# How to amplify function generator's output to several kV

Discussion in 'High Voltage' started by SaraJackson, Nov 25, 2011.

1. ### SaraJacksonNew Member

Joined:
Mar 8, 2011
Messages:
6
Likes:
0
Hi all,

I am gonna need a 2 to 3 kV AC, 1-100HZ voltage signal. It will be connected to an open circuit to make an electric field for my experiment.

I have a function generator with voltage output in the order of mV.
Any one knows how can I amplify that to several K?

Thanks,
Sara

Joined:
Dec 23, 2009
Messages:
6,777
Likes:
281
Location:
Cleveland, Ohio USA
Hi Sara

My first thought would be to find an old neon sign transformer and just drive it with a variac or find a high voltage transformer out of an old Hi Pot test set and drive it with a variac. Unfortunately getting a sine wave with the frequency range you want will not be easy at 2 or 3 kV. My suggestions would limit you to mains power frequency.

Ron

3. ### alec_tWell-Known MemberMost Helpful Member

Joined:
Jul 10, 2011
Messages:
9,254
Likes:
1,218
Location:
Cardiff, Wales
If a non-sinusoidal waveform (e.g. sawtooth) would suffice you could perhaps charge a capacitor from a 3kV DC source then discharge it and repeat the process at the desired frequency.

Joined:
Jan 12, 1997
Messages:
-
Likes:
0

5. ### duffyNew Member

Joined:
Dec 22, 2008
Messages:
2,618
Likes:
73
Location:
nortern 'sconsin

This isn't a simple problem.

First, you aren't going to be able to go directly from a few mV out on a signal generator to 2-3kV. You are going to need an amp to boost it into the tens or hundreds of volts, and then a transformer to get it the rest of the way into the kV range.

Second, if you expect it to generate kilovolt potentials at one Hz, you are really talking about a DC field generator. This would involve something like a Cockroft-Walton circuit, which unfortunately does not reverse its polarity.

Joined:
Jun 4, 2009
Messages:
152
Likes:
11
ONLINE
Look into modifying a camera-flash circuit (200 or so volts, I think) with a Cockroft-Walton multiplier, followed by a pull-down circuit. I've seen it done, giving 1000-1500 volts at microamps. Am not allowed to show the circuit, though: not mine.

Joined:
Oct 30, 2010
Messages:
9,916
Likes:
1,097

Joined:
Jun 4, 2009
Messages:
152
Likes:
11
ONLINE
9. ### MikebitsWell-Known Member

Joined:
May 24, 2008
Messages:
6,214
Likes:
174
Location:
San Diego, Ca
ONLINE
Are you sure about that function gen only outputting millivolts? Does not sound right as most function gens can output magnitudes greater than millivolts. You need to have the real numbers for your function gen before you proceed. Is this important? You bet, if your function gen outputs 1v into a 1000:1 step up system, you get 1000 volts give or take I2R losses. Now if it is only 1 millivolt from your generator, you only get a 1 volt output.

Last edited: Nov 26, 2011
10. ### saturationMember

Joined:
Dec 21, 2008
Messages:
45
Likes:
0
Location:
Kwajalein Atoll
Just on the top of my head, put an op amp in front of your FG to shield against your circuit. use this op amp to also adjust the output gain without having to load the function generator. Use 1000:1 or greater transformer to boost and isolate it from the low voltage circuit.

http://www.amazing1.com/transformers.htm

11. ### tcmtechWell-Known MemberMost Helpful Member

Joined:
Feb 22, 2009
Messages:
6,651
Likes:
429
Location:
Sawyer, North Dakota, USA
One less common approach would be to use good old vacuum tube tech to drive simple class A vacuum voltage amplifier circuit. 2 - 3 KV at up to few tens of milliamps is well within the working range of many common triode or similar tubes.

Last edited: Nov 28, 2011
12. ### duffyNew Member

Joined:
Dec 22, 2008
Messages:
2,618
Likes:
73
Location:
nortern 'sconsin
I like tcmtech's idea, something like an 813 beam power tube can handle a plate voltage well over 2kV, and will do it at frequencies right down to the 1Hz you want. They are still available new from Russia, for under \$30.

