Determining required head on existing water system where we can't measure equivalent length/friction

[Richard C]

New member here, if I'm posting in the wrong area please forgive my ignorance. Also, I'm an HVAC Contractor and not an engineer who is lost in my effort to obtain a formula or chart to help determine the heads of existing water circuits buried in the ground.

We have an existing older building where they buried 3/4" copper lines for recirculating potable hot water to obtain hot water availability at their facets / showers. Needless to say this facility hasn't any detailed prints for determining equivalent feet of piping. This system is similar to a hydronic heater closed system.
Given.
150 gallon hot water tank
Five (5) 3/4" copper buried circuits to bring back water from the loop to the tank
The five circuits are to be connected to the return of the pump(s)
We flushed each circuit and measured GPM at outlet of each circuit .
GPM test performed using 50 psi city water pressure.
Circuits:
1- 2.8gpm
2 4 gpm
3 5 gpm
4 8.6 gpm
5 10 gpm

Unfortunately have no other information besides GPM at the applied 50 psi pressure and don't know how to obtain anything else.
Is there a formula or chart or other tests we can use to determine the existing heads? Even if we were to install five individual pumps we have no idea what the required head is for each circuit. I have tried to obtain idea's from the manufactures but I haven't had success in reaching someone the typical customer service reps. Thank you.
Sincerely
Rich Close

 

[georgeverghese]

The pressure drop correlation for any piping system is

Delta p = (4fL/D). (rho.V^2)/2 . Presume you know what these terms mean

For any system, the first term is fixed (more or less) within a wide flow range. Use this to get the first term, and then you can derive the head required at the pump for any other flow rate.

Presume you've done a hydrotest on each of these circuits to confirm there are no significant leaks before the flow test.

 

[rotw]

Can we make a parabolic fit (Head vs. Flow) between two points, the one at zero flow and head (assuming only friction losses contributed to total head) and the point of measurement taken, then extrapolate a system curve from where the head required for the desired flowrate can be obtained?

 

[Artisi]

Was each cicuit tested separately or were they tested as a group, and what flow rate do you require on each circuit?
What is the vertical height of the highest point of each circuit?

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.)

 

[LittleInch]

Assuming each circuit was in parallel and you applied 50 psi to each one and the measured the flow.

The basic "rule" you can use which works within 25% of the measured value is that frictional pressure drop is proportional to the square of the fluid flow rate if you don't change anything else (diameter, length etc). This is what George is saying but in a more basic way.

So in your case in e.g. circuit 2 if you wanted 8 GPM you would need to apply ((8/4)^2) x 50 psi to the circuit ( so 200 psi).

Similarly if you only needed say 3 GPM then the pressure would only need to be (3/4)^2 x 50 psi = 28 psi.

As I said it more or less works within a factor of 4 so in circuit 2 if you wanted 20 GPM the pressure might be a lot higher than the proportion rule or if you only wanted 0.5 GPM it could be a lot less than the proportion^2.

Note that this assumes tat the inlet and outlet of your system was at the same elevation and that your system is a closed system under pressure.

Is that what you were looking for?



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