Co2 Laser PSU ideas needed / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Co2 Laser PSU ideas needed

Move Thread

LAN_403

cweeks

Tue Nov 20 2007, 12:36PM

Registered Member #497 Joined: Sun Nov 19 2006, 05:58PM
Location:
Posts: 3

Hi all,

For a little while now I have fancied having a crack at making a CO2 laser power supply, I have a commercial one and it appears to use two transformers in series that look very like TV flyback transformers, easy eh. I know "why make one if you have one", it just interests me, and as I can't find any designs for CO2 laser PSUs on the net that aren't based on NSTs I thought it would be good to start one.

The stumbling block for me is the design needs current and possibly voltage control. There are a lot of circuits out there for driving flybacks but none with current control. I was going to copy the UC3860 based design used by Project Thor

but it appears the UC3860 is now discontinued.

Anyway it needs a striking voltage of around 35Kv, and an operating voltage of 20Kv @ 30mA max. The current needs to be regulated and controlable. I am not sure about voltage control, only reason I can see for it is to stop the striking voltage being too high, but I can't see the laser tube minding a few extra Kv, only other benefit I can see is protecting the transformers from arcing and destroying themselves. So probably not essential.

I was thinking about a ZVS design with a series resonant load, but how do I regulate the current. I guess current can be measured from voltage drop across a resistor in the ground return line. Am I correct in thinking that output chopping would be a reasonable way of controlling it ? It looks like that is what the UC3860 in Project Thor did. There are other alternatives I have found to the UC3860 but none that when restarting the output after chopping synchronize it back into the ZVS cycle. Or does this not matter?

I am obviously not much of an expert and my experience of the finer details of such power supplies is nil, just what I have read on the net, I have made flyback drivers before but nothing with this level of control. Any comment greatly received.

Cam.

uzzors2k

Wed Nov 21 2007, 06:20PM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

Regulating current and voltage simultaneously is impossible so I'll assume you need constant current. With the specs you've supplied, you'll need to make a constant-current source with a max output voltage >35 kV. Once the tube strikes at around 35kV the resistance will drop considerably, so to keep the current at 20 mA the voltage will drop too, probably to 20kV.

I don't know anything about driving CO2 laser tubes, so I'll just assume you can use DC. That would allow you to use a current-sense resistor on the tube for direct current measurement. For power control a TL494 would work great. The electronics aren't too complicated, the hardest part will be the transformer. If you choose to use a half or full-bridge you'll need quite a few turns. Flyback topology would maybe be easier.

cweeks

Fri Nov 23 2007, 02:01PM

Registered Member #497 Joined: Sun Nov 19 2006, 05:58PM
Location:
Posts: 3

I wasn't meaning voltage and current regulating at the same time, I realize they a mutually exclusive. I meant (possibly) voltage limiting and current control. So if using the TL494 in a ZVS configuration and it's generating the required wave forms as so:

What will the TL494 do to the wave forms to regulate the output ?

uzzors2k

Fri Nov 23 2007, 04:43PM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

With current control no voltage control will be necessary. The rating 20kV 20mA implies that with 20mA flowing through the tube, the voltage across the tube will be 20kV. And with 20kV over the tube the current will be 20mA. So you can choose between voltage and current regulation, but current regulation would give you a free tube igniter. (during start-up the current would be 0, so the pulse width would be max, giving max voltage out)

ZVS with a TL494 is a bit over my skill level, but implies you are driving a load at resonance to get a sinusoidal voltage waveform. No matter what drive scheme you use, as long as you use output feedback regulation the TL494 will adjust the pulse width to keep the output constant. (PWM)

The required waveforms you show are for a SLR converter, is that what you plan on making? It's a much more complex converter than a standard forward or flyback converter. I'm not sure if it's even possible to regulate the output of a SLR converter, as they need 50% dead-time to remain in SLR mode which makes PWM difficult. I might be wrong though.

If I were to make a 35kV, 20mA power supply I would go with a forward converter (full-bridge). It will keep the TL494 design nice and easy. You will need a large ferrite core for all of the turns however, and HV diodes and capacitors for the output. The transformer should have a turns ratio for 40kV to ensure enough voltage and current to ignite the tube. Of course there may be easier ways, I'm in no way a professional.

Steve Ward

Fri Nov 23 2007, 08:57PM

Registered Member #146 Joined: Sun Feb 12 2006, 04:21AM
Location: Austin Tx
Posts: 1055

So you can choose between voltage and current regulation

Voltage AND current limiting is done often. You would want voltage limiting just in case your load doesnt act as expected (simple case would be if you lost connection to it).

You could always implement voltage limiting with a comparator that turns off the transformer driver temporarily. An over voltage condition is likely something that you would want the power supply to shut down for anyway. Otherwise, if you actually want voltage and current control, you need to use both error amps in the 494, one wired for your voltage the other for the current. As long as you dont need really tight regulation specs, then this is easily done.

To regulate an SLR converter you need to change the spacing between pulses (keeping the pulse widths the same!), with 50% duty cycle being the "max". One issue to watch for is hitting another resonant point. When i sweep the current control knob on my SLR converter, there are a few points where the output voltage really shoots up high. A closed loop system should naturally avoid these points unless it happens at the lowest possible frequency, in which case the feedback cant correct for it, and the power supply should just shut itself down. Certainly implementing voltage control for an SLR converter is no easy task, though current control isnt too bad.

Dalus

Fri Nov 23 2007, 10:10PM

Registered Member #639 Joined: Wed Apr 11 2007, 09:09PM
Location: The Netherlands, Herkenbosch
Posts: 512

Always use current and voltage regulation on high power lasers PSU's. I've seen a high power argon laser PSU's that developed a bad connection on a fuse-holder. The fuse holder and cable was reduced to a big drop of copper. Why, because the PSU kept ramping the voltage up to achieve a constant current.

cweeks

Mon Nov 26 2007, 07:45PM

Registered Member #497 Joined: Sun Nov 19 2006, 05:58PM
Location:
Posts: 3

Just what I was after knowing, thanks everyone. I might have a crack at an SLR based design and let you know how it goes. Although I was trying to avoid it I think I will have a poke around in my commercial PSU and see how they have it wired.

Cam.

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