Reforming Electrolytic Capacitors

[Whoops]

have heard stories of companies putting lytics on slo-bake for a while, don't know what temp, certainly do not want to boil them dry,

looks like quad 8 used phillips also inside this diy job we just popped the lid on while checking on the input iron,>

Yes, a lot of equipment and high end brand used Phillips capacitors. Neve gear for sure was full of them

 

[ln76d]

I can measure capacitance with my Fluke 179 meter , it measures from 1nf to 10.000uf

ESR I have a cheap chinese meter, it's not amazing but I can get some rough values, I'm never sure if I should trust it or not, but its what I have

http://hkbbs.leowood.net:88/mini_read.asp?id=9463898&page=1&property=0&ClassID=0

Both are sufficient.  You don't need extra occuracy, it's not about matching caps, so if there would be something wrong with caps you will get rather abnormal results with both meters.

 

[Whoops]

Hello,
first of all thank you all for the guidance.

So I had fun today testing some of the capacitors

Some of the Philips capacitors have the tolerance written, some don't.
The ones that have the tolerance written say -10% to +50%, so I took that as a reference.
Not all of them were Philips, I had also BC, Eiko and ITT

MEASUREMENTS

_______________________________________________________________________________________________________________
VALUE: 4.7uf 63V PHILIPS  (batch of 10)
Expected Tolerance: 4.23 to 7.05
Reference ESR:  2

Capacitance measurement FLUKE: from 8 to 9.8 (higher capacitance than tolerance)
Capacitance measurement CHINA: from 6.3 to 7.2 (most within tolerance but close to upper limit)
ESR measure: from 2 to 6.5 (most in the 4 region)

REPORT: Both Capacitance and ESR seem higher than they should. ESR much higher than reference
_______________________________________________________________________________________________________________

VALUE: 6.3uf 64V PHILIPS  (batch of 10)

Expected Tolerance: 6.12 to 10.2
Reference ESR:  2

Capacitance measurement FLUKE:  7.78 to 15.6 (most in the 9 region within tolerance)
Capacitance measurement CHINA: 6.88 to 9.65
ESR measure: 2.85 to 4.4 (some caps keep increave ESR while measuring)

REPORT:  Capacitance seems within tolerance but close to upper limit. ESR much higher than reference

_______________________________________________________________________________________________________________

VALUE: 10uf 40V EIKO Germany - BIPOLAR  (batch of 10)

Expected Tolerance: 9 -15
Reference ESR:  1.7 - 2

Capacitance measurement FLUKE:  10.5 - 12 (within tolerance)
Capacitance measurement CHINA: 10.1 - 11.61  (within tolerance)
ESR measure:  0.15 - 0.29

REPORT: Capacitance Good. ESR much lower than reference, these are EIKO Bipolar caps could these ones be lower ESR capacitors?

_______________________________________________________________________________________________________________

VALUE: 15uf 40V PHILIPS  (batch of 10)

Expected Tolerance: 13.5 - 22.5
Reference ESR:  1 - 2

Capacitance measurement FLUKE: 23 - 104 (most in the 25 region)
Capacitance measurement CHINA: 21.6 to 65.4 (most in the 22 region)
ESR measure: 0.45 - 1.83 (most of them measure 0.5 ESR)

REPORT:  Capacitance higer than tolerance, one cap measured 104uf. ESR values lower than reference

_______________________________________________________________________________________________________________

VALUE: 22uf 40V PHILIPS  (batch of 10)

Expected Tolerance: 19.8 - 33
Reference ESR:  0.8 - 1

Capacitance measurement FLUKE: 30.3 - 36.4 (higher than tolerance)
Capacitance measurement CHINA: 20.1 - 25.3 (within tolerance)
ESR measure: 1.45 - 3.69

REPORT:  If China meter is right Capacitance within tolerance. ESR much higher than expected

_______________________________________________________________________________________________________________

VALUE: 100uf 25V PHILIPS  (batch of 10)

Expected Tolerance: 90 - 150
Reference ESR:  1

Capacitance measurement FLUKE:  116 - 127
Capacitance measurement CHINA: 70.5 - 78
ESR measure: 0.92 to 2 (most ESR's measured 1.5)

