time, remove the Brewster stem covers so you don't break the stems, the wobble plates move the stem covers as well as the optic.

    Aligning a Coherent I-90 Argon Ion Laser

    (From: Sean (power@linklasers.freeserve.co.uk).)

    Firstly, power off laser. Check that (if fitted) the iris aperture is in the fully open "position-0". Also make sure that the cathode hasn't dropped or sagged. This can be checked by removing the output coupler from the laser and placing a piece of white paper or card an inch or so behind the rear Brewster window. If you shine a small flashlight on to the card to illuminate the card evenly, look through the other end of the laser where the beam would normally exit through the output coupler and see if you can see a clear hole diameter approximately 2.5 mm or 0.1" through the front Brewster window. If there are any obstructions or no light from the illuminated card, you either have a failing (sagging) cathode, mis-aligned iris aperture, or worst case, a cracked tungsten disk within the bore. To verify this, you can turn on the power supply BUT DO NOT AUTO or MANUALLY START the tube, this will allow the cathode to reach working temperature. Again look through the front Brewster window and check visually that the cathode is clear and not obstructing the optical axis of the laser cavity.

    If all this looks OK, then the lasing should be fairly easy to resolve:

    1. Run the tube at around 25 to 30 A. The tube voltage at 30 A should be 227 to 230 VDC.

    2. Replace the output coupler and remove the rear mirror.

    3. Place a piece of white paper or cardboard 2 to 3 feet away from the rear of the laser and allow the projected glow to illuminate the center section of your target paper. (This normally works best if you dim the lights in the room.)

      What you should see is a roundish discharge with a few diffraction rings super imposed, if you gently rock the front mirror plate by inserting a small flat bladed screw driver between the front mirror plates while adjusting the vertical and horizontal tuning screws you will eventually see a brighter dot move across the card. When you see this dot, you will need to steer the dot towards the center of the illuminated discharge on your target. Once found, you have set the front mirror to lase.

    4. Replace the rear optic and the intracavity PTFE seals from the rear of the laser. Adjust the tuning screws so that you can rock the vertical mirror plate back and forth so that the reflective discharge sweeps the rear Brewster window. Again, insert a small screwdriver between the rear mirror plates and rock gently and evenly the mirror plate while making small adjustments to the horizontal tuning screw. This process will sweep the optic in both directions and with a little luck you see should see the cavity flash greenish blue. You are then 99.5% there. Make smaller adjustments to both vertical and horizontal tuning screws until you achieve a stable lasing action.

    (From: Stephen Fels (stephen@fels.cc).)

    3 VAC for the filament is just about right, but current is what you actually want to measure. There should be a white sticker on the autofill with the cathode 'window' (a range such as 24 to 26 A) indicated. If the current is outside that range, the cathode transformer needs to be re-tapped for correct current. However, this shouldn't keep you from lasing, it just optimizes performance and extends the life of the cathode.

    Once the tube starts, make sure you are current regulating and requesting full current. The color and intensity coming out the output bezel will depend on the optic coatings more than anything so the actual color probably doesn't mean much.

    Once you've done a rough alignment, it will be necessary to do a 'vertical-search' to find flash and lase. You can verify that the output coupler is aligned properly by placing a white card with a 2 mm pinhole between the output coupler and the cathode Brewster window, then make sure that the retro-reflection is centered on the hole, when the plasma glow is centered around it. Then remove the rear dust shield and place a white card in the bottom of the head, under the rear Brewster. Position the card so the plasma glow reflection from the high reflector shines on the card (usually an inch or so forward of the Brewster window). Roll the vertical high reflector knob so that the high reflector plate tips down/forward and the reflection from the high reflector scans 'down' the image of the plasma glow (you should see a fuzzy oval of the Brewster window on the white card and a sharper, brighter image of the moving high reflector). Rock the high reflector plate back and watch the high reflector image 'scan' the plasma glow. Continue rocking the high reflector back and forth, while rotating the horizontal knob. If the image of the high reflector travels to the edge of the plasma glow, reverse direction. If you see a 'flash' of lasing, adjust the horizontal for the brightest flash, stop rocking and adjust the vertical knob until you achieve lase. You might try rocking a bit while adjusting the vertical, just to make sure you haven't gone too far, or were rocking on a diagonal.

    A few things to verify if the tube is ionizing:

    1. The shutter is open and the aperture is wide open 'OA'.

    2. The optics are a matched set (visible, UV, etc. if you have the manual, there will be a parts list which will give you the three digit number referenced on the optic (e.g., ####-004-##) and inserted the correct way (the ^ points toward the bore from each end).

    3. The "Visible/UV" switch is in the correct position.

    4. The cathode is not sagging into the beam.

    5. The plasma glow comes out of the output bezel without obstruction (should be round).

    6. The dust shields (and their linings) are not hanging in the way of the beam path.

    Aligning a Trimedyne 900 Argon Ion Laser

    (From: Steve Roberts (osteven@akrobiz.com).)

    NOT AN EASY LASER, RELAX!

    1. If you were doing alignment as with a small air-cooled ion laser, change methods.
    2. Relax.
    3. Laser Ionics mirror mounts stink like a raccoon who sucks eggs.
    4. It's a very critical confocal cavity.
    5. Make sure the intracavity shutter is open.
    6. Big bore lasers really suck to align.
    7. I have never seen so much backlash in a laser gear system as there is in a Ionics.
    8. I like Ionics when they lase, beaucoup output power, like 13 watts.
    9. Big laser optics have deliberate "wedge" in them, they are not parallel surfaces. This messes you up with the alignment laser.

    The following assumes the optics are clean, like very clean. The arrows point to the active side of the coatings.

    Get a darkened room, put the laser up on a table where you can work on it. Make it aligned and parallel to the long axis of the room, as your will be projecting a image. This works best if you can borrow a long focal length white light mirror from somebody that uses the same size. Amber theatrical gels have been known to help too.

    BIG LASER ALIGNMENT DOES NOT USUALLY USE A HENE!

    1. IMPORTANT: Remove the Brewster stem covers!
    2. Make sure you have clean Brewster windows and mirrors.
    3. Remove both mirror mounts.
    4. Identify the OC.
    5. Make sure it's clean, no films.
    6. Light the laser.
    7. Darken the room.
    8. Note the circles of plasma light from the bore on the wall coming from the Brewster windows.
    9. Put white screens at each bore light spot on the wall.
    10. Install the OC, but wiggle it loosely so you can see what its reflection looks like.
    11. As you move the OC around, you'll find it reflects plasma light.

    12. The idea is to center the reflection of the OC down the tube so you see a faint spot on the paper at the HR end. This is the reflection of the plasma light off the OC (hence why a white mirror is better). Center it in the plasma glow. I's a faint bluish white. Some people claim a yellow gel helps at this point. I don't use them.

    13. Stand at the rear of the laser, crank the horizontal control until the HR would be angled way to the left of where the laser would hit the wall. This is the rocking and scanning phase. Back off the vertical so you can stick a screwdriver in and rock the rear mirror plate violently (stem covers off!).

    14. Install the HR. If you can't wiggle the mount with a screwdriver, do it with your fingers. VERY VERY slowly crank the horizontal back toward center as you rock the the vertical. You should see a flash at some point. make sure you don't have crosstalk on the vertical movement as any unwanted horizontal motion when you rock the rear plate messes you up. When you see a flash, peak it on the horizontal while rocking, then slowly crank the vertical back in until you have steady lasing.

    15. When you have lasing, go to the section: Walking the Mirrors in External Mirror Lasers. If you have lasing, leave the OC alone, period.

    If it doesn't work, recenter the spot from the OC and do it again, some times it takes 2 to 3 tries. A piece of white card held near the Brewster with the optic loose will help you find the reflection.

    If I ever wrote a book on lasers, it would be titled: "Again, and again, and again"....



  • Back to Ar/Kr Ion Laser Testing, Maintenance, Repair Sub-Table of Contents.

    Additional Alignment Information

    Markus's Comments on Argon Laser Alignment

    (From: Markus Hakes (mah@josquin.pc.rwth-aachen.de).)

    Instead of dismounting both mirrors, aligning the plasma tube to an HeNe laser, adjusting the rear (HR) mirror followed by the front one as described in the procedure starting with the section: External Mirror Laser Cleaning and Alignment Techniques, I took off only the HR. After that, I put my HeNe in from the rear and aligned it so that the dot was centered coming out of the front mirror.

    Then I used the procedure from the section: Aligning the Rear Mirror but first on the front mirror, followed by the HR, which I aligned very carefully till the lasing began. I then adjusted both the front mirror and HR for maximum power.

    I found this to be an easier way especially because the front mirror was not that easy to remove (the photodiode light sensor is on it) and the adjusting of the argon and HeNe was a bit easier (the bore for the outgoing laser beam is much smaller than the end of the plasma tube, so it was easier to get the HeNe beam in the middle). Now at least I had only to remove one mirror. The mirrors are hygroscopic (they love water), so it is not good to remove them for a longer period than absolutely necessary.

    Stephen's Tips for Mirror Alignment Without Using an Alignment Laser

    (From: Stephen Fels (stephen@fels.cc).)

    Some tricks of the trade, which can save you having to use a HeNe alignment laser:

    For a sealed mirror tube:

    For a laser with a Brewster window tube:

    Vertical search tips:

    The idea behind a vertical search, is to make a search pattern of vertical lines with the rocking, combined with a horizontal motion, so the image of the mirror makes a sort of 'comb-tooth' pattern. If you're careful to rock the mirror on a true vertical and you see flash, take a moment to maximize the flash by continuing to rock and make smaller horizontal movements around the flash point, then you can find lasing by adjusting only the vertical control.

    Other 'Sanity Checks':

    These will make sure you are giving yourself the best chance of success:

    Once you're lasing, be sure to 'walk' the cavity for best alignment.

    Good luck and always be eye-safe!

    (From: Joshua Halpern (jhalpern@neteze.com).)

    We are increasingly using Webcams and other inexpensive video cameras for this sort of procedure as well as to take all the danger out of alignment procedures when laser is on. I've made it to 55 without losing any vision, and I want to make it to retirement. Webcams are cheap and make inexpensive alignment tools. They are also a great help when the technical support line tells you you are dreaming.

    Multi-line Tuning

    (From: L. Michael Roberts (NewsMail@laserfx.com).)

    Theoretically, an argon laser should produce 5 wavelengths: Two greens, two blues and a violet. Try passing the beam through a diffraction grating or a prism to observe if all these lines [colours] are present.

    If you laser is 'out of tune', the violet line will be the first to disappear so you could try adjusting the resonator screws ONE AT A TIME (and without adjusting the apex screw/nut) to see if you get more/less violet.

    DISCLAIMER: If this causes any problems/damage or reduced output - don't come crying to me Adjust the tuning screws/nuts at your own risk.

    Disks for Laser Alignment

    A Spectra-Physics tech just showed my friend at Sea World a neat trick. He had round disks with rims that slip over the optics without touching the face of the mirrors. Each disk has holes that are centered in the face of the mirror. You peak the laser with a disk with a large hole and keep switching to disk pairs that have smaller and smaller holes until the beams are centered on the mirrors, which results in maximum beam quality and power.

    This only works with medium and large frame lasers that have snap in optics in bayonet mounts that maintain alignment, air-cooled lasers with mirrors held by a clamp and "O" ring can't use this because they don't have repeatability in the optics mount, although if the laser is peaked you can take one mirror at a time out and it has a 90% probability of coming back lasing at least weakly, though its position will need to be adjusted.

    More on Mirror Adjustments

    Moving any adjustment more then 1/2 turn will probably kill lasing unless you have the nice scientific unit with the 80 pitch adjustment screws, so be extremely careful to remember what you just did, it doesn't hurt to take a pencil and mark a mark on the knob/nut so you can see where it was before you started adjusting. Bigger lasers have dial turn counters installed so you can see where your at. One of the tricks is to push on an end plate with a finger to see if the beam gets brighter or dimmer - you are changing the mirror's angle by a milliradian or so, and it makes a difference. This is a good quick check to see if something is messed up. If the setting is just slightly off, a small touchup on the adjustments should send the output power way up. It won't need much - just a fraction of a turn, as you move the screw back and forth. Power should peak and then go down again - that is the time to stop and go the other way, loss of alignment is a real pain to correct!!!!

    Looking at the laser you should see something like this:

           @ Vertical adjust
           |
           |
           |
           |
           X---------------------@ Horizontal adjust
         pivot 
    
    The pivot is never touched once the laser mirrors are being walked for best cavity alignment relative to the tube bore during initial alignment. Then, all the user ever has to do is slight touchups on the vertical and horizontal, a rule being:

    DO only BOTH verticals then do BOTH horizontals. Then do BOTH verticals again!!!!!!! (repeat the sequence).

    If you start playing with adjustments like "OK, I'll do the front vertical then the rear horizontal then the rear vertical then the front vertical again" you are not moving the axis of the lasing cavity relative to the tube, you are just misaligning your mirrors at random.

    Now it is not uncommon to just have to slightly touch just one vertical or horizontal adjustment on the laser from time to time, for a small fraction of a turn to get around a power loss from drift in the mounts or from vibration, slippage, settling in the mount springs, etc. Because the Brewster windows are oriented vertically, they tend to act as a non dispersing (not splitting light into its spectra, meaning beam bending) prisms inside the cavity, which means most often a vertical adjustment will be what is needed, in fact most large lasers have a knob or other means for rapidly scanning the vertical adjustment to find to reestablish lasing if its lost, cause the vertical alingment of the mirrors is the harder then horizontal.

