Cooling water consumers approx. 10-15m higher than cooling towers

[virk]

We are considering an installation with an open cooling tower mounted nearly on ground floor (pumps below) while the consumers will be elevated about 10-15m higher. Cooling water flow will be about 500 m3/h.

1) Concerning this I would like to know whether it is "allowed" to operate the return line to the cooling tower at below atmospheric pressure in the upper part or whether above atmospheric conditions should be maintained in the entire cooling water circuit.
(Of course I assume that pressure will be kept high enough to avoid cavitation at critical (high) points.)
2) If above atmospheric pressure should be maintained, is it a common method to "just" apply an adjustable valve at inlet of cooling tower and to adjust the required pressure drop with this valve?
3) Are there any other topics (degassing of water, etc.) which should/need to be considered in this special case?

Any hints to this topic are very welcome!

 

[lilliput1]

Instead consider closed loop evaporative cooling tower or cooling tower with water to water heat exchanger if you can't locate the cooling tower above the chiller.

 

[virk]

lilliput1, thanks for your answer! Of course we could consider other devices like you suggested but currently we want to stick to our system due to investment costs and "efficiency". So any further comments to my questions are very welcome

 

[lilliput1]

You can consider using a water back pressure valve or relief valve to hold back the column of water in the drop when the pump stops. However there is possibility of overflow if the valve fails. My previous suggestion is preferred.

 

[virk]

If "your previous suggestion is preferred", what problems are you solving with this suggestion? Why don't you prefer an installation like I described in my 1st post. That is why I put the three questions. I like to identify possible problems first.

 

[lilliput1]

virk I do't want to be blamed when overflow occurs because of eventual failure of you scheme. You are on your own, make the right decision for the client.

 

[SAK9]

Concerning this I would like to know whether it is "allowed" to operate the return line to the cooling tower at below atmospheric pressure in the upper part or whether above atmospheric conditions should be maintained in the entire cooling water circuit.
(Of course I assume that pressure will be kept high enough to avoid cavitation at critical (high) points.)

No.It should be above atmospheric pressure.

If above atmospheric pressure should be maintained, is it a common method to "just" apply an adjustable valve at inlet of cooling tower and to adjust the required pressure drop with this valve?

What you need a is a back pressure regulating valve that is normally closed by spring pressure and opens when pump pressure acts on it.Make sure that this valve provides positive shut off so that when the pump stops water does not drain into the cooling basin.The back pressure valve can be set say 0.5 bar above atmospheric.This valve can be installed at the tower inlet.

Are there any other topics (degassing of water, etc.) which should/need to be considered in this special case?

Provide air vents at high points as required

Another simple option is to have a static inverted U loop just in front of the tower inlet. The top of the loop will need to rise a little over the highest point in the tower piping. There will be a little of overflow but this limited to the down leg of the inverted U loop. Please bear in mind that inverted U loop can be a point where air might get trapped so automatic venting provisions will be required.

 

[LittleInch]

virk,

A sketch showing what you intend would have been useful, but looking at the OP my response would be

1) There is no absolute requirement to operate at a positive ( above atmospheric) pressure at all locations, but it is highly recommended. Operating lower than atmospheric pressure leaves you open to two phase flow / vacuum forming and this can lead to surging, bubbles where there shouldn't be any ,vibration and generally poor performance.

I would place the back pressure valve just before entry into the cooling towers at the low level.

2) A back pressure control valve which maintains a set minimum back pressure regardless of flow is a common procedure. Note however that over time, many do not seal 100% on no flow conditions.

3) As noted above, you need to work out what happens when the system starts and stops, which maybe is just making the pond under the tower big enough to accommodate more water from a partial drain down of the system or adding an actuated isolation valve on high level in the pond to avoid flooding the lower area.

I can't see many other requirements and 10-15m is only 1.5 bar extra.

