Delta-delta versus Delta-Wye

[Dumbo2929]

I apologize in advance if this has been answered before. I searched but did not find anything similar. I have a customer with a very old 1500kVA transformer (13.2kV-480V) delta-delta, with a true 3-wire delta service to their manufacturing facility. The transformer is being changed, but the main distribution panel is fine and remaining.

I recommended a delta-delta, but of course the contractor has a deal on a delta-wye; which he wants to hook up without the neutral connection. I know this will work and have a 30 degree phase shift.

Are there any negative issues anyone can suggest for applying a delta-wye in this application???

 

[waross]

Grounding the neutral is strongly recommended. You have a choice of grounding methods depending on your grounding philosophy. Solid ground, low resistance ground, high resistance ground.
respectfully

 

[davidbeach]

I'll second waross's recommendation to ground the neutral. High impedance grounding will provide all the advantages of ungrounded without the most serious disadvantages. New ungrounded installations should be completely avoided.

 

[Dumbo2929]

The customer using ground fault detection lamps on their delta service. Grounding the neutral of the transformer for a wye secondary will make that system ineffective.
The real question is what are the negative impacts of using a delta-(ungrounded) wye transformer to supply a 480V three-wire service (2000A)?

 

[davidbeach]

High resistance grounding of the neutral would not make ground fault detection ineffective. You would gain ground fault detection my measuring the current through the grounding resistor or the voltage across it, but would have conditions during a ground fault similar to those present with the delta secondary. What you wouldn't have with the high resistance grounded neutral is the severe transient overvoltages during arcing faults.

 

[Zogzog]

"The customer using ground fault detection lamps on their delta service. Grounding the neutral of the transformer for a wye secondary will make that system ineffective. "

It would also make that system unnecessary.

 

[tommom]

Grounding the system will reduce the possibility of a line-line fault going through their ground system. While a fault will trip a breaker off line, and thus shut down a piece of equipment or the System! Maintenance is much easier with a grounded system. Trying to find the fault when the lights come on before you get another one can be a real nightmare and very dangerous. Almost no one promotes ungrounded systems now.

Bear in mind - in that size, with grounded, ground fault should be added to the main and possibly to downstream breakers - 1000 amps and over this is a code issue. The equipment could be prone to arcing burndown without it.

 

[JensenDrive]

Sounds like some concerns about overvoltages on ungrounded systems are the biggest concers mentioned. If you do a web search on "ungrounded vs grounded power systems overvoltage" and you will find some stuff to read. Try

http://www.dymaxservice.com/technews/archive/fyi11.htm

for starts.

Assuming past success (which as a good stock broker would say, does not foretell the future) tells you to not worry about the overvoltage issues, some thoughts that come to mind:
I would think your old delta ground fault detection system will work OK; it will not know the neutral is floating, or that it even exists. Maybe you can connect your ground fault sensor to the neutral. High voltage at the neutral = ground fault. However, you will see third harmonic voltages in the neutral that you will need to ignore. In a delta-delta the third harmonic current required to excite the transformer circultates entirely in the two delta windings. In a delta-wye, the third harmonic now circulates in one winding, and creates a third harmonic voltage in the neutral.
Of course, the phase shift means there could be an issue of whether you will do a closed/open transition of the system voltages to a secondary source with another xfmr that has a different phase shift.
In a delta-delta xfmr, it is (remotely) feasible that the failure of any one leg will allow limping along with a two phase open delta. In a delta-wye ungrounded xfmr, the failure of any one leg means you loose too much to limp along.
The neutral in a wye winding may have reduced insulation levels near the neutral. You need to find out of the neutral insulation is rated for ungrounded operation.
In ungrounded systems, you are not required to have the ground fault overcurrent described in NEC 215.10. See 230.95 and 250.36. The 2005 NEC Handbook has some discussions in these two sections on the matter.

 

[peebee]

"High resistance grounding of the neutral would not make ground fault detection ineffective."

I agree. The existing lights should work just fine if you design the new system grounding correctly.

I generally agree with the other above posts that you should give system grounding another look.

 

[tinfoil]

In case it wasn't obvious from context above:

1) if you DO NOT ground the neutral point of the xmfr, then the secondary components must be rated for the FULL LINE-TO-LINE Voltage of the system.

2) No matter what, ensure that the xmfr is case grounded!

 

[ukzn]

ok before you all go on. what does this transformer feed. if motors or even other transformers are feeding of this board then a Star secondary will not suit your situation. also consider the that by incorprating a Star secondary that you will have to re-config the Restricted Earth Fault protection, by doing away with the NEC. and people dont take it for granted that a 480V secondary supply means it is a feed to a local/domestic usage.

" THINK BIG "

 

[davidbeach]

ukzn, not quite so fast. There would no problems with a wye secondary, it's done all the time.