Criteria for stopping cooling tower fans

Trial and error I think, never had to do it myself. Off hand can't think of any problems, unless you would like to highlight some? The only thing I can think of is restarting a fan, local inspection should indicate how fast a fan is running in reverse, experience should then guide you as to whether it trips on overload when restarting. This would possible suggest fans on the ends. But you will need to check.

Regards JD.

I think CW refers to COLD WATER, as opposed to chilled water from the condenser, rather than a clockwise rotation you might be more familiar with.

Rich

Thanks JDRETIRED for your reply. I meant to ask about some basis of selecting which fan to stop first, which second, third and so on. The cells are numbered A,B,C,D,E,F,G,H,I,J,K,L and 6 cells (A to F) are in one concrete block and remaining 6 on another. If we stop for example A, B this means 4 cells are operational on first block and 6 on another. Will it make a difference in draft conditions, evaporation rate etc?

<Will it make a difference in draft conditions, evaporation rate etc?>

It certainly will. Evaporation in the cells switched off will be much lower.

Why is there a bottom limit of 27.7degC? What happens to the processes connected to the tower when the local wet-bulb temperature drops lower than this figure?

Given your above statement and assuming that the tower water is corrected to the same headers in both sumps, wouldnt it be advisable to shut off one fan in each group, preferrably in the central portion of each sump to reduce end effects?

Good advice for the original poster.

27.7degC does seem unusually high, and unusually concise, for the lower-end temperature for a cooling tower. Usually the cooling loop approaches, or achieves at periods of low demand for 'coolth', the local wet-bulb temperature, so the tower is probaby located in low latitudes. Switching off a fan, or a group of fans, would mean that the cooling circuit won't be doing as much cooling as it could. Why is this under consideration (rhetorical question)?

27.7 degree entigrade is our design tempertaure for condenser back pressure of 65 mm of Hg. This corresponds to about 8% wetness in the steam at the last stage of the LP turbine, but if we allow the CW outlet temperature to go down below 27.7, wetness will increase which will damage the last stage blades.

The easier control then would be variable pump output. Probably much more responsive to have a 2 or 3 degree throttling range, maybe even on the hot side of the turbine, to improve the operation automatically. Either a throttling valve on the pump output, or a variable speed drive for the pump. Probably a lot cheaper to have a throttling valve near the pump, (less high pressure line). Hydraulic, pneumatic, stepper or voice coil control is a relatively cheap installation. Coil control is probably the more expensive to operate and highest failure rate, especially for a large system. You also might find that with better control, the equivalent of 30 degree water might be more efficient. Your steam dryness is based on the output temp of the hot water, and the quicker response of direct control may allow for safer, lower cost operation, without worrying about whether a fan is off or not.

Regulate your flow at the demand to regulate your steam chilling (Why?), and you could improve your efficiency and operational safety. Then you don't nrrd to disturb the engineers nap.

Rich

Yes makes sense now, Inspection of the stopped fan/s and your own personal experience I think will guide you to make the right decision.

Regards JD.

Thank you all for your good comments. Yes need for maintenance of fans will I think will be good basis for selection of fans. Also experience will guide us better.

The updraft of proximate fans will combine and be more efficient than distant updrafts.

Rich

The lower limit of 27.7C seems a bit of a challenge, I assume it to be tied up with efficiency, cost of cooling against load gain? Does this occur on load reduction, or is it hot well temp effecting the feed heaters? or some thing else? Yes it will effect evaporation, but why is that important in regarding to the final outlet temp? Is there a problem with make up water supply, leading to chemical and sludge problems?

Regards JD.

Giving a bit more thought to the 27.7c limit. I was wondering if a miss interpretation has occurred? Cooling fans take a heavy load when starting, and a temp of 27.7C may be necessary before starting to prevent the fans tripping on overload due to cold dense air? This then would only be a starting criteria and would not apply to having to stop them? Is this a large unit or are the fans common to a group of smaller uints.

Regards JD.

Sorry looked at post again, I had assumed power house operation. Which would put the lower temp requirements into a different light.

Regards JD.

Typically, multi-tower arrays are usually parallel operation, hot water to all towers through a manifold, cold return through a common return line. Shutting down one fan still leaves the hot water running through a tower, resulting in a n influx of hot, probably over 35° water mixing with your cold water, if you try to maintain 27, shutting down one fan out of four suddenly boosts your cold temp to 29. On/off cycles are going to be excessive, wasteful, and wearing. Are the motors 3Φ or single? Varying fan speed would generally be a better bet, un;ess you can shut down the water through a tower at the same time. That depends on whether or not 2 or 3 towers can supply enough water to your pumps.

Do all 8 towers supply a single chiller?If you are saving 5 kw shutting down one fan, a 1° lower cold side temp probably saves you 100 kw at the condenser. Try reading the electrical load of your condenser with a 27.7 CW, and then do the same with 26.7 CW. You might find that turning of a fan is a very costly thing to do, besides the added maintenance from start/stop cycles. And, you really don't save any water, if that is the concern, because hoter water evaporates more water than coole, and any water evaporating is contributing 1000 BTU/lb for it's evaporation, whereever it is occuring.

I really think shutting your fans down is a false economy of a non-engineer.

Rich

Probably more related to the ambient dew point than the temperature.

Rich

I had assumed a large cooling tower, but if it is a small aux, then there is one other consideration. In the power station we had a small cooling tower for the H2 plant, the policy was to shut down a single fan as H2 plant operation required, the result was that the idle fan bearings all became pitted due to vibration, and the policy had to be reviewed. Just thought I'd mention it in passing. My neighbour who has extensive experiance with chillers and associated cooling towers is away, roaming the outback of Australia, so I will keep an eye on the post.

Regards JD.

Is it possible thast your chiller (not the towers) are giving the 27.5 as design temperature for the rated output of the chiller? If that's the case, any lower cooling water temperature will increase the efficiency AND the output of the chiller, which should be substantially lower cost to run either centrifugal or piston compressors. The fan coil units, or process, will stop warming the chilled water if the temperature is below a set-point (thermostat setting) and your chilled water loop temperature will drop, dropping the hot water temp to the towers. The cooler water will not evaporate as much water as the ΔT to dewpoint is less. If you have let's say a 1000 kw chiller, the electrical draw at a cold water temp of 27.5 and a chilled water temp of 10, would be say 1300 A. If the cold water temp drops to 25, I'd guess the draw on the compressor would drop to 1100 A, and if the chilled water temp drops to 9 because of low use, the compressor would shut down, until the water out reached say 12. That would be 0 amps for a bit. Presuming a 5 kw fan, shutting down 1 fan @ 10 A would need to be off for 130 minutes to save the same as 1 minute of compressor off cycle. The compressor is chilled gas cooled, and will not suffer from short cycles, and are a really rare failure item with proper care.

I realize you probably have a pair of compressors, maybe one for each bank of towers, and if the lag chiller is not needed, a whole bank of towers will shut down. A 20 min off-cycle of the lag towers would need 80 minutes of 1 fan off-cycled, without considering the 1100 of 1300 A saved by a chiller cycling off. Here's a thought, set a temp sensor in the tower sump, 1 for each motor, motor A off at sump temp of 20, b at 18, c at 16, d at 15 etc. Using the same lead fan and lag fan will end with a fried fan on the lead more often. You will have to change out that motor more often. Cycling through the fans as lead/lag would result in all of the fans needing replacement at about the same time. My preference would be repeatedly replacing the lead fan motor several times before the #3 fan motor needed to be replaced, too.

Your choice.

Rich