Phantom Trips

Open Neutral

Senior Member
Location
Inside the Beltway
Occupation
Engineer
My client friend asked me about a phantom trip issue.

This is the house with PSE's new 100KVA transformer, a 600A fused main, and four 200A panels. Three of them are adjacent to the main, the last has a 200A disconnect there, and the panel itself is in the Mechanical Room at the other end of the house. All panels are new (well, when the job started a few years ago...) SquareD including the fused main.

A few days ago, the MR's panel main and the upstairs disconnect feeding it both tripped. No branch breaker was tripped. There was no unusual work going on that might instigate a trip.

The other three panels have some ilk of surge protectors (Imagery coming), this one does not. But in my thinking all four panels are obviously in parallel, and the three surge suppressors are, too.

The GC called the EC, and there it gets more interesting. They reported that they had recently seen similar anomalies on other SqD panels, and had one case where a shop vac started on another site would trip the SqD main but not a branch. This was repeatable. They've started using C-H panels instead.

I'm at a loss to explain this double-trip. The vacuum cleaner story makes me think "arc-fault grief" but these were {sub}main breakers, not branches.
Is it worth talking to SqD given a lack of data?
 

tortuga

Code Historian
Location
Oregon
Occupation
Electrical Design
My client friend asked me about a phantom trip issue.

This is the house with PSE's new 100KVA transformer, a 600A fused main, and four 200A panels. Three of them are adjacent to the main, the last has a 200A disconnect there, and the panel itself is in the Mechanical Room at the other end of the house. All panels are new (well, when the job started a few years ago...) SquareD including the fused main.

A few days ago, the MR's panel main and the upstairs disconnect feeding it both tripped. No branch breaker was tripped. There was no unusual work going on that might instigate a trip.

The other three panels have some ilk of surge protectors (Imagery coming), this one does not. But in my thinking all four panels are obviously in parallel, and the three surge suppressors are, too.

The GC called the EC, and there it gets more interesting. They reported that they had recently seen similar anomalies on other SqD panels, and had one case where a shop vac started on another site would trip the SqD main but not a branch. This was repeatable. They've started using C-H panels instead.

I'm at a loss to explain this double-trip. The vacuum cleaner story makes me think "arc-fault grief" but these were {sub}main breakers, not branches.
Is it worth talking to SqD given a lack of data?
What does PSE mean?
Sounds like a selective coordination issue.
Why does the mechanical room panel have a main, if there is a feeder main ahead of it?
Some circuit in your mechanical room panel might be oversized and before the branch breaker can trip the main's do.
 

synchro

Senior Member
Location
Chicago, IL
Occupation
EE
A few days ago, the MR's panel main and the upstairs disconnect feeding it both tripped. No branch breaker was tripped. There was no unusual work going on that might instigate a trip.

Is it worth talking to SqD given a lack of data?
It might be worth verifying with Square-D that the main breaker is not a counterfeit. And while you're at it you could see if they can offer any insight about a possible cause. I haven't heard a lot about counterfeit breakers lately but I know they are out there.
 

Open Neutral

Senior Member
Location
Inside the Beltway
Occupation
Engineer
It might be worth verifying with Square-D that the main breaker is not a counterfeit. And while you're at it you could see if they can offer any insight about a possible cause. I haven't heard a lot about counterfeit breakers lately but I know they are out there.
What does PSE mean?
Sounds like a selective coordination issue.
Why does the mechanical room panel have a main, if there is a feeder main ahead of it?
Some circuit in your mechanical room panel might be oversized and before the branch breaker can trip the main's do.
PSE == Puget Sound Energy

See crude 1-line.

There are three 200A PB's near the main. We needed to have fused main to solve AIC issues from the large xfmr.
The a,b, & c, are actually those with the {imho} ugly sllding exclusion PSE vs. Gen. breakers scheme.
d has a {sub}main because that's what the EC did; I can't object.

I had not considered counterfeit's but the EC is not a one-man here-today outfit.
The mech panel is new with new HVAC gear; and both upstream breakers tripped.
 

Attachments

tortuga

Code Historian
Location
Oregon
Occupation
Electrical Design
Gotchya, my guess remains its a two part problem,
a piece of mechanical equipment is getting a locked rotor or surge, and should be tripping its branch breaker.
Instead the feeder breakers are tripping before the branch.

When I ran into a similar situation, with large air compressors, the way I found it was with a moving around a data logger on the feeder and a few branches. Had to deal with the compressor and the breakers.

Are they perhaps frame breakers with adjustments?
600A fuses are something like 200,000
whats the voltage and model# and AIC rating of the 200A feeder breakers especially the ones in d?
 

Jraef

Moderator
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
Yep, first thing I would look into would be trip coordination. 600A service is more commercial/industrial where that would be a common practice, but someone who normally only does residential likely didn’t think of doing that. In addition, residential breakers are not known for being as accurate and repeatable when it comes to trip values, they tend to be factory fixed trips with a +- % that can get you into trouble on trying to coordinate large equipment to avoid nuisance tripping.
 

synchro

Senior Member
Location
Chicago, IL
Occupation
EE
My client friend asked me about a phantom trip issue.
...
The GC called the EC, and there it gets more interesting. They reported that they had recently seen similar anomalies on other SqD panels, and had one case where a shop vac started on another site would trip the SqD main but not a branch. This was repeatable. They've started using C-H panels instead.
I wonder if it's possible that "smart" switches/loads on multiple branch circuits could be falsely triggered by a shop vac, etc., and then collectively they might cause a large enough surge to trip the main breaker. That's a stretch, but the shop vac tripping the main is also an unusual situation.
I'm guessing that a new home with a 600A service probably has a lot of electronically controlled loads and so some strange things might be possible.
 

tortuga

Code Historian
Location
Oregon
Occupation
Electrical Design
Could an irratic phase to ground fault in a large load pull more fault current than say a locked rotor(L-L fault)?
My guess is a newer single phase pad mount 100kVA transformer would have 1.8Z ~46000 amps L-N and 23000 L-L available at the transformer.
If its a residential Square D QO then the mains are 22,000 short circuit current rating and the branches are 10k.
 

