Does system curve be different for 1 or 2 pumps? 2

[regina_casimiro]

thread378-458044

I was expecting that system curve did not change in case of multiple pumps. However in thread378-458044, for the same flow, the system curve shows higher head for 2 pumps in comparison to 1 pump.

I should be grateful to get an explanation
Many thanks

 

[EdStainless]

The system curve has to change when you go from one pump to two pumps.
You are putting more energy into moving the fluid so some balance of increased head and flow will be the result.
You can just add your pump curves, but the system curve shifts.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[LittleInch]

Regina,

Think about it. I'm assuming you mean parallel pumps here.

So lets assume your pups have flow to head curve like this (these are just random numbers to illustrate the point)
Flow Head
0 100
50 98
100 95
150 88
200 80
250 70

Now if you have flow at say 200 - one pump the head will be 80, but if this flow is divided into two pumps, the head will 95.

Does that make sense?

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

 

[Snickster]

I think she is talking about the flow versus pressure drop of the piping system which does not change no matter how many pumps (in regards to system curve changing). However the flow versus output pump head curve does change for say two pumps in parallel versus one, since two pumps will flow more than one at the same output head.

 

[Snickster]

If you double the flow you would require approximately 4x the head since pressure drop is proportional to velocity squared.

In any case the system pressure drop curve does not change it just extends out to higher flowrates.

 

[pierreick]

Hi,
The system (Head vs Qv) does not change, only the pump curves if you operate 2 pumps in //.
Pierre

 

[LittleInch]

Ah, misread the OP, but the linked post does not show any data so we can't answer the question without using epanet.

The system curve needs to be posted, but without more data we can't consent on why this apparent anomaly is there.

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

 

[Artisi]

Simply, if you add more flow to a system by adding an additional pump, the head imposed on the pumps will increase and the flow wil decrease.
Basic pump 101 lesson.

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.)

 

[1503-44]

EPANET is probably calculating the pump head that is needed for the increased flow rate caused by doubling the flow. Is it about 25-30% higher?

Found a better pipe flow calculator (that other one was rubbish) at

https://www.h2xengineering.com/pressure-drop-calculator/

Running it doubling flow rates, the exponent value is 1.8178, so required head is almost squared with flow.

Doubling the number of pumps would double the flow, but they will still be producing the same old head, so you won't get double the flow into the pipe system, unless you can increase pump RPM.

But adding the new pump doubles the original flow, which increases system resistance by almost the square. So its not working unless you almost double the head. That will require you increase pump speed by 50%, actually around 40%.

If you double the pump RPM, head will increase to around 4 times previous, so you only need to increase speed by 40% or so just to double it.
H2 = H1/(N1/N2)^2
H2 = 1/(1/1.40)^2
H2 = 1/(0.71)^2 = 2 x H1

That will increse flow again. You will then have almost 3 times the original flow. Probably not what you want. You only set out to double flow. Does your pipe have the capability to carry twice the original pressure at almost 3 times the original velocity?


--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[regina_casimiro]

Hi,
To be more specific: for the same total flow, does the system curve change for 1 or two pumps?
(please check the below data find in

https://www.eng-tips.com/viewthread.cfm?qid=458044

)
Many thanks for your help


Q System Curves for 1 Pump System Curves for 2 Pumps
(gpm) Min (ft)Avg (ft)Max (ft) Min (ft) Avg (ft) Max (ft)
0 80.00 100.00 120.00 80.00 100.00 120.00
200 81.60 101.60 121.60 84.42 104.42 124.42
400 85.94 105.94 125.94 96.13 116.13 136.13
600 92.81 112.81 132.81 114.38 134.38 154.38
800 102.08 122.08 142.08 138.84 158.84 178.84
1 000 113.70 133.70 153.70 169.28 189.28 209.28

 

[1503-44]

Looking at fel's input file.... Adding two pumps does not appear to be the reason for the change in head.

It seems that the system curve is changing because the suction and discharge reservoirs are not always the same, they have a high and a low alternative for each, and maybe both pumps do not run all the time. I can read some of that in fel's zip file. It also seems that pump 2 does not operate all the time.

"During the first and third scenarios, the pump(s) pump from the highGsuction reservoir to the low discharge reservoir (minimum static lift).EDuring the second and fourth scenarios, the pump(s) pump from the lowHsuction reservoir to the high discharge reservoir (maximum static lift)"

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[LittleInch]

Looking at that table it does look quite odd, but not sure what else changes in flow between one pump operation and two pump operation as Mr 44 says.

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

 

[1503-44]

It's unusual as normally with changing reservoirs often have different elevations and there are different static heads where the curves start at flow = 0, but both sets of these curves start with the same initial points. Maybe it's through two different pipelines. That would explain that. But I don't want to dig into somebody else's data files trying to make sense of data for a program that I don't use. Maybe fel will come around and explain.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[Snickster]

I all cases in the tables you provided above, at a given GPM the diffrence between 1 pump and 2 pump operational head is the same between Min., Avg, and Max heads. Since in reality there should be no difference, I believe there is some input in the program that is causing this to happen as for higher flowrates the head required increases in proportion to the increse in flowrates. In addition the increase in head with increasing flowrates is approximately equal to (Q2/Q1)² in accordance with how pressure drop does increase in piping with flow, which is odd that for two pumps versus 1 the increase in pressure drop is (Q2/Q1)² when it should be same. There is something in your program doing this. Do you perhaps have additional small diameter discharge piping with two pumps versus one that is adding this pressure drop? There is a difference between flow path directly at suction and discharge due to interconnecting of pumps for two pumps versus one so this may be causing the difference.

At 200 GPM delta H = 2.82 ft
At 400 GPM delta H = 10.19 ft
At 600 GPM delta H = 21.57 ft
At 800 GPM delta H = 36.76 ft
At 1000 GPM delta H = 55.58 ft

 

[Snickster]

Thinking about it a little longer I see that your tables above do not indicate difference if system curves but indicate difference is pump curves. For one pump it is just the pump curve of one pump. For two pumps it is the combined head versus flow curve of the two pumps in parallel.

If it were really the system curves then the difference in head between one flow and a higher flow would be (Q2/Q1)², which it is not but more closer to a linear relationship.

That is why in each case the increase in head required is equal for a given flow and equal to (Q2/Q1)².