Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.


shape of bhp vs flow curve for radial, axial, mixed flow pumps

shape of bhp vs flow curve for radial, axial, mixed flow pumps

shape of bhp vs flow curve for radial, axial, mixed flow pumps

In Mark's handbook (old edition) I see typical curves for radial and axial flow pumps. The axial flow gives requires max bhp at min flow and decreases as flow increases.  The radial flow gives min bhp at min flow and increases as flow increases.

Is this true in general?

I have seen a curve which started low bhp at low flow, increased to peak bhp near bep, then started decreasing again. Would that be a mixed flow?

Do bhp vs flow curves for fans have identical behavior (to pumps) for radial, axial, mixed?

RE: shape of bhp vs flow curve for radial, axial, mixed flow pumps

Come on you guys.... give it a try. The simple question is:

What is the general shape of BHP vs flow curve for axial, radial, and mixed flow pumps?  

True or False?
Is it monotonic increasing for radial flow?
Is it monotonic decreasing for axial flow?
Is there a maximum point in the middle of the curve for mixed flow?

RE: shape of bhp vs flow curve for radial, axial, mixed flow pumps

My recollection is that the configuration of the power and current curves depend more on the the impeller blade shape than on any undefinable radialness, or mixed-flowedness for non-axial impellers. Backward curved blades give non-overloading power curves which peak somewhere near the best efficiency flowrate. Forward curved blades produce an overloading characteristic that has power rising steadily from low to high flows. Straight radial blades give something in between but I'm not sure exactly how the power curve slopes. Since most sensible centrifugal pump designers use backward curved blades for low and medium specific speeds, most of the pump power curves seen by humans have a peak near the middle of the operating range. vanstoja

RE: shape of bhp vs flow curve for radial, axial, mixed flow pumps


Your statement on both radial and axial pumps is true. An axial pump consumes more power in the shutoff condition and should not be run in that condition for a long time as you suggested in other post. The power consumption decreases as you open up the valve.

In radial flow pumps it is the other way round, that is why you are always suggested to start the pump in shut off condition so that initial torque on motor will be low.

As for mixed flow pumps, forgive me, I have no practical experience.(But even for some radial flow pumps there is a maximum point on the BHP curve in the middle, not exactly I mean)

True, Fans and pumps behave in a same way.


Truth: Even the hardest of the problems will have atleast one simple solution. Mine may not be one.

RE: shape of bhp vs flow curve for radial, axial, mixed flow pumps

Thanks for all of the above responses.

I had a talk with a semi-knowledgeable pump engineer on this topic. He walked me through various types in a Worthington pump catalog and I believe that basically he agreed with the statements I made above.

From my limited understanding, most of the single-stage pumps will fall in the pure radial or pure axial categories which will have the steadily rising or falling BHP curve with no peaks or valley's in the range of the curve (BEP and beyond to runout). Most of the mutli-stage pumps will have a mixed characteristic with peaks, valleys and wiggles in the operating range.  Does this paragraph sound correct to you guys?

RE: shape of bhp vs flow curve for radial, axial, mixed flow pumps


with respect to your question about fans: They generally behave in a similar way to pumps because the work at low pressures (usually) and so the air behaves as an incompressible fluid. The info I have is as follows:

Centrifugal fans:
backward curved blade:power rises to max at the middle of flow range then falls at highest flow rates, this is known as a non-overloading power characteristic.
Paddle bladed: power rises continuously with flow
forward curved blades: as paddle bladed.

Axial flow fan: non overloading as per backward curved blade centrifugal.

Mixed flow:non overloading as per backward curved blade centrifugal & axial (obvious really!)

Reference - Air conditioning systems design commissioning and maintenance, Roger Colby Legg ISBN 0 7134 5644 2. I think its out of print now.

RE: shape of bhp vs flow curve for radial, axial, mixed flow pumps

Thanks Shaun.  It seems your discussion for fans is pretty consistent with Vanstoja's discussion for pumps, in terms of dependence on blade curve.

If I can paraphrase both of you are saying that:
- backwards curved blades gives bhp-vs-flow curve with  peak in the middle and low on high/low flows.
- forwards curved blades gives bhp-vs-flow curve continuously increasing.
I also understood van to say that the backwards curve was common on the low/med specific speed which generally corresponds to high/med actual speed.

