Voltage difference per leg question

Status
Not open for further replies.
J

joeyd233

Member
Location
Raleigh, NC, USA
Hello out there. Here is my situation: 1993 home, all original electrical. I have a 15A 110V circuit that seems to imbalance the per-leg voltage in the home. With the main off, I read 124V to N per leg, 238V leg to leg coming off the pole. With the suspect circuit off, and the main back on, I still see 123V to N each, and the 238 leg to leg. Nothing abnormal. Power up the suspect family room circuit, and I now have 128V to N, and 120V to N. Full running load on that circuit is about 5A, and it appears to be #10 or 12 conductor. (I will confirm, and is roughly 100 ft run). Load is a TV, fan with dimmable lights, modem, router, and a few PC's, tablet chargers, etc. I have been unable to isolate a single load causing the issue, so I am wondering if this is even a real problem, aka, is it within code? Owners are experiencing lights dimming and brightening throughout other circuits as appliances cycle, ie fridge, washer, etc. They say it is worse with colder temps outside, especially cold mornings. (30's and 40's F) Any thoughts are welcomed here.
 
joeyd233 said:
Hello out there. Here is my situation: 1993 home, all original electrical. I have a 15A 110V circuit that seems to imbalance the per-leg voltage in the home. With the main off, I read 124V to N per leg, 238V leg to leg coming off the pole. With the suspect circuit off, and the main back on, I still see 123V to N each, and the 238 leg to leg. Nothing abnormal. Power up the suspect family room circuit, and I now have 128V to N, and 120V to N. Full running load on that circuit is about 5A, and it appears to be #10 or 12 conductor. (I will confirm, and is roughly 100 ft run). Load is a TV, fan with dimmable lights, modem, router, and a few PC's, tablet chargers, etc. I have been unable to isolate a single load causing the issue, so I am wondering if this is even a real problem, aka, is it within code? Owners are experiencing lights dimming and brightening throughout other circuits as appliances cycle, ie fridge, washer, etc. They say it is worse with colder temps outside, especially cold mornings. (30's and 40's F) Any thoughts are welcomed here.
Click to expand...

How is this possible? 124+124=248.
 
161205-1414 EST

joeyd233:

It is almost certainly a neutral related problem at least.

I assume that all your measurements are at the main panel input terminals. Also it appears you may have typos, meter problem, or the typical power company variations. infinity's comment implies a typo.

You should probably use a 3.5 digit meter. Too hard to read a 4.5 digit meter with typical line fluctuations.

If I make measurements at my main panel I get some results like:

1. Phase A hot to neutral 123.76 V. But there is considerable variation with no specific load variations. At times I may find the fluctuation to be within about 0.1 V for several seconds. Somewhat longer times possibly 0.25 V.

2. Using a 100 W load, about 0.8 A, the change in voltage from this load is around 0.05 to 0.1 V. Hard to do this measurement because of the short time power company variations. A longer averaging time constant for the meter would be useful. The reason to use a 100 W load is discussed below.

3. Using a 1500 W heater, about 10 A, the change in voltage is about 1 V. Thus, the loop impedance of the tested hot, neutral, power meter, service lines, and 1/2 the transformer secondary is about 1/10 = 0.1 ohm.

4. If while changing the load on phase A you monitor the Phase B voltage, then you can expect to see about 1/2 the change that was seen on phase A, and it will be an increase. Further Phase A to Phase B will see the about 1/2 value and it will be a decrease.

5. Prior experiments on my panel indicate that phase B is about the same.

6. If you suspect a neutral problem, then start with the changing 100 W test load. This will reduce the possibility of creating an excessive voltage change.

7. If (5) seems reasonable, less than 0.5 V, then try the 10 A load.

8. The 1/2 figure above applies if the neutral wire size is the same as the hot size.

With the above as a guide provide more information.

.
 
Bad editing..

Bad editing..

Thanks for the replies. Yes, infinity, the total is 248V. 238 would not be possible with 124 each. Did the measuring yesterday, just bad editing today (or maybe none at all). Pton, I was thinking about the neutral issue at first to correct the dimming, but I have never seen a single circuit change the voltage balance like this. Trying to put out one (potential) fire at a time..
 
