We found high vibration on a turbine-generator unit. Vibration - vertical and horizontal 1×rpm - is high on turbine outboard bearing, and also high only horizontally on generator outboard bearing. ODS shows that moving starts from turbine, and shakes the whole unit horizontally, with nodal point at the coupling. Turbine vibration is not constant in time, rises after start from 2 to about 8 mm/s both directions. Support was injected with epoxy, former cracks were eliminated, vibration level on support dropped to normal level and support moving seems to be quite good. There was no significant change in vibration levels of machine points. Dial indicator measurements at turbine outboard foot shows a 0,25 mm axial difference at two sides (about 75 cm from each other) mostly during the whole casing expansion, and about 0,4 mm lateral (!) movement. Turbine outboard "left foot" pumps the oil under itself. Seems to bee not a simple expansion problem. Seems to be a thermal casing distortion. Seems to be a quite high angle misalignment between shaft and bearing. Clearance is about 0,2 mm as I know.

Am I right? Could it be serious? What additional measurements to do? What to do to solve this problem? Must to open the turbine housing? Could it be a crack in the housing? Could bearing misalignment be responsible for high vibration through uneven clearance?

Libra,

An ODS test does not reveal where vibration starts. It simply reveals the amplitude and phase of points on the structure.

If you have shaft probes, then filtered 1xSS amplitude and phase data during the speed runup and over time as load and temperature changes would be inportant information. A rotor thermal bow or unbalance with a possible rub could be the cause of the high 1xSS vibration.

Walt

I think it can't be clear unbalance because there is not too high vibration at critical, and just after startup vibration is quite low. Clear unbalance causes higher horizontal than vertical. It could be a shaft thermal bow (causing unbalance as well), but why moves the front foot laterally, why expands casing differently at the two sides? Axial vibration not so high, a bent shaft should cause high axial vibration. Bearing misalignmet also can cause rub.

And what to do with a bent shaft? Must to open the housing, I think.

It is not possible to use a shaft probe on the turbine side, no any place to do that, nor shaft relative probes are present. This machine is more than fifty years old.

Libra,

OK, so no shaft probes, then I would still get 1xSS amplitude and phase data during runup and over time at constant speed. There are simply too many possiblities that cause vibraton at 1xSS, and you need to reduce this list before opening up machine.

Walt

If you suspect casing distortion and thermal growth, optical alignment can help verify if this is the case. It uses theodolite and involves cold and hot survey on each bearing to determine thermal growth.

Thx for good idea! I'll try to organize this inspection.

Rgds

Libra

Libra,
I am assuming that this is steam turbine. From your description the machine runs up fine and fully loaded and it is only after being loaded for some time that you experience the increase in vibration on the turbine O/B and generator O/B bearings.
If you can get this far it probably is not a rotor bow or unbalance.
You should inspect the casing keys to ensure that they are not being "locked up" and have adequate clearance. Also, if this turbine has the inlet piping coming from one side it may be imparting a side load on the turbine.
The piping should be fixed at the turbine and should have expansion loops taking care of growth. Check the piping supports and hangers which may have become immobilized over time. Hangers also have hot and cold positions check these also.

I agree with Walt's approach take amplitude and phase data during start up and keep sampling as the amplitude increase. Look for any change in phase when the amplitude changes.

Regards,
James

Yes it is a steam turbine. There are steam inlet valves both sides, at startup works only one inlet valve, at 4,5 MW opens the other side one. The casing key is a cca 100 mmm long and 40 mm diameter cylindrical guide finger (sorry, if the term is not correct), and after warm up you can't find clearance at one side,and there is a clearance at the other side, but this clearance is not paralel but misaligned. the key was not locked up at this time, but machine was far from full load. As I read in the first message the machine moved laterally on the base plate.

I'll check the piping supports and watch the vibration increase vs foot displacement on baseplate.


Thx

Libra

Libra,

James comments were very good. It would appear that there are two possible causes of the problem:
1) Case thermal growth/distortion somehow causes a rotor rub at high load or over time
2) Differential thermal growth of bearing housings causes shaft misalignment.

To answer item 2, you would need data on the change in shaft alignment from cold to hot or viceversa plus the actual shaft alignment under stable cold conditions. There are several methods to measure the thermal change in shaft alignment including:
1) Laser-optical system like Permalign
2) Triangulation Bars to bearing housing targets (Eissenger)
3) Thermally stable stands with dial indicators (C. Jackson?)
4) Dynalign bar system (R. Dodd)
5) Optical or laser transit to targets on bearing housings
6) Measure vertical tenperature profile at each bearing and calculate vertical rise and plot result to scale

I have used Methods 2, 4, 5 and 6. My advice is that you should not rely on case movement measurements to assume that bearing housing is moving the same amount. You need to know how much all four bearing housings are moving.

Walt
w_f_strong [at] msn {dot} com