Re: Stacked film caps

Geoff@work <gwood@xxxxxxxxxxxxxxxxxxxxxxxxxx> wrote:
>
>"Scott Dorsey" <kludge@xxxxxxxxx> wrote in message news:dnah84
>>
>> Try using it as a coupling cap into a 1G source impedance (about typical
>> for a condenser mike input stage) and look at the 1 KHz square wave
>> response on a scope. On some caps it can get pretty interesting. My
>> suspicion is these problems are caused by mechanical deformation of the
>> dielectric as it charges and discharges.
>
>What sort of scope probe/input are you using. Wouldn't that be likely to
>totally swamp the DUT ?

Last time I did this, I had the calibration output of my old Tek 545
into the DUT, coupled through a 1G resistor and with 50VDC applied
across it through a second 1G resistor. Then the other side of the
DUT was connected to the gate of a 2SK170 FET rigged as a cathode
follower (with a 1G leak resistor to ground). I measured the output
of the cathode follower with a 10M scope plugin (I forget which one,
but it's the 2-channel differential one with the nuvistor/6DJ8 cascode
thing).

So there's substantial DC across the cap as well as the signal, which
may have been as much as 10V peak-peak.

You might blame the cathode follower for some of the problem, but it
changes when I change caps, and it goes away completely with an NPO
ceramic.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."
.