Introduction: PROOF OF CONCEPT: a Chilled Water Air Conditioning Prototype - DECOMMISSIONED.

I plan to use this 100watt compressor to chill water in an insulated tank while I'm at work. The energy for this will be via my home solar power system since I harvest a surplus of energy that I can't store for night use. I want chilled or iced water as the means to store that solar energy for later use. When I'm home at night I want this cold water pumped into the living room to provide cooling. This will reduce the drain from my solar battery bank since my air conditioner in that room is also solar powered.

Basically I want chilled water as the storage medium for my solar harvested energy.

Step 1: The Chiller.

The cold side of this former ice maker I have dipped into a bucket of water. During the indoor heat today it took 5 hours to chill the water to 18C with an ambient of 30C.

Mind you this is a plain open bucket and the temperature difference achieved is very good. I plan to use an insulated tank to store the valuable cold water.

Step 2: The Water to Air Exchanger.

I used a spare water pump to send cold water to an aluminum radiator with a 12volt pc fan attached. Clear pvc tubing was used to make the water loop.

Step 3: Thermal Performance.

Starting from the compressor to the water to air exchanger, the cooling performance is pretty good for a small scale prototype. The blower air is the same temperature as the water in the bucket. This is perfect since I live in a really hot climate.

I plan to use a bigger air exchanger for better heat extraction for the room. I like the idea of chilled water since it can easily be hooked up without having to worry about refrigerant lines and the hassles that go with them.

Stay tuned for the next version of this chilled water air conditioning unit!


betterways made it! (author)2017-01-20

Interesting. I want to have a cool cupboard for storage of food that doesn't need fridge temps & reduce need for giant fridge and have been idly looking for alternative methods

chanelok made it! (author)2016-04-18

A couple of tips from an old thermodynamics geezer. I designed and built chilled water systems for comfort air conditioning and process (chemical industry) systems for many years. I have a chilled water system in my home in south Florida. It works very well. First, it is important to understand that medium temp (regular air conditioning) compressors will move heat (in the range that you are contemplating) much more efficiently than will a low temp compressor. Secondly, moving heat with water is about 4 times as efficient as moving it with air. Therefore, if you can cool the system with water, your efficiency is greatly enhanced. I live next to a vast body of water that I use as a heat sink. Before I get any diatribes about thermal pollution, let me say that my system adds as much heat to the water in one season as does a 200 hp boat motor in 24 hours. Third, air handling units (kinda like automotive radiators) that are actually designed for chilled water are much more efficient/less bulky than discarded A/C evaporators (as suggested by rum100proof below). You can occasionally find them at a bargain on ebay. The A/C evaporators will definitely work, but the fans are not usually part of the package. Fourth, I always utilized propylene glycol in my systems. At a 50% mix, my efficiency loss was 6%, but the brine remained liquid down to -50 degrees F. That's pretty cheap insurance to keep from freeze-damaging your system and the propylene glycol is food safe (as opposed to automotive antifreeze).

Making ice with such a system does not improve efficiency. However, if you are producing an abundance of energy, this is a good way to use it. Liquid water absorbs one Btu per pound per degree F. rise in temperature. Ice melting absorbs 144 Btu per pound without a change in temperature. Obviously, the cooling capacity per pound of ice is much greater than that of liquid water.

NOTE: I have never posted here before. I am not sure how I might be contacted if any of you might have questions. You can email me at

rum100proof made it! (author)rum100proof2016-04-22 Mjtrinihobby said.... you are a guru!

My idea was to use a split airconditioner indoor unit as the AHU. These are (at least in my neck of the woods) complete units with fan, controllers and remotes. My idea was that these are aesthetically better and have all the controls, i.e. fan speed, temp regulators etc already built in. Am I wrong here? and am I thinking too simplistically? My main concern was, would the cooling coils ( which are designed for gas expansion) have the carrying capacity for enough water to provide sufficient cooling? Any insights would be helpful!

Also since I live in a 5thfloor apartment, I do not have the option of water cooling the return flow before it enters the chiller. My idea was to run the return line thru a cooling coil buried inside a swamp cooler. The swamp cooler, the chiller and possibly the pump would run off solar panels (depends on how much solar power I can generate) Running the whole system during sunlight hours, typically 8-12 hours as day during the warm season, would keep the apartment cool during the hottest part of the day and to use grid power only for the nights.

As a "thermodynamics guru" do u think my "plan in my head" makes any sense at all ? or am I barking up a pine tree in a tropical rain forest? ;-) . Mjtrinihobby has helped me to see the light with his proof of concept project but I would love to have any inputs you could offer.

BTW you mention............ "but the brine remained liquid down to" ........ do you use the propylene glycol in a brine solution or just water?

