On Monday, December 7, 2015 at 12:18:18 PM UTC-5, Spehro Pefhany wrote:
On Sun, 6 Dec 2015 07:37:11 -0800 (PST), rangerssuck
wrote:

On Saturday, December 5, 2015 at 11:41:05 PM UTC-5, Jon Elson wrote:
Larry Jaques wrote:


So, are you going to tell me that this is a coincidence, or are these
thieving bastids designing this crap to fail quickly, then making an
equal amount (to the sales profit) by servicing their dead crap? I'm
strongly guessing the latter. sigh
No, I think the evil *******s is pretty appropriate! On the other hand, it
may have just been ONE crappy relay. I bought a bunch of parts to put arc
suppressors on all the relays, but never did install them. So, there are
still 11 relays of two different types perking along merrily for the last 3+
years! I was expecting the rest of the relays to start failing, but that
hasn't happened.

Jon

It may not have been a bad relay, but may have been that the one relay that went bad was being asked to do something out of spec.

Life of a mechanical relay is typically only 50-100,000 operations at
full rated load. Often even less for motor rated relays. That's
completely specified, sometimes there is a curve showing typical life
at lower than rated current, and you can easily test for that (at 2
seconds per operaton 100,000 operations takes a couple days, and you
can test for 1,000,000 in a few weeks) . At only 10 operations per
hour in a product, 24/7, a 100k operations relay will last less than 2
years.

Here's one that's rated for only 25,000 operations at rated current
*resistive* load.
https://www.omron.com/ecb/products/p...pdf/en-g5q.pdf

Now as an engineer working for an appliance manufacturer, do you
recommend a $5 relay (longer life or heavily derated) instead of a 50
cent one to make it last 10-20 years rather than 5 average, knowing
that will increase the retail price by $50+, or do you use the cheaper
part? Do you make the same decision for all the **other** parts that
have a definite life span, and if so will you still have a job- or
will your product be affordable enough to sell in the required
quantities. It's not really evil, just an economic decision.

It seems like there ought to be solid state replacements for common mechanical relays. I've never had a failure of an SSR, and you can get them with built in snubbers and zero-cross switching. One would think it would be cheaper to mass produce SSRs vs mechanicals.

I've seen lots of failures of solid state relays- they almost always
fail 'on' and they do so more-or-less randomly rather than mechanical
relays that have a definite life. They also produce a lot of heat
requiring heat sinks, are more expensive and often less reliable in
the bathtub part of the life curve. They have notoriously poor
tolerance for overcurrent and overvoltage. There has been little
improvement over the years in SSRs or mechanical relays, though both
have gotten cheaper in real terms.

It's possible to make SSRs more robust by using much larger overrated
semiconductors and special I^2T fuses, but there's those pesky
economic trade-offs again..

--sp

--
Best regards,
Spehro Pefhany
Amazon link for AoE 3rd Edition: http://tinyurl.com/ntrpwu8
Microchip link for 2015 Masters in Phoenix: http://tinyurl.com/l7g2k48

Ease up, dude. Jon did NOT mention mechanical failure. He DID mention that the contacts were burned badly.

Now, as for the fifty-cent vs five-dollar decision, by the time the washer hits a retail store, the price difference might be $1,000 vs $1,020. I KNOW that given the choice, I'd spend the extra twenty bucks. Larry J. mentioned that in the next post.