Thanks to GW0OAJ for this article
I have recently finished making an ESR Meter and what a super item; a revelation. The best bit is that because of the nature of the beast, it can be used in circuits with no de-soldering required. Also, because the voltage level is very low, semiconductors are not driven into conduction, hence they are not affected. With my proclivity for old but high quality, valve stuff, I should have built one years ago!
Equivalent Series Resistance, ESR, represents those losses that might be seen as a series resistor connected to a perfect capacitor. It represents heating and all the badness that goes on with less than perfect components. It does not equate to the reactance component of the standard reactance or impedance calculations that are involved with the response of a perfect capacitor.
The markings were a bit shabby so I reworked them:
When fault tracing old kit, what is fairly certain is that:
1/ It is the electrolytics that fail; other stuff may be OK.
2/ Values below 1 μF just don't fail like those above 1 μF.
3/ Old "wet" types and any Tantalums are dead ringers.
4/ When they fail they really fail, no 'ish about them :D
5/ That failure may leave the capacitance value unchanged :(
6/ By any other means, high ESR is undetectable....
7/ ... and high ESR is baaaaad :(
8/ ... the sets performance is horrid....
9/ ... until there is a bang, a bad smell and no performance.
ESR Testers are easy to make. A number of patterns are out on the WWW.
To save a lot of looking, the best single page I found is:
The links are also what a lot of other searching will find you. Mine is chosen from here:
I modified it slightly. The guts are in four bits, interconnected by bits of electric spaghetti and the board was assembled in Ugly style. It works at a nominal 150 kHz, so no problems with signal paths :D
Top left of the PP3 battery is the 1uF input de-ballocking capacitor and back-to-back diodes for up to 400 volt surprises. Below left, the 74HC14 hex with one working as an oscillator and the rest strapped as a buffer. A power on LED, batt will be rechargeable type, hence an ext 2.1 mm connector. Zero adjust pot, signal conditioning; 2N2222 amp. The panel meter and directly on it, the signal bridge. Bottom right is a 78L05 to condition the voltage and assist with zero adjust.
It is hard to nail down solid information for exact values to expect. Electrolytic capacitor construction varies so much: chemistry, build quality, standards, technology, intended market, reliability required (Aerospace and Defence to cheap Chinese). This means there are no definitive standards for general values. Acceptable ESR also increases with voltage range and operating temperature. Large value caps have low (very) ESR and small caps have large ESR. For a specific type there surely is a standard.... but no exact guaranteed general purpose values.
A value judgement must be used with each test! The values I gleaned after considerable research, are at the top end of acceptable for semi-decent caps at a reasonable voltage. High quality caps should be much lower in ESR. Serviceable high voltage and/or high temp are allowed to be higher. Therefore general caps, for values as shown, should not really read higher, but may well be considerably less.
Typical de-couplers of, say, 0.01 μF, or even 0.1 μF hardly ever fail. What is a dead certain GO/NO GO is that *ANY* cap of 1 μF or larger that has an ESR of greater than 10 OHMS is truly knackered; wide as a barn door type fail..... even if measured capacitance is good!
Mine WORKS. I have found bad caps in new old stock already. I have not even looked at any in-situ on my old, duff gear, RA17, Cossor Scope, etc.
A couple of Youtube vids are linked to show how an ESR meter is used:
Not so well made, but to the point:
Copyright 2015 © Cwmbran and District Amateur Radio Society
Cwmbrân & District Amateur Radio Society
For access to this area email webmaster