Many thanx again


chanelok made it! (author)chanelok2016-04-22

Hey Rumi, I probably should clarify a couple of things. My system is a water cooled chilled water system. Both the condenser (condenses hot gas from compressor to hot liquid refrigerant) and the chiller (chills a glycol/water brine to about 35 deg. F) are tube-in-tube heat exchangers. The outer tube is steel and the inner tube is copper. Refrigerant flow is between the steel and the copper. Water flow is inside the copper. I use one circulating pump for the condenser to carry heat to the river. I use another circulating pump to move brine between the chiller and the fan coil units (closed loop). I have a cap tube metering device at the chiller. There are some other efficiency enhancements that I can get into in greater detail if you like.
Using a DX coil in place of a fan coil unit will work with some extensive copper tubing modifications. The metering device must be removed. U-bends will most likely have to be added to get adequate water flow. The airflow will be much greater than the heat transfer rate, which will waste energy. The fan should be slowed as much as possible. I will be happy to view pics of the AHU you intend to use and make suggestions.
If you live where the relative humidity is low, you can use a swamp cooler alone to chill the water without a refrigerant system. I designed many systems to do just that under certain conditions. Otherwise, you'll have to rely on an air cooled condenser. If so, the size of the condenser coil will determine efficiency... the bigger, the better.
I hope this helps. Good luck!

rum100proof made it! (author)rum100proof2016-04-22

Chanelok.......Wow, that is some detail you have gone into!...... I dont think tube in tube heat exchangers are within my capacity and capability. Though I do live in an area of relatively low humidity - around 50% most of the year..... using just a swamp cooler I dont think is really an option since temps regularly reach 35*c in the summers and last week we had three straight days of 40*c+. So I believe a chiller unit is called for. To extract some the heat from the return water(brine) my idea was to use the swamp cooler setup before it hits the chiller.

Your caution on the effectiveness of the aircon AHU is great! That is what I wanted/needed to know. I dont have one ( or the 5 I need) as yet and was trying to figure out if they would work before loading my home with more useless stuff than it needs to handle ;-)

About the chiller unit my plan was to use a solar powered, off-the-shelf, chest deep freezer which has the option of grid power in low/zero sunlight conditions. I have considered two options when using the deep freezer.......... 1. Use the storage cavity as a storage tank/chiller tank for the water ...or.... .... 2. Lining the inside of the cavity with copper tubing to work as a heat exchanger and making it a sealed system for the chilled water.

One more question, if I may, since the piping to the 5 AHU's will be considerable length, do I need to insulate the return line? My limited knowledge throws up two conflicting scenarios........ 1. If the water is still cold( i.e. less heat extracting happening if only 1 or 2 AHUs are working at the time), an un-insulated line extract heat from the ambient air and ... 2. If the line is insulated and all AHUs are online then there will be minimal heat loss during the return, which is counter productive. Which scenario would be the less damaging ( or more efficient) in your opinion? The third possibility I had considered was using a thermal valve to redirect the flow to the insulated line when below a set temp and to the uninsulated line above the set temp. Piping costs are manageable, but the thermal actuated valves are pricey and I was wondering if I need to go to such lengths?

What are your thoughts on this?

I will definitely take your offer and will try to send you some pics of the type of AHU I plan to experiment with. Dont have one on hand but will try to get pics of an existing installed one and mail them to you.

Appreciate your time and effort to share your experience and expertise.


Mjtrinihobby made it! (author)Mjtrinihobby2016-04-18

You are a guru! Thank you for that information. Can I private message you for further help as I test the revised version of my chilled water air conditioning prototype?

chanelok made it! (author)chanelok2016-04-22

Absolutely! I'll try to check this site more often. I'll be happy to help any way I can.

JohnG322 made it! (author)2016-04-05

Yes very clever.

I would be interested in knowing the temp difference between the water supply and return. The temps you have listed are of the heat exchanger and the compressor unit. That difference won't give a measure of cooling efficiency.

Measure the water temp before and after the heat exchanger. The delta of the water temp should allow a standard for measuring efficiency of design changes. The bigger the difference that can be continually sustained the better.

chanelok made it! (author)chanelok2016-04-22

Regarding efficiency, this is a step in the right direction. One must also determine the flow rate and the specific heat of the medium used to transfer the heat. I do not wish to go off the rails here but ultimately, efficiency is the watts moved versus the watts expended. If we expend 1000 watts to move 3000 watts, then our efficiency coefficient is 3. The higher the coefficient, the greater the efficiency. An analogous situation would be moving freight by truck or by rail. It takes about ten times the energy to move a given mass of anything by truck as it does by rail. Therefore, the efficiency coefficient of rail is ten times that of a truck.

Another efficiency consideration is that excess energy in one area (living space) might be moved to an area where it is desirable (hot water for showers, washing dishes, etc.) This can also be accomplished through judicious use of heat exchangers.

Mjtrinihobby made it! (author)Mjtrinihobby2016-04-05

hmm, I will do that! Good points! I'm building an improved system now so I will take measurements and post on the new instructable! Thanks!

rum100proof made it! (author)2016-04-17


This a wonderful proof-of-concept project and has helped me to take the decision to go ahead with a similar if more complicated project I have been working on (in my head) Your project saves me the trouble, expense and work on a POC of my own. Was always convinced it would work but was not sure of the energy saving potential.

I now have some questions which i wonder if you or anyone else could answer...

