Test capacitors without an ESR meter?

[cbmeeks]

Is there a way to get a good estimate on the life of a capacitor without an ESR meter?

Especially something a little more accurate than "it's good" or "it's bad".  Like something that would estimate a cap being within 10% of specs or 80% of specs, etc.

Thanks!

[setq]

A good thing to remember is that a low ESR doesn't imply that the capacitor is any good. I've had a couple of them read fine on an ESR meter and then blow up in my face. So the policy developed of ignoring the ESR meter and just replacing them in anything that comes through the door regardless of condition if it's more than about 5 years old.

Yes quite a few end up being binned even if good but I've been on the hot end of a Sprague Atom before and don't recommend arguing with them or taking a risk.

[cbmeeks]

Ah, good points.

I restore vintage computers.  Some of them are pushing 40 years old...but I admit that I don't always do a full recap on them.  Maybe I should replace them more often.

[Kleinstein]

It is possible to get an ESR test with an square wave generator a resistor (could be 50 Ohms output impedance) and a scope. The generator could be as simple as a NE555 or 74HC14 based one. Sometimes even the adjustment signal of the scope is good (strong) enough.

[helius]

Additionally, dielectric leakage of capacitors is another common failure mode that is separate from high-ESR. It can be tested with only DC instruments, while testing ESR requires an AC setup.

[kripton2035]

you can get plenty of easy to build esr meters from my repository web site (link below)
most of the time, they can be made with some left components in your drawers.

yes you can measure esr without an esr meter (ans there are also descriptions in my repository)
but it's easier with en esr meter. even a simple one.

[David Hess]

Is there a way to get a good estimate on the life of a capacitor without an ESR meter?

Especially something a little more accurate than "it's good" or "it's bad".  Like something that would estimate a cap being within 10% of specs or 80% of specs, etc.

Sure, use an impedance bridge.

[JacquesBBB]

Is there a special reason why you do not want to use an ESR meter ?

You can buy a  very simple but very efficient one for less than 10$ on ebay
as  (completely randomly)
http://www.ebay.com/itm/Mega328-Transistor-Tester-Diode-Triode-Capacitance-ESR-Meter-MOS-PNP-NPN-M328-/172326351345?hash=item281f7371f1:g:km4AAOSwdIFXx~zz

see also the very lengthy thread
https://www.eevblog.com/forum/testgear/$20-lcr-esr-transistor-checker-project/

These are of invaluable help when restoring old electronics.
It seems to me that all solutions that are proposed above are more  complicated/ or more expensive than using this very small,
battery powered, device.

But I may miss something.

[cbmeeks]

Is there a special reason why you do not want to use an ESR meter ?

I *want* to use one, just haven't talked myself into buying one yet.  :-)

I know I need one.  But, this is a hobby for me so I have to choose my purchases wisely.  Plus, I would have thought a $10 ESR meter is about as good as a $10 multi-meter.  Probably better than nothing, I suppose.

[vinlove]

I recall seeing someone using the Analogue Multi Meters  for checking capacitors.  They are on Youtube.

[bitseeker]

Is there a special reason why you do not want to use an ESR meter ?

I *want* to use one, just haven't talked myself into buying one yet.  :-)

I know I need one.  But, this is a hobby for me so I have to choose my purchases wisely.  Plus, I would have thought a $10 ESR meter is about as good as a $10 multi-meter.  Probably better than nothing, I suppose.

A $10 multimeter can measure a correct voltage, just as a $1000 one can. Of course, they're different in various ways, but it's much better having a $10 DMM than not having any. Similarly, you can make good use of a $10-20 component tester (and it's good for more than just ESR).

[dannyf]

Like something that would estimate a cap being within 10% of specs or 80% of specs,

No meter could do that, as it wouldn't know the spec of your capacitor. and many times, esr is not spec'd. even when it is spec'd, a capacitor may exhibit different esr under different test conditions.

Much simpler would be for a meter to measure the esr of a capacitor under a given condition, and you make a judgment whether that is good or bad.

