Generator Vibration 3

[Flashover]

Hi All,
interesting one for the board.
we have 150 Mwatt, 181.2 MVA generator (onto 110kV system), when operated at unity power pactor or slightly lagging we have no problems. however when run at full load and 0.85pf lagging we get vibration levels rising rapidily after about three hours (i.e there seems to be a thermal lag effect, everything reasonably steady then vibrations just take off)
having examined the vibrations on Bently Nevada scope we see that the vibration is 1 * Frequency, i.e 50Hz vibration which leads me to the conclusion of shaft imbalance brought on by shaft bending or rotor coils shifting.
has anybody experienced similar problem??
would love to hear views on the above.
Many thanks
Flashover

Flashover

 

[byrdj]

Does the vibration levels go down when load is reduced?
Many years ago, (~1970 vintage units same size as yours). units had similar increase in vibration brought on by exceeding a field current level (load and pf dependant). once vibration got rough, to reduce the vibration, the unit had to be taking off line and speed drooped to less than 500 rpms (back to gear for an hour was the practice). the problem was with the support for the end turns under the retaining ring. as the field bar expanded with current, the suport would stick and cause an endturn to buckle. The fix was replacing the end turn insulation (removing the retaining rings).

You may want to ask a repair facility if this could be a common problem for your design.

Hope your problem is not as costly.
Good luck

 

[Flashover]

The vibrations go back to normal operating levels when the power factor is restored to Unity. we can then continue to operate at full load.
we are now testing at 0.95, 0.90 etc
cheers

Flashover

 

[byrdj]

are you operating within design. I pulled a capibility curve I had. it showed MWs needed to be reduce to 90% when going from 90 to 85 Pf.

This would be an operating range where field heating is great. I don't have a answer, but suggest looking at the field cooling. If H2 cooled, besure the coolers are vented and flows set to balance delta T. changing the cooler exit temp may help. Purity and pressure should be verified also

 

[Flashover]

Hi all,
its an air cooled machine.
the gaurantee is 150 MW at 0.85pf.
any suggestions for further investigations that may help?
thanks

Flashover

 

[byrdj]

another comment
A Generator thermal unbalance vector is not uncommon. The balance shot called will give a compromise resolution that will reduce the high load (thermal induced) without making low load excessive. Is magnitude and phase data available at various loads?

 

[Flashover]

Latest development, is supplier believes that the cooling water for cooling air system should be increased. i.e it is an Air cooled generator (the air is cooled by closed cooling water system).
i am afraid this will just cover over the real problem of heating in the rotor. are there any tests in particular i should look for?
many thanks

Flashover

 

[ScottyUK]

A strong possibility is that you have a shorted turn on the rotor. You have all the classic symptoms:

- Apparent thermal lag
- Dependent on excitation
- Vibration signature

As the field current increases, so does the current in the shorted turn. The rotor will suffer a thermal bend as a result of heating in the area of the shorted turn. The rotor will then be out of balance. You need to address this with reasonable urgency.



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

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

 

[neps3]

Flashover

Quite agree with ScottyUK. Scotty is correct. Have given excerpts from our Consultancy Technical Newsletter on few issues on generators:

The phenomenon that is known to vary the rotor vibration as the field current is increased is “thermal sensitivity”. The main driver for this phenomenon is the large difference in coefficient of expansion (thermal) between the Copper winding and the rotor forging. Shorted turns issue is a reversible thermal sensitivity.

Shorted turns:

Shorted turns adjacent to a pole is most significant. Heat dissipated in the pole that has the shorted turns is less than that of the other pole. The temperature of the pole of the shorted turn, therefore, is less than the pole with good winding. This results in the expansion of the forging in the axial direction and hence a thermal bowing.

If vibration is detected in a generator rotor, tests can be carried out to determine whether the issue is due to shorted turns. The main electrical test is known “Recurrent Surge Oscillograph” (RSO) which detects shorted turns. This test is an “offline” test.

David Albright of Ohio (GeneratorTech) has developed flux probes which can be installed in the stator to monitor rotor winding condition. This is an “online” test. Quite an accurate test.

Cooling:

Differential expansion can be caused by uneven cooling as a result of blocked cooling ducts or cooling passages. Maintenance staff should carry out inspection during overhauls.

Slot wedge tightness:

Wedges tightness has to be uniform. If it does not remain uniform then the tight wedges can cause binding in the axial direction resulting in a bow. Soft footing of pedestal can add to severe vibration of the rotor. Operational test may reveal that the bowing is due to electrical problem, however, tests should be carried out to confirm whether the bowing is due to short of turn or any other problem.

End caps:

End caps shrink fit should be adequate and insufficient shrink fit can cause the end caps to move on the shrink fit and change the centre of mass resulting in vibration. Specialists and manufacturers or OEM should be consulted for correct shrink fit.

Your issue is seems to be a shorted of turns issue as it is a reversible phenomenon with the excitation. Check the rotor winding with a RSO instrument to detect any short of turn.

Wish to add that I have used Generator Tech rotor fluxprobes and am quite happy with it. It is an excellent equipment for online short of turns detection.

Thanks

Kantor

 

[Flashover]

Thanks very much all,
the improvement in cooling conditions has not resolved the problem. Big decisions required on what way to proceed

regards

Flashover

 

[ScottyUK]

Two main options were highlighted by Kantor: 1) An RSO should pick up the problem without having to remove the rotor. 2) For more definite tests, install an airgap search coil in the bore of the stator. There are quite a few companies with the equipment needed to analyse the results, but basically they are looking for assymetry in the rotor flux as it passes the coil. The shorted turn causes this assymetry unless you are dead unlucky and have two diametrically opposite faults. A 'fingerprint' of a good rotor is invaluable. The coils aren't too expensive, but the effort required to remove the rotor just to install it makes it impractical. The AGSC is a worthwhile installation if you have to take the rotor out for other reasons though - you never know when you might need one (like now...).

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

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

 

[edison123]

Another test to determine the rotor turn shorts is to measure the AC impedance of the rotor winding (at stand still)and compare it with the design value (provided by OEM). A lower impedance will confirm the turn short.

 

[Flashover]

Thanks all
Very much appreciated

Flashover