    So most of the time all an air-cooled laser needs is a touchup on the rear vertical mirror. If there is an intracavity (line selecting) prism, that complicates things so you might try the front vertical screw for maximum, once you get the feel, do the horizontals. It has to be learned by practice.

    Your eyeball is not a good laser power meter, there are usually jacks provided on the laser head to hook to a analog voltmeter, or in the case of the Spectra Physics or NEC lasers, the front panel meter on the power supply, to see when the >output is peaking. If you don't have a meter, the trick is to use a diffraction grating or prism to get the spectrum of the laser on the wall (assuming you have a multi-line laser) and peak the lowest gain lines lasing, which are the deep blues and violets, since they have so little gain, you can see their intensity go up and down due to small movements of the mirrors.

    Another trick is when the laser is in constant light mode, to monitor the tube current, as the cavity is peaked, the current will go down.

    Quick Alignment of Large-Frame Lasers on Optical Tables

    I (Sam) think the apparent benefits of this approach are more due to the ideal environment than to any particular details compared to "field" alignment. Heck, a chimpanzee can align any laser in under 1 minute flat if everything is on precision adjustable mounts bolted down to a 2 ton Newport optical table. :)

    (From: Steve Roberts (osteven@akrobiz.com).)

    For large frame laser systems in optics labs, this is the easy way. Call it the Russian method.

    You will need:

    1. A hot 5 mW (or more) HeNe laser on a pair of adjustable ring mounts.

    2. A white disk about 2 inches in diameter with a small hole in its center mounted on a adjustable height post.

    3. Two folding (planar) mirrors on Newport MM2 or similar mounts on adjustable height posts.

    4. An iris diaphragm mounted on an adjustable height post at the desired laser beam height.

    5. A beam block mounted at the exit end of the laser.

    Of course, if you have a BIG optical table, you will also have all the needed adjustable mounts and other precision hardware. :)

    In the system I dealt wtih, the laser was supported on blocks to raise the beam to 12 inches above the table.

    The HeNe alignment laser (call it the HeNe) is placed parallel to the laser to be aligned (call it the Ion Laser but of course could be a big HeNe, HeCd, or other long external mirrors laser) at the output end of the laser or a little beyond. Its beam is parallel to the chassis of the Argon. The beam then hits two fold mirrors to aim it through the bore at the high reflector end to the beam block target at the OC end. The iris is used to confirm that the HeNe beam is level and at the proper height all along its path. The HeNe shoots through the center of the 2" diameter white translucent plastic target mounted about 6 inches from the output end of the HeNe head.

    The Brewster covers of the Ion Laser are stripped off and the OC is mounted first, its reflected HeNe beam is redirected back through the fold mirrors to the target stuck in front of the HeNe. Unlike the conventional overlapping beam scheme used in most other alignment techniques, by the time the HeNe beam has returned through the bounce optics, it's a big bright dirty blob on the target disk and you can then measure the alignment error of the optic in beam diameters, i.e., you're not trying to place two tiny dots in the same spot, and NO post-alignment rocking of the mirror is required. This technique works with dirty optics as well. Stripping the Brewster covers lets you see if the error is in the optics holders gross position.

    Turn on the Ion Laser to 25 to 30 A (remember, we're talking LARGE lasers!), install the rear mirror holder and walk the ghost reflections back to the target disk. Slowly scan the returned beam across the disk with the rear horizontal knob while rocking the vertical knob fast. If you don't encounter lasing some place along the way, something is terribly wrong. If your rear optic mount doesn't have a hole for the beam, make one, or in the case of Lexel-type mounts, leave the rear cap off so the optic is still held by the springs, but is not compressed. Because you have a meter or more of lever arm, it's much easier to superimpose a fat blob over the hene beam, and the two fold mirrors make leveling the system much easier. Note that this doesn't help much for short lasers, but on a long frame with adjustable feet or up on blocks, it makes alignment a dream. Did it today, on laser which we had to find an optimal set of mirrors for Kr yellow, blue and red lines. We could flip in a set and in about 1 minute and be lasing again, even with a major wedge in the optics holder.

    Optical Monitoring to Assist in Mirror Alignment

    (From: Mike Harrison (mike@whitewing.co.uk).)

    Here's a handy new twist to the method for aligning external mirror resonators using a HeNe laser. Determining when the spot is shining exactly back at the HeNe can be tricky to see accurately, especially if you're standing at the other end of a 3 foot long resonator!

    Most HeNe lasers leak some light from the rear mirror, and the only way something happening at the OC-end should affect this is if the beam is being bounced straight back down the bore, i.e. perfectly aligned.

    So, in addition to the normal alignment setup, fix a photodiode or photovoltaic cell behind the HeNe's rear mirror, and place a mechanical beam chopper (a PC cooling fan is ideal!) between the HeNe's OC and the resonator being aligned, to chop the beam at at least a few hundred Hz.

    Connect the photocell to a scope (photovoltaic mode will usually work, so no supply needed). Align the resonator mirrors by adjusting to get the maximum AC amplitude of the chop pulses on the scope - i.e., the chopper is having the maximum effect on the rear mirror leakage, so the maximum amount of beam is heading straight back down the HeNe's bore. In addition to the chop frequency, you also see a lot of lower-frequency stuff due to interference effects, but this doesn't prevent the peak level being observed easily. In fact, vibration from the fan can be enough to shake up the interference patterns enough to make the peak level very easy to see. If however fan vibration is too much, you may need to suspend it from the roof to isolate it!

    And, if you don't have a scope, connect the photocell to an audio amplifier (MIC level input), and tweak alignment for maximum volume of the main fan-frequency tone. Laser alignment by ear! :)

    I'm still trying to get some life out of a huge Coherent krypton ion laser and I need to run it pulsed, so I'm trying to get the initial alignment as good as possible. This method is way easier than doing it by eye. Using a large-area photodiode (looks similar to the type used in solar calculators) on the back of a 1 mW HeNe tube, I get about 20 to 40 mV peak signal - easily enough to see on most scopes or for the MIC input of an audio amp. As you're only looking for the ac component (from the chopper), the signal is pretty easy to see even at very low levels, without any problem from ambient light.



  • Back to Sam's Laser FAQ Table of Contents.

    Maintenance, Alignment, and Modifications of the ALC-60X Laser Head

    ALC-60X Hands-On

    Actually having access to one of these small air-cooled ion lasers is definitely the best way to learn about them. Reading about these ion lasers in Sam's Laser FAQ or elsewhere (but I don't know of any 'elsewhere') can only go so far. You must have something on which to practice your laser skills. With a little experience (well, maybe actually quite a lot of experience!) you will become proficient at keeping them running and performing any required maintenance and repair including adjustment, alignment, and performance optimization.

    While prices for new ion lasers start at several thousand dollars and go way up, obtaining a used, but serviceable one on a more realistic budget isn't difficult. See the chapter: Argon/Krypton Ion Lasers. A complete working system - ALC-60X head, umbilical cable, and Omni-150 power supply - was recently offered on the Internet for $300. That rig was snatched up almost immediately. By spending some time searching for a good deal - and with bit of luck, your cost could be even less.

    If you are proficient in electronics and are willing to build your own power supply (which tend to be more expensive than the laser heads themselves - see the chapters starting with: Ar/Kr Ion Laser Power Supplies), your total investment, at least in terms of dollars, can be quite low. I (Sam) paid $100 for an as-is ALC-60X laser head (which turned out to just need cleaning and alignment) and built my own power supply mostly from parts I had collected over the years (finally some justification for all that clutter!).

    But keep in mind that what you will likely get your hands on is a 20 year old design. For a laser with external mirrors like the ALC-60X, you will need to stock a few things like gas chromatograph-grade (or better) acetone and methanol for cleaning the 5 active optical surfaces. Swabs in individual sterile packages with wood sticks are generally best for optics. Not all lens tissue is created equal, and some of it has hard particles that wouldn't hurt a camera lens, but could spell death to a soft coating on a laser optic. High grade Kodak tissue in the sealed envelopes is insisted on by most people in the laser refurb business but they prefer sealed sterile swabs for all cleaning. You use acetone first and then methanol and NEVER dip the swab or tissue in the bottle, always pour it out so you don't contaminate the fluid. Always keep a lid on the bottle to prevent air-born contamination from entering. If these fluids pick up water from the air it forms a white film on the optic that really zaps laser power.

    But this really isn't as bad as it may sound - once cleaned and aligned, an ALC-60X or similar laser can go for quite some time without additional maintenance. That is, of course, if you have enough discipline to keep you hands off of the adjustments after you get it running and have peaked its performance! :)

    Initial Inspection When You Get Your First ALC-60X

    (From: Steve Roberts (osteven@akrobiz.com).)

    Congratulations! You will soon find yourself shooting holograms, doing beam effects, and tomography of beer bottles. :-) Foggy nights will induce urges to shoot out the window, as it appears to stretch to infinity. (But avoid this temptation unless you want a visit from the FAA - and they won't be too happy!)

    The ALC-60X always reminds me of a quote from the Russian Aircraft Designer Tupolev: "Americans build airplanes like fine ladies watches. You drop the watch, watch stops. Russians build airplanes like Mickey Mouse clocks. You drop clock, it stops. You pick up clock, shake it, it starts again." :-) Welcome to the tough little laser that has a can-do attitude!

    In reading the description below, refer to the section: Photos of the Major Components of the ALC Model 60X Laser Head for parts identification.

    That little box is deceptive. The materials were carefully chosen for their expansion properties. Even though the thing looks like its machining was done in a high school shop by monkeys, they have to be some of the most consistent monkeys I have ever seen. Unlike some guns, cars, and many other things that are supposed to be standardized, you can take parts from any 2 of them and make it work. The rods are poured InVar. Everything on the rods is threaded 1/4-20. (Don't try to remove the rods, there are hidden setscrews.)

    The 5/8" nuts are the cavity adjustments and yes, aligning a brand new tube without making some special jigs is extremely difficult. Getting a bore centered on its true hot spot can take lots of patience. the big ones have fine adjust screws for that.

    Leave it be for now. Odds are it's workable even if it is only winking. A basic rule of thumb is never disassemble a partially working laser - you'll find only changing one variable at a time is a must. Unless someone else messed it up, right now it's probably aligned. Ignore the temptation to remove a mirror mount, although taking off the the light sampler is OK. (You'll find it keeps some of its mounting screws captive but don't lose the others!) Avoid putting any finger prints on its sampling disk. And, the cell is piezoelectric - squeezing it or slamming it will blow the op-amp in the light preamp. (Note the coated side of the disk goes towards the tube - otherwise it may sample too much light.)

    This would be a good time to power up the light sensor card in the head with 2 9V batteries and check it out with a flashlight. One of the causes of a winking laser on a power supply using light feedback is a blown light card with its output locked high.

    The cooling scheme we have backstage was developed by Mr Schweter and myself to meet the factory cooling profile, not too hot, nor too cold with a toasty core temp. Also, 10 to 15% of the heat must be shed through the baseplate. Running on a couple square foot of 1/4th aluminum base really adds to the lifetime. It doesn't take a piece much larger then the the laser to do the cooling. The holes in the bottom are for 1/4-24 screws.

    (Mr. Schweter compounds ceramics and glass for pigments for a living and had the tools to measure mass airflow and core/surface temps, If I can find his notes we can even tell you how many BTUs were measured based on tube current. The anode and cathode ends have slightly different airflows as well. The PVC ring is important, running without it will result in overtemp of the tube core.)

    The laser needs 4" to 6" of free space around it that has a large low resistance path to neutral room air, Otherwise it will breath its own exhaust. Attempting to reverse the flow for experiments will quickly overheat the laser, make sure you suck out the top.

    The connectors used are a series 3/5 hybrid of the AMP series 3 and series 5 CPCs (circular plastic connector).

    Watch your fingers with Mr. Patriot (the HUGE fan if that is what you have). It really deserves a guard ring as it must have been a chain saw or shark in a previous life. The fan will extend slightly over the lip where the cover ducks under the side plate at the top. Just bolt the fan down centered on the head and you'll find it easily slips on and off even with the overlap.

    Get some high pressure air, loosen the clamps on the anode and cathode heat sinks and take them off for cleaning, you'll find the metal webs are just joined together by a crimp, they pull apart and you can take them off without changing alignment or removing the tube. When you put the heat sinks back on, make sure there is a thin air gap between them and the tube clamps. Then carefully blow the dust out of the riser box fins with air. (I'm assuming you have American tube with a open riser box and not the interchangeable OMNI tube with the clamp-on solid aluminum heat sink in place of the hollow box.)

    Next step: Basic electrical checks. Put your multimeter on the Megohm scale. Scrape a bare spot on the base and make sure neither the cathode bell, anode bell, or cooling riser box reads less than infinity to the base plate - they should all be floating.

    There are fiber washers on the clamps that hold the tube so you should be reading infinity from the end-bells to the height adjusting 6-32 threaded rods. If you don't have this it's not a major problem, but it's a nice sanity check.

    Next up, unless you are using an 8 digit voltmeter, the cathode should be 0 ohms from lead to lead. You should also measure low ohms or a short from the cathode leads to the cathode end bell. This is the getter assembly on most air-cooleds. Having the end-bell short to ground and passing current through it will release a cloud of Ba, Ti, Sr, and some other gook, probably ruining the tube.

    Also look for bent pins, missing/broken wires stuff floating around the head, clean the dust out. A clean laser is a happy laser. :)

    The overtemp sensors are not very reliable, since they never cycle on and off, they just stick. :(

    If the 60X head originated from one of those copier things and someone hasn't already been in there, you may find a coating of black toner over EVERYTHING. If not, count yourself lucky. If the interior looks like a coal bin, see the section: Hands-On and General Cleaning.

