If all you have is a HeNe tube but no power supply, see the section: Testing a HeNe Laser Tube Without a Compatible Power Supply for ways to determine if the tube is good. The following applies to both bare HeNe tubes and laser heads though some of the inspection and/or tests will require removing the tube from any enclosure.
Several types of problems can prevent a HeNe tube from lasing properly or make it hard to start:
However, such damage could be an indication of a trauma that misaligned the mirrors - though this is quite unlikely - see the next paragraph.
Thus, if your HeNe tube appears to be glowing like a neon sign or fluorescent lamp (outside the bore) and starts at a very low voltage - perhaps half of the normal *operating* voltage specified for the tube - this may be the cause. A way of detecting it without powering the tube if the problem isn't obvious by inspection (it is hidden inside a laser head) or from the way the tube rattles, sight down the bore of the unpowered tube. In all likelihood, the capillary will now be misaligned enough such that one or both ends will be way off-center or not even visible. And, if it is still held in place by any metal spacer(s) that may be present, there will be no clear path through from one end to the other. Unfortunately, there is no way of repairing such damage. Where only a small part of the bore has broken off, the tube may still lase weakly if the broken part isn't blocking the internal beam path (or it can be jiggled such that this is the case). However, power will be way down.
Note: if you have a high power (long) tube, mirror alignment may not be correct until the tube warms up and/or external permanent adjusters may be required to stabilize the mirrors. Without these, there may be no, low, or fluctuating power. Very slightly pressing on the mirror mounts - or even on various parts of the tube itself - (with a well insulated tool!) will result in a significant variation in power. There may also be a "This Side Up" label on the tube or head indicating the proper orientation for optimal performance. Parts in the tube droop due to gravity (not the electrons, ions, or photons!). This probably applies mostly to HeNe tubes that are greater than 15 to 20 mW, are "other-color" (e.g., green) tubes, and possibly only some types and condition. However, there could be some less dramatic effects with shorter tubes. In addition, just touching one side of the tube with your hand will cool it which may result in a noticeable power change due to the slight contraction which results in a minute but significant bending of the tube and chance in mirror parallelism!
See the section: Checking and Correcting Mirror Alignment of Internal Mirror Laser Tubes for more information.
Aside from manufacturing defects, one way for such a failure to occur is for a power supply fault to drive grossly increased current through the HeNe tube. It is possible for this to result in an abrupt termination of the discharge inside the bore and an inductive kick and huge voltage spike due to the wiring. With the bore momentarily unavailable, the only other path is for an arc through the glass barrier. Like the failure of a MOSFET gate oxide due to electrostatic discharge, once any breech develops, it does not heal! The addition of a spark gap surge protector sized to break down at just over the specified starting voltage may represent a prudent precaution when driving large expensive higher power HeNe tubes. Figure about 25 kV per inch - though this can vary considerably depending on the shape of the electrodes and environmental conditions.
This is one reason not to use a power supply much larger than needed for your particular HeNe tube. I found out the hard way when while violating my recommendation not to use a microwave oven transformer, this happened with a large (35 mW) HeNe tube due to a wrong connection which bypassed the ballast resistor. It was not pretty :-(. The HeNe tube is now good as a sort of high tech neon sculpture but not much else. I even found a defective power supply brick - inadequate start voltage - that powers the sculpture just fine. Now to put it all on a nice polished wood base. :)
See Color of HeNe Laser Tube Discharge and Gas Fill for some not too terrible renderings of a normal tube's bore and some typical problems. (Of course, your computer monitor has to be reasonably well adjusted for these to be at all accurate.) Discharge viewing must be through a glass part of the tube, not the mirrors since their transmission wavelengths will dominate. For an enclosed laser head, it may be necessary to remove one of the plugs on the side or the anode end end-cap (taking care around the high voltage!). The comments about output apply to red HeNe tubes; orange, yellow, green, and near-IR HeNe tubes will likely produce no output at all unless the gas fill is nearly perfect. However, to maximize gain, "other color" HeNe laser tubes will likely have a slightly different discharge color due to modifications to the ratio of He:Ne, the isotopic purity of the gases, and other unknown factors. So, before you blame bad gas, make sure your tube is indeed the normal red variety. As examples of other color tubes:
Various shades of red, blue, and white are symptoms of gas fill problems. Since the total amount of helium and neon in a typical 1 mW HeNe tube is much less than a cubic cm - if returned to atmospheric pressure, almost any leakage or contamination is significant and will likely prevent lasing. Where the tube is 'up to air', no discharge will take place. And, a state of affairs anywhere in between is possible and especially common for old soft-seal tubes. Loss of helium through diffusion is can take place as well. Each of these cases is discussed below.
However, without a monochromator or optical spectrum analyzer, you won't be able to see slight changes in discharge color and these may be enough to kill lasing (though normally, they will be obvious). The only way to really determine if the color is correct where it looks correct and you happen not to have fancy instrumentation is to do a side-by-side comparison with an identical working HeNe tube. I say 'identical' because there can be subtle variations in the normal gas fill from different manufacturers (and from different color HeNe tubes). It may also be possible to take photos (digital or otherwise) of the two tubes (if you don't have two power supplies to run them simultaneously) and then compare those, but getting good color rendition may be a challenge.
Note also that the brightness of the discharge at the same current will almost always be lower with gas fill problems. This may not be immediately obvious unless a good and bad tube are run side-by-side but then it can be quite striking.
If the tube has a getter electrode (see the section: HeNe Laser Tubes and Laser Heads), check the color of the getter spot on the glass in its vicinity. The function of the getter spot is to combine with any unwanted non-noble gases (mostly oxygen and nitrogen) and should generally be black or metallic in appearance if still functional. A milky white, red, or brown color generally indicates that significant air leakage has occurred and the tube is probably no longer functional. Sure, it might be on the hairy edge but this isn't likely! (Note that sometimes a tube will be manufactured with a getter electrode but for whatever reason, it was never activated, the active material remained within its structure, or the active material is transparent. Thus, there is no getter spot, good or bad, and therefore no way to know - from this at least - whether there has been leakage. (For example, all normal (non-barcode scanner) Melles Griot HeNe laser tubes have a getter electrode but no getter spot regardless of gas-fill condition. So there's really no way to know their state of health for the getter.) It may be possible to reactivate the getter electrode by heating it by RF induction or some other means to drive off more getter material that may be present but (1) this is definitely for the advanced course and (2) the likelihood of helping the HeNe tube at this point is small unless the amount of leakage was very very infinitesimal.
(From: Don Klipstein (Don@misty.com).)
