YY Transformer with lost phase on the primary side

Status
Not open for further replies.
C

coop3339

Senior Member
Location
NJ
Hi

I have a situation where a grounded YY transformer is still producing 3 phase at the secondary side when a phase is disconnected on the primary side. The transformer is a 100kW 120/208Y to 277/480 delta, with a neutral connection at both sides. The utility is 120/208 3 Phase. The voltages on the 480V side with one leg disconnected on the 208 side are 277, 276, 268 and close to 480 between all. They also have the correct phase angle and frequency. There were no loads connected on the 480 side while testing. My question is where is the phase coming from. I think I know how its happening but I'm hoping someone who has more experience with transformer engineering can confirm it. I have attempted to explain it below. Please let me know if I am on the right track.

1. When the phase is disconnected on the 208v side the remaining 2 phases and the neutral make an open wye connection.

2. The open wye itself can't explain this because it is connected regular wye on the 480V side. This would not be a viable 3 phase configuration. If it were delta or open delta on the 480v side this would explain the voltage, in that case it would be a open wye/delta connection.

3. I think the common core of the transformer is acting as a delta connected winding and inducing the 3 phase voltage onto the 480V Y side, making up the 3rd phase.

4. I suspect if it were tested with a low impedance meter the voltage would drag down further on that phase.

The power is good enough for an inverter to qualify and connect to it even with the phase completely disconnected on the utility side. This prevents the inverter from meeting UL1741. I suspect a Delta/Y transformer would eliminate the problem but the utility will only approve a YY. I guess we could use a YY made from single phase transformers, but I have read that they have a lot of problems with harmonic currents and with the inverter this may be an issue. Any recommendation or further explanation would be greatly appreciated.

Thank you
 
correction

correction

sorry the first sentence should have been:

I have a situation where a grounded YY transformer is still producing 3 phase at the secondary side when a phase is disconnected on the primary side. The transformer is a 100kW 120/208Y to 277/480 wye, with a neutral connection at both sides.
 
Hi Phil,

Picture a 3 pole fused disconnect on the utility/208 side, with one fuse removed. All windings still intact.
 
yes primary neutral still connected at source and source is grounded Y120/208. This is a step up transformer, 120/208 to 277/480
 
I think you are right with the common core being the link here - if it were separate core coils and you lose an incoming phase you have no secondary output - see that all the time when POCO has pole mounted "banks", but lose an input phase on a single core three phase and you never know what kind of output you will get - but if delta primary it will run in "open delta" for a long time with no issues if load is not too high.

Did you try to load this in any way when missing one input line? I bet the more load you apply the more the voltages get out of normal ranges, but possibly will run some light load with no immediately noticeable problems.
 
Hi Kwired,

Thanks for taking a look at this. The strange part is that the mystery phase is at the correct phase angle. I think if it had a delta connection on the 208 side, it would not have the correct phase angle. It would be open delta if it had a burned open coil, but this has one feeder disconnected. I think with a delta connection on the 208 side, there would be something like 50%, 100%, 50% voltage levels on the 480v side, assuming no contribution from the 3 leg common core. The only thing connected on the 480v side is solar inverters and if they are turned on during testing they will qualify the power and connect.

Thanks,
 
I have attached a pic of a open Y/open delta connection. I think this is happening between the 208 windings and the 3 leg common core. The other phase is being created by the 3 leg common core inducing into the 480V Y secondary. At first I thought this would not work because of the 30% phase shift from Y to delta, but when it goes back from delta to Y, I think it goes back in phase. Notice the primary side connection in the pic, with just 2 phases and the neutral. This is exactly what is there when one phase is disconnected on the grounded Y primary.
Any thoughts on this?
 

Attachments

  • open wye.png
    open wye.png
    48.5 KB · Views: 7
coop3339 said:
I have attached a pic of a open Y/open delta connection. I think this is happening between the 208 windings and the 3 leg common core. The other phase is being created by the 3 leg common core inducing into the 480V Y secondary. At first I thought this would not work because of the 30% phase shift from Y to delta, but when it goes back from delta to Y, I think it goes back in phase. Notice the primary side connection in the pic, with just 2 phases and the neutral. This is exactly what is there when one phase is disconnected on the grounded Y primary.
Any thoughts on this?
Click to expand...

Did you say the voltage angle on the missing phase was correct when no load was connected?
 
There isn't realy any load on this just solar inverters feeding in at 480V. With the inverters disconnected and one phase disconnected on the 208 side, all the phases are still good on the 480V side. If the disconnect is closed to the inverters they see the all the phases as good and will connect. They actually exported power back through the transformers to the 2 phases and the neutral on the 208 side.
 
My calcs show that the voltage induced in the third leg of a common core would be a vector sum of the other two phases since there would be no flux produced that has the phase angle of the third leg. Typically, if all three phases are connected on the source side then the flux would sum to zero. Since you have a source missing then there should be a resultant flux on the third leg of the core. So, 277 @ 0deg plus 277 @ -120deg would equal 277 @ -60deg. This appears to be the same magnitude but opposite than expected phase angle. I would not expect to get the proper phase to phase voltages.

I would really like to see measurements (L-N and L-L) of the secondary terminals of the xfmr with no load connected.
 
That is what I had originally thought. The phase would be centered between the other 2 at about 60deg and something less in magnitude. My colleague tested with a power analyzer and says that all the phases are 120 apart and the derived leg is about 268v to ground/neutral and 480 or so phase to phase. I argued that you cant get a 3rd phase because it is not connected, but I later looked at the open wye/open delta connection and now I'm wondering if the core is acting like a delta winding.
 

Attachments

  • open wye.png
    open wye.png
    48.5 KB · Views: 1
coop3339 said:
That is what I had originally thought. The phase would be centered between the other 2 at about 60deg and something less in magnitude. My colleague tested with a power analyzer and says that all the phases are 120 apart and the derived leg is about 268v to ground/neutral and 480 or so phase to phase. I argued that you cant get a 3rd phase because it is not connected, but I later looked at the open wye/open delta connection and now I'm wondering if the core is acting like a delta winding.
Click to expand...

I just don't see how that is possible unless there is some type of buried winding like a delta tertiary or a zigzag secondary.
 
I don't know either. I am heading to the site tomorrow for testing, I will post what I find. I am very interested in seeing it for myself.
 
Status
Not open for further replies.
Top