These problems with wet electrolytes  should be a signal to anybody 
designing for reliability: avoid
wet electrolyte caps unless a limited llifetime is a workable solution. 
I try my best to use Tantalum
caps, even though the cost is twice as much.

I have a client that uses a 3300uF @ 35VDC radial used as a filter for a 
24VDC power supply in
a robot the does pipeline inspection. These caps have to be replaced 
every 2 years and sometimes
more often, as the environment gets hot (70C).  Of course there is no 
alternative.

There is one possible alternative- seal (pot) the wet cap. That prevents 
the electrolyte from leaking
but will  also  prevent it from venting safely in the event of a 
failure. But this seems  to work OK for
small caps used for audio purposes. You can also seal the venting end 
with  two coats of fingernail
polish, which adds a touch of class to an otherwise drab PCB...

--Bob

Russell McMahon wrote:

>>> Aluminium electrolytics should be run near their rated voltage so 25v
>>> is fine.
>>
>
>> Interesting, could you elaborate?
>
>
>> I guess it has to do with formation/maintenance of the dielectric, 
>> right? Just
>> wondering how much of a problem this really is, and what 
>> consideration to
>> give to it in a design.
>
>
> An aluminium "wet" electrolytic needs some potential to maintain 
> formation of its plates. At elevated temperatures such a cap will die 
> in far less than its rated lifetime if stored with no voltage applied. 
> ie although a cap may have a 2000 hour rated lifetime the actual 
> lifetime is usually far longer as it is run at lower than specified 
> temperature. but STORE the cap at 85C with no power applied and it 
> will last rather less than 3 months as it will dry out!
>
> I imagine how much voltage is needed to maintain formation depends on 
> design and maufacturer's skill and experience. It may well be that 5v 
> is enough for a 35v cap. But possibly you could expect a reduced 
> lifetime.
>
> The matter is not at all clear. Some web sources sugfgest lifetime 
> INCREASESt lower voltages.
>
>
>        RM
>


-- 
Note: To protect our network,
 attachments must be sent to
 attach@engineer.cotse.net .
 1-866-263-5745 Tollfree

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist