performance curve

[nileng]

Hi

I want to modify a pump charactristic curve which is gained from the experimental test.the head at shutoff is higher than the desired curve (required curve)but at bep and over bep flows the head is smaller.and the curve is steep. what shall I do in order to increase the head at BEP and at overload. and also I want to increase of efficiency?

 

[TenPenny]

You'll want to design a new impeller, because you want to change the shape of the curve.

 

[Artisi]

Can we assume the blue curve is the published curve "standard curve"

Is the impeller full diameter?
What impeller style is it, enclosed centrifugal, open type, semi open etc?

There are a few tricks to gain a little increase in flow and maybe in efficiency.

Can you post a picture of the impeller looking at the inlet and the discharge areas.

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.)

 

[nileng]

@Artisi

The blue curve is the standard curve and I want the red curve reach to the blue curve. the red curve is achieved from the test of the designed impeller. I want the head increse in BEP and overload flows.and the efficiency increase around the BEP.

The impeller is in full diameter and it is closed radial impeller.

I attached the meridional surface of the impeller.

 

[bimr]

It would be helpful to understand the details of the pump such as pump size, materials, age of pump, etc. With the information, you would obtain better advice.

 

[1gibson]

Is it underfiled?

 

[LittleInch]

The head of this pump is very low compared to the flow. Any minor error in the pressure readings or even if the guage is simply not at the same level as the desired curve - measured from the pump centerline?? could cause this error - i.e. if the guage is simply 1m higher then the reading will be 1m adrift.

You need to look at the specification for the pump - it could be that it is within its acceptable range.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[nileng]

@ tennypeny
Thanks for your response,by changing which parameters of the existing impeller can make the charactristic curves better ( near to desired curve)?

 

[Artisi]

What LittleInch has said is very valid, how accurate are the gauges, are they positioned correctly with appropriate height corrections allowed etc.

Think you need to expand your story a bit further, are you testing a new designed impeller against a theoretical curve or are you trying to reproduce a curve against an existing test curve. Bear in mind it is nearly impossible to test a pump to the accuracy of what came be achieved in a full-blown test facility. Just reviewing the 2 curves you have supplied and without going into too much detail, it is more than likely the pump performance you have tested would meet a test code spec.

The way to improve the performance a little is to clean-up and dress all the wetted surfaces to a smooth finish, polish where you can, check the leading edge of the blades for correct profile and smoothness, under-file the discharge edge of the impeller blades.

This can give a reasonable improvement to performance.

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.)

 

[nileng]

@bimr
Pump size: discharge: 100 mm,impeller diameter:225mm , the hydraulic of the pump is designed recently and the pump is new, the material of the casing is GG25, the impeller is made by 3D printer and its material is teflon (Abs)in order to test it.

 

[nileng]

@1gibson
No it is not underfilled

 

[nileng]

@littleinch
Thankyou, do you mean that I check the gauges of the tested pump to be same level with the desired curve gauges ?

 

[Artisi]

The gauges don't have to be at the same level, must must be corrected to the same level which should be referenced to the impeller centreline.

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.)

 

[Artisi]

If this is a recently designed pump, please explain the 2 curves, is one theoretical and the other actual, or are you copying an existing pump/ design?

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.)

 

[nileng]

@Artisi
Thanks for your responses, the blue curve relates to an actual pump of one of the pump manufacturing companies ( the blue curve is taken from the catalogue of that company) but the red curve relates to the pump that the hydraulic of it ( impeller and casing) is designed from the base by cfd and pump design softwares, it has been designed for the Bep of the blue curve. It means that the design point of the two pump is the same but the hydraulics are not same the red curve is my own design by using theories and software the blue curve is the charactristic curve of one of the pump manufactures. Now the aim is that to adabt the red curve to the blue one. To hava a good compatibility

 

[nileng]

The red curve is the test (experimental) results of my own design.

 

[Artisi]

If you are pirating pump design - sorry, no further assist will be forthcoming.

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.)

 

[nileng]

@artisi
I am not piroting, the company itself has given us the actual pump and its documents, and I am working on my own new design. Thats not piroting.

 

[Artisi]

Ok.
So you haven't physically tested the original pump, therefore you don't know the validity of its performance curve - you can only assume that your impeller / casing is correct, testing is not accurate, or the published curve is incorrect or a combination of all.

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.)

 

[Artisi]

Just to add, if the impeller is to design and you have a real shortfall, it is beyond the scope of this forum as you will need to get back to basic impeller design or employ a qualified hydraulic engineer to assist.

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.)