Esr & low value capacidence

[Bigdrums]

I am considering buying an ESR meter. I've read that ESR rises when the value of the capacitor goes down.
I'm wondering if this is always the case or if I absolutely need an ESR meter at all.
Also seems that prices vary greatly without clear benefits to the extra cost.

[Mick Bailey]

A capacitor can read spot on for value, but have high ESR. I get a lot of use out of my ESR meter and find it's a good time saver. I came to using one rather late on and wish I'd built one much earlier. ESR is just one parameter though, and it won't tell you if a cap leaks. Also a shorted cap will read good for ESR. A usual, they give a bench measurement and a capacitor in-service is subject to much higher voltage than the meter produces. Only you can decide if you really need one.

I have a $7 component analyzer that tests for ESR amongst other things. Probably far more useful overall than a dedicated meter. I built my own meter out of parts from an old computer power supply, a surplus meter and stuff lying around. I then calibrated the scale using fixed-value 1% resistors. Simple, but it works really well.

[Jazz P Bass]

"I have a $7 component analyzer that tests for ESR amongst other things"

I bought one of them myself.
https://www.ebay.com/itm/IR-Decoder-...81a35fc7b74dd6

A cute little gadget.

[olddawg]

I used to use an ESR meter a lot... especially for surface mount caps. After a while I got tired of call backs and simply replace any suspected cap. SUPPOSEDLY.. you can check ESR in circuit.... yeah sometimes. Usually you have to lift one lead anyway. I found it was more cost effective to keep a good supply of caps and replace anything suspect. They’re cheap.. time isn’t. I had 4 different ESR meters.

[Enzo]

Exactly, olddawg. If you are a hobbyist wanting to experiment, fine. But if you are a working tech, well, how much of my dollar a minute time should I spend testing a dollar part, especially if I have to lift it? If I doubt a part, I replace it.

[olddawg]

Exactly, olddawg. If you are a hobbyist wanting to experiment, fine. But if you are a working tech, well, how much of my dollar a minute time should I spend testing a dollar part, especially if I have to lift it? If I doubt a part, I replace it.

Yep... and how many times have you tested a cap... and it tested good... but when replaced solved the issue. Tremelo circuits come to mind. Very few cap testers will test a cap under operating conditions. I had a big expensive Sencore cap tester in the shop at one time. It took a ridiculous amount of time to set up and I got tired real fast of other techs blowing up caps. The only thing I used it for was testing large, expensive filter caps... and I always wore safety glasses

[Bigdrums]

Can this one be used in circuit?

[Enzo]

Most of them can be "used in a circuit". Problem is, the circuit around the parts can cloud your results. And you have to know when that is. And that comes from tech experience, not the meter.

Here is the deal, those meters are cool little things, we had nothing like that 50-60 years ago when I was beginning. You can learn things using them. On the other hand they will never tell you "this part is good and will cause no problems". So for the VERY inexpensive price of these, go ahead and try one. Over time you will learn what it is actually good for and what it really is not good for. Like any tool.

[The Dude]

In a way, they are like a tube tester. If they show a cap is high ESR, it's bad. If they don't, the cap could still be bad. IMO, an ESR meter can still be a useful tool and time saver.

[Mick Bailey]

My tester does in-circuit test just fine - I've never had to lift a leg to test. I have mine calibrated so that FSD is 30 Ohms, I use it for electrolytic caps and nothing else so that's a useful range. The test voltage is low enough not to turn on a semiconductor junction, and high enough frequency to overcome reactance. It's unusual for a parallel path with a capacitor to skew the results too much in audio circuits.You have to be aware that this is a possibility, though. Also, I can test a charged cap without accidental damage, but some meters don't have enough protection.