    A typical photocopier unit pulled at 5,000 hours will do about 65 mW at 9.5 A with a new multi-line TEM00 optics set installed (rather than the single line optics it likely had). While it is usually best to change both optics as they come in matched pairs, often you can do quite well by changing to a 60 cm radius broad-band Output Coupler (OC) which should be about $75 depending on from whom you buy it. The High Reflectors (HRs) installed in the single line lasers are in my experience almost always a broad-band optic. On single line lasers, the OC is coated to reject all undesired lines. The copier optics usually kill the 514.5 nm line, and are really optimized for the blue and violet lines. Newer copier optics will usually lase on 5 lines with even a older tube. Since there are so many surplus lasers out there, it's hard to tell if you have a single line or multi-line tube, they rarely match the optics numbers on the checkout sticker after they are rebuilt.

    Setup and Testing of a Newly Acquired ALC-60X

    Read through the following before powering up the laser though some of it applies only after the unit is lasing. I'm not quite sure what got into Steve when he wrote this. :)

    (From the X god Himself (Steve) with laser god approved spelling.)

    1. You must useth an isolated battery powered voltmeter set to DC Volts to make the current/output power measurements using the test jacks.

    2. Pushing the little white button on the side of the head enables the readings. This is done to force you to check your connections. Do not cross connect the amp jacks to the light jacks. That results in kaboom in thy supply and much crying. Having the meter across the ignite pulse fryeth the user and thy meter hence the other reason for the button.

    3. Some laser dealers think it is great folly to remove the light sensor and light card from the head thus forcing the PSU to run at its upper current limit. (One way thy will knoweth this is if the little square box at the output is missing, though the sensor could just have been disconnected internally.) Therefore the light board seeth no light and thus behold, the supply forceth the current higher in vain to make light, more light that cannot be made, The limit circuit then sets the maximum amperage, leading to overcooked pass-banks and damnation to the scrap pile for thy tube for it is running full out toward the high end of the exponential life curve, which is a very BAD thing. True current control only results when the light loop works, with one exception detailed in the fine print below. If this causeth thee angst, remember that it is written in the holy specification of the "X" by the original creator, Xerox, that this should occur, and that light and current should be crosslinked this way, and is thus a mystery but to them.

    4. Do not rely on the light jacks calibration, for it is sheer madness to think a cheap piece of selenium can be linear from .001 to 220 mW. It is folly even more to think that since the PSU needeth a certain range from the light sensor no matter what the design power of the head is, for interchangeability in the end users device, that the light jacks would ever be accurate above 35 mW, unless a factory tech was ordered to install a different disk inside the sensor box to change the calibration. This rarely happens for it is a time consuming expensive option from the factory. Better the user buyeth a true reading power meter and calibrate his equipment accordingly. But changeth not the head card setting unless it's too high and the supply oscillates. after all its just a 741, a few resistors, a sensor, one transistor and a pot, and therefore its expecting too much of it to be a light meter.

    5. Mirror alignment does change in shipping especially when the brokers think thy tuning bolts are part of the lid. Trusteth the broker not, for he be a destroyer who careth not about your misfortune, but laugheth all the way to the bank by cooperating with the good ole boys.

    6. The pot on the side of the head if you have one (some lasers for platemakers don't) - and it is connected - changes current, not light sensitivity, and there is a difference: When the current is set below the existing light level, the laser defaults to current and uses the light feedback only to cancel oscillations and noise in the beam. The pot on the light card inside the head is set to about 2.4 V of output at maximum light and more then 0.7 V at minimum light to be more friendly to an analog opamp with a limited upper output. That way, the opamp is biased half on at full drive, preserving linearity and does not have to go to the positive rail, which is difficult for that opamp to do and maintain bandwidth or not latch up. Note: European Xs generally have different pins in use on the connector for fans and interlocks and possibly the control pot, thus a 115 VAC head is not destroyed by a 230 VAC PSU and vice-versa.
    Behold, some platemaker PSUs were designed to run at constant current their whole life, and perhaps you have one of those, as a external AO modulator knocked down the power as required, and thus a light sensor was a unneeded expense. Therefore a current pot must be brought out, referenced to the internal supply control reference voltage, with a limit pot added and wired in series with the external control at the high end of the reference voltage. I've never had to do it, but a friend has, and it was not for the timid nor a trivial task, as it must be carefully insulated, as the controls can float 70 volts above case ground. Plus you have to make sure the external pot doesn't ever send a signal that causes more then 10 A through the tube, else kaboom, and damaged pass-bank transistors. Some newer smaller FET gold boxes that say "Landmark" on the side do not sustain more then 9 A without frying the FETs.

    Take the time to set it up right so you don't end up with a slag heap worth less then its weight in salt.

    General Cleaning of the ALC-60X

    If what you have is an ALC-60X (or similar laser head) from a xerographic copier/phototypesetter/whatever, there is a good chance that it is full of black sticky toner that somehow made its way into almost EVERYTHING. For a modest coating on metal parts, WD40 is fairly effective (yes WD40 is good for something). It loosens it up but doesn't really dissolve the toner itself. Avoid the use of strong solvents like acetone or xyline as they may damage some plastic or other on-metallic parts (not to mention your internal organs).

    Assuming the two desiccant flasks (they look like oversize CO2 cartridges) are present and their rubber tubes attaching them to the mirror assemblies are intact, none of this mess should have gotten inside to the optics - hopefully. If the rubber tubes were missing or disconnected, they should be replaced and the desiccant is probably fat and will need to be revived. See the section: ALC-60X Desiccant Bakeout Procedure.

    For general information on laser cleaning and alignment, see the section: External Mirror Laser Cleaning and Alignment Techniques and the instructions specifically for the ALC-60X in the section: ALC-60X Mirror Alignment Procedure.

    (From: Steve Roberts (osteven@akrobiz.com).)

    The only off-the-shelf hardware store solvent for toner that I have found is toluene. Then you have to wash off the oily toluene residue with methanol. I'd Imagine MEK or THF work as well, but might damage the Teflon or Nylon parts (like the Nylon insulators on the bottoms of the tube supports). The toner has glass microbeads in it. Avoid getting dissolved toner on your skin unless you like to painfully itch for a day or two. It feels like a burn when you first get it on. Only thing I have found that helps is to immediately wash in cold water and apply moisturizing cream or aloe.

    ALC-60X Desiccant Bakeout Procedure

    If you found the desiccant flasks disconnected or missing their rubber tubes, the silica gel inside is likely flat from absorbing moisture from the air for an unknown amount of time. The following will revive them as good as new.

    (From: Steve Roberts (osteven@akrobiz.com).)

    Get a piece of firebrick, a pair of gas pliers, a propane torch. An optional 6" length of Pyrex tubing that fits over the fitting on the end of the desiccant flask helps, as you can see the steam condense on it.

    Set the desiccant "hot dog" on the brick, light the propane torch to a medium flame, and stick the Pyrex tubing over the connector fitting on the desiccant. Slowly go back and forth over the desiccant till the metal starts to discolor a little under the flame as oxides form where the flame hits the copper. You can't hurt it. Silica gel won't melt under an air/propane flame, and the nub on the desiccant is brazed on, even though it looks like solder. Observe jets of steam condensing on the Pyrex tube, turn frequently, and shake it often. It takes about 5 minutes per desiccant tank to cook out the water. Remember, you have to get the whole thing inside and out well above 212 °F to get the silica to let go of the water. Probably like 500 °F to ensure a release.

    Next, set both desiccants on a clean tray in a clean oven. Your kitchen oven is just fine (silica gel is non-toxic none is going to come out anyhow). Use a temperature of 450 °F for 30 minutes. Immediately seal the ends of the desiccants when they cool down.

    Note: I (Steve) own a lot of these, and before somebody chimes in and says all you need is the oven, not the torch, this silica gel is treated to hold, and not release water under a low heat like the common stuff does.

    ALC-60X Optics Cleaning

    If you are going to all the trouble of aligning your laser as described in the sections below, cleaning the optics makes a lot of sense - especially if you don't know what it has been doing in its spare time. :) Who knows when the last cleaning was performed - if ever?! The general technique for cleaning optical surfaces using ultra pure acetone and methonal is explained in the section: Optics Cleaning Procedure. This should be followed once you have access to the surfaces of your optics (Brewster windows and mirrors).

    Cleaning should be performed at both the HR and OC ends of the laser unless you are sure that the optics are spotless. A spec of dust or nearly invisible film of condensed who-knows-what can cut the power way down or kill lasing entirely. It can also make the alignment much more difficult because the position of that smudge or dust spec can change and will also affect lasing!

    Here are specific notes on optics cleaning for the 60X. Refer to the following photos for parts identification:

    Now for the actual disassembly and cleaning:

    ALC-60X Mirror Alignment Procedure

    The following applies directly to the ALC-60X and Omni-532 and is a special case of the more general technique described in the section: External Mirror Laser Cleaning and Alignment Techniques. It is recommended that you study that set of instructions before what follows, and especially before touching anything on your laser!

    If it's lasing at all and the optics are clean, skip to the section: Step 6: Fine Tuning since the basic alignment is probably fine. However, if the beam is way off center after that procedure, the tube position may need to be adjusted. See the section: Step 2: Centering the Tube. And, if you are sure that the alignment at one end of the laser is fine, much of what follows can be skipped as well.

    After having done this basically from scratch, I can see why those laser techs get the big bucks. :) I hope they also have good health insurance that covers psychiatric counseling - I can't imagine doing this day in and day out though there is a sort of 'rush' when you get that first flash of coherent light. However, after doing it a few times using a rigidly constructed alignment jig (see below), it does get a lot easier and faster.

    You will need a red alignment laser (henceforth called the 'A-Laser'). A helium-neon (HeNe) laser producing a clean beam is ideal for this procedure though you may be able to use a laser pointer or collimated diode laser module if it produces a 1 mm or smaller diameter beam and that's all you have. An Inexpensive 2 to 3 mW HeNe laser tube or laser head and matching power supply should be adequate and can be purchased for $25 to $50 if you know where to look. It is well worth the investment and you can never have too many lasers! :-) See the chapter: Laser and Parts Sources.

    A rigid alignment jig will need to be constructed to allow the A-Laser and 60X head to be mounted precisely in line with each-other during the alignment procedure. Details of possible designs for these are given in the section: The Alignment Jig.

    The A-Laser itself needs to be mounted on a platform that allows adjustment of its height and orientation (three fined screws). One possible design is shown in Alignment Laser Three Screw Platform.

    A fluorescent orange (preferred) or yellow sticker (get these from office supply companies) with a clean 1 mm (.040") hole is then fastened over the end of the A-Laser so its beam passes through the exact center of the hole. (The precise hole size will depend on your particular A-Laser. 1 mm hole will clear the beam of a typical 2 to 3 mW HeNe laser. A 1/16th inch hole will also work.) There should be essentially no evidence of the beam on the sticker. If you don't have a Post-It Note(tm), substitute a business card or piece of white cardboard (which is what I have been using). Perhaps not quite as sensitive but quite close. :)

    Here are some basic rules that shouldn't be violated:

    Step 1: Setting Up the 60X and Alignment Laser

    The degree of perfection achieved for these initial steps will ultimately determine the ease (in a relative sort of way) with which final alignment is accomplished.

    Step 2: Centering the Tube

    Note: If the tube position is being adjusted on a laser that is already lasing, performing the centering in small increments, and then maximizing power and beam quality as described in the section: Step 6: Fine Tuning after each one may avoid the need to perform a total alignment from Step 1.

    WARNING: The following procedures may put you in contact with the electrically live parts of the tube. Make sure you have disconnected the umbilical cable, or unplugged the power supply from the wall and checked to make sure its main filter capacitors are fully discharged!

    The 60X tube is mounted on 4 posts using a pair of 6-32 nuts and lock washers on each. Each of the threaded posts on which the nuts are mounted has a hex hole in it to permit it to be turned. Assuming the posts are secure, don't touch them.

    The glass Brewster stems of the tube fit into an O-ring in a collar which in turn fits into another collar - part of the mirror mount assembly. This seals out room air and contaminants while providing some degree of compliance to allow the tube to be moved slightly up and down and from side-to-side.

    In order to adjust position, the top nut (on each post) needs to be totally loose to allow the bottom nut to be turned by hand (Y adjustment) and/or the tube to be moved slightly side-to-side (X adjustment). Your objective is to set tube position so that the beam from your A-Laser passes cleanly through the bore centered on the HR and OC optics. The result will be a nice bright symmetric well-formed red spot of light on a screen placed beyond the HR-end of the 60X. It should be nearly as bright and clean as if the 60X were not in the way. There should be no off-axis smudges or arcs and equal clearance top, bottom, left, and right. Hopefully, there is enough range in X not to require loosening the posts themselves. For Y, you will have to set the tube a bit high before tightening down the nuts as the lockwashers compress somewhat. It may take a few iterations but when you are finished, the result should be close to absolute and total perfection!

    WARNING: The glass of the Brewster stems is thin and fragile - take care when changing tube position that you don't apply excessive force to the glass! I like to keep the mirror mount assemblies slightly loose while adjusting the tube position - then you can confirm that it is not being stressed in any way by making sure you can jiggle them freely!

    Here are some additional comments on setting tube height:

    (From: Steve Roberts (osteven@akrobiz.com).)

