3 phase voltage fluctuations with current spikes

[Master_Krafter]

Service call for frequent shutdown of computers and lighting flickers.

3 phase service to an office warehouse. Mostly 129v circuits aside from basic hot water(30a 208v) and ac(17a 208v). Showing some wild readings.
LN 1 - G 134v
LN 2 - G 115v
LN 3 - G 111v
this is with basic operation.
Add a microwave on LN 3 and the voltage jumps
LN 1 - G 140v
LN 2 - G 118v
LN 3 - G 101 - 108v
utility provider is reporting 90a spikes but no significant voltage changes. Grounding and bonding is solid. Anyone have some knowledge to share?

 

[Tulsa Electrician]

Sounds like a neutral issue to me.
If you add all the voltages reading and divide by 1.73 ( square root of 3)or 208. Then I would say it's a neutral issue.

Pretty sure you meant 120 v circuits.

 

[jim dungar]

What are your Line-Neutral and your Neutral-Ground voltages?

 

[Master_Krafter]

Thanks for the reply Tulsa. Leaning back on the utility company to check their transformer. The ground and neutral are bonded.

 

[Little Bill]

Thanks for the reply Tulsa. Leaning back on the utility company to check their transformer. The ground and neutral are bonded.

You still need to check L-N and N-G.

 

[Master_Krafter]

Readings are the same L-N and N-G but we will double check

 

[Crash117]

Where’s your meter, it’s possible you have a bad connection in your meter or gear compartment? That service can’t come in straight from the utility transformer to that panel.

 

[ptonsparky]

I cast another vote for a Neutral issue.

 

[ptonsparky]

Why are all those whites pigtailed? Pigtailing the EGs or adding an EG bar is a better choice if you're short on spaces.

 

[Master_Krafter]

Where’s your meter, it’s possible you have a bad connection in your meter or gear compartment? That service can’t come in straight from the utility transformer to that panel.

I can upload a picture but everything seemed good at the meter

 

[Master_Krafter]

Why are all those whites pigtailed? Pigtailing the EGs or adding an EG bar is a better choice if you're short on spaces.

Fully agree, this is the way we found it. Added to the task list.

 

[Master_Krafter]

Also changed phases on a single phase sub panel from LN-2/LN-3 to LN-1/LN-2. That did seem to help the rise and drop in voltage slightly. Have not seen it over 136v LN-1 and not under 106v LN-3
Slight improvement but not solved

 

[winnie]

Voltage instability where the L-L is good but L-N voltage changes with L-N loads is a pretty solid indicator of an issue with the neutral.

The L-N voltages will be stable up to the point where the neutral is weak or open. If there are multiple neutral to ground bonds, then parallel current flow on the grounded metal can mask a problem with the neutral.

If you can get a large load that fluctuates in a well defined fashion (say a 500W incandescent lamp on a blinker) this can help trace issues.

 

[ptonsparky]

Underground?

 

[Crash117]

Underground?

Get the utility to run the beast of burden test at the meter. If the problem exists there, then it’s a problem on the power from the utility. It sure sounds like Winnie said, if you’ve got a neutral problem it can be masked or covered up by multiple bond locations.

 

[ptonsparky]

A hand held hair dryer or a couple milk house heaters are usually enough. All you need is a place to put your meters.

 

[Jraef]

I cast another vote for a Neutral issue.

Just had one recently that turned out to be a bad meter connection. Utility measured at their pole and proclaimed it was all good at their end, but we were seeing voltage fluctuations. Convinced the utility to pull the meter, there it was, hiding under there. Had to replace the meter socket.

User kept insisting this was a utility problem and wanted them to absorb the replacement costs for the socket and the damaged equipment (a couple of small VFDs and a fried PLC power supply)… I had to set him straight about who owns what. He was not a happy camper.

 

[ptonsparky]

Trust but verify:
POCO
OWNER
ENGINEER
EMPLOYEE
YOURSELF.

 

[Master_Krafter]

Voltage instability where the L-L is good but L-N voltage changes with L-N loads is a pretty solid indicator of an issue with the neutral.

The L-N voltages will be stable up to the point where the neutral is weak or open. If there are multiple neutral to ground bonds, then parallel current flow on the grounded metal can mask a problem with the neutral.

If you can get a large load that fluctuates in a well defined fashion (say a 500W incandescent lamp on a blinker) this can help trace issues.

Are you say potentially an internal neutral issu? Open or loose

 

[Master_Krafter]

A hand held hair dryer or a couple milk house heaters are usually enough. All you need is a place to put your meters.

We can repeat the voltage fluctuations adding load but not able to narrow it down yet.