[JacquesBBB]

The difference with a 10$ multimetrr is that this one is open source, open hardware
And fully documented
http://www.avrtester.tode.cz/upload/ttester_en.pdf#page1

I would recommend to get one where the IC is easily replacable/programmable
 As a bad move can fry the device.

Through hole with IC on a socket is my preference.

Although they are many with graphic screen, i am not sure they are any better than 2 lines
Lcd display.

[singapol]

Is there a special reason why you do not want to use an ESR meter ?

I *want* to use one, just haven't talked myself into buying one yet.  :-)

I know I need one.  But, this is a hobby for me so I have to choose my purchases wisely.  Plus, I would have thought a $10 ESR meter is about as good as a $10 multi-meter.  Probably better than nothing, I suppose.

To be honest in this hobby it's no fun if you don't have "meter" or instruments to measure electrical properties. You don't learn much! Since your budget is tight even a $10 all in one meter IS better than nothing. With respect to ESR , it is always good and essential to measure for repair and restoration but the final call is with you. The essential thing is knowledge about electrolytic capacitors, how it works and how it is manufactured will help you decide to replace a cap with regards to functionality of the circuit and the environment it is used in for reliability and lifespan.

The thing about E caps is it packs a lot of capacitance in a relatively small package compared to film caps which use the separation or gap between foil/electrode and a plastic insulator but an E cap uses a chemical
solution called an electrolyte to work much like electrolysis so the insulation layer is only chemical in nature
and is only molecules thick if you remember your physics class theory. This is called forming by the cap industry and you need to charge it with electricity or layman's term "powering" it in a piece of equipment.
The often given advice to power your equipment regularly has a basis in this fact.

Industry determination of useful lifespan is by theory and practical testing and interpolation of results
at the full rated voltage and temperature. As you know when there is a current flowing heat is produced
and it is this heat effect that eventually cause the majority of E cap failures. There are learned ways of calculation from the links below. In my honest opinion and also based on my own experience, this knowledge
helps you decide or make your decision to replace and scrap a cap or not. Also do note in consumer electronics the equipment is not used often unlike in a workplace eg. everyday so caps generally last longer.
there is one link from an expert that claims a 50 year lifespan. Ultimately it's your call so the more info you have about caps the better it is.

http://www.powerguru.org/electrolytic-capacitor-lifetime-estimation/

http://www.powerguru.org/influence-of-the-operating-voltage-in-the-useful-life/

http://www.powerguru.org/influence-of-the-operating-voltage-in-the-useful-life/

 http://www.rubycon.co.jp/en/products/alumi/pdf/Life.pdf

http://www.dfrsolutions.com/wp-content/uploads/2014/06/A-New-Method-for-Testing-Electrolytic-Capacitors-to-Compare-Life-Expectancy.pdf

http://bgavinsound.com/Reference/Electronics/Capacitor%20Replacement/Capacitor%20Life%20Expectancy.pdf

50 yrs. lifespan?
https://enphase.com/sites/default/files/Electrolytic_Capacitor_Expert_Report.pdf

[w2aew]

Here's a video I did showing how to measure ESR (only one measure of capacitor "goodness") with a scope and a function generator (square wave generator)

[lisafig]

You can buy a  very simple but very efficient one for less than 10$ on ebay

But @JacquesBBB! can you test the cap in-circuit with these?

[vk6zgo]

Is there a way to get a good estimate on the life of a capacitor without an ESR meter?

Especially something a little more accurate than "it's good" or "it's bad".  Like something that would estimate a cap being within 10% of specs or 80% of specs, etc.

Thanks!

Back in the day, we didn't have ESR meters.
An LCR bridge will tell you the capacitance----if that is very low, you don't have to mess around with ESR, just change the thing!

In later years, I was lucky enough to work where we had a digital LCR meter.
It measured capacitance, inductance & resistance.
As well as basic LCR. It could also determine the "Q" of an inductor, & what it called "loss" in a capacitor (basically another name for ESR).

