I have a lighting system for my home thats controlled by a circuit board and has ben in use for 20 years now. Recently, the lighting system switches can become un-responsive and if I remove the circuit board, disconnect all cables and power and leave it 'unplugged' for several hours, then reconnect everything it starts to function normally again for a day or two before becoming un-responsive again.

I know little or nothing about circuit boards but It seems like it may be a static charge issue. Is this basically a board that has reached the end of it's life or is there anything that can be done to fix this?

Any ideas would be gratefully accepted.

 

Welcome to AAC.

Be cause of the limited information you’ve provided I can only make a SWAG. My first suspect is an aging electrolytic capacitor or capacitors. This is a common failure mode for older boards. “Recapping” often gives a board new life.

 

Welcome to AAC.

Be cause of the limited information you’ve provided I can only make a SWAG. My first suspect is an aging electrolytic capacitor or capacitors. This is a common failure mode for older boards. “Recapping” often gives a board new life.

Thanks for your reply. As you have probably gathered, I’m no expert in these matters. I will look into having capacitors replaced. I’m the meantime, I’ve attached an image of the panel on the off chance it might reveal something.
thanks again.

 

I notice a clock chip (above the chip with label), one similar used in old PCs like the 486 series. It is possible the battery in it is dead. However, I don’t know if that would affect your operation or not. It also keeps the RAM alive I think…
C67, near that chip seems to be a bit swollen…
M48T18-100PC1 is the chip. They are not cheap…

 

I notice a clock chip (above the chip with label), one similar used in old PCs like the 486 series. It is possible the battery in it is dead. However, I don’t know if that would affect your operation or not. It also keeps the RAM alive I think…
C67, near that chip seems to be a bit swollen…
M48T18-100PC1 is the chip. They are not cheap…

Thanks for your help. I'll drop it into a specialist with your remarks and see what they say.

 

Seeming to me that a bad cap would be bad even after being unplugged for several hours, and would not function properly for 1 to 2 days before going wonky again. Also, though my first thought was bad switches, they don't go bad then go good after resting for several hours, then go bad again. And memory chips that depend on battery backup - a dead battery would also not fit the description of the 7 hour rest and 2 days functionality.

That's quite a complicated board for lighting control. Eight "Relay Drivers" suggest a fair bit of complication. 20 years use - you've gotten your money's worth out of it. Perhaps there's something on the market that does the same thing.

 

Seeming to me that a bad cap would be bad even after being unplugged for several hours, and would not function properly for 1 to 2 days before going wonky again. Also, though my first thought was bad switches, they don't go bad then go good after resting for several hours, then go bad again. And memory chips that depend on battery backup - a dead battery would also not fit the description of the 7 hour rest and 2 days functionality.

That's quite a complicated board for lighting control. Eight "Relay Drivers" suggest a fair bit of complication. 20 years use - you've gotten your money's worth out of it. Perhaps there's something on the market that does the same thing.

If the problem is temperature related it could act that way. Unless the TS has been rigorous in testing how long it takes to start working again, it might not be as long. But it is certainly possible it is some other component. All other things being equal, that’s where I would start—but not necessarily so if I had more information.

 

Looking closely at the board I didn't see any electrolytic caps that appear to have ruptured. Still, they don't need to rupture to go bad, they could simply be dry internally. Replacing caps is a cheap way to test for functionality, but given the complexity of the board changing caps might not be so easy to do. Quality soldering iron and proper technique would be the difference between success and failure - provided that "Caps" are the root cause.

I agree, proper testing is critical. One thing I often say is "Fully diagnose the problem before you start throwing parts at it."

 

If the problem is temperature related it could act that way.

If that's the case then a can of compressed air held upside down when spraying, spray various components one at a time to see if the functionality returns. Cold spray has been a good way to test for heat affected components in the past. Should still be true today as a method of locating an issue. But again, taking 2 days to overheat? Has me skeptical.

I'm bowing out of this thread because I don't see any possibility of ME drawing a conclusion or setting course for solving this other than what I have already posted. In other words - I'm out of guesses.

 

If that's the case then a can of compressed air held upside down when spraying, spray various components one at a time to see if the functionality returns. Cold spray has been a good way to test for heat affected components in the past. Should still be true today as a method of locating an issue. But again, taking 2 days to overheat? Has me skeptical.

Component cooler used to be a different can than the canned air...