Axial Load 3

[mottto]

Hi , is there a device to measure the axial thrust of a vertical pump? I know it can be done using axial load cell , but ....i need to measure the total thrust generated by a vertical pump. Thank you for any idea/ help

 

[JJPellin]

I am not aware of any way to measure the axial thrust. The pump manufacturer should provide data for shaft stretch if that is the concern. Otherwise, I would probably calculate the thrust based on pressure and impeller geometry. Simply summing up the pressures acting on the areas and adding in the weights, you should be able to estimate the thrust load with a reasonable accuracy depending on your need. Why do you need to know thrust load?

 

[EricZhao]

I don't see why you can't measure the load with a load cell. Is the vertical pump's gravity your concern? Could you just subtract off the gravity?

 

[sprintcar]

Eric
You could place a load cell under the thrust surface of the bearing, run the wire thru the housing, seal it against lube or process fluid leaks, install the pump and connect all the piping and wiring, calibrate the load cell, then run your test and tear everything down.

OR
As JJPellin says, the pump supplier should have the thrust values readily available for the actual duty conditions since these are required during the design process to calculate the bearing sizes. A simple email or phone call would save a lot of time!

"If A equals success, then the formula is: A = X + Y + Z, X is work. Y is play. Z is keep your mouth shut."
-- by Albert Einstein

 

[d23]

mottto

As a vertical pump manufacture I do this everyday! I would recommend you calculate thrust load based on worst-case scenario!

Worst case thrust load occurs at shut-it. Why not just calculate it? Get the impeller area from the manufacture and multiply it times the pressure. You will also need to include the shaft area and physical weight of the rotating assembly.

Note:

At shut-in there is no up-thrust or fluid velocity lifting effect.

D23

 

[mottto]

Thanks for all your replies ,
i did calculate the thrust load , but i wanted to check it by measuring it. I came across an old technical paper where hydraulic fluid was used along with a thrust gage, just like a piston.Strain gage and elctronics wont do it if you have a rotating shaft at 1700rpm.Thank you all for your inputs

 

[electricpete]

d23 - can you explain the components of vertical thrust?

So far I understand the biggest one is the difference in pressure*area, producing downthrust. Biggest when d/p is highest at shutoff. That part makes sense.

what is fluid velocity lifting effect? what are the other components and how do they vary with flow?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[JJPellin]

I am not sure how to calculate the lifting force caused by flow, but I have no doubt that is exists. There was a good paper at the Pump User's Symposium last year about upthrust in vertical turbine pumps. And we have seen this in our plant many times. The upthrust force is greatest at high flow which is also where the down thrust from discharge pressure is lowest. At very high flow rates, the upthrust can overcome the weight of the rotor and buckle the shaft, resulting in a permanent bend. The pump companies can make design changes to the impellers to try and counter this. The main thing they do is remove the back wear rings and balance holes so that full discharge pressure is acting on the entire back side of the impeller. But this can result in excessive down thrust at low flow. The best solution is to control the pump so that it never experiences run-out flow or dead-head flow. In the normal range of flows, thrust is not a major concern.

 

[vanstoja]

Axial hydraulic thrust measurements in centrifugal pumps with rolling element bearing motors have been made with axial load cells and sometimes strain gaged beam structures installed in direct contact with the rolling element bearings. This requires special modifications to motor end caps to insert the instumentation. Where access to the motor shaft non-drive end is available and the shaft end has axial internal threading (say for crane lifting of a heavy rotor)it is possible to install a thrust-measuring domelike structure with an axial bar for straingage mounting or with a piston device in a pressurized fluid cylinder that can measure axial thrust load by controlling fluid pressure to one side of the piston with a known pressure-loading area on the piston face. Possibly, load cells and/or strain gages could be mounted at or near the pump to motor coupling which is also accessible.
To answer electricpete's question, there are generally two main components of centrifugal pump axial hydraulic thrust loading, ie.,unbalanced pressure across unbalanced front and back face (shouds and hubs) and the fluid turning reaction force in the impeller channel. Equations are:
T=0.433*A*H + 0.00139*Q^2*sin theta/Ae where
T=Thrust(lbs), H=Head(ft.), Q=Flow (GPM), A=Unbalanced area(in.^2), Ae=Impeller eye area(in.^2) , theta =impeller channel turning angle (degrees). Calculation of impeller unbalaced pressure area becomes increasingly complicated as specific speed of the pump increases because of the curvature of the front shroud in shrouded impellers. Since the dominant factor in axial thrust is pump head the shape of the thrust curve versus flowrate, when normalized should be about the same as the pump head-flow curve. However flow instabilities in impeller channels often affect thrust more than head so dips in thrust curves often reveal unstable flow ranges not evident in the head curve.

 

[d23]

electricpete

Sorry for the delay. I’m kind of hit and miss for available time on the internet.

When I think of vertical pumps I tend to think closed type radial or mixed flow multi stage pumps. That may not be the case here.

With a closed type impeller in a vertical position there will be X-amount of thrust applied in an upwards fashion. This is due to the velocity or fluid momentum of the fluid entering the eye of the impeller. Guess a simple way to think about a single stage pump would be to consider what happens as fluid enters the eye for a closed impeller.

With a closed type centrifugal stage in a vertical position there will be three sources of thrust.

• Gravity is always downward.
• Down thrust resulting from the net pressure differential.
• Up thrust resulting from fluid momentum entering the eye of the stage.

Typically I expect my multi stage pumps to operate with a little down-thrust; however if one of my pumps (closed impeller type) operates at too high of a flow the fluid momentum entering the eye of the impeller will lift the impeller in an upward motion. In cases where we have operated with 0 head or open flow I have seen pumps self-destruct in up-thrust. They will try to lift the shaft and impeller out of the pump housing.

Sorry about the confusion. Mottto did not say anything about closed impellers. I made an assumption and you know where that left me!

D23.

 

[electricpete]

Thanks vanstoja (John) and d23.

I like the idea of downthrust curve looking like the head curve, makes it easy to remember.

Tell me if this is the correct way to understand the momentum part of it:

Water enters the impeller going up and leaves going at some angle away from up and toward radial.

F = M dv/dt = d/dt(mv) = rate of change of momentum

force exerted on fluid is change in momentum accross the pump

If fluid is changing it's momentom from upwards toward radially, then change in momentum is downwards. Downwards force on fluids requires equal and opposite upwards reaction force on the shaft.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[PUMPDESIGNER]

electricpete - Did you get d23s attachment I forwarded to you?

Your explanation is true for volute pumps, not true for turbines, axial flow, etc. There is no radial thrust on turbines (almost none), because the flow through the impeller is concentric, all radial thrust is balanced by the concentric flow.

PUMPDESIGNER

 

[d23]

PUMPDESIGNER,

Thanks!

I can explain a low easier with pictures.

D23

 

[electricpete]

Thanks. That was good information

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Enkidu]

Does anyone look at suction pressure? I always thought that was a main contributing factor.

 

[thewellguy]

To JJPellin:

I saw your post and I was wondering if you know where I can find a copy of the paper you referenced. Thank you