Doubt electrolytic capacitor

[cecil87]

Is there any study or analysis of electrolytic capacitor manufacturers showing how many minutes or hours are needed to completely reform electrolytic capacitors assembled in the circuit of electronic devices and these electronic devices were stored unused for a long period?

1 hour (60 minutes) is enough or not?

 

[D.A.(Tony)Stewart]

Reducing Failure Risk with Capacitor Reforming - Quad Plus

Spare capacitors are not enough to ensure quick recovery in the event of a failure. Periodic capacitor reforming allows for immediate use and quick recovery.

www.quadplus.com

"
A rigorous schedule for reforming capacitors should be incorporated in your facility maintenance schedules. This will prevent damage to spares and intermittently idle plant systems. Manufacturers of drive and other power supply equipment that uses large electrolytic capacitors include capacitor reforming schedules in the applicable documentation.

This may be annual in some cases.

--- Updated Oct 22, 2023 ---

Reforming may take up to an hour and is often done with the rated voltage and a large series R to limit the risetime of voltage and allow 10x the rated leakage current.

e.g. 1mF 50V Leakage
Kytocera REH series (10-120V) I≤0.01CV or 3μA, whichever is greater

1e-3* 50 = 50 mA ( This implies a worst case leakage decay time of 1 second = C * V/I. ) If I did this right.

Kyocera's accelerated high-temp aging leakage test is as follows; with an 80'C rise in ambient temp DC leakage is permitted to be no more than 5 times the rated leakage after 2 minutes charge up.

No method is given but current limiting the charge rate is expected.

 

[cecil87]

in my case it is without resistors and a controlled circuit

I just connect the electronic device to the electricity supply, but in this case I don't know how long it will take.

 

[D.A.(Tony)Stewart]

in my case it is without resistors and a controlled circuit

I just connect the electronic device to the electricity supply, but in this case I don't know how long it will take.

It ought to be a CC supply with CV to rated voltage. I would use the rated leakage current and expect the time it takes = dt = CdV/IL . I would use a series R that matches 1 to 5x the rated leakage resistance and required Pd so that the reforming and leakage reduction occur at the same time without shorting out.
What does your controlled circuit do?

 

[BradtheRad]

I think I read 1 mA is adequate, and can take several minutes to charge up.

An electrolytic with deteriorated electro-chemical barrier develops low ohms through its electrolyte. It tends to draw excessive current when plugged into full power. If it doesn't go pop then you're lucky it reformed quickly and sufficiently to survive. I started using a tape recorder that had been idle for years. I heard a pop and then a bubbling sound. Inside I found a capacitor burst and oozed clear gel.

 

[cecil87]

my doubt for electronic devices that have electrolytic capacitors soldered on the PCB

 

[KlausST]

in my case it is without resistors and a controlled circuit

I just connect the electronic device to the electricity supply, but in this case I don't know how long it will take.

my doubt for electronic devices that have electrolytic capacitors soldered on the PCB

So:
* You have a PCB with soldered capacitors
* in parallel there is a load that draws unknown current
* you don´t want to desolder the capacitors
* you don´t have control circuit, nor a series resistor.
* you just want to power it on ...

Isn´t this
1) ignoring the problem, (while you know the problem exits)
2) refusing a solution ... (because you don´t want to modify / solder)
3) but still expecting it to work?

How? It takes a magician.

Klaus

 

[FvM]

Realistically, it's not always possible to design protection means for long storage periods of built-in electrolytic capacitors into a product.
Epcos/TDK application note says you should be aware of up to factor 100 increased leakage currents. They'll return to normal values after 1 hour. Problems can be expected in this cases
- leakage currents cause capacitor overheating
- asymmetrical leakage currents of series connected capacitors involves imbalance and overvoltage

In the latter cases, a dedicated reforming procedure may be required, e.g. current limiting and applying a voltage ramp through variac for inverters and switching power supplies.

Most products don't seem to need a similar procedure. We know however that electrolytic capacitors are a common reason for product failure after 10 or 20 years of operation, mostly because they dried out. Hard to say if extended storage periods also play a role.

 

[cecil87]

If I desoldered all the electrolytic capacitors from all my electronic devices, there are many of them, it would take a long time and I don't have a variac or resistor, only electronic devices with electrolytic capacitors mounted in the circuit

 

[FvM]

The risc to damage capacitors by desolderng is higher than damaging it by just applying power to the device. That's the point of being realistic.

 

[cecil87]

How many minutes or hours should all electronic devices be turned on so that the electrolytic capacitors do not fail due to disuse?

 

[BradtheRad]

One easy thing to try is hook up an incandescent light bulb in the current loop, in order to reduce voltage going to your equipment. Incandescents bring fewer surprises compared to modern types. A light bulb gives an obvious visual cue as to how much power is going through it.

1) Say your device draws 7 Watts house voltage, then a 7W bulb reduces voltage by half.

2) Test by applying 5 seconds of power, then turn it off. Did your equipment show any distress?

3) If things seem OK then move up to a higher Watt bulb (or higher Watt load, or add bulbs in parallel.)

4) Repeat with increasingly higher Wattage.

Tests should be brief, a few seconds. The reason is because electronic devices are unpredictable when run below their specified voltage. Oscillators may stagnate instead of switching, or regulators may draw extreme current in an effort to supply sufficient power.

 

[cecil87]

In my case I only have electronic devices that use electrolytic capacitors, in my case I just connect the device to 220v
So no electrolytic capacitor manufacturer says when the interval for turning on the device is at least how many minutes or hours?

 

[cupoftea]

So no electrolytic capacitor manufacturer says when the interval for turning on the device is at least how many minutes or hours?

Sorry but unfortunately not....you just have to use "engineering judgement", kind of thing. Or literally have 20 years spare to do tests on large nuimbers of sample ciruits.

 

[D.A.(Tony)Stewart]

In my case I only have electronic devices that use electrolytic capacitors, in my case I just connect the device to 220v
So no electrolytic capacitor manufacturer says when the interval for turning on the device is at least how many minutes or hours?

Current limiting is essential. If you can't CC or constant current limit to 5x the leakage specs and the current to reduce such that it is in CV mode.

New shelf life may be 2 yrs but decline with wear. Reforming time may take 2 hrs or less if the leakage effective Resistance has risen to a stable reasonable value.

 

[cupoftea]

I have never heard of anyone , ever, putting in a circuit into a product, that would detect when the product had been disused for 2 years +...and then would auto switch in some circuitry to do the reformation of the electrolytic caps. It just doesnt happen.
If the customer doesnt use the product for 2 years, then suddenly starts using it, then its classed as the customer doesnt really need the product in the first place, so why take any special precautions for them?...it is tough luck to them. They get the magic smoke..and they buy a new product....nobody puts reforming cctry into any product.

 

[cecil87]

i tune on all electronics devices 6 months 1 hour (60 minutes) is good or bad for electrolytic capacitors? 1 hour sufficient?

 

[cupoftea]

..well it sounds better than doing nothing.
And yes, TBH, sounds like a good idea all round.
I'd be amazed if an hour wasnt long enough...but i suppose should check the official docs of the cap manufacs

 

[danadakk]

I have never heard of anyone , ever, putting in a circuit into a product, that would detect when the product had been disused for 2 years +...and then would auto switch in some circuitry to do the reformation of the electrolytic caps. It just doesnt happen.
If the customer doesnt use the product for 2 years, then suddenly starts using it, then its classed as the customer doesnt really need the product in the first place, so why take any special precautions for them?...it is tough luck to them. They get the magic smoke..and they buy a new product....nobody puts reforming cctry into any product.

Makes one wonder for space and planetary exploration what designers do.....or what one would
do for long life oceanography data loggers and sensors.....


Regards, Dana.