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

 

[virk]

Thank you SAK9 and LittleInch for your answers. I am still think about operating sub-atmospherical; by this I could save a lot of electricity; we need to pump about 500m3/h and 1,5 bar more or less is quite considerable.
I am also thinking about installing variable speed drive (frequency converters) for the cooling water pumps. In this case I could reduce speed and simultaneously adjust the "throttle valve" ("back pressure control valve", whatever) and by this slowly operate into the sub-atmospheric region. Just my thoughts currently....

 

[lilliput1]

Consider having 2 back pressure regulating valves pipes in parallel at the bottom of the riser both valved off, one open the other closed (backup). Provide overflow alarm so operator can open the backup valve and close the defective valve. Raise the piping up downstream of the chiller before the riser drop and put in automatic air vent at piping highest point to make sure condenser is always flooded. Provide check valve and isolation valve at pump discharge. Make sure available NPSH is greater than pump required NPSH.

 

[SAK9]

If you are the owner operator,you have the licence to experiment!Please refer to the attached sketch.I would recommend to have a tower bypass line that connects the return line to the pump suction. Both the tower bypass line and the tower inlet line will have motorised isolation valves .The pump suction line out of the tower basin also requires a motorised valve. Before the system needs to be shut down, the bypass valve will open and the tower inlet line valve and outlet valve also shall close. Please take into consideration valve opening times for 100% open and 100% closure. Once the system has become a closed loop pump and tower can shut down. This is a complex system that that has multiple points of failure.

 

[lukaiENG]

This all seems a bit over-complicated to me. I would back Lilliput1's original recommendations, specifically the second (use of a plate heat exchanger to hydraulically isolate the CT from the rest of the system).
You can select PHEs with a pretty tight approach for relatively little cost which means that your efficiency would not change too much.
I can envisage a failure situation where a valve fails and you empty your entire CHW system into your cooling tower yard causing flooding and significant downtime. adding a PHE seems like a no-brainer from a risk perspective.

 

[lilliput1]

Sak9 scheme can still result in overflow depending on the volume of water in the drop and if the pump flow can not keep up with the gravity flow of water in the drop upon start up.

 

[LittleInch]

Well you could set the back pressure valve to be 0 barg at the highest point or at the top of the downpipe back to your cooling tower.

You just need a big enough tank under the cooling towers to handle all the cooling water without overflowing, or have an overflow to somewhere it's not going to cause a problem.

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

 

[virk]

Planned operation time will be approx. 8760 h/a; this means fultime. I do not prefer to install plate heat exchangers to an open cooling tower system. I am afraid of downtime due to cleaning activities. And also the temperature loss is not negligible. System is designed for 28° wet bulb and 31/36°C cooling water temperatures. Even a loss of 1K would cause a redesign of process and equipment, etc.
If the system is started once, a stop of it will only rarely occur. So for going down manual valves for closing some lines would be sufficient; also draining the entire system due to an operation mistake would not be a problem.
"Wasting" energy is a bigger topic: 20 kW more or less results in 170.000 kWh annually, which amounts to f.e. 25000 $/a (depending on tariff). This is why I asked about operation at sub-atmospheric conditions (by avoiding cavitation). If I just kept the pressure at 0,1 bar absolute at the highest point I would/could avoid evaporation/cavitation of the water and would most probably not require the additional 20kW motor. This is why I put my questions: I wanted to know what problems could be expected while operating at such conditions: For example, if air enters somewhere via a leakage f.e., this air finally gets out of the system again, but the pressure will rise and by this the cooling water flow will decrease. If this leak gets very big it could end up in having 0 barg at a point where I intended to have f.e. 0,1 bara.
Further problem I can think about is that cooling water is enriched with air (dissolved in the water) in the cooling tower at 1 bara. Perhaps this air tends to get released of the water again at locations where the pressure is much lower like the 0,1 bara. If this can happen, pressure would rise also because of this and flow would decrease.
This is why I am collecting some thoughts from all of you; I appreciate this very much. Thanks a lot for your efforts! (I just read in an email that customer has decided to locate cooling towers on ground floor; so it will become reality )