    When aligning a newly installed plasma tube in a 60X or 532, sometimes you have to adjust the height of the two tube cradles which set the bore height. Obviously you want the bore centered in the mirrors. While I have made a gauge block with a small hole in it for setting the HeNe alignment laser beam height on the alignment table at precisely 2.25" inches, which is the center of the mirror mounts. An easy way to check this is while the tube is still out of the laser is to remove the mirror mounts and shoot the HeNe beam through the 6-32 threaded holes that hold the mounts to the end plate. This should give you a reference height for your plasma bore +/- .5 mm. This is close enough to give you a clear path down the bore and get decent power after walking the mirrors. You then slide the HeNe over and use it to install the tube. Because the cradles have lock washers and fiber insulators that crush when you tighten the cradles, you'll need to crank in a couple of turns of upward motion into the cradle support nuts. You can then fine tweak the cradle heights once your aligned and lasing, if you see any arcs or smears on the wall when you walk the optics and graze the bore walls.

    If you have a lathe, making metal rods that are the same diameter and size of your cavity mirrors with a millimeter or so hole through their exact center makes a good tool for aligning your alignment HeNe laser and checking your bore centering on lasers that have that adjustment. However with the 7.6 mm diameter mirrors this is probably a waste of time. Always remember to look at the lasers data sheet to find the correct beam height. (Right.... :) --- Sam.)

    Remember also to make sure the black nylon or fiber washers with protrusions to center the cradles are installed on the tube cradle screws, while the cradles themselves are insulated from the base, the second set of insulation provided by the fiber washers is a good additional measure to protect the PSU from a tube to ground short, something it is unlikely to survive. Checking from the anode and cathode metal end bells on the tube to the resonator rods with an ohmmeter is a good sanity check when installing a tube. It should read open or infinity. Remember to scrape any oxide off the metal with the meter probe to make sure your getting a good contact.

    Step 3: Aligning the HR-End

    It is best to do the HR alignment with the OC removed entirely but this isn't absolutely necessary. Just loosen the OC mirror mount screws 1 or 2 turns and wedge a piece of thin cardboard under one side, then tighten the screws just enough to hold it in place. This shim will deflect any reflections from the surfaces of the OC optic out and away from your HeNe laser faceplate without affecting the direction of the forward and return beams you will be using enough to matter. (However, if you intended to clean the OC and its Brewster window, now is the time to remove the OC mirror mount assembly.)

    Your objective is to adjust the HR-end mirror to get a clean reflection of the A-Laser's beam shooting straight back into its output aperture (the hole you made with that skinny bit, remember?). Since the HR is slightly concave, this turns out to be relatively easy (well relative compared to climbing Mount Everest, perhaps!) as it should keep the beam fairly tight. (Note: If someone replaced your HR with one having a shorter radius - say 45 cm instead of 60 cm which is the minimum for a 60X, the beam will come to a focus somewhere inside the tube and diverge again making alignment trickier.)

    Make sure your setup and the centering of the reflection are both perfect to 10 decimal places. :) If you do this, you are virtually guaranteed success in getting the tube to lase eventually even if the initial OC alignment leaves something to be desired.

    Step 4: Aligning the OC-End

    Without upsetting your head to A-Laser alignment, reinstall the OC mirror mount if you had removed it. Remove the cardboard shim and tighten the screws if you haven't. Use the same twisting torque as you did on the HR as you tighten down the screws.

    You are going to use the reflection from the inside surface of the OC (back to the A-Laser) as a guide to adjusting its alignment. In principle, if the spot falls exactly on the aperture of the A-Laser, the OC alignment should also be perfect.

    There will actually be 2 reflected spots back to the A-Laser (not couting the one all the way back from the HR - which will be much fainter): one from the front surface (which isn't particularly useful) and another from the inner surface (which is the important one). There are possibly two factors which will complicate matters: The OC optic is actually slgihtly curved and may also have some 'wedge' - its thickness might taper slightly across its surface. If the mirror is curved as is likely, the correct spot will be slightly spread out compared to the front reflection. You want to center it in the A-Laser aperture. The wedge may still result in a very slight offset due to refraction but this is hopefully not enough to matter.

    The reflections from the two surfaces of the mirror are shown in the photo: HeNe Laser with Reflected Dot. Two bright spots are visible. I didn't take this picture so I am not sure which is the one of interest. (However, what is clear is that the bottom mirror not needs to be turned slightly counterclockwise to raise the spots to the level of the A-Laser aperture.) One way to tell which spot is from the inner (curved) surface is to move the A-Laser a fraction of a mm from side-to-side. The spot reflected from the curved surface (the one you want) will move more than the other one.

    The theory sounds good. However, in practice, this indirect alignment isn't quite as accurate as what you did above for the HR - going all the way to the HR and back inside the bore. Reasons include the shorter optical lever arm for the reflection off the OC, its curved surface, and the wedge. But, it will get you close.

    In any case, you are now ready for the real excitement!

    Step 5: Hot Flashes

    Now that you have aligned the mirrors to your A-Laser beam, the moment of truth has arrived. Leave the A-Laser on and power up the 60X using current control only (assuming you have any sort of control). Obviously, light control won't work unless the laser is lasing at near proper power (and haven't reinstalled the light sampler box anyhow). Set the current as high as you are comfortable running extended periods of time - you may be at this for hours. Ideally, you want it as high as possible to increase the mirror adjustment range over which lasing will take place. I used only 5.5 A most of the time changing to 7.5 A only when I became depressed enough to think that perhaps the alignment had become too messed up and I would have to start over. :) If your power supply, tube, and cooling can handle higher current (up to 9.5 A), that will increase the sensitivity even further.

    With the A-Laser still on, the reflected spots from the OC will help to guide your efforts below and make them somewhat less than totally random.

    If you are really really lucky, there will be some sort of a beam when you switch it on. How lucky do you feel? Getting a beam at this point would be about the same as winning the Pennsylvania State Lottery Super Seven so don't be too disappointed if you are only greeted by that ugly diffuse purple glow. Then again, maybe the alignment is so perfect that the beam has been swallowed up by the A-Laser's aperture - check with a white card to be sure but don't be disappointed if there is no coherent light there. :) While the HR alignment is probably pretty good at this point, the OC alignment is likely far enough off that there will be no beam from the 60X.

    Before loosening anything, apply pressure on the non-nut corners of the mirror mount end-plates in both directions. Maybe you got a little lucky and the error is on the diagonal axis. Yeah, right. :) If you can get any sort of flash of laser light out by doing this, all you have to do is adjust the mount in the proper direction and you are almost done. If this doesn't work (don't expect it to), loosen the OC-end mirror mount assembly just a fraction of a turn on the cap-head screws (not the mirror itself, but the larger pair of screws).

    What you want to do is to wobble the mirror a few mR in all directions to see if you can spot flashes of coherent light indicating that you have passed through proper alignment - if only momentarily. If you are only real lucky, just loosening one of the screws will change the angle in X enough to get it going. But, not likely.

    The proper orientation will be near where the reflected spot(s) from the OC fall in the A-Laser aperture so use them as a guide. As soon as you see your first flash, you are home free! What this means is that the HR is aligned well enough so that you will succeed. DON'T touch the HR mirror alignment under any conditions until you have a steady beam!!! It's close enough for government work for now. :)

    If you see a flash, try to figure out how to get back to recreate the miracle. Where were the reflected A-Laser spots at the same time? Turning one of the two cap-head screws by the smallest amount and then rocking the mount vertically may allow you to home in on the needed tweak to the mirror adjusting nuts. Alternately tightening each cap-head screw and the mirror mount nut(s) to maintain the beam may take some time. The thing you DON'T want to do at this point is lose the beam entirely. Again, no matter how strong the urge, DON'T touch the HR-end nuts or screws at all!

    With a steady beam from the 60X, you can turn off the A-Laser.

    Step 6: Fine Tuning

    Once it's lasing, you can optimize the X and Y mirror adjustments. Concentrate on one at a time - alternatively go back and forth between the HR and OC nuts turning each a small fraction of a turn and then its same-axis mate to maximize power. No matter how perfect you thought you had the alignment jig set up, we're now talking about precisions that are much finer: 1/10th of a mR (1 part in 10,000, or around 1/100th of a turn on the adjusting nuts) can have a noticeable effect on power - not much room for error!

    There will be a range over which corresponding rotation of the nuts at each end by the same angle will keep the output more-or-less steady. (Rotating each pair of nuts by the same amount maintains the mirrors parallel in that axis.) You want to find the center of this range - which will be the global maximum for that axis. Once you have done X (say), do the same with Y. Then go back to X. However, DON'T attempt to adjust X and Y at one end and then go to the other end - stick with the same axis until it is as good as it gets!

    As you do this, the beam should increase in brightness as the mirrors are fine tuned to match the actual orientation of the bore. The spot projected on a matt surface (you definitely don't want to get hit by a reflection - this is a serious Class IIIb laser) should be nice and circular - TEM00 - once the power is maximized. What you may find is that the beam isn't quite in the center of the HR and/or OC. That's OK unless you are really fussy about appearances. You should still have found the 'hot spot' for the confocal 60X resonator - the tube is just slightly askew in its mount (I know, you thought its position had been set up perfectly!) but this won't affect output power or beam quality. Why? Because with the 60X's curved (spherical) optics, the mirror alignment has been fined tuned to compensate and fire straight down the bore. If you were to climb inside the resonator, you would experience a 'hall-of-mirrors' effect extending for as far as the eye could see!

    Congratulations! The first is always the hardest one. Next time, it will only take you 3 hours. :)

    Steve's Comments on Beam Optimization of ALC-60X Lasers

    The following comments were prompted by a request for information on getting rid of ghost beams and other artifacts from a high power (110 mW) ALC-60B laser (which is similar to an ALC-60X).

    (From: Steve Roberts (osteven@akrobiz.com).)

    Hum, if its a 60B, you should have brackets that hold the tube that are on riser screws and held in place with top and bottom nuts. You can adjust for best bore position using these, although if you have 110 mW I'd leave it alone. High power optics often result in using the whole bore space, and thus you will graze a little someplace no matter what. Bore grazing shows up as a distorted beam shape. It's a semi-confocal cavity (flat HR, curved OC), so it depends on where the flat HR is in respect to the OC. A photocopier 60X often has a 60 cm radius OC to change the beam profile to a "Top Hat" approximation of a TEM00 beam. This results in a clean low power beam, which is needed by optical systems. It gets worse if confocal optics are installed with a radius at both ends. A typical radius for a 60X optic is 100 or 200 cm for TEM00 with a flat HR. For really screaming power, i.e., 225 mW when the tube is new, both ends have a radius and the OC has a different transmission on the green lines.

    Expand the beam with a clean AR coated lens and check for circularity. walking the bore up or down should gain you only 10 mW or so. The mounts have plenty of range in their travel for all but the most poorly mounted tubes to lase at full potential.

    When you move the riser nuts, be aware that there are Delrin insulators to keep the brackets from conducting to the riser screws and you should always check for continuity from the bracket to the screw. If there is any, it could lead to problems. The Delrin spacers are a second line of defense in case the nylon spacers on the bottoms of the tube mounts fail, or in case there is an arcover from a dead or hard starting tube, they decrease the likelyhood of PSU pass-bank failure, which will happen if the anode or the cathode is ever allowed to conduct to the case.

    The purple bore light is is plasma light from the cathode glow region and the bore. (Anodes don't glow that much.). But, it is not well aligned to the bore relative to the lasing path. Don't use the purple glow as a check to see if your centered, it doesn't work on really short tubes like this, although it is used as a secondary alignment technique on larger lasers if a HeNe laser is not handy. The quality of alignment of the Brewster stems and end-bells is not great and therefore they are often off axis to the bore.

    The main thing which distorts a clear path down the bore are the Brewster windows. They are not always mounted perfectly on center, and on a rebuilt tube where the end has been cut off and rewelded to install a new cathode, they are almost always way off.

    Tilt a clean flat microscope slide in a beam and note how you can shift the beam around with it. It's this beam displacement I'm talking about, and it occurs naturally at both ends of the tube. The Brewster at one end bends the beam down relative to the mirror and the one at the other end beams it up. Theoretically these cancel out but result in the fact that the tube center is not necessarily the axis of the optic center. Having an off axis Brewster means that you have a bent path through the tube. Moving the tube around can compensate for this.

    Build a reference block with a tiny hole in it at 2.25 inches from the bottom - this is the beam height of a 60 series head. Put the laser on a flat surface such as a optical table. If you use an HeNe laser for alignment, use the reference block to set its height and then center the tube accordingly.

    Multiple ghosts are somewhat normal with high power optics, ghosts result from wedge in the optics, not from bore alignment. It is possible with a new high power tube like yours to get lasing on the high gain 488 line from almost any path down the bore, and I have often suspected from a single bounce off the bore, but it sounds like you're pretty close to maxed out. There are generally 2 or 3 different "sweet spots" where a 60X tube will work well and I think your near one of them now from your power, which is exceptionally high for a used X.

    Keep in mind the Brewster windows, being less then perfect, do deflect the beam path a little and having everything "eyeball" and "mechanically" centered is often not the center of gain for a given tube.

    Oops, almost forgot: Nearly every 60X that I have worked on does not reach maximum power while centered exactly on the mirrors, it helps if your near the center but it will probably not be spot on. The same goes for the level of the tube ends relative to the floor of the laser, it's rarely where the ruler puts it in the end. This is why large frame lasers have X-Y positioner adjustments on both ends of the tube, and on at least one I can think of, in the middle of the bore.

    The "little laser that could" doesn't have the best machining, that's why it's so affordable. It is quite stable, however.

