lights dimming when AC kicks on

Status
Not open for further replies.
S

Stevenfyeager

Senior Member
Location
United States, Indiana
Occupation
electrical contractor
I have read some responses in the past on this, sorry to ask again. But my instructor years ago told us to put heavy loads near the main breaker in a house to minimize dimming of lights when the heat pump or AC unit kicks on. I'm still practicing what my instructor taught but still am seeing some dimming in new houses. How do you minimize this ? Thank you.
 
When one circuit affects another like that, it must be by something they share; in this case, from the panel bus all the way back to the utility transformer. We only have control over part of that pathway. Of course, some momentary dimming is normal.

I, too, prefer placing the highest breakers nearest the main breaker or main lugs, and I also like stacking the 2-pole breakers on one side of the panel, and the 1-poles down the other.
 
I was at a new subdivision today on a call, homeowner said they only had half power. Sure enough, one leg was dead at the meter. About half the yard dug up where they added more field to the septic tank, started looking for the pad mount, nearest one was across the street, in line with where they had dug. Looked like that one pad mount fed four house’s each. Pretty long secondary. I imagine they have a lot of flickering light issues! (By the way, the plumbers did get a locate, but apparently broke a wire in the triplex. They claimed they didn’t hit it)
 
LarryFine said:
When one circuit affects another like that, it must be by something they share; in this case, from the panel bus all the way back to the utility transformer. We only have control over part of that pathway. Of course, some momentary dimming is normal.

I, too, prefer placing the highest breakers nearest the main breaker or main lugs, and I also like stacking the 2-pole breakers on one side of the panel, and the 1-poles down the other.
Click to expand...
Out of curiosity, what does stacking the 2 pole breakers on one side do? Thanks.
 
Undersized AC circuit.
Undersized service
Loose connection in service (anywhere from transformer to panel)
Bad start capacitor in AC
LEDs that are overly-sensitive to transient voltage drops.
 
Greentagger said:
Out of curiosity, what does stacking the 2 pole breakers on one side do? Thanks.
Click to expand...
1. It looks like I placed them on purpose.
2. It prevents two large breakers from sharing stabs.
3. The end caps of many panels have many small KOs on one (the neutral) side and fewer larger ones on the other side.

I like to make almost-full panels look almost empty:

DSC00850.JPG
 
LarryFine said:
1. It looks like I placed them on purpose.
2. It prevents two large breakers from sharing stabs.
3. The end caps of many panels have many small KOs on one (the neutral) side and fewer larger ones on the other side.

I like to make almost-full panels look almost empty:

View attachment 2553772
Click to expand...
No "curly tails" in there? Must have been before AFCIs were required because those panel don't appear to be plug-on neutrals
 
201003-1205

I suggest you will not reduce light dimming (flicker) of incandescent lights when an air conditioner turns on.

At my main panel voltage from phase 1 to the neutral bar drops by about 0.8 V with an approximate 10 A resistive load placed on that phase. The other phase increases by about 0.4 V for the same load change. Thus, both the hot and neutral conductors from the pole transformer each have an impedance of about 0.4/10 = 0.04 ohms.

If the starting current from the air conditioner is 100 A, then there is a 4 V drop on a hot wire. That alone is sufficient to produce incandescent flicker.

If the air condiioner is 120 V, then you will see 8 V change for loads on that phase. Very noticeable. If it is a 240 unit, then possibly not 100 A, and less hot line drop, and no neutral drop. But a 240 unit may be larger, and more power load so higher current, and 100 A may still be a useful criteria. Akso your source impedance may not be as low as mine.

In a copper bus bar main panel I don't see much reason for breaker position concern other than added heat. What is the resistance of a 1 ft long bus bar compared to that of 150 ft of service cable.

.
 
gar said:
201003-1205

I suggest you will not reduce light dimming (flicker) of incandescent lights when an air conditioner turns on.

At my main panel voltage from phase 1 to the neutral bar drops by about 0.8 V with an approximate 10 A resistive load placed on that phase. The other phase increases by about 0.4 V for the same load change. Thus, both the hot and neutral conductors from the pole transformer each have an impedance of about 0.4/10 = 0.04 ohms.

If the starting current from the air conditioner is 100 A, then there is a 4 V drop on a hot wire. That alone is sufficient to produce incandescent flicker.

