200 kV current limited SMPS / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

200 kV current limited SMPS

1 2 next

Move Thread

LAN_403

Dr. Dark Current

Thu Mar 31 2011, 10:00PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

Hi,
I was thinking about making a very high voltage power supply which could be used for drawing arcs. The first idea was a bridge inverter, two transformers with 50 kV output each (midpoint grounded) and a halfwave doubler on each transformer.
But then I realised this, when an arc forms on the output of the PSU, the doubler caps effectively get connected in parallel to the secondary windings. The dc voltage from the charged caps which shows up on the secondaries of the transformers could saturate the cores and blow the bridge.

Then I had this idea - what about making a SLR-type inverter but with resonant caps on the secondary side of the transformers? Because those caps would be much smaller than the "pure" doubler caps, core saturation should not occur when an arc forms. As the whole H.V. assembly would be put under transformer oil and the capacitances needed at xx kV are very small, I thought about making the caps just two plates under oil with adjustable distance between them.

What do you think about this concept? Would the oil dielectric H.V. caps stand to the kVAr's? Is such a power supply practically even doable in a "hobbyist environment"?

The schematic, just in case:

James

Thu Mar 31 2011, 10:44PM

Registered Member #3610 Joined: Thu Jan 13 2011, 03:29AM
Location: Seattle, WA
Posts: 506

What sort of output current did you have in mind? I think you would want more than just a single doubler stage as capacitors over 20kV start to get much more expensive. SMPS x-ray generators typically have 8 stages or so with a ~10kV output from the transformer. The heads are filled with mineral oil and that stands up fine. I've been thinking for a while of building a high power (several kW) ~120kVDC power supply using a similar topology, split or twin secondary windings feeding separate multipliers for + and -.

Dr. Dark Current

Fri Apr 01 2011, 07:28AM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

James, the short circuit output current I was thinking of would be somewhere around 20 mA.
As for the caps, like I said I would just use a simple plate capacitor under oil, the caps for the SLR topology (driving at f_res/2) could be very small at this voltage - somewhere in the few pF range.

Ash Small

Fri Apr 01 2011, 01:35PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Dr. Kilovolt wrote ...

.the capacitances needed at xx kV are very small, .

I can see a few potential problems here, but given that voltage makes no difference to the capacitance required (high frequency will allow the use of smaller caps), and the fact that it was already April 1st in the Czech Republic when this was posted.........(I also understand that multipliers/doublers aren't suitable for drawing arcs)

Dr. Dark Current

Fri Apr 01 2011, 01:56PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

Ash: Nope this is for real... You can draw arcs from a halfwave doubler on a high frequency transformer. The capacitance required would be a few pF.
However I think I'll ditch this idea... its starting to sound a bit unreal.

Ash Small

Fri Apr 01 2011, 02:15PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

What frequency do you estimate Fres to be? (ie what is the estimated inductance of the secondary)

Dr. Dark Current

Fri Apr 01 2011, 02:49PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

The transformer would be driven at some 40-60 kHz, which means the cap would need to resonate at 80-120 kHz. This also means the capacitance of the secondary should be so small that it would resonate with the leakage inductance of the transformer (+ additional series inductances if any) at more than 120 kHz, preferrably some 200 KHz... I'm afraid it's not real to design such a transformer with such output voltage.

Steve Conner

Fri Apr 01 2011, 03:28PM

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

It sounds like it would work better if you used a Tesla coil as the transformer.

Tesla's lab notes say that plate capacitors under oil can stand lots of HV/HF abuse.

James

Fri Apr 01 2011, 06:31PM

Registered Member #3610 Joined: Thu Jan 13 2011, 03:29AM
Location: Seattle, WA
Posts: 506

Well if it helps, I've attached the schematic I reverse engineered from a Kodak 2200 intraoral x-ray head. This is designed to produce selectable 60-70kV at 4-7mA. Operation is in the 85-95 kHz range, driven by a resonant converter IC. The transformer is a small ungapped ferrite about 30x50mm, capacitors are 15kV ceramic and diodes are 16kV. Relay selects the filament cup bias depending on whether 4mA or 7mA is selected. The transformer and multiplier are within the oil filled head while the inverter drive is in the wall mounted cabinet.

It should be relatively easy to scale this up to provide 20mA at 100kV, and then you just need to build a pair of them of opposing polarities.
]kodak_xray_generator.pdf[/file]

Dr. Dark Current

Fri Apr 01 2011, 06:59PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

James: Thanks, but the problem with multipliers is that you can't draw arcs from them without some kind of big HV resistors which waste power. I chose the single stage half wave double as you can still draw an arc from it when supplied from a high frequency transformer.

1 2 next

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