Breakers getting hot and tripping...

[tokenbanker]

Man, this is a pain in the ass. Before wiring in another panel I decided to try and add a 10ga neutral onto my 8ga. Ghetto yes..but I wanted to see if it would make any difference.

I took off the surge protectors and fired everything back up. Two of the breakers tripped in about 10min.

The symptom seems to be induction heating of the breakers...Nothing else gets hot. ALL of the wires are cold. Once I flip off the power the neutral bar is barely warm. The bus bars actually aren't that hot either.

My current plan is to just add another panel and have them wired so they *could* run 4 lights each. This panel seems to be happy enough with 4 lights.

what a PITA. Any other ideas for a quick fix?

Hmm...

Do you have access to a true RMS clamp-on ammeter? I'm thinking it would give you insight into the exact nature of your load on each of the conductors.

 

[Cam]

I went out and bought a TRMS ammeter.


Spent the last 30min clamping all the wires and monitoring it.

There is 27ish amps on each of the hot wires. And between 1-2 amps on the neutral.

sooooo....I'm guessing there is some effect on the breakers from the balancing thats taking place between the hots and the neutral.

The only time the lights behave perfectly is when the entire load is all on one wire. Then it all goes down the neutral. So I could probably install a second neutral bar and another wire for it and keep them separated. (I've never seen anyone do this though...I doubt theres enough room and it doesn't look like this will be doable.)

Its quickly looking like switching to 240V will be the solution. I'm going to try that when I get some free time.

Any thoughts? Anything else I should check with my new ammeter? I haven't got much experience using one.

 

[seeyouaunty]

If you could do something like a mspaint schematic of how you've got everything connected (wires, fuses, ballasts, etc) along with your amp readings we'll have a better idea of whats going on.

 

[Stress_test]

One of your outlets or a light is wired backwards. Either that or a ballast has a short in it. That's about the 2 things that could cause your problem.

OOops: 1 other things that could cause it would be a male end of a cord that's wired wrong.

 

[rives]

Any thoughts? Anything else I should check with my new ammeter? I haven't got much experience using one.

Have you tried killing all of the breakers but one, monitor that one with your new toy, and start flipping breakers back on one-by-one? It would be interesting to see at what point things go sour. I too would like to see a drawing of your installation.

 

[the_extremist]

My vote for fixing the issue altogether is to run three 240V core & coil ballasts on each of two different 20A 240V circuits.

Do you have any non-digi ballasts to try out?

 

[billbob]

fixed yet?
run a subpanel and go 240v. much less amps (half) over the wire. You might have a cut wire. Very easy to happen when you strip it and very hard to see.

 

[BlindDate]

Wow all these pages and all he has to do is clamp an amp meter onto the breaker wires. This guy is an idiot.

 

[seeyouaunty]

Wow all these pages and all he has to do is clamp an amp meter onto the breaker wires. This guy is an idiot.

Don't speak so fast, his latest measurements match the ones he took in the first post (2 x 27A = 6 x 9A) but Cam didn't give enough details to be sure, hence we are waiting for a circuit diagram.

 

[Cam]

Switching to 240V did the trick. Not sure why...it shouldn't have made a difference. I took it down and rewired the plugs. Breakers are 15, 15, 25, left to right in the pic. 40 Amp Square D contactor with trigger cable. Since I added a pretty heavy duty water pump on this line for my CO2 a 40A breaker is now feeding the panel. The three white ones are 240V. The grey one thats upside down runs my co2 system and is 120v. The ballasts are getting remounted on the walls this next week so they aren't above my co2 water supply. The lights are big kahuna's from HTGSupply. I love them. The lights are mounted to 2x4's that are raised up and down via an electric hoist. There is also a 14k btu LG Art Cool mini split in there you can't see.

As you can tell one side of the grow suffered because I left those lights unplugged for 2 weeks.

Thanks for the advice guys, peace.










The highest res this site would let me use was 800x800. Sorry.

 

[BYLROY]

To summarize this thread:
"blah blah blah"
"try switching to 240v"
"i don't understand why anyone would use 240v"
"blah blah blah"
"switching to 240v solved everything"

 

[RomulusUSA]

Can someone summarize why he needed to switch to 240? I understand the benefits, but after re-reading the thread, I'm still not entirely clear on what the problem was. Can an electrician explain it for me?

 

[the_extremist]

The fact that there are less terminations and less wires carrying current within the panelboard no doubt helps. When you have all those individual 15amp circuits feeding the lighting, the white wires also carry current inside that enclosure. This can contribute to heat problems around the breakers. The more wires carrying current inside an enclosure there are, the more heat builds up. Kind of like derating your ampacity when you have more than three current carrying conductors in a raceway. It's because those extra wires carrying current contribute to the heat within that raceway. Same principal.

My vote isn't for changing voltage as much as it was changing the actual ballasts themselves. Those ballasts might be poorly manufactured, built with cheap parts, they might be Friday models, who knows? This can result in an increased harmonic current flow on the return path. This will also contribute to the heat problem that was causing his breakers to trip.

I really like traditional core and coil ballasts. They're trouble free, reliable, affordable and the nicer ones can be run on 120v or 240v.

 

[Cam]

The wires themselves never got hot. When I left the cover off and had a fan on it full blast the breakers would still overheat.

My guess is that there was some strange effect going on with the load balancing between the neutral and hots. I would guess that the cheap ballasts were the cause of it.... BUT, I don't think it had anything to do with power harmonics after taking a look at the ammeter.

 

[Teslas Deathray]

Was the sub-panel plastic? A plastic box has less heat dissipating properties than a metal one. A 6 breaker box is quite small physically and would normally be used for small domestic set-ups. 6x16A breakers on continuous load all right next to each other will have each others heat adding together to form the ambient temperature within the box. Maybe a physically larger box would have done it. Even though the problem is solved now, it may help someone else at some stage. The switch to 240V would have reduced the current levels, and so the heating levels, to an ambient level within the box that would not cause the thermal trip element to operate.

(I know you had a fan on it, but this theory concerns conducted heat, and a breaker has alot more surface area in contact with the breakers on either side than exposed to the airflow)

 

[Cam]

It was in the exact same sub panel you see in the pic on page 6.

I actually had the same thought as you Teslas Deathray. I tried using two tandem breakers and two regular ones so I had two extra spaces in there. It didn't make any difference in the heat dissipation.

I'm gonna stick to the guess that there was some strange electromagnetic effect on the breakers from the loads balancing and the cheap ballasts I was using.. Although, in two years this is the only problem I've ever had with them.