Daytime: Solar->Ice-> Nighttime:Air conditioning

[Rick Glimmer]

I am trying to figure out how much ice i need to produce during daytime to cool a bedroom at nighttime.
Currently a 9000 BTU Air conditioner with a separate indoor and outdoor unit is more than sufficient.
The air conditioning is set to 24 degrees Celsius and the outdoor temperature goes from 32 to 28 degrees Celsius during the night.
The conditions where i live (Thailand) has very high humidity. Between 70-90% is very common.
At daytime i am outside not needing air conditioning, starting from around 19:00 when the house gets closed up the thermal storage from the walls, floors etc start to heat up the rooms because there is much less airflow. A simple fan relieves a little but not enough to live comfortably. Sweating does not help much because of the high humidity. After about 16:00 the wind practically stops blowing and closing up also prevents insects from entering.

I found a machine that makes about 100 Kg of ice with about 20 kWh of electrical energy.
My question is how much ice do i need to have the equivalent cooling power of a 9000 BTU air conditioner for about 8 hours. The room to be cooled is not leaking air, is isolated with about R15 and is about 32m3 (4x4x2) of which only the bottom half would need to be cooled efficiently as the higher air is not being felt when sleeping.

 

[brotherStefan]

The simple answer is that your 9000 BTU air conditioner would theoretically produce the equivalent of 3/4 of a ton (or 680 Kg) of ice (at 0ºC) per day (or about 230 Kg for 8 hours).

The more complicated question is, how will you melt that mass of ice in 8 hours? Considering that still air is a very good insulator, you will need to provide a constant flow of air across the ice (and the ice needs to be spread out because it only the exposed surface will melt). And the 230 Kg of ice, once melted, will leave you with 230 Kg of water at 0º. Do you keep the water until it warms to near room temperature before disposing of it, or do you drain it outside as it melts to avoid the additional humidity?

 

[Rick Glimmer]

Thank you for the answer.

I plan to circulate a fluid through copper/aluminum pipes that are surrounded by the ice (and in time melted ice water). This cooled down fluid will then be pumped through a modified internal airconditioning unit which will take the warmer air (1.8m high) and circulate it back into the room. Condensation will be on the aluminum 'radiator' and by gravity will be removed from the unit to an outside drain.
Basically i want to replace the outdoor unit by a box with ice and a small pump. At daytime i will have an excess of solar power that could be used to make ice.
I do not store daytime excess energy in batteries as they are expensive and need to be replaced several times over the solar panels lifespan. The biggest energy user is the air conditioner at night so that offers the biggest savings in energy use. I will not be off grid, just trying to lower the bills as close to zero as possible. :)

 

[brotherStefan]

Thank you for the answer.

I plan to circulate a fluid through copper/aluminum pipes that are surrounded by the ice (and in time melted ice water).

As I mentioned, air is a good insulator, so you will want to avoid the condition of the ice melting away from your network of tubing inside the ice box, leaving pockets of air. Once you have charged the ice container with ice, add enough initial water to cover the interior plumbing network so it always stays in contact with cold water.

 

[Rick Glimmer]

Yes, the pipes will be on the bottom parts, so there will be a layer of water to help with heat transfer.
I just realized that the air conditioner is not running full power all the time, i will buy a watt meter to see how much it consumes in one night. I expect that it is only on for about 20-40% of the time. If i just take the electricity bill it is about 600 kWh in one month. Without using the air conditioner it is about 150 kWh, so it is save to assume the air conditioner alone consumes about 450 kWh per month. Also the cooling 'power' of ice/water will be useful until it reaches about 18-20 degrees Celsius.
Not sure how much difference in water temperature and air temperature is needed to get a reasonable cooling effect. Obviously it will be a lot less when the water temperature approaches the air temperature. Also removing the condensed water from the air will make it feel cooler. All in all i hope to get get down to about 50-75 kilo of ice needed as that can be made with about 3Kw of solar power. I have 5Kw available so it will have about 2Kw spare to use for other stuff.
Some calculations if that is in the ballpark would help with further steps. I will need to use about the right size of insulated 'box' to store the ice and also need to modify a indoor unit. If those wished for numbers are out by an order of magnitude it would be nice to know.

 

[brotherStefan]

Somewhere you will need to determine the heat transfer coefficients for the tubing you plan to use in order to determine how much tubing needs to be in the ice box. And I'm not sure how you would control the heat transfer rate when the system is operational, nor do I have a good idea of how the efficiency for such an experimental unit would be calculated ahead of time.

 

[Rick Glimmer]

There are a lot of variables. Controlling the heat transfer from fluid in the tubes to the ice/water is mostly done with how fast the fluid is pumped. I plan to use copper tubing as copper has the best heat transfer capabilities. Once out of the icebox the fluid will go through a plastic tube with insulation to prevent heating up the fluid before it arrives in the indoor unit. The cold transfer to the air is influenced by how fast the fluid is pumped and how fast the warm air is blown over the 'radiator'.
Your answer giving about 250Kg of ice to be equivalent of running a 9000 BTU air conditioner for eight hours suggests it can be done.

I found a youtube video that uses the same principal here:
Their solution is to make ice during the night when electricity is cheaper. For me it is the opposite. Peak usage is during the night while the cheapest energy is at daytime. :) It shows it can be done, but i do not have enough knowledge to go into the numbers to see if it is feasible on a smaller scale.

In some time i will be able to measure my current air conditioning power consumption to get a better idea of how much ice/cold water is needed. Sofar it seems possible.