LASER problem

[itsmoked]

I'm working on a 100W Q-switched YAG LASER. It won't strike the pump tubes.

This thing is pumped by two Xenon arc tubes that are hooked electrically in series. Like all arc tubes they run at some reasonable voltage but must first be struck by an unreasonable voltage.

Typically a DC voltage is impressed across the tubes and a shunt inductor is put in place. Since the tubes represent a very high impedance the current runs thru the shunt. At some point the shunt circuit is interrupted. The forced collapse of the inductor's magnetic field causes a very high voltage spike. The spike is sufficient to ionize a path thru the tubes. Immediately the main supply current follows thru keeping the tubes ignited.

In this LASER's case the running voltage across the tubes is about 300VDC however during the ignition cycle it's raised to about 1800V. Then the Igniter starts cycling the inductor shunting cycles. It does this about every 4 seconds. Until the tubes ignite or it times out.

In my case the tubes are not igniting. The HV DC supply was replaced. The igniter was replaced and the tubes have been replaced.

Does anyone have any experience with this problem or suggestions on further troubleshooting methods.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[ScottyUK]

Not a laser expert and haven't played with a big one for over fifteen years so sorry if this a dumb question. What are you using as the interrupter? Is it fast enough or is there residual arcing which is draining energy from the inductor which should be going in to striking the tube?


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[Skogsgurra]

Adding to Scotty's question. Is the interrupter the original one? Was there a small HV capacitor involved? The capacitor is sometimes needed to avoid that an arc is formed when the 'points' when the switch opens. If it isn't there, most of the energy will be dissipated in the arc - as Scotty said. With the capacitor, there is enough separation when the voltage rises and finally strikes the tube.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[itsmoked]

I only had a few minutes to delve into the unit so far. The Power comes to the starter as two wires from the DC supply.

Then leaves as two wires to the Xenon tubes. I tell you this so we recognize that there is no other power or control signals to the starter. It just runs when the power is put to it.

It consists of a bunch of caps that look like your typical potted-in-blue rectangular box X or Y caps. They're about 1" x 1" x 0.5". I'd say there are about 5 shoulder to shoulder. There is also the inductor which is potted in the new unit and unpotted in the replaced unit. It's about 4" x 4" x 1".

There is one of those small arc suppressors in the middle of it all. You know, the little can with two axial leads sticking out of it. Inside will be two little domes separated by a few millimeters. And then it's wrapped with some sort of non-conductive can.

That arc suppressor is what flashes every 4 seconds. It looks like a tiny flash bulb. Flash....flash....flash...

I'm theorizing that the starter is working like a reflex oscillator. The caps charge up to the breakdown of the arc suppressor and BANG.
No start? Cycle repeats itself until the controller 10 feet away fails to see the required current or voltage drop caused by running tubes in the allotted starting period.

So Scotty I'm not actually seeing a clear interrupter.

Which means how is it actually supposed to pull this off?

Skogs; Umm OH! Yeah you mean like in an automotive ignition system. The condenser. If it's missing the ignition points erode quickly and the high voltage spark is anemic. I see what you mean.

Still not seeing a valid interrupter. Only a "shorter" in the arc-suppressor trigger.

I'll fish up some more info. Maybe I can get a picture of it so we can analyze it completely.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[MacGyverS2000]

No solution for you, Keith, but I find this thread interesting... I have been looking into making a 100W DPSS fiber laser.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[IRstuff]

Dan, a DPSS laser is substantially different than a flashlamp laser, so much of what's required for that is not required for a diode pump.

Keith, this out of my usual haunts, but the strike voltage seems low to me; I would have thought something in the mid to hi kV range would have been more likely, but that's not from any firm knowledge.

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[MacGyverS2000]

I know, IR, I was simply mentioning my curiosity in what happens here due to my interest in creating my own fiber laser.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[VEBill]

The typical consumer xenon strobe light uses a higher strike voltage (18kV sounds familiar for the ol' Radio Shack versions and related parts), but the trigger pulse is typically applied to a third terminal in the middle of the tube (thus a different design than this case).

