GE Isolation Transformer MV Cable Sizing Criteria

[kingpb]

Set-up:

GE Frame 7FA turbine, with Fast start Isolation Transformer, 5450KVA, 4160V HV side, with split secondary with a delta and Y winding both rated at 2725KV, 2080V.

This transformer feeds power to the GE LCI system. I am trying to determine how to come up with the criteria for properly sizing the MV cable on both the HV and LV sides. I have seen multiple installations and it appears every designer has a different approach (no consistency). I can't find any GE documentation other than one job it said it needed to be capable of 8000KVA and another that said 8500KVA for back to back starts. And they state some capacitance issues and criteria if there was no ac line reactor.

Anybody experienced with these cable requirements?

Would appreciate input.

 

[rcwilson]

GE has several documents for sizing the cables. Depending on the type of 7FA gas turbine the startup profile is a little different. Some have faster (or slower) start up sequences so the drive pulls more power for a shorter period of time. (Sorry, I can't share the document yet).

For those unfamiliar with GE Turbines, this "LCI" (Load Commutated Inverter) is a 4.16 kV VFD that starts the gas turbine by using the 18 kV generator as a synchronous motor. A gas turbine has to get up to speed to make its air intake compressor work before it can generate power. The LCI/generator combination does this.

The LCI /VFD rolls up the generator/turbine and holds about 30% speed to purge any residual fuel from the turbine and the downstream heat recovery equipment. (2-15 minutes) Then it coasts down to light off speed (14%) so the airflow doesn't blow out the pilot light. Once the flame is verified, it goes to full power and accelerates the TG to full speed, 3600 rpm, with the gas turbine finally helping out around 80% speed and taking over at about 90%. This 1-2 minute acceleration can take 11 MVA and about 9 MW peak. Most 7FA's are about 9-10 MVA and 7.2 MW peak.

Once at speed, the LCI is disconnected and the generator exciter turned back on to generate 18 kV, 60Hz.

The LCI -generator system has four sets of cables:
- 5 kV from plant switchgear to the isolation transformer,
- 2080V cables (really six- phase, 4.16 kV double wye) from isolation transformer to LCI input bridges,
- LCI to/from DC Link Reactor cables, and
- output cables from LCI output to disconnect at generator, (current to the stator = Is).

Due to the VFD waveforms, each cable set carries different currents.

I took GE's starting sequence withthe stator amps, calculated the I^2 T for each step, summed the total I^2t and calculated the equivalent constant current that would put the same heat into the cable over that time. I size the cables for that plus a safety margin. (GE recommends minimum 15%).

ETAP simulations of the starting profile and the cable temperatures in underground ducts and have proven that the cable doesn't overheat with back-to-back start attempts.

The calculated current ends up about the same as the FLA of the Isolation Transformer. (Makes sense, the transformer and cables both average out the heat input for a varying load.)

Size the isolation transformer feeder cable at 125% of the calculated isolation transformer FLA. (2 to four per phase).

DC link cables carry 1.2825 x Is, where Is is the stator current. (GE's figures).

Size the stator cables for the calculated current + safety factor (15-25%). Isolation transformer to LCI cables are same size, per GE.

Be sure the overcurrent relays and upstream equipment can handle that 1-2 minute 11 MVA+/- load without tripping.

We haven't melted any cables yet, and we've done many, many of these. We run Ampcalc or ETAP duct bank derating calculations on the cables and upsize as necessary for duct bank configuration, ambient, etc.

 

[kingpb]

Thanks much, that's what I was needing.:thumbsup:

 

[iceworm]

... For those unfamiliar with GE Turbines, this "LCI" (Load Commutated Inverter) is a 4.16 kV VFD that starts the gas turbine by using the 18 kV generator as a synchronous motor. A gas turbine has to get up to speed to make its air intake compressor work before it can generate power. The LCI/generator combination does this. ....

rc -
I'm not familiar with GE turbines. I've worked on Solar, Garret, Seimans, and a few different RollsRoyce models. All but one little Garrett has the compressor turbine separate from the power turbine. Since this GE can spin the generator and get the compressor shaft spun up, then it must only have one shaft. true?

ice

 

[rcwilson]

GE's smaller units, like the LM-2500, are double shaft with the turbine and compressor on seperate shafts running at different rpm. The larger units (LM600, Frame machines) are all single shaft with a solid coupling or a gear box to the generator. Most large CTG's turn at 3600 rpm (for 60Hz) with one shaft, direct coupling to the generator and no gearing.

For combined cycle operations, a single shaft machine line up with a 250MW gas turbine on one end, a 120 MW steam turbine on the other and a 400 MVA generator in the middle all coupled together is popular in Europe and the Mid East. Some suppliers have a clutch between the generator and the steam turbine. GE uses a solid coupling so the LCI starter has a lot of inertia to get rolling.

 

[iceworm]

GE's smaller units, like the LM-2500, are double shaft with the turbine and compressor on seperate shafts running at different rpm. The larger units (LM600, Frame machines) are all single shaft with a solid coupling or a gear box to the generator. Most large CTG's turn at 3600 rpm (for 60Hz) with one shaft, direct coupling to the generator and no gearing. ..

Really interesting. I don't know what sizes the GE frame numbers refer to. The biggest I have been around are in the 20MW range. I've never seen one without a gearbox. The PT's tend to turn in the 9000 - 1000 rpm range. The gens are, of course, 3600rpm. I'm guessing when power goes up, the PT speed slows.

As for the single shaft, I'm guessing they are designed to operate on-grid and won't operate well islanded. I'm guessing they won't hold frequency very well compared to one with split PT and compressor turbine shafts.

Q: What is the "C" in "CTG"?

ice

 

[rcwilson]

CTG= Combustion Turbine Generator. Used to be GTG for Gas Turbine, but "gas" turbines can also run on fuel oil and other strange products like steel mill off gas.

 

[kingpb]

Really interesting. I don't know what sizes the GE frame numbers refer to. The biggest I have been around are in the 20MW range. I've never seen one without a gearbox. The PT's tend to turn in the 9000 - 1000 rpm range. The gens are, of course, 3600rpm. I'm guessing when power goes up, the PT speed slows.

As for the single shaft, I'm guessing they are designed to operate on-grid and won't operate well islanded. I'm guessing they won't hold frequency very well compared to one with split PT and compressor turbine shafts.

Q: What is the "C" in "CTG"?

ice

The 7FA, which has been replaced by 7FB is the 60Hz large scale turbine capable of +\- 170MW. The 50Hz version is the 9FA or 9FB. The "A' or "B' has to do with efficiency.

These can be used in an island operation, however you would need either a lot of engine generator capacity or something like a LMS100 (GE 100MW machine which would still need engine generator to start) in order to provide the black start capability.