Verifying Phases in Electrical Panel

Status
Not open for further replies.

L1ght1ngGuy

Member
Location
SF Bay Area
Does anyone know how to best verify that the correct phases were landed to the correct main lug? I have a three phase four wire system and I want to verify that Phase A/B/C were actually landed correctly at the panel. Thanks.
 

Ponchik

Senior Member
Location
CA
Occupation
Electronologist
Does anyone know how to best verify that the correct phases were landed to the correct main lug? I have a three phase four wire system and I want to verify that Phase A/B/C were actually landed correctly at the panel. Thanks.

If I remember correctly, you take the circuit breaker position number, then divide it by 3 the remainder is the phase.

For example:
Black = 1
Red = 2
Blue = 3

#1: Circuit breaker position 16

16/3=5 remainder 1

So circuit breaker 16 is phase one.

#2: Position 7

11/3=3 remainder 2

This position is phase two
 

ActionDave

Chief Moderator
Staff member
Location
Durango, CO, 10 h 20 min from the winged horses.
Occupation
Licensed Electrician
Does anyone know how to best verify that the correct phases were landed to the correct main lug? I have a three phase four wire system and I want to verify that Phase A/B/C were actually landed correctly at the panel. Thanks.
Other than keeping rotation the same throughout a building I can't see why it would matter.

As far a checking the easiest would be to kill power, isolate the feeder wires, jumper each phase to ground and test with an ohm meter. If killing power is not an option then checking with an amp meter would work pretty good I imagine.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
161130-2429 EDT

Light1ngGuy:

To what do you want to reference? Its all relative. However, there is an absolute rotation direction (only two).

If you have some location in your facility that is to be the reference, then you can work from it. One way is with a small stepdown transformer. Use the transformer to create an aufio signal that modulates a transmitter, and then demodulate that audio signal with a receiver where you want to do the phase comparison.

I have never done this, so I need to try it. But not tonight.

.
 
Last edited:

Smart $

Esteemed Member
Location
Ohio
As asked by others, to what end? A, B, and C are arbitrary labels for Line 1, Line 2, and Line 3—which are arbitrary labels to some degree in and of themselves—representing the conventional connections further up line going back to the grid, the largest reference in the U.S. (eventually the power plants, but they all sync' to the grid).

Unless your POCO provides you with their distribution A, B, C, how the phases are connected to your service disconnecting means is all that matters for your site with the exception of rotation. To verify rotation, you either need a rotation meter which complies to NEMA standards, or a standard 3Ø NEMA motor, connected, bumped, and rotation verified (looking at end of output shaft, clockwise).

If rotation is CCW, you have to flop two lines, preferably B and C. Basically you want to keep POCO "A-phase" as your reference.
 

ron

Senior Member
For whatever reason you want it to be, if you want to match phase A of one panel to phase A of another panelboard, use a voltmeter and it should be 0V from A to A. Same for B to B and C to C.
 

iwire

Moderator
Staff member
Location
Massachusetts
For whatever reason you want it to be, if you want to match phase A of one panel to phase A of another panelboard, use a voltmeter and it should be 0V from A to A. Same for B to B and C to C.

Going to get a little funky if there are SDSs involved but still could be done.
 

L1ght1ngGuy

Member
Location
SF Bay Area
For whatever reason you want it to be, if you want to match phase A of one panel to phase A of another panelboard, use a voltmeter and it should be 0V from A to A. Same for B to B and C to C.


s This is what I was thinking. I am installing CTs to measure current flow for each individual circuit. I need to make sure I have a reference voltage for what's feeding each phase. I need to make sure the breaker I'm monitoring and the various phases I'm monitoring are correct. I know sometimes the phases are landed differently at the panel if they are not labeled correctly. Thanks.
 

ActionDave

Chief Moderator
Staff member
Location
Durango, CO, 10 h 20 min from the winged horses.
Occupation
Licensed Electrician
If the phases are consistent, and the transformers are all standard delta wye, each transformer will shift the same 30 degrees, so it should still work out.

