25 kV amplifier frequency dependence / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

25 kV amplifier frequency dependence

previous 1 2

Move Thread

LAN_403

Physikfan

Thu Jul 21 2016, 12:27PM

Registered Member #60240 Joined: Mon May 16 2016, 07:01PM
Location:
Posts: 304

Hi Patrick and Hazmatt

Here are the results of measurements of the curvature dependence of the load of this 25 kV amplifier at 1 Hz.
Equipment TEK 6015A and TEK 7854 with 7A26, Y: 5 kV/cm.

Output voltage at bei 1 Hz without load (only 100 MOhm of the 6015A probe):

Leybold10kV1HznoLoadparallel100M400x364

Output voltage at bei 1 Hz with 60 MOhm load parallel to 100 MOhm of the 6015A probe:

Leybold10kV1HzLoad60Mparallel100M400x377

The signal at 60 MOhm is nearly the same compared to the signal without load.
I think I need more measurements with more different loads to see what is really going on.

Regards

Physikfan

Sulaiman

Thu Jul 21 2016, 04:59PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

Guessing based on the data from the advert linked to above,
Input 35 VA
Output 25 kv no-load, 0.5 mA short circuit, 200 mJ maximum energy disharge.

Calculations
E = 0.5 x C x V^2
0.2 = 0.5 x C x (25000)^2
so C = 640 pF maximum, the 'smoothing'/'reservoir' output capacitor
clearly this unit is designed for operation by careless people. (students)

there could be an active output current feedback system
in which case 25 kV at 0.5 mA would be maximum power output (12.5 W)
but my guess is that for safety a 50 MOhm resistor is between the reservoir capacitor and the output terminal
maximum power output would be 3.125 W into 50 MOhm

So,
with a short circuit the falling part of the output sinewave has a frequency response of 5 Hz,
2.5 Hz into a 50 MOhm load.

There is probably 100 MOhm or so across the output capacitor to pull down the output when no load and for safely/slowly discharging the output when off.
This resistor will give more bandwidth by helping the load to pull down the voltage aross the capacitor.

I would have to open it up and look ... how have you resisted ?

P.S. I would recommend close inspection of any newly encountered H.V. supply at least for safety reasons
(unplug first and leave for a day before opening, look, don't touch, unless you are a 4HVer)

Patrick

Thu Jul 21 2016, 08:17PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Im thinking the best solution is to get a tek probe, then build your CW with resistive ladder, then verify its response with the p6015. The rest of this is getting complicated, which may lead you a stray.

Physikfan

Thu Jul 21 2016, 10:19PM

Registered Member #60240 Joined: Mon May 16 2016, 07:01PM
Location:
Posts: 304

Hi Patrick and Sulaiman

Ad Patrick: " Im thinking the best solution is to get a tek probe, then build your CW with resistive ladder, then verify its response with the P6015."

This is exactly the direction I want to follow.

My first steps will be:

1. Getting experience on commercial DC probes with poor AC properties up to 200 Hz.
2. Building a pure ohmic voltage divider up to 200 kV
3. Bulding a HV source around 20 kV from DC to 1 kHz or 10 kHz
4. Comparing the AC behaviour of this 200kV probe with a TEK P 6015A
5. Try to do a frequency compensation at least up to 10 or 100 kHz simply with a variable capacitor parallel to the low ohmic part of the divider.

" The rest of this is getting complicated, which may lead you a stray"

I think you are completely right but I will try to do my best.

Ad Sulaiman:

Thank you very much for your calculations.
I need more time to answer with a hopefully meaningful comment.

Regards

Physikfan

Hazmatt_(The Underdog)

Fri Jul 22 2016, 02:24AM

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

#3 is very difficult indeed.

To copy your power supply but increase it's slew rate is what you are asking for, but with a control voltage of 5V to control 25,000V, is a ratio of 5000:1 voltage gain, it's a pretty tough task unless you use a vacuum tube.

For a solid state approach, DC to maybe 10Hz could be doable if you had the schematic of your supply, but it will be noisy. You will have to sacrifice current for an increase in the voltage swing. Why? The output is filtered and stabilized with capacitance, and you are changing the voltage across that capacitor quickly which requires current, the more the filtering you have and faster response you demand, the more the current will be required.

For 20Hz to 10KHz, a 500W audio amplifier could be used, with a soft iron audio matching transformer, but it too would have a high ratio, and maybe develop 15mA at 25,000 VAC. Of course this is somewhat of a challenge to insulate as well, it would need to be under oil or fully potted.
The feedback from a divider would be fed into the amp to ensure the regulation.

Still higher frequency can be achieved with RF toroids, but the PA's are temperamental and hate mismatches and wide load swings.

Physikfan

Fri Jul 22 2016, 05:47AM

Registered Member #60240 Joined: Mon May 16 2016, 07:01PM
Location:
Posts: 304

Hi Hazmatt, Patrick and Sulaiman

I will investigate some high voltage DC power supplies up to 50 kV which should allow a modulation for plasma loud speakers.
Even if the modulation will not be 100% a variable frequency sine modulated signal could check the AC performance of a HV probe with a P6015A as reference.
This is my next step to proceed.

Ad Hazmatt "For 20Hz to 10KHz, a 500W audio amplifier could be used, with a soft iron audio matching transformer".
Please, what do you mean with a soft audio matching transformer?

" Of course this is somewhat of a challenge to insulate as well, it would need to be under oil or fully potted".

For the HV final stage I could use also an induction coil, what do you think?

Regards

Physikfan

Sulaiman

Fri Jul 22 2016, 08:46PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

before it gets blocked, download this book

Physikfan

Sat Jul 23 2016, 05:30PM

Registered Member #60240 Joined: Mon May 16 2016, 07:01PM
Location:
Posts: 304

Hi

Here is a picture of the item and also the technical data of the Leybold 25 kV power supply 522 41:

Leybold25kV1400x342

Regards

Physikfan

previous 1 2

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