Capacitor 'Life'

What temperature was the lifetime cited at? Electrolytic life is very temperature-dependent - this is from some Vishay literature:

Thanks John - the lifetime quoted in the Vishay advert was the top 105c which is pretty extreme for most electronic components. I hadn't found the reliability pages before and they are very interesting - especially when you see a standard aluminium electrolytic capacitor has a 'life' of only 3.6 years at 65c or less than 1 year at 85c. This is mainly due to the electrolyte drying out.

From the maintenance point of view it is obviously essential to keep up effective cooling for any electronic equipment which has electrolytic capacitors fitted. I wonder how many plant maintenance staff are aware of the limits and how much down-time has resulted over the years.

As you say...it is very important to keep electronics cool...

but it is also another thing to factor this intentionally to electronics to place a life expectancy...take computer motherboards for instance...My PC died because the capacitors dried out. I first removed them and found other rail voltages shorted to ground...so i gave up. When I observed the capacitor placement, i was astonished...they were all sitting right next to the CPU heatsink! I initially thought which idiot designer would put the caps there...the hottest part of the board...but then i thought...actually...it's clever design.

Just this morning I repaired a Mac G5 computer by replacing some bulging capoacitors. Unfortunately, the G5 series of Macs, now about 6 years old, are infamous for failing due to bad capacitors, and those capacitors come from at least two different, normally reliable, vendors.

As a colleague commonly reminds me, it is not uncommon for a 50-year-old mechanical device to continue to function normally. Why can't that be true for electronic devices? Of course 50 years ago, a 1 Farad capacitor would have required a forklift to move. Now I can place several in my pocket...

And you have to keep it in vertical position to avoid spilling of chemical.

I believe most quoted 'life' figures are wildly pessimistic as the manufacturer doesn't want to be bombarded with claims. We use small DC motors with a quoted brush life of something silly like 5,000 or 50,000 hours. I've never seen one of 'em with a worn out brush yet.
I ran some of my own test when we first started using 'em, the y bedded in nicely and that was it, no probs.
Del
_{(Of course the unwritten law of quality states that the phone will ring in the next half hour with a spate of brush problems)}

Even idle in storage also deteriorates the life of capacitors.

One other thing to remember is lifetime is not that the cap will fail. End of life is when the capacitor no longer meets it's spec. i.e. a 20% capacitor will be more then 20% from it's nominal rating. So depending on design this could be a major issue or not much of a concern. Electrolytic capacitors start to detoriate immediately after manufacture, there are many things that can effect the rate of detoriation and temperature is one of the major ones.

Shawn

Surely 'end of life' is when you define end of life with respect to the design in question, as generally one over rates both voltage and capacitance considerably...?

Perhaps I misspoke, I would agree end of life would be defined as the end of usability of design. I meant a rating of X number of hours does not necessarily mean that the device will fail in that number of hours. That would be more like a MTBF rating. The hours rating for caps is how long they will still meet their data sheet specifications.

Shawn