Measuring ESR of small capacitors

[mdijkens]

For a low noise power supply I'm building I need low ESR (<0.5ohm) capacitors of 1uF and 10uF
I have one of these cheap chinese component testers with ESR measurement that works quite well for bigger caps, but I have the feeling that the technique used to measure ESR is not suited for small capacitance?
All cheap ceramics have ESR's of 2 to 5 ohm although I read everywhere that the ESR should be very low. I've bought tantalum pearls but they are even higher?
I've been reading for hours and hours to figure out how to measure these smaller caps, but it's only getting more complex for me.
I've also tried with my function generator and oscilloscope but I think the frequency of the squarewave should be much higher for these caps but even at 8MHz (max I can go) the ESR is still high.

Do I have bad or inferior caps or do I measure it wrong?

[Kleinstein]

Measuring the ESR of small caps is difficult, as the measurement needs to be done at higher frequency. So the cheap ESR / capacitor tester might very well not be suitable. Even if the HW might work, it would likely need an extra calibration.

It would be unusual to get MLCC with an ESR in the 5 Ohms range, though there are special caps with intentional higher ESR.

With a square generator and scope, there is no need to go high in the frequency. One needs fast transitions and suitable probes. So it might depend on the quality of the generator. Also the resistor used as a reference impedance (e.g. the 50 Ohms output impedance of the generator) need to be of good quality or well characterized. With the somewhat nonlinear ceramic caps, the definition of ESR might get complicated and might not work very well.

[MosherIV]

Thanks Kleinstein, always so knowledgeable.

No here are my contributions:
ESR is not a characteristic that (many) manufactures use, they use Q factor. I cannot remember exactly what this means or signifies (I am sure someone will fill us in).
The point I am making is that you are unlikely to find official ESR values from manufactures.

ESR was something that some people discovered was a good indication of whether electrolytic capacitors had dried out or been damaged which would often lead to PSU failures.
It was relatively easy to knock up simple testers for it.
It has become this mythical thing that a lot of inexperienced new comers obsess about.
The use of low ESR capacitors is mainly important for PSU output smoothing, especially where there is going to be high ripple currents (correct me if I am wrong).

Most LCR bridges from the main manufacturers still do not measure ESR, they measure the Q factor instead (correct me if I am wrong).

If your capacitors are new and have not been sitting around for 10 years or more then they are probably OK to use.
Whether they are the right ones to use - another matter.

[mdijkens]

Thanks for the replies!

The ceramics I tested are the ones I normally use: These cheapest brown/orange disks (through hole) varying from 1pF to 10uF.
For my linear power supply the LT3083 specsheet demands low ESR (<0.5ohm) caps of 1uF and 10uF.
Will these cheap disc through hole ceramic caps have that low ESR ?

[mdijkens]

Anyone?

[tigr]

Ceramic capacitor 1uF.

[mdijkens]

According to the display that is 1046uF ??

[David Hess]

Most LCR bridges from the main manufacturers still do not measure ESR, they measure the Q factor instead (correct me if I am wrong).

They either measure D or Q but it is easy enough to convert between them or convert either to ESR or close enough anyway.

My ancient impedance bridge measures down to a D of 0.001 at 1kHz or roughly 0.16 ohms of a 1uF capacitor or 0.016 ohms of a 10uF capacitor so barely good enough for mdijkens' application.  It can see the difference between 0.1uF Z5U, X7R, and C0G capacitors but the absolute ESR values it returns are probably marginal.

A better option would be a real LCR meter like a DE-5000 which can operate at up to 100kHz and supports 4-wire measurements.

If I had to hack something together quickly, I would use a function generator, oscilloscope, and some parts to make a manual impedance bridge but with more time, there is probably a better way.

[oldway]

[MrAl]

Hi,

If you put a resistance in series with a capacitor and drive it with a square wave and measure the voltage across the cap, a pure capacitance will show an exponential curve that is rounded except where the square wave transitions from low to high or from high to low.  At those points the curve will change from concave up to concave down or vise versa, but during that tiny portion of the wave the amplitude will not change, so here are only two parts to the wave.

Now if you add a second small resistor in series with the cap and measure across both the small resistor and cap rather than just across the cap, that forms a voltage divider, so that when the input wave transitions we'll see an immediate voltage increase that does not take any time to change, followed by the typical capacitor curve as before.
Since this small resistor represents the ESR, that is the new wave shape we see with a given ESR equal to the small resistor value.  The ESR can then be determined by the amplitude of the part of the wave that goes straight up when the square wave drive transitions.