    Increasing Power Output of ALC-60X Lasers

    In addition to what is suggested below, if your laser head has a single line OC installed (the HR is probably already broad-band), then replacing it with a multi-line mirror can greatly increase power. See the section: Expected Output Power from Surplus or Previously Owned Ion Lasers. ALC will sell these optics in quantity 1 and may have the best prices unless you can find them used or surplus.

    (From: Steve Roberts (osteven@akrobiz.com).)

    The Xerox spec I had was a constant 23.7 mW over the life of the laser while in the copier. They design the tube to start its life at about 4.7 A and end at 8 A. I once met Xerox's factory tech - they have a rather odd way they tune the resonator over a certain current range to meet their specs so the laser doesn't need to get touched once installed. This tuning is usually way different from what you'd get if you tuned the laser for max power light show style. So you have a good chance of getting much more power or more lines with a simple tuning if the tube still has some life in it. The companies that refurbed the lasers for Xerox generally bought any used tube they could get their hands on from a bulk customer of the lasers and repumped them. So often times you can find a high power tube in a copier laser, tuned to act like a copier tube. Using a high power tube at low power also resulted in a major lifetime increase.

    See the section: External Mirror Laser Cleaning and Alignment Techniques to get started on peaking the laser. Often just a optics cleaning and a retuning result in a generous increase in power for these lasers.

    If you live near a laser refurb shop, take the unit in for a cleaning and power meter session. The fee won't be that much and you can get a good idea of what's going on. If you don't have access to gas chromatograph or better grade methanol, the cost of a professional cleaning can be cheaper then getting the good chemicals. Keep in mind that the head photocell is only a rough indicator and is especially inaccurate at higher power levels. And, if the laser was set up for say, a 35 mW tube, a 100 mW tube, or a 180 mW multimode system, its calibration factor will be different from whats stamped on the side of the head.

    Also measure your tube voltage at around 10 A. (See the section: Measurements of Current and Voltage in Ar/Kr Ion Laser Power Supplies.) That's a good measure of gain and possible improvements in output power. For example, a reading of 104 V at 9.3 A is reasonable for a middle-of-life tube, maybe just a bit on the low side. But the tube should still last a long time with good cooling and sensible use of the PSU's current control. Also see the section: ALC-60X Tube Voltage and Life Expectancy.

    A cheap piece of diffraction grating or a CD disc (another good alternative use for those AOL coasters) can be a decent method of seeing if more than one line is lasing. You're well above the 10 mW range if you get 5 lines going - probably at least 20 to 30 mW minimum. A violet line is also a good sign. A high mileage tube (e.g., 5,000 hours) really needs a good cleaning/tweaking. Does it sting the soft tissue on the inside of your wrist? (Don't try this with a 10 W laser!). Can you pop a orange or red balloon? Try a black plastic floppy disk as well. These tests are done close to the head at the beam waist where it sort of comes close to a focus (or with a positive lens to focus it to a point). Is the beam fuzzy around the edges or clean like a HeNe?

    Using an Oudin Coil to Start a 60X Tube

    This is what should be done when a 60X or similar tube just flashes but won't stay lit using a known good power supply set up correctly, the wiring has been checked, and the filament is hot. The cause is likely high pressure due to not being run frequently enough.

    The procedure is a little tricky, but not overly so, for you have the top cover off and need to ignite the tube, then get the cover on in 20 to 30 seconds so you have cooling. The cover doesn't have to be bolted on, just loose.

    Refilling ALC-60X Tubes?

    In five words: It probably isn't worth it.

    In a nutshell: If you have a few good 60X cores (e.g., working, structurally sound cathode, clean undamaged bore, unsputtered Brewster windows) on the shelf and are willing to spend a few weeks or months experimenting on them with the understanding that you may revive a small percentage but ruin the rest, by all means go ahead. It isn't just a matter of filing a hole in the exhaust tube and pumping in some fresh gas! And, if you were thinking about totally rebuilding 60Xs for profit, you're probably on some sort of mind altering drug. :)

    Also see the sections: About the High Cost of Refills and Refurbs and Ion Tube Rebuilding in Your Basement?.

    In more detail:

    (From: Steve Roberts (osteven@akrobiz.com).)

    I've got turbo pump, cap manometer, the pure gases, access to the original designer, the pinch off tool (only one company on the planet makes these), process notebooks from a company that refurbed several hundred 60X tubes a year, and two years of working closely with another established fellow in refurbing. I fixed his power supplies and spend more time working on 60Xs then anybody not manufacturing or rebuilding them in bulk. And, if I'm only getting a 50% yield on 60X tubes using good cores, how are you going to take over the world?

    It costs nearly as much to do a short tube as a long one. And maybe you won't be disappointed when you've spent a day or two of work to set up a tube for pumping, after scrubbing it for two hours in toluene by hand to get rid of the toner, and it leaks on station and is beyond recovery.

    You're not going to easily breath any life into a X. If you are knowledgeable about your 60X history you would know that most of them out there have already been rebuilt 2 or 3 times as they are recycled at 5,000 hours or so under contract by large purchasers. They go to the public when they are shot for refurb. Get a row of them lined up and look at the welds, you can tell with practice tell who rebuilt them from the weld quality. During the repump, one way or another you have to expose a porous cathode to nearly atmospheric pressure when you open it up, and this means a heck of a lot of work to recover the emitting surface on the cathode. It isn't just a matter of bake out and backfill, even on a chop and pump. In addition, there is a getter in there too. You would have to saw a tube open to find the getter and it can cross contaminate the cathode during processing.

    Also keep in mind the cavity optics play a part here. Brewster windows have a finite lifetime and are quite expensive.

    As if this isn't enough, a standard 60X power supply (e.g., Omni-150) will not stand up to the strains of heating up tubes during processing. They tend to go boom - at $150 or so worth of parts and a half day of labor - when the tube gets out of its compliance range. So, you either need a special power supply or added protection for your existing one.

    Yeah I had the same idea as you, lets undercut everybody part time in the basement, having found a turbopump new in the crate for $200 with driver. I used to curse the people who did refurb commercially after calling them for quotes. Forgive me guys, now I know better.

    It isn't that easy and one other thing I know from repairing small ion lasers for many years, and having quite a client base, nobody's going to pay $300 to have a chop and pump with a 50 percent success rate after they hear about 2 or 3 failures of your systems when shipped. I could go into the details here in much more detail, but all I'd be doing is encouraging you to try. Sure, experiment on your own units, but don't open the doors to the public until you're sure you can recover 75% of a given make and model, and make sure you tell em up front it's a 50/50 proposition with no warranty.

    Unless you have a pile of cores to work with, your recovery rate is going to be nil on the air-cooleds. Rebuilders only do them if they are large air-cooled lasers or they have a contract to do them in bulk, for a reason.

    Steve's Comments on ALC-60X Cooling, Line Drop Out, and Old or Rebuilt Tubes

    The comments below were in reply to this request for help:

    (From: Jeff Keyzer (jkeyzer@ucsd.edu).)

    "I tried replacing the big blower on my ALC-60X head today with a Rotron Major (Patriot with flat sides) fan, which should be about 235 cfm from their Web site. Patriot fans seem to be very popular with 60X owners. I built a 1 inch riser for the fan and secured it to the head, and RTV'd the gaps between the riser and the flat sides of the fan. Overall the installation went well and looks very nice. However, I had some *bad* results with it:

    1. The exhaust air is substantially less in volume than my blower (the blower has got to be around 400 to 500 cfm), and much warmer. The fan itself gets too hot to touch after 5 to 10 minutes of operation at 9.5 A tube current.

    2. This is the disturbing part - my multiline head loses up to 3 weak lines after 5-10 minutes at 9.5 A current! The 457, 475 (or so) 488, 500 (or so) and 515 lines remain but I lose some of the weak lines (blues and greens) in between those lines. With the blower these lines remain indefinitely. (I need to let it run for an hour or so and see what happens, but it's definitely still lasing on 8 lines 15 to 30 minutes later.) To me this is a very bad thing and I'm guessing it is the tube running too hot and the pressure changing inside. Note that I can run all day using my higher power blower without losing any lines.

    However, maybe this is a normal, documented behavior since so many people use Patriot fans. I couldn't find any mention of this in the FAQ. Perhaps my tube has junk in it that gets baked into the walls of the tube at the higher temperatures and stops lasing on some of those extra lines? I don't have a power meter at home so it's hard for me to quantify the difference in output power. I'm guessing it would be a very noticeable dropoff.

    I have pretty much ruled out alignment related problems, as I tried tweaking the resonator X and Y adjustments and couldn't get the lines back again. However running the tube at 5 A for a few minutes and then cranking it back to 9.5 A gets back the lines until they disappear again in another 5 to 10 minutes.

    My tube is an older one with catenodes and two heatsink rings in the riser box, putting out about 60 mW at 9.5 A current after a 30 minute warmup with the blower (not the Patriot). No idea how many hours are on the tube or how many (if any) rebuilds it's had. The head has in excess of 4000 on the hour meter. Omni-150R power supply. I have never cleaned the optics but I have adjusted the alignment to get 60 mW before.

    I am now almost positive that the problem is that with my particular (old) tube, the pressure differential due to the dense heatsinks in the head is too much for the poor Rotron fan and its cfm drops dramatically when placed onto the head. The head was likely being severely undercooled. For now I'll be sticking with a blower!"

    The following is a short summation of my fan test experience, I think you may have either too short a riser, or a major with bad bearings.

    Your Major is 15% smaller in air flow then a Patriot and if used may have bad bearings. Make sure air is exhausting out the top of the head, not the sides. The problem with lasing lines dropping out is not normally a function of the fan, the tube is overfilled. Yes, the Patriot gets too hot to touch, thats normal. The alternative for many people was guessing with muffin fans, that's why we researched the Patriot. My one suggestion is raise the riser to two or three inches and see if things are cooler, and make sure air is exhausting up high enough that it doesn't get sucked back into the side cooling fins and thus reheating the core. Also, a few early heads that were designed to run at lower currents have more dense fins, and if the fins are full of dirt, blow them out with high pressure air.

    As for the lines dropping off, that means your tube pressure is climbing too high, you may have a older tube that was not designed for high current operation or somebody has overfilled the tube to the point that its nearly on its high end inversion point, where if the pressure is much higher the arc will go out. Until I see a curve of tube volts versus current taken at six or more points, I'd derate that laser to 30 mW. Also at what current does the discharge drop out at ????

    It's never really been a issue with the Patriot, a couple of dozen simillar fans are in existence with many different hobby and biomedical users, and they are very similar to the fans that National laser, American, and Melles Griot now put on their lasers. One commercial maker of laser shows for clubs using 60Xs pushed to the limit switched to the Patriot for two hundred units with a good increase in lifetime. The engineering for the Patriot fan was done by comparing several heads with American's and Omni's factory cooling with thermocouples and sensitive air flow meters, and despite what you would think compared to a water cooled laser, air cooled lasers run with the tube core very hot by design.

    Where the problem comes in is rebuilds, what many rebuilders did is drill out the bore on older tubes from the 0.45 or 0.50 mm that it normal to the larger 0.55 used in higher power heads. What they didn't do is drill out the corresponding gas returns. A ceramic shroud inside the tube prevents easy access to the return holes when the cathode assembly is cut off. The drilling cleaned the contaminants from the bore making it much easier and quicker to reprocess the tube, at the cost of taking the tube away from its design characteristics and possibly getting drilled material stuck in the gas return tubes. On tubes designed for say 60 or more mW, this didn't matter, but on older low power tubes, as the gas returns get blocked from sputtered material, pressure builds up at one end of the tube and lines drop out. So as your tube got hotter, the wires stuck down the gas returns to prevent ignition of the plasma in the returns expanded and cut off even more of the return flow causing even more problems.

    You just hit the unlucky set of circumstances to end up with a tube that is pressure sensitive at the high end instead of the low end. It's a quirk of microeconomics with the reprocessors using every trick they could to make their process more attractive to the bulk end users. Some rebuilders bought huge amounts of tubes, and didn't care what went to whom or about where a given head came from. As long as the finished rebuild passed its warranty lifetime conditions of the new end user, they didn't care about details like rated design power. As a general rule, I'd be much happier with a high pressure tube then a low pressure one. Especially if it's doing the violet lines. Clean the optics with high grade chemicals, derate to an upper limit of say, 8 A (especially if it already has over 4,000 hours - keep in mind that rebuilders nearly always replace the hour meters so thats probably a true clock), and be happy with slightly less power!

    Also about 15% of the heat flow for a head should go through the baseplate. High power heads need a aluminum base to set on and dissipate some of the frame heat. I'd get it on a decent aluminum baseplate to spread the heat. This also aids in improving the life of older tubes.

    Dave's Comments on Testing and Tweaking of Omni-532/ALC-60X Laser Heads with Omni-150 Power Supplies

    (From Dave (Ws407c@aol.com).)

    A word of warning: A laser head does need to be "matched" to the PSU as far as current limits and light card pot setting. Other settings are pretty much set and forget. In my opinion, an accurate way to keep an eye on tube current is a MUST - and at all times.

