This winter I installed a 16-circuit subpanel and rewired the second floor. The sub is fed from the main via a 60A 250v circuit over 6/3. Everything has been working fine ever since, but I haven't really been pulling much on these circuits until this week.
Today I'm noticing that when one of my window A/C's compressor kicks on it is flicking the lights. But these lights are on a different circuit (same panel). I did not expect this. The A/C is pulling about 10A (120v) on a 20A dedicated circuit. Other power being drawn from this panel includes:
-computer: ~5A on a dedicated 20A circuit
-computer and extras: ~5A on a dedicated circuit
-another A/C: ~8A on a dedicated circuit
-lights: ~? (1A maybe) on another circuit
So total I doubt I'm drawing more than 30A of 120V, so I can't be maxing out the feed, right? That feed should give me 96A of 120V (80% of 60A at 240V).
So what's the deal? Why would any lights on a different circuit flicker when the compressor kicks on on another circuit? To be fair, the flicker is subtle. It's not huge, but it's there.
The compressor may be pulling as much as 40 or 50 amps upon startup which would cause your other circuits to be affected. Incidentally #6 wire is only good for 50 amps. You should have used #4 for a 60 or 70 amp subpanel feed.
Hi, A/C is classical inductive load like many other electrical load we use. Being that it's surge current when starting is way more that the running current craw. Split second light flicher is considered normal. If you understand basic Ohm's law, you can see why it happens.
Hi, Two things, momentary starting surge current is much higher. #4 wire was better choice. Doing all this work, did you you take out a permit? Just wondering. Where I live we need a DIY permit from city and two inspections, during, after the work is done.
#6 wire has about .00046 ohms/ft. Double that for return and call it 0.001 ohms/ft. Assume 50 ft of wire to subpanel means 0.05 ohms. Thus a 10 A load would only drop about .5V. To test this get a digital voltmeter and measure the voltage at a plug of interest. Plug a load like a toaster in the other plug and see how much drop you get. Do this on both sides of the
240 circuit. If the drop is much more than calculated above then something is amiss.
Thanks everyone for the insights. I hadn't thought about the increased load on startup and how that could be enough of a pull to affect other circuits in the panel. I'll put my kill-a-watt on the A/ C and see if I can see what it draws during the startup (it might not be fast enough to see, though).
To answer some questions:
dimming when a motor starts, it's normal
I suppose technically it's a dim. I just wanted to distinguish it as a very quick dimming (quick in that it dims down and then back up very quickly) from what I used to have when the floor was all one 15A circuit---back then the A/C would come on and everything would dim and stay dimmed until the compressor shut off.
would cause your other circuits to be affected.
Ok, so since this is a 120V A/C, I suppose I should only be seeing a dim on other circuits that share that same side of the panel (e.g. the same single hot wire from the main)? I'll have to take a look and see if that's indeed the case. Or is the 40-50A 120V pull on one leg enough to affect the other side of the panel?
out a permit? Just wondering. Where I live we need a DIY permit from city and two inspections, during, after the work is done.
I did take out a permit. The inspector was great. We went over my plans before I started, he came out early on before I'd finished rough- in to make sure I was doing things correctly, then he checked the full rough-in, then a final. He approved the #6 for 60A. As I recall it rates for around 55A technically but the NEC allows you to round that up. I also had a long thread on this group discussing the feeder wire back then:
formatting link
The run is about 35' from the main. Part of the reason I wanted to go with #6 was flexibility in getting from the basement to the attic.
Looking back at my old thread, it makes me realize I should probably check and make sure I balanced the loads well on the two sides of the panel. I meant to, but I'll double check.
240 circuit. If the drop is much more than calculated above then something is amiss.
Thanks for this. I'll try this out and see what I find. I'll also look up Ohm's law...
well, the A/C is on a different side of the panel as the lights I was watching, so I guess it's enough to affect both sides of the panel. i did do a good job of balancing loads (for example, there are two a/ c's, one on each leg).
6/3 Romex, copper, has a 60 degree rating of 55 amps. 4/3 Romex, copper has a 60 degree rating of 70 amps. # 4 copper can only be used at a 100 amp rating for service entrance and main feeder
Just want to be sure, for balancing purposes, that you understand typical panel buss is not left side on one leg and right side on the other. They alternate in most panels
Hi, If you really want to do the test. Ordinary meter is too slow as well as human eyes. Really oscilloscope is needed or a meter which can read the instantaneous voltage and freeze the display.
Do you have a citation of this? I keep reading these stories out here, but I've never seen anything more this sort of third hand anecdotal evidence. I've read over my homeowner's policy (Texas form HO-B) and I can't see anything that I interpret would allow them to deny a claim on this basis.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.