Me and my Tan Theta Meter (Capacitor tester)

[Mooly]

I am going to link a couple of capacitor threads to this. This meter was based on a design published by Cyril Bateman around 10 years ago. It measures the "goodness" of electroylitics, not the value, not the E.S.R. , it's much better than that.
So what is Tan Theta. Well Tan Theta = ESR/Xc put simply.

Some results then,
A "Nover" 47 mfd 16 volt cap gives 0.107. A Panasonic ECA 47 mfd 16 volt gives 0.068.
A Jamicon47 mfd 50 volt gives 0.035. A "value" 47 mfd 16 volt gives 0.106. A rubycon 47 mfd 100 volt gives 0.018.
A Gemcon 220 mfd at 16 volt gives 0.056 and a Panasonic ECA 220 mfd at 35 volts gives 0.056.
The lower the better !!! So those Panasonics are pretty good, that Rubycon exceptional.

[groovebot]

nice

[forr]

Hi Mooly,
I also built one, I found so few bad electolytics that Cyril sent me some to convince me that they exist.

[unclejed613]

does it measure in-circuit? i currently use an ESR tester that uses a 100khz signal rather than 50khz, and follow up with an ohmmeter (i had a cap once that measured low ESR, but it was still bad.... it was shorted). i think i'm going to build a scope attachment that works like a huntron tracker. i've found curve tracers to be very effective at spotting a lot of defective components a lot more efficiently than an ohmmeter. another device i think i'll eventually put together is a zener tester. i've seen a lot of zeners that measured ok with a meter, but had shifted their breakdown voltage more than 20% or acted shorted once full volage was applied.

if i build a curve tracer, i think i'll go the huntron one better and add a 100khz test mode. another good companion to a tracker type device is a magnetic probe so you can apply the tracker signal at for instance a power supply rail, and ust the mag probe to find a short by following the traces that use the most current (and create the strongest magnetic field). u used to have to find out which of 50 TTL chips was shorted. with a tracker it was simple. apply the tracker signal to the +5 rail and ground. then place the mag probe on top of each chip in the center of the chip. whichever chip showed a large current on the mag probe was the shorted one.

[Mooly]

Hello Forr and Unclejed,
It's a smart piece of kit for sure and yes it does test in circuit. A pity I don't use it more really.
It really is one of the few "original ideas"

[Lumba Ogir]

Mooly,
low "goodness"...Oh My Goodness Gracious...

[lineup]

Originally posted by Mooly

A "Nover" 47 mfd 16 volt cap gives 0.107. A Panasonic ECA 47 mfd 16 volt gives 0.068.
A Jamicon47 mfd 50 volt gives 0.035. A "value" 47 mfd 16 volt gives 0.106. A rubycon 47 mfd 100 volt gives 0.018.
A Gemcon 220 mfd at 16 volt gives 0.056 and a Panasonic ECA 220 mfd at 35 volts gives 0.056.
The lower the better !!!

Is not surprising one 100 Volt cap is better than one 16 Volt.
The Bigger the Better!
About 6 times bigger ...

Test one 16 Volt rubycon aganst one 100 Volt Panasonic

[Mooly]

Originally posted by Lumba Ogir
Mooly,
low "goodness"...Oh My Goodness Gracious...

me ?

[Lumba Ogir]

Mooly,

me ?

My Goodness!
ESR is one of many features, should have low goodness, ideally zero at all frequencies. In audio it`s more about high goodness, like linearity, dielectric losses and operating frequency range.
Thanks to Goodness.

[unclejed613]

did you say it was a kit or a public domain design? either way i'd like to build one...