Open tank or buffer vessel 5

[Thermo01]

Is there a reason why in process cooling, open tanks (sometimes with Weir plates) tend to be used rather than a low loss header or buffer vessel?

We have a site where we have been asked to combine 3 separate chilled water systems into one (2 x 40kw and 1 x 60kw chiller). All 3 systems have their own pipework pumps, heat exchangers, etc.

The plan would be to have 1 Chiller (140kw of cooling) with primary pumps, feeding a LLH or buffer vessel. The individual areas would then take off the LLH or buffer vessel with their own pumps. Whenever I come across open tanks, they always seem to have issues with the pumps cavitating (I understand they probably haven't worked out the NPSH to prevent this) but there must be a reason why an open tank has been used rather than a pressurised system.

Any suggestions would be appreciated

Thanks

 

[1503-44]

Don't know, but might there be advantages to reducing pressure to atmos, having some evaporation and allowing excess heat to escape at the open tank? Like a cooling tower ... without the tower?

If you think I am wrong about anything I say, then please do correct me, but please also refrain from condescending lectures to myself and others here, in the preschool level details of chem, math and physics. Thank you. I will try to do the same.

 

[Rputvin]

When you're not changing elevation it's a lot easier to have an open tank (or "nearly" open as the case may be) as your controlled low-pressure location instead of pressurizing, dealing with volume changes as temperatures swing, having pressurized makeup available, code/safety relief valves or the added headache of a central safety system, introducing air from servicing sections of the line, not being able to read a pressure gauge without finding the "zero" gauge to get your delta, etc.

The atmospheric tank simplifies a lot of those issues, and most companies won't really have design or code requirements if there's no pressurization happening other than some minimal pump head, so the cost of the install comes down significantly as well.

The more complicated you make it, the more expensive it becomes to bid, engineer, build, etc.

And keep in mind not everything likes to live on a pressurized loop. Most of the equipment I deal with has a backpressure requirement of <5psig, so an open tank is basically a requirement.

 

[LittleInch]

Well with an open tank you know for definite what the pressure will be.

Why your pumps cavitiate or ones you've seen I don't know, but there are many badly designed HVAC systems around so maybe they didn't understand the basics of liquid flow - who knows?

What's a LLH? It's always best to spell out abbreviations in full first as we all use different ones sometimes.

The thing you really need to watch when doing something like this is whether your one new big machine can handle the turndown and not start short cycling.

Sometimes more smaller units give you better flexibility and a more even temperature in your chilled water.

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

 

[1503-44]

Rvputvin, great post. I read an article last week about how the world is so in to solutions that always add things and making them more complex, rather than taking things away, or building simple things that work well. Like everything has to have all the latest lights, bells and whistles. Perfect example. Thanks.

If you think I am wrong about anything I say, then please do correct me, but please also refrain from condescending lectures to myself and others here, in the preschool level details of chem, math and physics. Thank you. I will try to do the same.

 

[FacEngrPE]

In the HVAC world a water tank can be sized to buffer swings in cooling load. and time shift loads. A decoupling manifold will work fine if you do not need thermal storage.
Thermal storage can also be sized to minimize short cycling of chillers.