I have rejuvenated a couple soft-seal HeNe tubes by heating the getters, either with a glow discharge or a Solar furnace made with an overhead projector Fresnel lens.
(From: Sam.)
I have also revived both a red and a green Melles Griot HeNe laser tube using a jerry-rigged Solar furnace made from a $1, 7" x 10" plastic Fresnel lens intended as a reading magnifier. See the sections: HeNe Tube Lases but Color of Discharge Changes Along Length of Bore and Melles Griot GreNe with No Output for details.
Any source of RF power can be used to determine if a bare tube still has a reasonably low internal pressure (but not if it will lase). However, RF excitation cannot be used to test enclosed laser heads because it is generally not possible to view the inside of the actual HeNe tube and the (metal) case would prevent RF penetration or create other problems.
This approach can sometimes be an effective way for starting some HeNe tubes (even one that is normally hard-to-start) if the ionization reaches enough of the bore. It should certainly be able to substitute for the normal high voltage starting circuits for exposed capillary type HeNe tubes like those in laboratory lasers like the Spectra-Physics 124 and its cousins.
Note: For metal enclosed laser heads, it will likely be necessary to remove one of the end-caps and the wire connection to the tube before being able to apply RF to it from a low current source like an Oudin coil even if the cable is known to be good. Otherwise, the capacitance of the cable will greatly reduce what reaches the tube and there may be no glow even if it is perfectly healthy.
CAUTION: Too much voltage will puncture the glass and ruin the tube. If using an HV or RF source that has an output terminal like an Oudin coil, it is best to attach a couple inches of wire to a metal part of the HeNe laser tube (like a mirror mount) and then touch that rather than the glass. And, use the lowest setting. If it doesn't glow with that, it's not going to do any better at 1,000,000 V. :) DO NOT apply it to the glass unless you are sure the voltage is less than about 10 kV. The dielectric breakdown voltage for the glass of HeNe laser tubes may not be that high! Use the metal parts and wire extension to be safe.
CAUTION: Damage may occur to the HeNe tube if the glow continues for more than a couple of seconds. Don't ask me how I found out (portions of the glass became hot enough to crack). Damage may also occur to you if your parents find out you were using the family microwave for this purpose. :-(
If the color is more toward the pink, lavender, or white, the gas fill may be incorrect or some air may have leaked in. Or, the tube may be end-of-life with significant sputtering around the cathode. See the additional paragraphs on gas-fill problems, below.
More extensive testing and even partial resuscitation of some HeNe tubes may also be possible while heating your hot chocolate. See the section: Using a Microwave Oven to Evaluate and Revive HeNe Laser Tubes for the exciting, but risky, details. :)
Note: In case your were wondering, this is not an effective way of exciting the tube to lase as the discharge intensity inside the narrow bore (capillary) where it counts is way too low. See the section: RF or Microwave Power Supply for HeNe Laser?. As a point of interest, the inventors of the HeNe laser, Ali Javan, William R. Bennett, Jr., and Donald Herriott, of Bell Labs, attempted to use a magnetron for excitation of their original laser in 1960 - and the quartz tube melted! This approach would probably have been quite effective for their wide-bore design if it were not for this minor amount of collateral damage.
However, since you will no doubt insist on experimenting, (1) do so with something other than the family microwave and (2) consider using a Variac to drive the primary of ONLY the high voltage transformer of the microwave generator (fed from the microwave oven's controller). For safety, DON'T attach it externally, DON'T bypass or disable any door interlocks, and make sure the cooling fan is always powered from the full line voltage. This modification will allow some control of power (relatively safely) so that your experiments will be at least less likely to destroy things too quickly. (However, note that the filament of the magnetron is also powered from the HV transformer, so this will limit the useful range and result in some time delay for power to stabilize.) My guess is that adjusting the knob somewhere between 60 and 80 percent, and full voltage will result in 0 to 100 percent of microwave power (the magnetron is a non-linear device which has a threshold voltage below which no output is generated). Then, after you have tried basic nuking of your sacrificial HeNe tube, see what effect a short length of wire attached to the anode (to act as an antenna) will have on excitation of the central bore, add shielding, adjust tube position, etc. Have fun but take care!
Note that if you can sustain a discharge but it is the wrong color, you may have one of those really old Epoxy sealed tubes that leak and air has leaked in. The tube is probably not worth repairing but might make an interesting wall hanging (power optional).
If you have a spectroscope (see the section: Instant Spectroscope for Viewing Lines in HeNe Discharge), it is easy to see if this is the case as the neon lines in Bright Line Spectra of Helium and Neon will be predominant.
One quick test that can be performed visually with a simple diffraction grating to compare the brightness of the neon 585.25 nm line and the helium 587.56 nm line. These are (or should be) two bright adjacent yellow lines. If the mix is correct, these two lines should appear equal in brightness. OK, maybe it isn't so simple since finding those lines by eye could be a rather large challenge. :) However, they can be seen in Bright Line Spectra of Helium and Neon although the helium line is much brighter in this rendering.
It doesn't take too many of those nasty H2, N2, or O2 molecules to affect lasing adversely since they have many energy level transitions much lower than the helium transition to get excited. With just a small amount of unwanted gases, there may still be an output beam, though it will probably be much weaker than expected. One unusual characteristic of such a tube may be that the discharge color is correct at the anode-end of the bore but wrong toward the cathode or may vary in some other way. See the section: HeNe Tube Lases but Color of Discharge Changes Along Length of Bore.
With incorrect pressure and unwanted gases, the tube voltage could be quite different than normal (low or high). The tube cited above had a slightly lower (perhaps 100 to 200 V) operating voltage before having its getter activated. Where the discharge voltage has increased, the tube will dissipate more power while operating, and thus may also run hotter than normal HeNe tubes. Small amounts of oxygen and nitrogen may increase the starting voltage substantially as well. If you can measure tube voltage (see the section: Making Measurements on HeNe Laser Power Supplies), compare it with your tube's specs (see the section: Internal Mirror HeNe Tubes up to 35 mW - Red and Other Colors).
And, I've seen exactly one (1) HeNe tube that had the mirror coating on one (1) of its mirrors totally blown away, most likely due to damage resulting from a lose cathode-mirror mount connection and the discharge taking place inside the mirror mount tube itself.
See the section: Damage to Mirror Coatings of Internal Mirror Laser Tubes for more details.