Open Neutral

Senior Member
Location
Inside the Beltway
Occupation
Engineer
All the panels are SqD. The 600A fused disconnect exists as the "series coordination" given the AIC exceeds the ratings of the varied panel-mains. The two breakers that opened were new, 200A, SqD, the ?Q? flavor, not Homeline or whatever it is.

No loads on that panel are "smart" -- all are GSHP or related.
 

tortuga

Code Historian
Location
Oregon
Occupation
Electrical Design
Do you know the details on the transformer? Is it a typical 100kVA pad mount, ug service?
Lateral distance and size?
 

tortuga

Code Historian
Location
Oregon
Occupation
Electrical Design
Thanks for the photos, I love it when we get photos, makes it easier.
Just back of the napkin numbers I only come up with about 18500 Amps at the service disconnect and around 5300 in the mechanical room panel thats a 100kVA 1.8Z transformer. The square D QO main breakers come with a 22k rated main, but I might be way off.
I'd ammertre the loads, cycle all the breakers and good old smell test that panel D. have a look in that large tap box, check the temp of the terminations etc.
Could be a bad branch breaker, loose connection.
What are the big motor loads in the mechanical room?
 

Open Neutral

Senior Member
Location
Inside the Beltway
Occupation
Engineer
There are 3 GSHP's, one w/air handler, a sewage lift station, and several 3HP pumps for the koi pond.

These are not your average resi electricians. We compete for their time with the nearby large Bremerton naval base, and their rates show it. The day of the transformer cutovers, I'm told they had 10+ folks on site (in addition to the PSE contractors.) Not only was the 500 ft. primary pull done, and the dual 350MCM, and the CT box & main to the PB's in the house, but also the replacement run to the second building, from the original 37.5KVA xfmr & its work.

My SWAG on the AIC was many times yours, but it was the first time I've ever done one... I don't recall what PSE & the EC said but it was close to mine and several x the SqD rating.
 

synchro

Senior Member
Location
Chicago, IL
Occupation
EE
Do the GSHP's and/or other motors have VFD's or soft starters? If not, then if several motors just happen to start about the same time (an unlikely but still possible occurrence) then the main breaker might trip. Just a thought.
 

Open Neutral

Senior Member
Location
Inside the Beltway
Occupation
Engineer
Not as far as I know. I'll have instrumentation on them Real Soon Now to track usage (see my rant about The Energy Detective folks..) but that's months away. The three GSHP's are water->air for summer AC, water->water for domestic HW, and water-> water for the hydronic floors. {& spa}.
 

tortuga

Code Historian
Location
Oregon
Occupation
Electrical Design
Are you the EOR on the job or just helping your friend?
What size/type of wire goes the 180' to panel D.
There are 3 GSHP's, one w/air handler, a sewage lift station, and several 3HP pumps for the koi pond.
Yeah could be those electricians need to move a load off that Panel D.
I would not rely on energy use meters, rather use a known good data logger, I use an old Hobo, I bet those electricians have something.

If those ground source heat pumps are large they could have MCA upwards of 35 Amps each @240 single phase.
The 3 HP motors would be 17 each
Sewage pump might be the same or slightly less
that could be 160 Amps of load right there.
Whats the NEC calculated load on panel D?

As an aside I ran the fault calcs at the service a few different ways to refresh my memory, and with the given
100kVA
1.8Z
60 FT
2 Conductors
350kCMIL AL
Magnetic conduit or not I dont come up with more than 18000 Amps, I did not do motor contribution though.
Curious what the number was
 

Open Neutral

Senior Member
Location
Inside the Beltway
Occupation
Engineer
As I said in my initial post... in WA state, I'm licensed to ride the ferry and buy dinner.

It's a 200A panel, but I strongly doubt that all three HP's could be running at the same time. Among other things, the air unit is used for summer cooling so it would not be enabled now. And the 3HP pumps are running at less than 50% speed. The third is off, it's for backflushing.

The EC is staking their claim with a surge being the cause. They're adding a surge eater at D.

If it repeats, we'll take further actions.
 

kingpb

Senior Member
All the panels are SqD. The 600A fused disconnect exists as the "series coordination" given the AIC exceeds the ratings of the varied panel-mains. The two breakers that opened were new, 200A, SqD, the ?Q? flavor, not Homeline or whatever it is.

No loads on that panel are "smart" -- all are GSHP or related.
Are the devices actually listed as rated as a Series Combination? What are the actual device model numbers? Has a time current coordination curve been developed to help understand problem? BTW: you cannot use series breakers if the the rated motor load on the panel exceeds 1% of the panel AIC rating.
 

Open Neutral

Senior Member
Location
Inside the Beltway
Occupation
Engineer
Good questions, I'm sure. But given that L&I [state AHJ] has been out and not red-tagged anything, I assume the best. There are Series labels on the cabinets; I don't if L&I required them. I know the SqD fused disconnect was purchased to fulfill the series coordination requirement.
 
Top