# Transformer voltage stress when feeding a Cockroft Walton Multiplier

#### SiCEngineer

Joined May 22, 2019
311
Hello,

I am designing an SMPS and wondered what the voltage stress would be on a transformer that was feeding a cockroft walton multiplier circuit? I have a high voltage output requirement but I want to have a transformer acting as galvanic isolation and achieve the high step-up using a cockroft walton circuit (or other voltage multiplier) instead, so that the voltage stress on the transformer is much smaller than the output voltage. Is this the case with the cockroft walton multiplier? If not, are there any other ways a low voltage stress can be imposed on an isolating transformer and the step up can be performed afterwards?

#### Sensacell

Joined Jun 19, 2012
2,890
One end of the input to a Cockroft Walton Multiplier is typically at ground potential, so the transformer would see no more voltage stress than the output voltage it produces.

#### SiCEngineer

Joined May 22, 2019
311
One end of the input to a Cockroft Walton Multiplier is typically at ground potential, so the transformer would see no more voltage stress than the output voltage it produces.

View attachment 180372
Ah, that's what I thought. However, that's not what I want. I want the transformer to see less than the final output voltage, because my final output voltage is 10kV and I'm trying to find a way to use a transformer of low secondary winding voltage. What about the parallel cockroft walton? Ideally I'd want the transformer voltage stress to be limited to the actual transformer voltage, if that is possible?

#### KeithWalker

Joined Jul 10, 2017
1,911
Ah, that's what I thought. However, that's not what I want. I want the transformer to see less than the final output voltage, because my final output voltage is 10kV and I'm trying to find a way to use a transformer of low secondary winding voltage. What about the parallel cockroft walton? Ideally I'd want the transformer voltage stress to be limited to the actual transformer voltage, if that is possible?
Read the answer from Sensacell again carefully. The transformer does not "see" the final output voltage, only the secondary voltage it produces.

#### SiCEngineer

Joined May 22, 2019
311
Read the answer from Sensacell again carefully. The transformer does not "see" the final output voltage, only the secondary voltage it produces.
So the voltage stress across my transformer will not be the final multiplied and rectified DC voltage, but simply the secondary transformer AC value? Say for example I had my secondary transformer winding at 300V, and then used a cockroft walton multiplier so that I had a final DC output of 900V, my transformer would only have to be rated for the 300V?

#### KeithWalker

Joined Jul 10, 2017
1,911
So the voltage stress across my transformer will not be the final multiplied and rectified DC voltage, but simply the secondary transformer AC value? Say for example I had my secondary transformer winding at 300V, and then used a cockroft walton multiplier so that I had a final DC output of 900V, my transformer would only have to be rated for the 300V?
Yes, as long as the secondary winding is not connected directly to a voltage that exceeds the maximum rated insulation breakdown voltage between it and the primary winding and frame of the transformer.
For example, if the frame is grounded and you connect the high voltage output from the multiplier circuit to ground, , then there would be 900V between the secondary winding and the frame.
If one end of the secondary is connected to ground, then there will only be 300V AC on the other end of the secondary with respect to the frame..

#### SiCEngineer

Joined May 22, 2019
311
Yes, as long as the secondary winding is not connected directly to a voltage that exceeds the maximum rated insulation breakdown voltage between it and the primary winding and frame of the transformer.
For example, if the frame is grounded and you connect the high voltage output from the multiplier circuit to ground, , then there would be 900V between the secondary winding and the frame.
If one end of the secondary is connected to ground, then there will only be 300V AC on the other end of the secondary with respect to the frame..
So if my load was connected this way, would the insulation breakdown voltage between it and the primary winding and frame of the transformer have to be rated at the high output voltage? Since I would be connecting my HV output from the multiplier circuit to ground? Because even in this situation, one end of the secondary is grounded.

