PDA

View Full Version : Need Help with cold water tank design



ac_sheldon
13-11-2006, 06:33 PM
Let me start by saying that I am not a refrigeration engineer. I am trying to develop an application for my company and came across this forum. I am hoping that someone here can point me in the right direction. Here goes.

I need an insulated tank of FRESH water (approx. 48x48x48") in which I can vary the water temperature between 0 and 35 deg. C.

The warming operation is easy but I am having trouble sourcing a system that can do the cold part. I have to maintain water temperature when in cold mode between 0 and 1 deg. C. It will be 100% recirculated water - I just need a tank of cold water. In the past we have filled the tank with ice and let it melt - using the slush layer at the bottom as our 0-1 deg. C water. Not the most elegant solution.

The problem I am finding is that most chillers cannot go this low without freezing up unless you add salt or antifreeze. Unfortunately, I have to have fresh water.

Does anyone know of an off the shelf or semi-off the shelf solution for this system? Cost is a factor as I am trying to limit the total cost of the system to $3000 or less.

Any advice or help is appreciated. As an electrical engineer, I am out of my area of expertise and need some help!!

Thanks in advance!!

NH3LVR
13-11-2006, 07:39 PM
Welcome to the Forum!
You are indeed correct about the problems with Chillers that near the freezing point.
Ice is the easiest solution, but as you mentioned it is not very elegant.
I have worked on systems that would do what you needed. They involved Vertical Chillers using plates or refrigerant chilled tubes mounted above a tank. The water is recirculated to the top of the Chiller and runs by gravity back to the tank. It works very well. If it should freeze it will cause no damage. In fact you will want it to build a bit of ice. One common application is in cooling Cherries. Chester Jensen makes this kind of equipment.
http://www.chester-jensen.com/
But this is going to be above your budget.
Another possibility, if you had a maintenance shop at your location would be to fabricate a copper coil in a housing. This is not a large project. It would require a Tech to charge it legally.
The heat load is important here. Do you have any idea how much heat is put into your process?
Others who work these calculation all the time can work such things out faster than I. I would consider using a Hot Gas Bypass System to maintain a stable suction.
When you want the Temp to rise would you just shut off the Cooling or do you need to maintain varying temperatures?

LRAC
13-11-2006, 08:17 PM
Hi NH3LVR

Not sure if you have these in the states, python beer coolers complete with refrigeration system and water circulation pump. This is a stand alone unit and should meet your requirements.

Plenty available in the UK not sure about USA.

Kind regards
Lrac

ac_sheldon
13-11-2006, 08:24 PM
Thanks for the replies. To answer your questions -

1. There is no heat load other than room ambient (21C). We are testing lifejacket inflators and just drop them in the cold or hot water to make sure they still work. It all happens in less than 30 seconds.

2. It will not need to constantly vary the temperature between hot and cold. It would be either cold (with the conditioner system on) or hot (with the heating element on) but not on the same day.

Also, I am not concerned about how long it takes to get to setpoint. As much as 24 hours would be fine. I just need tight control once I get there.

monkey spanners
14-11-2006, 12:09 AM
Check out www.packo.com (http://www.packo.com) and look up ice water systems in the dairy section. All you need is a suitably sized coil of pipe in the bottom, or down one side of your tank to form ice on with enough free space for you testing. Add an air blower and spray bar at the bottom of the tank to ensure even temperature. The condensing unit could be mounted on one side off the tank like an old fullwood 90 gallon milk tank.
This is how all the old milk cooling tanks used to work in the UK until direct expansion took over in the nineties. Will have a think about it tomorrow and maybe come up with some more ideas.

Cheers Jon

taz24
14-11-2006, 01:21 AM
Check out www.packo.com (http://www.packo.com) The condensing unit could be mounted on one side off the tank like an old fullwood 90 gallon milk tank.
This is how all the old milk cooling tanks used to work in the UK until direct expansion took over in the nineties. Will have a think about it tomorrow and maybe come up with some more ideas.

Cheers Jon


You beat me to it after years off working on milk tanks I thought that the design would suit this aplication right down to the ground.

Cheers taz.

US Iceman
14-11-2006, 01:31 AM
Add an air blower and spray bar at the bottom of the tank to ensure even temperature.


I llike the idea of the air sparger to provide some agitation of the water. A cheap little air compressor might be suitable for this purpose. The air pipe would have to be trapped above the top of the tank to prevent water from siphoning back onto the air compressor. Or, mount the air compressor above the water level.

A simple plate heat exchanger (Direct expansion to keep it simple) or two sized properly to have about a 12 hour pull down should help to keep the refrigeration system reasonably small.

Here is a source you can review for the plate heat exchangers. This are similar to those described by NH3LVR also.

http://us.tranter.com/phe/platecoil/plateapp.htm

Between one or two plates, the refrigeration system, and some misc. work you are getting close to your budget. Perhaps a little more, but... it does follow the KISS principle quite well.:cool:

TXiceman
14-11-2006, 03:50 AM
You will not be able to get your water all the way to 0 dC with pure water. With a falling -film chiller like made by Chester Jensen, 33 dF is about as close as you can get. The problem with getting so close to the freeze point of the water is that the refrigerant on the other side of the cooler is 3 to 5 dC colder than the leaving water. You heat transfer wall temperature can be below freezing and will have some localized freezing. You may gt it ti run this close in a steady state with a shell and tube or a brazed plate heat exchanger, but the slightest upset will allow some freezing of the water in low flow or localized areas.

I would stick with the falling film style evaporator.

Ken

Jolly George
21-11-2008, 10:18 PM
Let me start by saying that I am not a refrigeration engineer. I am trying to develop an application for my company and came across this forum. I am hoping that someone here can point me in the right direction. Here goes.

I need an insulated tank of FRESH water (approx. 48x48x48") in which I can vary the water temperature between 0 and 35 deg. C.

The warming operation is easy but I am having trouble sourcing a system that can do the cold part. I have to maintain water temperature when in cold mode between 0 and 1 deg. C. It will be 100% recirculated water - I just need a tank of cold water. In the past we have filled the tank with ice and let it melt - using the slush layer at the bottom as our 0-1 deg. C water. Not the most elegant solution.

The problem I am finding is that most chillers cannot go this low without freezing up unless you add salt or antifreeze. Unfortunately, I have to have fresh water.

Does anyone know of an off the shelf or semi-off the shelf solution for this system? Cost is a factor as I am trying to limit the total cost of the system to $3000 or less.

Any advice or help is appreciated. As an electrical engineer, I am out of my area of expertise and need some help!!

Thanks in advance!!
Hi Im no refrigeration engineer either, but at 0 degree's your not going to get a lot of ice if you stir the liquid constantly, so it can't solidify. Refrigerant passing through a coil, seems like a good idea, but as a guy who is designing heating elements for a living, I would purchase a flow (through) heater and use a temp measurement device to open a valve to allow liquid to flow out of the tank into the heater (when required) and return back. Does that make sense?

Brian_UK
22-11-2008, 12:24 AM
Hi Jolly George, you need to keep an eye on the dates of some posts, this one is two years old and the original poster hasn't been back with an update. :)