REPORT:  If Fluke 179 is right Capacitance within tolerance. ESR higher than expected

_______________________________________________________________________________________________________________

VALUE: 100uf 40V PHILIPS  (batch of 10)

Expected Tolerance: 90 - 150
Reference ESR:  0.3 - 0.5

Capacitance measurement FLUKE:  130 - 152
Capacitance measurement CHINA: 85.5 - 106.6
ESR measure:  0.7 - 1.76

REPORT:  Capacitance within tolerance. ESR higher than expected

_______________________________________________________________________________________________________________

VALUE: 1000uf 40V PHILIPS  (batch of 10)

Expected Tolerance: 900 - 1500
Reference ESR:  0.04

Capacitance measurement FLUKE:  1077 - 1100
Capacitance measurement CHINA: 1037 - 1072
ESR measure:  0.04

REPORT:  Capacitance and ESR all GOOD. Fluke and China measurements become closer

_______________________________________________________________________________________________________________

VALUE: 1500uf 25V PHILIPS  (batch of 4)

Expected Tolerance: 1350 - 2250
Reference ESR:  0.05 - 0.1

Capacitance measurement FLUKE:  1495 - 1631
Capacitance measurement CHINA: 1496 - 1620
ESR measure:  0.07 - 0.08

REPORT:  Capacitance and ESR all GOOD. Fluke and China measurements are the same

_______________________________________________________________________________________________________________

VALUE: 3300uf 25V PHILIPS  (batch of 4)

Expected Tolerance: 2970 - 4950
Reference ESR:  0.05 - 0.1

Capacitance measurement FLUKE: 3095 - 3566
Capacitance measurement CHINA: 3110 - 3570
ESR measure:  0.04

REPORT:  Capacitance and ESR all GOOD.

_______________________________________________________________________________________________________________

VALUE: 4700uf 50V  (batch of 4)

Expected Tolerance: 4230 - 7050
Reference ESR:  0.05

Capacitance measurement FLUKE:  4850 - 4890
Capacitance measurement CHINA: 4790 - 4800
ESR measure:  0.11

REPORT:  Capacitance GOOD. ESR higher than espected

_______________________________________________________________________________________________________________

VALUE: 6800uf  35V    ITT  (batch of 3)

Expected Tolerance:  6120 - 10200
Reference ESR:  0.03 - 0.05

Capacitance measurement FLUKE:  7632 - 7874
Capacitance measurement CHINA: 7540 - 7790
ESR measure: 0.02 - 0.04

REPORT:  Capacitance and ESR all GOOD.

_______________________________________________________________________________________________________________

 

[emrr]

Am I remembering there's one end of life mechanism that can make electrolytic value go high? 

 

[JohnRoberts]

Thank you Gene.

What does it mean "zero cap volts"?

Probably means at initial start up, so cap is fully discharged to 0V,  limiting full supply voltage to deliver just 1 mA. (A 15v supply with 15k resistor in series would deliver 1mA to a zero voltage cap.  As the cap charges up, current falls proportionately.

1mA may be low for larger caps. I seem to recall capacitor leakage specs that were a function of C x terminal voltage, + a little. 

Start messing with it and you'll figure it out... They may not even need to be reformed if quality parts.

JR

PS: I never heard of cooking caps and since time and temperature are what cause them to loose electrolyte that doesn't seem like the best idea. Maybe a pass/fail test to look for any that leak when heated, indicating a bad bung seal, unless they already dried out.

BTW over the years I have taken a lot of caps apart when troubleshooting production problems, at Peavey we even has an xray machine so I could get a crude idea of the insides without sacrificing the parts, but nothing like taking one apart to get to really know them. 

PPS: I had one interesting batch of bad caps where the swaging tool that attached the lead to the capacitor electrode at the capacitor factory was broken or out of adjustment so the cap leads were making weak/intermittent contact. The only way to find that was by taking a few apart.

 

[JohnRoberts]

Am I remembering there's one end of life mechanism that can make electrolytic value go high?

