When circuit breakers collide.

[wbsettle]

Trying to keep this simple. I went to replace the motor on a gable vent fan over the bonus room/garage (died a couple of years ago, but recent record heat finally motivated me to fix). After running myself ragged trying to identify which circuit was feeding the fan (flip breaker/climb ladder/inductance tester beeps/curse/repeat), I discovered I had to flip *two* breakers to cut power to the fan. Everything else connected to those two breakers seems to function normally...lights go on/off with one breaker, receptacles go on/off with the other...no evidence of anything else being crossed, but I didn't search extensively. These breakers are on the same side of the panel (phase?), but not directly adjacent.

How worried should I be about this? Suggestion on where to look for the cross connection?

For background. We've been in the house 7 years and I'm just now discovering this. House is roughly 17 years old, but the bonus room appears to have been finished some time after the main house was built with the finish work clearly not up to the main house standards. I have discovered a few other anomalies over the years...a GFI that was wired so that it *was not* protecting the outlets downstream and a 15A receptacle or two that were wired with 10g instead of the expected 14g.

Thanks!

-Brent

[LCHIEN]

10 ga is OK instead of 14 ga., as long as the receptacle terminal screws can get around the 10 ga wire... You can always use a larger wire (smaller gauge no.) to carry the current that requires the smaller wire, as a minimum.

As for the motor that runs on two breakers...
One wild guess would be that its a 220V attic fan (rare?) connected across opposite phases to get 220V... that would require the two hot wires from each circuit. Not really safe since the breakers aren't ganged. tripping or turning off one breaker still leaves the device "hot". Still a 220V attic fan would be odd here. Does it have a label stating the voltage?

So that brings the next question - how were you determining that it took two breakers to cut the power? You said inductance tester??? Is that one of those things that you hold next to a wire and it indicates if the wire is hot or not?
Since I'd expect a 220V attic fan to be rare, I'm wondering how good you r method for checking a hot wire was.

The best thing in this case is to cut the main breaker so that all your house is completely dead, then go disconnect the fan, then turn on the house power and use a voltmeter to see if its live then try and see what breaker its on if that's important to you. The fix the fan. Easy to say when its 75 degrees here at my computer, I know.

[LinuxRandal]

I don't know what they were thinking, but when I was a kid, when dad went to do anything like hanging a ceiling fan, plan on a new watch. It seems that whoever wired the house had smoked crack first, and ran two separate circuits through a junction box (light fixture box), so there was residual power that shown on the test lights (and he would hit a live wire and zap a watch).
Just a thought.

Thankfully I never took after him (why my mom didn't want power tools around the house), as he had issues with plumbing, vehicle repair, electrical work and a few other things (ended up with an axe in the head).
I compare him to Red Green mixed with Tim Taylor.

[wbsettle]

Loring, I don't think there's anything wrong with my method for troubleshooting. It's a standard 120v fan motor. Yes, I used a non-contact voltage sensor (pencil type) at the fan (allows confirmation the motor and not thermostat is bad)...I said inductance, but maybe that's not the principle they work on. I identified the pair of breakers needed to kill power by trial and error. When doing only singles didn't kill the fan after going through the entire box, I started shotgunning multiples. When that approach eventually cut power to the fan, I worked backwards to find the magic combination, eventually narrowing it down to two breakers that, conveniently, were labeled to handle the garage and bonus room power. As far as I can tell, the receptacle and light circuits in the garage/bonus room work as expected per their labelling...single breaker on/off, expected circuit killed...only the attic fan seems to be cross connected somewhere. Granted I haven't checked every single outlet to make sure they're all actually dead when the ones in use are killed.

The fan motor has been replaced and is working normally. My question is how worried should I be about finding the crosswired point since it presumably has been so for ~17 years without any apparent harm? Google turns up a very few hits for similar questions, but never a clear explanation of risks or resolution strategies...well, nothing beyond the obvious fact that one could be in for a shock working on a circuit that is presumed dead, but is actually hot. ;-)

I'm also aware the larger gauge is acceptable. I was just illustrating the point that I have found a couple of anomalous conditions within the house wiring prior to this, such as using 10g for at least one outlet, despite using 14g for the bulk that I've checked (replaced a string of outlets a couple of years ago since the house was built using those push connect outlets that are prone to failure over time, per internet lore). Since, presumably, the original electrician ran out of 14g and found a hank of 10g on the truck, it raises the question of sloppy work in my mind.

-Brent

[LCHIEN]

since your initial isolation mthod used an "inductance" tester (I'm suspecting its an eletric field detector), the results are suspect.

Does the now working attic fan turn off with either breaker? If only one turns it off then that's the breaker. If they both turn it off then you have an apparent problem, still.

I just don't see how such a thing can happen.

[vaking]

I am with Loring. If you had this fan cross-wired from 2 separate breakers it would mean this fan gets power from 2 separate circuits. In this case power from one circuit would be feeding into the other circuit thru the fan connections. In this case turning off one breaker would not have turned off outlets or lights, you would have had plenty of cross-wired things in your house. Since you say everything else worked properly - I don't think you have cross-wired fan.

[woodturner]

since your initial isolation mthod used an "inductance" tester (I'm suspecting its an eletric field detector), the results are suspect.

It sounds like he is using a voltage detector "pen" such as
http://www.toolexperts.com/santronic...n-contact.html

My first guess, assuming his testing is accurate, is that the hots from the two breakers are connected together at some point. Easy to do when rewiring if one is not very careful.

[LCHIEN]

It sounds like he is using a voltage detector "pen" such as
http://www.toolexperts.com/santronic...n-contact.html

My first guess, assuming his testing is accurate, is that the hots from the two breakers are connected together at some point. Easy to do when rewiring if one is not very careful.

I have one of those pens, they work by detecting the electric field. Because they can be affected by other objects and nearby wires, they can give false positives, altho they don't seem to give false negatives.

If the hots of two circuits were wired together then the other lights and outlets would exhibit the same problem... would require both breakers to be off - which is not the case the OP stated.

I'm thinking the least reliable observation/measurement made here is the pen detector - which I asked if had confirmed by actual operation of the breakers but there has been no reply so far.

[wbsettle]

I'm thinking the least reliable observation/measurement made here is the pen detector - which I asked if had confirmed by actual operation of the breakers but there has been no reply so far.

Wow. Tough room. Some of us do have day jobs that keep us from jumping as soon as someone posts a response to a question.

While I'd never heard of it before, it appears I have a multiwire branch circuit. Two 120 volt circuits are allowed to share a single neutral as long as the hots are on opposite phases. In my case, the two circuit breakers are not in adjacent slots, as was required by later NEC updates, but they are on opposite phases. Amazingly, to some extent, this circuit is still allowed by code today. With a MWBC, one can get a ghost voltage on the dead circuit through the neutral from the live circuit. Apparently, the pen tester is sensitive enough to pick up this residual voltage. Given how this fan was installed, the pen tester was really the best tool since I had no exposed connections for contact probes...fan power junction was inside the thermostat and line side feed runs off into a wall...I had no desire to go prying off the cover of the metal t-stat box when the pen said the line was hot. I found a thread where someone measured 28v on the dead half of a MWBC circuit...not enough to actually power a real load, but enough to give you a good tingle and show up on a DMM. I'll still investigate this a bit further at some point, but I'm now comfortable that the circuits are functioning as intended and safe per 1994 NEC.

-Brent

[LCHIEN]

sorry i didn't mean it as criticism or chastisement, it was just a fact; the deciding test (or at least to decide in my mind) had not yet been reported.