NOT bonding a delta/wye secondary XO

[Hltracey]

I am relocating an existing 13.2kv to 480 volt delta/wye transformer. It is a 225 kva and feeds a 480 volt three phase 100 HP fire pump.

Currently the XO is NOT bonded.

Is this a case where you DON'T want to bond the neutral as there is no neutral load?

Would bonding the neutral possibly adversely affect the fire pump?

Because I am relocating it, it will now be my baby. Please offer some discussion and possible solutions.

Thanks
Howard

 

[GoldDigger]

I am relocating an existing 13.2kv to 480 volt delta/wye transformer. It is a 225 kva and feeds a 480 volt three phase 100 HP fire pump.

Currently the XO is NOT bonded.

Is this a case where you DON'T want to bond the neutral as there is no neutral load?

Would bonding the neutral possibly adversely affect the fire pump?

Because I am relocating it, it will now be my baby. Please offer some discussion and possible solutions.

Thanks
Howard

If you do not bond the neutral OR corner ground the "delta" by grounding one of the phase lines, you will have an ungrounded delta, even though the transformer secondary is actually wye wound.
Making the fire pump an ungrounded system might actually increase its reliability, but for a normal circuit this would require the presence of a ground detector. Also for a normal circuit the center point of a wye would be grounded, but since the phase to ground voltages would be above 150V and the neutral connection of the wye is not used as a circuit conductor (it is not, right?) then you are allowed to run ungrounded.

If the system has a ground detector and you ground the X0, you will generate an alarm condition.
If the fire pump was designed to be supplied by an ungrounded delta, then you will not necessarily harm it, but you will make the fire pump system more vulnerable to being disrupted by a single fault.
If there was no ground detector and there is an existing fault in the fire pump, then grounding the X0 will cause sparks and smoke.

 

[augie47]

I agree with Golddigger other than I find no exception to the 250.21 (B) requirement for a ground detector.

 

[texie]

I am relocating an existing 13.2kv to 480 volt delta/wye transformer. It is a 225 kva and feeds a 480 volt three phase 100 HP fire pump.

Currently the XO is NOT bonded.

Is this a case where you DON'T want to bond the neutral as there is no neutral load?

Would bonding the neutral possibly adversely affect the fire pump?

Because I am relocating it, it will now be my baby. Please offer some discussion and possible solutions.

Thanks
Howard

IMO it would not be the norm to run a system like this ungrounded. As mentioned you would need a ground detector. The norm would be to run this as a grounded system with the fire pump controller being the service disconnect. All full service fire pump controllers are SUSE rated. You then must run the grounded neutral to the fire pump controller and bond it there even though it is not used as this is the fault return path.
Is this a POCO owned service transformer?

 

[Hltracey]

Their is no neutral load being fed by this transformer, just the 480v 3PH 3W pump.

This is a customer owned transformer

 

[jim dungar]

I would not purposely install an ungrounded 480V system. There are many more negatives to it than there are positives.
As evidenced by questions on this forum, many electricians seem to struggle with troubleshooting when the system is not a solidly grounded wye.

 

[tom baker]

250.20 requires a wye 277/480 transfromer to be grounded, in the event of a line to case fault the system bonding jumpers provides the path back to the transformer windings.

 

[jim dungar]

250.20 requires a wye 277/480 transfromer to be grounded....

Only if the neutral is used as a circuit conductor, per 250.20(B)(2).
In this case, the OP is feeding a 3W load and does not need the neutral, so grounding would not be required if there is no conductor or jumper on the X0 terminal.
Evidently the existing installation has been ungrounded.

 

[GoldDigger]

Evidently the existing installation has been ungrounded.

Whether deliberately or not.

Can you think of any reason that a ground detector would not be required, or will Howard have to choose grounding the system or installing a ground detector (if there is not one currently)?
I would not expect lack of a ground detector to be grandfathered in.

 

[Hltracey]

Not sure I have ever come across a ground detector, can you explain?

A couple of engineers that looked at this in the past felt that bonding could affect the arc fault rating of existing OCP. They also suggested that if it were a phase to ground fault the pump would still be able to operate until in eventually became a phase to phase fault and blew fuses. After looking at this they still recommended leaving the bond off.

The he windings still have a reference to ground as I can read voltage with a tester.

I was in agreement that with no neutral load the bond wasn't necessary. Not so sure now?!

 

[GoldDigger]

Not sure I have ever come across a ground detector, can you explain?

A couple of engineers that looked at this in the past felt that bonding could affect the arc fault rating of existing OCP. They also suggested that if it were a phase to ground fault the pump would still be able to operate until in eventually became a phase to phase fault and blew fuses. After looking at this they still recommended leaving the bond off.

The he windings still have a reference to ground as I can read voltage with a tester.

I was in agreement that with no neutral load the bond wasn't necessary. Not so sure now?!

The basic principle is that having the equipment continuing to work through an initial ground fault is not much help if you do not know about it so that you can fix that fault. If you do not know about the first fault the second fault will come along and stop your equipment with no warning.

But that is primarily oriented to process equipment that makes a big mess if it shuts down in an unexpected and uncontrolled way.
Since the fire pump will not be operated routinely, there will not that opportunity to repair the first fault if the fault is in the pump motor rather than the wiring system. That makes the need for knowing about the first fault somewhat moot.

A ground detector can be something as simple as a light bulb connected from a phase to ground. If it lights up, you have a fault.
For three phase, the equivalent is three light bulbs to ground. Initially they will all light dimly. If one goes out and the other two get brighter, you know that you have a fault to ground, and you know which phase it is on.

 

[jim dungar]

Can you think of any reason that a ground detector would not be required...

I believe one would be required.

 

[jim dungar]

Not sure I have ever come across a ground detector, can you explain?

Ungrounded systems and ground detectors should only be employed where they are maintained by qualified persons.

 

[jim dungar]

A couple of engineers that looked at this in the past felt that bonding could affect the arc fault rating of existing OCP.

Do you mean Amps Interrupting Rating (AIR or maybe AIC)? In general, 3-phase faults are easier to manage with solidly grounded systems than with ungrounded ones (equipment availability and cost among other things). I am not aware of any equipment that is rated for ungrounded systems that cannot also be used on grounded ones.