Will the voltage continue to increase if we continue to connect these lockroft-walton voltage multipliers? What is the problem if the voltage does not accumulate at some point? In fact, I have already confirmed the results of the question.



I tried to make a 5 to 1kV DC/DC Converter. I made the same circuit as the picture and used a 5V/15mA supply. The part that i used in the Voltage Multiplier are 10nF/500V capacitors and 1kV diodes. But the output voltage was only 500V. So I added three more stages of multiplier. But the voltage has hardly accumulated since stage 5. It was not much different from the previous 500V output voltage.

Also i used a circuit below.

The right side of the transformer functions as an oscillator. Approximately 250Vac is output from the transformer pin 8 and 9. The output voltage of 750V came out in the 3rd stage of Multiplier. So I added two more stages of multiplier. However, the voltage did not accumulate from the third stage, and all the third, fourth, and fifth stages had similar voltages.



Why doesn't the voltage stack more than a certain level? How can I increase the voltage more and more?

 

The voltage will increase but there are diminishing returns that come from adding more stages. There is leakage and the available output current decreases with each added stage.

Here is a simulation with some useable results:

 

The voltage will increase but there are diminishing returns that come from adding more stages. There is leakage and the available output current decreases with each added stage.

Thank you, does it mean that the voltage does not accumulate more than a certain level because the power of the supply is insufficient??

 

As you add more stages the output impedance increases until leakage becomes significant. Also note that connecting a meter to measure the voltage will reduce the output voltage.

 

How much current will you draw from the output?

 

How much current will you draw from the output?

I want to draw about 350uA of current from the output.

 

As others have stated, there is a diminishing return with increases in stages in a classic Cockroft-Walton multiplier. This is because of a counterintuitive time sharing of the diodes that happens. If you capture the current waveform across each diode, you will see something unexpected. Each diode sees only a portion of each half cycle. Add enough stages and the time slice each diode gets, starts to approach the diode reverse recovery time (trr) which can be particularly long for HV diode junction stacks. Needless to say, when the time slice equals trr, there is no further gain to be had. That's not the only reason, but it is a major and often overlooked factor.