Autotransformer with floating neutral: current imbalance?

[Rampage_Rick]

It's been a long day and my brain has long since departed. Here's the latest issue I'm working on:

I've got a building with a 400A 120/208Y service

They have a water pump that was recently replaced, now popping the breaker.

Pump is 15HP 3PH 575V 19A nameplate, running about 17A

It's connected via a 30KVA 208Y/600Y autotransformer, with X0 left floating

Now I calculate that with the pump running around 17A the autotransformer should be drawing about 48A, yet it's tripping a 60A breaker after about 10 minutes.

The only thing I can thing of is that perhaps there is a voltage imbalance on the source side. If the legs are unevenly loaded ahead of the autotransformer, could that create enough voltage imbalance and/or current imbalance to pop the breaker? Is it worth the effort to replace the 3C6 cable feeding the autotransformer with 4C6?

 

[GoldDigger]

It's been a long day and my brain has long since departed. Here's the latest issue I'm working on:

I've got a building with a 400A 120/208Y service

They have a water pump that was recently replaced, now popping the breaker.

Pump is 15HP 3PH 575V 19A nameplate, running about 17A

It's connected via a 30KVA 208Y/600Y autotransformer, with X0 left floating

Now I calculate that with the pump running around 17A the autotransformer should be drawing about 48A, yet it's tripping a 60A breaker after about 10 minutes.

The only thing I can thing of is that perhaps there is a voltage imbalance on the source side. If the legs are unevenly loaded ahead of the autotransformer, could that create enough voltage imbalance and/or current imbalance to pop the breaker? Is it worth the effort to replace the 3C6 cable feeding the autotransformer with 4C6?

Since you have a wye-wye transformer, it is not mandatory to leave the X0 floating. Any voltage imbalance across the three lines on the primary side will simply be carries across to the secondary side without the risk of circulating currents. With the X0 open, the result will be a voltage and current balance in the secondary, but at the risk of a current imbalance in the primary instead, I think.

But either way, there could end up being a current imbalance that trips one pole of the three-pole breaker. Put an amp clamp on all three secondary and all three primary wires. (one at a time, of course.)

 

[Rampage_Rick]

I did throw a clamp-on meter on the legs, got 50-51 amps on two of them and 53 on the third.

Also took measurements with the pump disconnected (no load on the autotransformer) and saw 4 amps on that third leg.

The current imbalance in the primary is the one I'm wondering about. If there's some other appliance that's starting up and pushing things further out of balance... I saw a single pole breaker for an oven, and there's a ton of single-pole circuits for fridges, beer coolers, etc. The amusing part is I just ordered a bluetooth adapter for my Agilent DMM that will allow me to do data logging on my phone, but that won't be here for another week.

 

[augie47]

IMO, your breaker is simply undersized.
You are measuring a 53-54 amp load.
Taking transformer losses into account, once the pump with the 19 amp nameplate rating is fully loaded I would expect to see a primary amperage a bit higher than you are measuring.
The 30 KVA 208 transformer would be allowed a primary breaker rating more in the 100 amp range. Of curse, your conductors would need to match.
Unsure of the type "cable" you have for your primary, but sized at #6 it should allow you to at least change your breaker to a 70 amp. Since the 60 is holding for 10 min., a 70 might solve the problem.
That would be my 1st step.

 

[Smart $]

IMO, your breaker is simply undersized.

Perhaps it is the start up current combined with steady state current which is tripping it...

The 30 KVA 208 transformer would be allowed a primary breaker rating more in the 100 amp range. Of curse, your conductors would need to match.

Perhaps not... for a motor-only load. The autotransformer aspect is throwing a curve in my recollection at the moment.

Unsure of the type "cable" you have for your primary, but sized at #6 it should allow you to at least change your breaker to a 70 amp. Since the 60 is holding for 10 min., a 70 might solve the problem.
That would be my 1st step.

That or perhaps even higher. Code permits up to 250% for motor start up... which comes out to a max' 110A standard rating.

 

[Phil Corso]

Rampage...

1) Current measurements indicate some imbalance, but insignificant impact at this time thus far. What are Ph-Ph, Ph-N and Ph-G measurements for both 208V and 600V systems?

2) Is Clamp-on meter a 'True' RMS type ?

3) Regarding single 4-Amp measurement... can you measure current in 208V system EGC?

Regards, Phil Corso

 

[Jraef]

This may be one of those problems that can only be exposed via data logging, as you already suspect. Tripping 10 minutes after starting is an eon in electrical terms. Although it's unusual for a basic TM breaker to trip on current imbalance unless designed to do so, you haven't told us what the 60A breaker is, and I know that some are indeed designed to do so. A larger single phase load, such as an HVAC, refrigeration or air compressor that comes on periodically and causes that severe imbalance could be a culprit and if it does not happen while you are standing there holding the clamp-on when it happens, you will not see it. But the problem I have with that is that this is a pump, which means it has a starter of some sort, which means an overload relay and a better than 50% chance that whatever that OL relay is will have current imbalance protection, which SHOULD trip quicker than the breaker. And also to that point, ANY long term load related issues SHOULD be getting picked up by that OL relay long before a breaker trips. If you can, please describe the rest of the components, in gory detail.

So if not that, then there are three other, maybe more likely, scenarios in my opinion.

1) as pointed out by augie47, a 60A breaker on a 53 or 54A load is too small. You are not supposed to load that breaker beyond 80%, so 48A. Sure, we know the breaker CAN handle 60A continuously, but only in controlled circumstances, packaging etc. you are most likely outside of that envelope.

2) after 10 minutes of running your transformer is heating up*, increasing its internal resistance, dropping the output voltage, which increases the motor current, which heats up the transformer more. Normally this would only take place up to a point of stasis, but the breaker it tripping before you get there. * Transformers have a positive temperature coefficient of resistance, meaning as the temperature rises, so does the resistance in the copper windings. It's usually not much, but is more pronounced in autotransformer configurations because the primary and secondary use the same copper. That's why motor starting autotransformers are equipped with thermal cutouts that prevent rapid restarts.

3) the new motor is now "energy efficient" whereas the old one was not. EE motors are known to have much higher magnetic inrush than older designs as a result of some of the steps to improve RUNNING efficiency. There are documented reports of up to 2000% of FLA. that, combined with the transformer having magnetic inrush as well, is stressing the already marginally sized breaker as discussed in #1 to allow it to trip more readily on otherwise marginally acceptable circumstances.

Or all of the above...

 

[Rampage_Rick]

I guess what broke my brain is the fact that the previous 15HP pump ran fine but this one had issues. The pump supplier's tech was flummoxed, but then he couldn't even figure out why he was seeing a 50 amps at the panel and 17 at the pump...

I ended up swapping the NA3P60 breaker for a 70, and I replaced the aluminum 3C6 with 4C4. That was adventure in of itself, involving 4-6" of sewage in the crawlspace, hipwaders, a large hose, and possibly hepatitis. (caused by another pump issue: septic clogged with tampon applicators, etc)

On the up side, my voltages at the pump went up by ~20V, the current dropped ~3.5A, and I'm now seeing different amp/volt readings between the phases. This is after 6 hours runtime:

 

[GoldDigger]

A 20% drop in secondary current should correspond to roughly a 20% drop in primary current. There is your margin back. I would not be surprised if a good 60A breaker would not hold, although the original might still have issues.
If the entire pump was replaced and not just the motor, there might actually be more water pumped and therefore more mechanical load on the new motor than on the old one.
Even if the GPM or even the entire pump curve for the new pump matched the old one, impeller wear or other factors might have had the old pump moving less water.