Hi,
If you use cooling water it's like you are draining your system from the chemicals used to protect your equipment ( anti corrosion anti scaling agents) .
Think twice before proceeding.
Pierre

Hmm, but can't those simply be be topped up? We already add those in regularly so I suppose all we'd have to do is increase the frequency.

Effects:
1/ Heat transfer rate
Heat transfer coefficient for steam condensing is ~100 times higher than that for water cooling.
2/ Mean temperature difference
Steam has much more temperature difference than water and therefore heat transfer force.
3/ Reynolds
Reverse cooling water has low pressure and can not provide high pressure drop across heat recipient. High pressure drop is required for turbulence and high heat transfer coefficient.
4/ Thermal resistance
Cooling water is much more dirty than steam and therefore cooling water side of heat exchanger has a layer of scale with high thermal resistance.

General equation
Q=U*A*ΔT where
Q - heat transfered, W
U - overall heat transfer coefficient, W/m^2*°C
A - heat transfer area, m^2
ΔT - mean temperature difference, °C

In case of cooling water:
Q is not changed as defined by recipient requirements
U is much lower because of (1), (3) and (4)
ΔT is much lower because of (2)
It means that A must be much much higher to compensate heat transfer inefficiency. It means that reverse cooling water can be used for heating of recipients but it will require much much more expensive heat exchangers than those heating by steam.

This is is a noble aim and in theory should be feasible.

The issues tend to be based around what happens when different scenarios occur in terms of max and min flows for the different elements and how that impacts the performance.

So e.g. the cooler aims to reduce temperature by a certain amount. Does your heating requirement end up with the same temperatures? Does it cool the water enough to be the same as the towers?

Is the inlet temp to the towers high enough for your heating load or does it need more heat which you need to input?

What happens if one section doesn't take water, e.g. if the unit creating heat isn't working, does the other system still need heat? So you end up with a dual system?

Can you use the cooling water direct or do you need to segregate it and need a water/water HX. E.g. you probably can't use cooling water for hot water for personal use (showers, wash basins etc), but if you're just pushing it through a closed radiator system then you should be ok.

The devil is in the detail of temperatures, flows, distance, how much turndown can you towers with stand, how to control it etc.

A good project for sure, but it cannot just be assumed without some engineering.


Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

Hi,
Seems to me you have little experience about cooling towers operations and cost associated with wrong doing ( direct and indirect costs) .
As I said you need to be extremely careful with this kind of ideas which can killed your entire operation .
Very little to do with heat transfer equation , much more on the chemistry of the water and the consequences to have a system unbalanced .
At the end of the day it's your decision .
Pierre .

Thanks for your advice. I'll look into what you guys brought up.

I remember seeing something like this at a large milk processing plant. There was demand for both heat and cooling, at relatively mild temperatures, and someone had worked out that the duty of each was about the same, but only when averaged over a few hours.

So they installed a huge tank that had hot inlets and outlets up top, and cold inlets and outlets at the bottom. And presumably some internals to stop convection currents from mixing everything up. So if something needed heating, they would draw from the top of the tank, and return to the bottom. And for mild cooling duties, would draw from the bottom and return to the top. So their process was constantly adding heat to the top, and removing it from the bottom.

Any difference between the heating and cooling duties would start shifting the gradient within the tank, but because it usually evened out over time it would never get to the point where either temperature was too far out of spec. And there was trim heating and cooling at all the endpoints anyway.

I could never figure out why they didn't just do two tanks. Perhaps it was someone's way to get around a footprint issue. The only time I really talked to someone about it, they seemed to think it was a bit of a dud.

What you can do, without mucking around with water chemistry, etc., is to simply use the cooling loop as is and add an inline heat exchanger to pre-heat the inlet water into the steam heating. This would keep the cooling loop relatively unchanged, other than adding the heat exchanger, and does not require you to adjust the chemistry. Likewise, your inlet water remains unchanged, other than passing through the preheat first.

While, in principle, one could use the inlet water as the coolant, but then what happens if the inlet water isn't required for some reason but you still need to cool? This would require adding a bypass to run the inlet water into the original cooling tower, and you have to dink with the chemicals, etc. to keep the cooling tower clean.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert! https://www.youtube.com/watch?v=BKorP55Aqvg
FAQ731-376: Eng-Tips.com Forum Policies forum1529: Translation Assistance for Engineers Entire Forum list http://www.eng-tips.com/forumlist.cfm