Centrifugal Pumps - Torque-Speed Curve

Check the pump curve for your size of pump

The curve itself will not change. As p911 says, the position on the curve will change.

Well that does depend on how you increase this flow through the pump. If you are increasing the flow rate though the pump by increasing the pump shaft torque then the new torque/flow combination should land exactly on the pump curve. (Within a reasonable measurement of error.) If instead another device is forcing more fluid through the pump then the torque to flow curve relationship at that pump will change. At least this is how I understand your question.

Dear Mr. redfred,

You are absolutely right.

DHAYANANDHAN.S

Look at your pump performance curve. The flow at the head required to generate enough pressure for your system is generally shown with the corresponding required horsepower ( based on a 1.0 Specific Gravity for water.)

Depending on the S.G. of the fluid being pumped, your pressure generated and your HP requirement will be a direct ratio of the specific gravity of the fluid to that of water. Since pressure= .434 x h x sg the point on the curve to generate the necessary head for the volume will show the required HP for water and multiply that by your SG to get the required HP for your application. The further out you go the more HP will be required. If there is a change in the SG of the pumped fluid this also can drastically change your HP requirement.

Referring back to your previous thread - here . The T-S curve gives very general information on starting torque as a percentage of full-load torque vs. percentage of full speed. It does not give absolute figures. It is not a particularly useful curve for this type of centrif, is rarely used or seen, and is not usually required for pump/motor selection. The pump performance curve is far more useful and important as it shows flow, pressure, absorbed power, so you can see what the pump will do and what motor is required.

Referring to your original question, no, the T-S curve will not change if you increase the duty flowrate. You will notice that your T-S curve covers three different pump sizes, and by implication, any impeller diameter and any duty speed for these pumps, and therefore any duty flowrate. (The power absorbed at duty point will increase - refer to the pump performance curve).

Different size impellars are sometimes included on pump curves, not all the times.

And depending on a situation.....usually unknowns, fault out due to over current.

#5 answered your question!

If you increase the flow by any means: Speed, Impeller diam. or less friction losses/lower discharge head, you will increase the power absorbed by the motor and delivered to the pump shaft. this could lead to overcurrent tripping if the motor becomes overloaded ...

Also, increasing the flow ... increases the torque required in general.

thanks, mine was more a statement than a question

Serious process pumps are usually designed for a range of impeller trims, and published curves will always show these. We can see from the OPs T-S curve that the pump in question is a KSB model RPK, probably size 40-200. He should be using the relevant pump performance curve when considering changes to the duty, motor sizing, etc. Some users size their motors for End of Curve, so they will not overload (unless the fluid is changed).

Thanks Holzfeller, I ran into you before ......think it was pump issues also:

I normally worked with G&H before they merge with Triclover to become Alfa Laval. My early experiences comes from bad experiences with sanitary pumps. Some call it practical....sure didn't seem practical at the time.

Really can't see why you are so concerned about the torque-speed curve. Practically all 3 phase induction motors have sufficient speed-torque characteristics to start and accelerate centrifugal pumps up to the operating point. The flow is constantly increasing as the pump accelerates as is the head being generated and as is the BHP or KW draw. If the increase in flow didn't require more "hydraulic?" power then something has changed within your system like removing a heat exchanger and its associated pressure drop. When the pressure drop of the system is reduced the flow will increase out to a new operating point. The power draw would also increase since you are moving further out on the operating curve. Then you either throttle back to maintain less flow or operate out at that new point with the increased power requirement.

Why are you interested in torque? Presumably this is a fixed-speed pump, with head/flow determined by the pump curve and the system characteristics. The pump supplier should also provide a curve of power vs flow, which will tell you whether your motor has enough power for the increased flow. On the power vs flow curve you could divide the power figures by the (constant) speed to get a curve of torque vs flow, but it isn't normally done, and wouldn't help answer the question.

Presumably this is a fixed-speed pump, with .....

can't assume that, The reasoning why........he may be having problems and is at loss to where to go

But if the speed changes everything changes. It seemed reasonable to think he would have said so if that's the case.

The revised duty of this pump can be visualised as increase in flow and decrease in head achievable on the same curve i.e. no change in impeller dia. The new duty point has a lower efficiency and therefore, the rated power is getting increased from 46.4kW to 76.4kW. We wish to replace existing 75kW motor with a new 90kW motor.

It is the starting torque required by the pump that concerns motor vendors and hence the need for Torque-Speed characteristic of the pump for buying a new motor.

As we know,

Torque (τ) = moment of inertia (I) X angular acceleration (α)

In this case I is not changing as we are running the same impeller, but, i think α should increase as we want more flow from the pump? If α increases, so does the torque and hence a new T-S curve.

OK, that confirms it's a fixed-speed pump.

Spinco in #13 has already answered this. Starting torque is not an issue with centrif pumps. The pump starting torque is close to zero at zero speed, and as the speed increases the available motor torque is higher than required torque up to and including operating speed, and even then the motor is likely to have a margin of power and torque.

It must be a big change in duty point to go from 46.4kW to 76.4kW, but if that's what the curves say.......

The required HP is being increased because you are pumping more fluid at a lower efficiency point (probably well beyond the pump's BEP) even though you are producing less head. You will not be doing any of those things though until you have accelerated up to the rated speed, already past the question of starting torque required. Starting a 75 KW pump-motor is generally always done with a throttled discharge valve and the HP-KW load is minimum since there is little to no flow. The moment of inertia and angular acceleration aren't changing from your original conditions since you state that you have NOT changing the impeller diameter or increasing it's weight so its moment of inertia is not affected. The conditions you state about increased flow at reduced head at a lower efficiency point ONLY occur once you are up and running at the specified speed. Can't imagine why your motor vendors have raised the issue of starting torque.

If α increases, so does the torque and hence a new T-S curve.

No! Did you not read what I wrote earlier in post #5?

Referring to your original question, no, the T-S curve will not change if you increase the duty flowrate. You will notice that your T-S curve covers three different pump sizes, and by implication, any impeller diameter and any duty speed for these pumps, and therefore any duty flowrate. (The power absorbed at duty point will increase - refer to the pump performance curve).

I concur with Spinco and Codemaster, the T-S curve is for starting conditions only which, for electric motors, is never an issue with centrifugal pumps. I don't know why you (or your motor supplier) are so obsessed with the T-S curve. I worked with centrifs for ten years and never once needed one. The pump performance curve, on the other hand, is invaluable.

Just noticed the 2nd part of this post. Angular acceleration has nothing to do with flow, increased or otherwise. When the pump is running normally, at rated speed, angular acceleration is zero.

If you want to know angular acceleration e.g. to estimate time to reach full speed (and know the M of I of the pump, motor rotor, coupling etc), at any point on the speed/torque curve, the net torque to give acceleration is the difference between motor torque and pump torque. But as I and others have said, it's not an issue with the vast majority of centrifs.

<...The increase in flow is causing an increase in hydraulic power...>

As the power required to move the fluid is the increase in head provided by the device multiplied by the volumetric flowrate through it, that observation is not really surprising.