I hope this note from my own 130 repair is interesting. First time I've had trouble from a cap that seemed okay.
SYMPTOM: 10pF reads high on 30pF range.
CAUSE: Dielectric Absorption in C92 .0047/400 charge pump cap.
NOTE: The cap is not leaky, it just has high DA.
It's a Sprague 160P DiFilm.
Measure C at two different frequencies. (I used my GR 1680-A.)
6.5nF at 1kHz, 7.5nF at 400Hz.
An ideal cap will be constant C, a cap with DA will read increasing as f decreases.
Can measure in-circuit if 130 range switch is OFF.
The change in C vs f makes the charge pump move more charge at low rep rate (1/3 scale) than high rep rate (full scale).
FIX: Replaced with polypropylene. Also replaced C93 .015/400 10pF range. Then recalibrated both ranges.
Joke's on me - now 10pF reads a bit *low* on 30pF range!
Maybe I should have used polyester, which has more DA than polypropylene.
I bet those 160P's had some DA when new, when Tek drew the meter scale.
Oh well, |error| is smaller now at least.
Replace all Sprague DiFilm in positions sensitive to Dielectric Absorption.
But don't use _too good_ a replacement.
From: TekScopes@groups.io <TekScopes@groups.io> on behalf of Tom Lee via groups.io <firstname.lastname@example.org>
Sent: Friday, December 11, 2020 3:31 PM
Subject: Re: [TekScopes] 130 LC meter
Looking forward to seeing the details when you’re ready! I very much prefer the 130’s approach to measurement over the RC-based ones that dominate ebay offerings.
As to the custom meter, I’m sure you’re correct. The relationship between frequency shift and component value is only approximately linear. Tek had to have made the scale a little nonlinear to compensate. HP routinely did the same for their instruments, as did many other companies.
Sent from my iThing, so please forgive brevity and typos
On Dec 11, 2020, at 13:51, SCMenasian <email@example.com> wrote: