Is it ok to reverse polarize an electrolytic cap upto 1V ?

[Kevin.D]

Has title says really. Is it OK to reverse bias a polarized Aluminium electrolytic by 1 volt ?.
I cant get a definite answer from my web search some pages say they are OK up to
2 volts reverse .What effect will it have on  long term performance and life span .?

Thanks

[sacherjj]

I'll let those more knowledgeable than me answer this one.  My understanding is that even 1V will start the redox process that will remove the oxide layer and make the cap short out.

I have heard that it is ok to put two electrolytic caps together in series, with similar polarity leads touching, you have a non-polarized cap at 1/2 capacitance.  To take advantage of the higher voltage of the series combination, you need some high value resistors in parallel with each cap.  You can also purchase non-polarized electrolytic caps, which have oxide layer on each side of the foil.

[Psi]

This page http://my.execpc.com/~endlr/reliability.html says

"Don't subject aluminum capacitors to more than about 1 volt reverse polarity at 25C, and more than about 0.4 volt at 85C"

So one would assume that the reliability is fine as long as you don't break that rule.
However i guess you could question the reliability of the webpage


I also found this pdf http://electrochem.cwru.edu/encycl/misc/c04-appguide.pdf

Reverse voltage
Aluminum electrolytic capacitors are polarized and must be
connected in the correct polarity. They can withstand
reverse voltages up to 1.5 V. Higher reverse voltage can
cause failure by pressure build up and rupture of the capacitor’s
safety vent structure. Non-polar and semi-polar
devices are available that can withstand reverse voltage.

[EEVblog]

Yes, I was of the understanding that 1V was the maximum rule of thumb for reverse voltage without any lasting effect. But that will certainly be device/manufacturer/process dependent though.

Dave.

[Mechatrommer]

"Don't subject aluminum capacitors to more than about 1 volt reverse polarity at 25C, and more than about 0.4 volt at 85C"

my understanding is dont do it at all. if do, then my understanding is "not so good" practice.

[nukie]

What are you trying to do? Save yourself trouble, they are polarised for a reason! There's another type of unpolarised called bipolar electrolytic. Use those!

[T4P]

What are you trying to do? Save yourself trouble, they are polarised for a reason! There's another type of unpolarised called bipolar electrolytic. Use those!

A hell lot cheaper to make your own bipolar electrolytics .
Maybe trying to cut corners ?

[grumpydoc]

Why do you want to reverse polarise it?

If you're going to put a constant DC across it then just turn it round so that it isn't reverse polarised.

If you're going to put AC across it with a DC offset but where the AC+DC voltage can swing down to -1V you should be OK if the average potential is positive.

If you're going to put AC (+/- DC offset) across it but the average is going to be -1V then turn it round.

I suspect more info about what you're going to do with said capacitor, what value and what voltage waveforms you expect to put across it is needed.

There are also non-polarised alternatives, as others have said, up to at least several hundred micro-farads

[T4P]

Nah , easier to roll your own bipolars .

[SeanB]

Capacitors used in switching supplies in base drive do get reverse bias, and fail in a few years. If you have the need to have random reverse voltage better to use the 2 in series approach, at least then you will have some improvement, but better is to get a proper bipolar, even though they are not common and are much bigger with only a few values and voltages.

[Fraser]

If making your own bipolar capacitor using the crude but apparently effective two polar capacitor in 'series opposing polarities' method you should ensure that the total ESR is not going to effect your design. If a design is 'mission critical' always use the correct component for the job and not a compromised solution. If you are just prototyping or using a component outside design spec just for tests..... no worries....play to your hearts content 

In the audio world (not audiophiles) it is not uncommon to see the 'series opposing polarities' method employed in crossovers of cheap systems. It works for audio and the failure rate appears low in this non critical application.

Aurora

[Zero999]

I would think it depends on the capacitor's voltage rating. -25V across a 400V electrolytic capacitor is probably no worse than -1V across a 16V capacitor.

[sacherjj]

I would think it depends on the capacitor's voltage rating. -25V across a 400V electrolytic capacitor is probably no worse than -1V across a 16V capacitor.

The issue with reverse voltage is that the oxide layer on the aluminum on only on one side.  (Bi-polar electrolytics have both sides oxide coated.)  The voltage that it takes to start redoxing this oxidation layer is around 1 V, regardless of the voltage rating.  The voltage rating is a property of the dielectric strength.  However, going past 1V on any voltage electrolytic will start to redox the cap and eventually make it short out.

[IanB]

I would think it depends on the capacitor's voltage rating. -25V across a 400V electrolytic capacitor is probably no worse than -1V across a 16V capacitor.

I doubt it. The "electrolytic" in the capacitor description describes how an electrochemical forming process is used to produce the insulating barrier by electrolysis of the electrolyte. The electrochemical potentials required for the forming reaction to take place or be reversed are a function only of the materials and electrolyte used. With an aqueous electrolyte any adverse potential difference above about 1 V is going to produce destructive effects. (This is why lithium ion cells have a non-aqueous electrolyte. An aqueous electrolyte would decompose at the ~4 V they operate at.)

[Greate]

How about doing an actual experiment to check it out. I know capacitor can be checked if they become faulty but is there a way to check if they are degrading?

Sent from my A100 using Tapatalk

[SeanB]

If you can measure ESR then simply take 10 identical units ( preferably small ones if they go bang) and reverse bias then for 5 hours via a current limiting resistor at various voltages. Start at 1V to 5V in 1V steps. A simple way is to use 4 100R resistors in series to make a divider across 5V and connect each capacitor in reverse via the 1k resistor to each tap. After an hour measure the voltage across each 1k resistor to get the leakage current in each cap, and again after every hour. At the end check capacitance again to compare with the original value and check the ESR compared to original. This will give you an idea as to how they respond.

Do not do this with Tantalum capacitors, they have a nasty habit of letting smoke out, unless they are a glass frit seal unit, when they go bang with a loud noise and send a slug across the room.

[T4P]

Do not do this with Tantalum capacitors, they have a nasty habit of letting smoke out, unless they are a glass frit seal unit, when they go bang with a loud noise and send a slug across the room.

Yep tantalums are tantalum dry slugs , they are so susceptible to blowing up that if you use them to filter power their lack of ESR will blow them up (Ironic , it's very main attractive point blows itself up)
I do not know what you mean by glass frit seal unit (?)

[SeanB]

Look at solid tantalum slug capacitors for the glass sealed units. High reliability aerospace units. Kind of the best units you can get. Mostly made by Vishay.

www.vishay.com/capacitors/list/product-42073/

[T4P]

Right , glass sealed ones are great .
However wet slug and dry slug will blow into your face a day sooner or later if they were used in place of a alu cap where they were supposed to be , power supply filtering ! ( If you don't add enough resistance )

[SeanB]

I used those in power supplies in aircraft, where they worked well for many years. They never popped, except that they eventually died from old age, as they used a cheaper seal instead of the glass seal, and used a rubber compression seal which eventually failed and leaked ( admittedly after 40 years and hundreds of hours at 100C plus) which made the caps go high resistance. Capacitance was still in tolerance, but ESR was in the kilo ohm range. The only failures were when I put them in the wrong way round and turned on the power.