Wye-Delta Transformer testing

Thread Starter

notoriusjt2

Joined Feb 4, 2010
209
say we have a 2400V to 480/277V dry type delta-wye transformer. The high side is protected by 80A cutout fuses, and the low side goes directly to a distribution breaker panel (600A main breaker) for a whole crap-ton of heaters.

One day a single high side fuse blows. Upon arriving at the scene you see that not one low side breaker has tripped and the transformer visually seems ok.

After removing all energy sources from the transformer, what types of testing should be done?
 

timrobbins

Joined Aug 29, 2009
318
Probably a full suite of insulation and winding resistance and impedance tests on all windings to check if the transformer itself is not to blame. That won't check thermal causes, or load causes (overload imbalance), or incoming over-voltage causes, or aged fuse.
 

Thread Starter

notoriusjt2

Joined Feb 4, 2010
209
we performed a transformer turns ratio test on all three phases and they came out OK. we also performed a megger test between the high and low side windings and got a very high resistance, between the high side terminals and ground and got a very high resistance, and between the low side terminals and ground and got next to no resistance.

I understand that a 4-wire wye system has a ground reference, but how is it that each low side terminal is basically connected to ground and nothing shorts out?
 

marshallf3

Joined Jul 26, 2010
2,358
There's very little way to test those transformers. Get a good clamp-on ammeter and see what's going in and out when it's working, then evaluate the loads you've got on the secondary side to make sure you're fairly balanced and that you don't have some huge motor that might have a winding starting to go out.

Ends up as being a long matter of elimination but the transformers themselves rarely fail. Luckily I've got the huge ones outside the building and are the property of the electric company so if one fails it's their problem, I get straight 480 in at about 6,000A per leg from each of them but I rarely have to get even close to tapping that.
 

marshallf3

Joined Jul 26, 2010
2,358
I understand that a 4-wire wye system has a ground reference, but how is it that each low side terminal is basically connected to ground and nothing shorts out?

They aren't connected to ground but the huge amount of copper in those may lead you to believe that they are. If you're getting somewhere between 440 - 500 out between phases then you may have to call your electric company who will usually gladly install some monitoring devices at absolutely no charge to you to help pinpoint the problem.

[EDIT:] Added the words "between phases"
 
Last edited:

timrobbins

Joined Aug 29, 2009
318
Wouldn't the star of the Wye go out as neutral and be directly bonded to PE - possibly surreptitiously - which is why you get a low ohm reading.

I guess your turns-ratio test would have identified a shorted turn type situation (?).

When you energise the primary again then check the priamry mag currents with no load - and compare with a known satisfactory level.
 

Kermit2

Joined Feb 5, 2010
4,162
.
When you energize the primary again then check the primary mag currents with no load - and compare with a known satisfactory level.
transformer losses with an open secondary are core losses, the transformer has a rating for core loss but I don't know if such will be present on the name plate.
 

marshallf3

Joined Jul 26, 2010
2,358
At the input and output voltages he's running any faults should present themselves in short order. My guess is there's something down the line that's hitting it with a far more than an unusual load and he'll have to trace that down.
 

Thread Starter

notoriusjt2

Joined Feb 4, 2010
209
At the input and output voltages he's running any faults should present themselves in short order. My guess is there's something down the line that's hitting it with a far more than an unusual load and he'll have to trace that down.
yea but wouldn't you think that the distribution panel individual branch circuit breaker or even the main breaker would have tripped first?
 

marshallf3

Joined Jul 26, 2010
2,358
Hard to predict. He indicates heaters which have a huge amount of draw until they come up to temp (think tungsten lamps since we don't know what they are) and he also mentions fuses instead of circuit breakers which have a far better ability to absorb peak currents until they trip.

Without actually being there with meters I'm just guessing in the dark so I can only offer a tiny bit of advice.

If there are a lot of heating circuits involved I'd turn them on a few at a time, for all we know a new shift comes in there at 8 AM and turns everything on all at once. Time delay fuses may help but that isn't really the answer - or it might be.
 

marshallf3

Joined Jul 26, 2010
2,358
Not necessarily. Even if he has a large circuit breaker on the main input it's going to be far more forgiving and is probably rated to handle more than just this one transformer.

I've got my two 480V mains with 4,000A and 2,500A breakers and only once has one tripped - and that was because an employee who should not have been in that room in the first place was apparently poking around and hit the wrong button on the distribution panel that tripped it off. Breakers this big take a manual rewind of 10 times on a lever then it's a huge bang when you can finally hit the reset button. I wish I had doors on that room but I don't and all I can do is put up signs for the time being to keep your damn hands off. I know that the 2,500A one wasn't exceeded, that leg of the building is lucky to use more than 500A max.
 
Last edited:

Thread Starter

notoriusjt2

Joined Feb 4, 2010
209
Hard to predict. He indicates heaters which have a huge amount of draw until they come up to temp (think tungsten lamps since we don't know what they are) and he also mentions fuses instead of circuit breakers which have a far better ability to absorb peak currents until they trip.

Without actually being there with meters I'm just guessing in the dark so I can only offer a tiny bit of advice.

If there are a lot of heating circuits involved I'd turn them on a few at a time, for all we know a new shift comes in there at 8 AM and turns everything on all at once. Time delay fuses may help but that isn't really the answer - or it might be.
well while this transformer was offline all the heaters were off. The building dropped to 48 degrees, and when the transformer was turned back on, all the breakers were closed within 1 minute of each other. No trippings, no problems
 

Kermit2

Joined Feb 5, 2010
4,162
Perhaps a bird or squirrel 'bought the farm' on the power connections just outside the building?

If the line had been momentarily shorted by a surge, there would be created large enough circulating currents in the line to blow any fuse that was loaded close to max at the moment of the surge.
 

marshallf3

Joined Jul 26, 2010
2,358
well while this transformer was offline all the heaters were off. The building dropped to 48 degrees, and when the transformer was turned back on, all the breakers were closed within 1 minute of each other. No trippings, no problems
That should tell you something, and even by code I believe you can use slow blow (time delay) fuses of the proper ratings. Most of my distribution panels have circuit breakers but some still have fuses, if I run into a problem I try my best to simulate conditions then if I need to I'll go with a time delay fuse. Of course I start with 480 coming in, I wouldn't feel safe working with any higher voltages. There's enough copper in this building that if you ever pulled it all and sold it you could probably finance a major league baseball team for a year.
 
Top