As you recall my observation was that:
- pure axial flow (common in single stage) gives bhp-vs-flow curve continously decreasing
- mixed flow (common in multi-stage) gives bhp-vs-flow curve with peak in the middle and low on high/low flows.
- pure radial flow (common in single stage) gives bhp-vs-flow curve continously decreasing

I have not looked closely at the curvedness of the blades for pumps curves I checked. Something does not quite add up. One thing is that your formulations do not provide any room for a pump with continuously decreasing bhp vs flow, but I have seen that. I will review a little more and see if I can find an example.

In the meantime, any other comments?

RE: shape of bhp vs flow curve for radial, axial, mixed flow pumps

There is an excellent book on the subject: Design of Centrifugal and Axial Pumps by Stepanoff, also the Pump Handbook by Karassik, Fraser, Messina and Krutzsch.
Left the Stepanoff book at home (different country) but as far as I recall... there is a math derivation showing that decreasing power vs. flow characteristic for axial pumps.
In layman's terms (I hope)... an axial pump will require more power at very low flows (with constant speed) because the impeller will create eddies... the efficiency goes down the drain so to create very little flow you require an awful lot of power.
As the flow increases so does the efficiency and more than compensates for the increased flow...therefore the power requirements go down.
Do I make sense of am I full of myself?

RE: shape of bhp vs flow curve for radial, axial, mixed flow pumps

On further review, I'd have to say that you are generally right about what you saw which may have been a repro of Fig. 9.3 in Stepanoff which is sitting on my lap as we speak. He has 7 brakehorsepower curves ranging in specific speed from 900 (double suction) to 9200 (single suction). The latter, #7 drops continuously from 25 to about 130% capacity. #6 at 5200 also continuously drops though at a shallower slope. Numbers 1 through 5 all show continuously rising curves. Meanwhile all the head curves are dropping with increased capacity and the two lowest 900 and 1500 droop below 50% capacity. I'm inclined to believe that these repesentations of power curves are somewhat idealized and may not hold up some cases. I deal mainly with pumps ranging from 2000 to 6600 in specific speed and can't recall having seen any with constantly  rising power curves though we do run them out to capacities approching 150% which is well beyond Stepanoff's Fig. 9.3 plot. The blade curvature effect I mentined earlier shows up in Stepanoff's Fig. 3.5 which is for an idealized pump. Blade discharge angles (beta 2) less than 90 degrees give a peaked power vs capacity curve. Beta 2 = 90 deg. is a 45 degreee constantly rising straight line. More than 90 deg.Beta 2 shows a constantlly rising and steepening curve which Stepanoff says "is never realized in actual pumping machinery." Figure 9.8 shows how shock and friction losses turn the idealized straight line high to low head curve into a parabola over the entire flow range. Since power consumption depends on head these same losses will drive the power down at both high and low flows. Only the highest specific speed pumps seem able to maintain a rising characteristic toward shutoff flow. Most of the others succumb to head loss effects so power drops off especially in the highest flow ranges. Not being a pump designer but rather a pump design EVALUATOR, this is about the best I can do theory-wise. Would you consider trading in your Electrical Engineers license for a Mechanical Engineering certificate, Pete, so we can get this pump stuff straightened out once and for all.

RE: shape of bhp vs flow curve for radial, axial, mixed flow pumps

Just a word of warning:

I have come across mixed flow pumps that are overloading at closed valve head.

RE: shape of bhp vs flow curve for radial, axial, mixed flow pumps

To sow,
I wouldn't doubt that mixed flow pumps in the high speed end would approach axial pump' rising characteristic toward shutoff. I just looked up Mixed Flow pump range in the Hydraulics Institute Standard and was a bit surprised to find that officially it extends from 4200 to 9000 (RPM-GPM-Ft). They don't recognize the so-called Francis-Vane pumps in the 2000-4200 range and just lump them in with radial pumps from zero to 4200. Francis vane pumps which generally predominate in the nuclear power industry main coolant pumps behave much more like mixed flow then radial pumps with lots of three-dimensional, secondary flow action not seen very much in mostly two-dimensional radial flow pumps below 1500 specific speed. Perhaps we need an official category recognizing the in-between Francis-vane pumps.

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!


White Paper: Introduction to Multi-patterning
As the semiconductor industry races to keep pace with Moore’s Law, which regards the doubling of transistors per square inch on integrated circuits, lithography hardware systems are not always able to provide a quality image for creating circuits. Download Now

Close Box

Join Eng-Tips® Today!

Join your peers on the Internet's largest technical engineering professional community.
It's easy to join and it's free.

Here's Why Members Love Eng-Tips Forums:

Register now while it's still free!

Already a member? Close this window and log in.

Join Us             Close