Auto-Synchronizing GE Frame 7EA


Thread Starter


It has been several years since we have had the Mark IV control system for our GE Frame 7EA combustion turbines. I remember that they auto-synched very quickly (5-10 seconds) after FSNL was reached.

From what I can tell reading the old elementaries, the frequency window was >.12hz and <.24hz differential as compared to line frequency which correlates to a window >100.2% speed and <100.4% speed (assuming line freq was 60hz). FSNL setpoint was 100.3% speed so that falls right in the middle. I realize that you want to synch a little fast but this seems a little too fast.

With our new control system, the time to auto-synch after FSNL takes much longer than it used to(30-40 seconds). The new freq window is >0hz and <.1hz differential as compared to line freq. This correlates to a window of >100.0% speed and <100.17% speed. FSNL setpoint is 100.1% speed which barely falls into this window and sometimes requires speed decreases to satisfy the auto-synchronizer.

My main question is: Is it safe to synchronize these units at a frequecy as much as .2hz above line frequency. It seems that it was doing that for 25 years before we replaced the control system. Can this be allowed because they are such small generators(80 mw)? I would like to widen our frequency differential window to >0hz and <.2hz on our new system to allow a faster auto-synch.
Basically we would still auto synch at 100.1% speed or 60.06hz (assuming line freq is 60hz) but a wider window will hopefully eliminate speed decreases. Thanks for your help.

Hmmm. It's noted that you didn't mention what your new turbine control system was nor what help they supplier has offered with this problem.

Auto-synchronizing involves two things: voltage matching and speed (frequency) matching. Well, truth be told, both generator voltage and generator speed (frequency) are usually adjusted to be just slightly higher than grid/bus voltage and -frequency but it's just a way of describing the process of automatically adjusting voltage and speed so that when the generator breaker closes the reactive current is, or should be, very low and the real power flow is very slight positive.

From your description, it seems the new control system isn't doing speed matching very well if an operator has to manually issue lowers (or raises) to get the generator frequency inside the synchronization window. And this may be the problem (as it is frequently on Speedtronic turbine control systems): the allowable window for voltage and/or frequency is just slightly larger than the synchronization window so situations can arise when no adjustment to voltage or speed will be automatically made because while the synchronization window is not satisfied the voltage/speed window is satisfied, requiring an operator to make a small adjustment to get the parameter inside the allowable synch window.

What I'm trying to say is that quite often the allowable synch window and the voltage/frequency permissive windows are not properly "aligned". Let me try to draw a picture.<pre>
Allowable Frequency Allowable Frequency
Minimum Maximum
| |
| | | X |
| |
Synch Relay Synch Relay
Minimum Frequency Maximum Frequency</pre>

'X' represents the actual generator frequency, which, in this example, is greater than the Synch Relay Maximum Frequency setting, so the control system won't lower the frequency because it's within the allowable frequency difference window even though it's outside the Synch Relay Maximum Frequency setpoint and the synchronization relay won't close the generator breaker.

So, I'm suggesting that you review the configuration/programming of the new control system to determine if something similar is happening, and then work to resolve it.

Now, as for the 100.3% speed setpoint versus the 100.1% setpoint. For an 80 MW machine with 4% Droop, 0.3% speed error represents (0.3 / 4 = 7.5% * 80 MW) 6 MW. This means that if the unit were synchronized with a speed error of 0.3% the load on the machine would be 6.0 MW upon breaker closure (which is pretty close to the default Spinning Reserve setpoint for most GE-design heavy duty gas turbines).

Now, that's a lot of load for many people, and not so much for others--it's all about experience and risk tolerance. The ONLY reason for having any positive speed error is to ensure the power is flowing OUT of the generator-set and not into (reverse power). The actual value can be anything that's acceptable to the owner/operator and/or the utility being synchronized to. (Many utilities don't want so much power on breaker closure.)

There's no set synchronization value for a GE-design Frame 7EA heavy duty gas turbine--it's all up to the power system designers and utility regulators at the time of construction, or at any time there is a new analysis/study of the power system in the area. The default synchronization values for all GE-design heavy duty gas turbines are all the same--and subject to change during commissioning or at any time afterward. I've seen utilities come back after 20 years and want to change synchronization parameters because of changes in the area power system (new generation and new loads nearby; new switchyards nearby; etc.).

So, again, from the information provided it would seem the new control system is either not automatically matching frequency (not likely) or the allowable frequency window is larger than the allowable synchronization frequency window (more likely).

Opening the Synch Relay Frequency window is going to cause more wear and tear on the generator-set load coupling. This is because it will allow more angle between the stator magnetic field and the generator magnetic field, and when the generator breaker closes the generator rotor will have to travel a longer distance--and then stop--to get into synch, and that means more physical "bump" on the load coupling when slowing down the generator-set to get the generator rotor to synchronous speed (grid frequency). Remember: The generator breaker can be closed at any time--but the closer the synch scope is to 12 o'clock then the "smoother" the physical bump to the generator set will be; the further away from 12 o'clock the breaker closes the more violent the bump will be. A couple of degrees isn't much, but there are great physical forces at work during synchronizing (stopping the generator--and turbine--rotors when the breaker closes). The idea is to try to minimize the physical forces as much as possible.
Thank you for your reply. Let me try to explain a little better.
First of all, saying that We have 80 mw machines was just a generalization. The droop setting is 4% based on a 72 mw base load.