161205-1752 EST

With the measurements I outlined above you can estimate the neutral impedance vs a hot leg impedance. An assumption to this is that the greater transformer impedance is associated with the secondary, primary to secondary leakage inductance combined with secondary resistance.

.
 
joeyd233 said:
Hello out there. Here is my situation: 1993 home, all original electrical. I have a 15A 110V circuit that seems to imbalance the per-leg voltage in the home. With the main off, I read 124V to N per leg, 238V leg to leg coming off the pole. With the suspect circuit off, and the main back on, I still see 123V to N each, and the 238 leg to leg. Nothing abnormal. Power up the suspect family room circuit, and I now have 128V to N, and 120V to N. Full running load on that circuit is about 5A, and it appears to be #10 or 12 conductor. (I will confirm, and is roughly 100 ft run). Load is a TV, fan with dimmable lights, modem, router, and a few PC's, tablet chargers, etc. I have been unable to isolate a single load causing the issue, so I am wondering if this is even a real problem, aka, is it within code? Owners are experiencing lights dimming and brightening throughout other circuits as appliances cycle, ie fridge, washer, etc. They say it is worse with colder temps outside, especially cold mornings. (30's and 40's F) Any thoughts are welcomed here.
Click to expand...
Call the poco, especially if there is an overhead service involved. I would normally tell a customer to go ahead and call them if I'm told these are the symptoms, but my hunch is even stronger with the cold weather effect you describe.
 
Seems to be N related.

Seems to be N related.

All,
Thanks for the ideas on what to look for. My consensus is that the questionable circuit has a bad N. When this branch is off, the home voltages are balanced, and the dimming seems to go away. When that circuit is on, I see the change in voltage per leg. Just need to buckle down and find the culprit.:thumbsup: Is there any harm in changing the breaker to an AFCI -- just in case?
 
joeyd233 said:
All,
Thanks for the ideas on what to look for. My consensus is that the questionable circuit has a bad N. When this branch is off, the home voltages are balanced, and the dimming seems to go away. When that circuit is on, I see the change in voltage per leg. Just need to buckle down and find the culprit.:thumbsup: Is there any harm in changing the breaker to an AFCI -- just in case?
Click to expand...

I don't think the one circuit has a bad neutral. If that was the case you would just not have a circuit, unless it is a MWBC. You need to check the neutral from the weatherhead/meter all the way to the panel.

ETA: The longer you can stay away from an AFCI breaker the better, IMO!
 
joeyd233 said:
All,
Thanks for the ideas on what to look for. My consensus is that the questionable circuit has a bad N. When this branch is off, the home voltages are balanced, and the dimming seems to go away. When that circuit is on, I see the change in voltage per leg. Just need to buckle down and find the culprit.:thumbsup: Is there any harm in changing the breaker to an AFCI -- just in case?
Click to expand...

Loads between the two ungrounded lines are apparently balanced until you throw the loads on this one circuit on one side and unbalance the system.

Bad neutral on just that one two wire circuit will not cause problems for the rest of the system - just that one circuit. Bad neutral ahead of multiple circuits that are balanced by that neutral will impact all of the downstream circuits from the bad neutral point.

Best way to check is to put a heavy load from ungrounded conductor to neutral, 1500 watt heat gun works great, hair dryer also works. Connect it at the main panel (I have a pigtail with alligator clips to connect such a temporary load for testing such things) and watch what voltage does when you switch load on/off. Better yet is to turn off the main so you don't damage sensitive electronics should you throw the voltage to neutral off too far with your testing and maybe apply a 100 watt light bulb load to the other line at the same time. If the light gets brighter when you turn on the heater you probably don't even need to know what the actual voltage was - this is a pretty good sign the neutral is bad somewhere between your test point and the source.
 
ptonsparky said:
The only good thing about putting in an AFCI is that, with luck, there would be one less AFCI in the world.
Click to expand...
Yup.

joeyd223, this is not a problem an AFCI is intended to protect you from, the AFCI itself could possibly be damaged from this condition as well.
 
joeyd233 said:
Owners are experiencing lights dimming and brightening throughout other circuits as appliances cycle, ie fridge, washer, etc. They say it is worse with colder temps outside, especially cold mornings. (30's and 40's F) Any thoughts are welcomed here.
Click to expand...

