3
\$\begingroup\$

I have a microwave oven transformer. The cap that is with it is rated for 2100 volts so I assume the transformer has an output close to that. I know the diode and cap make a voltage doubler which should boost that to around 4200 volts.

I have 4 1300 volt 5 uf film caps in series which gives a voltage of 5200. I have charged these caps with a low powered hv power supply without problems. Output of the HV supply was well over 5200 volts. As long as I did not leave it on too long the caps were fine. I did burn up a few from too much voltage. I knew it might happen when I was pushing it.

The MOT will only charge the cap bank up to the output voltage so without the doubler I can charge to 2100 or 4200 with the doubler correct? I know the diode has to be used to create DC pulse current to charge.

Since the MOT does not have a current limiting feature, will this harm the cap bank? I believe caps act as direct shorts at first and then build resistance until they no longer take on a charge.

\$\endgroup\$
  • \$\begingroup\$ Do you know that what you do is very dangerous? The kind of question you ask makes me fear you don't. 1uF (=5uF/5) at 5kV is 125J, quite enough to kill you. \$\endgroup\$ – Vladimir Cravero Nov 2 '15 at 21:11
  • \$\begingroup\$ Yes I know it is very dangerous. \$\endgroup\$ – grantr Nov 3 '15 at 13:14
2
\$\begingroup\$

The MOT is sort of current limited because of the way it is constructed .To summarise this its got more leakage inductance than a normal transformer.This is also seen on battery chargers and welders ,its a very old tried and trusted technique.You dont know if the caps are charge and discharge proof and you dont know the fault current of the MOT.If you are popping caps you could experiment with placing a lamp ballast choke in series with your incoming mains .Lamp ballast chokes are easy to find and give reasonably predictable current limiting.If you cant find a lamp ballast choke you could do a half wave doubler which is also called a diode pump using a small HV cap on the AC side for current limit and placing your cap to be charged on the DC side.A ballpark starting value of this current limiting capacitance could be say 10% of the output capacitance .PS I hope you know what you are doing because you do have high volts and high currents and high fault energy .

\$\endgroup\$
  • \$\begingroup\$ I used a 15kv power supply running off of 2 aa batteries with low output current to charge these caps. That works fine. I purposely fried those caps trying to get more out of them. I knew it was likely to happen before I heard the magic pop. The question I need to know is will the MOT connected directly to the caps burn them up when the power is switched on? I know the voltage output will not exceed the voltage rating of my caps. \$\endgroup\$ – grantr Nov 2 '15 at 20:15
1
\$\begingroup\$

I have 4 1300 volt 5 uf film caps in series which gives a voltage of 5200. I have charged these caps with a low powered hv power supply without problems. Output of the HV supply was well over 5200 volts. As long as I did not leave it on too long the caps were fine.

So, for simplicity, let's say you have two 50V caps in series and one, due to tolerances is 4uF and the other is 6uF. OK so far?

You feed this with current or voltage or whatever and what you will find is that the 4uF charges to 60% of the overall terminal voltage whilst the other only attains 40%. Do you see the problem?

People tend to counter this by charging with a ramp voltage and having bleed resistors across each cap - this tends to equalize the voltage to the same across each cap but that's messy at 5200 volts - even 1Mohm will dissipate 27 watts.

This is the big problem with series capacitors being charged together.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.