Fluctuating power generation in Hydro generator 1

[bartb102]

Weird situation. Antique hydroelectric plant, three 750KW 40Hz, 120RPM generators driving a 2.5MW 40Hz. motor, shaft coupled to a 60Hz. generator. Yes, 40Hz. generators.

Power from ONE generator fluctuates 200 - 300KW when paralleled with one or both other generators. Stable when only the one generator is driving the motor. The other two are reasonably stable. Variation is fairly rapid, as fast as can be observed on the GPS relay. Increasing field (over exciting it) seems to stablize power, but increases stator current. Motor power does not fluctuate outside of what is expected.

This is a relatively new condition, did not exist a year ago.

 

[waross]

The only way that the power can fluctuate much is if the input energy is fluctuating. If the water flow is stable, there may be a failing CT or instrument giving a false indication of power fluctuation. If the water flow is fluctuating the governor and/or flow control system need an overhaul.

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

 

[ScottyUK]

Do the governor settings match? What are they?


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[rasevskii]

Well, I know what plant that is, but won't mention it here.

Does the gateshaft position vary on the unit or is it stable? If it is varying, check out the governor: Air in the oil or dashpot, belt drive slipping, loose linkages in the position feedback...are these the original governors or have they been replaced?

What about the DC excitaion? Is the original AVR still in use? Does the DC excitation current vary? If the governor is acting in an unstable manner this can also cause momentary fluctuations in the field current, and large fluctuations in the stator current. Can you change over to manual excitation and see if that stabilizes the unit?

If it is an excitation problem, check thr PTs for loose connections on the HV and LV sides as well as for bad fuses.

Is the original metering still in use, with the old analog meters one can see what is varying easily. If these were replaced with some kind of digital system, it may be difficult to see what is going on on an instantaneous basis.

just some ideas...

rasevskii

 

[rasevskii]

If the gateshaft position is in fact, not varying, and all the instrumentation seems to be correct, then a rather far-out possibility is a turbine problem: I believe in this plant there are two or four francis turbines on the same shaft driving the generator through a bulkhead. The turbines are in a flooded headrace. The control gateshaft operates the wicket gates on each turbine via a linkage. All of this is under water. Could it just be that one of the underwater linkages has come loose or adrift, resulting in a power fluctuation even though the governor is doing nothing...?

rasevskii

 

[rasevskii]

In fact, a bit of further research on this plant shows that there are no governors anymore, only an actuator for each turbine, connected to the gates via a mechanical cable. Has anything gone wrong with this?

Also that the exciters are a static system. Has this been checked out, especially regarding PT and CT connections?

rasevskii

 

[bartb102]

Mr. rasevskii, you know the plant.
Gates are not moving.
Excitation varies a fraction of an amp, out of 90ADC. Probably due to brush loss. Drive to analog output control module is stable, fluctuates slightly to regulate KVAR's. I should force it to a constant value, see if I still get power fluctuation.
Generation is stable when problem generator is the only unit driving the 40Hz. motor.
Francis wheels are not in the best condition, but should cause equal trouble when operating as only unit.
Agreed, fluctuating horsepower applied to the shaft should be the only thing that can affect output KW, unless excitation is too low for the load.

 

[rasevskii]

Could this possibly be a cavitation problem in the turbine? At low loads francis turbines can come into a cavitation area, where power fluctuates widely even though the gate position remains constant. Usually it is not allowed to run a turbine in this load area. Can it be that when the other units are paralleled together that the load on the problem unit becomes so low that cavitation occurs?

The turbines, being old, will have likely different cavitation areas.

The regulating system should have a lower load limit that prevents this happening, set differently for each unit, activated only when the unit CB is closed, overridden by a protection trip of course.

BTW can the frequency changer be started from the 40hz side or only from the 60hz side? Does the syn motor have fixed excitation or an AVR?

rasevskii

 

[rasevskii]

Hang on, here is another thought: Is the submergence of the turbines in the pit sufficient, is air being taken in? Is the intake gate or are the stop logs fully open? Now it is winter there what about floating ice...

rasevskii

 

[bartb102]

Problem occurs at all gate positions. Generally the units are operated for maximum power output, with adequate dead band & time delay to prevent repeatedly stroking the gates.

