Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

donate to eham
   Home   Help Search  
Pages: [1]   Go Down
  Print  
Author Topic: Electrolytic cap life measured in hours?  (Read 1162 times)
KF6IIU
Member

Posts: 293




Ignore
« on: May 11, 2008, 02:21:15 PM »

I am browsing through the Digikey catalog looking for new HV electrolytic caps for my oscilloscope, and I am seeing the lifetime of most electrolytic caps specified as 1000 to 2000 hours, with some rated as a few thousand more.

Less than three months of continuous duty?

I know electrolytic caps are usually the first thing to go in a circuit, and I'm sure the spec is for maximum temperature and voltage and minimum leakage and ESR, but is the average lifetime of electrolytic caps only a few thousand hours? Or is this some kind of 99.9 percentile figure? If caps fail under their ESR and leakage spec in just a few thousand hours in a critical design, how do designers work around this?
 
I guess that's why (other than the maximum standard voltage rating for electrolytics being 450V) the voltage doubler caps in my scope are four identical ones in series :-)
Logged
KB1LKR
Member

Posts: 1897




Ignore
« Reply #1 on: May 11, 2008, 02:50:36 PM »

It may be MTBF (Mean Time Between/Before Failures/Failure) at rated voltage and max rated temperature. For an electrolytic the life will increase roughly linearly with decreasing voltage and increase  exponentially with decreasing temperature. Ripple current is also a factor.

Note that if it is indeed MTBF and if the failure distribution function is exponentially distributed, 63.2% should have failed at that time. Failure may not be absolute either, it could be they no longer meet published specs -- you'd have to review th emfg's data sheets to be sure though.  
Logged
W5FYI
Member

Posts: 1054




Ignore
« Reply #2 on: May 11, 2008, 06:03:27 PM »

I think you should look at the manufacturer's data sheet to get the facts. I looked at one manufacturer's product, and the short life was for a ripple test at 85°C (185°F) for a change in capacitance of ±20%. The life test was for the same temperature and tolerance conditions. What surprised me was the shelf life test. It was only 500 hours!

Coincidentally, I was reading an article earlier today that explained how electrolytic capacitors can degrade simply by sitting on the shelf. The electrolyte slowly dissolves the insulating layer on its plates. Chemists have a general rule that most chemical reaction speeds roughly double for every 10°C increase in temperature, and 85°C is about six doublings from a warmish 25°C or so operating temperature.

According to the article's author (C.G. Cunningham, Popular Electronics, July 1966), an electrolytic that has been sitting on the shelf for a long time will lose its ability to hold its rated voltage. However, electrolytics can be reformed by applying voltage and slowly bringing it up to the rated figure, something akin to what photographers had to do to their early electronic flash units which hadn't been operated for a while.

Thanks to your question I learned something new today.

FYI - Stew
Logged
KF6IIU
Member

Posts: 293




Ignore
« Reply #3 on: May 11, 2008, 09:00:51 PM »

I'll look at some data sheets to see how they test these things.

Although it's nice to have a junk box full of electrolytics - it's use 'em or lose 'em.

Logged
W8JI
Member

Posts: 9749


WWW

Ignore
« Reply #4 on: May 12, 2008, 07:02:41 AM »

The hours of life for an electrolytic normally require multiplying by a formula. They are never (so far as I have seen) a directly used number.

There is a formula that enters core temperature of the capacitor and voltage and uses the hours rating as part of the equation.
Overall it is a lengthy process because you have to know the dissipation in side the capacitor, ambient temperature, and even terminal voltage at certain ripple currents.

The result is also not the life of each capacitor, but a mean time before failure. Different manufacturers use different formulas and they may have different formulas for different parts inside one manufacturer, so you really need to read the manufacturer's application sheets.

73 Tom
Logged
G3RZP
Member

Posts: 8143




Ignore
« Reply #5 on: May 12, 2008, 09:29:21 AM »

I have a power supply happily working with electrolytics dating from 1946.........but YMMV.

I suspect we tend to run at higher ripple currents and temperatures these days
Logged
KB1LKR
Member

Posts: 1897




Ignore
« Reply #6 on: May 12, 2008, 10:50:29 AM »

At a quick look I see Cornell Dupilier computer grade electrolytics rated 4000h at rated temp (+85°C) and rated voltage, w/ failure as: delta C +/-20%, ESR 200% of limit, DCL 100% of limit, so there you go. Keep'm well below max DCWV and as much cool as possible any they'll be good for a lot longer.

See also:  http://www.cde.com/tech/selectinvcap.pdf
"Selecting Capacitors for Inverter Applications" for further discussion of life and other issues.
Logged
KC2MMI
Member

Posts: 812




Ignore
« Reply #7 on: June 12, 2008, 04:59:09 PM »

"rated 4000h at rated temp (+85°C) "

That 85C rating might be the key here. The rule of thumb I was taught is to derate everything by 2x per 10C change in temperature.

So from 85C to 65C to 45C to 25C, which would be a warm room, you've got a 2^3 (16-fold) change in life expectancy and 4000 hours becomes 64,000 hours or seven years of continuous 24x7 operation.

If I did my math right. Anyway, that's why cooling fans are CRITICAL. Proper design parameters mean keeping things cool enough to hold against your bare skin--even if we can't build computer CPUs that way yet.<G>
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!