undercooling of service fluid during chiller shut down 1

[leo31]

Hi there, I went to a learning visit to a small plant where it was intended to use a chiller to cool down water used in a heat exchanger; from what I've gathered they had to stop using the refrigerant unit altogether as they switched their productions from continuous to little batch productions, resulting in having to turn off the unit multiple times per day causing the down below problem:

from what I understood during the shutting down phase the water sub cools as it has nothing to exchange heat with (as the batch has already ended and the heat exchanger is empty). This sub cooling effect makes the fluid unsuitable for the next batch and allegedly it's the reason why they are not using the unit anymore.

I was wondering if there’s some way to resolve this, they said they thought about using more water as the Q removed would be the same and since there’d be more flow the ΔT would have to be higher, resulting in less sub cooling. I was wondering if this could really be the answer, as they haven’t yet made any changes and the chiller remains unused.

 

[HVAC-Novice]

If I understand correctly. the chilled water is used to produce a product. When they stop the machine that uses the chilled water, the chiller runs a little bit longer since it is impractical to literally shut them off the same time.

what fluid is made unsuitable? is the chilled water cooling a fluid like an oil or what?

Some better control, modulating chiller or a buffer reservoir are options to solve the problem. How much time does the chiller run after the machine shuts off? Can this be shortened?
if it is 10 minutes, you need a buffer that stores 10= minutes of water. Maybe a buffer on that "fluid" side so it keep circulating even if the machine is shut off.

Need much more information onw hat is going on.

 

[MintJulep]

A sketch of the system and a process sequence diagram showing energy flow vs. time would be so helpful.

 

[leo31]

Sorry for not being clear before. The chilled water is used to cool down a product stream using a heat exchanger, it goes in at 7°C and gets out at 12°C, after that it goes back into the chiller unit (Refrigeration capacity 300 kW) where is cooled back from 12°C to 7°C making it suitable to go back to the heat exchanger, this works seamlessly during normal batch operation (30 min circa).

After the normal production there's no more product stream inside the heat exchanger tubes, so the chilled water gets in the heat exchanger at 7°C but also gets out at 7°C, it travels then back to the chiller that will cool the chilled water under 7°C, since it will exchange the same heat as before. This is what causes the sub cooling of the chilled water during the shutting down phase (3-5 min from what i know).

Apparently no buffer fluid can be used as a replacement of the product stream in the heat exchanger tube side during the shutting down phase as it could mix with residues form the product and cause problems for the next batch.

They said they thought of introducing more water to the chilled water already present during the shutting down phase, I don't really understand how that would work in practice.
I hope I haven't made things even more confusing !

 

[HVAC-Novice]

So is the chilled water getting too cold the concern, or the product getting too cold? Your OP says they have to discard the product.

on the chilled water side they have to tighten up control to shut off the chiller faster. and add a buffer tank to bridge the gap for when they are not bale to shut off the chiller fast enough. How fast a chiller shuts off depends on the chiller, controls etc. a HVAC engineer should look at the specifics. A modern modulating chiller may handle that better than an old single speed chiller.

 

[leo31]

The chilled water, the product is not present at the shutting down phase as the batch has already ended and the product is already stocked.
How would the buffer tank work ? thanks for your answers !

 

[HVAC-Novice]

I think this in your OP threw me off as it sounds like they have to discard of the sub-cooled fluid "This sub cooling effect makes the fluid unsuitable for the next batch". Obviously chilled water that is too cold can be re-used.

anyway, a buffertank on the chiller return side would still supply 12°C water to the chiller till it shuts down. How quick your chiller can shut down or modulate determines the volume needed. if it takes 4 minutes to shut down after the load is removed, and you have 100 gpm flow, you need 400 gallons in the return pipe + buffer tank. More = better. if your chiller can modulate down, it can be less. So it really depends on the equipment and system and isn't something the internet can tell you.

I bet someone knowledgeable looking at the chiller, controls and your system can make wit work as is. or design an upgrade to make it work.

 

[leo31]

I perhaps could've phrased that better, but yes they did say that the problem is the chilled water getting too cold, but from what you're saying that shouldn't actually be a problem. Also does the water in the buffer tank need to be at 12°C or could it be warmer? (i.e ambient temperature ). Thanks again, I'll try to have a talk with the supplier HVAC engineer.

 

[HVAC-Novice]

The return water temperature can be anything you want except too cold (or boiling, I guess). When a chiller system starts up, all water is at whatever room temperature. If your production run is short (like 30 minutes) you don't want to oversize the buffer tank, though. or it will take too long to actually get 7°C. Insulate the buffer tank. that way after an hour of no production, it still is "cool".

The reason the chiller should NOT get too cold water is that then can get close to freezing temperature. Like your chiller produces a dT of 5°V. if your return is 7°C, it would bring it to 2°C. that means the surface of the evaporator will be under 0°C and can ice up. the chiller has safety controls to prevent that. the math is a bit more complex, this is just to demonstrate. There also are glycol systems that can get colder.

The chiller manual will provide engineering data. Don't just buy a buffer tank based on what I said. an HVAC engineer needs to evaluate the specifics, how often and how long it is on an doff, what ambient temperature and so on.