1. I plan to use a solar powered domestic chest freezer as the chiller unit.

Question: Do I need to use the freezer as a water tank/make up unit, or could

I make a copper cooling coil that runs along the walls of the freezer

This in my mind achieves two objectives- 1. I can have a sealed

system with no air getting in to cause air-blocks and corrosion.

2. The remaining space in the freezer can still be used

2. I plan to use junked/recycled split airconditioning AHU's to extract the heat in the rooms. The reason? Aesthetics, already installed control systems with remotes etc.

These AHU's should be cheap enough to acquire since according to my experience faulty air conditioners are usually due to the compressor /condenser side problems.

Question: 1. Can anyone see why this would not work?

2. Are the evaporator coil pipes large enough diameter to carry water

instead of refrigerant.

Any other cautions, inputs, suggestions are welcome.

Once again thanx and good luck with your project!



Mjtrinihobby made it! (author)Mjtrinihobby2016-04-17

Ah you read my Mind! How did you do that? I will use a chest freezer and simply fill it with engine Coolant to allow below 0C temperatures. I will drill an access port in the lid to allow tubing, power and sensor lines. My setup can freeze water but it is too low powered to freeze a bucket during the solar harvesting hours. I have however modified it slightly to chill a 2inch pvc tube with a tiny water pump and successful cooled my bedroom with it. As I type this to you I'm installing wall brackets to mount the small compressor outside on the wall.

The chest freezer is perfect for having an insulated tank of chilled water. That will be a later stage of my chilled water air conditioning prototype which will be controlled by my home automation system. I basically want on demand cooling and stored chilled water. The unit I'm installing now will be for on demand. The chest freezer for storage of coldness.

Using old air handlers are perfect! I don't see why the tubing size would stop you from going ahead. I used a 6inch square pc water cooling radiator and that worked well to cool my bedroom which is 10ft by 12ft. Mind you it took about an hour to remove the heat of 28C to a comfortable 23C.

I think we both are on the right track and I will keep you informed of my progress. Private message me anytime to exchange notes!

rum100proof made it! (author)rum100proof2016-04-17

Thanx for your response and thoughts.......... I am barely competant at reading books, so reading minds is far beyond my metaphysical capabilities :-) ;-).

What does your experience tell you about my idea of wrapping a custom made cooling coil on the inner walls of the freezer rather than free coolant. My thought process is this where free liquid in the freezer cavity is concerned....

1. The return liquid entering the freezer cavity will warm up the already cooled liquid partially defeating and adding unnecessary load. This could be mitigated by horizontally baffling the cavity so the flow of warmer water is dropped on top and slows the mixing with the colder lower layers. The liquid pick up line would need to access this layer.

2. We would need some kind of agitator to mix the central column of liquid with the wall column else the central column will be warmer. This could defeat the natural layering of warm and cold liquid so could it land up to be self defeating?

I assume when you say engine coolant you mean antifreeze mixed in water. Or are you referring to the new "waterless-coolant" available? What is the cost effectiveness of using "waterless-coolant" and what is the chemical footprint?

I am not an engineer by training or experience but do have an aptitude for understanding by observation, however do you think I am "overthinking" the whole process???

Appreciate your inputs and sharing your experience and knowledge.

BTW my system (theoretical) needs to cool approx 1500 sq ft of living space so I need to ramp up my system quite a bit!

Mjtrinihobby made it! (author)Mjtrinihobby2016-04-18

My best advice for a diy system is to build it and test it in a smaller room. For me I intend to simply fill the chest freezer with antifreeze Coolant, no internal agitation. The chemical content I have no idea but yes it will be more expensive than plain water. You can try water first but it freezes too fast.

1500sqft is alot of space to cool and I'm thinking an inverter class air conditioning unit be the best way for you. You can still use the chest freezer experiment to assist the commercial AC unit.

My aim is to have my surplus solar power converted to ice and also have a low powered air conditioner as well. I will keep you updated on my progress.

rum100proof made it! (author)rum100proof2016-04-18

Thanx for your time and effort to give me these insights. Will keep you updated on my progress ........if and when! ;-) In the meanwhile i will follow your progress reports with keen interest

Mjtrinihobby made it! (author)Mjtrinihobby2016-04-18

yes stayed tuned for the newer version of my water chiller. I'm currently doing real world testing plus tweaking.

rum100proof made it! (author)rum100proof2016-04-18

Thanx for your time and effort to give me these insights. Will keep you updated on my progress ........if and when! ;-) In the meanwhile i will follow your progress reports with keen interest

DIY Hacks and How Tos made it! (author)2016-04-02

Clever DIY air conditioner system.

Mjtrinihobby made it! (author)Mjtrinihobby2016-04-02


About This Instructable




Bio: As of April 2017 I have decided to no longer post on instructables. The fact that several of my published works have been removed without ... More »
More by Mjtrinihobby:Replacing the starter and overload in a fridge. Fixing a Slack Transmission Dipstick. How to Send Bold and Italics Text in Whatsapp
Add instructable to: