ESR, how low is too low?

[zeus]

I got my ESR meter last Friday and have been comparing old lytics with new ones to get an idea on numbers.
I found a 27yr old ELNA 47uf, 160v, 85c, that measured 0.19 ohms and should be around 1.0 ohm.

Why is this lytic's ESR so low?
Is this considered a bad lytic?

[PeteS in CA]

Re: ESR, how low is too low?

Hmmm ... I'm a little surprised that ESR would be specified for an 85C cap. That 1.0 ohm will be a maximum rating, of course. Two Qs might account for the another part of difference. Is the rated 1.0 ohm ESR at 120Hz? And what is the measurement frequency of you ESR meter? If the rating is at 120Hz and the meter is measuring at 10KHz-100KHz, that would pretty much explain the difference.

All that said, did you check to see if the cap is shorted?

[PCBONEZ]

Re: ESR, how low is too low?

It might also be partially shorted from overheating.
That's why you check with an ohm meter as well.

.

[zeus]

Re: ESR, how low is too low?

Correct PeteS, 85c lytics do not have specified ESR ratings. I got the 1.0 ohm number from a
chart on the web that said this chart was a "rough guide only". Considering ESR is a
rather important spec, Why manufactures don't put it on data sheets for all series
lytics I'll never know. I guess they want end users to reinvent the wheel every week.

The chart didn't say what freq the ratings were measured.
My meter only measures at 100KHz.
This cap does not seem to be shorted.

What reading should I get for a new 47uF, 160v, 85c lytic cap tested at 100KHz?
Is there a good reliable ESR chart for GP lytics out on the web?
Is there a formula I should be using?

[PCBONEZ]

Re: ESR, how low is too low?

GP lytics are not generally used in applications where ESR matters all that much which is why it's not in the data sheets.

Or the inverse.

If ESR matters they don't use GP caps.

.

[zeus]

Re: ESR, how low is too low?

PCBONEZ, I ohmed this ELNA cap and got 19.36 Mohms using a Fluke 289.
So its not shorted.

[zeus]

Re: ESR, how low is too low?

GP lytics are not generally used in applications where ESR matters all that much which is why it's not in the data sheets.

Or the inverse.

If ESR matters they don't use GP caps.

I understand that, I'm trying to learn when a GP cap has bad ESR numbers.
I need to compare good numbers with bad. I found a ELNA 47uF, 160v, 85c
lytic cap that seems to have an ESR number that is too low (0.19ohms).

[zeus]

Re: ESR, how low is too low?

GP lytics are not generally used in applications where ESR matters all that much which is why it's not in the data sheets.

Or the inverse.

If ESR matters they don't use GP caps.

So, you're saying GP lytics don't need to be tested for ESR at all?
How do you properly test a GP lytic to see if it's good or bad then?

[PCBONEZ]

Re: ESR, how low is too low?

I agree that 0.19 ohms sounds wrong but:

You have no point of reference.
-
There is no data for it.
-
You can't compare it to other manufacturers or series or caps because they will all be different just as they are in low ESR series and makes.
-
The only way to get a point of reference is to compare it to several ELNA 47uF, 160v, 85c from the exact same series. (Actually any in the same series with the same can size should be nearly the same.)
-
Lacking that, you can test all you want but the numbers don't really mean anything.

.

Also, ESR should go up as caps age, not down.

Are you sure that meter is operating properly?

.

[PeteS in CA]

Re: ESR, how low is too low?

using a Fluke 289

I have DMM envy.

[PCBONEZ]

Re: ESR, how low is too low?

"DMM envy" ???

Is that like ....... oh never mind.

.

[PeteS in CA]

Re: ESR, how low is too low?

Fluke's 289 is their newest high end DMM. Put one of those and a Fluke 199C Scopemeter in a brief case and you practically have a portable engineering lab. Think of it as techno-macho.

[PCBONEZ]

Re: ESR, how low is too low?

No, I meant more like 'probe' envy.

.

[zeus]

Re: ESR, how low is too low?

Ok, so I will assume this cap is bad. I guess I'll be testing every old and new lytic cap that crosses
my path and log it's measurements to come up with some sort of average to compare to.

I received my 289FVF a week and a half ago, and I can't put it down (little kid in a candy store).
I did spend the extra $70 for the extra goodies just to see what the Fluke View software was like.
I must say the 289 is night and day compared to my 179. I now have some components that test ok,
that tested bad on the 179 do to the 289's accuracy. It was often hard to get a 0.27ohm resistor to
measure in tolerance with the 179.

I never thought I'd spend $600 for a DMM, but hey, I sold some crap on Craig's List, and walla,
I'm in DMM heaven. LOL!

[PCBONEZ]

Re: ESR, how low is too low?

Now that there is a nice meter.
NICE.

.

[willawake]

Re: ESR, how low is too low?

damn you have more pens than me also

[PCBONEZ]

Re: ESR, how low is too low?

Back to the original question.
You have to look at what the cap is doing in the circuit.
-
Also have to consider that a cap's ESR changes with frequency. The data tables only show ESR at one frequency. You have to get into the graphs and/or math to figure it out at other frequencies.

If it's being used to pass ripple out of a DC circuit as on a motherboard then the lower ESR is at the expected ripple frequency the better off you are.

If it's being used to tune something to some frequency then the ESR at 'some' frequency may be critical to keeping the overall circuit working in spec but I think ESR matters much more as the frequency goes up into the GHz and higher ranges. I don't recall ESR ever coming up when I went through Navy 'radio school' as a precursor to other electronics training. (But that was 27 years ago too. Old technology and old memory.)
.