#### Attachments

• 10.5 KB Views: 5

#### MisterBill2

Joined Jan 23, 2018
9,225
I suggest that a switch-mode power supply is not at all a good choice for driving such a voltage multiplier circuit because of the transformer waveform. You will have a much better chance of success with an inverter type of circuit tat will deliver a more symetric wave form. The multiplier circuit will work much better that way. Also, switching power supplies are rather complex circuits and I do not predict success in creating one. Copy an existing inverter circuit and it should work quite well.
And a question now is what will the use be for the high voltage? Will it need to be regulated? And how much current will the application require? The regulation of voltage multipliers is usually quite poor, so you do need to know that in advance.
And what sort of design engineer are you?

#### SiCEngineer

Joined May 22, 2019
311
I suggest that a switch-mode power supply is not at all a good choice for driving such a voltage multiplier circuit because of the transformer waveform. You will have a much better chance of success with an inverter type of circuit tat will deliver a more symetric wave form. The multiplier circuit will work much better that way. Also, switching power supplies are rather complex circuits and I do not predict success in creating one. Copy an existing inverter circuit and it should work quite well.
And a question now is what will the use be for the high voltage? Will it need to be regulated? And how much current will the application require? The regulation of voltage multipliers is usually quite poor, so you do need to know that in advance.
And what sort of design engineer are you?
I am a PhD student at the moment, so my experience is a little bit limited. At the moment for my high voltage supply I am probably going to use a full bridge LCC/parallel resonant tank to account for the high voltage transformer parasitics. I am just looking for other ways that I can deal with the issue of high voltage transformers, such as for example using a voltage multiplier circuit fed by the transformer, rather than using the transformer itself to achieve the step up. The voltage does need to be regulated, current is couple hundreds of amperes. I did read that the regulation is quite poor for series voltage multipliers but for parallel multipliers the regulation is much better.

Thanks

#### Sensacell

Joined Jun 19, 2012
2,890
COUPLE HUNDRED AMPERES!!!!!

(duck and cover)

#### Doros

Joined Dec 17, 2013
137
Sorry to intervene, you want 10kV with couple hundreds of amperes?

Is it correct? This is huge energy

#### SiCEngineer

Joined May 22, 2019
311
Huge energy in general, or for the multiplier circuit? I stated 10kV as I was simulating for that voltage - in reality it will be more like 5kV.

#### Doros

Joined Dec 17, 2013
137
Without being an expert I calculate V X I = 1.000.000 W, (1MW)

This is the power for the residential needs of a small town (not being an expert). Others in the forum know better than me

So it is huge in gneral anf for the multiplier.

#### SiCEngineer

Joined May 22, 2019
311
Without being an expert I calculate V X I = 1.000.000 W, (1MW)

This is the power for the residential needs of a small town (not being an expert). Others in the forum know better than me

So it is huge in gneral anf for the multiplier.
For V = 10kV and I = 300mA (lets say)
VI = 3000W.

What you have calculated seems to be the load resistance.

#### Doros

Joined Dec 17, 2013
137
You have mentioned "current is couple hundreds of amperes"

#### SiCEngineer

Joined May 22, 2019
311
You have mentioned "current is couple hundreds of amperes"
Oops, missing a small "m" in front of that

My bad guys, I meant milliamps. Couple hundred MILLIamps

#### Doros

Joined Dec 17, 2013
137
Then I think it is a lot for the multiplier. I haven't seen any arrangemment giving this current with HV.

I believe in the forum are people can help

#### KeithWalker

Joined Jul 10, 2017
1,911
I did not realize that you planned on generating such a large amount of dangerous power. The only recommendation I will make, considering your lack of experience, is to purchase a ready made and tested supply. Any other approach has the potential of resulting in devastating and maybe even lethal failure.

#### BobTPH

Joined Jun 5, 2013
3,649
A 3 KW 5 KV supply is not something for an inexperienced person to design.

If you go ahead with this, keep you hands in your you pockets and wear safety goggles when you turn it on!

Bob

#### MisterBill2

Joined Jan 23, 2018
9,225
A 3 KW 5 KV supply is not something for an inexperienced person to design.

If you go ahead with this, keep you hands in your you pockets and wear safety goggles when you turn it on!

Bob
a supply like that requires a whole lot of safety protocols indeed. While it will not evaporate the wrench out of your hand that level of voltage does demand a lot of extra effort to keep things safe, since there are no minor shocks at that voltage level.