I have removed some capacitors from disposable camera flash units. They are 80 uF, 100 uF and 120 uF at 330 volts. Some of the say "photo" or "flash" on them. I want to use them in a bench power supply with a maximum of 90 volts on them.

These caps are small compared to "normal" caps of these values. Can anyone think of a reason not to use them in a non-flash circuit. Will they fail young if a voltage is applied to them for long periods, for instance.

 

The dielectric breakdown voltages for a capacitor always need to be higher than the potential difference in the circuit -- so you're fine there -- provided that you mind the polarity. If you reverse polarity, you'll figure it out the moment they explode and the neighbors think you're firing a gun indoors.

I assume you're using them in parallel with the output of your supply to smooth out some ripple? The real concern is that they are relatively low capacitance. For less than $5, you can buy something new whose capacitance is 100 or 1000x higher.

 

Hi,
A Filter capacitor at the secondary side (power supply circuit),the value is depends on the current of the loads.
If a high current lets say 3A TO 5A load is required then a 1000 to 2000 uf and the voltage rating
must a higher than the output dc voltage.

 

I assume you're using them in parallel with the output of your supply to smooth out some ripple? The real concern is that they are relatively low capacitance. For less than $5, you can buy something new whose capacitance is 100 or 1000x higher.

Actually the intent is to use them as the part of a filter on the power to an analog regulator. The power for the analog regulator comes from a switching pre-regulator through a 100 uH coil. The caps have to have a low impedance to the switching frequency and also supply the instantaneous current that the analog regulator is delivering to the load. (I am wondering if the 100 uH coil is too much inductance but that is another matter I will deal with later if needed...)

Right now I only have a 47 uF bulk capacitance and I am thinking it is way too low. I am shooting for more like 400 uF. The way I built the circuit it may actually be easier to mount 4 small caps that to mount one big one. In fact I have some larger caps both in value and size but I will have a very hard time fitting them in.

Also, I am trying to be cheap and the thought of 4 "free" (and available) caps has a lot of appeal.

 

Actually the intent

Dang it -- I knew I should have asked instead of just assume Sorry about that.

Hey -- can you provide a schematic of your circuit & frequency of operation?

 

Actually the intent is to use them as the part of a filter on the power to an analog regulator. The power for the analog regulator comes from a switching pre-regulator through a 100 uH coil. The caps have to have a low impedance to the switching frequency and also supply the instantaneous current that the analog regulator is delivering to the load. (I am wondering if the 100 uH coil is too much inductance but that is another matter I will deal with later if needed...)

Right now I only have a 47 uF bulk capacitance and I am thinking it is way too low. I am shooting for more like 400 uF. The way I built the circuit it may actually be easier to mount 4 small caps that to mount one big one. In fact I have some larger caps both in value and size but I will have a very hard time fitting them in.

Also, I am trying to be cheap and the thought of 4 "free" (and available) caps has a lot of appeal.

The only thing I can think of is ESR. Too high and the cap will start heating if significant ripple is applied. If you can measure it, you can calculate the heat dissipated.

Also, the small size that you described may mean it can handle fewer watts than most filter caps and you may see temp rise more than expected for a given watts of heat losses.

 

Ditto GopherT on the ESR. I would think that a disposable camera capacitor is of the cheapest possible type, with ESR not even a consideration. You may wish to parallel 10 uF of ceramic capacitor with that photoflash capacitor. The ceramic will help absorb the sharp edges from the switching supply.

 

The only thing I can think of is ESR. Too high and the cap will start heating if significant ripple is applied. If you can measure it, you can calculate the heat dissipated.

In the reason I started on this wild ride of learning electronics at my age, a DIY electrical discharge machine, many of the published circuits use flash caps. And the use them for the low ESR.

"Compared with electrolytic capacitors usually used for power supply filtering at power frequency, a photoflash capacitor is designed to have lower ESR, ESL, and capacitance manufacturing tolerance, but does not need as high a working temperature ". From, https://en.wikipedia.org/wiki/Photoflash_capacitor

 

In the reason I started on this wild ride of learning electronics at my age, a DIY electrical discharge machine, many of the published circuits use flash caps. And the use them for the low ESR.

"Compared with electrolytic capacitors usually used for power supply filtering at power frequency, a photoflash capacitor is designed to have lower ESR, ESL, and capacitance manufacturing tolerance, but does not need as high a working temperature ". From, https://en.wikipedia.org/wiki/Photoflash_capacitor

Nice article. The temp rating may be an issue if other electronics are heating in the neighborhood on the PCB or the enclosure. Apparently ESR is low so the heating won't likely come from the cap.

from the referenced wikipedia article:

Photoflash capacitors are not subject to the high temperatures of cased electronic equipment in continuous operation, with nearby components and sometimes the capacitors themselves dissipating heat; they are often rated at a maximum operating temperature rate of typically 55 °C, compared to 85 °C–105 °C or more for capacitors for continuous use in electronic equipment.​

 

Okay, the caps should be fine as far as ESR is concerned (I will put a few microfarads of ceramic caps in parallel anyway). I am a bit concerned about the working temperature since the cap is near the linear regulator's series pass transistor. The transistor is only intended to dissipate about 16 watts and it is on a fairly large extruded heatsink (1.5 degC/watt) -- but still worth considering.

The low temperature and the low duty cycle of usage may explain the small size. The small size will not dissipate as much heat as a larger size. I found a _much_ bigger 600 uF, 330 volt cap in another photoflash unit I have. I may try making it fit. It may still have the temperature limitations, however. It is a brand name cap -- Elna if I am not mistaken -- so I at least have a chance of finding data on it.

Thanks for all of your excellent input so far.

 

I have removed some capacitors from disposable camera flash units. They are 80 uF, 100 uF and 120 uF at 330 volts. Some of the say "photo" or "flash" on them. I want to use them in a bench power supply with a maximum of 90 volts on them.

These caps are small compared to "normal" caps of these values. Can anyone think of a reason not to use them in a non-flash circuit. Will they fail young if a voltage is applied to them for long periods, for instance.

They are probably optimised for heavy pulse discharge - but other than that; they're pretty much just electrolytics.

You could make an "atom bomb to kill an ant hill " CDI - but you'd need a serious inverter to pump it up fast enough for engine RPM.