    I believe many people that are new to these argon systems have their lasers die early due to the many critical settings involved as well as the minimum amount of time to recover from a high current overload situation that goes un-noticed and causes damage without the user knowing until it is too late. I believe this is why MWK insists on using current control only on the lasers he sells, even though they may live a shorter life due to plasma oscillations. I only use light control. But when in light mode, if an unmatched head/PSU is powered up and there is no lasing, the current could possibly be pinned at the max current limit well beyond 10 A with a long time trying to find the beam, thus cooking the tube in the process. So I guess his thinking (MWK and others) is leave it in current mode only and less risk of a return due to runaway high current in light mode with an inexperienced user. I am not defending MWK but was always curious as to why he and others insist on current only and the FAQ as well as other sources say light. Now I know why, after running several heads and PSU combos. :-)

    Another thing I think most people are not aware of is parasitic oscillations within the Head/PSU combined. I have run a small bore 60X tube and when the current goes past about 8 A the tube voltage jumps way up and in my AM radio receiver nearby (always have one on for listening to RFI, very useful tool!) I hear a violent hissing white light noise. This noise is something oscillating and the user would never know anything was stressing unless tube voltage is monitored or a radio was on and the user was keen to the RFI signature of what was taking place. So, the user would continue to run the laser unaware of what was happening and crank away until something blows.

    So what I do when testing a newly acquired 60X or 532 head from eBay or other source:

    1. Put the PSU into standby mode.

    2. Fire it up and watch the current like a hawk.

    3. Confirm that the discharge is lit and current is within acceptable limits.

    4. If there is a beam, I walk and tweak, or find a beam and do the same safely at low current.

    5. Switch into light mode and bring the current up a bit say to 5 or 6 A, then tweak the mirrors for the lowest current reading (adjust carefully for hard to turn nuts can shoot current way up real fast), then let it cook for a bit.

    As it is said elsewhere in this document, these lasers are mission critical and there is not much room for errors. :-)

    I have 3 Omni-150-series PSUs on my bench and they are all different in many ways yet the same thing.

    I guess no two are alike!

    Yup, the argon laser is a must have for the hands on learning experience for sure, I am enjoying it quite nicely. :-)

    Dave's Comments on Optics Cleaning and Tube Swapping

    For optics cleaning, I use the usual high quality solvents and top grade sterile swabs, but I remove the entire mirror assembly including the dust shields if I am only doing the Brewsters in a 60X. That way, I have a beam without alignment when I put it back. And it seems to work with me between my 2 60Xs. I can remove the tube and replace it or swap with another tube (if carefully centered) and not loose a beam as long as the mirrors are not loosened from their cells.

    To remove a 60X tube and keep a beam, mark the cells and then remove both of them completely with dust shields. Plug or clamp the hoses and plug each cell or install a standby tube to keep them clean. Remove the anode wire and the anode heatsink completely, disconnect the filament wires, undo the 4 cap screws making sure the center screws don't turn, lift the cathode-end first and push back the anode Brewster through the faceplate, and lift the tube out of the frame. Reverse order to install or replace tube, and you'll have a beam that just needs fine tuning when power is applied. :-)

    While I had my tube out, it went outside to the hose and I really washed it. :-). You need good high pressure water to get that toner and other dusty junk out of the riserbox heatsinks. After I hosed it down, I gave the Brewsters a quick wipe, then used a hair drier on it to prevent rust. Now this tube runs a lot cooler.

    (From: Sam.)

    Maybe use the dishwasher next time. :)

    Swapping Mirrors in ALC-60X Lasers

    The usual desire would be to replace boring single line 488 nm optics with those for full grand and glorious all line operation. Modify as appropriate for going the other way.

    Put the power supply into current mode (NOT light mode) so that the current won't go haywire if there is no lasing.

    There are two ways to go about it:

    1. Just swap the mirror optic itself. This requires removing the beam pickup assembly (4 screws) and then the mirror retaining ring (2 cap screws). The mirror will either stick to the O-ring on the retaining ring or remain in the housing. In the latter case, use a piece of masking tape to grab it if it won't come out easily. Take care not to touch the mirror surface!!!

    2. Swap the entire mirror mount. This requires removing the laser head cover to be able to pull off the rubber tube going to the dessicant flask. Then, remove the 2 cap screws holding the mirror mount in place and *gently* twist it and rock it back and forth to get it off the Brewster stem. Take care not to apply too much force to the Brewster stems.

    Keep the Brewster stems exposed to the room air for as little time as possible to avoid collecting dust.

    In both cases, some mirror alignment may be needed. DO NOT touch the adjustments on the rear mirror until you have lasing. With luck, it will lase as soon as the new front mirror is in place. But, some fiddling may be required.

    Once you get it lasing, walk the mirrors for maximum power. Then see if you have all the lines you want. A lively tube with clean optics will produce multiple lines at very low current - maybe 5 A. If not, you'll have to do the rear mirror as well in a similar manner. Just don't do them both at the same time since total realignment may be required.



  • Back to Ar/Kr Ion Laser Testing, Maintenance, Repair Sub-Table of Contents.

    Maintenance Information on Specific Ion Lasers

    Notes on the American Laser Model 68 Argon Ion Laser (ALC-68)

    (From: Dean Glassburn (dglassburn@mindspring.com).)

    I have regassed many 68 systems and typical voltages are 180 to 190 V at 20 amps. A tube with high voltage probably has sat just too long. Running it will help reduce the pressure and clean up the bore. Don't run very high current immediately, creep it up over a four hour period. Make sure you have good air flow, not just the little fans on the side. This laser needs large cooling blowers to remove the heat. The square ports are where the heat is removed via suction. That tube could be as old as 18 to 20 years. I still manufacture reconditioned tubes, but due to the price of ceramic, new tubes are not cheap, $4,700. Reworked units are cheaper. The other problem involves either overheating of the units, (not enough cooling) or running the tube at low pressure which tends to promote ignition through the bypass track. Usually you don't know this is occurring. The starter stops, current is supplied (4 to 5 amps) but the tube is not ionized down the main bore. Running it like this will promote metal vaporization down the bypass. Once that happens, the tube cannot be reprocessed with reliable starting. It wants to fire down the side as opposed to the bore. Low pressure is very bad to this type of tube. Typical cathode power is about 100 watts.

    Spectra-Physics Argon Ion Laser (SP-162) - Discussion

    (All replies from: Frank Tompkins (frank@Uakron.edu).)

    "Hello, I recently picked up a Spectra-Physics Model 162A-07 Argon Laser head and 262A power supply. I It was a pull from a Hell Drum Scanner after the scanner was hit with a forklift. The problem is is the umbilical that hooked the laser to the scanner. I have no idea what the wires do. There is no output power control pot on the front of the panel, just a hole with the markings for it."
    (Pulling out my manual) The power control is a 50 ohm linear pot, going to points N, O, and P on the circuit board (N is the wiper, O is the ccw extreme, and P is the cw extreme)
    "Inside there is a binding block where the pot was and wires leading to the umbilical. The umbilical terminates in a AMP plastic twist lock connector. Is this standard on these supplies? Where might I be able to find a schematic?"

    My 162A has an "older style" rectangular connector that has flat pins oriented either horizontally or vertically to provide polarization. Sorry I don't know the brand, but it's definitely not an AMP twist lock. The manual shows the type of connector I see on mine, so you must have an OEM version.

    "When I got it, the laser would put out maybe 2 or 3 mW at 488nm at about 3.5 A tube current in light feedback control mode. If you move your hand in front of the beam you can see that the light is pulsed. When I switch to current feedback mode the laser goes to full power to about 12 mw at about 10 A.

    I adjusted the mirrors. I am now getting about 22 mW (so says the meter on the front) of the blue/green line. It just needed some adjustment of the vertical adjustments (It just has allen screws. One thing though, If I move the rear vertical adjustment I get the other lines. I guess it is a beam selecting prism. I called SP today and they are looking up the info on the multi-line optics. Since I can tune the laser to the desired line, does this mean I will only need to get a different high reflector and not a different output coupler?

    What I am thinking is that the guy that had it before me (who knew next to nothing about lasers or electronics) played with the pots and variable capacitors on the control board. Does anyone know the procedure to reset these settings to normal?"

    There is some calibration and alignment info in the manual.....

    "What is the maximum tube current the head can handle?"

    Depends on the cooling fan on top. The small fan units are limited to 9 A, the large fan unit is limited to 12 A.

    "Last but not least is the optics themselves. In place of a high reflector there is a wavelength selecting prism. Where can I find multi-mode optics to replace these?"

    The manual does list part numbers for the optics, but considering it is a 17 year old laser, the parts are probably discontinued by now.....

    "What is the maximum output I can expect from this head (label says .12 W, yeah right...) It was manufactured in '79/'80."

    The manual is not real clear on this, but from my experience, about 25 mW (all lines) seems about right. That 0.12 W (120 mW) seems high.

    "FWIW, I only paid $50 for the set, so I am not too worried about doing to much to it."

    Good deal - I paid $90 for a head, and had to build my own supply! Got any more???

    P.S. Watch your eyes - 25 mW of argon laser light can do damage quickly. Also, watch out around that power supply, 140 V at 10 A can kill!

    Britt Pulsed Argon Ion Lasers

    More info including photos of a Britt pulsed argon ion laser in action can be found in the Laser Equipment Gallery (Version 1.97 or higher) in the "Assorted Argon/Krypton Ion Lasers" wing.

    Description and Comments for Model 3260

    Here is a description of one model of a Britt ion laser (from a unit that was on the Ebay Auction, November 1998:
    BRITT model 3260 High Power Argon Laser for sale. Pulsed output selectable from control panel 20, 50, 100, 200 milliseconds and paint mode. Although pulsed it's still great for Laser light show special effects or scientific applications (spectroscopy). Control panel has 2 VU meters (laser power, 2 scales to 30 watts) and (Laser gas supply), Panel also has standby/ready switch, Aiming beam power level control, Laser power level control, cool/thermal mode switch and Emergency shutdown switch. Laser head is air cooled and contains power supply. Unit is Multi-wavelength (5 color, violet,blue,magenta,yellow,green) and multi voltage selectable ( Runs from 108 to 240 VAC ). Unit can be cranked up to 1.7 Watts or momentarily kicked up to 20 watts using foot pedal for high power mode of operation. Head also contains a built in Ultra High vacuum valve to ease regassing. Total unit weight (Laser head, control panel and foot switch, long power cables) about 122 lbs before packing. Buyer pays for actual shipping cost prepaid. Will ship air freight insured against loss or damage.

    (I don't know if it is the same as the one discussed in the next section: Frank's Acquisition of a Britt Ion Laser.)

    (From: Dean Glassburn (Dean@niteliteproducts.com).)

    This is a pulsed laser so it can only be used for light shows as long as the beam is not moving - else it strobes as a series of dots. When run in high power mode the unit will thermally shut down to prevent destruction of itself. The Britt was designed only for medical applications where a few hundred milliwatts are required for a 1/10 to 1/2 of a second. The tube in thermal mode will run 750 pps and deliver about 2.0 watts predominately blue and green, but no yellow. This is assuming it's a 152 model tube. the 150 and 100 model has very different specifications with the rep rate and pulse energy. Your eye will integrate the peak 20 watt pulse over time and still seem like about a 1 watt laser. The divergence is terrible and the large bore gives about a 1/8 inch beam to start with.

    Frank's Acquisition of a Britt Ion Laser

    (From: Frank Roberts (Frank_Roberts@klru.pbs.org).)

    I have come across a Britt medical argon laser which appears to be a pulsed design. It is about 1 meter in length, runs off 240 VAC single-phase AC and is air-cooled. The power supply is integral to the laser head itself, and appears to be a flashlamp type design. What I mean here is that the 240 VAC is rectified to about 340 VDC and applied directly across the tube and a considerable capacitor bank for energy storage. The laser is fired with a high voltage trigger pulse applied to a trigger electrode wrapped around the tube bore itself. All the electronics appear to be functional in that indeed the potential measured from anode to cathode is about 340 volts DC. A string of 3 100 W lamps burns brightly across the supply, so it is sourcing current as well as voltage. The cathode filament does light up as expected and a high voltage pulse of several hundred PPS is applied to the trigger electrode. This pulse easily jumps an air gap of about 1/4", so I assume it's healthy also. Only one problem, no discharge. I applied a 7.5 kV neon sign transformer across the tube to check tube integrity and the tube did light up with a characteristic argon color. I know the difference between nitrogen purple and argon lavender, and the discharge color was indeed the color of argon. One added note: The getter deposit inside the tube is shiny, indicating no air leaking into the envelope.

    Discussion on Gas Adjustment

    (Continuation from: Frank Roberts (Frank_Roberts@klru.pbs.org).)

    There is a gas recharge system built into this laser and the vacuum gage on it seems to indicate an overpressure condition. This may be the only thing keeping this laser from firing. Methinks that somebody got a little punch-happy with the recharge button. Each time it's pressed, a tiny amount of fresh argon is metered into the tube.

    I'm sorry to be so wordy, but I'm trying to come to some kind of decision as to what to do with this laser, and would appreciate some input from my fellow laserists out there. I do not have a vacuum system readily available to me at this time. I do have a three-stage pump capable of pulling a vacuum of 6 microns. The decal on the tube indicates the operating pressure of the laser to be 40 microns. It's theoretically possible to pump the tube down below it's operating pressure and then refill the tube from the recharge supply.

    (From: Steve Roberts (osteven@akrobiz.com).)

    You'd need a 10-8 Torr vacuum or better.

    Attempting to repump a ion laser with getters without rebuilding the tube is a no-no. The getters will powder and end up in the bore, destabilizing even pulsed lasing, then it will find its way to the most electrically neutral spot in the tube, the Brewster windows via Murphy's law, thus suppressing or killing lasing. You really need a turbo pump to clean the water vapor out of the tube, else lifetime is nil and the cathode will be shot. The Britt is pulsed but it has a 2-3 mW quasi cw simmer mode for aiming for 15 seconds at a time. The pressure in a pulsed argon is much much lower then the similar sized CW argon which would have a pressure of around a torr and needs a much cleaner gas fill.

