Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations IDS on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Centrifugal pump duty point prediction

Status
Not open for further replies.

fd74

Chemical
Jul 17, 2003
10
A centrifugal pump was selected for a certain application for an expexted duty point of 30 m3/h @ 20 m (for those of you not familiar with metric units, don't bother because numbers are not so relevant for the final question).

Owing to modifications to the piping during construction, pump is now running at 21 m3/h @ 22 m.

I've been asked to evaluate a mean to increase capacity to a value near 30 m3/h (say 28, in any case it's a non-critical transfer operation so capacity would only affect total time required for the operation), without replacing the pump with a new one of course.

Let's say we substitute the impeller with a slightly bigger one (operating curves for various impeller dimensions are available from the pump menufacturer).

How can I predict the new operating point and choose the right impeller?

I mean, is there a way other than the theorethical, to evaluate circuit H-Q curve?

Thank you all in advance for your replies!

fd74
 
Replies continue below

Recommended for you

For geometrically similar impellers, you can very well use affinity laws.

Q1/Q2 = D1/D2, H1/H2 = (D1/D2)2 and
P1/P2 = (D1/D2)3

Your new impeller will be (30/21) = 1.43times the old impeller and if your numericals are true, I doubt whether can get such a huge margin in your casing. The new differential head will be 45m approx. Now, calculate resistance with 30m3/hr flowrate and draw the system curve on the new pump curve. The intersection will give you the new operating point.

This is all easier said than done. So, you should do some hardwork. Come back if you require further help.

PS: Check your existing prime mover for the new power requirement.

Regards,


 
First of all thank you quark.

I'm aware of the existence of affinity laws, which can be used, being known one operating curve, to estimate pump behaviour in different conditions (speed, impeller diameter, and s.o.).

We are lucky enough to already have those curves from the manufacturer.

As you said the hardwork is evaluating the system curve!

Is there an empirical way to determine this curve?

Regards.
 
short of using affinity laws, call the pump mfg and provide them the new process conditions. they will gladly generate new performance curves for you. provided the existing pump has the capability of installing larger diameter impeller.

good luck!
-pmover
 

A quick assessment for a flow increased from 21 to 30 m3/h, assuming an original discharge head (not differential head) of 20 m totally devoted to friction drop, ie, no elevation components, would show a new discharge head (= system resistance) of (30/21)2 [×] 20 ~ 41 m. If there is an elevation difference, it should be taken out of the 20 m, and added again after estimation of the new friction drop.
 
fd74,

Emperical formula used for system curve is Darcy's equation, as you might knew it already. Your question is not very clear. Do you mean about the procedure to draw a system curve? If so, let me know and I will write down the procedure.

Regards,


 
It is very easy to plot the system curve on the pump curve using the method outlined by 25362. For these purposes a simple calc like this is usually all you need. Just remember to separate the effects of static height and friction.

If you get to 30m3/h @ 41 m head you will be using double the pumps original rated power. Sometimes this can have a huge knock-on effect because you have to upgrade switchgear and cables as well. In this case it is a small pump and existing electric may be OK. But it needs checking.

You should go back to the original design and see what has changed. Do the construction mods account for all the change? Is there more static head than expected? Is the line longer than expected? Are there additional valves and fittings? Are the liquid's physical properties very different from expected? Remember that friction head varies with the 5th power of the pipe diameter, so changing the pipe size could dramatically reduce your pressure drop if it is not due to static head.

It could be a construction error. We have all seen wrong gaskets installed or other debris left in the line. Check all this before you make major mods.


Good luck - let us know when you find the answer.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor