Head loss calculation while entering and exiting a common header 1

[TeaPotPilot]

Hello everyone,
I am trying to calculate the flow rate at the outlets from a common header which receives water from 3 identical pumps through 3 identical risers. I have attached a diagram here for clarification of the case here.
1. The output of pump is 100 cubic meters per hour at 15.54m head.
2. Diameter of pump outlet is 65mm.
3. Diameter of riser is 200mm and length is 3m.
4. Diameter of Common Header is 250 mm.
5. Diameter of branches from header is 200mm.

What I am not being able to figure out is what kind of head loss occurs at the entrances to common header and exits from the common header and how can I calculate the losses?

 

[Artisi]

School project?

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.)

 

[TeaPotPilot]

No, its not a school project. I am new to the field and I came across this problem at my job. I don't know what happens around the header section. Never came across such problem during my undergraduate study.
Any help would be appreciated.
Thank you.

 

[Artisi]

Are the 3 branches in operation together?
Yes / no?
what length of pipe of each branch and is each the same and what is the static head?

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.)

 

[TeaPotPilot]

Yes, the 3 branches as well as the the 3 pumps are in operation together.
All the risers carrying water to the header are 3m long placed vertical. The three branches are to carry water to three chillers. For simplicity, lets say the length of each branches is 1m placed horizontal.

The static head is 3.125m.

 

[LittleInch]

I think you're looking for precision where none exists.

Fist your pumps are not truly identical and the performance looks like a conversion from imperial?

So you won't have the same flow coming in from all three branches. maybe within 95% of each other, but not 100. So if you're trying to calculate flow to two decimal places - the pumps won't give you that.

Second if the flow out of your three pipes equals flow in what is this header doing? Why not pipe direct?

Assuming your outflows can vary a bit(??) then there will be some flow horizontally in the header.

How much losses?

I would simply add on the equivalent of another 1m of your 200mm pipe for the head losses.

But first you need to explain / think about what the purpose of the header actually is.

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

 

[Artisi]

And explain the 100m3h (each pump?) and the 15.54m head.



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.)

 

[TeaPotPilot]

RESPONSE TO @LittleInch:
FIRST OF ALL, THANK YOU FOR THE RESPONSE!
The pump performance is given in the product catalog by the manufacturer in the global itself. And I am trying to calculate the flow rate as accurately as I can, at least theoretically. What I am not being to think through is how would the entrance to and exit from the header resist the flow and how I could calculate the loss.

"Assuming your outflows can vary a bit(??) then there will be some flow horizontally in the header."
Yes, outflows can vary.

"I would simply add on the equivalent of another 1m of your 200mm pipe for the head losses."
1m for each inlet to and outlet from header?

 

[TeaPotPilot]

RESPONSE TO @Artisi
Yes, 101.8 cubic meters per hour at 15.54m head by each pump (GIVEN IN THE PRODUCT CATALOG BY THE PUMP MANUFACTURER).

 

[Artisi]

To establish the flowrate for each pump you will need to attach the pump performance curve, the catalogue data is of no value to establish the operating flow rate.
What is the static head from the supply source to the discharge point plus the head losses thru' the chillers?

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.)

 

[SandCounter]

Assuming the diagram is somewhat representative of the actual system, that is, manifold outlets are directly across from the inlets, and the pumps are all running the same. The out-flows of each will be essentially the same and the fluid between the inlet/outlet sections of pipe would be more or less static. If the pumps are centrifugal, as indicated by the diagram, you will need at least one additional operating point to build a curve. If they are positive displacement pumps and a Newtonian fluid, use Darcy-Weisbach to calculate major losses in the pipe and then flow-through Tee coefficient for the manifold.

 

[LittleInch]

TePotPilot,

what Artisi is saying is correct. The data given by the vendor represents only one point point on a curve. For a centrifugal pump, which I can only assume this is, the flow and head varies for the same speed of the pump depending on what the system pressure / flow curve lokos like.

If, as you imply, *(but haven't told us), the flow is variable due to the varying flow from a chilled water system, then the flow and head of the pumps will change also.

So yes, there will be some losses if you pump from one pump into a different header, but at those sorts of flows and velocities, those losses are very low and completely swamped by the other changes.

You really need to think of the rated duty of the pump as a guide as to the performance of a pump, but it's actual performance is dependent on factors outside of the pumps control and actual flow and head could vary considerable. So find the pump curve.

This is a it like saying the economy of your car is 50MPG at 56 MPH. However the car can also drive at 65 MPG @ 20MPH or 35 MPG@75MPH. It's vary rare to be able to drive at 56MPH constantly on a road.

Same thing here - it is quite rare in a cooling system to be doing the same flow at the same pressure drop 24/7.

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