TexasMaster said:
Call the poco, especially if there is an overhead service involved. I would normally tell a customer to go ahead and call them if I'm told these are the symptoms, but my hunch is even stronger with the cold weather effect you describe.
Click to expand...


I agree to call the POCO. There is a bad neutral and the only place you can check is the neutral connection in the panel. Could be the weatherhead, splices in the cable, tap at transformer or bad connection at the meter base. You do need the power company there to cut the meter seal to start to check things out.

Had one like this last summer and they decided just to replace the cable as it had been spliced many times and didn't look to be in very good condition. It didn't matter to me as the problem went away with a new cable. The homeowner was happy.
 
joeyd233:

I thought you had a good idea about the problem until you brought up AFCI.

I made the assumption that your voltage measuremernts were at the main panel and hopefully at the main lugs into the panel. Now I am not sure of this.

The measurements need to be directly on the conductor into the panel, not on the lug. Does it make a difference? Possibly. When you are testing why take a chance of the problem being at the lug? Just get directly on the wire.

Uploading photo does not work today.

Some bench tests on a small transformer where the measurements are directly at the transformer.

Source voltage was maintained at 124 V 60 Hz.

The transformer is a 175 VA "Signal Transformer Co",
One primary --- wired 120 V, DC resistance about 1.5 ohms. (Two coils in parallel.)
Two secondaries wired series adding,
each secondary --- nominal 12 V, DC resistance about 0.05 ohms.
measured turns ratio primary to secondary via no load voltage measurement 124/14.05 = 8.826 .

Test load applied to only one secondary about 2 ohms.

Pri V ---- Sec 1 V ---- Sec 2 V --- Sec 1 + Sec 2
124.0 --- 14.05 ------ 14.06 ----- 28.14 ------------ No load on either secondary
124.0 --- 13.51 ------ 13.86 ----- 27.38 ------------ Sec 1 loaded with 2 ohms, I = 6.7 A.
Difference
1.0 ------ 0.54 ------- 0.20 ------ 0.72

Load current reflected to primary is about 6.7/8.826 = 0.76 A.
1.5 ohms * 0.76 A = 1.14 V. Thus, the effective primary voltage is 124 - 1.14 = 122.9 V into an ideal transformer of ratio 8.826. An unloaded output secondary should be 13.9 V and I measured 13.86 V. A good correlation.

In this experiment the voltage change of the unloaded secondary is primarily a result of the common impedance of the primary, the 1.5 ohm primary DC resistance..

When you add external resistance to the transformer sedondaries, i.e., distribution wires, then the voltage changes change. and for the more loaded side the change is bigger in the down direction.

As the distribution impedance increases then the change in voltage of the least loaded side may increase as the loaded side becomes more loaded.

You definitely have a neutral problem somewhere, and most likely a power company problem. You have been told this several times now. Do your measurements at the input to the main panel, and tell us that is where the measurements were made.

If you get good measurements at the input wires to the main panel, meaning you see the voltage problem at that point, then you can be quite sure it is a power company problem, and you can prove it. Could you have some other problem? Possibly.

.

.
 
Problem Solved

Problem Solved

Final problem ended up (as suspected) a bad neutral connection. Called out the poco, an it turned out the bad connection location was at their side N wire landing lug in the meter box. Although technically a "customer" side problem, since he had the wire loose to diagnose, he was nice enough to clean up the minor corrosion, grease it up, and reconnect. Voltage imbalance gone. No more dimming, happy homeowner. He did tell me that some imbalance is acceptable, but neither leg should drop below 110V under a 10A load. Anything lower, issues are present. Thanks for all the feedback!!
 
joeyd233 said:
He did tell me that some imbalance is acceptable, but neither leg should drop below 110V under a 10A load. Anything lower, issues are present. Thanks for all the feedback!!
Click to expand...
Normal voltage drop in the neutral conductor is the cause of such imbalance, but should be minimal. I was called to investigate issues with about a 1900 foot run of conductors that were not oversized for VD and the end result looked like a classic case of open neutral yet we confirmed there was no issue with the neutral other then the long undersized run.
 
Status
Not open for further replies.
Top