Bearing arrangement for vertical motors

[Cheetos]

Have a few questions about the bearing configuration for vertical motors. Just in case, DE - drive end. NDE - non-drive end

1) Typically, I've read that the NDE bearing takes the thrust load, and the DE bearing is more of a guide bearing. Why is that? Can DE bearing take the load and NDE bearing as the guide bearing?

2) In horizontal motors, typically, DE bearing is the locating bearing and NDE bearing has a wave washer to allow for shaft thermal expansion. This is to prevent the shaft thermal growth towards DE because you don't want to put strain on customer's coupling. In vertical motors for pump applications, if the NDE bearing takes the thrust, then you need to fix the NDE. Does that mean DE floats and the thermal growth will go toward the customer's coupling?

 

[BrianE22]

The DE bearing reacts more of the radial load. To keep the bearings nearly the same size, let the NDE bearing handle the thrust loads.

 

[JJPellin]

As a general rule, it is usually preferred to have a rotating shaft in tension, rather than compression. Placing the thrust bearing at the top accomplishes this. Many vertical motors are driving vertical pumps. Some vertical pumps have a lot of weight supported on the motor thrust bearing (vertical turbine pumps, deep well water pumps, etc.). Many of these are rigidly coupled from the pump to the motor. It is better to have the thrust bearing at the top to place the entire shaft in tension and support the weight of the pump rotor.

Johnny Pellin

 

[Cheetos]

Thanks guys for the response. I think it started to make sense to me. Do you guys know how vertical motors address the issues of thermal growth and which direction the shaft is allowed to grow?

 

[JJPellin]

If the coupling is flexible, it can be pre-stretched to allow for thermal growth. If it is rigidly coupled, the driver machine is usually not sensitive to axial thermal growth.

Johnny Pellin

 

[TugboatEng]

All of the motors and many of the gearboxes I work with are in a horizontal orientation and have the thrust bearing on the drive end side of the shaft as you described. There is an inherent weakness to this design because the opposite drive end bearing must have clearance in its housing so it can float with thermal expansion. Eventually, the outer race of these bearing begins to rotate and wears out the housing which starts causing alignment problems. It seems like it would be ideal to have the coupling manage the axial growth as it can be better engineered to do so without wear.

A slender shaft in tension may be flexible enough to allow the use of a cylindrical roller bearing to allow for thermal growth while a rigid shaft and compression would require a spherical roller bearing.

 

[Daren Samy]

My working experience with horizonal centrifugal pump motor is to keep the float on DE side of motor. NDE side of motor is kept fixed. DE side uses deep groove ball bearing with 0.1~0.2 paper gasket between bearing outer race and end plate collr. NDE uses only Roller bearing.

Daren!

 

[electricpete]

I’ll probably repeat what others said but here’s my two cents.

1) Typically, I've read that the NDE bearing takes the thrust load, and the DE bearing is more of a guide bearing.

Among large vertical motors (let’s say above 250hp), it’s fairly common that the NDE takes the downthrust (which is typically larger) and the DE takes the upthrust (which is often smaller and/or momentary). It is equivalent to what you’d call a “cross locating” arrangement if the machine were vertical. I’d say this vertical cross locating arrangement is more common in large motors than an arrangement where the upper bearing takes both directions of thrust.

Can DE bearing take the load and NDE bearing as the guide bearing?

Yes. We have a family of motors where the DE takes both directions of thrust load, including the downthrust which is the larger load. So the lower bearing is a fixed bearing (back to back angle contact) and the upper bearing is a floating bearing (deep groove loose in the housing).

2) In horizontal motors, typically, DE bearing is the locating bearing and NDE bearing has a wave washer to allow for shaft thermal expansion. This is to prevent the shaft thermal growth towards DE because you don't want to put strain on customer's coupling. In vertical motors for pump applications, if the NDE bearing takes the thrust, then you need to fix the NDE. Does that mean DE floats and the thermal growth will go toward the customer's coupling

Yes thermal growth would affect positioning of the pump. Typically the amount of movement is small in comparison with the available clearances especially if pump is positioned in the middle of those clearances.

As an interesting aside, the growth would not necessarily be toward the pump because the rotor is not the only component growing, the stator is growing as well. Since the vertical motor is supported at the bottom, if the stator and rotor both grew by the same amount then there would be no change in pump position. So what is relevant is the differential growth. Maybe you have an intuition that the structural components of the stator are not growing as much as structural components of rotor… that was my intuition as well until I was proven wrong by one of our vertical motor families. The vertical motor in question was what I called above the vertical version of a cross locating arrangement. Specifically a spherical roller bearing on top took downthrust, while a deep groove on bottom took any momentary upthrust. During refurbishment of these motors, we set the endplay to 0.010” (We didn’t realize at the time but the factory endplay setting was much larger). We did this for 6 motors. About 10 years later we had a bad lower bearing degradation event that we investigated. Examination of the lower bearing showed it had clearly been in an upthrust condition for most of its life (which is strange because any upthrust should only be momentary). We are lucky enough to have continuous temperature monitoring on these motors. Examination of the trends showed the lower bearing temperature went up after we refurbished (and reduced the endplay) on every one of these motors. Later we ended up setting the endplay to 0.025” and the lower bearing temperature went back down. What had happened was the stator was thermally expanding more than the rotor and eating up the 0.010” room temperature endplay to create a tension that loaded the lower bearing against the upper bearing. The upper bearing was a lot beefier and had no problem but the lower bearing coulnd’t handle it.


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