What needs to be understood is that an ESR test is just one parameter you're checking and it needs to be an appropriate test for the particular circuit. For example, a trem circuit would not be something I'd use an ESR tester for because a higher than usual ESR reading for the phase-shift caps won't affect the operation and the usual failure for these is DC leakage (or a cap is open). So the context of the test is important - you test for ESR where it's appropriate to the likely failure mode of the circuit. I had a high-end studio limiter in for repair and calibration recently and one of the reasons for issues with this particular unit can be the electrolytic caps giving a higher than usual ESR and it's the cumulative effect that gives issues. They're not dollar parts in this case, and the time and cost to replace 29 of them is greater than the three minutes it took to check them all for ESR. As it happens they all tested good. I look for high ESR where its relevant to the circuit, not as an assurance whether a cap is good in all respects.

However, an ESR reading can give you an idea of a caps condition before it fails through DC leakage or shorting. If an expected reading is (say) 1 or 2 ohms and you're seeing 15 then the cap is heading in the wrong direction, yet that same cap may read fine for leakage at full operating voltage, and still be operational.

It's down to personal preference and what works for you. I find my meter to be an invaluable time saver, but it's just one of a number of test items I use. The same goes for my ring tester - It takes a shorter time than it takes me to write this sentence to test an output transformer in-circuit and tell if its good, bad or still working but has a problem.

[Helmholtz]

I have a $7 component analyzer that tests for ESR amongst other things. Probably far more useful overall than a dedicated meter. I built my own meter out of parts from an old computer power supply, a surplus meter and stuff lying around. I then calibrated the scale using fixed-value 1% resistors. Simple, but it works really well.

So your preferred ESR meter is your own design?
What is FSD?

[g1]

I think FSD would be 'full scale deflection' of the meter.

[Enzo]

Oh, I wasn't coming up with full scale deflection either, but I suspect you are right. That is one of those terms lost on millennials. None of their test gear uses a moving needle. It is on the list with "dial" phone or "carbon paper". I bet "mirrored scale" might be lost as well.

[g1]

Some of them would probably tell you 'scales are for weed, man'.

[olddawg]

My tester does in-circuit test just fine - I've never had to lift a leg to test. I have mine calibrated so that FSD is 30 Ohms, I use it for electrolytic caps and nothing else so that's a useful range. The test voltage is low enough not to turn on a semiconductor junction, and high enough frequency to overcome reactance. It's unusual for a parallel path with a capacitor to skew the results too much in audio circuits.You have to be aware that this is a possibility, though. Also, I can test a charged cap without accidental damage, but some meters don't have enough protection.

What needs to be understood is that an ESR test is just one parameter you're checking and it needs to be an appropriate test for the particular circuit. For example, a trem circuit would not be something I'd use an ESR tester for because a higher than usual ESR reading for the phase-shift caps won't affect the operation and the usual failure for these is DC leakage (or a cap is open). So the context of the test is important - you test for ESR where it's appropriate to the likely failure mode of the circuit. I had a high-end studio limiter in for repair and calibration recently and one of the reasons for issues with this particular unit can be the electrolytic caps giving a higher than usual ESR and it's the cumulative effect that gives issues. They're not dollar parts in this case, and the time and cost to replace 29 of them is greater than the three minutes it took to check them all for ESR. As it happens they all tested good. I look for high ESR where its relevant to the circuit, not as an assurance whether a cap is good in all respects.

However, an ESR reading can give you an idea of a caps condition before it fails through DC leakage or shorting. If an expected reading is (say) 1 or 2 ohms and you're seeing 15 then the cap is heading in the wrong direction, yet that same cap may read fine for leakage at full operating voltage, and still be operational.

It's down to personal preference and what works for you. I find my meter to be an invaluable time saver, but it's just one of a number of test items I use. The same goes for my ring tester - It takes a shorter time than it takes me to write this sentence to test an output transformer in-circuit and tell if its good, bad or still working but has a problem.

I used to believe all of that... but I spent too many years working on pro and consumer camcorders back in the day... I would check a whole board of surface mount caps and replace the ones with bad ESR.. what I found over time was that if a couple of the same value cap were bad... the rest would probably be bad soon. So I would “stuff” the boards with dozens of caps. It was tedious and time consuming, but the repair was reliable. All of the major manufacturers offered “kits” for various models. I guess my point is I no longer believe that in circuit ESR testing will reliably identify a “marginal” cap.