    Hint, hint: Might I suggest building a styrofoam reservoir around the gas bottle and ever so slowly chill it down with LN2, then pulse the solenoid till the pressure goes down using Mr. Boyles and Mr. Charles's laws for gases in a constant volume enclosure. You are not trying to freeze the argon, only to do sorption pumping onto the glass and volume reduction per basic physics. Otherwise you'd have to do a thorough the reservoir repump, as soon as you cracked the seal on the tube you'd suck in enough dirt from the atmosphere to fog the windows big time even in a clean room.

    Note I'm leaving out details of how you'd have to go about reprocessing a cathode if it ever sees nearly atmospheric pressure in argon etc because I work closely with a refurb shop and can't give away all the secrets etc. I will say the LN2 trick works however.

    (From: Frank Roberts (Frank_Roberts@klru.pbs.org).)

    My god Steve, the LN2 idea's so simple I'm ashamed that I didn't think of that. A neat little variation on the cryo pump. According to Merck, argon's normal boiling point is 87.28 K while LN2 boils at 77.36 K. LN2 should cause the argon in the reservoir to sweat the inside of the metal tank. (Now to calculate the vapor pressure of liquid argon at 77.36 K.) To add insult to injury, I was a physics major the first time around. I should have caught that one, thanks for the tip. I'll let you know if it works.

    (From: Dean Glassburn (nitelite@concentric.net).)

    Actually the Britt laser comes with a cryo attached to the tube on the end of the copper pipe. pull it away from the tube while the head is suspended upside down and place the cryo into a cup of LN. wait until the copper tube frosts up, about 10 minutes, then open the large valve connected to the assembly, about 1 turn. This will pull the tube below the operating range. on the end of the laser put the switch into the standby range and turn on the system. this will outgas the cathode. remember to keep the cup filled with LN. if you have a gauge it should be reading zero. Shut the valve, remove the LN, and allow the temp of the cryo to come up to room temp. At that point you can turn on the system, add gas back to the tube using the refill button until about 25 milliTorr is there. You should get the tube to ignite. Keep and eye on the pressure as it will use gas very quickly initially. Once stabilized, you can fire off the getter using a 12 volt transformer and a variac to heat up one of the getter filaments to a dull orange. Do this only if the gas has a pinkish discharge around the cathode neck and anode.

    (From: Steve Roberts (osteven@akrobiz.com).)

    I have only once seen a Britt on a trip to a conference in Canada, now I know why they last so long. A built-in pump, a built-in fill system, and a built-in cleanup system, and not to mention some of the most beautiful glasswork ever done. Too bad these features don't come on CW systems. :-)

    (From: Andreas G. Nowatzyk (agn@acm.org).)

    This method may work on lasers with a large volume reservoir, but it does NOT work on Coherent lasers that have a fairly small volume, tubular, high pressure Ar reservoir. At LN temperature, Ar condenses (even freezes), but it has a vapor pressure that is much higher than the operating pressure. Thus the Ar reservoir pressure will stay higher than the tube pressure even at the temperature of LN (77 °K).

    I just tried this on a Coherent IONOVA 100 (20 W) low hour tube that had been unused for over 10 years. It fired up fine, but gave only 4 W in multiline mode (it should be a lot closer to 20 W). At 55 A, it should have a tube voltage of 520 V, but the power supply topped out at 52 A with 607 V. The controller also issued a tube over-pressure warning. I tried cooling the reservoir by immersing it in LN and operated the fill valves manually. This actually made it worse. It seems that Coherent puts quite a bit of Ar into the reservoir.

    This might work if there is a large volume reservoir, that is at *slightly* higher than operating pressure. In this case, the reservoir pressure can go down via cooling enough.

    Unfortunately, Coherent uses a small steel tube as Ar reservoir, less than 1/2" in diameter and about 15" long. The Ar in it is under considerably higher pressure, so that p*v = k*t doesn't do enough.

    Comments on Laser Ionics Ion Lasers

    This is in response to someone who claimed they were junk.

    (From: Aron Bacs, Jr. (aronbjr@bellsouth.net).)

    Actually, if you treat the tube correctly, measure the tube voltage, current, filament voltage, the tube last quite a long time. We have quite a few that are mid 80s vintage. The refill system could be better, but unlike any other system, if you over fill the tube you can surround the fill reserve tank with liquid nitrogen and re-evacuate the tube. Sometimes this process actually improves laser performance, we did this many times and it works like a charm. Again since we used so many, we had a good amount of knowledge on there upkeep as well as an excellent source of mirrors (100s) and spare parts. I will agree that the newer Ionics lasers were not made as well as the older ones. Thus they tended to be more difficult to upkeep.

    As for the supplies, they were brute force linear supplies and pretty simple to fix and work on. Again little things like running mild acid through the supply removed corrosion from the resistor block, and proper upkeep of the pass-bank made the supplies reliable in our (my) experience.

    We have had some really good techs over the years, and one thing is true. If you put effort into proper maintenance and keep good records over time of the your lasers AND the power supplies (via serial numbers, which supply best suits which head, etc.) These laser systems can last a really long time, almost 20 years for us. But I believe this to be true with all laser systems.

    And while the newer Chroma series from Spectra Physics are outstanding, the only repair options are replacement of the tubes and power supply PCBs. We have many of these as well to maintain. But that was how they were designed, and again they work well for us. We have a fair number of 171s, and there is no substitute for proper care and use of these lasers as well. While not as tough as the Ionics, 171s will last a long time. I witnessed an amazing demo once on this subject by Bill Smith president of Ionics, when he lifted the front of a 554 Ar+ laser head (running at full power) of the test bench by about 20" and then let go! I thought I would wet myself just from the noise, but the laser kept running with no ill effects, it convinced me as well as the shipping agent that said that lasers were too delicate for them to cover! The insurance agent pulled out his check book on the spot and wrote us a check for our claim right there! Any way back to the 171 Power supplies. They are linear too, it may be my age, but I can repair a linear supply in 1/10th the time of a switcher!

    Just my 2 cents worth...

    Tests of Some JDS Uniphase/Cyonics Argon Ion Lasers

    While done with multiline tubes, the general characteristics of other Cyonics/Uniphase should be similar in terms of tube voltage versus tube current.

    (From: Piotr Kucharski (Piotr.Kucharski@streamcn.pl).)

    These tests were performed consistently on single JDS supply (except for Head4 that refused to run CW), using the same time frame procedure, same measuring tools, same environmental conditions. Each head was tuned for best performance (HR mirror tweaked).

    The power supply used for testing heads 1 to 3 was a JDS Uniphase 2111-65MLQYV with the remote Cyonics 2500 (LCD).

    The 2111 series power supplies have an advantage over 2114 series supplies that they provide 110 VAC power to the head to run the fan atop the head. 110 VAC is fed through the otherwise not used (and sometimes nonexistent) pins 4 and 7 of the power connector. 110 VAC is not insulated from 230 VAC input line! When checked at the PSU connector (power off!) shows resistance of 5 to 6 ohms (probably additional winding on filament transformer).

    Cathode measurements were made with an AC/(DC) clamp meter and multimeter. The out output power was measured using Ophir analog display with Coherent thermal head.

    Head1 (6,500 hours):

    Filament current 15.7 AAC at 0 A anode current, 2.86 VAC; 16.7 AAC at 4 A anode current.

     Anode Current  Anode Voltage  Power  Comment
    ----------------------------------------------------------------------------
          4              97.8        6.5  5 lines lasing at idle current (4 A)
          5              99.3       18    6 lines lasing (2nd green appeared)
          6             100.8       30
          7             102.4       44
          8             104.4       66
          9             107.2       90
          10            109.3      115
          11            111.2      140
          11.8          112.4      150    Max amps for PSU, temp trip in 2 min
    

    Head2 (500 hours):

    Filament current: 16.7 AAC at 0 A anode, 2.78 VAC

     Anode Current  Anode Voltage  Power  Comments
    --------------------------------------------------------------------------
          4             101.0         7   5 lines lasing at idle 4 A current
          5             102.2        16
          6             104.0        28   6 lines lasing (2nd green appeared)
          7             106.0        44
          8             108.0        64
          9             110.0        85
          10            112.0       105   Quite fast voltage drop
          11            112.7       130
          11.8          114.0       150   Max amps for PSU
    

    Head3 (2,500 hours):

    Filament current: 16.7 AAC at 0 A anode current, 2.91 VAC.

     Anode Current  Anode Voltage  Power  Comments
    -------------------------------------------------------
           4             99.9       10.5  Idle current
           5            101.2       24
           6            103.0       42
           7            105.1       64
           8            107.3       85
           9            109.6      115
          10            111.4      145
          11            113.2      170
          11.8          114.3      200    Max amps for PSU
    

    Head4 (9999 hours):

    Filament current: 17.0 AAC at 0 A anode, 2.89 VAC.

     Anode Current  Anode Voltage  Power  Comments
    ------------------------------------------------------------
           5             94.5        20
           6             96.2        35
           7             98.0        50
           8             101.4       72
           9             105.5       90
          10             110.0      115  Still amazing results!
    

    Head4 tests were performed on DIY switching supply, since head failed to run CW on the JDSU supply. This may be due to low idle voltage (gas depletion).

    A significant fact is that the tube with 2,500 hours was peaking in power. This may indicate that JDSU makes their tubes slightly overpressure, letting them first go up, then down in power, prolonging their life, as ALC used to in the past.

    Omni-543 and Omni-643 Laser Tube Voltage

    (From: Dale Harder (dale@hhr-lasers.com).)

    The test current is 10 amps.

    However, very few if any power supplies will drive these tubes other than those from Omni.

    Discussion about Ion Laser Tube Condition and Operation

    The following was prompted by a posting to the USENET newsgroup alt.lasers (indented) asking about what the tube voltage/current behavior implied about eh performance of two ALC-909 large frame argon ion lasers. Replies are from Steve Roberts.

      ALC909-1:
     Current   PO    Voltage
    -------------------------
       15A    1.0W    202V
       20A    1.9W    199V
       25A    2.9W    198V
       30A    3.8W    197V
    
    

    ALC909-1 is very low pressure, and is reaching the inversion point, a region of operation where the number of ions in the tube equals the number of electrons, instead of exceeding them, as in normal operation. Tube number one is at the point where you need to consider repumping it, running it will result in cathode damage and sputtering as well as faster gas cleanup. Put a ammeter on the cathode, I'd be willing to bet you have a negative or zero delta "T" when the plasma is lit to 22 amps

     ALC909-2:
     Current   PO    Voltage
    -------------------------
       15A    1.1W    224V
       20A    2.0W    222V
       25A    3.1W    220V
       30A    3.9W    215V
    

    ALC909-2 is a middle of life happy tube, doing quite well, probably with very clean optics.

    As you can see, the voltage goes DOWN when increasing current ( really strange in my eyes )...normally the voltage increases when increasing the tube current

    Correct, you are a reaching a pressure region where the laser will lase at reduced power, 488 nm becomes the dominate line and green lines are falling off fast. Now is the time to consider repumping it. You have crossed the inversion point some place between 25 and 30 A.

    I also noticed another very strange effect: when I increase the current beyond 30A (up to 35A for short time. A few seconds .....the ALC909 is rated to 32A ) the output power does NOT increase in a large step. It only increases to about 4 or 4.1 W.

    You have reached maximum laser plasma density for that design. The gas returns can no longer return gas to the cathode fast enough to make up for what is pumped down the bore, the cathode is starved of gas to ionize and you have reached the limit for that tube. I would NOT run it there again, you are risking catastrophic failure.

    Run these at a watt and a half, two watts for a best lifetime/power tradeoff Do not ever expect a ion tube to have a long life at the manufacturer's rated multiline power. Always derate the laser's capability, see the life chart in the FAQ for a 532 air-cooled, on longer tubes like these, the curve is not as steep, and there is a happy region where they will last a long, long, time. Judging from the curves you give, I'd say its 2 watts or so for your good tube.

    Don't expect to find what I have said here to be backed up in books, unless you can find some really early 1960s papers on tube design in a university library.

    The laser also has around 7500 hours on it, it is an ex-medical laser, a HGM-5.

    HGM was formerly was part of ALC. HGM-5's have shorter Brewster stems, a bigger reservoir and are adapted to pulse duty. When an HGM-5 gets so it only starts low, that's about the middle of its life in medical service. The PSU is designed to start at low currents and stay idling to produce a aiming beam, then ramp up to power on demand, so usually all the tech does when the tube reaches this point is set the idle adjust in the power supply lower so the tube reliably starts, and then tells the doctor he's on the downward slope of the life versus power curve. At 7500 hours, that becomes about a 275-350 mW cW laser if you want it to last any length of time in CW service. Make sure it still has a positive "delta T" in operation before you buy.(see faq) usually the medical people do not clean the optics frequently like a lab person does, so don't be disappointed if it has low power and glowing Brewster stems when you get it.

    The other thing is that I don't think its been fired up that much, and hence could be over-pressure. If it still fires, I'm under the understanding that I can burn the excess gas back into the tube material?

    Not as much on this tube as on others, if it were 2000 hours I'd say yes, but not at 7500. Now you may get lucky, they don't usually change the hour meters when tubes are exchanged, and you may actually get a low hours tube, then again....

    Steve's Search for 752 nm

    The 752 nm line is extremely useful in Raman Spectroscopy as it is close enough to the visible to use a standard detector and yet far enough away to minimize sample fluorescence. 752 is specified by many laser manufacturers as a "special" line, i.e., special optics and testing required at time of delivery. While a new fresh 1 watt krypton ion laser straight from the factory with new optics will do ~40-100 mW of 752, it is a very weak gain line and very, very, hard to find, as tubes get older, it dies first, as fast as if not faster then the argon UV lines.

