Direction of Thrust. 1

[poppeye]

I have a two stage high head split case pump in for repair. Operating conditions 900gpm @ 1200ft. Pump has a 3' suction lift. Impellers are douple suction fully enclosed type. Pump inlet on outboard end of pump (thrust end)with the discharge on opposite side at coupling end. My delema is the pump is fitted with MRC 8309 Angular Contact Thrust bearings. This type of bearing is directional sensative and must be installed properly in accordance with direction of primary thrust.
Since both impellers are fully enclosed double suction, these impellers should be hydraulicly balance with no major thrusting. From what I have given, which direction would be the direction of the primary thrust. Pump wearing ring clearnaces and innerstage bushings are at factory recommended clearances, with the pump hydralulcs operating 1.5% just to the left of peak efficiency on the curve

Thanks
Poppeye

 

[goutam_freelance]

Hello popeye,

I think the best course of action here is to consult the manufacturer's maintenance manual. The thrust should hydraulically balance and the small residual thrust should depend on manufacturer's design.

Regards,

goutamiam

 

[poppeye]

goutamiam,
Unfortunately, the manufacurer Flowserve, has been unable to respond to my inquiries about this pump. Cause of failure to this unit was not associated with the thrust bearing, but during the disassembly and inspection phase, I noticed that the thrust bearing was installed with primary thrust indicated by the direction of the bearing as thrusting from outboard (suction & thrust end) to coupling end. This pump was designed and manufactured at a Flowserve facility that is no longer in existance. So trying to find someone that is familiar with this design is really gettng frustrating. All of the factory manuals I have on this particular model of pump do not show this type of bearing configuration, nor do they address the direction of their installation or pump thrusting. I know that my manuals for this pump are up to date since my company was a Flowserve Representative untill a year and a half ago. I tried to use some of my back door contacts at the factory to no avail.
I stumbled on this web site last week, and thought I would throw this situation to the engineering experts. The model of pump is a Flowserve 4GTR. I just want to be 100% sure of the bearing direction of installation before I give it back to a customer that depends on it's performance and dependability.

Thanks,
Poppeye

 

[Pumpenhirsch]

Popeye,

you are right, normally one stage pumps with double suction impellers are balanced. But some pump manufactures are installing wear rings with different diameters to get a defined axial force in one direction.

At your pump with 2 stages and a very low specific speed ns of approx. 670 there happens something special in the areas between impeller and casing. Normally you have a leckage flow from the outer diameter of the impeller to the inner diameter and then through the wear ring back to the suction. But at a 2 stage pump you have the bushing between the both stages. Through this bushing you have a leckage flow from the second stage to the first stage. This means that you have a leckage flow from the inner diameter to the outer diameter impeller at one side of the first stage and at the other side the leckage flow is as usual. But leckage flow from inner to outer diameter reduces the the rotational speed of the fluid ring between impeller and casing !! And thus increases the pressure distrubution in this area. This means that you have at one side of the impeller a higher pressure than at the other side and so an axial force !

So the result in your case is that the second stage impeller is balanced, but the first stage produces an axial force. This effect increases with low specific speed.


Regards,
Pumpenhirsch

 

[smckennz]

Hi Popeye

perhaps you should think about the effect of the coupling, and thermal growth of the motor shaft and perhaps the pump shaft.
Forget about hydraulic thrust for the moment. We can come back to it.
You said the thrust bearing is at the non drive end and its principal thrust direction is towards the drive (coupling end. With a (non solid) coupling,the face to face clearance is provided to allow for expansion of the shafts, among other things. With the bearing arranged to thrust toward the coupling, growth can be absorbed by the coupling and things can work to a rational design. If the thrust face was in the opposite direction, then shaft growth would tend to unload the thrust bearing, so the thrust bearing would wind up doing nothing useful.

The thrust caused by the balanced impeller hydrodynamics is a little unpredictable, unless a thrust bias is designed in by the use of unequal sized wear rings/inlet areas. It is more normal to use a bidirectional thrust assembly such as a deep groove ball bearing, or dual thrust bearings back to back. You will need to check the ring diameters to find whether or not a thrust bias has been designed in. However it sounds like the pump designer has assumed negligible hydraulic axial thrust; a little scary.
If a compression spring is fitted between the coupling faces (to expand it), then things would make some sense.
My bet is that the "as-found" primary thrust direction towards the coupling is correct.

Cheers

Steve McKenzie

 

[poppeye]

Thanks Guys,
From what I am understanding the following is correct.
1. Leakage from 2nd stage to 1st stage through the innerstage bushing will cause hydraulic thrusting on the impeller shroud that sees the leakage against. This would cause the shafting to thrust from 2nd stage to 1st stage.

2. Thermal growth of the pump shaft is not a concern as the fluid being pumped is from a salt pond that fluctuates at ambiant. Coupling is grid falk type with .250 between pump shaft and motor shaft.

3. Impeller wear rings are of equal size and diameter. So from what I get the only thrusting would be from leakage against the inboard impeller shroud of the 1st stage.

Thanks
Poppeye

 

[JJPellin]

I would take a slightly different approach to this one. First, we have about a dozen pumps of this configuration. All of them are in cold crude oil service. The original documentation for all of them indicates a matched pair of 7308 angular contact thrust bearings. (Our pumps are slightly smaller than yours) The configuration of the pumps makes it impossible to predict net thrust. It is even possible that the net thrust could change based on flow rate or wear ring leakage as the rings wear. Just because the pump was found with an 8309 (PumpPac) bearing set, I would not assume that this is the correct bearing. With no clearly defined net thrust, this is a poor bearing choice. A matched pair of 40 degree contact angle 7309 bearings should work very well. If there was a problem with ball skidding on one bearing, I would consider a pair of 15 degree contact angle (DiamondPac) bearings instead. The net thrust of this pump is going to depend on leakage across the center bushing, subtle differences in wear ring clearance, mechanical seal diamaters and casting irregularities. Trying to predice net thrust seems like a exercise in futility.

 

[dryrain]

Well about the thrust we use a lot multistage vertical centrifugal pumps and the thrust load ( the load along the shaft line) is downwards. ie towards the suction. As the fluid is pumped against pressure the shaft is loaded by this thrust load and it increases with pressure (with considerable shaft strain). The bearings are indeed sensitive and the 'fat' coller of the bearing must be towards the rotor of the motor( if thats the prime mover). The bearing collar that has more delta r( difference of radius) . About the shaft strain u must contact the manufacturer

dry

 

[flowman]

poppeye,
You must have qty. 2, Thrust Bearings. If you have a cross-sectional of the pump you will probably notice the bearings are installed face to face or back to back. This allows for thrust abillity in either direction. Typically you will want only .005 lateral movment of the bearings in the bearing housing, adjusting with gasket thickness. You will want to specify flush ground set when you purchase the bearings.
Flowman

 

[rdcowan]

Can someone help me with the torque sequence and values for the 1.125" and 0.875" nuts/studs that hold the pump halves together? (We can't seem to locate a detailed maintenance manual)

THANKS!

Robert

 

[rmw]

rdcowan, this is a good question, but off topic for this thread. Recommend you start a new thread on this one.

rmw

 

[quadtracker]

I agree with the use of a pair of 7309 angular contact bearings mounted back to back, this should look after axial thrust in both directions. These are often used in most Union Pump horizontal multi-stage pumps with shaft end play at about .005" adjusted by gasket/shims on the bearing housing.