freeze protection of piping systems

[jav2]

I have a client that would like to provide freeze protection for a cooling tower condenser water system by mainataining a minimum flow rate through the system. Does any one have a readily available reference document/source I can use to calculate the required flow rate.
Thanks
JAV

 

[sterl]

Marley has references on operating a Cooling Tower in freezing weather, but that has nothing to do with the pipes...Which appears to be the topic of address here.

The Steam Tracing or Cable Tracing people have lots of calculation basis for the rate at which they apply heat to assure that staganant water in pipes doesn't freeze...Most are going to incorporate insulation values, but for smaller pipes, the electric folks give a lot of data on bare pipe wattage requirements.

Are you going to short circuit the tower itself and just run domestic water through the pipes and then to drain? If this is to be a closed loop, how are you going to keep the water heated?

 

[walkes]

I would think that for freezing temperature operation a closed system with a 50% propylene (or ethylene depending on the application)glycol solution would be preferable. The pumping power requirements would increase and bypass control valves would be needed to maintain your condenser water temperature through the chiller in an appropriate range.

If the tower is a closed circuit unit not being used in the winter a 30% solution for burst protection could be used. (Fluid remains in the pipe, but no flow requirement)

 

[krb]

For heating coils, ASHRAE recommends a minimum velocity of 3 fps, I am not sure if that applies to cooling towers as well but you could use it as a start to determine a minimum flow in your system.

You can also look at sump heaters in addition to the antifreeze and heat tracing as suggested. It is unclear what your loop is going to be, but I would check with the chiller manufacturer if you are planning on running the water through an offline chiller, there may be issues with low refrigerant temps.

KRB

 

[tombmech]

I have used trickle flows, sometimes combined with a steam heat exchanger, with success. However, these are usually on chilled water supply systems whose volumes made glycol prohibitive.

Walkes has a reasonable suggestion about a glycol-charged drycooler, but in my experience the maintenance on these are worse than the cooling tower. Certain computer room A/C drycoolers (Liebert) are highly reliable, but that is a special case. Anyway, you can't charge an actual cooling tower with glycol - which is why he suggests a "closed system". In a cooling tower, the glycol percentage would constantly increase, and significant blowdown cycles would be needed. This is wasteful and can have serious environmental consequences, too. So, glycol should be restricted to closed systems.

As for the cooling tower, if it's in operation, then the rejected heat is usually more than adequate to prevent freezing. The normal control mode is to bypass directly to the basin, though. Water trickling up top and through the fill can freeze on the edges, tearing up the fill and drift eliminators.

If the tower is not in operation, or operation is sporadic, then the basin is what's in danger of freezing. There are many methods of basin freeze protection. If you are still worried about the pipe in those cases, insulation may be your best bet. A well insulated pipe can provide protection well below freezing for quite awhile. If that is combined with basin heat and some semi-regular operation in winter, that is often all that's needed. You should probably put management systems in place to drain the system if things are shut down over the holidays, though.

These days I try to steer away from heat tracing systems unless it's absolutely necessary. They are reliable when they work, but alarms in case of failure are usually left out.