Most common end of life mechanism for typical electrolytic caps is low C and high ESR from loss of electrolyte.

JR

 

[Rob Flinn]

Your results don't surprise me.  When I test old caps it's nomally the low value capacitance ones that are most flaky.

 

[Whoops]

Thanks JR.

So as you pointed out some of the Caps, measure higher Capacitance than the rated value specially smaller caps.
ESR is also higher than the expected value, for the exception of the larger caps 1000uf and higher.
Although ESR is high is not as higher as the readings I get from the stressed and used caps I've been replacing in Power supplies.
I replaced all the caps in 2 Mackie PSU's today, and most of the caps measure 1/4 of the capacitance value and the ESR value skyrocketed.

Caps of 1000uf or higher measure fine in terms of capacitance or ESR.

Could the Caps that have an higher capacitance reading than rated mean that those caps are Leaky and need to be reformed? would that improve the ESR reading also?

I was reading this in another website:

"A leaky Cap (as in, capacitor with a high leakage current… not one that is physically leaking electrolyte) will usually trick your ESR meter to show lower ESR than what the capacitor may have. So if the cap has gone high ESR, your meter may not show it and you might end up putting a faulty cap back in service. To avoid this, check the capacitance of the cap. If it is higher than 20% of its specified capacitance, it is likely leaky and it is time to reform it. "

Also it's strange how I get really different readings with the Fluke 179 meter and the China Cap meter,  the China meter always measures lower. The diference between the 2 is much higher in caps under 1000uf, above that value the 2 meters measure almost the same reading.
I don't know which of them is correct but the China meter readings are always closer to the rated value.

 

[JohnRoberts]

Thanks JR.

So as you pointed out some of the Caps, measure higher Capacitance than the rated value specially smaller caps.

I did?

ESR is also higher than the expected value, for the exception of the larger caps 1000uf and higher.
Although ESR is high is not as higher as the readings I get from the stressed and used caps I've been replacing in Power supplies.
I replaced all the caps in 2 Mackie PSU's today, and most of the caps measure 1/4 of the capacitance value and the ESR value skyrocketed.

Caps of 1000uf or higher measure fine in terms of capacitance or ESR.

Could the Caps that have an higher capacitance reading than rated mean that those caps are Leaky and need to be reformed? would that improve the ESR reading also?

I don't know, high capacitance is not a failure mode I am aware of. Note some electrolytic caps are +80% -20% tolerance, do you have spec sheets for those caps?

I was reading this in another website:

"A leaky Cap (as in, capacitor with a high leakage current… not one that is physically leaking electrolyte) will usually trick your ESR meter to show lower ESR than what the capacitor may have. So if the cap has gone high ESR, your meter may not show it and you might end up putting a faulty cap back in service. To avoid this, check the capacitance of the cap. If it is higher than 20% of its specified capacitance, it is likely leaky and it is time to reform it. "

may be a quirk of how some capacitance meters work... perhaps not really high capacitance, just leaking current and fooling the capacitance meter that probably looks for rate of change when driven by a unit current.  We didn't have capacitance meters when I was a bench tech..

Also it's strange how I get really different readings with the Fluke 179 meter and the China Cap meter,  the China meter always measures lower. The diference between the 2 is much higher in caps under 1000uf, above that value the 2 meters measure almost the same reading.
I don't know which of them is correct but the China meter readings are always closer to the rated value.

Old saying, man with one meter knows the measurement, man with two meters never knows true result. 

JR

 

[Whoops]

Old saying, man with one meter knows the measurement, man with two meters never knows true result.

LOLOLOL



Well Put!
I will not forget that.

Well I will just use China cap meter from now on to measure capacitance because it makes my NOS caps look better.
Done

 

[emrr]

My mention of a high capacitance being an end of life indication may be a wrong memory, I feel like I've only read that once, and recently, with a description of the 'why'.   

Like JR said, you go back far enough and tolerance could be something like -20/+100%, and a lot of really old equipment measures nearly 2x with caps like that. 

What you REALLY need is 3 meters. 

 

[Whoops]

My mention of a high capacitance being an end of life indication may be a wrong memory, I feel like I've only read that once, and recently, with a description of the 'why'.   

