Open Delta PT vs Wye PT 1

[TxPowerEE]

Hello all,
My first post here, so if I leave anything out let me know. I am dealing with a utility that is demanding we install wye-wye pts for protective relaying in our gear. We are connecting to the utlity with 5 2.2MW gens at 13.2kV. We have capacity for up to 12. The system will be paralleled for several minutes but will never export power.

The utility is demanding the y-y PT because of a requirement in IEEE 1547. Their argument is that if a phase upstream of my facility (on their grid) goes to ground and the gens are online the y-y pt is required to capture a voltage sag due to the ground fault. The protective relay is a GE750. I get the sense they are demanding this only because its in 1547, they can't explain to me why the y-y is superior in this instance to the open delta.

My questions
1) Is there any advantage of the y-y over the open delta?
2) Looking for a voltage drop to clear a ground fault seems like a rather round about way of clearing a ground fault. Wouldn't is just be better to use the 51g setting in the relay to clear the ground fault? Or possibly the residual ground fault settings?

 

[davidbeach]

Open delta is a relic of electromechanical relays. Never, never, use open delta with numeric relays for new applications and try to replace them with wye when you can for retrofit projects. Open delta can not measure V0 or any phase-ground voltage, you are just throwing information away. Looking at voltage on an ungrounded system is about the only way to detect a ground fault; you've said nothing about the system configuration, perhaps phase voltage is the only way to detect a ground fault when the utility breaker has opened - open delta can't do it.

 

[vima]

open delta generally used in high impedance grounding system or ungrounded system.
Is it just Y-Y PT or earthed through neutral?? Explain how your generator neutral earthing and the grid earthing was carried out?

 

[davidbeach]

No, open delta (two transformers) is worthless in high impedance grounded systems. What is usually used in high impedance grounded systems is broken delta (three transformers). Open delta provides phase-phase voltage only. Broken delta provides 3V0 only. Connected to a numeric relay wye-wye VTs provide phase-phase, phase-ground, (+) sequence, (-) sequence, and (0) sequence voltages.

 

[slavag]

Hi.
Very shortly.
100%+.. agree with David.
Best Regards.
Slava

 

[dpc]

There's very little chance of winning this argument with the utility, so you may as well figure out a way to give them what they want.

I'm not sure I fully agree that open-delta PTs are a relic of electro-mechanical relays. This arrangement is commonly used in metalclad switchgear mainly because two PTs are cheaper than three. The open delta works OK for metering which is what they were generally used for.

But I agree that for protection in distributed generation applications, three PTs are generally going to be necessary to get useful relaying information for the reasons David Beach gave.

 

[slavag]

Yes Dave. It's issue
"This arrangement is commonly used in metalclad switchgear mainly because two PTs are cheaper than three. The open delta works OK for metering which is what they were generally used for."
But we can agree, for the 3xPT per BB, price isn't important . And today many designers would like used voltage information not only for metering , also for the NPS and PPS voltage meas.
Regards.
Slava

 

[TxPowerEE]

Bear in mind that this system can accomodate up to 12 gens and with 4 utility feeds, plus gen tie breakers. This adds up to 26 sets of PT's, which are used for paralleling controls protective and metering information. The costs of PT's is substantial. However, the project was specified, bid, and accpetted by the utility with Open Delta. Now 2 years later they are demanding Y-Y.

DPC is 100% on the money the utility is going to shove this down my throat, but their protection engineers could not tell me why they wanted Y PT's. The just kept saying "because IEEE 1547 says its required." Incidentally, IEEE 1547 isn't required by the Utility commission, whose interconnect standard does not mention PT configuration.

DB - the system is grounded, but the utility configuration is unknown. I'll know more after a meeting today.

 

[dpc]

I don't see why the utility would require wye-wye PTs for each generator. As long as you provide them at the utility interconnections and the protection isolates your generators from their system when there is a fault on their system.

Maybe I'm not understanding the configuration.

 

[davidbeach]

The price adder for 3-VTs rather than 2-VTs is what, 0.0001 times the cost of the gen set? How can that be a substantial cost?

 

[TxPowerEE]

The system is designed for Utility parallel, so if you used opend delta on the Gens and Y-Y on the utility you would have to accomodate for the phase shift. What we are going to do is install Open delta PT's with a a 1:1 ratio on the secondary of the Y-Y. This is the most cost effective option and the least painless option.

DB substantial now if you had to replace 26 PT's after the fact, parts and materials not to mention lengthy outages while the PT's are replaced. If it was done up front the cost would be minimal.

 

[dpc]

The aux PTs on the Y-Y PTs should work fine for synchronizing function, although you really only need this on one phase.

I'm still not clear on why 26 PTs have to be replaced?

 

[TxPowerEE]

DPC
If we didn't install the aux pt's the utility signal would be shifted 30 degrees due to the Y-Y vs Open Delta and all of the generators, Utility board buses and Utility breaker PT's feed the synchronizing controls. The synch controls ,possibley, could be adjusted to account for this shift but it is not the optimal solution and could create maintenance/functionality issues down the road.

 

[dpc]

I understand the need for aux PTs to correct the phase shift for synchronizing. But you don't need aux PTs on every set of PTs to accomplish this, or at least I wouldn't think so. But I don't your one-line so I'm probably missing something.

But it sounds like you already have a solution.

 

[slavag]

Hi.
Intresting thread.
From technical point of view.
Y-Y connection is more better, BUT NOT MUST for the genrator protection, synchronazing and metering.
As I understand, you have open delta connection on the BB too and sinchronazed genset to BB.
I think you need aux, PT for the BB PT only ( I don't know how sections you have).
voltage connection before auv PT you will use for the genset sinchro, after aux PT for the utilities infeeds.
Regards.
Slava

 

[davidbeach]

Many synchronizers and sync check relays can be set to expect a 30[°] phase difference at synchronism (actually any multiple of 30[°]) - you shouldn't need aux VTs just for that.

 

[slavag]

Hi.
If TxPowerEE have " synchronizers and sync check relays can be set to expect a 30° phase difference at synchronism (actually any multiple of 30°) " it's solve all problems ( I'm hope).
Good Luck.
Slava.
P.S. In your new projects, please try use Y-Y.

 

[dpc]

Better do a slow and careful phasing check.

 

[alehman]

This is a good discussion, but I think to provide a more definitive response we need know more about how your system is configured and the utility requirements.

Open delta gives you accurate V1 and V2, but not V0 as stated above. If it is important to know V0, such as for ground detection, then YY is required. Do you have utility ground detection presently and how is it accomplished? Does the utility have a standard for this? It is unusual for a utility distribution circuit in the U.S. to be anything but solidly grounded (there are exceptions).

Since this is a retrofit, it would be preferable to minimize the cost and retrofit YY only where required. I think it is unreasonable for the utility to say you must upgrade every PT in this existing system unless there is demonstrated a need.

I wouldn't go so far as to say open delta is a relic, there are many still being installed and they work fine for many purposes. YY is a minimal cost add for a new systems and probably worthwhile for the added flexibility even if there is not an immediate need.

 

[REM76]

I don't know alot about your system, But I would think that you should be able to detect ground faults on the utility side using a 67N function (directional over current) at the point of common coupling (PCC). In this case, I believe you still need a Y-Y PT only at the PCC for accurate polarization using calculated 3Vo.