The ESR of an electrolytic capacitor is normally just a small fraction of an Ohm for a high capacitance, low voltage capacitor (such as a 1000µF, 16V cap), and can be as high as two or three Ohm for a low capacitance, high voltage cap (1uF, 450V). When the capacitor ages, this resistance increases, and it often does so in such a dramatic way that the equipment completely ceases to function or even blows up semiconductors. It's very common to find capacitors that have degraded to 100 times their normal resistance, while their capacitance remains fine! On a typical capacitance meter they will measure close to their correct values, but they are completely bad! This is where the ESR meter comes in: It measures the equivalent series resistance of the capacitor, almost independently of its capacitance.
An additional beauty of an ESR meter is that in almost all cases it
can check capacitors while they are in the circuit! This is so because
a good capacitor would measure almost like a short circuit, and so any
other parts connected in parallel will have minimal influence on the measurement.
These are the features that make an ESR meter an irreplaceable tool for
troubleshooting electronic equipment.
Here is the schematic, which you can click on to get a larger version for printing.
One section of a dual low power operational amplifier is used as a square wave oscillator. A small ferrite core transformer is used to step down the voltage and provide the necessary low impedance output. A 10 Ohm resistor loads the output to absorb inductive spikes from the transformer, which could cause a false reading for low value capacitors. The other section of the op amp amplifies the signal that gets through the capacitor being tested, and its output is rectified and applied to a 50µA galvanometer through a calibration potentiometer. A small 5 Volt regulator maintains the supply constant while the instrument is being powered from anything between about 7 and 15 Volt. I power the meter from the 13.8V bus which I have in my workshop, but if you prefer, you can use a 9V battery instead, connected through a switch. The power consumption of this circuit is so low that a 9V battery should last at least 100 hours.
Building this ESR meter is simple and straightforward. I assembled the circuit on a scrap piece of project board, and used a small plastic box to install the board and the meter. The only part that could pose problems to inexperienced builders is the transformer. I made mine using an Amidon ferrite core, type EA-77-188, which is a tiny double-E core having a cross section of 22mm2, and external dimensions of about 19x16x5mm total. I used the nylon bobbin that Amidon delivers with it, wound a primary winding consisting of 400 turns of AWG #36 wire, and as secondary I wound 20 turns of AWG #26 wire. If you have a larger or smaller core, you can adjust the turn numbers in inverse proportion to the cross section area. The wire size isn't critical - the gauges I used are about 3 or 4 numbers thicker than necessary, while at the same time this bobbin has room for wire at least two numbers thicker than the ones I used. Thus, you can choose from about 6 different wire gauges for each winding, with negligible impact on the performance.
Considering that the transformer is so uncritical (because it runs at very low power), feel free to use any small ferrite core you have on hand, as long as it has no air gap. Dead PC power supplies and old monitors or TVs are great sources for such cores. Do not use an iron core, because it would probably have far too much loss at 50kHz.
The test leads are soldered into the circuit, and fixed in place using hot melt glue. Soldering them is much preferable over using any sort of connectors, because this meter easily detects resistance as low as 0.1 Ohm, and a connector can easily vary its resistance more than that! By the way, this set of test leads was bought as standard tester replacement leads, for very little money.
The meter is a reasonably good one rated at 50µA full scale, which
I had on hand. If you find a cheap VU meter that works well, you can use
it, of course. If you prefer to use a 100µA meter, change R11 to
50k. I used a trimpot for R11, but you might want to use a panel-mount
potentiometer instead, which would allow fine adjusting the full-scale
point if your meter happens to be unstable. If you use a cheap meter I
would recommend this.
He powered it from a 9V battery and even added a LED!
Curt used an existing transformer taken out of an old PC power supply. It has much lower impedance, and slightly lower turns ratio than my transformer, but works well enough.
Instead of the TL062, he used a TL084. That's a change I would not recommend, because the TL084 is not rated to work at the low voltage used here, so it's a matter of luck and tolerances that it works at all! Also, given the much higher saturation voltage of that opamp's output section, Curt had to modify the value of R8 to get a reasonable scaling on the meter. Before that, his instrument was driven into saturation, compressing the scale. But finally it worked for him, showing that builders can take quite a bit of artistic liberty and still get a useful result!
Curt used flat-style dead-bug construction on an unetched piece of PCB,
obtaining a low profile circuit.
do these two tantalum caps have? If you look at them, you will notice that
they look alike almost as much as two eggs laid by the same hen. The two
come from the same factory, and should belong to the same manufacturing
batch. The two came in a bag of 20 equal ones I bought at a local store.
Measured on a capacitance meter, both have almost exactly the same capacitance,
very close to their rated 47µF. Measured with a common ohmmeter,
both have essentially infinite insulation resistance. But when measuring
their ESR with my newly built meter, I was in for a surprise: The one at
right, like the other 18 caps not shown in this picture, has an ESR of
about 0.2 Ohm, which is quite normal for a tantalum cap of this size. But
the one on the left is a bad apple (or egg?) and has a whopping 15 Ohm
ESR!!! It has a manufacturing defect, impossible to tell without measuring
the ESR. If I had placed this capacitor in a circuit that exposes it to
significant ripple current, it would have exploded, which is a trick tantalum
caps just love to perform. Worse than that, if I had used it to filter
a signal, it simply wouldn't have worked well, and I would have never suspected
why my circuit didn't perform as calculated! After this surprise, I must
advice you to test every component before putting it into a circuit, and
never assume that a part that comes fresh from the factory is actually
a good part!
The first true troubleshooting use for my new ESR meter was in rejuvenating my dear oscilloscope. That one is a 30 year old Tektronix, which I bought in bad condition 20 years ago and fixed. For those 20 years I hadn't done much to it, and over the years several minor functions had failed, but the biggest problem was that the trace had degraded to a fuzzy broad strip, its intensity modulated by hum and by the signal being measured! This really smelled like degrading electrolytic capacitors, no wonder after 30 years.
With my ESR meter I quickly found five electrolytic capacitors which
had degraded a lot. Interestingly, four of them were of the same brand
and can size, even if of two different values - probably there was a sealing
problem! Of these capacitors, one had 4 Ohm ESR, one had 6 Ohm, one 7 Ohm,
and the other two had such high ESR that the meter wouldn't even deflect!
The other electrolytic caps in the oscilloscope were all still healthy,
with ESR values well below one Ohm for most, and close to one Ohm for a
few low capacitance, high voltage ones. I replaced the bad ones, and this
gave me back a well defined and stable trace! Also one of the dead functions
revived. The other dead functions were due to two open resistors, one value-shifted
resistor, and two open transistors, which I found by conventional troubleshooting.
After fixing all that, and spending two days completely realigning the
instrument, it's again as good as new! But that's a matter for another