This seems to be particularly common with Melles Griot (and similar) tubes using a "start-tape" running from the anode almost to the cathode along the cylinder wall. Any condensation will cause problems. I recommend disconnecting the start tape from the anode and removing it entirely by carefully tearing it or pushing it way back inside so the end is at least an inch away from the anode connection. I have never seen the start tape make any difference in starting time though Melles Griot claims some statistical benefit. :)
Similarly, the dropout current may have increased to the point where the power supply current setting is marginal. Increasing the operating current may allow the tube to continue to operate for awhile but it's probably nearing the end of its useful life.
For more on measurable parameters of a HeNe laser tube that can be used to assess its health, see the sections starting with: HeNe Tube Use and Life Expectancy.
It probably doesn't make sense to spend a lot of effort, time, or money to revive a 1 mW HeNe tube that can be replaced for $15. However, if you are ambitious or a new tube cannot be substituted easily (e.g., due to mounting restrictions), see the sections starting with: Repairing Leaky or Broken HeNe Tubes.
First, you need to determine the tube's power connections. See the section: Identifying Connections to Unmarked HeNe Tube or Laser Head if you aren't sure.
There are many ways to power a HeNe tube for the purposes of seeing if it produces a beam. Almost anything that can provide enough voltage to get a few mA through the tube will result in at least a momentary flash of laser light out the end if the tube is good. There won't be any way of determining output power or whether the tube meets specs, but the knowledge that it lases at all may be enough to take the next step - the purchase or construction of a proper power supply.
It is easy to use the family microwave to see if the tube is gas-intact if the tube will fit inside. See the section: Using a Microwave Oven to Evaluate and Revive HeNe Laser Tubes. While this won't tell you if the tube lases, if it fails this test, there is no need to go further.
To test for lasing, current must be passed through the bore of the tube. A couple of options for a quick test power supply are:
Even a high voltage AC supply with appropriate current limiting can be used safely for a few seconds only. (I've been sent HeNe laser tubes which have been operated on AC because the owner copied some power supply design off the Web and didn't know any better. The output power (what of it there was) gradually declined over a few minutes and then there was none.) And even with the rectifier voltage, the tube will be restarting once per cycle which is hard on it so don't run that for too long either. None of these are suitable to operate a HeNe tube continuously unless proper filtering and starting circuitry is added to turn it into a proper HeNe laser power supply.
Don't go overboard though: Too high a voltage applied in the wrong place can arc straight though the glass at which point you have a rather boring high-tech sculpture. :( A very high current can also damage the tube very quickly, thus the need for the current limiting ballast resistance.
With these power supplies driving the tube, if there is any output beam, even if it is weak or in the form of short flashes, the tube is probably good. However, there is no way to tell if it meets specs since HeNe laser output power is only maximum over a narrow range of tube current and these quick test power supplies are at most controlling only average current, not instantaneous current as would be the case with a real HeNe power supply. But, at least you know the tube isn't dead.
It consists of the following components in series:
Wire the output of the transformer in series with the rectifier(s) and ballast resistors. The positive output goes to the anode of the HeNe head or tube; the negative to the cathode. It doesn't matter whether the laser has an internal ballast resistor. Insulate everything VERY well. :)
Powering the laser should result in flashes of coherent light, probably at the power line frequency (60 or 50 Hz). The amount of light will not be that impressive even with a perfectly good high power laser since the current is nowhere near optimal for any length of time, if ever. However, the presence of laser output would confirm that there is life.
WARNING: Since centertap of transformer secondary should be grounded, both outputs of the power supply will be floating with respect to ground. Take care.
In this condition, the tube still lased at a power level which relative to its rated output, is approximately proportional to how much of the bore has the correct color. In this sample, about 2 mW for a tube specified at 4 to 5 mW. I don't believe the starting or operating voltage has been affected very much.
The explanation that makes the most sense is that due to the discharge current in the bore, the few N2 and O2 atoms (and any other party poopers that may have entered without an invitation) are being ionized and pushed toward the cathode of the tube leaving the desired helium and neon atoms to play at the anode-end. The contamination, whether due to a manufacturing problem or an air leak, is so marginal that nearly all of the unwanted atoms are swept from about half the length of the bore. However, the other HeNe tube I have like this had the color change in the exact opposite direct - correct at the cathode but blue-ish-pink at the anode, also reduced power. I now suspect that it may have been internal contamination. More research is needed. :)
Another unusual characteristic of the Northern Lights tube was that the output power (what of it there is) peaked at a current somewhat higher than expected (8 mA as opposed to the 6 or 6.5 mA typical of this size tube). I don't know whether this is simply due to the overall contamination or that more of the nasty unwanted ions being swept from the bore when running at a higher current.
This tube had an unfired getter which provided a means of cleaning up the contamination without a refill. A few weeks later, I got around to making the attempt. And the results are.... See the section: Repairing the Northern Lights Tube.
However, it could also be that your power supply operating voltage, ballast resistor, and other factors may need modification. Of course, if the system used to work reliably and suddenly died, an actual power supply or wiring problem is most likely though a dead HeNe tube is also possible especially if the system has been unused for several years. But don't overlook the unlikely, but not impossible situation where your line voltage is low for some reason! Check it first. The discussion below is somewhat oriented to the situation where a HeNe tube or laser head is being assembled with a power supply (or parts have been replaced) and the combination just doesn't want to work properly. However, some of it also applies to actual failures as well. Where the power supply itself is suspect, see the section: Power Supply Measurements, Testing, Repair.
There are several types of possible behavior depending on how well the power supply, ballast resistance, and HeNe tube are matched up, and if any of these as well as the wiring, are faulty. You first need to determine if the discharge is being initiated at all. If the starting voltage is adequate, there will be momentary flashes that may be extremely short and weak and only visible in a darkened room but operating current may not actually follow. Under marginal conditions, operating current will flow in response to the starting voltage but won't be maintained. These flashes will be brighter and longer in duration. The result may be a nice flashing laser. In fact, this progression is exactly what will be seen when operating a HeNe laser tube from a power supply on a Variac as the voltage is increased: Short flashes followed by longer flashes and at some point, a steady beam.
WARNING: If your HeNe tube doesn't start after a reasonable length of time (like a minute), don't leave the power supply on overnight in a futile attempt to get it going. Starting is a stressful time for power supply components, especially some wide compliance designs, and an extended period with the very high starting voltage on parts of the circuitry may result in total failure. It could also result in electrical breakdown (arcing) inside the laser head or cable. If the laser is flashing, this may be ultimately bad for the tube as well. Turn it off, step back, and try to determine what is wrong.
