High Voltage, Low Current, Low Frequency Voltage Multiplier Issue

Thread Starter

Terroman

Joined Aug 12, 2016
12
Hi,

I'm constructing a high voltage equipment to be used for electrostatic experiments. Thus I would like to have this source to be positive / negative and have the option to linearly control the output level.

I have managed to control the output linearly using a feature in a 20W power supply (6000V with 3.3mA max). With the intent of augmenting this output voltage through a voltage multiplier, I have created a stacked solid-state relay switching circuit in order to obtain a pulse train between 0V and 6kV at 30Hz. This switching also works correctly.

I have a problem with the voltage multiplier (I'm using ceramic 1nF capacitors and BY724 diodes). According to my simulations, the circuit should operate with double the input voltage but I am not getting a voltage increase. I'm also not obtaining the expecting rectification.

Could you please highlight any possible issues you can think of? I'm thinking that maybe I do not have enough current for the diodes - maybe I need to change the diodes with better ones?

Thanks for your replies!
 

Thread Starter

Terroman

Joined Aug 12, 2016
12
Hi Bertus,

Thank you for the list! However, I would like to try and make this circuit to work before I change the topology. I've consumed too many resources in this project to give up now - at least I would like to understand what exactly is happening and why, for my knowledge.

Please find the voltage multiplier schematic attached. I am using the "opposite_terminal" in order to be able to translate to a negative voltage (in this case, the input would be connected to "opposite_terminal" and the ground would be connected to the marked "output" terminal).

Voltage_Multiplier.JPG

Thank you for any insights!

Moderators note : shown image full size
 
Last edited by a moderator:

Standisher

Joined Jan 16, 2015
156
Might have nothing to do with this but the schematic you attach shows diodes in a different orientation from positive input than the standard voltage doubler (see attachment). The capacitors also seem to be very low values but perhaps that is intentional

standisher_doubler.jpg

Moderators note : adapted circuit to fit ToS
 
Last edited by a moderator:

Thread Starter

Terroman

Joined Aug 12, 2016
12
Yes exactly, thank you for you help! The circuit I'd shown is able to output +kV or -kV depending on where the input and ground are connected as stated above.

Update: After some tweaking with the timers of the microcontroller, I was able to obtain a frequency of 490Hz for the Solid-State Relays and switching still occurred correctly. Still, the increased frequency was not significant for voltage multiplication to occur. I have replicated the circuit on a circuit using a signal of 5V. I have observed very similar results when the capacitors were 1nF. However, when I increased them to 1uF, voltage multiplication occurred and obtained 8.2V (10V minus the potential barriers of the diodes).

Therefore, I'm now searching online for 1uF capacitors which have a voltage rating higher than 6kV. I have managed to find 0.1uF, 10kV capacitors, but I'm not sure if this 100nF would be enough. Last solution would be to try to construct them myself. However, how can voltage rating of a capacitor be measured once I construct a capacitor?

As always, thank you for your comments.
 

tcmtech

Joined Nov 4, 2013
2,867
Hi,

I'm constructing a high voltage equipment to be used for electrostatic experiments. Thus I would like to have this source to be positive / negative and have the option to linearly control the output level.

I have managed to control the output linearly using a feature in a 20W power supply (6000V with 3.3mA max). With the intent of augmenting this output voltage through a voltage multiplier, I have created a stacked solid-state relay switching circuit in order to obtain a pulse train between 0V and 6kV at 30Hz. This switching also works correctly.

I have a problem with the voltage multiplier (I'm using ceramic 1nF capacitors and BY724 diodes). According to my simulations, the circuit should operate with double the input voltage but I am not getting a voltage increase. I'm also not obtaining the expecting rectification.

Could you please highlight any possible issues you can think of? I'm thinking that maybe I do not have enough current for the diodes - maybe I need to change the diodes with better ones?

Thanks for your replies!
You would likely be time and money ahead to have a common neon sign transformer for your base power supply to drive the voltage multipliers instead of a solid state relay system.

I've played with that design many times and getting several tens s of KV DC output with minimal losses was pretty easy.
 

ian field

Joined Oct 27, 2012
6,536
Hi Bertus,

Thank you for the list! However, I would like to try and make this circuit to work before I change the topology. I've consumed too many resources in this project to give up now - at least I would like to understand what exactly is happening and why, for my knowledge.

Please find the voltage multiplier schematic attached. I am using the "opposite_terminal" in order to be able to translate to a negative voltage (in this case, the input would be connected to "opposite_terminal" and the ground would be connected to the marked "output" terminal).

View attachment 113960


Thank you for any insights!

Moderators note : shown image full size
You can get away with small capacitors if you use higher frequency - CRT colour TVs run at just over 15kHz, the CRT cathode current can be up to 100mA per gun. At that frequency; you need fast diodes - you might get away with 70nS, but 35 would be much better.

Very old TV flyback transformers had an 8kV overwind to drive a simple tripler that was a separate component.

Most black & white portables use a straight un-multiplied EHT transformer - somewhere between 6 - 12kV is typical. Moulded in diodes became popular, but older types had an external rectifier stick on an insulated cradle.

An old favourite CTV of mine was the Bush A823 - the multiplier was potted in RTV sealant, so it was easy to recover the individual components.
 
Top