2 pumps in series - different flow rates

[desann]

Hello, I have a general question regarding 2 different pumps in series.

In this situation the head is added and flow remains the same Q = Q pump1 = Q pump2.
Also I read if the capacities of the pumps are different the flow is determided by the smallest pump.

But in practical situation , what about if the 1st pump is delivering less water than the 2nd pump? I found that the 2nd pump will suffer and cavitation will occur, I am right?

Is there a guideline how big this difference must be to get this problem? Or how to calculate it?

Can this be prevented by simply reducing the power of the second pump by the frequency driver to match the flow rate?

Thank you!

 

[LittleInch]

What type of pumps?

For centrifugal pumps you want the first one to be higher capacity. But you need to create a joint pump curve and see where it intersects the system curve.

 

[desann]

yes they are centrifugal pumps. The system should feed another technology which needs 8-10 m3/h flow and 4 bar pressure at its inlet.

First pump PSP 18-22
Second pump Ebara Matrix 10-5


This is the system curve from the first pump to the suction of the second. The Q will be 12 m3/h, which is lower than the maximum flow of 2nd pump.




Since the required pressure at the end is 4 bar, for total performance I add this pressure to the system curve. The total system curve at the end + two pumps in seriers



Edit:

The technology that comes after my setup is high pressure fogging system requiring 8 m3/h flow at 4 bars. Its recomended by its manufacturer to oversize the supply line, being able to pump 10-12 m3/h. However in reality it wont happen because the last pump (fogging system) will determine the flow, which will be always 8 m3/h.

So the first pump will deliver 12 m3/h, but the second pump on it's 100% power needs 15 m3/h of water supply, which won't happen in my case because as said the first pump will only do 12 m3/h.
The fogging system will limit the flow to it's requirement, so only 8 m3/h at pressure 4 bar, this won't change.

So in other words, it's just another resistance to the system, which further limits the flow? Such as in this next graph?


And here comes the tricky thing which is beyond my knowledge. If the flow of the system is restricted by the pump at the very end (fogging system pump), does the pump in the middle (Ebara 10-5) have enough water if it's pumping at it's 100% power?
I would say yes, it has because all the pressure at the end it's just the same as pumping into a high static head. Am I right?

 

[goutam_freelance]

Centrifugal pump is a dynamic machine where there is no definite flow or head.

The combined Q-H curve is to be obtained by adding the heads generated by two pumps at any particular flow as @LittleInch said.

what about if the 1st pump is delivering less water than the 2nd pump?

Both pumps will deliver the same flow, which is determined by the combined Q-H curve and system resistance curve. If this flow is low, then cavitation in both pumps is possible

The Q will be 12 m3/h, which is lower than the maximum flow of 2nd pump.

When the pumps are in series, you do not consider them as suppliers of flow; rather, you consider them additions of energy.
.

 

[pierreick]

Hi,
As you said the flow is dictated by the user (8m3/h), this is what continuity is about: Mass in = Mass out.
Pump in series will generate head (TDH). Plot Head vs Q for pumps in series and System. The intersection is your operating point.
Hydraulic calculation will tell you what you will get from your set up.
Pay attention in particular to the NPSHA of the second pump to mitigate the risk of cavitation.
Get support from mechanical or chemical engineer.
Good luck
Pierre

 

[LittleInch]

So the first pump will deliver 12 m3/h, but the second pump on it's 100% power needs 15 m3/h of water supply, which won't happen in my case because as said the first pump will only do 12 m3/h.

This is where I think you are getting confused. The first pump on its own could deliver 12m3/hr, but its flow when you have liquid is limited by the end user flow.

Note also your pump curves are very steep and most pump curves do not drop off as steeply as that.

If the flow of the system is restricted by the pump at the very end (fogging system pump), does the pump in the middle (Ebara 10-5) have enough water if it's pumping at it's 100% power?
I would say yes, it has because all the pressure at the end it's just the same as pumping into a high static head. Am I right?

The pump will not be pumping at 100% because the flow through the pump is not 100%.

So long as the pump at the beginning is operating within its capacity (about 15-16 m3/hr) then it is all ok.

The way you have drawn it is correct - the combined pump curve will deliver the flow where it intersects the system curve, which appears to be 8 or 10m3/hr depending which graph you use.

Does that help?