We have a Selco T4500 auto-synchronizer that performs the synch function. It provides contact outputs for voltage raise/lower and speed raise/lower to the Emerson Ovation system. The max voltage differential is set at 10% and voltage matching is not a problem. The max frequency differential is set at .1hz and must be positive to avoid a reverse power condition. The operator does not administer speed raises or lowers to line the unit up. This is performed by the Selco T4500 contact outputs wired to Emerson inputs. When auto-sync was satisfied, the Selco T4500 would calculate when 0 phase occurs and issue a breaker close signal.

If we convert speeds to frequency and assume line freq is 60hz at the time of synchronization, the Mark IV min/max freq window was 60.12hz - 60.24hz. The FSNL setpoint was 60.18hz. This would load the unit to 5.4 mw at breaker closure and the speed of the turbine would immediately jump to 60hz.

Assuming again that line freq is 60hz at the time of synchronization, the Emerson min/max freq window is 60.0hz - 60.1hz, but as I mentioned, the freq error must be positive. The FSNL setpoint is 60.06hz. This would load the unit to 1.8 mw at breaker closure and speed would immediately jump to 60hz.

The Emerson freq window is lower and smaller than the Mark IV window used to be. I think this causes the Selco synchronizer to issue more speed raises/lowers to line speed up. I don't want to change where we synchronize (60.06hz). I just want to widen the allowable window from a max of 60.1hz to a max of 60.2hz. I think the synchronizer would then not need to issue any raise/lowers to be in the window and would issue a breaker closed command sooner.

The problem I am running into is that our engineering experts say that if we allow the generator to synch at a frequency as high as .2hz above line freq, it could slip poles and damage the generator. GE didn't seemed concerned with this as their max freq was .24hz above line freq. Of course if our speed controls are working properly, the synch would occur at .06hz above line freq.

Do you see that synchronizing as high as .2hz above line freq could damage the generator?

Thanks very much for your input.
No; that doesn't seem like it would harm the unit.

But, from the information provided it also seems like the synchronizer isn't working correctly as programmed.
I can see why you would say that the auto-synchronizer is not working as designed. If everything lined up perfectly and FSNL speed lined up perfectly at 60.06hz, it would be in the window and synchronize.

What seems to happen is that with a speed control error deadband of .1% speed, you might end up a little high on speed when you reach FSNL. This might put you outside the allowable freq differential window and the auto-synchronizer must now issue some speed lower pulses. I have also noticed that the unit might come up to speed at just below the window and the synchronizer still issues a couple speed increase pulses which puts you outside the window so it has to turn around and issue speed decrease pulses.

I wonder if a couple of speed increase pulses are inherent to the auto-synchronizer program to insure that freq error is positive. I will be contacting the manufacturer with that question.

It boils down to that I think the freq diff window is too tight and if I open it up just a little, it might sync faster. The Mark IV synchronized at 60.18hz (100.3% speed, 3610.8 rpm's, 5.4mw's). All I want to do is sync at the current Emerson design of 60.06hz (100.1% speed, 3603.6 rpm's, 1.8 mw's) without hunting around. Of course those speeds assume 60hz line freq.

I just have to convince our support engineers that widening the diff freq window that allows a worst case condition of .2hz above line freq wont damage the generator. The worst case in the Mark IV was .24hz above line freq. I can understand their concern but maybe it's because they are used to working with large turbine/generators (up to 800 mw's) and ours is of course much smaller. Maybe the smaller mass is why we should be able to be a little looser on the window. I thank you again for your input to my questions.

You may have something with the synchronizer needing to make at least one adjustment to "feel good" about the readings it's seeing. I have seen this before, but it was a long time when digital control systems weren't as sophisticated and well-trusted as they are today.

The generator isn't going to slip a pole at 0.2 Hz--unless the excitation is well below rated. Slipping a pole is more about the torque from the prime mover overcoming the magnetic forces of attraction because the excitation is too low OR the torque is too high--and that's not going to happen when the synchroscope is turning at normal synch speeds (which is generally one rotation ever 6-10 seconds, or 6-10 revolutions per minute).

Some autosynchronizers can close much faster than this rate--which is fine; it just means that there will be more real power after the breaker closes. But, at 0.1 or 0.2 Hz, slipping a pole is not a realistic concern--as long as excitation is at or near rated.

Your response to the manual synch question will be very interesting. If the unit is ever synchronized manually, how often is it manually synchronized? And, can any operator manually synchronize a generator-set to the grid?
I can't remember the last time we synced it manually other than to a dead bus for black start testing. Manual synching wouldn't be an option because we have to meet a 10 minute start requirement.