Seawater Cooling system Network Balance

[Statesman]

Gents,

At present we are working on an older plant with very little PID Information. We have two pumps, each with the following duty (Design Rate 387 m^3/Hr Head:90M of water (8.82 bar))
They are connected in parallel, with only one pump running at a time (second as standby). They discharge to a 10" Header which has the following lines coming off it :-

2 No. 8" Line
3 No. 6" Line
1 No. 4" Line
1 No. 3" Line

After these take-offs, the 10" Line eventually reduces down to 3" and runs as a partial ring main, feeding the plant and eventually being discharged into drains (The majority of plant on this main is now redundant)

Off of one of the 6" Lines comes another 3" Module Header, which is, effectively used to supply a 1/2" Heater feed. (Again,Due to various plant redundancies over the years)This Module Header has, according to Plant Line list the following operating conditions (Op Press 7.7 Bar,Op Temp Amb,Design Press 19 Bar, Design Temp 38°C)
We propose to run another 3" Header from this 3" Module Header, this line will supply a couple of 1" Connections.
Due to the fact that a lot of the branches are now,in effect,redundant , when I come to work out my mass flow rate at my new 3" Header, is it safe to assume that the mass flow of the pump discharge will be split (Proportionally) between all of the branches off the 10" Header, bearing in mind that the information for a full network balance is not readily available, and therefore loading equivalents for these branches isnt available.

Any guidance would be greatly appreciated

 

[rbcoulter]

Proportional to what? Pipe diameter? Flow will distribute itself in a manner favoring the path of least resistance. This is affected by pipe diameter, pipe length, pipe roughness, pipe fittings, valves, etc.
An equivalent K value would need to be calculated for all the branches to determine how the flow will distribute.

 

[ENC]

I agree with rbcoulter that the K value is needed for all the branches for distribution. If it is possible you may want to consider using a portable flow meter that would allow you to establish branch flows in a proportional basis. Another approach would be to measure pressure in each branch, which can be done from a drain or vent valve. Then from the dischage of the pump to the PI, you would have the Delta P of each branch to a specific location. Flow will be proportional to Delta P in the system.

I hope this helps.

 

[Statesman]

Thanks for your comments but, as stated in the original post, we dont have enough information to work out the k values for each of the branches and as the plant is over 300 miles away, it isnt easy to measure the branch flows,or the pressure in the line.
We require this information for a 'first run' sizing of a new heat exchancher that we are installing.
So, if I was to assume an equivalent pipe length in each branch (Apart from the ring main which would be longer) a friction factor being the same for all lines (same material, age etc) would I be wrong in assuming the flow is proportional in each branch (to the diameter?)

Or, going at the proplem from another angle :-

Is there any way that I can deduce the mass flow from the operating conditions in the line I am tie-ing into ?
i.e. 7.7 Bar and Seawater running within a 3" pipe ?
I would have thought that knowing the pressure, size of line and the density of the fluid, I would be able to deduce the velocity of the fluid in the pipe (And hence work out my mass flow rate from that)?
To at least give me a ballpark figure for my available Seawater cooling rate ?

 

[ENC]

If you reference "CAMERON HYDRAULIC DATA" the flow through orifices, nozzles and venturi tubes is provided. In your case you will have to assume the value H = difference in head between upstream end and throat in feet. This should give a good approximation, from the max to the min.

From flow data - Nozzles a 3" diam wiht 110 psi pressure will provide 2820 US Gallons per Minute. NFPA has a pocket guide with the same flow information in it.

Good Luck.... If you need furthere assistance email me at fjstanton@cs.com

 

[rbcoulter]

If we assume that each branch is about the same length, pipe roughness, have the same pressure drop from pump discharge to end of pipe (is it atmospheric discharge?) and other restrictions in the pipe, valves, etc. are negligible, then Darcy's equation would say that the 5th power of the inside diameter is proportional to the 2nd power of the flow rate. So flow can be proportioned based on the inside diameter raised to the 2.5 power.

See the Crane Handbook, page 3-2.