Axial shaft movement turbogenerator (Cyclic movement)

[tiagoandrei]

Hello guys,

We are facing an interesting problem in a synchronous generator (4 Poles , 1800 RPM , nominal power 6.5 MW) driven by a gas turbine and gearbox. The shaft is presenting an axial movement (~1,5 Hz) when the load is about 1.6 MW.

Apparentely, the load (active or reactive) in the generator doesn't have any relation with frequency of the movement or its amplitude. We have conducted a lot of tests to comprove this theory.

The axial bearing was damaged by a continous operation of this turbogenerator on this condition. Actually is an single sleeve radial-axial bearing.

We have checked for misalignment between generator and gearbox. A little deviation from manufactur recommendations were verified. Although alignment were corrected, the problem still persist.

We also have conducted a lot of inspections in coupling (gear coupling) looking for something wrong but nothing were observed.

I have read some threads in this forum about the "hunting magnetic center" phenomenon present in eletric motor, but nothing regarding generators. It can also happen ?

 

[edison123]

Axial as well as radial magnetic centering requirement applies to all AC and DC machines, both motors and generators.

In your case, the axial magnetic center issue could be causing this hunting.

Muthu

http://www.edison.co.in

 

[electricpete]

From the video, you have enough movement to be easily visible..

If you can see the shaft on the other sideof that coupling (gearbox side) and it's not moving the same amount, then I would think your gear coupling is in question from the excessive sliding over a period of time. How that fits into the rest of the picture (cause, effect, unimportant), I'm not sure.

The axial bearing was damaged by a continous operation of this turbogenerator on this condition

Which axial bearing are you talking about? One in the gearbox? Or the momentary-use thrust bumper (shaft shoulder against bearing edge) in the electrical machine?


=====================================
(2B)+(2B)' ?

 

[electricpete]

Which axial bearing are you talking about? One in the gearbox? Or the momentary-use thrust bumper (shaft shoulder against bearing edge) in the electrical machine?

If it was the bearing in the electrical machine bearing ( momentary-use thrust bumper / shaft shoulder against bearing edge) that was damaged, then your coupling was not set up properly. The endfloat of the coupling should be limited to prevent this type of contact.

=====================================
(2B)+(2B)' ?

 

[tiagoandrei]

Axial as well as radial magnetic centering requirement applies to all AC and DC machines, both motors and generators.

In your case, the axial magnetic center issue could be causing this hunting.

Which are the possible problems for a issue in a magnetic center since this generator have been operating for 10 ou more years ?

I was thinking that magnetic center is a project concept, regarding equilibrium of magnetic forces. Can it change with a long operation time ?

Which axial bearing are you talking about? One in the gearbox? Or the momentary-use thrust bumper (shaft shoulder against bearing edge) in the electrical machine?

I was talking about the one in eletrical generator.

I am not sure why you used the therm: "Momentary-use bumper".

If it was the bearing in the electrical machine bearing ( momentary-use thrust bumper / shaft shoulder against bearing edge) that was damaged, then your coupling was not set up properly. The endfloat of the coupling should be limited to prevent this type of contact.

We have another 2 generators like this one. None of them have never presented this problem before, neither the one with the same coupling. Why only this generator is working like this ?

The other ones has a axial movement when it is excited in the start up process. But it's a "one-way" movement in the gearbox direction.

What I can't understand is why this specific generator is hunting for the magnet center and doesn't stop.
Our coupling has parallel gear, how it is possible to it limit the movement ?

 

[edison123]

In the normal running position, the rotor core and the stator core must align axially as close as possible. This is axial magnetic centering, which is fixed for life for a machine.

See this thread for earlier discussion on this issue.

https://www.eng-tips.com/viewthread.cfm?qid=444858

Muthu

http://www.edison.co.in

 

[electricpete]

I am not sure why you used the term: "Momentary-use bumper"

The large electric machines that I am familiar with have sleeve bearings with no dedicated thrust (axial) bearing other than the axial edge of the sleeve bearing which makes contact with a shoulder on the shaft (that's what I call a thrust bumper). This type of thrust bearing is not designed for continuous axial contact, only momentary. These types of machines (without any continuously-rated thrust bearing) also have an endplay of approx 0.5 inch (12 mm) and require the rotor to be held in a mechanically centered position (centered enough to avoid continuous contact with thrust bumper) by the thrust bearing in the connected equipment, through a limited endlfoat coupling.

I suspect your machine is the same but I'm not positive. Does your machine have a large endplay / endfloat when uncoupled?


=====================================
(2B)+(2B)' ?

 

[tiagoandrei]

In the normal running position, the rotor core and the stator core must align axially as close as possible. This is axial magnetic centering, which is fixed for life for a machine.

Thank you Edison123 for the link.