Where the power supply components and/or wiring is exposed and subject to dirt and grime, first, carefully clean everything to eliminate possible sources of electrical leakage, which can affect operation, particularly the very low current starting circuit. As an experiment, try warming up the unit (which drives off conductive moisture) with a hair dryer or heat gun on the 'low heat' setting. This may enable it to start more easily confirming the need for some housekeeping. :)
First, vacuum and/or dust it off with a soft brush, then use mild detergent and water followed by isopropyl alcohol (rubbing or medicinal is fine as long as there are no additives). Give it ample time to dry completely. The hair dryer or heat gun can be used to help it along. You may now find that your starting problems have disappeared!
If your tube or head has an external starting loop or tape (see the section: Power Requirements for HeNe Lasers), it must be cleaned thoroughly as well (or maybe it has become disconnected, is broken, or has shorted to the case!).
There is also a possibility that something else is shorting out the power supply, possibly only when enough voltage is applied so it won't show up with an ohmmeter test. Sometimes, the ballast resistor inside cylindrical laser heads will arc to the case. This can be checked with an HV insulation tester or more easily for most people, by removing the end-cap(s) and visually inspecting (as well as smelling!) for evidence of arcing, or by disconnecting the anode wire and driving the tube directly from the power supply with an external ballast resistor.
Assuming none of this helps, there are three types of behavior: (1) No action of any kind, (2) an occasional flash possibly at random intervals, and (3) a periodic flashing laser which never settles down to normal steady operation. However, the behaviors and their causes are not really always independent so read through all of the possibilities before replacing components or ripping your system apart!
This generally means that the starting voltage is inadequate for the tube or isn't reaching it, there are other circuit problems, or the tube is bad. In rare cases, shining a light into the tube will allow it to start. Tubes with longer and narrower bores (capillaries) will generally require greater starting voltage and your power supply may just not be up to the task. While tube manufacturers generally specify a starting voltage of 7 to 10 kV (or higher), typical tubes will fire with 3 to 5 times their operating voltage. Thus, a tube that runs on 1,700 VDC will probably start on 5,400 to 8,500 VDC.
In the case of an enclosed laser head with a HV (e.g, Alden) connector, HV cable, and internal (potted) ballast resistor, there may be a breakdown in one of these components and it may only show up when starting voltage is applied (not with an ohmmeter).
Allow the laser to attempt to start for 15 or 20 seconds and turn off the power supply. Immediately pull the Alden connector out of the power supply and discharge it on a metal surface. A nice long (e.g., 1/4") spark indicates that the starting voltage is probably adequate from the power supply and breakdown in the wiring is not likely. If there is little or no spark, either the starting voltage is low or zero, or there is a broken connection between the connector and the tube resulting in not much capacitance to store a charge.
Here are two more tests for this situation:
If the tube now starts, one of the original components was faulty (most likely the potted ballast resistor assembly if the negative connection runs through it) and this will need to be replaced.
Assuming the power supply and wiring check out and the tube is good, the only solution is to boost the starting voltage or use a different type of starting circuit (inverter instead of voltage multiplier, for example).
Note that newly manufactured tubes requiring more than a second or so to start using a compatible power supply are usually rejected as defective and may end up in the hands of surplus dealers who may sell them as 'new' even though they don't meet specs. Thus, you may be more likely to end up with one of these hard starting tubes!
And, it may not only be high mileage tubes. I recently discombobulated (translation: disassembled to harvest its organs) a vintage HeNe laser-based LaserDisc player and found a little incandescent lamp buried near the bore of the laser tube. It is even documented in the service manual (which includes the assembly procedure for the optical pickup), but there is no discussion as to its purpose. However, the way it's wired in suggests that the lamp is not in the original design. The only possible explanation is that it was there to help a possibly hard starting HeNe laser to start. It must have been included to be able to use tubes that otherwise would fail to start quickly enough. I've seen this before with an HP-5501B, and later with several HP-5517Bs, that had a similar quirk: With the cover off, they would start. But with the cover on, they could take a minute or more to start. Nonetheless, for a manufacturer to deliberately add a light bulb to aid starting in a production unit is so strange. Well, I did have a professor whose motto was "If it works, use it". :) That's also my motto for repairing the HP lasers!
Based on tubes I have tested, the starting voltage is much lower with the anode and cathode connections interchanged. However, the voltage drop across the tube when running with reverse polarity is much higher than with correct polarity. Thus, the tube may not run within the normal operating voltage range of your power supply even if the discharge is initiated - more likely it will just pulse.
Nonetheless, even if it just pulses, at least you know the tube is not totally dead. If the tube is otherwise undamaged, there should also be an indication of (at least weak) laser output from the business end of the tube. Perhaps, all you need is a power supply with higher starting and/or operating voltage. An inverter type starter using a flyback transformer appears to be particularly good for hard-to-start tubes. Unfortunately, I do not know of any reliable way of determining the likelihood of success without actually trying it.
I have one 5 mW HeNe tube that requires (depending on its mood) as much as 15 to 20 kV to start (it should be less than about 10 kV). However, once started, it runs with a normal operating voltage of about 1,800 VDC.
WARNING: Do not let the HeNe tube run for any length of time with reverse polarity as damage may occur due to heating and sputtering at the anode end of the tube.
This sort of behavior is probably more likely with a pulse type starter but can occur with other types as well. What is likely happening is that the energy is insufficient to fully ionize the gas inside the bore of the HeNe tube so the discharge doesn't 'catch'.
In addition to the other possibilities listed above and below:
What happens is that the discharge is initiated but the voltage drops too much at the tube anode and the discharge goes out. This cycle repeats resulting in a flashing HeNe laser.
To produce a stable discharge, the following must be satisfied:
These factors are not independent. Since the negative resistance and sustaining voltage of the tube are not normally specified and depend on current, some amount of trial and error may be required to achieve consistent stable operation but in most cases it really is very easy.
Cycling behavior can be due to several factors:
If the transformer or inverter drops too much under load, the tube voltage may fall below the minimum for the tube/ballast combination as soon as it starts. This cycle will repeat continuously or it occasionally may catch.
Use a higher voltage and larger ballast resistor, and/or increase the uF value of the main filter capacitor (and/or the one in the DC supply to an inverter type supply as well if it isn't regulated).
Minimum capacitor values for less than 5 percent voltage ripple (typical voltage and current requirements):
Actual ripple in the current to the tube may be several times greater than this since it depends on the change in voltage with respect to the total effective resistance of the PS+tube+ballast resistor combination). However, the resulting ripple in the optical output power will be 2 to 10 times lower than the ripple in the current depending on operating point. The lowest will occur around the tube's optimal current specification.
For an unregulated power supply, increase the operating voltage and/or decrease the ballast resistance.
