mark32
Senior Member
- Location
- Currently in NJ
Hello all,
I am trying to troubleshoot this circuit, as the breaker will trip after a few seconds, but I am unsure what I'm supposed to be looking for. In this photo, (Which was taken after we started troubleshooting) you will see the black from the clock's pigtail is together with a lead that goes to that relay. These tie together with the feed (black 120vac) coming in and going out to the rest of the clocks. We are getting continuity to ground on one of the red conductors. (I believe this red conductor is the one that is used to synchronize the clocks). So okay, now we have something to look for. However, what is actually suppose to take place here? I am told 24vdc is applied to the red conductor, (for synchronizing purposes) but how does this occur? You can see some LV wiring on the bottom of the relay, which I assume originates at the master clock two floors above. The relay itself is labeled as such, the "input" is the LV and the "output" are the black and white/red conductors. Initially, I thought once the control wiring closed the relays contacts, 120v would be applied to the white/red conductor, but this is not the case, at least I don't think it is. But, it must be the case, which is why the breaker is tripping. Can anyone shed some light on this for me?
I am trying to troubleshoot this circuit, as the breaker will trip after a few seconds, but I am unsure what I'm supposed to be looking for. In this photo, (Which was taken after we started troubleshooting) you will see the black from the clock's pigtail is together with a lead that goes to that relay. These tie together with the feed (black 120vac) coming in and going out to the rest of the clocks. We are getting continuity to ground on one of the red conductors. (I believe this red conductor is the one that is used to synchronize the clocks). So okay, now we have something to look for. However, what is actually suppose to take place here? I am told 24vdc is applied to the red conductor, (for synchronizing purposes) but how does this occur? You can see some LV wiring on the bottom of the relay, which I assume originates at the master clock two floors above. The relay itself is labeled as such, the "input" is the LV and the "output" are the black and white/red conductors. Initially, I thought once the control wiring closed the relays contacts, 120v would be applied to the white/red conductor, but this is not the case, at least I don't think it is. But, it must be the case, which is why the breaker is tripping. Can anyone shed some light on this for me?
Attachments
Last edited: