Re: capacitor

[Sam Sampere <sampere@suhep.phy.syr.edu>]

On Sat, 8 Feb 1997 kowalskil@alpha.montclair.edu wrote:

>  
> As you can verify, the initial slope gives R*C of about 1000 seconds.
> This means that R is 20000 mega-ohms. I call it "the lower limit" 
> because the leakage resistance of my C is in parallel with the probe. 
> The input impedence may actually be much higher than R.
>
> Note that I am showing only the beginning of the discharge curve. The
> semi-log plot of all data is not a straight line indicating that R 
> grows rapidly as the voltage is lowered. The capacitor may or may not 
> be responsible. Does it make sense to think that the input impedence
> itself is voltage-dependent?
>                                         Ludwik Kowalski
You did not specify the type of capacitor used.  My bet is that you used
an electrolytic capacitor.  Try this experiment with the same value
capacitor, but this time use a mylar cap.  Compare the two log curves.
The mylar cap is linear to almost the limits of the ULI resolution.  The
electrolytic diverges after only a few time constants.   I think this
explains your 'voltage dependant resistance'.  Those deviations are
capacitor dependant, not voltage dependant.

Sam