HOW TO CONSIDER THE HEAD-FLOW CURVE OF A PUMP, IF USING DIFFERENT PIPES 2

[CHK07]

thread194-343587
I'm working in a swimming pool construction company as a mechanical engineer.

Actually I'm not clear about how the head-flow curve mentioned for a pump.. The pump is coming with a standard connection of 2" or 3" likewise. so I think the head loss curve only means for the connection pipes ( ie, for 3" connection pumps, the head loss curve is for 3" pipe only). if we are using another pipes, we can't consider the head-flow curve ( ie, for example if using 2" pipes for 3" connection pipes).

Normally I'm using different sized pipes in a system, but how can I consider the head-flow curve for sizing a pump?

 

[LittleInch]

CHK07,,
The pump curve as it's known is valid for the pump and should plot differential head across the pump vs flowrate through the pump.

For any particular system, the actual flow you get will vary depending on the size and length of your pipes, any head difference and any head loss across e.g. a filter.

So ideally you need to work out a "system curve". This is the head required at the pump location to get a certain flow.

So assuming there is no elevation change anywhere, at 0 flow there is 0 head loss. Then for the inlet pie and outlet pipe calcualte the head loss (friction) as you increase flow so say at 20,40,60,80 100% of max flow.

This generates a head required vs flow graph.

where the pump curve intersects the system curve, this is the flow you will achieve.

Note: there is a massive difference in head loss per unit length between a 3" pipe and a 2" pipe. For same flow a 3" head loss is about 25% of a 2" pipe head loss.



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

 

[fel3]

Also pay attention to minor losses and actual vs nominal diameters. Many pumping systems (probably including this one) have short pipes with lots of bends and fittings. In these cases, minor head losses can be significantly more than pipe head losses, whereas in water distribution systems it's the other way around. Many pipe types have significant differences between actual and nominal diameter (e.g. each of the different flavors of PVC pipe). If this is the case, and if you have long enough piping runs to worry about, then I suggest using the actual diameters instead of the nominal diameters. The differences in cross-sectional area can be surprising.

============
"Is it the only lesson of history that mankind is unteachable?"
--Winston S. Churchill

 

[CHK07]

LittleInch , Fel3...

Thanks for your valuable responses....
I got your point, that pump curve doesn't depend on the piping system whatever we are doing. we just want to find the head loss occur in our piping system at a required flow rate and find the pump that produce that much energy by looking the pump curve.

Actually here I have 2 self priming pumps of same brand but 2 models. I'm attaching the pump curve of those models. here if we look, the 3HP pump provides a head of maximum 22M ( Euroswim 300). but the other model which is 5.5HP (KPA-550) providing only maximum of 17M head. Obviously, the pump having 5.5HP should provide more head than 3HP. Only the difference is, the KPA model have 3" connection and Euroswim model have 2" connection. That's why I got that doubt, maybe the pump curve depends on the connection pipes. here if we are looking for a wide range of pumps, the pump having 3" connections are providing low heads comparing with pumps having 2" connection.

 

[fel3]

CHK07…

The maximum heads for each pump don't actually matter for your question. What you need to do is develop the system curves that each pump will "see" so that you can determine where on their pump curves they will operate. If the pumps are plumbed in parallel, then you need at least three system curves: Euroswim 300 alone, KPA-550 alone, and both pumps operating together (if the pumps and their piping were identical, then you would only need two system curves because one pump alone would be the same as the other pump alone). If your inlet head and/or outlet head can vary more than just a little bit, then you need six system curves. The first three would be ones I listed above and assuming minimum suction head to maximum discharge head; the second three would be similar, but assuming maximum suction head to minimum discharge head.

All this is easily accomplished using water system modeling software. For freeware, I suggest EPANET (see

https://www.epa.gov/water-research/epanet),

which just happens to be the calculating engine behind some of the commercial programs. Here is a thread where I explain how to model a system like this to develop system curves:

https://www.eng-tips.com/viewthread.cfm?qid=420577

At some time in distant past, I may have uploaded a couple files to help create a simple pumping station model in EPANET. However, I didn't find the thread. Instead, I updated my files and attached them here.

Finally, here are two more threads that might be useful to you:

https://www.eng-tips.com/viewthread.cfm?qid=325862

https://www.eng-tips.com/viewthread.cfm?qid=320430

============
"Is it the only lesson of history that mankind is unteachable?"
--Winston S. Churchill

 

[FacEngrPE]

You can not assume the pump with a higher power motor has a higher discharge head. This really depends on the individual pump curve.

The Pumps - Affinity Laws provide generalizations that are useful for estimating the relative importance of the various aspects of pump selection.

 

[bimr]

The head loss shown by the pump is the resistance to flow from the piping. Pumps don't make head loss, they create flow.

For a small system use one of the online head loss calculators to calculate the headloss in the piping. Here is an example:

Add another 15% headloss for the miscellaneous fittings such as elbows and 10 psi for the swimming pool filter headloss.

Then, ask a pump supplier to supply a pump for the conditions (flow and headloss).

 

[LittleInch]

Chk07,

As others have pointed out, power of a pump is not always linked to head, but all lithe things being equal (especially efficiency) power is protons to head times flow.

Pump vendors will know their own market and it is reasonable that a 3" network will have lower head than a 2"network.

So the euroswim 300 is aimed at a flow of around 25 to 35 m3/hr at about 15m head, the 550 aimed at 60 to 70 m3/ hr at about 12m head.

This equates well with having a 2" or 3" system respectively.

These are unlikely to be self priming but need a flooded suction.

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