Progessively Worsening Transformer Inrush?

[big john]

Had an unusual trouble shooting call for unexplained breaker that began occasionally tripping after transition from generator power. The frequency of the trips began increasing from once every few months to weekly.

No intentional load changes and no obvious ground faults in the circuit. 225A thermal magnetic breaker replaced and tripping persisted.

Put current monitors on the circuit at the 75kVA 480Δ:208 transformer immediately downstream. With no load on the secondary we got an inrush of 2,300A. I've heard of inrush in the 15 range, but this was severe enough to actually cause cable slap in the conduits, which is definitely unusual for a transformer this small.

Further, for whatever reason the problem was suddenly so bad we couldn't even re-energize without tripping even after repeated attempts and the transformer was removed from service.

The voltages on this transformer were perfectly normal, no ground faults, and in the past once successfully energized there were no overload or overheating conditions.

What could cause a progressively worsening inrush current? I'm guessing some sort of failure within the iron, but I cannot explain it for the life of me.

Any theories?

 

[templdl]

Had an unusual trouble shooting call for unexplained breaker that began occasionally tripping after transition from generator power. The frequency of the trips began increasing from once every few months to weekly.

No intentional load changes and no obvious ground faults in the circuit. 225A thermal magnetic breaker replaced and tripping persisted.

Put current monitors on the circuit at the 75kVA 480Δ:208 transformer immediately downstream. With no load on the secondary we got an inrush of 2,300A. I've heard of inrush in the 15 range, but this was severe enough to actually cause cable slap in the conduits, which is definitely unusual for a transformer this small.

Further, for whatever reason the problem was suddenly so bad we couldn't even re-energize without tripping even after repeated attempts and the transformer was removed from service.

The voltages on this transformer were perfectly normal, no ground faults, and in the past once successfully energized there were no overload or overheating conditions.

What could cause a progressively worsening inrush current? I'm guessing some sort of failure within the iron, but I cannot explain it for the life of me.

Any theories?

Can I assume that you have already run a test for a winding failure?
If not what you have described may be as the result a winding or windings that are progesively failing.

 

[kwired]

Turn to turn fault lessening the total impedance?

Had you measured voltage when you could get it to energize and hold? That could have possibly been an indicator of some problems

 

[big john]

Can I assume that you have already run a test for a winding failure?
If not what you have described may be as the result a winding or windings that are progesively failing.

Only a simple 1kV megger.

I have seen a similar failure before with a simple PT where it would pass a TTR and insulation resistance test, but would blow fuses every single time it was energized. Couldn't explain that either.

Turn to turn fault lessening the total impedance?

Had you measured voltage when you could get it to energize and hold? That could have possibly been an indicator of some problems

When it held we had good voltages. I can't quote them exactly, there was nothing that made me suspect inter-turn failures.

I would be surprised to learn that you could have a turn-to-turn failure severe enough to cause 25x overcurrent, but not bad enough to affect voltage. Also remember that the running current seemed reasonable, only inrush was a problem.

 

[Tony S]

A bit of a long shot. I had a 1850kVA 11/3.3kV transformer doing the same trick, not bad enough to trip the OCB on I_i but bad enough to get me worried closing the OCB.

Due to various faults we had to de-tank the core giving me the chance to re-torque the lamination ties. It dropped the inrush enough to stop me getting nervous when I switched it in to service each week.

 

[mbrooke]

Wild guess: turn to turn shorts or shorts in the lamination increasing eddy currents. Keep in mind that small transformers have mediocre protection compared to large transformers where current differential picks up A LOT that fuses or simple time current curves can not. I can not speak for small transformers, but I know in large ones hard bolted faults are rare compared to other failure modes that are not picked up by pure over current in service.

 

[big john]

...Due to various faults we had to de-tank the core giving me the chance to re-torque the lamination ties. It dropped the inrush enough to stop me getting nervous when I switched it in to service each week.

Wild guess: turn to turn shorts or shorts in the lamination increasing eddy currents...

Yeah, that's where we're leaning, but I can't prove it and don't know enough theory about magnetic circuits to be able to speculate.

I'm almost tempted to do something silly like just take a small control transformer and drill through the laminations to see if deliberately trying to short them out causes a measurable increase in excitation.

We were discussing it, and this is probably the 3rd little dry transformer in a couple years that we've seen where it passed insulation resistance, and it passed TTR, but went into overcurrent when energized. These guys are usually far too small to justify hauling excitation testers out to each one, but I'm interested if there's any other field test to prove this?