Getting a long enough meter leads lead can be another issue. We had to check phasing at a cell site one time so the boss made a 500' jumper out of a reel of wire and alligator clips to go from the service to the shelter.
 

jumper

Senior Member
If I remember correctly, you take the circuit breaker position number, then divide it by 3 the remainder is the phase.

For example:
Black = 1
Red = 2
Blue = 3

#1: Circuit breaker position 16

16/3=5 remainder 1

So circuit breaker 16 is phase one.

#2: Position 7

11/3=3 remainder 2

This position is phase two

Nope.:happyno:

Divide breaker position number by 6.

Remainder:

1 or 2 = P1 Black/Brown
3 or 4 = P2 Red/Orange(Purple)
5 or 0 = P3 Blue/Yellow
 

Fnewman

Senior Member
Location
Dublin, GA
Occupation
Sr. Electrical Engineering Manager at Larson Engineering
The voltmeter test will work, provided the possible voltage difference will be within the meter's range. Old school electricians used several light bulbs in series - in fact, a light bulb arrangement was used for synchronizing a generator to the bus before the days of digital controllers. Some of those are likely still in use.
 

JFletcher

Senior Member
Location
Williamsburg, VA
If the main concern is phase rotation, hook up a small 3ph motor and check its rotation (you can buy a used one from ebay ~$125). All of the phases should have been taped (or colored jackets) prior to the pull, so "A" phase in the MDP should be "A" to the panel it feeds, and so on. If the wires coming from the MDP are say black for A, red for B, and blue for C, they could be landed ABC, BCA, or CAB, the rotation will remain unchanged. ABC to BAC, ACB, or CBA will be reversed. or this:

"Getting a long enough meter leads lead can be another issue. We had to check phasing at a cell site one time so the boss made a 500' jumper out of a reel of wire and alligator clips to go from the service to the shelter."

Checking A to A, B to B, C to C would show ~0V, or continuity (eta: what Ron wrote). 500' of 18ga wire and a few alligator clips would be cheaper than a 3ph motor.
 

Ponchik

Senior Member
Location
CA
Occupation
Electronologist
Nope.:happyno:

Divide breaker position number by 6.

Remainder:

1 or 2 = P1 Black/Brown
3 or 4 = P2 Red/Orange(Purple)
5 or 0 = P3 Blue/Yellow

It seems like 3 works as well as the 6. :happyyes:

I tried 4 different examples and they all came out the same phase. Or maybe I am doing something wrong.
 

ActionDave

Chief Moderator
Staff member
Location
Durango, CO, 10 h 20 min from the winged horses.
Occupation
Licensed Electrician
s This is what I was thinking. I am installing CTs to measure current flow for each individual circuit. I need to make sure I have a reference voltage for what's feeding each phase. I need to make sure the breaker I'm monitoring and the various phases I'm monitoring are correct. I know sometimes the phases are landed differently at the panel if they are not labeled correctly. Thanks.
Huh? What do labels have to do with anything?

There is a wire landed on a lug that feeds A phase, that wire heats up the A phase buss, same thing happens for B phase and C phase.
 

Smart $

Esteemed Member
Location
Ohio
It seems like 3 works as well as the 6. :happyyes:

I tried 4 different examples and they all came out the same phase. Or maybe I am doing something wrong.
The rule changes a bit...

Division by 3:
Remainder of 1 -> ØA
Remainder of 2 -> ØB
Remainder of 0 -> ØC
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
If a panel is numbered with all the odd numbers on the left and all the even numbers on the right, then you have to divide by six since both breakers in the same horizontal row are on the same stab.
It is even more complicated if the panel is also numbered on the basis of half width breakers so that, for example, 1 and 3 as well as 2 and 4 are all on the same stab. For that you would divide by 12.

Sent from my XT1585 using Tapatalk
 

Smart $

Esteemed Member
Location
Ohio
...
It is even more complicated if the panel is also numbered on the basis of half width breakers so that, for example, 1 and 3 as well as 2 and 4 are all on the same stab. For that you would divide by 12.
The most accepted convention I've seen for tandem breaker numbering is number by space rather than by circuit and appending "a" and "b" to the space number to reflect the tandem breaker circuits. This way the "divide the circuit number by 6" paradigm remains true to its origin.
 
Status
Not open for further replies.
Top