If the air condiioner is 120 V, then you will see 8 V change for loads on that phase. Very noticeable. If it is a 240 unit, then possibly not 100 A, and less hot line drop, and no neutral drop. But a 240 unit may be larger, and more power load so higher current, and 100 A may still be a useful criteria. Akso your source impedance may not be as low as mine.

In a copper bus bar main panel I don't see much reason for breaker position concern other than added heat. What is the resistance of a 1 ft long bus bar compared to that of 150 ft of service cable.

.
Click to expand...
I will say Gar is correct but you can " Minimize " the effect and lower the inrush with correctly applied " Single Phase hard Start Gear " for your respective condensing unit.
This is: Specified Potential Relay and Start Capacitor, to go with the Run capacitor already in the cabinet. This is assuming you have a standard basic unit without any new generation electronics running the compressor load. The inrush can be significantly trimmed with this maneuver and I have seen it and done so hundreds of times. Pton is correct. The other thing is to attempt to educate the customer that the world is not coming to and end every time the AC comes on under peak load conditions. Certain compressors have terrible starting characteristics and they are not all made equal. This was seen in the emergent " Scroll Technology " to the extreme in the early 90s.
 
LarryFine said:
I, too, prefer placing the highest breakers nearest the main breaker or main lugs,
Click to expand...

I don’t quite understand that since VD affects the entire bus, rather than the area between the main and the breaker the overload is on.
IMO it’s the way I see a lot of them, and it looks good, but it does no good for flicker.
 
Hv&Lv said:
I don’t quite understand that since VD affects the entire bus, rather than the area between the main and the breaker the overload is on.
IMO it’s the way I see a lot of them, and it looks good, but it does no good for flicker.
Click to expand...
I agree that voltage drop occurs everywhere along the current pathway, and we can't do anything about the shared supply. But, for theory's sake, picture a panel bus as a series string of resistors, one between each adjacent pair of stabs.

A large load at the top will contribute flicker to all loads, due to to voltage drop of the supply current pathway. But small loads below that point will only exhibit the unavoidable drop that the large load experiences, and little more.

A large load at the bottom will add the additional voltage drop along the bus to the drop it would cause at the top, and every small load above the large one will experience the bus's drop in addition to the rest, proportionately to its position.

The difference may be small, but, like loading the dishwasher or sorting laundry, adding a method can made a mundane chore less mundane and even interesting. We have to load the panel; why not use a method to base the decisions on?

And, yeah, it looks cool to me. :cool:
 
LarryFine said:
I agree that voltage drop occurs everywhere along the current pathway, and we can't do anything about the shared supply. But, for theory's sake, picture a panel bus as a series string of resistors, one between each adjacent pair of stabs.

A large load at the top will contribute flicker to all loads, due to to voltage drop of the supply current pathway. But small loads below that point will only exhibit the unavoidable drop that the large load experiences, and little more.

A large load at the bottom will add the additional voltage drop along the bus to the drop it would cause at the top, and every small load above the large one will experience the bus's drop in addition to the rest, proportionately to its position.

The difference may be small, but, like loading the dishwasher or sorting laundry, adding a method can made a mundane chore less mundane and even interesting. We have to load the panel; why not use a method to base the decisions on?

And, yeah, it looks cool to me. :cool:
Click to expand...
So what are you suggesting, or recommending, large loads top or bottom?
Any justification for one over the other or time that you would have to use one over the other?
And would one way lesson the dimming more than the other?
Or are you simply saying it wouldn't matter? What you do won't change what is coming from the utility. Six of one half dozen the other.
I had one journeyman when I was training tell me to always put large loads on bottom, no explanations as to why, and I never considered it.
 
I usually put the large loads on the top, but not for these reasons. I do so so any heat they generate won't affect any smaller breakers that are above them. I've seen stove breakers heat up and the 15a lighting circuit above it, with the heat it generates itself, cause it to trip.
 
201004-1421 EDT

480sparky has a valid reason for putting a highly loaded, meaning on average is producing a lot of heat, breaker at the top.

I ran a simple experiment on my main panel (SqD QO 200 A, copper bus bars). Voltmeter across length of a bar. A change of about 10 A produced a change of about 1 mV. Need a much more controlled experiment. Thus, about 10 mV at 100 A. You will not see light flicker from a 120 V incandescent bulb from a 10 mV change in applied nominal voltage.

.
 
Status
Not open for further replies.
Top