Anyone have any resources that might give a good case-study on this?

 

[kwired]

Yeah, that's where we're leaning, but I can't prove it and don't know enough theory about magnetic circuits to be able to speculate.

I'm almost tempted to do something silly like just take a small control transformer and drill through the laminations to see if deliberately trying to short them out causes a measurable increase in excitation.

We were discussing it, and this is probably the 3rd little dry transformer in a couple years that we've seen where it passed insulation resistance, and it passed TTR, but went into overcurrent when energized. These guys are usually far too small to justify hauling excitation testers out to each one, but I'm interested if there's any other field test to prove this?

Anyone have any resources that might give a good case-study on this?

Now you possibly opened this conversation to a whole lot of questions about the installation. Types of loads, high harmonic loads, deteriorating agents that may be present, load conditions, transformer rating as well as other design specs, ambient temps, possibly many other questions.

 

[big john]

Now it's running on a temporary 75kVA transformer that looks like it was salvaged from a junkyard. If this were load related, I think the overcurrents would have continued.

This problem and the similar ones I mentioned are unit specific, because swapping trannys has always fixed it. I'm just at a loss to explain what is going on inside a relatively simple system of copper, paper, and iron that would cause this.

 

[Tony S]

Yeah, that's where we're leaning, but I can't prove it and don't know enough theory about magnetic circuits to be able to speculate.

I'm almost tempted to do something silly like just take a small control transformer and drill through the laminations to see if deliberately trying to short them out causes a measurable increase in excitation.

We were discussing it, and this is probably the 3rd little dry transformer in a couple years that we've seen where it passed insulation resistance, and it passed TTR, but went into overcurrent when energized. These guys are usually far too small to justify hauling excitation testers out to each one, but I'm interested if there's any other field test to prove this?

Anyone have any resources that might give a good case-study on this?

I’ll save you wrecking a transformer.

It does cause higher inrush and over heating if enough laminations are shorted. I sent an inductor core to have the mating faces machined, instead of using a slab mill along the grain they went across burring the laminations.

 

[mbrooke]

Yeah, that's where we're leaning, but I can't prove it and don't know enough theory about magnetic circuits to be able to speculate.

I'm almost tempted to do something silly like just take a small control transformer and drill through the laminations to see if deliberately trying to short them out causes a measurable increase in excitation.

We were discussing it, and this is probably the 3rd little dry transformer in a couple years that we've seen where it passed insulation resistance, and it passed TTR, but went into overcurrent when energized. These guys are usually far too small to justify hauling excitation testers out to each one, but I'm interested if there's any other field test to prove this?

Anyone have any resources that might give a good case-study on this?

This might be a bit over kill, but I'd think this is a tool worth its weight in gold:


https://www.youtube.com/watch?v=eel...ion_3763327879&feature=iv&src_vid=O8wYhLnhzO8


How it responds to small transformers I have no idea, but everything you describe is a legit concern for utilities and they often invest in tool to detect it.

 

[mbrooke]

How the logic is gathered and interpreted if curious:



https://www.youtube.com/watch?v=KS0t3-VmwpA

 

[gadfly56]

Now you possibly opened this conversation to a whole lot of questions about the installation. Types of loads, high harmonic loads, deteriorating agents that may be present, load conditions, transformer rating as well as other design specs, ambient temps, possibly many other questions.

Note that the inrush issue is present with no loads on the secondary.

 

[big john]

...It does cause higher inrush and over heating if enough laminations are shorted. I sent an inductor core to have the mating faces machined, instead of using a slab mill along the grain they went across burring the laminations.

Did you see the effects in service, or did y'all discover the mistake prior to being energized?

 

[ATSman]

Keep in mind that the primary windings are under tremendous stress and move when energized. An intermittent turn to turn or phase to ground short may only show up under these conditions; something that test equipment cannot duplicate.
I have never done it but while under a megger test, hit the core laminations with a sledge hammer and see what happens

 

[kwired]

Note that the inrush issue is present with no loads on the secondary.

yes, but loading conditions possibly could lead to insulation deterioration and possibly turn to turn shorts is why I mentioned those things.

 

[Tony S]

Did you see the effects in service, or did y'all discover the mistake prior to being energized?

It was put in to service after I’d left the company. The shift engineer phoned me about it, Oh how I laughed. I’d made the right move, the coil and yoke weigh 2½ tons.

I returned to the company two years later, it was still sat languishing in a corner. A new yoke was shipped by the manufactures from the US to the UK, it didn’t fit the coil.

It became a catalogue of disasters