Shaft size estimation

[shahyar]

Hi friends,
Is there any formula/thumb rule to estimate pump shaft size roughly?
Assuming we know flowrate and head of pump.

Thanks

 

[Montemayor]

shahyar:

The way a practical engineer does it is he/she look in the pump catalog listing and find what the manufacturer picked as the shaft size. There is no better practical and horse-sensical method.

Unless you plan to be building pumps, why would you want to be able to calculate the best/optimum/desirable/practical/etc. shaft size? And if you don't know how to do it now, why go into pump fabrication? Manufacturers select their shaft size from practical/economical reasons. They can calculate the theoretical size that will do a given job, but then they have to consider other things: other impeller/hp uses of the same pump shaft, surges, contingencies, etc. This is the real side to engineering: the real life applications - of which there may be more than one.

 

[Artisi]

Don't forget that the bearing selection to suit the loading of the pump hydraulics also plays a major part in shaft sizing. As already pointed out, the shaft design is a complex but also a very practical science.

Naresuan University
Phitsanulok
Thailand

 

[Holly41]

I don't know of any way to calculate shaft size using only GPM and head. There are some very good reasons to calculate shaft size though. The biggest one is to minimize shaft deflection. Depending on the type of pump, this could be the single most contributing factor in seal and bearing failure. If you are talking about an end suction centrifugal pump, use the L3/D4 ratio calculation. This would be the lenght from the bearing center line to the impeller, cubed; divided by the diameter of the shaft to the 4th power. If this number is less than 60, then the shaft size and diameter are OK. If it is higher, then the amount of shaft deflection will, over time, shorten the life of your bearings and seals.

 

[sprintcar]

The best way to determine pump shaft size is to ask the pump vendor. They did all the work - don't try and second guess them. If shaft failure is an issue (there's another thread in here about that) first look at your system conditions and see what changed since initial installation. Then look at the shaft for changes (material, vendor, etc.) and let your suppliers help you solve any problems.

To do the actual calculations, you need all the hydraulic characteristics of the pump, bearing dimensions, shaft dimensions, plus motor and drive details.

"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

 

[GoncaloBaptista]

The answer to your question is No. You need further information such as rotational speed in order to calculate impeller diameter and you need to define what material you are going to make your shaft of. Once you have the impeller diameter you can then calculate torque and the stresses.

 

[checman]

I agree with Holly41, There are two many different design factors that go into designing a pump to give you the flow and head you are looking for. Stress on the shaft very with the design such as impeller diameter and rpm. L3/D4 ("L-cubed over D-to-the-fourth") is a measure of pump rotor stiffness, it's ability to resist radial load and to minimize deflection.
It comes from the basic cantilevered beam deflection formula, which you can find in any book on mechanics: y = F x L3 / (3 x E X I), where F is radial load, L is cantilevered length, E modulus of the elasticity of the material, and I is moment of inertia.
The lower L3D4, the less is shaft deflection, which is better for mechanical seal alignment, vibration and shaft whip. If L3D4 becomes too large, a pump shaft can snap, especially if operated close to shut-off, where hydraulic radial loads are excessive and not ballanced on volute type casings. ANSI pumps have L3D4 ratios range from 20 to 120, but newer designs have been introduced with this ratio below 10.0.

Regards Checman

 

[Artisi]

I would suggest that if you know the head, flow and rotational speed you could "estimate" with a lot calculation and research the impeller diameter and blade width with a reasonable degree of accuracy, from there it would be possible to estimate a shaft size after setting a lot of parameters such as bearing spacing, impeller overhang (if an end suction pump) shaft materials etc, etc,. - but why bother - just apply the thoughts of Montemayor - simple and practical.

Naresuan University
Phitsanulok
Thailand