For a regulated power supply, decrease the ballast resistance so that the voltage for the desired operating current falls within its compliance range.
Shorten the wiring - minimize the distance between the power supply and ballast resistor, the ballast resistor, and tube anode, and don't use long runs of high voltage coax (which may have higher capacitance). Increasing the energy of the starting circuit slightly may help as well.
Also see the sections: How Can I Tell if My Tube is Good?, About Hard Start HeNe Tubes, Testing a HeNe Laser Power Supply, Power Supply Construction Considerations, and Adding a Start Wire.
As far as I can determine, the fundamental physics behind this phenomenon may not even be well understood by the major laser companies. The only meaningful data is statistical, because even a give tube with a given power supply will have dramatically different start times from attempt to attempt, as will tubes built side-by-side through the entire production process.
Tubes that are kept in dark cold environments for long periods of time don't tend to start well. But, once one of these tubes is started successfully, restarts will likely be instantaneous, or at least reasonably quick. However, left overnight, they will revert to being uncooperative.
Also lower fill pressures and cleaner tubes make for hard starting - not to mention power supply variables.
Some manufacturers (e.g., Melles Griot) use a conductive 'start-tape' running the length of the tube attached to the anode electrode to aid in starting. It's not even really proven that this improves performance (and I've found that it can be a source of electrical breakdown problems. I've never noticed any difference in the speed of starting after removing the start-tape). Uniphase had a pointed electrode inside the anode mirror sleeve to aid in starting but it isn't obvious that it made any statistical difference either. There has even been talk of using a trace of radioactive gas (as used to be common in neon indicator lamps and glow tube fluorescent starters), but this of course would probably not be a popular idea today!
A given production line may still have hard-start related yield problems from time to time (which kind of suggests the Ph.D physicists don't understand it). Funny thing is, no one can tell anything that's different on a hard-starter versus a regular one.
For some hard start tubes that otherwise run well without current drop out problems, adding a start wire directly from the anode with a few turns wrapped around the tube near the cathode may help starting dramatically. For example, on Spectra-Physics side-arm laser tubes, there is a section next to where the cathode attaches to the main bore where adding a start wire may work quite well. This allowed an SP-155 which would almost never start at normal line voltage to start instantly first time every time.
Use a well insulated wire connected before the ballast resistors. Wrap the end a few times around the narrowest section of the tube near the cathode or use a metal clip. Some experimentation may be required. Just try not to zap yourself excessively during testing. :)
However, if used with a some linear power supplies having a voltage multiplier starter (probably without a final diode/filter stage), the tube must run stably substantially above its lower current dropout point or else the start wire will tend to turn the tube off as well as turn it on and the result will be a flashing laser, which is usually not a good thing. Adding a HV diode in series with the start wire might help, but I haven't tried it.
As added protection for you and the power supply, instead of connecting the start wire directly to the raw high voltage output, install a series capacitor with bleeder. I would suggest 1 nF with a 100M ohm resistor across it. Both should be rated for the peak starting voltage available from the power supply, typically 8 to 15 kV.
And, for other-color HeNe tubes which have much lower gain for a given length than red HeNes, all of the above may apply. The following comments were prompted by questions about a non-lasing short green HeNe tube (similar to a Melles Griot 05-LGR-024, 215 mm in length:
(From: Lynn Strickland (stricks760@earthlink.net).)
Those things are touchy, touchy, little SOBs. They usually have an almost flat HR and OC combination. If it does lase, it will probably be a few tenths of a mW at best. Probably have to walk the beam AND tweak both ends for any hope. Try some magnets too, for 3.39 micron suppression. In general, low power greens are a bitch to tweak.
Note that the green discharge is more 'pink' (red tubes more 'orange'). Fill mixture is a little different, but the different color mostly due to lower fill pressure - which is why greens have shorter lifetimes than red.
For example, I found that some recent samples of the popular Melles Griot 05-LHR-911 HeNe laser head, rated at 1 mW minimum power output, were all made with neutral density filters to assure that the maximum power output was less than 1.5 mW. With the filters removed, it jumped to between 1.8 and 2.1 mW! Apparently, the filters were individually selected to get the lasers as close as possible to 1.5 mW without exceeding it since their attenuations were not all the same and the weakest laser in the batch (with the filter) actually ended up having the hottest tube.
More likely, the manufacturer accidentally used too large a bore for the length of the resonator and the mirror curvature. For example, if this is a green (543.5 nm) HeNe laser, they may have used a bore sized for a red (632.8 nm) HeNe laser by mistake resulting in a mode diameter that is too large. Or, it might have been designed on the hairy edge, size-wise, in an attempt to get as much power as possible out of the tube and the engineers weren't lucky that day. With a correctly sized bore, slightly incorrect mirror alignment would result in lower power but maintain a TEM00 beam profile.
If you had been the original owner, the laser might have been replaced under warranty. As it is, you now have what I generally call an "interesting" laser. :) Or, since the specifications are often only with respect to "95% mode purity", if the hole represents less than 5 percent of the power, maybe it's considered acceptable, though I can't imagine anyone being entirely happy with a laser that's supposed to be TEM00 having a hole in the middle of the beam unless all they care about is the number of photons per second!
However, even with too wide a bore, it may be possible to adjust the mirror alignment to obtain a TEM00 beam. Gain access to the cathode-end mirror mount (to avoid a shocking experience) and allow the laser to warm up completely in the orientation that will be used. Aim the laser at a screen of some sort like a white business card, and gently press sideways on the mirror mount from orientations every 45 degrees or so (top, bottom, left, right, and the diagonals). Closely examine the spot shape to see if it pops into TEM00 at any time while still maintaining near-maximum power. If it does, it should be possible to adjust the alignment as described in the sections starting with: Problems with Mirror Alignment. A power meter will be useful to assure that the output power is still near maximum. It may in fact end up being maximum with the TEM00 beam, though some other beam profile when cold. Very likely, the alignment will be very critical. Several attempts with multiple warmup cycles (to allow the metal mirror mount stems to settle in) may be needed to achieve consistent behavior. I adjusted a green HeNe laser head that formerly had a doughnut beam. It now starts out TEM01 but becomes a beautiful TEM00 once it warms up, and significantly exceeds spec'd power. But I haven't decided whether a consistently interesting doughnut beam or a boring high quality TEM00 beam only when warmed up is preferable. :) CAUTION: You could make the laser worse or dead by attempting mirror alignment. So unless you've done this successfully before, it may be best to leave well enough alone and enjoy the unusual behavior. There's nothing else you can do about it!
See the section: Basic HeNe Laser Power Supply Considerations.
