Polystyrene caps wayyyy off

[cloudscapes]

Heya,

Wondering if anyone has experience with measuring polystyrene caps. I got a batch from mouser and ebay for this project:
http://jhaible.com/legacy/frequency_shifter_fs1a/fs1a

And I'm finding the values to be super way off the marked/labeled values. And not by a bit. The closest match I have is a bag of 10nF being measured anywhere as between 18 and 22nF. Another example I tested was a bag of 8200pF being measured at around 18.5nF. And a 330pF cap measures at over 10nF, which is 30 times its expected value.

Is there a special way to measure these kinds of caps, with a multimeter in capacitor mode? I measure film/ceramics and those measure fine. So I have to assume there's a special way to measure polystyrene, considering all the various caps from multiple sources are all wrong..

Any ideas?

[EBK]

Perhaps replace the batteries in your meter?  Sometimes weak batteries can account for odd measurements.  Just a wild guess.

What are you using to measure them, by the way?

[davent]

Just measured what i have and all within the second decimal place,  nothing special for measuring.

I've been getting coupling caps for my tube projects from this guy in Toronto for a good number of years and he has pretty extensive number of values in 5% and 1% polystyrenes. A real person to deal with help you out, $25 minimum plus shipping.

https://www.justradios.com/cart.html
dave

[Marcos - Munky]

Perhaps replace the batteries in your meter?  Sometimes weak batteries can account for odd measurements.  Just a wild guess.

I'd try that. Meters go wild on measurements when the battery is weak, despite the "weak battery" icon still not being visible.

[cloudscapes]

I'll try the new battery tomorrow (gotta go out and buy some), good idea.

I'm walking back on a bit what I said, it seems it's having trouble measuring some small value film/ceramic/mlcc caps as well. It had no trouble nailing 100nF and higher, but was completely off on the 10nF and lower. So that probably nails down either battery or the meter itself. It's not one of those cheap $20 meters, it served me well for a decade, but it's no Fluke either. UNI-T UT60A

Thanks for the link Dave! I'll try them soon!

[EBK]

The manual for your meter says this:

"To increase the accuracy of capacitance measurement especially when measuring under400nF, use REL mode to automatically subtract theMeter built-in equalized capacitance and residualcapacitance of test leads from the result."

Maybe that would help too.  Also, the lowest range on your meter is 40nF.  I would expect that it wouldn't measure pF-range caps very well.  I had similar limitations with my meter too.  That's why I bought myself a Peak Atlas LCR45 to measure caps.  Not cheap, but quite handy for identifying caps with mysterious labeling.

[amptramp]

Just measured what i have and all within the second decimal place,  nothing special for measuring.

I've been getting coupling caps for my tube projects from this guy in Toronto for a good number of years and he has pretty extensive number of values in 5% and 1% polystyrenes. A real person to deal with help you out, $25 minimum plus shipping.

https://www.justradios.com/cart.html
dave

I have two boxes of his yellow capacitors beside me.  The proprietor, Dave Cantelon, is a fellow member of the London Vintage Radio Club and a lot of us use his capacitors.

[mozz]

Polystyrene caps should be very close to the stamped value. I was going through some last night for a build as they are used in old wah pedals. The dissipation and R are about the lowest you will find in a cap and mica would be next in line to that.

[Rob Strand]

Apart from the battery flat issue:

When you measure caps you have zero the meter, or use the Rel function on your meter.  Cap meters will measure *all* capacitance including the leads.  You cannot touch the probe tips, even holding the probes can have a small effect but not to the level you are seeing.   It's best use clips and zero with the clips not connected to a cap.  For small values you cannot let the leads move after the zeroing the leads need to be kept in the same position.

If your meter does not have a zero then you have to read the display with no cap connected.    Then measure the cap.  Then to get the true value subtract the zero measurement from the displayed measurement.

On some meters you have to do that on the exact same range as the cap is measured.  That means for autoranging meters you have to force or hold the range.   You have to know what your meter needs.

10nF being measured anywhere as between 18 and 22nF.
8200pF being measured at around 18.5nF.
And a 330pF cap measures at over 10nF,
which is 30 times its expected value.

If you look at the measurements they are all 8 to 10nF too high.
That's more like a zero offset than a multiplicative error.

[mozz]

If the offset is 10,000pf high, there are other problems. I think the meter was meant to only measure higher values. You can get a lab grade LCR fairly cheap on Ebay, a lot of quality stuff was used in schools and production lines that didn't have a well known name such as HP or Textronix.

[Rob Strand]

If the offset is 10,000pf high, there are other problems. I think the meter was meant to only measure higher values. You can get a lab grade LCR fairly cheap on Ebay, a lot of quality stuff was used in schools and production lines that didn't have a well known name such as HP or Textronix.

Definitely not a normal case.  I'd say 50pF to 100pF would be typical for leads.  You might see 1nF on some meters since they can have added capacitance inside the meter.

That's why I'm thinking the OP was *touching* the cap leads.   You probably only need to touch one to stuff-up the measurement on some meters.    If the zero is off the offset could be artificially high.

--------------------
One other thing:

If you don't know the quirks of your meter  you could measure some polyester caps/green caps or ceramics of the same value and see what you get.    Measure in the same way.

That should show up if it's the way the caps are measured or it's the meter.

[duck_arse]

what we all really want is to see some pics of the caps causing all the trouble. please.