At higher frequencies the inductance will start to show an effect too though.  Now we have a resistor and an inductor in series with the cap, and when we look on the scope w see the effects of both the inductor and the ESR during the transition.  The inductor adds the new effect where the wave shoots up quickly and then may take a little time to come back down.  That will interfere with the reading we get from the ESR somewhat.
Depending on the ratio of the component values and the frequency, we may or may not get a good reading on the ESR this way.  This is when we have to switch to a method that tests for the energy loss in the circuit and deduce the ESR from that, because a pure inductance and pure capacitance does not absorb energy while a  pure resistance does.

That's probably as deep as you need to get but then there is radiation.  If the circuit radiates significant energy then that starts to look like increased energy loss, and then the measurement also starts to depend on frequency.

So there's no perfect rule, except for one, and that is where the cap is actually placed into the end application circuit and the entire circuit tested as whole.  This means that the cap will be operating at the required frequencies of the application and that's the best test.  We could then look for signs of higher than normal ESR, or in some cases even lower than normal ESR which could mess up some circuits like boost regulators.

After all is said and done though, i would think for capacitors 1uf and above you should be able to get usable results using the square wave and scope method because the ratio of capacitance to inductance should be high enough.  If you have doubts, we can roll some numbers and see what we get.  The circuit is simple enough, a resistor in series with a small inductor and small resistor and in series with a pure capacitance, and take measurements across the inductor and small resistor and capacitor.

[oldway]

The problem of ESR has become crucial with the appearance of switch mode power supplies.
Before, no one was worried about the ESR.

So, for someone like me who lived the era before the switch mode power supplies, this ESR phobia seems ridiculous and demonstrate that the person does not understand anything about an electronic circuit.

We even arrive at a totally absurd situation of wanting to measure the ESR of capacitors in a vacuum tube circuit ... for me, it deserves a visit to the psychiatrist ... it is similar to some ramblings of the audiophools, the miraculous audio cables of $ 1500 and other bullshit of the genre.

In general, we can just measure the ripple ... if the ripple is within the specifications, no need to measure the ESR.

In switched mode power supplies, the same principle of measuring ripple value applies.

One thing we often forget is that an electrolytic capacitor has a maximum RMS current that should not be exceeded ... nobody seems to remember this.
Yet, it was a problem we were often faced with filters of fast SCR's inverters projects.

[The Electrician]

In general, we can just measure the ripple ... if the ripple is within the specifications, no need to measure the ESR.

One thing we often forget is that an electrolytic capacitor has a maximum RMS current that should not be exceeded ... nobody seems to remember this.

This thread is not about the heating of a capacitor caused by the ripple current in the ESR.

It's about the effect of ESR on stability and performance in low dropout regulators.

[ahbushnell]

In general, we can just measure the ripple ... if the ripple is within the specifications, no need to measure the ESR.

One thing we often forget is that an electrolytic capacitor has a maximum RMS current that should not be exceeded ... nobody seems to remember this.

This thread is not about the heating of a capacitor caused by the ripple current in the ESR.

It's about the effect of ESR on stability and performance in low dropout regulators.

ESR in filter capacitors can impact control loop stability in power supplies.  So it's just about heating or ripple. 

Andy

[The Electrician]

In general, we can just measure the ripple ... if the ripple is within the specifications, no need to measure the ESR.

One thing we often forget is that an electrolytic capacitor has a maximum RMS current that should not be exceeded ... nobody seems to remember this.

This thread is not about the heating of a capacitor caused by the ripple current in the ESR.

It's about the effect of ESR on stability and performance in low dropout regulators.

ESR in filter capacitors can impact control loop stability in power supplies.  So it's just about heating or ripple. 

Andy

The OP started this thread to ask about aspects of the power supply he is building using an LT3083.  I don't think heating or ripple in the output capacitor is of concern since the voltage there is almost pure DC.  And, yes, as I already said, the ESR has an effect on stability, which is really the only reason to be concerned about the ESR in this application.

[oldway]

The OP started this thread to ask about aspects of the power supply he is building using an LT3083.  I don't think heating or ripple in the output capacitor is of concern since the voltage there is almost pure DC.  And, yes, as I already said, the ESR has an effect on stability, which is really the only reason to be concerned about the ESR in this application.

We are in the beginners section and the title of the subject is: "Measuring ESR of small capacitors"

It is therefore normal to speak on all aspects of ESR measurement of capacitors and related subjects.

The beginners section has a didactic purpose, not just to answer a particular case.

[tigr]

According to the display that is 1046uF ??

No.The capacitor is in a circuit with other capacitors.

