Pump Head Calculation - 1 Pump pumping to 2 locations

[jkate]

Hoping for a little help. I have a system that will be using 1 pump that will pump to two branches. In this system, we’re pumping a total of 42gpm of water, 40 gpm flows through 2" SCH80 pipe to a scrubbing tower. There is a tee off of the 2" pipe that flows 2 gpm through 1" SCH80 pipe that goes to a waste tank. See attached sketch. Some information:

SUCTION SIDE
Static head on suction side = 2ft.
5ft of 2" SCH80 pipe
(3) 90elbows, (1) flow thru run tee, (1) ball valve

DISCHARGE SIDE
40GPM BRANCH

Height of discharge point = 17ft
25ft of 2" SCH80 pipe
(5) 90elbow, (4) 45elbow, (2) flow thru run tee, (1) flow thru branch tee, (1) check valve, (1) ball valve
flow passes through a heat exchanger with pressure drop of 2 psi
at discharge point is a spray nozzle that requires 7psi​

2GPM BRANCH

elevation of branch off of 2" pipe from centerline of pump = 4ft
Height of discharge point from centerline of pump = 22ft
322ft of 1" SCH80 pipe
(6) 90elbow, (1) flow thru branch tee​

I'm having a hard time calculating what the TDH for the pump in this system need to be due to the two different flows in two different size pipes,so I can provide this information to a pump supplier.

Any help would be greatly appreciated.

 

[LittleInch]

FIrst just do your calcs based on the 40GPM case.

Work out what the head / pressure is at the junction with the 2GPM line. Assume you're running both at the same time?

Then see whether that head is enough to drive 2GPM through your 1" line. You don't mention any end pressure drops for the 1" line?

If its not enough head for 2GPM then increase the head at the tee location and add that increase to the head required for the pump. Note you will then need to put flow control on the 2" line.

If it's more than enough then you need to put flow control on the 1" line.

Then just make your pump 42GPM @ whatever head you calculate

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

 

[jkate]

LittleInch, thanks for your reply. As far as pressure drops in the 1" line, the only losses are the friction through the pipe and fitting. If I find that there isn't enough head at the 2"/1" junction, how do I increase the head at that tee location?

 

[fel3]

As LittleInch explained, this should be a simple hand calculation. However, it's often just as quick to model a pumping system like this using software and you would be able to handle pump selection as well. For simple systems, I usually use EPANET (which is freeware from the US Environmental Protection Agency). I have set up a prototypical "simple pumping station" model in EPANET that you could modify for your system. There is an accompanying Excel spreadsheet for plotting system curves and pump curves for pump selection. This model is attached to the last message in this thread:

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

You can download EPANET here:

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

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

 

[LittleInch]

You increase the pumping head from the pump!

But what that means is that you will need to control the flow in the 2" line as otherwise it will pump more water down that line until the pump curve meets the system curve. So an extra pressure drop before your HX.

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

 

[jkate]

LittleInch...got it.

 

[jkate]

fel3, thanks for your input. I'll check out EPANET and the thread you mentioned.

 

[georgeverghese]

Agreed, you have to install an FIC-FCV somewhere on the 2inch line to create sufficient backpressure to allow 40gpm to flow into the scrubber. Without the control valve, almost all flow will go to the waste tank. We are assuming flow to the 2 destinations occurs simultaneously.
For the waste tank, what matters is the minimum level in this tank, not the elevation of the discharge point at this tank. What is min. pressure in this tank - atmospheric ? This has to be taken into account also.

 

[MortenA]

Im inclined to say "do your own homework". But my 2 cents:

1) Remember that the outlet pressure from the pump is determined by the pump - and if the flow is supposed to be 42 gpm then find your upstream pressure by this.

2) based on your description its two lines with no (significant) common flow. So calculate your dP for each line (and if this is not homework then checalc.com has some nice simple tools for this. Use equvalent lengths for your fittings

3) One of these two line will have the lowest pressure drop and will thus get too much flow - and as littleinch says will require active flow control. The other line will most likely also.

--- Best regards, Morten Andersen

 

[pierreick]

Hi,
This is not too complicated! I will use excel and solver to find the solutions.
What is important is to well describe your system with the right set of equations.
Consider the examples attached to support your work.
As others suggested you will need control valve + flow meters to adjust the flow in the branches.
For the calculation you will need to adjust the Cv of the valves to match the requirement in flow rates.
Interesting excercise.
Good luck
Pierre

 

[goutam_freelance]

I find that your 1" line has much higher length (322 ft vs 25 ft for 2") and also higher discharge elevation (22 ft vs 17 ft for 2").
It is wasteful to increase pump head corresponding to total flow (42 gpm) only for higher head requirement for 2 gpm flow.

I suggest you design for full 42 GPM flow bigger pump at lower head as required for pressure drop for 40 gpm flow and additional 2 gpm flow booster pump for waste tank.

Engineers, think what we have done to the environment !

https://www.linkedin.com/in/goutam-das-59743b30/

 

[LittleInch]

Just do the numbers first.

Yes it's a lot longer but its velocity is a lot lower.

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

 

[goutam_freelance]

Hi LittleInch,

Yes, you are right! The actual friction drops for 2" and 1" lines need to be calculated. If 1" line has large friction drop compared to 2" pipe, then the scheme proposed by me is relevant.

However, too low a pipe velocity in 1" line(approx. 0.26 m/s here) may result in deposition of dirt inside.

Engineers, think what we have done to the environment !

https://www.linkedin.com/in/goutam-das-59743b30/

 

[Snickster]

I agree with little inch in that you would do a pressure drop calculation first in the 2" line back to the pump. You would then calculate the pressure at the junction of the 2" and 1". Since the 2" line is the critical process line that has the most flow I would insure that you have the correct flow to that segment in any case. Then if the 1" line does not equal a flow of 2 gpm at calcualted junction pressure I would resize that line so that it does - go up to 1 1/4" or 1 1/2". In any case in order to maintain the desired flow splits I believe you would need flow control somewhere, either on the 2" line or the 1" line. The line that flow control is put on is the line that you can't tolerate the flow to decrease from the design flow rate and the other line it does not matter if the flow rate is higher or lower.