Pump Design Question 1

[gun66]

So the problem I am having is when you design an impeller you use specific speed term. From this you get different coefficients that have been determined through experimental data that give the dimensions for D2 impeller diameter and b2 impeller exit width. So now let’s say I have a set of initial conditions, flow, head, rpm = 6000 gpm, 1000 rpm, 160 ft head. Using data from any book or charts available I will get a specified D2 and b2. Now let’s say I want to increase b2 but hold D2 constant to increase the flow. I am aware my pump curve will get flatter and percent head rise will decrease, and thats fine. What I did now what select a larger b2 and worked backwards to see what my flow would be in the equation...however this might be the wrong approach as the specific speed is still the same.

The other option would be to select a gpm and calculate a new specific speed with a new b2 dimension. Both ways are come up with similar answers since the Ku constant for flow is pretty flat around these specific speeds.

So now my b2 and D2 dimensions are kind of based off of different specific speeds….1600 or 2000 let’s say. So what specific speed do I use for future calculations, such as the volute constants I will be required to look up on the casing design?
I feel as I am overlooking something really simple, any help would be appreciated.
Nick

 

[BigInch]

How'd you get 1000 rpm?

 

[gun66]

1000 rpm is a given based on our performance chart and where we want the pump to operate.

 

[Artisi]

"1000 rpm is a given based on our performance chart and where we want the pump to operate."

Are you discussing an existing pump/impeller design or are you designing a new impeller.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[1gibson]

I think the question trying to be asked is: How will you achieve 1000 rpm?

1) Diesel or tubine? OK
2) 6 pole 60hz motor on VFD? OK... but why not design for 1180 rpm and ditch the VFD.
3) 6 pole 50hz motor? Not OK. You need to account for the motor slip, your actual speed will be 985 rpm or so depending on how close to full load on the motor.


Also, if you are designing a new impeller, what exactly is the "performance chart" you refer to? Trying to put this delicately: are you attempting to create a high flow impeller by using the manufacturer's performance curve and physical dimensions from a low flow impeller (that you currently have on hand) ?

Artisi, I think the answer to your question might be yes to both.

 

[gun66]

This is a new impeller.

These are either direct drive, or belt driven...belt most of the time which is why 1000 rpm is not an issue sheaves can give you any speed you want. The rpm is not the issue, you can pick any rpm you want and design the pump based of that. The only difference will be where excatly the BEP will end up lying on the pump curve.

This is a very general basis for any pump design
givens: rpm, flow, and head which gives us specific speed.
Unknowns: Impeller OD, Impeller Width, Entrance/Exit Angles, Cutwater OD, Hydraulic losses, vane shaping, volute shaping, shaft diameter ect ect.

No I am not using an old impeller and increasing the width, this is an easy modification that only increases efficiency slightly and reduces percent head rise. With a brand new impeller and different specific speed than the previous model so all parameters will change.

Specific speed is a dimensionless paramter used for pump design.

Anyhow I think I have it all figured out at this point, thanks for your help guys if you have any additional comments feel free would love to hear any ideas on this still.