[MrAl]

The problem of ESR has become crucial with the appearance of switch mode power supplies.
Before, no one was worried about the ESR.

So, for someone like me who lived the era before the switch mode power supplies, this ESR phobia seems ridiculous and demonstrate that the person does not understand anything about an electronic circuit.

We even arrive at a totally absurd situation of wanting to measure the ESR of capacitors in a vacuum tube circuit ... for me, it deserves a visit to the psychiatrist ... it is similar to some ramblings of the audiophools, the miraculous audio cables of $ 1500 and other bullshit of the genre.

In general, we can just measure the ripple ... if the ripple is within the specifications, no need to measure the ESR.

In switched mode power supplies, the same principle of measuring ripple value applies.

One thing we often forget is that an electrolytic capacitor has a maximum RMS current that should not be exceeded ... nobody seems to remember this.
Yet, it was a problem we were often faced with filters of fast SCR's inverters projects.

Hi,

I wish it was that simple.

ESR in the output filter cap has been recognized as a factor in the stability of a regular boost converter some years ago.  Some values of ESR that are TOO LOW or too high can cause the output to oscillate either continually or for too many cycles.  There is stuff on the web able this.  Quite a while back i created a Windows application for analyzing the boost converter and ESR and related, and found this to be true.
Also, the (positive) linear regulators with PNP output (for example) can have this problem too so for some regulators the data sheet mentions a range of cap values that are acceptable.
So it's not really a phobia it's a real issue.

Measuring the ripple just tells us if the design is stable at that measuring point and passes a given specification.  It doesnt tell us that changing the load or input voltage or something else will cause an oscillation or not.

It takes a lot of deep analysis to figure out why the boost converter ESR is so picky.  The simplest explanation is that a pole in the right hand plane of the complex frequency plane appears and must be dealt with as a proper range of allowable ESR.

[oldway]

Honestly, a project that depends on a factor as variable in time (AND TEMPERATURE) as the ESR of an electrolytic capacitor is a bad project and must be avoided. 

Learn how to make projects that are not going to ruin your business into support and guarantee.

[The Electrician]

For a low noise power supply I'm building I need low ESR (<0.5ohm) capacitors of 1uF and 10uF
I have one of these cheap chinese component testers with ESR measurement that works quite well for bigger caps, but I have the feeling that the technique used to measure ESR is not suited for small capacitance?
All cheap ceramics have ESR's of 2 to 5 ohm although I read everywhere that the ESR should be very low. I've bought tantalum pearls but they are even higher?
I've been reading for hours and hours to figure out how to measure these smaller caps, but it's only getting more complex for me.
I've also tried with my function generator and oscilloscope but I think the frequency of the squarewave should be much higher for these caps but even at 8MHz (max I can go) the ESR is still high.

Do I have bad or inferior caps or do I measure it wrong?

So far no one has given any practical advice for your question.

As you have realized, the cheap Chinese ESR meters don't work well for low value capacitors.  They mostly use the technique described here: http://geoffg.net/Measuring_ESR.html

They apply a square pulse and measure the impedance, and that requires that the reactance of the capacitor at the measuring frequency be less than the ESR.  This doesn't work well for small capacitors because their reactance can be greater than the ESR, so that this type of meter isn't measuring the ESR, it's measuring the impedance.

The traditional method is to apply a sine wave of voltage and measure the resulting sine wave of current.  The phase angle between the applied voltage and the resultant current allows calculation of the ESR.

Professional instruments to do this tend to be expensive.  However, in the last few years, low cost instruments using the sine wave method have appeared on the market.  One of the favorites on this forum is the DE-5000.  If you search for that on the forum, you'll find many threads about it.  It can usually be had for about U.S. $100 on eBay.

That would solve your problem, and it's a worthwhile instrument to own.

[G0HZU]

The other way to do it would be with a clean sine wave test signal. Feed this to the capacitor via a sense resistor and use a decent (AC Vrms) DMM to measure the voltages around the sense resistor. From these voltage readings you can calculate the ESR and the capacitance from basic theory. It's not going to be as good as the very best $$$ LCR meters but I'd expect it to compete well against the modern $100 meters and it costs nothing if you already have the signal source and a decent DMM

[tigr]

And it's even better to make an ESR meter for $ 3.
 

[oldway]

You can not only take the ESR into consideration, a capacitor also has a series inductance.

The impedance of a capacitor varies with frequency as Dave had shown in the video on bypass capacitor tutorial.

More info in this document:
https://www.illinoiscapacitor.com/pdf/Papers/impendance_dissipation_factor_ESR.pdf