 

[controlsdude]

"That arc suppressor is what flashes every 4 seconds. It looks like a tiny flash bulb. Flash....flash....flash..."

We called this a spark gap


back in my navy radar days
I had one of these in a radar high voltage modulator, if there was an overvoltage condition in the modulator, this spark gap would fire to prevent the modulator from over charging. There was a thyristor tube that did the firing of the inductor/cap tuned circuit, which gave you a pulsed output.

But this spark gap was for safety in this modulator, the thyristor tube did the actually firing.
But after it fired it always had the chance of firing at a lower voltage and kept firing like your seeing. After we fixed the problem with the modulator, we had to replace the spark gap.

Now there are modulators that use a spark gap tube to fire the modulator to give a pulse to a either TWT (traveling wave tube, CFA (cross field amp) type device. These were lower power type modulators.

Maybe the problem is that the spark gap is firing at too low a voltage to fire the tube? Is there a heater plate attached to this spark gap? Maybe there is a resistor adjustment to heat the spark gap or decrease the heat? The more modern ones use solid state type tubes (scrs back to back)

I would look this up on the web
Spark Gap Modulators

 

[Compositepro]

All the flash tubes that I have seen have a third "trigger" electrode on the outside of the tube. Sometimes this is just a fine gauge wire wrapped around the tube. Perhaps yours is missing. You could try wrapping a grounded wire around the tube that comes near the high voltage electrode. This concentrates the electric field to help initiate ionization of the gas in the tube.

 

[itsmoked]

controlsdude:
I'll look up, "Spark Gap Modulators"
And yes! That thing is a "spark gap".

After some thought, if that spark gap is truly the trigger for ignition it's a "shorting device" not an inductor's, "opening device".

It would seem to me that this would mean things are different. Perhaps the inductor actually is a trigger transformer and the spark gap is dumping some cap's charge thru the primary of that upon sparking.

But alas gentlemen this has all been put on the back burner because they just informed me that, "it seems to work now". I requested from my contact pictures for us to analyze. When he went to the laser room to get them he discovered it was re-assembled. He hunted down the laser guy who said, "it just started working!". So I'm off the task for the immediate future.

No doubt I will get the panicky plea for help eventually - when their demand for that tool is at some sort of peak.

Stay tuned.

Thanks a bunch everyone.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[controlsdude]

Thinking back on the problems I had with a modulator and your problem.

Usually if the switch was not firing correctly.
Adjustment on heater for the grid circuit on the switch or spark gap.

Tube could have failed, this is usually the case if it continuously fires or does not fire at all. If it fires all the time you get an overload (continous short) or does not fire at all.

Since its high voltage sometimes the wiring would fail, symptons was a heavy ozone smell, arcing (lightning), etc....

But if there was a mismatch in impedence between the cap/inductor (pfn - pulse forming network) and the amplifier (your ionized tube laser) you would see symptons where it fires irregular. Only way we could find this problem was put a resistor bank in place of the amplifier. Of course the resistor bank was massive and had a blower on it for the heat.

Never had the pfn go bad, but if it did, I heard the container for the pfn would usually kind of explode or expand due to heating.

 

[MacGyverS2000]

Since we're talking about lasers working/not working, and I have no problems side-tracking a thread... ;-)

I recently lost use of my system for several weeks. It would randomly stop firing in the middle of a job, so some lines would cut, others not... I would get cuts around an object that looked like a dashed line. Not happy.

Turned out to be EMI issues, of all things. It seems the Y-axis motor controller card will, at some random time (the system has been working fine for several years) decide to throw off enough EMI to throw the "fire" signal into a conniption. My guess is a reg. cap was drying out. Luckily the company was aware of this being a possible issue as it came as a huge surprise to me when they suggested sending out an updated control board. Lo and behold, I swapped out the board and the laser was hunky dory. The new board was exactly the same, layout and components, except a choke had been added to each line heading to the motor.

