Appliance failures - relays

even using relays with quick-connect termination or with the switch terminals at the opposite end to the coil terminals would help.

Nice clues, thanks. The above two show a broken fillet, lifted, with mushroomed edges. I take that as proof of horizontal motion in the through-hole, and possibly of the component swaying from side to side atop the board.

The first picture shows solder preferentially piled up on the left side, which probably means something too.

Those are distinct from the other failure pictures in this thread, and in fact I think it's pretty clear that there cannot have been lateral motion in the Acura-pic failures, otherwise the edges of the broken fillets would be bent, mushroomed, or otherwise deformed. They weren't. They were thin, clean, and sharp. Virgin.

That looks almost like a poorly wetted joint, but could also be a break under axial load.

Here's another picture of failed automotive joints, also of relays(!):

Unfortunately for the car-owner, the poster doesn't seem to have noticed the incipient cracks forming on the same pins of the relay to the left of the unit marked. The lower left pin's joint looks like it's already cracked clockwise from about 4 o'clock to about 1 o'clock.

(I wouldn't have noticed this except for having seen such cracks with my own eyes--that's what they look like.)

Cheers, James Arthur

GM had a similar failure on their turck wiper motors. the connector pins would develope cracks around the solder joints and the wiper would fail to work properly. Fixed mine by bending the pins onto the pcb, there were no mechanical mounting points on the connector. (much like the relay)

Cheers

Build something that lasts five years, and these days the consumers will think it lasted well, and buy another from you. Build it to last ten, and you've simply done yourself out of a sale.

Sylvia.

Having recently shopped for an appliance, I'd gladly pay extra for something if I was sure it was good.

I'm hot-rodding a fridge right now with extra insulation. In writing a spreadsheet to estimate the thermal performance you really get to see the design trade-offs, i.e., how adding insulation here or there saves energy, but not enough to ever recover the cost.

If it were up to me the walls would be thicker, the compartments smaller, and energy consumption about 30% lower. (I think simple mods will get me close, repaying the investment in a couple years.) (Of course it looks nerdy having insulation on the outside, but fortunately, for me that's a bonus.)

I try to design stuff to last as long as reasonably possible. Electrolytics are often limiters.

Cheers, James Arthur

Makes sense. Thanks. I usually go solid-state, but I've fielded lots of relay designs with no regard for orientation. Next time, I won't.

s.e.d. is cool.

Cheers, James Arthur

I now see the case for pins with flanges that mechanically rest on the pcb. I always assumed that set insertion depth, but it would also help mechanically.

All these examples remind us never to use solder for mechanical support.

Cheers, James Arthur

That philosophy has now migrated over to the industrial sector.

We yanks like to use NEC stuff, generally built very well however, in the last recent years corporations are getting cheap and now we're seeing IEC cheap cheap junk appearing in the common area of industrial applications. Needless to say, the "CRAP" does not hold a candle to the old NEC stuff for the most part. Some of the common items on the list are so poorly made compared to what the old timers are used to that many of them are embarrassed to even tell any one they're using IEC.

This goes along with modular electronics being made in these so called other places in the world, like inverter drives for example. We now have a a slew of a particular drive brand in use that has earned its nick name "The blue flame", cause they are commonly known to blow out the guts of the Igbt modules in welding slag all over the cabinet. If it isn't that, the caps are going bad! We also call them bean counter drives. 400 Volt land drives reconfigured software only to operate on 480 systems thus barely making the max voltage rating, one little tickle and it's all over.

Jamie

I've had to repair an industrial diswasher that regularly burnt out relays because, although they were used within their design ratings, they were mounted with the gaps horizontal.

Beware of some pre-punched relay panels for wall-mounting enclosures which have the relay socket fixing-hole pattern the wrong way around.

Any crud that gets in there soon falls out, their relays also had dual contact points, which would mitigate the effect of minor crud infestations.

I have a similar problem with my 2000 ford ranger occasionally failing to start. No click from the solenoid, just total silence. Replaced the clutch switch and interchanged the relays with no help. Sometimes I just hold the key turned for a few seconds and it magically starts. The ford agency said the problem was probably not the starter or solenoid, so maybe it's a bad solder connection to the starting relay?

-Bill

--- news://freenews.netfront.net/ - complaints: snipped-for-privacy@netfront.net ---

I've heard of some people using freezers as fridges, by adding an additional thermostat. I'm not sure whether that is optimal as the gas pressures and compressor size are perhaps not ideal for fridge temperatures, but the insulation is likely a lot better than the manufacturers can get away with in a normal fridge.

Whilst you are at it you can slightly modulate the thermostat setting based on when you want to consume most of your power. This helps where the utility will buy your solar power during the day, but for a third of the price that they charge when they sell it back to you at night. If you can find a phase change material that freezes around 4 deg C and put a tank of it at the back of your fridge, you could do most of your cooling at the time of day that you choose, and even based on the generation and other appliances' consumption at the time.

Chris

a modern 300 liter A++ rated fridge can be down to something like 100kWh a year, that doesn't make much of a difference for a normal household

-Lasse

This is the mother of those inventions (and my inspiration as well):

He mods a chest-freezer to run at refrigerator temps, claims 36kWhr/year.(!)

Chest conversions don't lose their cold air when you open, don't trade humid air and condense as much water, and have other technical advantages.

But, they're not as convenient for accessing your food, and I'd still need a separate freezer, which nearly negates the energy savings. Not to mention the added floor space needed.

So far, a single 1/2" layer of foil-backed polyisocyanurate on the sides and top of a conventional top-freezer combo unit gets a 4 deg C advantage (measured) on those surfaces, costs $8, reduces estimated consumption about 20%, and saves $10 / year. That's pretty easy.

Slightly more trouble would be increasing the passive coil spacing from the refrigerator's back, so I could squeeze in a sheet of polyiso there. That would save an estimated $4.48 annually.

Filling the freezer and fridge door compartments has great potential. Loosely packing plastic grocery-store vegetable bags into another of same makes fat insulating bats, food safe, of high insulating value. That saves another $6 a year. Together, my spreadsheet model estimates those save 36%.

I probably won't do all that--I'm just playing for the easy improvement. But I'll probably do most--it's free. Just throwing a comforter over the top and left/right sides would be well worth it.

Yes, the better insulation is the active ingredient.

How about a phase-change material that freezes around 0 deg C?

Energy Star-rated 21 ft^3 freezer/fridges in the U.S. run about 420kWh / year.

Cheers, James Arthur

what does NEC stand for this this case?

I meant to say NEMA.. I must of been drunk! :)

Sorry..

Jamie

You're absolutely correct, that's what you get when you're knee high in that crap all day

Jamie

Hiya Bill!

Could be. The Acura's "main" relay powers the ECU and the fuel pump, so Ye Olde Acura would crank, but not catch, or catch, vibrate, then kerklunk.

Schematic:

Al's Internet shows examples of bad solder joints in Fords, vibration failures, they say.

They're hard to spot. Here's a pic:

Happy hunting!

James

On a sunny day (Thu, 21 Aug 2014 16:22:20 -0400) it happened Neon John wrote in :

Right, I agree, circuit makes no sense whatsoever.