I also had an adaptor for a DMM which read capacitance.
Over time, I noticed that this device would show a cap with high ESR as being low in capacitance, so that was useful if someone else was using the LCR meter.

Fixing stuff up for work, it took more time to remove, test, & resolder  electrolytic caps with stubby leads than to just fit new ones, with leads you could cut to length, & make nice new solder joints.

Time being money, & electros being cheap, it was my usual practice to change them "scattergun" style.
Being able to test unusual ones was very useful,though.

[MrAl]

Hi,

You can test electrolytic cap ESR by energizing it with a square wave through a resistor and monitoring the current through the resistor and the voltage across the cap with a scope.

The ESR is detectable during the first few instants of the square wave rise (and fall).  That's because ESR is nothing more than a resistance, so you will see the voltage across the cap jump up by a very small amount for a 'good' cap, then it will start to ramp up (or down).
During the first instant, if the voltage rises by 'v' volts and the current is 'i' amps, then the ESR is:
ESR=v/i
and note that the current 'i' does not change much with a good cap because there is a resistor in series with the source and for a small voltage change across the resistor and cap the current is relatively constant.  With very bad caps you'll see 'i' change too, but then the ESR becomes less interesting because you know it's a bad cap right away.

The other thing you can test is the capacitance using:
dv=i*dt/C

but this is done AFTER the first few instants after the voltage across the cap starts to ramp up (or down).
If you take two readings v1 at time t1 and v2 at time t2 (during the ramp) you then have:
dv=i*dt/C
v2-v1=i*(t2-t1)/C

and so you can estimate C as:
C=i*(t2-t1)/(v2-v1)

In the past i have measure several caps after having problems with the devices they had been installed in.  For example a computer powe supply where the electrolytic cap values and ESR changed markedly causing the computer to shut down for apparently no reason.  After the caps where changed, the power supply worked normally again.
I found that both the cap ESR rises very significantly and the capacitance went down a lot.  Both of these things cause problems with the ripple which is detected as either over or under voltage and so the power supply shuts down.

I have some graphic depictions of scope shots too if you want to see them.

[spec]

There are three checks you need to make on an electrolytic capacitor:

• Capacitance value
• Leakage current (at full working voltage, if possible)
• ESR

[rhb]

The BSIDE ESR02 Pro comes in a nice very solid case.  I just put a 20% 220 uF electrolytic in mine.

BSIDE                       229.0 uF   0.31 ohms   Vloss 1.4%

Peak Atlas ESR70     233 uF    0.16 ohm   no Vloss measurement.

The BSIDE was $20.  It also tests resistors, inductors and transistors.  For those I bought a Peak Atlas DCA75 and LCR45.  So probably 15x as much for the Peak Atlas kit.  In all fairness, the DCA75 also provides curve tracing if plugged in to a PC which the BSIDE does not.  Also the BSIDE measures the turn on Vbe whereas the DCA75 measures the saturation Vbe and Vce.

I'd bought the Peak Atlas stuff because it has a good reputation and at the time all the Chinese testers were bare boards without a case.

Both the DCA75 and the BSIDE report the transistor type, leads and hFe.

[bitseeker]

For more about the BSIDE ESR02 Pro including a look inside:

https://www.eevblog.com/forum/testgear/bside-esr02-pro-transistor-tester-(teardown-quick-test)/

[Shock]

Buy or make a cheap Chinese AVR tester if you need something to get you by.

Then later a Bob Parker designed ESR meter with input protection mod for frequent in circuit testing. Then a DER EE DE5000 LCR meter for out of circuit testing at different frequencies.

[rhb]

FWIW the link @bitseeker posted was what led me to order one.   After a $20  DMM, I think it's the best investment you can make.

[bitseeker]

And not to be forgotten, if you have an analog panel meter lying around, build the 5-transistor ESR meter:

https://www.eevblog.com/forum/projects/5-transistor-esr-meter-design/

[Shock]

If you want to use a multimeter as the display use Jay_Diddy_Bs other design.

https://www.eevblog.com/forum/projects/esr-meter-adapter-design-and-construction/