From: pcdhSpamMeSenseless@electrooptical.net   
      
   On 2020-05-28 17:48, Ralph Mowery wrote:   
   > In article <041d25c9-3829-9f33-f1cb-506ac85a75dd@electrooptical.net>,   
   > pcdhSpamMeSenseless@electrooptical.net says...   
   >>   
   >>> I was using a Fluke 87,  a LCR meter from China, an older component   
   >>> tester and the new component tester.   
   >>>   
   >>> The first capacitor was a Sprague .06 uF 600V.  Two China testers showed   
   >>> near the value.  Within the 10%  tollorence,    The LCR tester showed it   
   >>> to be .08 and the Fluke as .1 uF.   
   >>> This is a new,but very old capacitor.   
   >>>   
   >>> Next capacitor was a 20 year old no name of .068 of 50 V made with the   
   >>> Poly something dielectric.  All meters were with in less than 10 %.  Ok   
   >>> here.   
   >>> Same results with a newer one of .01 uF .   
   >>>   
   >>> Next came a Silver mica.  It is .01 at 600 V. Fluke shows up at .0150,   
   >>> LCR at  .0120.  Two component testers were close and in spec.   
   >>>   
   >>>   
   >>> What gives with some capacitors checking like they should and some being   
   >>> way off, not just 10 % or so ?  I ran the tests several times on each   
   >>> capacitor to see if maybe the leads were not making good contact and any   
   >>> other similar thing I may have missed like having my fingers across the   
   >>> leads.   
   >>>   
   >>> Ralph ku4pt   
   >>>   
   >>   
   >> Might be bad soakage--micas are horrible for that.  If you stick a 1-Hz   
   >> square wave into it and look at the voltage across a the 1-M input   
   >> impedance of your scope, you might see something interesting.   
   >>   
   >> Scoping what the meters are doing to the cap would be interesting too.   
   >>   
   >>   
   >>   
   >   
   > I have a decent test bench.  Not lab quality,but not too bad.   
   >   
   > I hooked a function generator set as square waves to 2 1000 ohm   
   > resistors for isolation.  From each resistor I went to a capacitor.  One   
   > was the Sprague  and the other was the poly something capacitor.  Then   
   > back to the ground side of the generator.   
   >   
   >   A dual track Hanteck 200 mhz scope with 10:1 probes were hooked across   
   > the capacitors.  I started out at .1 Hz and went up to around 10,000 Hz.   
   > At all times the traces were almost identical.  They started out as   
   > almost perfect square waves as expected .  At a couple of hundred cycles   
   > the leading edge started to show a rounding off near the rise of the   
   > cycle (top of the trace)  and same for the negative part of the cycle.   
   > As I increased the frequency they started resembling sine waves.  About   
   > like I expected.  At no time was there any measurable difference in the   
   > waveformes.   
      
   My suggestion was to look at the current through the capacitors going   
   into a high impedance.  That way you don't load down the generator.   
   >   
   > I measured them again on the capacitor checkers and same results. The   
   > poly capacitor was very close on the meters and the Spraque was showing   
   > .1 instead of .06 or close to .06 on hte Fluke and .08 on another   
   > tester.  Had it shown .07 or .05 I would have called it meter tollorance   
   > but not almost double.   
   >   
   >   
   > I don't have a capacitor tester that I can put any high voltage on them   
   > like 500 or so volts.   
   >   
   > When I have more time I may try scoping the meters and see what they are   
   > putting across the capacitors.   
   >   
      
   Hmm, dunno.  The meters may be using different schemes.   
      
   Cheers   
      
   Phil Hobbs   
      
   --   
   Dr Philip C D Hobbs   
   Principal Consultant   
   ElectroOptical Innovations LLC / Hobbs ElectroOptics   
   Optics, Electro-optics, Photonics, Analog Electronics   
   Briarcliff Manor NY 10510   
      
   http://electrooptical.net   
   http://hobbs-eo.com   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)