smart relay contacts mysteriously burning up

[thekman]

208 3phase w/ neutral service. 3hp 3ph 208-230v motor rarely if ever used to charge hydraulic cylinder. Smart relay control, typical setup used in 100's of locations with similar or identical scenario.

2 outputs (dry contact relay) wired to run traffic lights (120v 60W bulbs) from of 1 leg (through smart relay output) and return to source neutral. Technician verfied .5a on both hot and neutral to light that's on.

4 identical systems, 2 sets of 2 within a few hundred yards of each other, all burning up the plc outputs' contacts.. ..JUST the 120vac (voltage in = L3-N) output that is on is burning up. These (nor any of the others) are being turned on or off, it's pretty much just sitting there on all the time.
The location is within a mile of the ocean, so I'm thinking the air has something to do with it, and maybe a combination of the salty humidity mixed with some harmonics on the neutral, although not sure what might generate harmonics if nothign is sharing source.....?
To my knowledge, there's nothing else on the source, and the only other things on the 120vac circuit are some solenoid valves and a contactor, yet none of which are used.
It's as if the made contacts are just somehow building up contaminant, pushing the contacts apart just enough to arc the 120.
Any thoughts?

 

[thekman]

oh, relay contacts rated for 250vac, 10a.

 

[Comcokid]

Are the bulbs incadescent, halogen, or one of the more modern LED types?

 

[thekman]

they are standard incandescents 120v 60W

 

[VEBill]

It might be worth having a proper failure analysis done on the failed contacts. This would confirm or refute various theories. Contamination should be visible. Overcurrent should leave clues.

In general, there's a possibility that the root cause will eventually turn out to be something not mentioned, and not derivable from the information provided.

 

[thekman]

I tore apart one of the through hole spst relays inside the unit and found corrosion and burnt contacts. When it's still hooked up, you can still hear the relay firing and the contacts moving, but they don't make, from the corrosion.
Probly should have mentioned that too huh. sorry about that.

 

[MikeHalloran]

Is this PLC or whatever mounted in a proper box, or is it just nailed to a piece of plywood, or a telephone pole, or what?
Indoors? Sheltered at all?



Mike Halloran
Pembroke Pines, FL, USA

 

[Skogsgurra]

The inrush current of a 120 V 60 W incandescent bulb is rather high. It is well known that traffic lights eat relay contacts if nothing is done to prevent it.

The easiest and most effective remedy is to preheat the filament by letting around 20 percent of rated current flow when relay is open. It is being done all the time.

Saves the bulbs too.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[thekman]

it's mounted on dinrail in a metal enclosure which is inside a bigger metal enclosure. There is no extreme temperature involved, yet I wonder about the salty humidity.

 

[IRstuff]

Perhaps you need to replace them with something that's sealed?

TTFN

FAQ731-376

 

[ScottyUK]

Solid state relay - suitably over-rated - might be a reasonable solution. Better still would be to prevent salty contamination entering in the first place.

What is you ventilation like? Enclosures within enclosures suggest it is not very good: the rating of your relay contacts will be based on some expected heat rejection to ambient, but if they can't get rid of heat then failure will occur. Heat accelerates corrosion too, doubling the rate for every 10°C rise.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[thekman]

I agree with the potential for heat to be a factor, however, I've got many identical setups used similarly wihtout incident.
The inrush for an incandescent would be a great clue, but the output burns up after a period of being on for several months.
I have a feeling there is some information I'm not getting.

 

[jraef]

I would also look carefully at the operating sequence. For example if there is some combination of events that can cause the relay to chatter (on-off rapidly), that will eat contacts in very short order. You may have a command signal of some sort that does not have enough hysteresis built-in and under just the right circumstances, i.e. a limit switch in just the right position with a vibration, or a float switch that flutters in turbulence, it gives a command to the smart relay that creates that rapid cycling. One easy way to tell if you can't see anything is to associate a counter to that output so that every time the output is commanded, the counter increments. Go back after a period of time and see if the count value matches what you expect it to be. I have found a couple of issues like that. Sometimes it's hard to see the problem if you are not right there when it happens, but if you can prove that it happens, you know you are missing something.

"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[BrianE22]

Are you using any coil voltage spike suppression? Some methods cause the coil current to decrease too slowly and the contacts to release too slow causing burning of the contacts.

 

[potteryshard]

Becasue this issue is confined to one small area, I'd look into the power source. Any relay which is held on continuously that experiences sustained low voltage is likely to heat up the coil, and arc damage the contacts.

If contaminated moisture were the problem, it seems like it would be more prevalent on the de-energized, and consequently unheated, relays.