Pumps performance

[Bob511]

Hi,

There is one thing I am still struggling to understand about pumps which is the relation between BHP and capacity of pumps.

I know that on radial flow centrifugal pumps the higher the head the lower the capacity on a specific pump the less amps it draws and the lower the BHP requirement. What I do not understand is why the axial flow pumps behave in the opposite way. Is there a description in terms of fluid mechanics as to why is that? I mean what I understand is more flow means more fluid to be moved means higher BHP and more amps. The opposite is true for axial flow pumps.

 

[BigInch]

OK, first that's not always true. They sometimes have a differently shaped head curve. As this link explains, curves in between the three that are shown are still possible. Power at any point is the result of multiplying the pump's curve coordinates (Q,H) together, then dividing by efficiency coordinate. If the curve has a little wave in it, power results may not be always continue going higher as you move left to right, or right to left for that matter.

This should explain. If it doesn't, tell us what you didn't understand here,

http://www.gouldspumps.com/cpf_0009.html

"We have a leadership style that is too directive and doesn't listen sufficiently well. The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward CEO, BP

**********************
"Being GREEN isn't easy" ..Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://vir

 

[25362]

Following BigInch's clear explanation about the fact that power is not only a function of Q, but of the product of Q and H, one may say that the power consumption of a pump with a relatively flat QH curve would usually increase continuously with the flow rate, and this may lead to overloading the power supply in the event of a rupture in a discharge pipe or some similar accident. A pump with a very steep QH curve, however, may overload the driver at reduced flow rates.

 

[rmw]

I worked with some big (96 inch was big to me) mixed flow pumps and the curve was almost flat throughout the whole pumping range. But when the suction and discharge gages were read with accuracy and the motor volts, amps and PF read, the pumps were right where the curve said they should be with respect to power consumption (and double checked with other flow measuring devices).

I still scratch my head, knowing more that it is true than exactly why it is true. Those were about 5K HP, so you couldn't monkey around much.

rmw

 

[Bob511]

Thank you for the explanation you provided. I already saw the link and I am still looking at an explenation in terms of fluid mechanics and pump dynamics. Is there some mathematical descriptive analysis of what is going on? I mean like in terms of impeller shape or the like.

 

[BigInch]

Its mostly various components of friction that are the cause of the curve being something other than flat. In the regions of the curve towards the left, bearing, seal and stuffing friction components are relatively high while fluid friction entering and moving through the pump is still low. As you move to the right fluid friction becomes a major component and its friction is squared with flow, so the typical centrifugal pump curve falls off rapidly as fluid flow increases and its friction^2 takes over. Other friction components are more linear with pump speed. There is also the "bite" of the impeller, which at a given rpm will be feeding the impeller with exactly the volume it wants at some flowrates, but at other flowrates the impeller tends to eat some of the volume and "throw the rest away", or rather more technically, recirculate it back to towards the suction end. Sometimes this is enough to cause a wave in the curve when the impinging flowrate exactly matches the impeller's "bite size and swallowing capacity".

"We have a leadership style that is too directive and doesn't listen sufficiently well. The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward CEO, BP

**********************
"Being GREEN isn't easy" ..Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://vir

 

[Artisi]

Bob511,
Is this what you are looking for in mathematical term -specifc speed - Ns

Ns = rpm x square root of flow @ BEP / H^3/4; this gives the impeller type.

Radial flow Ns is upto approx 1500
Francis vane upto approx 4500
Mixed flow upto approx 8000
Axial flow from 9000

 

[BigInch]

Artisi, while Ns does relate to impeller shape, IOM it doesn't explain much about "what's going on" (although that term is a bit subjective).

Continuing along with my explanation, I'd say mathematically what's "going on" can be simplified to an equation for pump differential head vs flow, Q.

Head = A * Q^2 + B * Q + C
A is a coefficient relating H to fluid friction effect,
B is a coefficient relating H to linear friction effects, as Q relates directly to pump rotational speed, thereby also to friction of stuffings and bearings.
C is shutoff head (head at zero flow)

There are still friction effects when flow is zero; the pump is rotating, in which case it's the net effect H-H_{friction Q=0} = C, is what appears on the pump curve as shutoff head. (see attached)


For pumps with secondary curve components making those little waves, additional terms to those in the simple equation have to be included.

"We have a leadership style that is too directive and doesn't listen sufficiently well. The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward CEO, BP

**********************
"Being GREEN isn't easy" ..Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://vir