GSU Delta/Wye Floating Neutral & Line Imbalance 2

[Ceast]

I need some recommendations regarding high side voltage imbalance. The system is a 2.5MW 2400 delta gen feeding a delta/wye 2400/12,470 GSU. The neutral is floating on the highside wye. I have 15% ph-N voltage imbalance when running the generator isolated. I was called to the customers site to troubleshoot a failed DECS 200. I found the A-ph PT failed as well as the DECS. This is the second PT failure in less the a year.

THe sensing PT's are 60:1 wye.

The generator is used mostly for interruptable duty and runs minimally.

Why is the neutral floating? Recommended solution?

Thanks

 

[waross]

Not good. If a branch, bird or small creature contacts a phase conductor, the neutral voltage to ground may rise from zero to 7200 Volts and the voltage on the other phases may rise from 7200 Volts to 12470 Volts. That may take out a PT and the death throws of the PT may take out the DECS. The unbalance may be an indication of a failing PT.
Consider grounding the neutral.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jghrist]

Is the 12.47 kV system normally fed by a utility source with a solidly grounded neutral? The neutral is probably floated to avoid having the transformer be a ground source. If the system is run connected to the 12.47 kV system without connection to a grounded utility source, then a ground source is needed or you will have the voltage unbalance problems noted. Either ground the transformer neutral or provide a separate grounding transformer if you run with the generator isolated and connected to the 12.47 kV system.

 

[Ceast]

jghrist, yes the 12.47 utility system is grounded. I'm having a real problem with gen voltage stability and the fact that the voltage matching potentials are Aph to Aph makes paralleling a bit of a challenge. All of the metering is Y connected and the DECS is ph-ph sensing.

I believe you are correct I need to run a ground strap from the xfmr Xo terminal. Funny thing is they have had this unit in service for 5 years. I don't know how they ever got it paralleled! They had to have had voltage offset issues at start-up.

 

[Ceast]

I'm sure the high voltage on Aph is killing the Y connected PT and causing grief for the regulator. Once the PT dies (loss of sensing)the DECS drives the gen into overvoltage and trips.

 

[alehman]

If the PT's are connected line-ground, they should be rated for at least line-line voltage.

Should definitely add grounding for island mode. Resistance grounding may be workable for both interconnected and island operation.

Alan
----
"It’s always fun to do the impossible." - Walt Disney

 

[waross]

The X0 should be connected to a ground grid, not the generator frame. The generator and the X0 may be independently connected to the same ground grid.
In the event of a ground fault on the line, ground currents will be much higher. Should ground fault current damage the connection between the generator and the grid, the touch voltage on the generator may rise to lethal levels.
Connect the X0 directly to to grid.It is still possible for fault current to damage a connection to ground, but it is better to kill a PT as is happening now than to kill an employee.
I was looking at a large diesel installation in the third world. There was a large gauge cable from the station ground grid to the generator. The connections and workmanship looked good. However the cable was connected to the radiator assembly which was completely rubber mounted.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jghrist]

Paralleling shouldn't be a problem on the low side. The problem is when the high side of the transformer is energized from the low side when the 12.47 kV system is not connected to a grounded utility source. Without grounding X0, the 12.47 kV system is ungrounded and Ø-n voltages are unstable.

If X0 is connected to the 12.47 kV neutral, then the transformer will be a ground source and current will flow in the transformer windings during a 12.47 kV system ground fault and if the 12.47 kV system has unbalanced loads. These currents will flow with the generator off if the 12.47 kV system is connected to the utility source. This may or may not be acceptable, depending on the source and transformer impedances, and on the transformer protection, but needs to be considered.

 

[Ceast]

We solidly grounded X0 to the substation grid. Problem solved.

 

[waross]

Sounds good. You will have a better feel for it being on site, but there is a possibility that one or more PTs has been damaged by overvoltage. I would keep this in mind. If you see unbalanced voltages in the future consider that a PT may be failing and verify the health of your PTs before worrying about possible system problems.
Thank you for sharing the final resolution with us.
Yours
Bill

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Ceast]

Wella, I need to back up a bit. I have to say I was miss informed. Much confusion here.

Problem not solved. I reviewed the system and this AM and discovered the config is not as discribed. X0 not grounded, and it shouldn't be!

I spent today trying to troubleshoot the voltage imbalance/instability.

To begin with the GSU is a delta/delta and the PT's are Y. Gen is 2400Vac delta. GSU is 2400/12470. Utility nominal voltage is 12,900.

While gen is running. Primary=12,800Vac Secondary PT voltages are A-N =142 B-N =112 C-N =118 60:1 PT's.

