flow rates calculation

[chemroopa]

Hello Everyone.

i am a newbie here. Could anyone please suggest me how to find flow rates in a pipe network from the pump outlet. This pump pumps acetone at 100 gpm through a 3" pipe. After 100' of 3" pipe, the line splits to 2" and 1". I have calculated friction losses but i dunno how to correlate this with flow rate. Given size, pump discharge rate, friction losses how can I can estimate flow rate? Velocity is also unknown in 3" ,2" and 1" pipline. Your help is greatly appreciated.

 

[dickon17]

How do you know the flow rate will be 100gpm? Flow rate will probably depend upon system pressure loss. How much 1 inch pipe and how much 2 inch pipe? Are these 2 discharge pipes open ended? Is there an operator at the end of each line who can manually control the flow demand?

 

[25362]

Velocities multiplied by cross-sections gives volumetric flow rates.

 

[chemroopa]

I figured out that flow rate from volume change in the tank volume from 3" discharge pipe. Recently, 2" and 1" lines were added to feed acetone to other tanks that are maintained at 7 psi pressure. 1" pipe is 48' feet long while 2' pipe is 39' feet long. We just look at level in the tank and close automated valves on these lines.

25362,
I do not know velocities in those pipe line to figure out flow rates. Also plz suggest me which book would very handy for such calculations

 

[BigInch]

To calculate flowrate in a pipe the calculation generally requires that you know the values of the "primary variables, inlet pressure, outlet pressure, the inside pipe diameter and the length of pipe, Q, Pi, Po, D, L, respectively, along with several other "minor variables", the pipe's roughness, viscosity of the fluid, etc. Leaving the minor variables aside, if you know 4 of the 5 primary variables, you can calculate the unknown variable. You can generalize the equation for a pipe by using dP = (Po-Pi) or the equivalent in terms of head, H, dH = Ho-Hi.

Calculating flows in a network requires similar knowledge of each pipe. For each pipe in the network you must know 4 of the Q, Pi, Po, D, L, or, 3 of the 4 variables, Q, dP, D, L, or Q, dH, D, L. If a pump is included, you must use an equivalent expression for the pump relating the pressure (or head) increase with the flowrate, instead of the pressure (or head) decrease for a pipe.


To calculate flowrates in your network, generally you would have to know the inlet pressure (or head) at the pump inlet.
The pump, where the equation is for dH of the pump relating to Q of the pump, the equation of the pump curve. And a flow loss equation for each pipe in the network, also an equation where dH proportional to Q for each pipe. Therefore you must also know either the outflow of each pipe out of the network, or alternatively the outlet pressure (or head) at each pipe outlet from the network.

Do you have an inlet pressure at the pump suction and do you know the outlet pressures at each pipe's discharge point from the network?

If you have those values, then you can use an iterative procedure to find the flowrate in each pipe, where the pump suction pressure + pump added pressure - pressure loss along the pipes to any outlet must equal the discharge pressure. There are two other equations you must employ for networks. At all junctions, the pressures must be equal at all connecting pipes and the flows into the junction must equal the flow out of the junction.

http://virtualpipeline.spaces.msn.com

 

[25362]

Can you pls tell us how have you calculated friction losses ?

 

[chemroopa]

Thank You Biginch. My pump discharge pressure is 40psig. I can calculate suction pressure from suction head or TDH minus discharge pressure( I guess).

25362,
I got resistance coefficient values for fittings from Crane technical book and density of acetone is 49pcf. With know pipe sizes, elevation I found pressure loss for fittings, strainers, pipes and control valve. Ain't I right?

 

[BigInch]

First problem is you don't know the pump discharge pressure, unless you will control the pump to that pressure, in which case you need an additional control valve equation relating dH to Q.

You normally have the pump differential head vs. pump flowrate given by the pump curve, not its discharge pressure. The typical solution is to set some minimum suction pressure according to the NPSH required by the pump, then add the pumps differential pressure at a given flowrate to the suction pressure. That will equal a discharge pressure.

Then you have a temporary discharge pressure, which you must proceed to check. When the pump is connected to a pipe, you must subtract the pipe Head Loss equation (dH or dP) for that pipe, (the pipe inlet pressure is typically calculated from a known outflow pressure). With those two differential pressures and the known suction and pipe outlet pressures, you can iterate to solve for flow.

If your pump dH for a given flowrate = pressure (or head) drop in the pipeline, then the assumed flowrate was correct. Else, return to above.

http://virtualpipeline.spaces.msn.com

 

[dickon17]

chemroopa, do you have an actual question which needs some help to answer, or are you just anxious for knowledge?

Analyzing flow distribution in a multiple outlet pipeline system is not difficult, just tedious. You take a first guess, run some quick calculations and see if you are close. Then make a correction and recalculate the whole thing, each time getting closer and closer to the answer.

 

[chemroopa]

dickon17, this is an actual question from my plant, Nippon chemicals. We use 1" inch line to wash after completing a certain grade of polymer. I am trying to find out the flow rate through this line when we use 2" pipe simultaneously.

Thanks dickon17 and BigInch for solving my questions.