Modulated HV power supply (ideas needed)

[LithiumOverdosE]

I'm asking for a bit of brainstorming of the honourable members of Sonsivri.  

I have two insulated HV plate electrodes which are place into ionic solution (FeCl3).
The goal is to observe the action of the ions due to electrostatic action and it works fine.

That HV DC signal (30 kV) have to be modulated with the signal whose amplitude should be 1000 V pk-pk (alternatively 750 V) and with frequency range of 1 Hz to 40 kHz.
For the experiments I use a simple Villard voltage mulitplier and it performs just fine with very little ripple (<5V).  
The capacity of the electrodes is <50 pF.


Originally I thought of using a custom made HV audio transformer and modulate output directly.
However, very high voltages would require tremendously big and complex transformer due to insulation limitations.

The second idea was to introduce one large DC cap and directly drive the output with the appropriate audio transformer in place.
The concept immediately fell apart because the low end of required spectrum would require very big cap with appropriate HV rating.

At this point I'm out of idea and any help would be appreciated.  

[DreamCat]

I think the flyback transformer from old big CRT TV is good choice, or you can custom one.
Here is the structure I recommand to you.

[LithiumOverdosE]

I considered the same approach but the capacity necessary for low frequency range would be quite large (too large for HV rating capacitors).

[optikon]

Transformers are such a natural choice for this task. Did you actually explore the size required? Seems to me, since the power is low, they might not be prohibitively large. I think there are ways transformers can be wound for high voltage so that wire insulation size is minimized thus resulting in a smallish design...

[DreamCat]

I considered the same approach but the capacity necessary for low frequency range would be quite large (too large for HV rating capacitors).

according the structor in my image, the main voltage no need very large capacity. it has fixed frequency, you can set it to 20k~30kHz, but this frequency is nothing to do with modulate voltage.  the capacity determine by its load.
you may note that the main voltage is DC voltage, the modulate voltage superimposed on a DC voltage, so the 1kv transformer no need very high insulation class(grade).

[LithiumOverdosE]

@optikon

Core size due to power is not the problem.
However, winding audio frequency transformer includes multiple, sandwiched layers which makes it larger and especially if the HV insulation is requirement (not to 30 kV though) so the required core is usually much bigger compared to narrow bandwidth transformer.

@DreamCat

You're correct. I missed the location of grounding.

This seems fine as a concept.
I have to do a bit of experimenting to see what happens in real-life.

[user77]

I'm thinking, and if you try to use a coil sparkle? you can modulate the frequency, and I think also the voltage maybe with a pwm, but the output hv normally is 12-15kv, with a hv  doubler  maybe it's ok to you ... it's just an idea ...

[LithiumOverdosE]

Did you mean ignition coil?

It would require non-commonly available HV parts and besides the ignition coil is limited in maximal frequency and introduce high inductance leakage.

I prefer working with ferrite cores and commonly available components.
This means I'll probably make custom made transformer and use resonant voltage multiplier.

[Ichan]

Brainstorming here, seems the requirement have some similarities with DC excited CO2 laser power supply.



Usually the output is around 20 - 30 KV DC, and has analog and pwm input to control the output - i do not know much about the detail.

-ichan.

[LabVIEWguru]

Lithium overdose:

http://www.repairfaq.org/sam/lasersam.htm

This is Sam's Laser Faq. I don't know if you've ever see it before, but it is a very, very large FAQ concerning lasers.

[kreutz]

I don't see any reference to the ionic current you use, but depending on that current (if it is a low current) you could use a high impedance 30 Kv dc generator (flyback with voltage multiplier +  series resistor is ok) and a variable parallel load (current modulated using a HV vacuum tube with grid).

[sadman]

hi friend LithiumOverdose

Please have a look to this may be help full

https://hackaday.io/project/1787-High-Voltage-Regulated-Power-Supply

http://www.hardhack.org.au/hv_reg_power

sadman

[hef4015]

I didn't follow all the links, but here comes a quick suggestion:
* make or buy a 30 kV DC power supply 
* use a transformer to transfer some energy to 30 kV potential 
* build a HV generator with 1 kV amplitude using MOSFETs, can be controlled by an arbitrary waveform generator, module connected to the 30 kV potential 

I've seen similar systems before. Just make sure you do not touch the high voltages... 

[LithiumOverdosE]

Please have a look to this may be help full

https://hackaday.io/project/1787-High-Voltage-Regulated-Power-Supply

http://www.hardhack.org.au/hv_reg_power

Thx sadman, but the voltage is quite a bit lower than required.

and a variable parallel load (current modulated using a HV vacuum tube with grid).

Parallel load at 30 kV?
Did you had any particular tube in mind?

* build a HV generator with 1 kV amplitude using MOSFETs, can be controlled by an arbitrary waveform generator, module connected to the 30 kV potential 

I've seen similar systems before. Just make sure you do not touch the high voltages... 

Could you please elaborate on this idea?
I didn't quite get it.

[borberk]

How much current draws your electrostatic electrodes at 30kV to be in doubt about value of capacitor in HV source? What is DC equivalent resistance of electrodes emerged in water solution of FeCl3?

[kreutz]

 

Parallel load at 30 kV?
 
Did you had any particular tube in mind?

for example: http://www.jogis-roehrenbude.de/Russian/GP5/GP5.pdf

Just search on the internet. H.V. Shunt regulators were frequently employed a few years ago, many were originally used for Color TV HV regulation, they are still used on some pulsed Xray generators (up to 150 kv).

kreutz

[kreutz]

I found this info, it might help depending on your requirements. It is an audio modulated flyback HV p.s. I suggest increasing the flyback switching frequency to at least 80 kHz ( 2x audio frequency bandwidth (=40khz as per your messages) as per Nyquist), it will be necessary to redesign the flyback transformer for this purpose.

http://boginjr.com/electronics/hv/flyback-driver-2/