Low Freq. Capacitor Charger

"Static" == "DC" so what *exactly* do you mean by a "10 Hz" pulse?

What voltage waveform do you want to appear on the capacitor?

If you just want to end up with a charged capacitor you probably want a flyback convertor circuit.

This is a dangerous circuit. Defibrillators work both ways you know...

Before you build the supply, you should do some math. Do you care how accurate the initial voltage? What's your definition of "Hold the charge for at least 24 hours?"

If you're content to have a third of the voltage after 24 hours, that's one time constant. 86400 seconds / 100uF --> 864 MegOhms of equivalent leakage resistance. Even if you ignore the leakage in the cap and the surrounding circuit, depending on the environment surrounding your plate electrodes, you might find that difficult to obtain. If you need more than a third left, you need proportionally higher equivalent leakage resistance.

Look into the technology used for radiation dosimeters.

As for the power supply... Last I checked, places that develop pictures from disposable cameras got paid 22-cents for each used camera they returned to the factory for refurbishing.

My local Walgreens sold me 10 of them for $3. The flash power unit is just about what you need.

mike

Sorry, if I did not explain this fully. The 10Hz charging circuit would be disconnected from the cap after charging, thereby leaving the isolated cap holding a static charge.

Jeff Niles

The charge circuit will be disconnected from the cap after charging, thereby leaving the cap fully isolated. I was also planning to use a low leakage variety, if they are available at 250V rating. Given this, what would be the longest interval I could expect before the cap discharges by half?

For this application, it must be charged with the 10Hz oscillator as described in my OP. It's part of an experiment.

Thank you for your reply,

Jeff Niles

Unless it is in evacuated chamber it will be gradually attracting any charged dust and ions from the air to the open capacitor electrodes - in much the same way that old CRT screens do. How long it will hold its charge depends on the leakage rate of the storage capacitor, quality of construction of the jig and the relative humidity amongst other things.

A 100uF capacitor at 250v will give you a nasty bite. You should make provision to earth the thing when not in use with a few meg across it.

Why 10Hz? It would be easier with a faster oscillator and a flyback step up. And do you want regulated 250V DC on the plates at the outset?

I suspect you will find capacitor leakage current a problem. Why can't the plates be actively maintained at 250V DC ?

Regards, Martin Brown

Film capacitors - mylar, polycarb, like that - can have self-discharge time constants of years. But 100 uF is a huge heap of film caps. A 1 uF cap would be about right if the purpose is to hold a charge on well insulated plates.

Why 10 Hz?

John

Sounds like we are being asked to assist with a homework (or extra credit) problem... Typical student type (or Physicist) with little understanding of practical electronics.

Oppie

Sorry Jeff, but specifying "10Hz" with no other contraints is non-sensical.

"10Hz" could be a single high energy pulse charging it in a few ms, or a train of current pulses taking an hour - indistinguishable from a linear ramp. I doubt anybody is going to take this requirement seriously unless you are able to explain better what it is you want, or perhaps explain the background to the "experiment" so we can work it out ourselves.

ETI and EA magazine both did zener diode testers using a circuit very similar to what you seem to be suggesting (their output was clamped at

Others have suggested the right types of caps to use.

I get that feeling too. It's straightforward to build or get a photoflash-type switching power supply, but the "10 Hz" business just gets in the way. Normally, one does step-up applications like this at a few KHz, using a tiny transformer.

John Nagle

Jeff Niles schrieb:

Hello,

you should do the necessary math before. Calculate the needed current to drive a 100 µF capacitor to a voltage of 240 V 10 Hz sinus and calculate the necessary primary current to the transformer to get the secondary current to the capacitor. 10 % duty cycle requires even more current.

Bye

John Nagle schrieb:

Hello,

a minatur audio transformer is not useful with 10 Hz anyway.

Bye

They work fine with short pulses repeating at 10Hz