Ungrounded Delta 480V/3-Phase/3-Wire System

[Shujinko]

I have a project where the existing pad mounted utility transformer is configured as delta-delta. The utility company says it is an un-grounded delta secondary at 480V/3-phase/3-wire system. I asked them to take voltage readings between phase-phase and phase-ground just to make sure it is an un-grounded delta secondary. They sent me back readings for phase-phase at 480V, but phase-ground measured approximately 277V for each leg. I was expecting the phase-ground measurement to be 0V on an un-grounded delta system. The utility company says it is definitely not a 480Y/277V system and the interior of the transformer has no neutral connection.

I have heard of lightly loaded utility transformers having other than 0V on phase-ground for un-grounded delta systems as the there is a capacitive effect between phase-ground.

Anyone have some insight on this issue as to why there would be a 277V measurement on phase-ground in-lieu of 0V? The utility company is pretty unresponsive.

 

[peter d]

It's completely impossible to read 277 volts on an ungrounded delta secondary. Something is lost in the translation here. I would ask to see a picture of the transformer data plate.

It could be that the reading was simply a capacitive measurement and they rounded it to 277 volts because that is a nominal voltage.

 

[Jraef]

Or, someone has connected VFD(s) to the secondary side and they have surge protection devices (MOVs) that are in a Wye configuration referenced to ground. You are supposed to remove the ground reference on those, but most people don't read the installation instructions, or if they do, don't know that they have an ungrounded delta system. So when you measure from phase to ground anywhere on the system, you are measuring through those MOVs. The first time there is a grounded fault anywhere on that system, those MOVs will attempt to (very briefly) become the Wye point for the entire system, then vaporize and likely damage the rest of the VFD. That's why they tell you to remove the ground reference connection.



Because many drives are made for the world-wide market where they don't use delta power systems, they may not even have a diagram like this, they simply instruct you to connect your drive to a solidly grounded Wye system, not mentioning the consequences of not doing so or not knowing what you have.

 

[gar]

170316-1632 EDT

Why would you not expect 3 approximately equal voltages when reading from any line to ground of an unloaded and floating (ungrounded secondary) 3 phase delta transformer? I am surprised they are all very close to each other, but I would expect some reasonable balance.

Assuming this is an unloaded transformer, no wires connected to the secondary, then put some reasonable resistive load from one transformer secondary terminal to earth (ground) then measure the three voltages to earth. A convenient resistive load might be 3 100 W incandescent bulbs in series. Now I would expect a considerable change in the three readings. The resistive loaded voltage going down and the other two up. Relative impedances will determine the balance.

If there is a long cable or just wires connected to the transformer, but no other load, then it is quite likely the three voltages to ground would be close.

If you did not read approximately equal voltages to ground what would you expect to read? Excluding outside factors you could read anywhere from 0 to 480 V depending upon what residual loads were on the secondary.

.

 

[Shujinko]

It's completely impossible to read 277 volts on an ungrounded delta secondary. Something is lost in the translation here. I would ask to see a picture of the transformer data plate.

It could be that the reading was simply a capacitive measurement and they rounded it to 277 volts because that is a nominal voltage.

The transformer nameplate says it's a delta-delta transformer I have a picture of it.

The readings the power company sent me were as follow:

225KVA Transformer (4160-480V)
A-B = 476 V
B-C = 476 V
C-A = 479 V
A-G = 277 V
B-G = 271 V
C-G = 278 V

 

[__dan]

I have a project where the existing pad mounted utility transformer is configured as delta-delta. The utility company says it is an un-grounded delta secondary at 480V/3-phase/3-wire system. I asked them to take voltage readings between phase-phase and phase-ground just to make sure it is an un-grounded delta secondary. They sent me back readings for phase-phase at 480V, but phase-ground measured approximately 277V for each leg. I was expecting the phase-ground measurement to be 0V on an un-grounded delta system. The utility company says it is definitely not a 480Y/277V system and the interior of the transformer has no neutral connection.

I have heard of lightly loaded utility transformers having other than 0V on phase-ground for un-grounded delta systems as the there is a capacitive effect between phase-ground.

Anyone have some insight on this issue as to why there would be a 277V measurement on phase-ground in-lieu of 0V? The utility company is pretty unresponsive.

Before doing anything, look at the transformer name plate and I would put money the secondary is Y on the nameplate, floating ungrounded Y. Y secondary is a four terminal output, only three terminals have a connection on them.

