Got drawn into a discussion today, and I thought I knew the answer, but now am not sure. A supervisor said that pumps are built in such a way that the dynamic forces are balanced radially, such there is no net load on the pump bearings. Further he stated, the weight of the pump rotor is actually compensated for such that there is not even a static load on the pump bearings. That is, there is more 'up force' on the impeller than 'down force'. He said that fans are designed the same way.

I'm pretty sure that's not accurate, but hey, I'm no expert. What do ya'll think about this (or better yet, what do you know)

I am not a pump designer but it is common knowledge that pumps are designed to have impeller radial forces balanced at Best efficiency point. We see vibration when a pump is off it's curve or the volute is severely erroded. I do not know for a fact that gravity is countered by a radial force, but if this is true bearings would have to be designed different. Ball bearings would skid and sleeve bearings would be unstable. Myth Busted .

I'm not sure that the radial forces are balanced at BEP, but they are certainly lowest.

And yes, I agree that if there were effectively NO radial forces on the impeller, you would have all sorts of problems with bearings. Especially spherical rollers, given that many pumps will have a significant axial load.

I'm aware serious pumps and fans sometimes have fins on the back of impellers to help reduce >>axial< forces.

I believe It would take a clever non-traditional design feature to oppose gravity and "cancel" the weight in a horizontal shaft pump.

uniform pressure in the volute, thus no radial loads due to pressure is theoretically correct at Best efficiency. But the The cutwater or tongue has influence even at BEP.

http://books.google.com/books?id=4ktxIf91viUC&pg=PA121&...sSHnsmzRpgaY#PPP2,M1

This site confirms no radial forces, but some of their explanations of other are phenomenon are not to my liking. http://www.mcnallyinstitute.com/04-html/4-12.html

Basically a single volute will have odd number of vanes while a double will have even number of vanes for balance or to acheive best balance. Discharge may not be on the top. Cutwater clearence should be 10-12% or not less than 8% as a rule of thumb. There are techniques to reduce hammer at the volute lip passing. But to expect all balanced - have Peter Pan rebuild your pump.

The seals/wear rings may help support the gravity load, assuming good alignment internally. The wear rings do act like bearings. I can't see this for fans.

Of course the overall up force (static) is balanced with the down force or the impellor would not stay in one place. With fluid film bearings one can look for radial position as an indicator of net force on the bearings; this may be somewhat more difficult for rolling element bearings. Often, there is a large side load on a pump at off peak efficiency points.

Double volutes or diffuser arrangements can cancel a number of vane pass harmonics.

Flow out of the impellor vanes can never be exactly balanced. This results in a 1X force; one should consider how good of a mechanical balance is needed due to this unknown force, which is flow dependent.

Rusty,

I'll tell you what I "think" about it because I don't really "know" much.

I think that you can design until you are blue in the face, but if you get "no" load on a bearing, you probably did it by some freak combination of errors.

It is not correct to beleive that the bearings do not deal with load. When the pump is in operation forces are exerted in two directions. In the axial direction as well as in the radial direction. Thrust bearings take care of forces in the axial direction. Because of rotation of pump radial forces are generated and the radial bearings takes care of these radial forces.
I have seen design engineer apply forces using hydraulic jacks and monitor the bearing temperatures after placing the pump to rotate using temporary fixtures.
Mechanical seals are not used to take shaft loads.
Bigger pumps bigger bearings bigger load carrying capacity. Does that make sense guys.

quote:

Further he stated, the weight of the pump rotor is actually compensated for such that there is not even a static load on the pump bearings.

I thought if I lived long enough some of the Star Trek stuff would come true. Did not know that Anti-Grav was available.
It seems that most agree that there has to be load on pump bearings. I worked on a lot of pumps and compressors before doing vibration work and that is what the bearings are for, to handle the load. It certainly could not be the seals or the wear rings. Not sure where this was coming from. Rusty, this guy was not suffering from a Tequila Sunrise was he?

Radial forces are almost balanced at best efficiency points. Never heard of pumps which will have no static loading on bearings. At other than best efficiency points, radial forces are not balanced. That's why pump shafts have to be designed taking into consideration their deflection in running condition to minimise it. Almost perfect radial balance can be achieved by providing double volute casings.

quote:

Almost perfect radial balance can be achieved by providing double volute casings.

Sounds good and looks good on paprer. But 30 plus
years working with double volute pumps I find this is seldem true.

Forces on the bearing and pump efficiency are two different things.

Pump efficiency is calculated using this formula:

Pump Efficiency = Water horse power divided by the brake horse power.

Water horse power = Q X H X Sp. Gr divided by 3960 where Q is flow in GPM, H is head in feet

And B H P is the power delivered to the motor for above operating condition.

So the point I am making here is that the balancing situation or loading on the bearing has nothing to do with pump efficiency.

quote:

Originally posted by Riyaz Ahmad Dave:
Forces on the bearing and pump efficiency are two different things.

Pump efficiency is calculated using this formula:

Pump Efficiency = Water horse power divided by the brake horse power.

Water horse power = Q X H X Sp. Gr divided by 3960 where Q is flow in GPM, H is head in feet

And B H P is the power delivered to the motor for above operating condition.

So the point I am making here is that the balancing situation or loading on the bearing has nothing to do with pump efficiency.

Riyaz,
You have misunderstood my statement.Force on pump bearing and best efficiency points are two related things. Every pump is designed to run at a certain discharge and head, i.e., where the system resistance curve intersects the pump characteristic curve. This is the pump best efficiency point. At this point, the efficiency of the pump is highest and radial thrust on the pump shaft will be minimum. If radial thrust is minimum, forces on bearing will also be minimum. When the pump operates at any point other than BEP(Best Efficiency Point), loading on bearings will be more due to higher radial thrust.
Regards.
Irshad

quote:

Originally posted by Barry:

quote:

Almost perfect radial balance can be achieved by providing double volute casings.

Sounds good and looks good on paprer. But 30 plus
years working with double volute pumps I find this is seldem true.

Barry,
Will you pl. explain your statement because I still see so many critical pumps in our plant and everywhere having double volute to reduce radial thrust.
Regards.
Irshad

Akhtar,

You are right. Forces on impeller are minimum at BEP.

The intersection of system head curve and pump curve may or may not be at the BEP.

And by the way where in India are you. I was born and raised and got all my education in Mumbai.

This is my first post on this board. As a pump engineer I think I can help out a bit. A volute pump will always have some residual radial loads. These will be at a minimum at the Best Efficiency Point (BEP). In addition as you add volutes, double volute, triple volute (rare but does exist) the offdesign radial loads drop. This is why diffuser pumps have low radial loads. However, you still have the weight of the rotor. Note that for the volute pumps the angle at which the load is applied changes as you change the operating point. In some cases if the radial load magnitude is greater than the rotor weight you can have the shaft running on the upper part of a sleeve bearing.
For axial loads there are methods to balance these out and theoretically a balancing disk does this but it does not react well to tranisient conditions.

Regards,
Dan