ECN Forum Forums Technical Discussion Electrical Theory and Applications Phase shift on transformers

As a lot of people on this forum are on-the-job trained electrians as opposed to engineers, I went ahead and drafted up a few phasor diagrams, hopefully they won't be too hard to understand!

These are Phasor Diagrams drawn in polar coordinates. (A phasor is basically a voltage represented as a vector.) Vectors represent voltage; angle represents the sine wave and phase shift.

1 full rotation represents 1 cycle, or about 1/60th of a second. If this were viewed in real time, you would see the vectors rotating counter-clockwise 60 times a second. The distance between the vectors is 1/3 of a cycle, or 120 degrees and represents the phase shift between each of the 3-phases.

This is what a normal Wye (star) looks like as a phasor diagram. Each leg is 120V long, and each is rotated 120 degrees apart.


Delta is derived by going line-to-line. The phasor diagram makes it easy to see the relationship between the wye and delta voltages. Mathematically, this is how the difference in voltages is derived.


In this diagram, the vectors we drew in the above diagram have been shifted on the graph so they're all measured relative to 0V. You can see that the delta voltages are rotated 30 degrees ahead of the wye voltages- this is what's referred to when people say delta is leading wye by 30 degrees or vice versa. The square root of 3 (1.73x) difference is easily visible here, too.


In this diagram, the blue represents the delta primary on a delta/wye transformer. The green phasors represent the wye created on the wye secondary. As the wye secondary is wrapped around the same coils as the delta primary, it essentially is taking the new wye and forcing it onto the delta. And since that delta was leading the old wye by 30 degrees...
Well, it's hard to explain this in a way that doesn't come off as incomprehensible gibberish, but hopefully staring at the pictures long enough will make sense


[This message has been edited by SteveFehr (edited 10-28-2006).]

Those sure are pretty, but all wrong down here in the great state of Alabama. We're ACB rotation.

Steve, thanks for the post.

It is memmbers like this, all of you, who make this forum great.

I am just "scrathing the surface" on these phasor diagrams, and the math involved, so if my questions seem stupid to some, I apologize.

Another way for this phase shift to make your life miserable is if some one rolls the primary rotation on one of two transformers that need to be paralleled. This will lead to a 60% phase shift on the secondary between the two transformers. Not a happy scenario when the high side is a welded 138 Kv bus.

Karma already caught up with me- I took the PE exam today, and there were 2 questions dealing specifically with this, which I was really not expecting! So taking the time to think it through again well enough to explain it has already paid off

I just wish there had been more NEC questions and less freaking high-voltage fault analysis... this fall's test did not play to my strengths, that's for sure.

WFO: we had that happen once a couple years ago on a 480V plant- breakers tripping left and right, popping like popcorn! Was bad news. Turns out, the contractor, despite reassuring us TWICE that he had metered and verified rotation, had indeed swapped phases at the UPS bypass switch. The only good news was it wasn't my fault, nor my problem, as I didn't work there yet. But oh, I hear about it every time anyone suggests an online test vice scheduling an outage...

[This message has been edited by SteveFehr (edited 10-27-2006).]

I remember my first OT callout to a substation. The utility claimed to be finished repairing our 12.6KV AC1 line. I called our Power Controller and stood clear as I asked him to close the AC breaker. KABOOM! Once my pulse settled, I asked him to open the AC Bus Tie and try closing AC1 again. There weren't any loud noises that time. Since each AC bus has its own 120/208Y aux transformer, I went to an ATS fed by both for answers. I measured A to A, B to B, and C to C, not expecting any value, just that they should all be about the same. They weren't. The next step was to jog a compressor fed by a panel off of that ATS, transfer, and jog again. Seeing the pulley rotating the wrong direction was the final proof I needed to call the LD (Load Dispatcher) and ask him to please have the lines switched back the way they were. As a new guy, I wanted to be really sure before I told the utility that they had things a tad backwards.
Joe

Steve,

A phase rotation check will never determine a "rolled" primary. All a rotation check does is look at combinations like X1->X2->X3 or H1->H3->H2 it never determines which conductor is actually A phase. The only way to determine a rolled primary is to physically trace it out or to connect it up and compare the ouput voltages to a known source.

I have a location with six transformer all verified (using a TTR) as Dyn01 conncted units. But after they were installed and phase rotation was corrected, voltage measurements discovered a rolled primary on two of the units.

What is a "rolled" primary?

It's when the 3 phases are still in order, but attached to the wrong terminals. EG, instead of ABC on consecutive terminals, they might have CAB. So, if you, say, had an undervoltage on phase A, that undervoltage would show up at the B terminal instead of the A terminal. Not an issue for most power distribution, but it can be an issue for parallel systems and networks.

And yeah, my example wasn't rolling the phases, it was transposed phases and reversed rotation.

[This message has been edited by SteveFehr (edited 10-28-2006).]

Is there such a thing as a "rolled" secondary?

Like most high leg deltas are?