Like JR said, you go back far enough and tolerance could be something like -20/+100%, and a lot of really old equipment measures nearly 2x with caps like that. 

What you REALLY need is 3 meters.

LOLOLOLOL
3 meters would be quite a party.

I did some searching on this Philips axial caps, and when there's info/specs or description on tolerance it always says -10% / + 50%

So let's take that as the reference

 

[ln76d]

It surprise me Guys, because the oldest cap i measured always had higher capacitance than stated tolerance.
Like in Whoops measurements. I newer found lower capacitance always higher.

Whoops look for specs of both meters. Chinese can be  more accurate than Fluke. Especially with lower capacitance values.
Usual multimeters have worse tolerance for capacitance measurements.

 

[Whoops]

It surprise me Guys, because the oldest cap i measured always had higher capacitance than stated tolerance.
Like in Whoops measurements. I newer found lower capacitance always higher.

Whoops look for specs of both meters. Chinese can be  more accurate than Fluke. Especially with lower capacitance values.
Usual multimeters have worse tolerance for capacitance measurements.

So that could be because it has high leakage?

Basically if a cap measures lower capacity than it should we know it's bad, if it measures higher than tolerance we dont know if it's good or bad?

 

[ln76d]

Basing on my measurements (i really often measure caps) i always was sure that capacitance go much higher when condition of capacitor is worse due to aging. Even with really old caps from the 50's (i have bunch of old Siemens, Hydra and many others) i never measured lower capacitance than tolerance shows. Always is higher. Doesn't matter which meter am using.


After a little thinking a little edit

I never used totally out of specs caps but there's a question how capacitance would look after spending few hours in the circuit?
Someday i will probably make that experiment but now have no time.
If you have some option for test, use the worst specs cap in some circuit after some time discharge it and measure again.
There's probability that readings will be really different.

 

[Whoops]

I see.
Might do that test in the future.

As for Bad caps, I recently replaced all the Electrolytic Caps in the PSU's of 2 different Mackie Mixers, as I told in a previous post.

One Mixer a big Onyx one has heating problems in the PSU area, so Caps blowing or gone bad it's a common problem.

The other fixer had voltage problems, the leds were flickering. Probably the caps used originally were low quality, or the voltage rating were on the limit to save costs , or too much heat, or just time made the caps go bad.

I measured the replaced caps,  for the exception of some that measured fine in capacitance and ESR values, all the other ones had much smaller capacitance than rated and really high ESR. None of them measured higher capacitance than rated or above tolerance.

 

[ln76d]

I measure the replaced caps,  for the exception of some of them that measured fine in capacitance and ESR values, all the other ones had much smaller capacitance than rated and really high ESR. None of them measured higher capacitance than rated or above tolerance.

That should be valid. With desoldered (used) old caps rather capacitance is always lower.
I had in mind only NOS caps with higher capacitance

 

[Whoops]

Whoops look for specs of both meters. Chinese can be  more accurate than Fluke. Especially with lower capacitance values.
Usual multimeters have worse tolerance for capacitance measurements.

Tried to find specs just for fun, on the Fluke 179 meter specs they state + - 1,2%  accuracy.

On the Chinese Meter I have no specs but the similar models I find online state 0.5% for values under 200uf:

Accuracy                         
  ±0.5%    (<2000uF)
  ±1.0%    (2000uF)
±2.0%    (20mF)

200.0pF (0.5%),
2nF, 20nF, 200nF, 2µF, 20µF, 200µF (0.5%),
2,000µF(2%)
20,000µF(4%)


So this might explain the China meter reading lower than the Fluke under 1000uf, and the values being similar above that

 

[emrr]

here's a veer.....can you slow bake the absorbed moisture (noise) back out of NOS carbon resistors?   

 

[JohnRoberts]

here's a veer.....can you slow bake the absorbed moisture (noise) back out of NOS carbon resistors? 

I am not aware of that being a thing... Would that be like "excess" noise (noise that is modulated by current flow).

It seems like something you could experiment with (or just use better quality modern resistors).

JR