Unusual case with film capacitor from the power supply.

Thread Starter

bob2

Joined Jun 15, 2019
180
Now with the LCR meter in use, you can check any components.
Before, I didn’t check such capacitors, I couldn’t even think that they can change their capacitance. Such a loss of capacitance is typical for electrolytic capacitors.
But when repairing a small power supply unit I discovered that the capacitance of the 470nF film capacitor decreased to 280nF.
Given the tolerance of 20%, this is much more.
Can this capacitor still be considered able to work, or should it be replaced?
What are the reasons for reducing the capacitance in such capacitors?
 

Wolframore

Joined Jan 21, 2019
1,748
metallized film capacitors generally lose capacitance due to loss of the conductive surface. Losses can happen due to over voltage, corrosion, self healing and partial discharge. They have a limited life. Whether you need to change this particular capacitor is how critical this value is and how it's being used. There are some situation where loss of half the capacitance will make little difference and others... where it is part of a RC circuit, where it may make a significant difference.
 

Danko

Joined Nov 22, 2017
941
Often it happens when germetization is damaged and moisture eats metallization on film surface.
EDIT: Too slow typing:(
 
Last edited:

AlbertHall

Joined Jun 4, 2014
10,068
If it's part of the snubber circuit, a low value will result in increased voltage and possible failure of the switching transistor/MOSFET.
 

Thread Starter

bob2

Joined Jun 15, 2019
180
This is the capacitor of the input filter of the power supply unit. Apparently, the loss of capacity did not occur simultaneously, but gradually under the influence of voltage surges.
Do you think that having lost almost half of its capacity, will it effectively perform its function or does it need replacement?
It would also be interesting to know at what minimum capacitance is it still considered to be able to work?
 

narkeleptk

Joined Mar 11, 2019
465
Following your other thread where all testing was done in-circuit, I'm curious... Was this done in-circuit when you got this reading and if so, did you remove it from circuit and re-check it?
 

Wolframore

Joined Jan 21, 2019
1,748
That’s awfully small for input cap. Must be just for filtering high frequency. If you change it I would get one rated for higher voltage.
 

Thread Starter

bob2

Joined Jun 15, 2019
180
Following your other thread where all testing was done in-circuit, I'm curious... Was this done in-circuit when you got this reading and if so, did you remove it from circuit and re-check it?
I did all the measurements again.
The results are identical and do not depend on the test voltage 0.2.0.5.1V.
In-circuit measurements.
10kHz-250nF
100kHz-138nF
Removed from the board.
10kHz-250nF
100kHz-138nF
That is, in fact, the capacitor has lost about 70% of its capacitance.
With LCR Pro1 Plus, it became possible to check the parameters of such capacitors at different frequencies, despite the presence of other components.
In this particular case, there was no influence.
 

Thread Starter

bob2

Joined Jun 15, 2019
180
Also very important secondary parameters are D and ESR for assessing the quality of a capacitor.
In this particular case, these parameters were much worse than the new one.
A multimeter in our case could not measure the capacitance of the capacitor, because it is shunted by an inductor.
Given the combination of the above factors, it is possible to check the filter capacitors for three parameters and make a decision about its further operation.
 

MisterBill2

Joined Jan 23, 2018
6,354
You identified the problem. Thank you.
But. Not indicated ways to test it. What thoughts do you have about this?
There are some ways to check the D and ESR qualitatively, which can tell if they are able to do what they need to do, but not what the actual values are. If the circuit is operating but either the dissipation factor or the ESR are excessive, the capacitor will feel warmer then the surroundings. So a calibrated feel with the fingers will tell if it is OK. But this is not an infallible test, it is just usually right. Measuring the AC component across the capacitor will tell a lot about the ESR, since the ESR is in series with the effective capacitance as well as the source impedance of the AC component of the voltage. Some capacitance meters also display the "D" factor, I do not recall the brand of the one that I used that did provide that reading. There are some quite simple published circuits for evaluating the ESR, often by the application of an AC voltage and measuring the current drawn, the better ones supply a constant AC current and measure the voltage developed.
 

narkeleptk

Joined Mar 11, 2019
465
@bob2, I get the feeling your a salesman or developer for this device, which is fine but I just can't believe that its the ultimate in-circuit testing device. As far as test equipment goes its not a big expense which makes it even more disbelieving that its accurate enough to take it seriously when testing in-circuit. I probably would get one to play with but I don't like tweezers that much and prefer probes so not sure it would be worth it just to have another lcr meter.

So far, from using this device can you give a few cons about it?
 
Last edited:

Thread Starter

bob2

Joined Jun 15, 2019
180
There are some quite simple published circuits for evaluating the ESR, often by the application of an AC voltage and measuring the current drawn, the better ones supply a constant AC current and measure the voltage developed.
Yes, but this has to do with electrolytic capacitors, not film capacitors. Or maybe you have such a circuit?
 

Thread Starter

bob2

Joined Jun 15, 2019
180
some of those look like poly film... they can corrode and have various other issues...
Yes, in several situations, I became convinced that film capacitors could lose capacity for various reasons.
That's just recently, there was a situation with the film capacitor 1uF.
The capacitance is 558nF. That is, it has decreased by almost half. At the same time, the D of the capacitors with a reduced capacity and the new capacitor are almost the same. But the ESR differs significantly, several times.
Apparently the reason for this is due to the different processes that occur in the capacitor. Somewhere more prevails D, somewhere ESR.
 
Last edited:
Top