PHASE-TO-GROUND NO VOLTAGE (I'M NEW TO THE INDUSTRY)

[kwired]

Hi to All!

I apologize for taking to long to respond.

I've conducted test at loadside of the xfrmr, as you have suggested, and here are the results:

Vab = 240V
Vbc = 240V
Vca = 240V

Vag = 240V
Vbg = 0V
Vcg = 240V

Connection wiring of 500kVA, 480V/240V Dry Type Xfrmr: Delta (P) and Ungrounded Wye (S)

We then traced the connections of the secondary, we found that phase B was body grounded. Is this what may have caused grounding of Phase B? Or are there external factors that contributes to this grounded phase?

We tried disconnecting the panel where we found a grounded phase B, and readings were back to normal (expected readings). Then tried to connect it again, at first all is well but after a couple of minutes, phase B was again grounded.

Also, yes we patterned our code to the NEC with certain ammendments.

I take the portion I bolded to mean there was a factory link or jumper from B phase to the transformer case/housing. If so that is an intentional bond to create a corner grounded system and the voltage readings you posted are exactly what should be expected.

I am a little confused at what you intend to mean with this "Connection wiring of 500kVA, 480V/240V Dry Type Xfrmr: Delta (P) and Ungrounded Wye (S)" Though you apparently don't exactly follow NEC. In general there is no such thing as an ungrounded wye system if you are using NEC. I also not sure if 480 is primary or secondary voltage though it seems it likely is primary voltage, or what is wye and what is delta, or if there even is a wye side to the transformer. Sounds like whichever side you are measuring is a 240 volt corner grounded system though.

So what I am seeing so far is that you do have a corner grounded delta supply, you just apparently haven't seen one before and don't fully understand what you have or how to work with it. You can supply 240 volt loads (single or three phase) with it. One conductor happens to be grounded, but is still the same potential to the other conductors of the system and the load(s) only care about receiving proper voltage and could care less which conductor is grounded. (get into some electronics and digital signaling circuits and this gets more complicated but a simple three phase motor or resistive load won't care what is grounded or what isn't) There is no 120 volt taps to this system, if you need 120 volts for something you will need to separately derive it.

 

[Johnnybob]

Hi to All!

I apologize for taking to long to respond.

I've conducted test at loadside of the xfrmr, as you have suggested, and here are the results:

Vab = 240V
Vbc = 240V
Vca = 240V

Vag = 240V
Vbg = 0V
Vcg = 240V

Connection wiring of 500kVA, 480V/240V Dry Type Xfrmr: Delta (P) and Ungrounded Wye (S)

We then traced the connections of the secondary, we found that phase B was body grounded. Is this what may have caused grounding of Phase B? Or are there external factors that contributes to this grounded phase?

We tried disconnecting the panel where we found a grounded phase B, and readings were back to normal (expected readings). Then tried to connect it again, at first all is well but after a couple of minutes, phase B was again grounded.

Also, yes we patterned our code to the NEC with certain ammendments.

Umm, if you are pulling readings from the load (secondary) side, and if it was a Y wound secondary, you would not get the same readings phase to phase and phase to ground should differ by the sq rt of 3, or 1.732. Someone mis-wired the secondary in a delta configuration with a corner ground.

 

[kwired]

Umm, if you are pulling readings from the load (secondary) side, and if it was a Y wound secondary, you would not get the same readings phase to phase and phase to ground should differ by the sq rt of 3, or 1.732. Someone mis-wired the secondary in a delta configuration with a corner ground.

If you had 240 volt secondary coils and connected them in Y but made no additional connections to the neutral point, only brought the three phase conductors out for availability to use and grounded one of them it would be easy to assume it was a corner grounded delta. This is not too likely to happen on a single core factory wound unit though and would be more likely to happen with a three phase "bank" composed of what could otherwise be stand alone single phase units.

Information OP has given though still points to this being a corner grounded delta secondary, it maybe wasn't supposed to be that but sure looks like it is.

 

[ggunn]

We tried disconnecting the panel where we found a grounded phase B, and readings were back to normal (expected readings). Then tried to connect it again, at first all is well but after a couple of minutes, phase B was again grounded.

What seems strange to me is that B became grounded after a couple of minutes instead of instantaneously.

 

[Fulthrotl]

What seems strange to me is that B became grounded after a couple of minutes instead of instantaneously.

to the OP....

well, it's a corner grounded delta, just not intentionally.
something out in the field is going to ground, and grounding the
entire phase, intermittently.

as there were no grounded corners, any single phase going
to ground won't fault the OCPD.

shut off all the branch circuits, and you should have an ungrounded
system. close one at a time till you find out where the fault is.

if it's randomly going to ground with no loads closed on the panelboard,
then you have a feeder or the panelboard intermittently connecting to
ground.

i suppose it's possible to have the secondary of the transformer intermittently
going to ground, but i've never seen it happen.

it's quite possible not to have any arcing and minimal current flow when this
intermittent closes. you wouldn't have any issue until something on another
phase went to ground, and then there would be current flow.

what would be interesting is when you have zero volts on B phase, measure
the resistance of B to ground. if it's solidly grounded, you should have less
than 1 ohm.

 

[mivey]

What seems strange to me is that B became grounded after a couple of minutes instead of instantaneously.

Isn't it though? Some kind of high impedance fault that takes a moment to bleed down or to "weld" back?