Saved me from having to send the cartridge back for a rebuild, which isn't exactly chump change. Sometimes it's the oddest stuff...

Dan - Owner

http://www.Hi-TecDesigns.com

 

[itsmoked]

Thanks Dude.

Interesting Dan.

Reminds me I have a 1W blue diode laser I want to run on my CNC router. Just need some time..

Keith Cress
kcress -

http://www.flaminsystems.com

 

[zzzjim]

When high impulse currents start flowing , we are entering into into the Marconi age .
Due to the RF oscillation nature of these events , Stray capacitances
can play a part , Earths near wiring either added or lost .

( Have seen external light conditions change the firing voltage in an non laser gas tube system , light makes it lower )

I was wondering if testing with some covers off ,changed the capacitances and let some RFI energy out !

I also like controlsdude's suggestion of Loss in wiring .

The thrill is in the chase !

 

[itsmoked]

zzz; The chase. I agree!

I also suspect you might be correct. I believe the original tubes may have been bad. They put in new ones and tried it with the covers off. The tubes fired occasionally and seemed to fire during attempts to measure the voltage across them,(a capacitance change). At some point they fired consecutively a couple of times. They then put the ton of sheetmetal back on and bingo, it starts reliably just as you suggest.

I've seen a lot of fluorescent lights that wouldn't light if the sheetmetal below them was missing. Same sort of thing.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[ScottI2R]

Hi Keith,
First and foremost, I would go with another set of flashlamps. I have personally witnessed bad "lots" from mfrs. Resulting in multiple lamp changes at a pm. Most of my systems had a seperate PFN for each lamp (2). That generally made for easy troubleshooting as each fault was flagged per bank. I take it your trigger system is up close to the lamps. Our trigger xfmr was right outside of the pumping chamber. Potted cube with a single HV lead that connected in the center of the chamber. On the inside, under the bottom gold plated reflector was a stainless steel plate that ran almost the length of the lamps. This assisted breakdown ionization of the gas, at which point the lamps would (should) simmer at a low current of around 90-100v and 1 amp on our systems. You should be able to hear the trigger circuit firing in the chamber, not just at your spark gaps.

Things to check: Like I stated, lamps. Ensure all of your connections and HV cable in and around the pumping chamber.INNCLUDING the integrity of the insulation. We did have one laser that was an old Raytheon SS500 ( The SS stood for sprinkler system ) that had two lamps in series. Aside from always having water leaks, this thing would forever arc thru the end caps of the flowtubes that house the lamps, causing my preionization trigger pulse to go to ground ensuring my lamps would never simmer. Boy, I really do not miss working on that old dung heap.

Since Dan mentioned EMI, I had a strange intermittant problem with one of these lasers a couple of years ago. When you would first start the HV, PFN voltage would rise, then the trigger circuit would fire. Enter the problem: With every pulse of the trigger, the lasers safety shutter would open. At this point the uPC would see an incorrect shutter state and fault and shutdown. I was able to see the shutter open trigger pulse coming from the micro with every trigger pulse on a scope. Darn transients! Let me end this with just saying, I was never able to resove this before it too, mysteriously went away. Our work around was to turn off the shutter enable key, thereby eliminating power to the shutter control pcb until the HV started.

Just wondering, what mfr and model is yours?

...and good luck with the intermittants. They keep life interesting.

Scott

I really am a good egg, I'm just a little scrambled!

 

[itsmoked]

Thanks for your experiences! They're filling in a lot of questions and giving me lots more details to check out, if/when they call me back.


As for type, this is a ROFIN SINAR. They can sell you parts but they say, "they have no one who works there who understands the ancient lasers....wanna buy a new one?"

Keith Cress
kcress -

http://www.flaminsystems.com

 

[ScottI2R]

Wow, I was a FSE for them from 92-96. Mostly the transverse flow CO2 lasers.

Scott

I really am a good egg, I'm just a little scrambled!

 

[itsmoked]

As far as I can tell they don't even pedal a YAG anymore.

Keith Cress
kcress -

http://www.flaminsystems.com