1.2 MV Marx Generator / Projects / Forums

Forums

4hv.org :: Forums :: Projects

« Previous topic | Next topic »

1.2 MV Marx Generator

1 2 3 4 next

Move Thread

LAN_403

Dr. ISOTOP

Tue Dec 21 2010, 11:45PM

Registered Member #2919 Joined: Fri Jun 11 2010, 06:30PM
Location: Cambridge, MA
Posts: 652

A continuation of the discussion in my "I have some capacitors..." thread.
I have a 16 6.66 nF high-voltage pulse capacitors, good for 200pps@50KV, a low rep rate at 80KV, and single-shot operation at 100KV, with polypropelene dielectric. The current plan is to build a 16-stage Marx generator charged to 80KV/stage using an x-ray transformer. Charging will be with inductors, since I can't find power resistors capable of withstanding 80KV.
Since I can't build anything right now (the caps are at college and I'm at home), I might as well plan this out nicely. Anything I should watch out for?
Should I make this a bipolar Marx instead?

mikeselectricstuff

Wed Dec 22 2010, 12:36AM

Registered Member #311 Joined: Sun Mar 12 2006, 08:28PM
Location:
Posts: 253

Bipolar has advantages in reducing voltage above ground, reducing corona issues, and also allowing use in a smaller space - less ground clearance required. 6.66nf at 80KV/stage is going to make an almighty bang...
I suspect resistors could be the biggest challenge.

Dr. ISOTOP

Wed Dec 22 2010, 01:24AM

Registered Member #2919 Joined: Fri Jun 11 2010, 06:30PM
Location: Cambridge, MA
Posts: 652

If I were to charge using inductors, what inductance should I aim for? From what I can find, it should be around 1 mH.

klugesmith

Wed Dec 22 2010, 03:50AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

bwang wrote ...
Charging will be with inductors, since I can't find power resistors capable of withstanding 80KV.
.... Anything I should watch out for?

Each inductor will need to withstand 80 kV between its terminals when the bank fires.

Pinky's Brain

Wed Dec 22 2010, 03:52AM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

I think it depends on what you are discharging into. AFAICS L should be significantly greater than discharge time squared divided by C ... 1 mH should do that for a spark.

It's also possible to build a Marx with diodes and one single inductor BTW. The inductor has to hold off the full output voltage though, and it sees a smaller capacitance on discharge (the series capacitance) so it needs a higher inductance too (so basically all the inductors you didn't put in between the stages end up strung together at the end).

Dr. Shark

Wed Dec 22 2010, 04:17AM

Registered Member #75 Joined: Thu Feb 09 2006, 09:30AM
Location: Montana, USA
Posts: 711

Nice project, and exactly what you should do with those capacitors! You might want to reconsider using resistors for charging though. I am not sure how you would figure out the correct inductance, Pinkys "discharge time squared divided by C" probably derives from Ï‰ = sqrt(1/LC) and should put you into the micro or even nano-henrys for any reasonably fast discharge. But if it's way up at 1mH, you are basically Tesla Coil secondary territory. Winding all these coils would be a major PITA. I also vaguely recall (and I thought Mike had tried this) that people here were having trouble with resonance between charging inductors and the capacitors, leading to voltage spikes that destroyed some of the capacitors. You may want to take precautions with safety gaps or start out at a much lower than rated voltage.

Dr. ISOTOP

Wed Dec 22 2010, 05:08AM

Registered Member #2919 Joined: Fri Jun 11 2010, 06:30PM
Location: Cambridge, MA
Posts: 652

Hmm...for a 1 microsecond discharge time, Pinky's formula gives me 0.1 mH, which seems to be much more reasonable to wind than 1 mH.

Dr. Shark

Thu Dec 23 2010, 01:10AM

Registered Member #75 Joined: Thu Feb 09 2006, 09:30AM
Location: Montana, USA
Posts: 711

If you want to get serious with the Marx and e.g. fry all your home electronics rather than just creating a loud bang, I think you would be looking at discharge time in the nanoseconds. Your capacitors should be fully capable of that, it's just a question of getting the geometry right. But even a conservative 100ÂµH could be as low as you would want to go since the coils need to be physically large enough to withstand 100kV across them.

ScotchTapeLord

Thu Dec 23 2010, 02:14AM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

Your neighbors are going to love this...

The pulse time will be governed by the esl of your capacitors in series and your layout...
You might need to take some measures to suppress corona.

And the inductors forming a resonant circuit with the capacitors could be a real problem, as someone mentioned. You may want to use your capacitors at a lower voltage to account for possible voltage excursions...

klugesmith

Thu Dec 23 2010, 02:38AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Dr. Shark wrote ...
100ÂµH could be as low as you would want to go since the coils need to be physically large enough to withstand 100kV across them.

For reference, here's a 50 ÂµH single-layer coil wound with AWG10 wire (during development of a one-off LISN).
It's ugly 'cause hot melt glue softened during a test at 50 amps.

1293071742 2099 FT104049 Lisn Coil 49 Cr

As for resonances: a SPICE simulation would be easy to set up and might be very enlightening.

1 2 3 4 next

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