Determing voltage on secondary of Delta-Wye transformer with primary phase missing

[mull982]

I am trying to confirm what the secondary readings would be on a 480V Delta Primary to 480V Wye connected isolation transformer when one of the primary phases is lost. Here is what I think:

Because this is an isolation transformer and the secondary wye connected L-L voltage is 480V then the L-N voltage across each winding must be 277V. So the 480V across each primary delta connected winding divided by the 277V across each secondary winding gives a winding ratio of 1.73 between the primary and secondary.

Now with a phase disconnected on the primary delta winding you will have one primary winding still reading 480V between the two remaining phases, and the other two windings would only have 240V across each of them since they are now in series between the remaining two phases. So if you take this 240V and divide it by the 1.73 winding ratio you will have aprox 138V L-N on two of the secondary phases and will have 277V L-N on the third phase on the secondary. You will also have the same L-G readings for a solidly grounded secondary. Are these expected readings correct.

Now the part that I cannot seem to see is what you would have for L-L readings on the secondary. I know you would have to subtract the vectors as you typically would but dont know what the phase angle of the 3 phases would be. Would the one full voltage 277V secondary phase be 180deg apart from the other two phases that are at 138V instead of 120deg since this is now a single phase condition?

Even if the secondary is feeding a load is it possible to read 277V L-G on the secondary through backfeeding or is this not possible since each phase still has a reduced voltage supply and in not really being backfed?

 

[LarryFine]

If I'm not mistaken, you'd have full voltages, but at 57% capacity.

Added: Maybe that's open-Y/open-Delta. Now I'm not sure. Anyone?

 

[rattus]

If I'm not mistaken, you'd have full voltages, but at 57% capacity.

Added: Maybe that's open-Y/open-Delta. Now I'm not sure. Anyone?

Larry, you got it. Furthermore, if the missing phase is due to an open line, the open delta would backfeed to the primary.

 

[jim dungar]

Added: Maybe that's open-Y/open-Delta. Now I'm not sure. Anyone?

It all depends on if you are describing the effects of loosing a single phase conductor, or a complete winding.

If you loose a winding the transformer configuration becomes an open-delta to an open-wye. However if you only loose a phase conductor, the transformer configuration remains a delta-wye.

With the loss of of a single primary phase conductor, the transformer will now be fed with only a single L-L voltage, and you will have the condition usually refered to as 'single-phasing'.

 

[mull982]

I apoligize I should have been more specific. I am referring to a case where one of the phase conductors is lost. With this being the case I dont believe the primary can be considered an open delta but rather this would be a single phase condition as jim mentioned and act as a typical delta-wye transformer. Wouldn't this single phase delta-wye transformer have to have reduced voltages similar to as I described since the primary would no longer have 480V across each of its windings?

 

[gar]

100712-2238 EST

Mull982:

Draw your delta and Y windings and place dots to indicate primary to secondary phasing.

A transformer reflects a load to the primary as modified by the turns ratio. If you short the secondary of a transformer, then reflected to the primary is a moderately low impedance determined by the short and the series leakage inductance and winding resistances. Ideally a shorted secondary could be considered a short on the primary.

The single primary with 480 will produce 277 on its secondary for any reasonable load.

Note how the phasings of the primary are related. A dot is connected to a non-dot all the way around the primaries.

When the two primaries in series fall in parallel with the single primary, then their phasing is opposite that of the single primary. So the two secondaries from the two series-ed primaries will have a 180 deg phase shift from the secondary of the single primary relative to the neutral of the secondaries.

The voltage on the two secondaries will depend upon the reflected load. So short one secondary, then the other secondary will have 277 V relative to neutral but opposite in phase from the single primary's secondary.

Equal loads and you will get your equal secondary voltages of about 277/2.

.

 

[mull982]

100712-2238 EST

Mull982:

Draw your delta and Y windings and place dots to indicate primary to secondary phasing.

A transformer reflects a load to the primary as modified by the turns ratio. If you short the secondary of a transformer, then reflected to the primary is a moderately low impedance determined by the short and the series leakage inductance and winding resistances. Ideally a shorted secondary could be considered a short on the primary.

The single primary with 480 will produce 277 on its secondary for any reasonable load.

Note how the phasings of the primary are related. A dot is connected to a non-dot all the way around the primaries.

When the two primaries in series fall in parallel with the single primary, then their phasing is opposite that of the single primary. So the two secondaries from the two series-ed primaries will have a 180 deg phase shift from the secondary of the single primary relative to the neutral of the secondaries.

The voltage on the two secondaries will depend upon the reflected load. So short one secondary, then the other secondary will have 277 V relative to neutral but opposite in phase from the single primary's secondary.

Equal loads and you will get your equal secondary voltages of about 277/2.

.

So then assuming balanced load on secondary we should see readings of 138.5V, 138.5V, and 277V L-N and L-G on all secondary phases?

 

[gar]

100713-1437 EST

mull982:

Yes, approximately in the real world, and exactly in theory with ideal components. Try three comparable control transformers, like 250 VA, and run the experiment. Also create an unbalanced load and observe the effect.

The two 138 outputs will be 180 deg out of phase with the 277 relative to the neutral point.

.

 

[mull982]

100713-1437 EST
Yes, approximately in the real world, and exactly in theory with ideal components. Try three comparable control transformers, like 250 VA, and run the experiment. Also create an unbalanced load and observe the effect.

If this was connected to a vfd would measuring at the input of the VFD also give these expected results?

The two 138 outputs will be 180 deg out of phase with the 277 relative to the neutral point.

What you would you expect if you took L-L readings between these three? Would you have to subtract using an angle of 0deg for the 277 leg and 180deg for the other two 138V legs?

 

[gar]

100713-2002 EST

mull982:

Where is the VFD connected? As viewed from the input terminals to the VFD it probably looks like a rectifier followed by a capacitor input filter. If this is a three phase input and is fed from the Y secondary that is now very unbalanced in output voltage and phase angles, then what happens will be dependent on the rectifier and filter configuration.

If I have my coils phased correctly, then when the hot end of the 277 secondary is at a peak + value the hot end of either other secondary will be maximum -. So the voltages add.

.