Cooling Tower Discharge Pipe Sizing

[BronYrAur]

Is there a design guide or rule of thumb for sizing the line coming from a tower sump to the pump? In my specific example, I have a small tower that will need 180 GPM. My two choices for pump location are either directly under the tower or a more preferred location in the building. If I put the pump in the building, I will have a 4' drop from the tower at which point I will enter the building basement and run approximately 50' horizontally (with only room for a very slight pitch). At this point, I can drop down another 6' to reach my pump suction.

For 180 GPM, my pressurized piping will be 4", but what about this gravity return line back to the pump? I'm sure it should be at least a 5" but what is the proper method to size? Is there a friction factor that I should try to hit? If I did go 5", the friction loss would be around 0.6' per 100' of pipe.

 

[berkshire]

BronYrAur (Mechanical)
You also have this posted in local ventilation. Maybe you should kill that one, and try Pipelines, Piping & Fluid Mechanics Engineering.
B.E.

The good engineer does not need to memorize every formula; he just needs to know where he can find them when he needs them. Old professor

 

[MintJulep]

Size so that the NPSH head at the pump suction is what the pump needs.

In other words, sufficiently large so that the pump can't suck water out of the bottom of the pipe faster than it can enter by gravity at the top.

 

[cdxx139]

You have to size the NPSH as stated above. This is a problem that comes up more than it needs to. If you have to much pressure drop in the pipe from the CT to the Pump, then increase the pipe size.

knowledge is power

 

[BronYrAur]

I understand the NPSH requirement, but the elevation and atmospheric pressure alone are going to give me enough NPSH, provided that I don't have too much loss in the pipe.

I won't since I am up-sizing the pipe more than I would for a pressurized system. I just don't know how to do that properly. It there a friction factor or velocity that I should shoot for on a more-or-less gravity drain system?

 

[MintJulep]

If I've sketched your system properly in my head you have:

10 feet of head due to gravity.

60 feet of piping at 0.6 feet/100 feet loss plus a hand full of elbows.

So about 9 feet head at the pump suction. If the pump is happy with that then you're good.

 

[TD2K]

Rule of thumb is to keep piping suction velocities below 3 ft/sec for a bubble point liquid. Now, this isn't a bubble point liquid so you can go up from there.

Using the pressure drop tables in Crane, with a 4" you are looking at about 1.8' head loss/100', less than the elevation change you are looking at. Velocity in a 4" line is about 4.5 ft/sec. Also, since you say this is off a cooling tower in your title, you have 30' + of NPSHA just from atmospheric pressure before you start adding elevation changes and subtracting line losses.

I'd likely pick a 4" line. If you think that you might need more water in the future, run a 6" but it's not needed for 180 gpm.

 

[BronYrAur]

Thanks all. I'm running 4" on my pressurized piping, but I'm going to bump it up to 5" on the return from the tower. That keeps me under 3 fps.

 

[TD2K]

I know 5" pipe is listed on the various piping tables but I've never actually seen 5" pipe used at least in the services I work with. It's like 2 1/2" piping or class 400 flanges.

No reason you can't use 5" pipe, just wonder what fitting availability is like.

 

[BronYrAur]

5" is not a common as 4" or 6" for sure, but it is readily available. Same with 2-1/2".

It's 3-1/2" that you see in the tables but no one stocks.