Or you could use do the same thing with a high-voltage mosfet, like the IXTV03N400S (4kV, about \$20). http://ixdev.ixys.com/DataSheet/DS100214-%28IXTH_V03N400_S%29.pdf

Or an IGBT, like the IXBT12N300 (3kV, \$26). http://www.electro-tech-online.com/custompdfs/2011/11/DS10012028IXBH-T12N30029.pdf

In all these cases, you supply your class-A amp with that Cockroft-Walton multiplier.

Last edited: Nov 28, 2011
13. ### tcmtechWell-Known MemberMost Helpful Member

Joined:
Feb 22, 2009
Messages:
6,651
Likes:
429
Location:
Sawyer, North Dakota, USA
Or a small microwave oven power transformer. They usualy work in that 2 - 3 KV range once rectified and filtered and just use a small variac to tweek the output voltage down if its too high.

Joined:
Dec 23, 2009
Messages:
6,777
Likes:
281
Location:
Cleveland, Ohio USA
The biggest problem I see is the low end frequency of 1 Hz. Ten Hertz maybe but 1 Hz?

Ron

15. ### unclejed613Well-Known Member

Joined:
Apr 3, 2009
Messages:
1,828
Likes:
99
so here's an amplifier you might find useful. you will need the +/-800V supplies, but it will do exactly what you are trying to do... the transistors are cascaded in series, so that no output transistor has more than a fraction of it's Vce-max across it.

#### Attached Files:

• ###### hv-amp.jpg
File size:
83.8 KB
Views:
1,026
Last edited: Nov 28, 2011
16. ### duffyNew Member

Joined:
Dec 22, 2008
Messages:
2,618
Likes:
73
Location:
nortern 'sconsin
Yes. If you remove it from an oven, be sure to shake all the microwaves out of it.

17. ### jpanhaltWell-Known MemberMost Helpful Member

Joined:
Jun 21, 2006
Messages:
5,896
Likes:
502
Location:
Cleveland, OH, USA
ONLINE
18. ### geraldfryjrMember

Joined:
Sep 19, 2011
Messages:
76
Likes:
3
I am in the finishing stages of a power supply that can do what you require.
It is variable from 0V to 13.25Kv and can be modulated as I have tried modulating it.
It takes 28 stages of voltage multiplication to get that high and you wont need but 6 or 8 stages for your voltage range.

I have modulated it up about 70Hz and the removal of a few filter caps should allow me to go much higher.
It will definitely do 10Hz as well as sub 1Hz for that matter.
Once I get the voltage feedback regulation circuit done I will be able to modulate it to its full voltage range.
I have it fully documented here,

http://www.diyaudio.com/forums/plan...tor-insulation-mylar-coating.html#post2531218

So have a look at maybe it will give you a good idea of how to go about this.
I have a nicer and much simpler version in mind as well and is mentioned in the thread and I will be working on that version soon.
Besides what I have designed it to do,the fact that it can be modulated opens up a whole new world IMO.

If you have any questions about it I will be happy to explain as it is a pretty straight forward and classic design.

enjoy !!

jer

P.S. Please read the whole thread as the link was just the basic schematic and when it is finished I will present the complete as finished schematic.

Last edited: Nov 29, 2011
19. ### geraldfryjrMember

Joined:
Sep 19, 2011
Messages:
76
Likes:
3
I have finally finished this supply and it is working flawlessly.
It is regulated,variable and is extermely stable.
The complete and final schematic with details have been post at the thread to the previous posted link.

It can be modulated but I have not optimised it yet for such use so its upper frequency of modulation is quite limited.
The highest frequency I had got without trying was around 70hz.

With the removal of a couple of capacitors and a buffer on the optional input I am sure one could get a very good modulation range out of it.
It tracks linearly at a rate of 1Kv out per .5v of control voltage in.
It is slightly non linear above 13kv as 13.8kv is its maximum output voltage,as well as below 1kv it is only very slightly non linear.
This is acceptable as it is extremely stable and linear between these two ranges.

I figured that I would post this update for any one interested in such a device as it is a very simple but versatile circuit.

jer

Last edited: Jan 7, 2012