This information is provided 'as is'. Readers should only attempt to experiment with this stuff if they are competent at electronics work and understand the dangers of mains voltages. No responsibility is accepted for any damage, injury or death (however serious) or disruption of the space-time continuum caused by anything remotely connected to this website. It is assumed that users are familiar with the potential hazards of lasers - if not, please go to http://www.laserfaq.com first and read the safety information there. Many thanks to Eric King at Laser Innovations, and Steve Roberts at the University of Akron for helping with info about this type of laser.
A while ago, I got hold of some very large ion laser tubes, from a friend who rescued them from a scrapyard - apparently they were from prototype medical laser systems. These are seriously BIG lasers - 3 feet long, requiring water cooling and several kilowatts of input power to produce a few watts of laser output. Two of the tubes were fairly quickly diagnosed as being non-starters (although they would probably be rebuildable by a professional laser refurbisher). The third, a Krypton tube, looked more promising though, as the filament was intact and there was no sign of physical damage. After some research, I identified it as a Coherent Innova 90 tube, requiring 230VDC at 25-35 amps, a filament supply of 3V at about 35 amps, and a starting impulse of several KV. Fortunately the latter two requirements were not a problem as the filament transformer and high voltage igniter unit were still attached to the huge hunk of aluminium that this and one of the other tubes was bolted to, but powering this sucker was still going to be something of a challenge.....!
I started off with a simple pulsed supply, charging a bank of four 3300uF 400V electrolytics, then discharging through the tube by operating the igniter. This conformed that the tube was gas-intact, giving a bright purple plasma flash, but no sign of lasing, presumably due to the mirrors not being in alignment. Aligning in pulse mode was going to be pretty near impossible, as I could only easily produce one pulse per second - it would take forever to find the correct alignment at that rate! I did try alignment with an external helium-neon laser, but still no hint of lasing...
So now it was time to get serious with the power supply..... Large ion tubes like this use solenoids around the tube to produce a magnetic field to concentrate the plasma, and the solenoid on this tube used very heavy-gauge wire, and was designed to be connected in series with the 25-35A anode supply. This had 2 advantages for building a crude PSU - firstly I didn't need a seperate supply for it, and secondly, it would act as a ballast resistance, giving a reasonable amount of control of the tube current by varying the DC supply voltage. Fortunately, when I built my workshop a few years ago, I had the foresight to lay a 10mm2 mains supply cable, good for at least 50 amps. I also had a huge dual-ganged variac (which coincidentally was originally used to control the intensity of two halogen lamps in a very old YAG laser), and four 3300uF 400V electrolytics I bought very cheap from a radio rally (hamfest). All I had to buy were a couple of 400v 35 amp bridge rectifiers - I used two, one fed from each gang of the variac, partly to spread the load, and partly to get a more equal current through each gang of the variac, as the voltages from each side varied by a volt or two depending on the setting. A bit of extra wiring from the breaker box, and a couple of uprated circuit breakers later, I was ready to shove some serious power into this tube....
After hooking up a cooling water supply, I applied some heater current, fired up the variac and hit the igniter button. The tube started drawing a couple of amps with a nice purple glow visible from each end - so far, so good... Gently cranked up the current to 20A, 25A..30A... Loud hum from variac, lights dimming slightly, and that familiar smell of 'hot electrics' from the variac..... I figured that the alignment probably wasn't too far out, due to the previous work with the helium-neon tube, and the OC (output cpupler) end was probably the most likely to be misaligned (as the distance from the alignment laser was shorter) so the easiest thing to do was to remove the retaining cap to loosen off the OC mirror, and wiggle it about, in a highly unscientific fashion.... After a few seconds' fairly random movement, I saw the tiniest momentary red flash - this instantly confirmed that the Beast is Alive and would lase!!! The smell of Hot Variac was getting pretty strong by now, so I switched off to let things cool off a bit, and replaced the mirror retaining cap. The mirror mounting and alignment system on this laser is quite well designed, allowing quick searching for a rough alignment point by rocking the end plate back and forth against spring tension, so you can traverse a reasonable adjustment range very quickly. Powering up again after things had cooled down a bit, it only took a couple of minutes to get some initial lasing - this then allowed me to reduce the current and to peak the output with less chance of things smoking - by this time, even the cable I'd run to the breaker box was getting worryingly warm!
After carefully cleaning the optics, the output is now seriously bright - even at 25 amps tube current, the beam will burn holes in black paper. I've not yet managed to photograph this, as the intensity of the beam completely dazzles the digital camera!
>Running the laser at the 2001 Cambridge teslathon, just
before all the lights went out....!
<Overall view of
> Initial alignment using Helium-Neon tube, mounted on two X/Y
micrometer mounts from a surplus fibre optic fusion splicer.
<The Variac From Hell! Two 10 amp variacs, ganged together, and grossly overloaded. Loud hum and smell of burning varnish....!
>Two 35 amp bridge rectifiers (one fed from each variac) on heatsinks,
and two of the four 3300uF 400V electrolytic smoothing caps.
< Tube Current measurement via clamp-meter (I originally used a current-sensing shunt but switched to the clampmeter when it started glowing!). Red button on right is used to manually discharge the smoothing caps via the big green resistor to the right of the clampmeter. Filament transformer bottom-right.
> Igniter unit - this generates a high voltage impulse to strike the
tube. Brown coil on top passes the 30 amp anode current, and a high voltage generator and
spark gap injects a HV impulse across this inductance to ionise the gas.