    Most people can faintly see it, but it has been my experience that 50% of the staff in the lab can only see it when it is peaked and at high power. For most people it is too weak to find the faint alignment flashes. It is very dim to the eye, making it dangerous. A tuning prism if used must have well adjusted horizontal tracking to move from the red to the IR. On some lasers prism height compared to the bore center line to the bore can be a issue as there may not be enough tuning range with the factory prism tilt. It can not be said enough that very clean optics are required. We had to change the prism centering.

    For a example "Lexel" 8195KR, the rear optic is a high reflecting flat and the front is either 2 meters or 4 meters radius with a ~.7% transmission. A few lasers were set up with dual coated optics, meaning 752 and 647 will lase off the same high reflector, so you align using a red OC and and then change to the IR OC. Our set made for a "RamanIon" by "Lexel" will also sustain bright 647 and weak 676 lasing off the OC.

    If you have a new tube, proper prism alignment, and scientific quality replaceable mirror mounts, finding 752 is not that bad in a dark room once you are peaked for 647 red, but you must crank up the tube current to maximum and resonator drift can make life very difficult.

    I spent a half a week trying to get this line on a medium power krypton. call after call to the factory (and more then just one laser factory at that) said it lased at normal pressure and normal magnet. Well, yes it will on a very fresh tube almost overdriven. We could get 32-33 mW at about 32-33 amps. The professor's goal was 100 mW, but at 32 A and up tube life would be very short. The other tech finally found it on my day off, but lasing was weak. We re centered the optics and tweaked the prism as much as possible, but it was VERY easy to lose lasing. He found it when the professor walked by and said can't you see that weak red line faintly flashing? The other tech could not see it, so the prof walked the laser, a skill he rarely uses. 35 mW at 33 amps. Not acceptable for lifetime issues.

    The next day I lost lasing trying to clean a now dusty front optic and could not get it back. I was getting desperate, 2 hours at high current and still no near IR. In desperation I clipped the magnetic field wire with a pair of snips. Zing, the current dropped to a max of 12.5 Amps, the tube voltage soared through the roof and 752 sprang up enough to be visible, on the order of 120 mW. However NO MAGNET is one of the worst things you can do to a modern tube. The plasma will eat the bore and cook the anode. The anode relies on the pinch to force the plasma into the correct part of the hollow anode. A call to Brian at Cambridge resulted in the suggestion that you need at least 50% magnet. (Thanks Brian, and also major Thanks to Steve at Cambridge for teaching me about prism height)

    We badly needed that line to run a sample for 24 hours in a time crunch. Our argon and krypton are on the same cooling loop, and cross interlocked so you can only run one at a time no matter what. I needed a resistor at ~70 ohms to knock the magnet down to half current. Jump wiring the argon magnet in series with the krypton magnet did the trick until I can build a magnet regulator. After all, it was cooled and not being used. I removed the argon interlock chain relay for safety just in case. With half field and moderate current, I got 250 mW of the IR line easily.

    The moral of the story is some krypton lines really need variable field, but don't knock off too much magnet if you want your tube to have a long, healthy life. The sad part of the story is that many so called experts had not used this line in so long that details of it are sparse at best. I haven't tired 799 yet, but since it is very low gain under normal conditions, I probably wont bother.

    Dave's Four Color RYGB ILT Laser Saga

    (From: Dave.)

    Back in February of '05, I did a barter deal for a whitelight laser I saw on eBay. It was an ILT-5470K-00 multiline argon/krypton ion laser with 50 hours on it. This laser is specified to do 15 mW total output on 4 lines: RYG and B for its running life. The deal was done and the laser was purchased on eBay for $2000.00 and then traded to me (I love trades). :-)

    I have always wanted a small full color multiline laser and the plus with this model is the 120 VAC power input requirement. Red, Yellow, Green and Blue all in one beam from the convenience of any standard 120 VAC outlet, sweet! I do not know of any other laser or manufacturer of lasers that has a laser that produces RYGB from a standard 120 VAC outlet with a nice high quality TEM00 beam.

    The terms of the auction stated that 7 days were had to try out the laser and could be returned if not happy with it.

    When I received the laser, it was in immaculate condition, and indeed from what I saw it had next to no hours on it, judging by dust buildup and the general all around condition of the unit. I fired it up and as sometimes happens with the shipment of lasers found that it did not lase and was out of alignment. After fiddling and adjusting, I got a beam and walked and tweaked it up to max power.

    I was kind of bummed out because, all I had was blue?? Nice! I thought, 2 grand for a 5 mW blue ion laser. :-(

    I let the thing run for awhile hoping that maybe from sitting it would come around and give me some more lines but to no avail, just blue. 3 lines but blue.

    OK, time for a mirror/Brewster cleaning. I figure it really did not need it just by how clean it was but went ahead and cleaned it anyway.

    Put it all back together and did a optics alignment with my bare-bones autocollimator setup, fired it up and tweaked it in and walked it up to max and now had 4 lines, 3 blues and 1 red line. Both colors were equally bright and way above spec for power but no Yellow or Green? It had a white beam and a brilliant white spot on the wall with only these two colors. I believe this is how the laser was running when the seller photographed the laser for the auction I saw.

    Now with time running out for a return for a refund within 7 days, I began to make decisions whether to keep it or return it. I kept thinking if I run it it will come back, but it never budged.

    As I was running the laser, I would monitor the individual beams on a far wall through a prism. One of the times I powered down the laser , I noticed all the colors flash on the wall for an instant just at the key switch was turned off??

    This had me baffled and I still wonder what was going on for it to do this. I restarted the laser and ran it for a few minutes and shut it down several times and the effect repeated itself the same way every way. Red/Blue while running and RYGB all at once for an instant at the moment of shutdown??

    Seeing that all the lines could be produced and the laser did seem to run perfectly, well almost, Red/Blue only while running, I decided to keep the laser and hope. :-)

    Time went by and I was not happy with my 2 color laser and decided to pull the cover and get a close look at the Brewster windows and optics again and see if I cleaned them well enough the first time I cleaned them.

    The horror took me over when I closely inspected the HR end Brewster window! There was a considerable buildup of dust on the INSIDE of the window!! Now I was bummed and knew that I was not going to get any yellow or green ever from this tube and pretty much threw it up on a shelf and forgot about it for a while. :-( :-(

    Oh well, let's see if I can find an off-the-shelf tube for this laser. I started to hunt the net. I contacted a gentleman on the west coast that rebuilds lasers and he did not have off-the-shelf tubes for ILT and was reluctant to even rebuild it saying it is too difficult to get the green out. :-( This particular person would rather sell me a Lexel but I really wanted the small air-cooled tabletop 120 VAC full color! :-)

    I finally found a company on the net that dealt with ILT and they wanted $3000.00 for the tube. Yikes!!

    I also found a company "Midwest laser products" that sells this laser fully refurbed for $6000.00. Yikes again. :-( (This company is excellent by the way and I've had several very pleasant transactions including the finale of this saga.)

    Well, no luck finding an off-the-shelf tube and I pretty much figured I wouldn't anyway.

    Time went by and then I stumbled across an eBay auction that stated "I will completely rebuild your small air-cooled laser to brand new condition better than factory" and the price seemed right. The auction had ended and no more were posted but I did a search on the seller and found that he worked for a laser rebuilding company in Connecticut.

    I read this company's Web page and felt maybe their prices would be OK for a rebuild. I called them and said what I had and was given a quote for $1400.00 for a complete rebuild including new cathode and Brewsters and a warranty and would be equal or better than original.

    I sent it in and waited.

    Waited, and waited.

    I called and was told that they couldn't get it to lase and they had to re-rebuild it?? :-( :-(

    I waited, and waited.................

    I called again and was told that they couldn't get the color balance right and would probably send it back????? Send it back I said to myself?? Again saying to myself if a rush of thoughts. Send it back how? Not working, half working and what would be the charge?? Yikes, this laser has been a bad luck streak since the beginning. :-(

    I in a semi pleading voice asked "I had really anticipated getting this laser back to spec and if I could talk to the technician". The tech came on the phone and began to explain how much trouble this type of laser is and would be willing to keep trying if I gave him time. So, on the back burner it went and I forgot all about it.....

    Almost a year went by and one of my friends that was admiring on of many of my argon lasers that I have accumulated on ebay asked "what about that whitelight you had ?"

    I replied in a foggy remembrance, oh yeah, I'll have to call 'em and see. My friend couldn't believe I had forgotten about it and angrily said "if it were me, I would have given 'em a ration a long time ago!!"

    That night I had a worry that maybe there was a policy of no customer response within a year the customer might forfeit the laser !!

    I called the next morning with my opening statement "it's been almost a year and what's happening with MY laser???"

    I was told by the woman that answered, "I will look into this and get back to you."

    The next day at my work , I received a FAX stating I was charged for the rebuild and the laser was shipped and a tracking # was provided.

    ???? I was thinking, Wow, I am all set, it's on the way and I will finally have a nice 4 color laser. :-)

    With $3500.00 invested, I open the boxes that arrived and eagerly setup the laser and power it up.........

    She comes right up and was producing a brilliant WHITE beam and an irritatingly bright WHITE spot on the far wall. Finally, I have the laser that I wanted and was happy. :-)

    NOPE! Not yet :-) I did this test at my work where I have all my packages sent and when I got it home and inserted a prism in-line, RYB, no GREEN!

    They RYB'ed it. I do remember telling them I wanted my "whitelight" laser rebuilt and they did do a fantastic job and it does do a perfect white light but NO GREEN and not to original factory spec. The laser actually did well over factory spec - approximately 50 mW - and was happy with that but not happy because I wanted the "4" colors.

    After working with the laser for a while , I noticed I needed to run it at ~9 A for a nice white and this seemed too high because when backed down below 7 A , it produced only Yellow.

    Now I am wondering about the optics and if they were swapped out or if the gas mix was off.

    I proceeded with extra cavity optics experiments to see if I could get the green out. :-) Nothing, not a peep of green.

    Oh well, 5 lines: 3 Blue, a Yellow, and a Red.

    It went back on the shelf and sat, Had way too many other lasers I was working on at the time.

    Ebay, you know, eBay is a great place to get lasers and for the most part. I have been very lucky and with my seemingly high price paid into this ILT, I really did not pay that much depending on how it is looked at. :-)

    I take chances all the time on eBay buying lasers in unknown condition form many that do not have a clue what they are or do not know how to test.

    I take the chance, buy (pennies on the dollar), make it work (most of the time) and resell working for quite a bit more than I paid and everybody seems happy :-)

    Yes, I do get junk lasers and some I have paid hundreds of dollars for, BUT every decent laser I do get makes up for the junkers. :-)

    One day not too long ago , I stumbled across a matched set of large frame Spectra Physics whitelight optics (the big ones:-) and bid. And was very surprised to win them for $68.00. I plan to build a He-Se laser and these would definitely help get many lines out, especially the RED ones. :-)

    Not to go off the subject, I set up the SP OC in front of the ILT's OC in intracavity mode and was very surprised to see "8" lines !! 2 Reds, Yellow, 2 GREENS, :-) and 3 Blues !! WOW, 8 lines and with 2 greens at that ! Yes they were low power and blurry as with a 3rd mirror jury rigged setup but nonetheless 8 lines !

    OK, this tells me that the gas-fill on the ILT seems OK and I am going to take a chance and I did. So I emailed Midwest laser products and asked if they could sell me a set of mirrors for the ILT-5470K-00 laser that they are selling.

    The next day they replied: "will check".

    That night they replied: "Yes and they will cost just short of $500.00 for the set". I replied to them and said: "GO AHEAD and place my order. I want them."

    The mirror set arrived yesterday. I put them in today and YES, Finally, the laser is working the way it is supposed to do ! 7 lines 2 Reds, Yellow, Green, 3 Blues! The beam has a yellowish-white color to it and this is more like it should be due to the yellow and green together. 4 lines at idle and ALL lines at 5 A. :-)

    Nice tight TEM00 beam and a final investment of $4000.00 for this laser so far.

    Krypton lasers are very finicky from what I have read and heard from many people. This RYGB laser that I have now and the way it operates is by pure luck and I really do not think I can get another one to perform the same way this one does. :-)

    Yup, total cost $4000.00 for a "new" RYGB ion laser, not too bad, I challenge anybody to find a new one for less.

    I am happy. :-)

    Well almost not quite done yet ,

    I now know why people do not like ILT lasers! They have the worst designed mirror adjusting scheme going to date!! The ALC 60X is a precision instrument in comparison. The ILT mirror mount/adjustment will break off after repeated back and forth adjustments. :-(

    Now I need to find a junker ILT for the mirror cell that broke in my laser. :-(

    I am sure I can find that for cheap,

    All in all, the thing to think about with mixed-gas lasers is that they are very expensive to buy and get fixed!! White light optics are rare and expensive unless you get lucky. :-)

    The cheapest and most bang for the buck in a small low power whitelight is a Omni/ALC 60X mixed with a HeNe or Red diode. For the cost of this whitelight, I have amassed 5 complete Argon/HeNe whitelight setups that have much higher power output!



  • Back to Sam's Laser FAQ Table of Contents.
  • Back to Ar/Kr Ion Laser Testing, Maintenance, Repair Sub-Table of Contents.
  • Forward to Ar/Kr Ion Laser Power Supplies.


    Sam's Laser FAQ, Copyright © 1994-2009, Samuel M. Goldwasser, All Rights Reserved.
    I may be contacted via the
    Sci.Electronics.Repair FAQ Email Links Page.