thrust runner design

[ginsoakedboy]

The results of sectional-3D FEA on a Thrust Runner show that the Runner "bows" under the thrust load. The axial displacement of the periphery of the Runner is ~60 µm. The thickness of fluid-film between the Thrust Runner and Tilting Pads is estimated to be ~10 µm.

Radial interference of 5 mils/inch diameter is modeled between the Shaft and the Runner. The applied loads are: (i) Interference, (ii) Centrifugal, and (iii) Axial Thrust (mapped from 1 Tilting Pad on to the Runner Surface).

I am concerned that the excessive deformation on the periphery of the Runner will cause the lubricant film to be unsustainable, and drastically reduce the load-carrying capacity.

My question is, how much displacement is allowable on the Runner?

 

[ginsoakedboy]

Please help.

 

[MikeHalloran]

Do the pads tilt in a direction and to a degree sufficient to allow a uniform film thickness? Does your fluid film model account for the possibility of one surface being conical?



Mike Halloran
Pembroke Pines, FL, USA

 

[unclesyd]

Have you thought about using a different lubricant in the model?

Some EP additives and viscosity modifiers have a pronounced effect on the displacement of an oil film.

 

[ginsoakedboy]

The tilt pads should allow rotation about any axis passing through the pivot (as in the pad has 2 rotational degrees of freedom).

Mike,

We use pressure mapping obtained from one of the bearing analysis codes to generate the FEA pressure loads on the face of the pad. So, there is no "fluid-film model".

Are you suggesting that it may be possible to perform the entire fluid-structure interaction simulation in some kind of a software package?

Can you send me more information on such a simulation?

 

[MikeHalloran]

>>>Are you suggesting that it may be possible to perform the entire fluid-structure interaction simulation in some kind of a software package?
<<<

I'm suggesting that it _should_ be possible, given any of today's "multiphysics" solvers, and enough computer. to analyze the flow through a bearing with flexible pads.

I'm not suggesting it will be easy or straightforward.



Mike Halloran
Pembroke Pines, FL, USA

 

[ginsoakedboy]

I have looked into the multi-physics simulations and gather that it would be a lot of effort and the results may not be worth the time and money spent.

What would be the largest permissible axial deflection at the OD of Thrust Runner? The working surface on the Thrust Runner is ~15" OD x 6.5" ID.

 

[beretta24]

Have you tried looking at the problem from a different angle such as questioning what the minimum effective area is for the bearing? What happens to the bearing analysis if you reduce the radial size of the pad in the area that's experiencing questionable deflection?

Can you increase the thickness of the thrust runner or change materials to decrease the deflection?

 

[racookpe1978]

What's the shape of the thrust runner?

Can you add stiffeners or ribs behind (on the back side) of the thrust runner to prevent movement?

Seems like you got a more of "structural" strength (bending-related) problem due to high loads trying to go through the metal runner, rather than a lubricant-caused problem between the runner and the fixed surface.

 

[electricpete]

Radial interference of 5 mils/inch diameter is modeled between the Shaft and the Runner

Is such a tight interference required? I'd think if the runner is sandwiched axially between a shoulder and a nut, (so the nut works with the fit to carry the load), you should be able to relax your interference fit considerably and reduce resulting distortion. Just a thought fwiw (I'm no rotating equipment designer).

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

 

[unclesyd]

The Rotordynamics Software that maybe of help. If I recall they will do a one off analysis or if needed they will be sell you the software.

This program was used to correct some bearing problems in our HNO3 plant compressor which allowed us to operate a higher RPM from a safe 10,000 RPM to a safe 12,500 rpm, a previously unsteady operating point. 25,000 hp likes steady state operation.

http://www.rsr.com/index.html

 

[electricpete]

For my own info, I would be interested to hear if 5 mils per inch interference is normal. It sounds very high to me.

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

 

[MikeHalloran]

It's huge.


Mike Halloran
Pembroke Pines, FL, USA

 

[tmoritz]

Hmm...a 5 mil interference fit would be a loose fit between the thrust collar and shaft on one of our hydro units. That leads me to suspect the scale of ginsoakedboy's thrust bearing is a bit smaller than what I'm used to working with.

GSB, can you provide any information on the dimensions of the thrust runner, axial load, shaft rpm, etc. that would give me and others a better idea of what you are up against. I mean a 60µm deflection isn't all that big if your thrust runner is 120 inches in diameter.

Regards,

Tom Moritz
US Bureau of Reclamation

Tom Moritz
Mechanical Engineer
US Bureau of Reclamation

 

[electricpete]

Hmm...a 5 mil interference fit would be a loose fit between the thrust collar and shaft on one of our hydro units.

Did you mean 5 mils or 5 mils per inch of diameter?
We've been talking 5 mils per inch

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

 

[tmoritz]

My oversight. I didn't see it was five mils/inch.

Tom Moritz
Mechanical Engineer
US Bureau of Reclamation