However, a faint clicking or snapping sound may actually be normal during starting if the power supply uses a pulse starting technique or is cycling a PWM controller attempting to start.
Also see the sections: How Can I Tell if My Tube is Good? and Starting Problems and Hard-to-Start Tubes.
A different power supply or slight adjustments or modifications may make your HeNe tube happy, at least temporarily. However, where the HeNe tube is an inexpensive vanilla flavored variety, replacement may be the easiest solution if it turns out to be marginal. :-)
The symptoms are that the tube may start normally but then go off and restart, possibly quickly and unpredictably. One possible cause is a bad internal connection between the cathode can and its attachment to the mirror mount where the negative lead of the power supply is hooked up. The type of construction susceptible to this malady is where a 'nipple' on the end of the aluminum cathode can is swaged (pressed/squished) into the mirror mount rather than actually being attached by spot welding or via a spring contact. After many thermal cycles, the swage can loosen resulting in intermittent contact especially as the tube heats and parts expand. (This is sort of the same problem that aluminum house wiring can have if improper termination techniques or devices are used, but in that case, the consequences can be much more disastrous!) Any sort of high resistance increases the required tube voltage since the mirror mount has a higher 'cathode fall' voltage drop. The discharge will likely go out and the power supply will then attempt to restart. In some cases, the discharge may strike to the mirror mount itself (look for a glow near the mirror) and if this persists, will eventually destroy the mirror. (See the section: Damage to Mirror Coatings of Internal Mirror Laser Tubes) After the tube warms up sufficiently, since aluminum expands faster than steel or Kovar, the problem may disappear once the connection tightens. However, until then, the intermittent contact and many restarts is hard on the power supply and nearby mirror.
Assuming the power supply and tube are properly matched and the power supply isn't defective, this is a defective HeNe tube. No cure is possible. This is a relatively unusual problem (I've only seen it in two (2) HeNe laser tubes so far) so first check external connections and make sure your HeNe tube and power supply are properly matched. If its maximum voltage is marginal, as the tube heats up, the voltage drop may increase just enough to result in erratic behavior. However, one possible difference between this and a bad cathode connection is that with the latter, the condition may clear up once the tube heats up since the expansion of the aluminum cathode will improve contact. The marginal voltage situation will just get worse. The power supply itself could also be defective. The easiest way to determine which is at fault is to swap the PSU and/or tube with known good units.
I've seen this malady on a few older Melles Griot HeNe lasers. On newer ones, a thin metal conductor has been added - presumably in response to this type of failure - connecting the spider/bore support (which is in good contact with the cathode can) directly to the end-cap. It is spot welded at both ends.
Also see the section: Unstable or Flickering HeNe Tube.
The male pins can be cleaned with a file or fine sandpaper. The female contacts can be tightened by wedging a small flat-blade screwdriver inside the connector between the side and the plastic contact support. Entire replacement Alden connectors are readily available or can be salvaged from dead laser heads or power supplies. To avoid the possibility of arcing or a shocking experience, use at least 2 layers of heat-shrink tubing for insulation with a minimum of 3/4" beyond the bare wire sections, and stagger the splices. Add another layer of heat-shrink over the completed splices.
Note that if the discharge is actually going on and off, the cause is entirely different - an incompatibility with the power supply, incorrect ballast resistor, low line voltage, etc. See the section: Unstable or Flickering HeNe Tube.
However, sometimes you will find a laser that exhibits significant periodic variations in output intensity even where the discharge is perfectly stable. There are two types of phenomena depending on the period of the power cycles:
These result in fewer longitudinal modes having sufficient gain to sustain the lasing process. As the resonator length changes, these lines move with respect to the gain curve of the lasing medium. Where there is cyclic variation in output power, only a very few lines are of sufficient gain to sustain lasing and then only when they are near the peak of the gain curve. The tube stops lasing entirely when there are no lines with sufficient gain to sustain oscillation. See the section: Longitudinal Modes of Operation.
High mileage tubes with low gas pressure and tubes that are leaky (usually soft-seal but not always) with a contaminated gas fill may produce a very weak beam that comes and goes in a similar manner. (Such a tube may also be hard starting or erratic on its normal power supply independent of the slow fluctuation in in output power.) Very short and very long tubes are more susceptible to these effects. Short tubes have fewer possible longitudinal modes available so as the gain falls off with use, the variations become more pronounced. Similar behavior may be present with some yellow and green tubes since their gain is so low to start with and everything is critical.
For longer HeNe lasers, in addition to the mode sweeping at the output wavelength, there may be a longer period power variation due to power stealing by the unwanted 3.39 um line if it isn't adequately suppressed by bore/mirror design or magnets. This would occur at a rate of 0.632.8/3.391 as fast as the 632.8 nm mode cycling (for a 632.8 nm laser). If the laser output power is recorded over time, one would see the effect of the 3.391 um superimposed on the shorter one but it won't show up as a smooth variation - more like mountain peaks appearing within rolling hills. :)
To confirm, try adding some medium strength magnets along the tube or head. Experiment with the number and orientation of the magnets but a half dozen with alternating polarities along one side of the tube are typically adequate. If the magnets reduce the amplitude of the 3.391 um fluctuations (and probably increase the average output power by up to 25 percent or more), poor design is the likely cause. Among other things, the mirrors are too reflective at 3.391 um. Aside from installing the magnets permanently, there isn't much that can be done.
A similar sort of varying intensity behavior will result if a polarizing filter is placed in the output beam of a randomly polarized HeNe tube or a HeNe tube that is supposed to be linearly polarized but isn't working properly because its internal Brewster plate has fallen off or its polarizing magnets have weakened or are mispositioned. However, in this case, what happens is that as the laser switches between longitudinal modes and/or the mirror alignment shifts ever so slightly, the polarization angle and thus the output intensity of the beam may change significantly. This is perfectly normal for a randomly polarized tube but indicates a problem with one that is supposed to be linearly polarized. See the section: Unrandomizing the Polarization of a Randomly Polarized HeNe Tube.
(From: Daniel Lang (dbl@anemos.caltech.edu).)
