Cooling tower bypass

[OCHVAC714]

FYI I do not have too much experience doing cooling tower designs. I'm looking to replace an existing cooling tower (250 tons) and add a bypass valve for the condensing water.

What should the bypass valve and pipe be sized for? (full flow of the cooling tower, minimum flow of the chiller, or ??)

When going into bypass mode how should the pumping problems be addressed when changes in head occur (open loop to closed loop)? The condensing pumps have VFDs and the new cooling tower will also have a VFD.

[nique]

if it were me I would set the condenser pump vfd (started by chiller) for proper gpm thru the chiller, let the chiller (if capable) control the tower vfd and forget the condenser bypass. just my opinion but that is how I set up all the systems I do.

[emcontrols]

Full flow, same pipe size as tower supply and return and oh yea bypass to sump if possible to prevent cond pump cavitation

[ckartson]

I'm not sure I understand, If you have vfd's why do you need a bypass? If the tower fans are on vfd's that sould be good enough No?

[motoguy128]

I thought you wanted to maintian full scondenser flow unless you were at a part load condition on the chiller.

I've seen simple PID controllers connected to a 3 way valve to control condenser loop temps. One chiller we have uses river water for cooling so a 3 way valve is a must have 6 months out of a year just to keep water temps even in the 60's.

[OCHVAC714]

The building is a commercial building in Southern California. They are looking to add the bypass to address issues with the chiller that they are having at low ambient conditions and early startup.

The chiller, cooling tower and condenser water pumps are all located on the roof. So I'm thinking of specifying a diverting 3-way valve at the cooling tower header and having a full flow sized pipe (8") to tap into the return (pump suction side). This is what they have expressed that they wanted.

emcontrols - I see what you mean by piping to the sump to keep the same conditions, but this might not work well in this situations.

What else can be done to address the changes in head when in bypass mode and is this really going to be a problem? I've read that placing the 3-way valve at the pump suction (mixing valve) that this can create problems on the pump suction and create vacuum conditions. But what about in my situation with a 3-way diverting valve prior to the cooling tower inlet. How do I calculate the change in head in bypass mode?

[emcontrols]

What typically happens is when the bypass is full flow to the tower the bypass line drains out and becomes a great place to store air(dont know why but it happens all the time and I have repiped many many bypass lines to sumps to solve problem, If pumps are above the bypass line the problem in my experience doesnt happen), so when it goes to bypass it sends the air to the pump and bingo cavitation. so if you dump it to the sump the air disipates through the water in the sump and the condenser pump always has water as long as the sump stays at proper levels. This only applies to a tower that the water is open to atmosphere. In addition you also get better mixing with the return cond water which makes for a more consistant water temp for control purposes.

Cond pump VFD? - Is it to be controlled or for balancing only (IE set cond VFD pump at required GPM at leave it alone)? Need to know that before we can thow out some control sequence options!

Chiller? - What type of chiller 1.Screw, Recip, Centrifugal? Reason being each type of compressor manufacture will have a factory recommended head pressure to maintain for peak performance and there are a few choice methods to control, each unique to the compressor style some can have slow stroke time on reaction to temp change others need extremely fast stroke time. (example: tower bypass valve 3-way, cond leaving 2-way with minumum, cond pump VFD)(controlled by chiller head pressure transducer, field added head pressure transducer, temperature sensor CW entering, temp sensor CW entering & wet bulb-Which I personally dont like, and CW sump temp - usually low ambient applications with remote pit(sump) and probablly a few more.

After you answer we can pick a good control sequence.

Is there a problem with chiller head on low ambient and is that the reason for the new tower and control sequence? Just curious

[Six]

I have a tower with by pass, VFD on pump and VFD on tiwer fan. SP for by pass is 75 degrees closed at 85.

Its for low ambienr start up.

Bypass is same size as supply and return and pipes to sump with pump below tower. Ive nevwr had issues with air.

Tower VFD its it 2 to10 input from a Honeywell controler and pump runs at 60hz.

Ive never a an issue with this set up with the exeption of the ocvasional drive fault.

Tower fan is not enough to get water temp up fast enough.

[dlove]

We have lost a lot of our overnight heat load in our building over the last couple years. We are going to be constantly dumping heat into the condenser loop via the boiler/heat exchanger set up this winter it looks like. The set point that has been there for years has be 60dF for when to start adding heat to the loop.

Anyway.. There bypass valve in the building before the pipes go underground out to the tower, there is no control over this as it is just manual valve. I imagine opening it up partially would be a good idea to reduce cost on the boiler dumping it's heat into the loop? I know that if we get warm days during the winter it would need to be closed again but for the most part i think it would work. Any advice on this?

Also, the pumps and fans are on VFD's, although the VFD pumps are just mainly used as a soft start. I could add a output signal to the VFD and just drop the speed of the VFD at night, this would achieve the same results correct? On a closed heat pump loop i have always thought and read that it is best to just set it and forget it.

Thanks

you just need to be careful you have enough flow and heat out to the tower to keep the bundles from freezing. Personally, I wouldn't do it.

[Gnr027]

I live in Canada and the building I take care of the cooling towers are shut down in the winter and chiller isolated to go threw the boilers cooling towers are drained and cleaned at that time

[dlove]

I have all the inlet and outlet temp sensors out at the towers alarmed so I may just try it for a day or two or more and keep a close eye on it. I am talking opening it maybe a 1/8 to start, maybe a 1/4 open to see the effects on the loop temp.

As long as you have flow and water temp above 55 you should fine. But i would curious how how you will save. let us know if delta T spread changes. I'm thinking the slower water higher spread.