Cooling Tower Return to Building / Pump Suction Question

[RobsVette]

Hey guys,

I am an engineer in the NYC area working in the HVAC business and have seen a few things lately that have challenged my understanding of how to pipe condenser water systems properly.

Many of the chiller plants we have worked on lately have the condenser water pumps sucking through the chillers, rather than pushing through.

I would imagine the arguement for this is that if you ever dead head the system you are adding the dead head of the pump to the effective pressure the chiller / cooling tower will see.

I always thought that the reason you are supposed to come from the cooling tower straight into the pump is to prevent air bubbles from forming that can damage the pumps if ingested. This becomes a problem as you get into systems with multiple chillers and pumps, were the starting of a second pump can cause an air bubble to be sucked in due to the sudden increase in velocity. This is of course assuming that you dont have a VSD on the pump or other slow start device. I would imagine that the coming down from the tower, into a chiller inlet header, going down into the chiller, then going up into a chiller outlet/pump suction header is not a good idea because you can collect air bubbles in the pump suction header and starve the pump of water.

I understand not wanting to subject the chiller to excess pressure if something goes wrong and saving money by not getting connections rated for this pressure, but to me this just seems like a bad idea. Am I crazy? or do I need to get with the 21st century and this is how it is done nowadays?

 

[DLite30]

One possible reason is this. All pumps will had some amout of thermal energy to the fluid depending on their efficiency. So if the pump was upstream of the chiller, all of the chill water would be prematurely heated prior to reacher the heat exchanger.

 

[pipesnpumps]

First off any connection on a chiller is going to be rated for more than the dead head of a condenser pump... You are talking about less than 80 ft of head for a properly designed system.

I don't see any advantage to chiller first arrangement, and in fact a big negative is air. Not from bubbles traveling down from the cooling tower, but from air going into the system at flanges or other leak points that are under vacuum and suck air in.

On the chilled water side of the chiller it's a closed system, so placement is not that important.. But it is still best to push through the chiller in that case, because the air separator on the suction of the pump is at the warmest spot in the system (the return) where the dissolved air breaks out...

But I have yet to see that arrangement with condenser water pumps and I deal with them all the time. Very odd. pump heat is nothing compared to fan heat so I don't think that has any bearing. Not sure why someone would arrange it that way.

Real world knowledge doesn't fall out of the sky on a parachute, but rather is gained in small increments during moments of panic or curiosity.