Frequency changer is started using one of the units on the 40Hz. side. Has same excitation system as the units, operated to maintain a constant voltage. Generators cannot swing the bus voltage very much. We do not see the power variation in the 40Hz. motor. By the way, these generators are ungrounded wye, motor is grounded wye.

Problem happens regardless of weather. We can tell when the units are sucking air due to low intake water level- output power drops and air is discharged downstream. Power fluctuation due to low level is slower than what we observe on the problem units. Intake & turbine configuration hasn't changed in 100 years. Hmm, I'll take that back, trash racks were changed 3 years ago. Problem was first observed some months after that.

 

[rasevskii]

Now this sounds more and more like a metering problem, that is, the power (kW) is not actually fluctuating. Are the old original analog meters (GE's finest) still in operation on the switchboard, or can you only read the kw from the GPS relay? What sort of relay is the GPS? It may be something strange in the PT connections (earth fault) in view that the units have ungrounded neutrals. What is the stator voltage (I think I read 550V somewhere but imagine 2300 is more likely).

Does the stator current as seen on the actual ammeter and the kw as seen on the actual kw meter fluctuate? What is the frequency of this fluctuation? It can be that the old meters are too heavily damped to see any fast oscillation.

It could also be turbulance in the turbine pit due the new trashracks, but nobody noticed it before.

rasevskii

 

[rasevskii]

Well here is yet another idea: The units are tending to fall out of step with each other because there are no damper windings on the rotor (is this true)? But the freq changer motor would have to have a damper wdg, otherwise it could not be started as an induction motor. Or do you start it back to back with one unit (both excited)and bring up the speed manually? Then sync the 60hz Gen to the grid.

IOW the excitation can go too far into the underexcited region at times.

Solution would be to run all the units at unity PF including the FC motor. No sense in having a lot or VARS circulating around.

Are there any VAR or PF meters on the panel...

rasevskii

 

[bartb102]

We start back to back, synching at 60Hz. Then bring in the other two gen's at 40Hz. Generators are quite simple, no damper, one winding per slot, 120RPM.

Generators are 12,000V nominal. Make 36A at rated power. Yes, we do try to run at 0 KVAR.

Metering is a Basler GPS100.

 

[rasevskii]

If I understand correctly, the Basler GPS100 display is the only metering that is used, and the old analog meters on the swbd are no longer connected or used for readings? If so then the digital display may be jumping around for some reason.

I still suggest to look at the analog meters (if still in use) to verify any oscillation or hunting in kw or stator amps.

rasevskii

 

[bartb102]

Still doesn't answer why when troubled unit is only one on line it's power is stable. Adding one or two more units starts the problem.

First time it was ever noticed unit B was fluctuating, unit A was stable. Either by itself was stable. Put C on with B, no problem. Put C on with A, problem. Turned out unit A had a bad waterwheel.

Yes, units have been inspected.

 

[uhpo]

maybe some reactive between generators?? but this is not about kw, but yes about currents, cause if you got a better situation when you play with excitation, seems to be a problem with the reactive between both gensets... cheek the voltage on no load, and then start the motor, with one generator, and check the voltage again, so in all the gensets, if the voltage on no load is diferent or when the work alone with the load, for sure you got a problem with the reactive.....

i hope i could help, from my poor point of view.

 

[bartb102]

Thanks to everyone who contributed to this thread. Everything seems to come back to it being a mechanical/water flow issue, not really an electrical problem.

 

[rasevskii]

Yes it looks that way. From the available photos and videos, it appears that these turbines have cylinder gates rather than wicket gates, which may cause more unstable flow regimes across the runners.

The sample rate and filtering of the digital metering may be just that unfavorable so that he readout goes to extreme values.

rasevskii

 

[bartb102]

The GPS samples each physical parameter 12 times per cycle, sample rate is adjusted to match measured frequency. But I suppose a really poor waveform could upset the process.

 

[waross]

Sounds like a unique installation. Are there any online links that you are allowed to share with us that describe the station?
Thanks.

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