[sdiy] Caps, nerds, New Year's Eve and soakage

[Roman Sowa]

Good point about temperature!
Ambient temperature during 1st measurement was 16.1C, and during the 2nd 
one 20.3C. Although the termometer was about 1.5m away from the cap so 
this is just aproximate.
Also, the temperature changed during that period from 16 up to 20, then 
down to 14 and up again to 20.3C

Typical tempco of polypropylene cap is -200ppm and dielectrical 
absorption 0.05% - numbers taken from Vishay whitepaper about caps.
So that makes in theory 11mV rise from absorption (32V down to 10V) and 
8.4mV rise from temperature change for total of 19.4mV.
That's still about 7mV less than measured, so aparently the cap was not 
that good. Assuming -250ppm tempco and 0.08% DA it all matches.

Another source of error is multimeter input capacitance of 370pF, so at 
the moment of probing the cap with previously discharged probes (meter 
showing 0V) voltage drops by about 2.5mV as it forms capacitive divider 
with tested cap charging meter's input.

Anyway, it was good fun!

As for higher voltage - you dare me to it. Maybe I'll try to charge it 
to over 300V from AC outlet with a resistor and diode.

Roman

W dniu 2018-01-03 o 20:01, Richie Burnett pisze:
> Interesting experiments Roman! ...but you bottled out only charging that 
> 400v cap to just 32 volts! :-)
> 
> How constant do you think the temperature was during your experiment?...
> 
> I wonder if the voltage changes you measured could be attributed to a 
> change of capacitance with ambient temperature. It is charge that is 
> conserved, so terminal voltage would rise and fall slightly with 
> temperature unless the capacitor has zero temp coefficient.
> 
> -Richie,
> 
> Sent from my Xperia SP on O2
> 
> ---- SHELLY wrote ----
> 
> Roman —
> 
> I have also found unexpected soakage in polypropylene.
> 
> http://ijfritz.byethost4.com/super_sh_rand_web.pdf
> 
> Ian
> 
> 
> Sent from XFINITY Connect Mobile App
> 
> 
> ------ Original Message ------
> 
> From: Roman Sowa
> To: synth-diy at synth-diy.org
> Sent: January 3, 2018 at 8:15 AM
> Subject: [sdiy] Caps, nerds, New Year's Eve and soakage
> 
> As a last activity in the office the last day of 2017 (Friday 29th) I
> have charged a bulky polypropylene capacitor up to 32.1V with the
> intention to measure it after going back on first Monday of 2018 (here's
> the New Year's Eve part of the subject).
> The cap is 1.5uF/400V from year 1987.
> To my surprise, after almost 4 days, or more precisely 88 hours, it was
> still charged to more or less the same voltage. Unfortunately my meter's
> input resistance is 10Mohm for ranges above 10V, so it started quickly
> discharging during measurements. First reading was not too big as it
> started in the middle of converter cycle, and next one was 31.2329V with
> the next ones dropping by about 0.5V each. So assuming the meter was
> connected about 1.5 cycle earlier, that indicates it was really about
> 32V at the start. No drop whatsoever during 88 hours!
> To make more precise measurements, I then charged it to 10V, in order to
> take advantage 1Gohm loading of the meter, and left it for another 29 hours.
> Yesterday's reading - 10.0004V
> Today's reading - 10.0277V
> Not only it did not drop, but voltage rised by 27mV! I think that's
> dielectric absorption. After all, it was charged to 32V for 4 days, so
> it could have soaked a bit. OTOH I wasn't expecting any absorption
> effect to be aparent in polypropylene capacitor.
> So now I discharged it and let it rest till tomorrow to see how high it
> can go by itself.
> 
> Roman
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at synth-diy.org
> http://synth-diy.org/mailman/listinfo/synth-diy