If I manually raise the gen voltage up to match the Utility voltage it becomes very unstable and I recorded secondary peaks of almost 300Vac. If I drop the voltage down to 12,700 everything looks normal other than the seconday voltage imbalance. The Regulator output goes nutty once the voltage hits 12,900!

Am I seeing ferroresonance? Delta/delta GSU and Y PT's.
Shouldn't the PT's be Open Delta?

I'm at my wits end on this one!?!?!?

East

 

[davidbeach]

Bad luck there. You will find it next to impossible have a grounded system and an ungrounded system line up sufficiently to allow synchronization. You'll need to provide a ground bank to provide a stable system until synchronized with the utility. You could then disconnect the ground bank if desired. Of course if it were my utility, you'd never have had a delta-delta GSU signed off.

 

[jghrist]

You can't have Y connected VTs on the ungrounded system, but if you used open-delta VTs, I don't know why it couldn't be synchronized to a grounded system. David has more generation experience than me, so I will defer to his expertise, however.

You will have trouble trying to serve Ø-grd loads on the 12.9 kV system if it is isolated from the grounded utility source. You shouldn't try to serve the system islanded without a grounding bank.

 

[Bronzeado]

Hi Ceast,

Could you send to us a single line diagram of your system, including the power transformer and VTs conections and, also, the voltages waveforms during the gen tests (voltage changes)?

Best regards,

H. Bronzeado

 

[davidbeach]

Y connected VTs on an ungrounded system is not a problem if you understand what you are getting. Connected to a numeric relay you'd only use the phase-phase voltages and the sequence voltages.

Trying to synch an ungrounded system to an other system, grounded or not, will be frought with difficulties as you won't be able to adjust out the standing voltages to ground of the two different systems. If you have a lowside breaker you might try synchronizing there, the voltage balance may be better.

For many reasons the highside should be a grounded wye, but it is too late for that now, so a grounding bank that let's you synch two grounded systems is your best bet. If the utility truly doesn't want you to be a ground source, then you can switch it out after closing the breaker.

 

[waross]

David, will you comment on using 60:1 (120;7200V) PTs wye connected on an ungrounded 12470 V system. As per Ceast's
21 Sep 09 20:31 post. At an indicated A-N =142 (118%) are we getting close to saturation of the PT?
Thanks
Bill

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[davidbeach]

Wye connected VTs used on anything less than a solidly grounded system should be rated L-L on the primary. thread238-173479 has a rather interesting discussion of what can happen when L-N rated VTs are used on an ungrounded system and they're allowed to saturate every half cycle.

 

[odlanor]

For ungrounded system you should use TPs group 1. They have different connection and insulation. See IEEE TP standard.

 

[Ceast]

Just to clearify, the gen is a Cooper-Bessmer recip 2400Vac Delta. The GSU is a 2400/12470 delta/delta. The distribution system is 12470 grounded wye. I'm picking up the pieces of the engineering debacle. It's in a small municipal.

The gen is direct coupled to the GSU. Also this particular unit has a direct coupled 2400Vac 250 HP blower motor used for combustion air. The motor must begin to spin as the engine is started. The motor may be causing us problems. I'll get back to that in a minute.

The VT's are wye connected 7200:120 ratio. I believe they should have installed VT's rated for line to line. Ungrounded phases are subjected to 1.73 times rated in the event of a ground fault on the delta system. How does that square when the gen is paralleled to the wye system?

I agree with M. Beach it's just wrong to parallel a delta/delta GSU to a wye system.

It's my opinion that the A phase VT is saturating at around 12.9kV line to line volts. I recorded over 150 VT secondary volts at 12.9kV. The VT voltage is very irractic (saturation) above 12.9.

What would happen if the combustion air motor had a low grade ground fault?

I'm going to test the motor in the next few days.

BTW I will also download the 1/4 cycle events from the blue box. The voltage phase angles should indicate presence of a ground fault. Right?

Thanks for the input.

 

[waross]

The motor on the 2400V side of the transformer should not affect the 12,470V side of the transformer.
Another issue that may arise is synchronization. If you have much leakage to ground on the wrong phase it may put your phase reference out by as much as 30 degrees.
If you have access to some distribution transformers, connect the primaries in wye to the 12,470V system and connect the secondaries in delta. This will give you a stable neutral point on the 12,470V system. The wye point should be grounded to maintain the neutral point at ground potential. This is a good example of an instance where neutral and ground are NOT interchangeable. There are serious safety issues if the neutral rises above ground potential.

Bill
--------------------
"Why not the best?"
Jimmy Carter