advertisement


Siemens Sikorel 105 - 22000uf/40v

T

thomasp

pfm Member
I bought these power caps back in the 90s, I think they were made in 1989. I have been using it on and off until now. Are they still any good? I want to post a pic., just can't find the attachment button.
 
'What do you want to use them for?', is really the question.

If they have been used on and off, chances are they will be good yet.
Do you have a bench power supply..?, or small PSu - even a small wallwart type capable of 30-40vDC output? If so it would be easy to re-form them, measure the leakage current, and that is a proxy for all you really want to know.


PS you cannot post pics directly to pfm - you have to host the pic yourself somewhere, then post a link.
 
Yes, I have various small PS and I can reform them first. But forgive me for asking, how do you measure the leakage current?
 
When reforming you use a large -value resistor in series between the PSU and one terminal of the capacitor, such as 100Kohm (0.25w metal film, or whatever you have to hand of around that value; it is not remotely critical, and does not need to dissipate any heat)

Use your DVM to measure the voltage across this resistor through the process, and see what it reads at the end after several hours of applied voltage, up to a day or so:

100mV across 100K is 1uA leakage. It's a very sensitive test, even if your DVM is nothing special.
 
Just bear in mind that 100k x 22000uF has a time constant of 2200 seconds or 37 minutes. Multiply this by 6 to get to the pretty much fully charged state and that is 3.5 hours. You may also find that you get 'strange' results for leakage current because of fluctuations in the PSU output voltage. If this drops with the cap at full charge you may see 'negative' leakage current as the cap is now discharging back into the supply. If you can rig a LM317 to give a nice stable voltage it is worth the effort for measuring leakage. Also bear in mind that leakage current is dependent on applied voltage as well as capacitance value...
 
"100mV across 100K is 1uA leakage"
I charged 1 cap with a 24vdc wallwart (it actually go up to 32 vdc unload) with a 100k resistor. After 8 hrs, I measured 200mv across the resistor which is 2 uA, I presume. Is this acceptable for deployment?
 
You must log in or register to reply here.


advertisement


Back
Top