"The typical HeNe laser's gain curve is wide enough for 2 or 3 longitudinal modes to oscillate simultaneously. As the laser warms up, the cavity expands, causing the modes to decrease in frequency. When a mode gets too low with respect to the HeNe linewidth, it goes out and after a bit, a new one appears on the high side of the linewidth. This typically has a period of 3 to 10 seconds. I suspect that an old laser that is doing this is down to 1 or 2 modes due to reduced gain and may be approaching 0 or 1 mode, causing a visible intensity modulation.I noted a similar problem when using a HeNe for Laser Doppler Velocimetry. In this case we were seeing a low level intensity modulation that would start at approximately 60 Khz, sweep through zero and back to 60 Khz and then disappear for several seconds before starting again. The entire cycle repeated in approximately 5 to 10 seconds. The longitudinal mode spacing for our laser was 385 MHz. The sweep between 0 and 60 kHz only appeared when the laser was operating in 3 modes. The frequency difference between modes 1 and 2 was not quite the same as the difference between modes 2 & 3 except when exactly symmetrical (amplitude of mode 1 = amplitude of mode 3). We were seeing the difference of the differences! The longer interval free of intensity modulation occurred when only 2 modes were oscillating."
For more information on this phenomenon, see the section: Longitudinal Mode Pulling.
A simple test to confirm thermal gradients as the likely cause is to gently press on each mirror mount (careful: high voltage!), or perhaps even in the center of the tube if it is supported at each end. If power can be restored to near normal no matter what its value by doing this (the direction and force required will not be constant), it is likely a thermal problem.
Therefore, it is important to mount long higher power HeNe tubes both at the recommended locations (usually by gently clamping the glass near the ends of the tube) and in a case to promote temperature uniformity and isolate it from convection currents. The alternative is messy: Active feedback to monitor output power and tweak the mirror mounts with a servo system. :) Long yellow and green HeNe laser tubes are particularly susceptible to very erratic behavior as a result of thermal effects. If not mounted in a suitable enclosure, it may not be possible to achieve mirror alignment that results in stable output power and fluctuations of 100 percent could result. In other words, the beam may vary in output power even to the point of disappearing entirely over a period of a few minutes. In fact, Melles Griot will not even sell yellow, green, or other low gain HeNe laser tubes by themselves (not mounted in an enclosure) as standard products, at least in part for this reason.
If you are experiencing excessively short life (e.g., a month instead of years), the first things to check are operating current and polarity. See the section: Making Measurements on HeNe Laser Power Supplies. Of course, if you omitted the ballast resistor, life will likely be very short. :-(
If the HeNe tube and power supply are mismatched, one can damage the other. For example, running a 1 mW HeNe tube on a power supply designed for a 35 mW HeNe tube may not only result in too high a current by design (e.g., 8 mA instead of 3 mA) but may also result in much higher current if the compliance range of the power supply is exceeded (i.e., the voltage across the HeNe tube is much lower than the power supply can handle). Conversely, attempting to power a 5 mW HeNe tube using the power supply from a barcode scanner (designed for a .5 to 1 mW HeNe tube) will likely result in a blown power supply. Just because the high voltage connectors mate and/or the tube lights up doesn't imply anything about compatibility! Also note that maximum optical output occurs at the optimum operating current - too high or too low and it goes down. (Operating current for yellow, orange, and green HeNe tubes is even more critical than for the common red variety so setting these up with an adjustable power supply or adjusting the ballast resistance for maximum output is recommended.)
New and even used HeNe tubes and power supplies from reputable surplus dealers will generally last a long time if not abused. But, much of what you get at swap meets and hamfests has been pulled from equipment for one reason or another. So, the problems you are experiencing may have nothing to do with your setup!
(From: Lynn Strickland (stricks760@earthlink.net).)
Speaking as a non-physicist....
There are so many variables in a gas discharge, it's a game of averages. That's why the power supply business can be so tricky - and why, for the power supplies you can look inside of, you see so many modifications. That, and the rate at which electronic components go obsolete keeps it in a continuous state of flux (no pun intended).
Reasons for the variability in lifetime and failure mechanisms from design to design revolve around design fill pressure and gas mix, operating current, distance from capillary bore-end to cathode, optical design (some designs are more sensitive to misalignment than others). Also power supply variability, ballast resistor value differences, operating current tolerances (often set at, say, +/-0.2 mA).
Gas lasers can be a pain, but for a lot of applications, they're still the most cost effective solution -- in some cases the only solution.
If the HR mirror substrate was ground without any wedge, the coated mirror surface and the outer surface of the HR mirror glass will form a Fabry-Perot etalon or second resonant cavity whose transmission will depend both on the exact lasing wavelength (which is changing due to mode sweep) and its thickness (which is changing due to thermal expansion. When the distance between the two surfaces of the HR mirror is a multiple of 1/2 wavelength (possibly plus 1/4 wavelength since the outer surface is to a lower index of refraction) of a lasing mode, the reflection will be a maximum. This will modulate the effective mirror reflectivity by a surprisingly large amount resulting in a corresponding variation in the waste beam power, as well as an inverse variation in main beam power. For a normal red HeNe laser where the waste beam is perhaps 30 uW and the main beam is 1 mW, if the waste beam power changes by 100 percent (30 uW to 60 uW), the main beam power will change by approximately 3 percent. So, this might go unnoticed under normal conditions. However, some applications use the waste beam to monitor the lasing modes, and for those, a tube with this malady is next to useless.
So, now for the analysis. (Don't worry, this isn't too bad.) As the laser tube heats up and expands, there will be the normal mode sweep for the resonator formed by the two mirrors as the longitudinal modes drift under the gain curve. This results in a change of frequency within the Doppler-broadened neon gain curve over a range of about 1.5 GHz corresponding to a change in wavelength of roughly 2 picometers. The total power of the main (output) beam then varies slightly due to differences in gain for the modes depending on their position on the gain curve.
But, there will also be a smaller increase in the thickness of the HR mirror glass so there will be its own slower change in behavior. For most HeNe lasers, the HR mirror has a reflectivity very close to 1 - 99.9% is typical. The uncoated outer surface will have a reflectivity of around 4%. If there is wedge, then the 4% reflection from the outer surface does little more than create the ghost beam. But if these two surfaces are parallel, they form a Fabry-Perot etalon which can have an effective variation in transmission of up to 2.25:1 (if everything is perfect, more below), and this will cause a similar power variation in the waste beam and inverse power variation in the main beam.
The relevant calculations are (1) to determine how the transmission function of the weak etalon of the HR mirror varies with temperature and (2) to determine the approximate number of variation cycles based on temperature rise.
(1-R1)(1-R2)
T = ----------------------------------
[1-(R1R2)1/2]2+4(R1R2)1/2sin2(phi)
Where:
This equation reduces to the more common one found via a Web search if R=R1=R2. And indeed, since Tmax/Tmin for an etalon with R1 very close to 1 and an arbitrary R2 is the same as for an etalon with the reflectivity of both mirrors equal to R=sqrt(R2), solving one of those equations for R=sqrt(R2) will return a nearly identical result (though, of course, the actual transmission will differ dramatically!).