The large electric machines that I am familiar with have sleeve bearings with no dedicated thrust (axial) bearing other than the axial edge of the sleeve bearing which makes contact with a shoulder on the shaft (that's what I call a thrust bumper). This type of thrust bearing is not designed for continuous axial contact, only momentary. These types of machines (without any continuously-rated thrust bearing) also have an endplay of approx 0.5 inch (12 mm) and require the rotor to be held in a mechanically centered position (centered enough to avoid continuous contact with thrust bumper) by the thrust bearing in the connected equipment, through a limited endlfoat coupling.

After your explanation I pretty sure that our bearing is this "Momentary-use bumper" you are talking about. I mean, for sure it was not design for a continuous axial contact and your description fit on it (See the file attached)

Does your machine have a large endplay / endfloat when uncoupled?

We didn't check for it yet because is difficult to us to force a movement on this shaft (too heavy).
But for measurements analyses on the bearings and (also in shaft) it can be seen that the bearing's float is smaller than the endplay coupling. So in cases of an axial movement on the shaft the bearing will responsible to limit this and consequently will be damaged if a "large" cycle movement happens (it's happening).

So I am concluding, as you said, that my coupling should be responsible to limit the axial movement...

Another information that I forgot to mention, is that no kind of anormally in generator voltage, current, frequency was seen. No additional frequency is been generated. We also have look for faults in excitation system but nothing was found.


So can I say for sure that the axial movement is caused by the high endplay coupling ? It's a mechanical problem ?

 

[electricpete]

So can I say for sure that the axial movement is caused by the high endplay coupling ? It's a mechanical problem ?

I'm cautious of saying anything for sure, especially about causes (there may be multiple factors at work).

One thing for sure is that if you have excessive float in the coupling, then that does need to be addressed. You may be able to reduce the endplay of existing gear coupling by adding a plate or a button (according to

http://www.couplinganswers.com/2015/01/gear-coupling-tutorial-part-ii.html

)
... maybe that part has been left out during asesmbly. Adding the button/plate should dramatically reduce the magnitude of this axial movement to within the new range of float (but I don't think you can set gear coupling float all the way to zero and still be able to accomodate misalignment).

Other things you might look at include axial magnetic centering as edison mentioned. i.e. you would prefer to have the coupling hold the rotor at the position of the generator stator's electromagnetic center, so that the ability of the magnetic force to restrain movement is maximized. In motors we can determine the magnetic center by running uncoupled and seeing where the rotor goes. In generators (assuming you can't run it as a motor) you have to rely on info from the OEM. If oem info is not available and you have to try to guess magnetic center, you could shoot for the location where the ends of the rotor iron are axially centered with respect to the ends of the the stator iron.

And I think misalignment and possibly gear coupling tooth condition can contribute to axial movement. If it were my plant, I'd be nervious about the condition of the couplnig anyway, based on what it's been through (I'm surprised the visual inspection suggested the teeth were ok).


=====================================
(2B)+(2B)' ?

 

[edison123]

You could see if it is mechanical or electrical problem by observing the shaft movement with and without generator excited. If the shaft axially moves only when the generator is excited, it is most likely magnetic center issue.

The coupling by itself is unlikely to create such an axial movement. As Pete says, there are other reasons also.

Muthu

http://www.edison.co.in

 

[tiagoandrei]

Thank you for your comments edison123 and electricpete . I really appreciate them.

You could see if it is mechanical or electrical problem by observing the shaft movement with and without generator excited. If the shaft axially moves only when the generator is excited, it is most likely magnetic center issue.

We already have done this test. Although the generator is not excited the axial movement is still present.
We are thinking that this situation (with or without excitation) is not valid for a true test because residual magnetic forces can still remain and it is forcing the generator to find a magnetic center.

What do you think ? This residual magnetism in the rotor is sufficient to create a false analysis?

Just to exemplify, in the condition of no current or voltage excitation, the generator create a output voltage of 1000 V(nominal 13.8 kV).

 

[edison123]

Nope, residual magnetism is too weak to cause the axial swing. I would check the alignment, coupling and the generator leveling.

Muthu

http://www.edison.co.in

 

[electricpete]

To my thinking the magnetic force plays more a role to restrain oscillation (from other causes) than to cause oscillation. It is the motors with weak (not strong) magnetic centering forces that are more susceptible to axial oscillation. (he magnetic force acts like a nonlinear spring in parallel with any spring action from the coupling. Why do we care about magnetic centering then? Because when at magnetic center, the magnetic force-per-displacement (effective magnetic spring constant) is largest and therefore the ability of the magnetic force to restrain axial oscillation is maximized. I came to this realization as documented in wandering comments in this thread: THREAD237-413657

If you look at the pdf files I attached in that thread, I gave several . I looked at expressions for magnetic force vs displacement. The slope of that force vs displacement curve is the effective spring constant of the magnetic force. That slope (spring constant) is maximized when the rotor is centered.

Regardless of whether you subscribe to my explanation above or not, you clearly have a mechanical problem, because residual magnetism is quite a small effect. The airgap flux density from residual magnetism is at least an order of magnitude smaller than airgap flux density from normal operation.

=====================================
(2B)+(2B)' ?