SMT750 AVR relay weld fault

[Anonymous user]

I got this error after replacing the RB48 battery after 6 years of trouble-free service. Internet chatter says this is a common problem, often when replacing the battery, and the relays aren't very robust. I'm fully able to replace the AVR relay if that is in fact the problem, but it would be a lot easier if I could find the circuit diagram for the SMT750 and/or specs and/or industry part number for the relay. Can anyone help?

[Anonymous user]

Well, the relays test fine on the bench, but since they're only $0.64 apiece (maybe there's a lesson to be learned here), I'll replace them all and see what happens.

[Anonymous user · Accepted solution]

Replaced them all with no change in behavior. Any ideas out there?

[Aged]

Is it possible that you can provide me with part numbers for the relays so I can get replacements? Do you happen to know which one(s) is for the avr?

[Zorac]

Did you have any luck finding schematics/part numbers? I am having the welded relay fault as well. Updated firmware and no luck. Opened it up and found a blown component (c910). Some of the traces were pulled off the board, but luckily, they are still usable. Figuring this blown component is probably the problem. I guess it is some sort of Y capacitor, but weirdly, I think it has 3 legs unless the traces are in worse shape than I think.

[Zorac]

Update: So, probing around a bit with my multimeter and with the help of a picture of a good board I found on eBay, I managed to get it back together and running. Time will be the test, though. I'm pretty sure c910 is a filter or decoupling cap that blew up. The trace goes under and connects with one of the legs, so only 2 contact points. I happened to have some pretty generic 50v 104 (100nF) ceramic caps in my parts bin from another project, so I rebuilt the trace with a combination of wire and the legs on the ceramic cap. Ran it all day with no load in the garage and I didn't get the error, so I have moved it over to my server rack now.  As a bonus, the unit will enter green mode as well, it wouldn't do that before.

For reference I have attached 3 pictures, one of what the board should look like, one of the damanged cap, and one of my repair (excuse the blob of solder mask 🙂  ). The question is why did the cap go? Was it age, manufactuer fault, heat (I did buy it used so I can't comment on the conditions it came from), or is there an underlying issue with the unit that caused the cap to let go. Hopefully a transiant issue and not a latent issue with the unit.