Switching HV with IGBT's? / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Switching HV with IGBT's?

previous 1 2 3 next

Move Thread

LAN_403

dan

Wed Mar 22 2006, 05:08PM

Registered Member #223 Joined: Mon Feb 20 2006, 06:42PM
Location: Ottawa, Ontario, Canada
Posts: 125

The problem with using a variac and a NST is that you need to have the variac at full so the NST can strike the laser tube. Then bring the variac to the power level you want. If you are using a rectified NST to power the laser you could also just use a light dimmer to vary the power to the to the NST. To strike the laser at the desired power level you may need a 'start' button, which will temporally allow mains to bypass the variac or dimmer. A simple circuit could be made to do this automatically.

codesuidae

Wed Mar 22 2006, 05:16PM

Registered Member #337 Joined: Wed Mar 22 2006, 02:52AM
Location:
Posts: 5

Part of the problem with reducing the voltage using a variac is that some materials require a high peak power to be marked, but lower average power, so its necessary to use lots of fast, short high power bursts.

Thats why I am looking at controlling the rectified NST output, it just seemed like the most direct route to getting that kind control.

Certainly controlling input to the NST would be easier, I just don't know yet if it will give me the ability to generate the control that I need to run the laser this way.

dan

Wed Mar 22 2006, 05:22PM

Registered Member #223 Joined: Mon Feb 20 2006, 06:42PM
Location: Ottawa, Ontario, Canada
Posts: 125

Why not use a triac or IGBT and some control circuitry to control the duty cycle of the NST?

codesuidae

Wed Mar 22 2006, 05:47PM

Registered Member #337 Joined: Wed Mar 22 2006, 02:52AM
Location:
Posts: 5

I'm concerned about the limit of the PWM freqency I can use if I control power to the NST primary.

If I need a 1kHz PWM output from the laser how would this affect the NST? Can I feed the primary with a 120v 1kHz 10% PWM signal (for example) and expect a reasonable output? What about higher frequencies, say, 25kHz?

Steve Conner

Wed Mar 22 2006, 06:09PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

The NST won't work well if at all, with frequencies above 60Hz. It won't do sh*t at 1kHz or 25kHz. So to do PWM on the input side, you'd need to run the laser directly off the unrectified AC output, and chop out cycles of 60Hz with a triac. I guess 60Hz is a low enough frequency that it would punch a string of holes when the head was moving fast, instead of cutting smoothly.

Phase angle control (like a regular lamp dimmer) might work better, but you can't turn it down that far before the peak power starts to get affected, and the power's still going to be pulsing at 120Hz, unless you rectify and filter the NST output, but then you'll not get the peak power.

You could investigate those high frequency electronic NSTs, they could maybe be modified for PWM control.

EastVoltResearch used to have a natty circuit on his site for a high voltage series MOSFET string, but I can't find it any more. Pulsing a directly excited CO2 laser tube accurately to do marking and cutting is really pretty much the same problem as modulating a magnetron or TWT, so I guess his approach is relevant.

The words "cheap" and "CNC laser cutter" don't belong in the same sentence hee hee grin

Marko

Wed Mar 22 2006, 07:46PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Maybe some huge, HF vacuum tube could do, primitive but today they are still used in high power transmitters.
And there would be trouble finding one too.

You could investigate those high frequency electronic NSTs, they could maybe be modified for PWM control.

I doubt they are powerful enough, its like a small flyback with driver in the same box ready to be hooked to mains (since neon signs don't require kilowatts but just few hundred watts at most)

For pulse transformer, as I said, problem is finding the appropriate core.
High-frequency solid state MOTs can be boosted to few kW and it might be enough.

If not paralell few cores and get a monster power supply like thhis one

cheers...

vasil

Wed Mar 22 2006, 08:08PM

Registered Member #229 Joined: Tue Feb 21 2006, 07:33PM
Location: Romania
Posts: 506

Do you think that the laser tube will last long with these power input variations? The discharge parameters have to be mostly constant. Generally , CO2 lasers, have CW emission, and you modulate the light beam outside the tube. You can deflect a percent of the beam, to get a few power ranges (40%, 60%...etc).

codesuidae

Wed Mar 22 2006, 08:50PM

Registered Member #337 Joined: Wed Mar 22 2006, 02:52AM
Location:
Posts: 5

Steve Conner wrote ...

You could investigate those high frequency electronic NSTs, they could maybe be modified for PWM control.

I should mention that I'm looking to run a laser of no more than 50W, and half that would probably 'cut it', so to speak. I'll most likely be cutting things like acrylic and fabric. I don't really have any reason to cut sheet metal, but I suppose I'll think of something if I have that capability.

Steve Conner wrote ...

The words "cheap" and "CNC laser cutter" don't belong in the same sentence hee hee grin

The motion control parts are no problem and classify as cheap for what I'm doing, its just the laser bits I'm concerned about. I figure I could pick up a sealed unit with everything done for me, but that wouldn't be as much fun and I wouldn't learn much, so I'm hoping to get into it a bit deeper.

vasil wrote ...

Do you think that the laser tube will last long with these power input variations?

I have no idea, I've seen others refering to controlling CO2 laser power with HV PWM, and I've seen that sealed units use logic level PWM for power control as well. I don't know enough about CO2 lasers to know what is appropriate.

vasil wrote ...

The discharge parameters have to be mostly constant. Generally , CO2 lasers, have CW emission, and you modulate the light beam outside the tube. You can deflect a percent of the beam, to get a few power ranges (40%, 60%...etc).

I was curious if that could be done, but I hadn't yet run across any information about it. Considering the relatively low power level I'm interested in I think that this would be a good option, and probably much simpler to impliment. I'll keep it in mind.

vasil

Wed Mar 22 2006, 09:08PM

Registered Member #229 Joined: Tue Feb 21 2006, 07:33PM
Location: Romania
Posts: 506

I am thinking for example at a ceramic/graphyte disc, with holes on the border (like a Nipkov wheel), that will allow / obturate the beam with a frequency related with the speed rotation. The stuff will behave like a "beam power dimmer", adjusting the energy relised in the focal point.

Hazmatt_(The Underdog)

Wed Mar 22 2006, 10:14PM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

Why not just use a really beefy Photoflash circuit. Easy design, you can strobe them really fast with a large enough capacitor, voltages are managable ~450V topps, It just seems like a good direction.

And you can make your own pulse transfromer with a ferrite rod. Just get the old pulse transformer, inspect it and duplicate its design concepts, then implement that in your own pulse transformer. Its just that easy!

Mots and NST's are likely to damage the tube with their high average power.

previous 1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob