Yeah, that's what I was worried about as well. I will change the triac and hope that it works. I have a way to test these on the bench.
The other repaired unit is being tested on the water heater. The one mfg'd in 2011. At first, the burner would not turn on. So I took apart the OVD again and it seemed that there was no AC on the burner wire. Seemed that the third micro switch wasn't up against the cam enough. I adjusted it and also repaired a through hole connection with a piece of wire, thinking that was the break in continuity. Finally I got a good test on the bench and put it back in service. Still no burner turn on. Further investigation traced to a separate thermal switch on the stack that was open. It was resettable, and that was the final problem. Now it's working. I'll let it run through a few cycles to make sure it's working properly. At least I know I can repair the old blue units made in 2011.
I had another thought about a design improvement. Since this 2011 unit failed due to one of the micro switch contacts becoming welded shut, perhaps adding snubber capacitors across each switch might extend the life. In addition to adding fusing and MOV surge protection.
The other repaired unit is being tested on the water heater. The one mfg'd in 2011. At first, the burner would not turn on. So I took apart the OVD again and it seemed that there was no AC on the burner wire. Seemed that the third micro switch wasn't up against the cam enough. I adjusted it and also repaired a through hole connection with a piece of wire, thinking that was the break in continuity. Finally I got a good test on the bench and put it back in service. Still no burner turn on. Further investigation traced to a separate thermal switch on the stack that was open. It was resettable, and that was the final problem. Now it's working. I'll let it run through a few cycles to make sure it's working properly. At least I know I can repair the old blue units made in 2011.
I had another thought about a design improvement. Since this 2011 unit failed due to one of the micro switch contacts becoming welded shut, perhaps adding snubber capacitors across each switch might extend the life. In addition to adding fusing and MOV surge protection.
Last edited:
After successfully repairing the older OVD module, I tried replacing the triac on this newer model (the one with the 16-pin IC).
The module used to do nothing but now it's doing something, but that something isn't right. It rotated the motor to the closed position and close the contactor to feed power to the burner, which is obviously wrong. It should open the vent damper and then close the contacts to the burner.
Prior to replacing the triac, it was totally dead--no contact closure to the burner. Maybe the program in the microcontroller is corrupted. The damper motor is only supposed to rotate in 90° increments (open to close to open), but I watched it rotate 180° when I applied AC and made contact from the sense wire (normally goes to the aquastat) to the AC hot to simulate a call for heat. Then I disconnected the sense wire to simulate high limit temp cutoff and waited five minutes. The damper is supposed to rotate to the closed position, but nothing happened. I waited 9 minutes and still nothing.
So I guess this unit is kaput. Damn that micro controller! If I could rewire it to the old circuit, I might be able to make it functional.
The module used to do nothing but now it's doing something, but that something isn't right. It rotated the motor to the closed position and close the contactor to feed power to the burner, which is obviously wrong. It should open the vent damper and then close the contacts to the burner.
Prior to replacing the triac, it was totally dead--no contact closure to the burner. Maybe the program in the microcontroller is corrupted. The damper motor is only supposed to rotate in 90° increments (open to close to open), but I watched it rotate 180° when I applied AC and made contact from the sense wire (normally goes to the aquastat) to the AC hot to simulate a call for heat. Then I disconnected the sense wire to simulate high limit temp cutoff and waited five minutes. The damper is supposed to rotate to the closed position, but nothing happened. I waited 9 minutes and still nothing.
So I guess this unit is kaput. Damn that micro controller! If I could rewire it to the old circuit, I might be able to make it functional.
Well, you may be able to do that if you add a bodge card (perf board works fine) with a through-hole 2904 and the appropriate satellite parts. In essence, duplicating the circuit of the 2011 board.
Could you just duplicate the 2011 board with someone like JLPCB? It looks pretty simple and probably 2 layer. I think you know all the parts now. Another option would be a total redo using an arduino or pi-zero. Pretty simple logic to duplicate. I have used pi's for quite a number of projects and find the biggest advantage is the networking. All of the projects report back to a central command to keep tabs on everything from the pool to the AC/Heat performance. In your case you could add in some thermistors in key places. Maybe even get a CO sensor to add a second verification that the flue is operating correctly.
You could take it further, using Hall effect sensors / reed relays / proximity switches, and a switching IC like the 4066 / 4093, or similar.
Then no wear problems with switches.
And a 555 as timer.
No need for Pi and so on.
You could even take it commercial if you feel there is a market.
Then no wear problems with switches.
And a 555 as timer.
No need for Pi and so on.
You could even take it commercial if you feel there is a market.
Part of me wants to make a commercial competing product just to stick it to them for being so uncooperative and also for selling a shoddy product, but the other part of me is saying "I'm too old for that.. let someone younger and more ambitious do it."
Lots of good ideas. The making of another PCB would require precise dimension for the microswitches to line up properly on the cam. That rules out perf board.
I supposed I COULD strip down one PCB and make a scan of the board itself and do careful measurements then replicate it in ExpressPCB Plus or some such. But at the moment, I just want to have my water heater work and I'm there and have more pressing matters to deal with (such as dealing with defective materials on my roof and dealing with that company').
My solar stuff is networked, at least as far as the charge controllers go (wish I could access my three BMS'es via something more range capable than bluetooth, but networking an OVD seems unnecessary.
Lots of good ideas. The making of another PCB would require precise dimension for the microswitches to line up properly on the cam. That rules out perf board.
I supposed I COULD strip down one PCB and make a scan of the board itself and do careful measurements then replicate it in ExpressPCB Plus or some such. But at the moment, I just want to have my water heater work and I'm there and have more pressing matters to deal with (such as dealing with defective materials on my roof and dealing with that company').
My solar stuff is networked, at least as far as the charge controllers go (wish I could access my three BMS'es via something more range capable than bluetooth, but networking an OVD seems unnecessary.
I "get it". I was so angry after my blower went on after 3 years on a 12 grand AC system and labor to replace the "free" part was 425. So when the condenser fan went out (one of those fancy multi-speed ones) I thought am I going to pay some guy 500 to install a new "free" motor or am I going to install a fixed speed 200 dollar motor. I opted for the 200. The HVAC industry thrives on unreplaceable parts. And I know the motor was actually ok as I have removed the controller part from the motor exposing the 3 motor terminals and have successfully gotten it to spin with a controller I've been building up. ECM motors are interesting animals. My end goal is to be able to replicate the controller and replace the fixed speed motor with the original motor and new controller. Of course carrier will not sell just the controller. Buggers. Needless to say but for being basically retired I have spent way more in time than I saved in money. Way way more. As I also have reverse engineered their communications protocol so in theory I could remove the TStat and replace it with a program of my own. Their protocol signalling method is brain dead. No wonder they don't open it up. Too embarrassing to publish.