If it's padmount, you may have to ask the utility to open the cover to see if it has a fourth unused secondary bushing. I would verify this before taking any additional step. Delta secondary would probably show you red leg voltages to ground, even floating. Not 277 unless it's Y.

 

[electrofelon]

They were just using a high impedence meter. This is expected.

 

[Shujinko]

Before doing anything, look at the transformer name plate and I would put money the secondary is Y on the nameplate, floating ungrounded Y. Y secondary is a four terminal output, only three terminals have a connection on them.

If it's padmount, you may have to ask the utility to open the cover to see if it has a fourth unused secondary bushing. I would verify this before taking any additional step. Delta secondary would probably show you red leg voltages to ground, even floating. Not 277 unless it's Y.

I have a pic that says it's a delta-delta transformer and another pic that shows three connection points on the secondary side X1, X2, X3.

 

[Shujinko]

Or, someone has connected VFD(s) to the secondary side and they have surge protection devices (MOVs) that are in a Wye configuration referenced to ground. You are supposed to remove the ground reference on those, but most people don't read the installation instructions, or if they do, don't know that they have an ungrounded delta system. So when you measure from phase to ground anywhere on the system, you are measuring through those MOVs. The first time there is a grounded fault anywhere on that system, those MOVs will attempt to (very briefly) become the Wye point for the entire system, then vaporize and likely damage the rest of the VFD. That's why they tell you to remove the ground reference connection.

View attachment 17072

Because many drives are made for the world-wide market where they don't use delta power systems, they may not even have a diagram like this, they simply instruct you to connect your drive to a solidly grounded Wye system, not mentioning the consequences of not doing so or not knowing what you have.

No VFD's on this system. The utility transformer feeds a motor control center that has some motor starters on the branch circuits.

 

[__dan]

No VFD's on this system. The utility transformer feeds a motor control center that has some motor starters on the branch circuits.

Check the MCC for switching transient suppressor capacitors, TVSS, or some PFC caps, sometimes as an accessory box on the top or side. The MCC may have a small cap box attached to it, that may be Y connected internally.

 

[winnie]

I volt that electrofelon and gar have it.

There is no such thing as an ungrounded system. What we call ungrounded, if there are no faults, is really _capacitively_ grounded through all of the unintentional capacitors formed between the circuit conductors and ground.

A wire in a conduit is a capacitor. The magnet wires in a motor and the steel of the motor? Another capacitor.

The (very, very small) leakage through the insulation and the air around the conductors? Resistors.

An 'ungrounded' system in perfect condition will have very small _wye_ currents to ground, through the above 'parasitic' components. If the system is in perfect condition, and you measure with a high impedance meter, then I'd expect voltages quite similar to those from a wye system.

If you measure with a low impedance meter, or connect a load from phase to ground and measure voltage across that load, then the voltage to ground will be much lower than 277V.

If there are faults or intentional wye resistances (for example old school ground fault detection would be with a set of lamps connected wye to ground) then you might need a very low impedance meter to get a low voltage measurement.

-Jon

 

[Carultch]

There is no such thing as an ungrounded system. What we call ungrounded, if there are no faults, is really _capacitively_ grounded through all of the unintentional capacitors formed between the circuit conductors and ground.

Since an ideal capacitor acts as an open circuit when subject to DC, does what you say still apply to what we call "ungrounded DC systems"?

 

[peter d]

I volt that electrofelon and gar have it.

-Jon

Didn't I say that in the first reply? :huh:

It could be that the reading was simply a capacitive measurement and they rounded it to 277 volts because that is a nominal voltage.

 

[winnie]

Didn't I say that in the first reply? :huh:

That too

 

[peter d]

That too

Not nearly as detailed and in depth as yours, of course. :lol:

 

[winnie]

Since an ideal capacitor acts as an open circuit when subject to DC, does what you say still apply to what we call "ungrounded DC systems"?

Not nearly as well.

You still have the resistive leakage through the insulation, but this is probably small with modern insulation materials.

-Jon

 

[ptonsparky]

Didn't older ungrounded systems use lamps for ground fault detection? Wouldn't these be the resistive element to ground?

 

[iwire]

Didn't older ungrounded systems use lamps for ground fault detection?

Yes.

Wouldn't these be the resistive element to ground?

Yes.

 

[ptonsparky]

I would say that his system is fault free and working as it should then.

 

[iwire]

I would say that his system is fault free and working as it should then.

I do not think every ungrounded system had these, I think in the past they were optional.