So, for R1 being the HR mirror with 99.9%R and R2 being the outer surface with around 4%R, the equation reduces to:
0.001 * 0.96 0.00096
T = ---------------------------------------------- = ------------------
[1-(0.999*0.04)1/2]2+4(0.999*0.04)1/2sin2(phi) 0.64+0.8*sin2(phi)
So, T varies by a factor of about 2.25 (Tmax/Tmin) due to the sin function going from 0 to 1 as phi changes due to thermal expansion of the mirror glass. Note that the exact reflectivity of the HR doesn't alter this result by a significant amount as long as R1 is close to 1. Thus, that ratio of 2.25 for the maximum variation in HR transmission (1-R) will be essentially the same for any HeNe laser with a non-wedged but ground and polished HR mirror.
Condition Relative T Condition Transmission
----------------------------------
Min (phi=90°) 0.67
No Etalon 1.00
Max (phi=0°) 1.50
But depending on the actual parallellism of the HR surfaces, and the condition of its outer surface, the ratio of 2.25:1 may not be achieved.
Note that where the reflectivity of the HR is much closer to 1 than the reflectivity of the OC as with most red (632.8 nm) HeNe lasers, the power of the small waste beam will vary by a ratio of up to 2.25:1 but the power of the much larger main beam will only show very small inverse ripples. And the total power from both ends will be essentially constant. However, if the reflectivities of the HR and OC are similar - the power from both ends will vary significantly, though by much less than that ratio of 2.25:1. For a very low gain laser, the total power will also get somewhat smaller as the HR transmission gets higher since it is running very near the lasing threshold even under the best of conditions.
Lm * n * (Cex + Cn) * (Tf - Ti)
N = ---------------------------------
2 * Lambda
Where:
Assuming the temperature of the mirror climbs from 20 °C to 70 °C during warmup, for a 4 mm thick mirror substrate, the optical length will increase by about 2.76 um or 8.7 half-wavelengths at 633 nm.
Using a similar approach, the number of mode cycles for the main tube will be:
Lc * Cex * (Tf - Ti)
Number of Mode Cycles = ----------------------
2 * Lambda
Where:
For the same temperature rise, a laser head like a 05-LHR-171 with Lc of 400 mm, the distance between the mirrors will increase by about 65 um, corresponding to 205 cycles at 633 nm.
For a bare tube, the temperature rise will be less, so both the number of mode sweep cycles and power variation cycles will be smaller.
It's easy to measure the power variation with any sort of optical power meter or even simply a silicon photodiode and a multimeter on the uA range. There is no need to wait for the tube to warm up on its own to do this. A small blow-dryer can be used alternately on the heat and no-heat settings to vary the temperature of the HR mirror mount up and down. Or, if the waste beam is from the cathode-end of the tube, even finger warmth will work. But trying this on the anode-end will result in a shocking experience!
I built a little widget with a ThermoElectric Cooler (TEC, Peltier device) to do this more precisely for barcode scanner tubes, which tend to suffer from lack of HR wedge much more so than higher quality and larger HeNe laser tubes. The TEC is clamped between a small plate which is soldered to a mirror mount clip, and a transistor heatsink with small fan attached. A 10K ohm thermistor is glued against the clip to monitor the temperature and the outside of the clip is covered with hot-melt glue to add some thermal insulation. Everything weighs in at only 20 or 30 grams and clips on the mirror mount without detectably changing alignment. Currently, this rig can only be used on the cathode end mirror because it isn't electrically insulated for the high voltage present at the anode. But this is sufficient for dealing with the HR mirrors of barcode scanner tubes, which are all anode-end output. (At least the ones I have are.)
Applying negative or positive current via a pair of switches (one for on/off and the other for polarity) to the TEC allows the mirror's temperature to be ramped up and down. It should be possible to use a TEC controller to maintain a constant temperature, though I haven't tried this.
Here are some specific examples of tubes with major waste beam power variation.
Siemens LGR-7641 red HeNe laser tubes are apparently made without wedge in the HR mirror (at least some of them). For a red tube with its much higher gain and lower OC mirror reflectance, there will be virtually no detectable variation in output power due to interference in the HR mirror, but waste beam power could still change significantly.
Plot of Siemens LGR-7641 HeNe Laser Tube With Waste Beam Power Variation During Warmup (Uncorrected) shows the behavior of a healthy fully to spec tube. The tube was enclosed in a thermal blanket (a bunch of thin packing foam) so that its temperature increase would be higher and similar to that of a cylindrical laser head. Several complete cycles of dramatic power variation is clearly evident.
The cause being due to the etalon was confirmed by putting a dab of 5 minute Epoxy on the outer surface of the mirror. The Epoxy is smooth and clear enough to pass sufficient power for the photodiodes (though the power is lower). But the Epoxy is lumpy enough to greatly reduce the power variation. The glass and Epoxy are fairly closely index matched so that the dominant reflection is no longer from the planar glass surface but from the lumpy surface of the Epoxy. So there is minimal reflection directly back along the optical axis and thus minimal etalon effect. There is still a small amount of variation that doesn't track the output power in the main beam but it is greatly reduced. The residual long term fluctuations are at least in part due to the imperfect index matching of the glass to the Epoxy.
I expected that an optical wedge would virtually elminate it. But, adding a Brewster plate from a Melles Griot polarized HeNe laser tube glued to the HR at an angle using Norland 63 UV cure optical cement to attach it and fill the gap resulted in almost no change. The wedge assembly looks a lot better than a glob of Epoxy, but doesn't work any better and may in fact be very slightly worse. Admittedly, the index of refraction of Norland 63 (n = 1.56) isn't quite the same as that of optical glass (n = 1.50 to 1.53) so some improvement should still be possible.
Running the numbers for a few residual reflectance values using an Excel spreadsheet, it can be seen that even a 0.04 percent reflectivity for the outer surface of the HR would result in around 8 percent variation in waste beam power, similar to what is shown in the two plots of the tube with correction:
Residual R Pmax/Pmin
------------------------
10.00% 3.7028
4.00% 2.2491
1.00% 1.4935
0.40% 1.2881
0.10% 1.1348
0.04% 1.0833
0.01% 1.0408
0.004% 1.0256
0.001% 1.0127
The Fresnel equation for normal incidence reflection for materials with index of refraction n1 and n2 is:
n1-n2
R = (------)2
n1+n2
Plugging in n1=1.50 and n2=1.56, the result is just about 0.04 percent. How about that. :) So, going to Norland 65 with n=