Main header in primary loop

[hollandhvac]

Having to design the main pipe in the primary loop of a chilled water plant.

I want to connect 5 chillers to the primary loop of a chiller system.

3 units will be connected immediately and 2 units will be future connection.

The connection of the chiller is a 6 inch victaulic connection.

In the case of the 5 units connected, I have a water take off from the main pipe of 535.5 l/s (8588 Gallons/min).

I have taken a max speed through the pipe of 1,5 m/s (295 feet/min).

Which would give me 14 inch pipe and I take the decoupler with the same size.

As I have not done this before, do the number taken and results look reasonable or can I expect problems?

What do I need to think of after this?

Thanks

 

[KiwiMace]

If each chiller is 107 l/s then a 6" pipe is a bit small... 2.2m/s. Might be ok for the chiller connection, but doesn't mean you pump it around in that size pipe.

Your flow calcs are erroneous. In a 14" pipe, you'll be over 5.5m/s. You want a header somewhere around 600mm (24") for 535L/s.

I would check that all five chillers need to be running at once, it would be typical to have at least one in standby on a system this size.

Rough in pipe around 300Pa/m, headers around 50-80Pa/m. Then check velocity.

 

[Drazen]

Yep, something is really wrong, with your data velocity comes over 5 m/s.

can you check flow that you gave once more, it is very, very large, you will not be able to find any reasonable hydronic valve with such sizes.

 

[hollandhvac]

KiwiMace; Your comment is correct. The 6 inch is the chiller connection. The chiller will be immediately connected to the main pipe, so as short as possible. The main header it self will be either a 26 or even 28 inch as you have already indicated.

The intention is that it will be 4 run 1 standby, but I know that at the end they will all be running as the system we have now, all the untis are running.

Drazen, as you may understand yuo are correct as well.

The flows are correct, as I have to supply a lot of buildings with this system and have to be prepared for future as well.

Actaully we need 3 units now and keep 2 spare locations, but I calculate already for all 5 units running.

I hope with this you can help me further.

 

[Drazen]

Well, I can admit that such figures are quite confusing for me.

To have reasonable speed of 1,5 m/s you would need 32" pipe, which hardly belongs to ordinary hydronic installation anymore, you would need special measures to avoid hydraulic hammer.

For 5 degrees C temp. difference, total capacity comes up to 11,2 MW of cooling capacity. I would have to conduct deep study of such special circumstances to give some useful advice, which falls far beyond scope of tip-based discussion.

 

[TD2K]

1.5 m/sec is pretty low for those size of lines. The effect of velocity on pressure drop goes down as the line size goes up.

You may have a velocity spec that you have to meet but I would also be looking at the pressure drop profile through your system and see how much a higher velocity affects that. I would think the savings of smaller pipe would offset a higher pump head.

 

[GMcD]

Google the Trane Multiple Chiller Technical handbook or talk to your local Trane saleman and get a copy. Doing engineering design over the internet without visuals is just going to cost you time and money later on using the trial and error, and opinion method of design.