I recently was having some HVAC issues with a heat pump blower motor not running. Being comfortable using a multimeter, I opened up the panel to the furnace and was checking voltages on the capacitor and blower motor when one of the hot power wires arced by touching the metal furnace. I now have a complete loss of power to the furnace. Immediately after the arc, the dual gang 35 amp breakers tripped. I reset the breaker to find that the power was still not reaching the furnace at all. I checked the wires coming out of the breaker with a non-contact voltage meter and confirmed voltage is coming out of the breakers.
There is only about 8 feet of wire from the main electrical panel where the breakers are and the 8 gauge wire coming into the furnace electrical panel. The wire coming out of the breaker leads goes up in my attic into nearly inaccessible areas before coming down through the ceiling into the furnace panel. I checked the resistance between the breaker and the input of the furnace power and there was no connectivity. The ohm meter was 0.L, if the wires were good, there should be connectivity.

QUESTION: Is it possible for both 8 gauge wires coming out of the two 35 amp breakers to completely burn out so as to not be able to feed power to the furnace because one of them arc'ed? If so, is the 8 gauge wires being burned and not providing current the only scenario or could something else be amiss with the wires feeding the furnace from the breaker?

I went up in the attic and tried to investigate any visual failures from the 8 gauge wire but I was not able to find any nor could I get a non-contact voltage detector to work on any of the 8 gauge wires since the sheath is too thick.

enter image description here

  • What brand is the breaker? And is it in a panel it was made for?
    – Tyson
    Dec 17 '17 at 21:05
  • It’s a GE 35 amp standard D breaker and it is the correct fit for the panel. I replaced it with a brand new 40 amp breaker which made no difference to proving current to the furnace. Dec 17 '17 at 21:18
  • Given all that I have read here, I also vote for an all new branch circuit cable. Copper is always a better choice of conductor type.
    – Paul Logan
    Dec 18 '17 at 7:59

In short, it is possible. It's not very likely and whether this is actually what happened, it's pretty hard to say for sure without more information.

Can you try to further diagnose the current problem by getting yourself a contact voltmeter and testing:

  • voltage at the breaker between phases (black to red)
  • voltage at the breaker between each phase and ground
  • voltage at the furnace between phases
  • voltage at the furnace between each phase and ground

The fusing current of 8 gauge copper (you didn't specify copper; I'm assuming) is 472A. However, mistreated (sharp bends, kinks, fastener penetrations, over-tensioned wire, etc) wire may have a lower fusing current. Note that only one of the conductors would have to fail since that would be sufficient to interrupt the circuit. The additional testing described above may tell you which of the wires has failed (if any).

The particular brand and model of breaker you have will determine what its expected performance is like - in particular, how quickly it will trip (and so how long the wire is exposed to over-current). All listed breakers should perform roughly the same in this regard if they are not defective or in a failed state. Note that most breakers are rated only for a limited number of full-spec trips (often two) before de-rating in some way.

A defective breaker or one which has experienced enough trips may lead to a condition where the wire could be exposed to over-current long enough to melt. However, breakers are made to de-rate in the direction of more safety (ie, trip at lower-than-rated currents) and not less so I wouldn't expect this to be a likely scenario.

An alternate explanation could be that you have a junction box out of sight and a (weaker) connection between two different pieces of #8 wire has failed.

Since you mention the current wire is only about 8 feet long it seems like the thing to do would be to surface wire a replacement from the circuit box to the furnace box and write off the old wire (which you may want to cut back to the wall, or remove completely if you can to prevent future use; at least nut the ends and label it as failed somehow).

  • 1
    Excellent answer but don't forget to mention that a non-contact meter will not work in this situation. Kevin needs to get a voltmeter and check voltages at the breaker both across the phases and to ground to see if the breaker or its connections are good. Then he needs to check the same thing out at the connection of the furnace. If you have a burnt wire in the cable it will most certainly be burnt through the sheath and there will be a burnt smell. I would bet on breaker or a connection failure.. Dec 17 '17 at 19:22
  • I measured the voltage coming out of each breaker to be 123v from wire to ground. There is 240v between the two breakers. There is no voltage at the furnace at either wire. Also, I temporarily replaced the 35 amp breakers last night with brand new 40 amp breakers (all that Home Depot has) and I still had no power at the furnace. When investigating in the attic, I saw no signs of melted sheathing or burning smells. As soon as the wire arc’ed, the breaker tripped so the wire wasn’t overloaded for a long duration. Dec 17 '17 at 20:53
  • So a major revelation occurred when testing the terminals of both ends of the wires. I noticed that the wires coming out of the breaker are 6 AWG aluminum while the wires coming into the furnace are 8 AWG copper. This definitely supports the junction box hypothesis. While the wires only travel a very short distance in the attic, their initial emergence into the attic is hidden behind a plywood sheet and I can only start to see them about a foot or two from where they emerge into the attic. I’m very surprised by this given their short distance of travel. Dec 17 '17 at 22:23
  • Splices from aluminum to copper are more suspect than others. And if the splice is not accessible (ie, in a box behind a plywood sheet) then it is not to code and that might be a hint that it was done inexpertly. I'd definitely go with a new all-copper run. Dec 17 '17 at 22:28
  • @Kevin Devine, so what did you find behind the plywood? Where was the splice connection of the #6 Al to the #8 Cu and how was it connected? Dec 19 '17 at 14:21

There is no chance of the entire wire burning up. The wire failed at the terminations. Unscrew the wire from the breaker and furnace, and closely inspect both sets of wire ends, and the screw terminals on which they terminate. You will find either arc damage, or poor workmanship, or both.

Also check for any intermediate splices. There is nothing wrong with those if they are done properly.

It's also possble it's a failed breaker, but I wouldn't count on that.

  • Federal pacific breakers fail. Our company still finds several panels a year.....
    – Tyson
    Dec 17 '17 at 21:08
  • All terminals look good with no damage. The breaker screws look go as new and all wires look fine. The only damage spot is the actual terminal end that arc’ed that is connected to the capacitor deep within the furnace. But there is no power coming in well before that point. Dec 17 '17 at 21:21
  • 1
    @KevinDevine "look good"? Visual inspection is not enough. You have to disassemble them and closely inspect the parts taken apart. A side effect is checking the torque on termination screws. Dec 17 '17 at 21:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.