Nightime AC by way of daytime freezing using dump load - is my plan crazy?

[Sparky]

I have about 2kw of solar cells on my roof. My housing situation is stationary, but I can limit my insulated space to an 8' cube. I have some batteries, but I don't want to depend on cycling them them to run my A/C unit.

I'm thinking I'll get a standard 120v deep freezer and use it to chill a solution of salt water or glycol/antifreeze to near-freezing temps. I'll then pump the solution through a radiator & fan and bask in the awesomeness of cold air.

My assumptions are that

1) the electricity required to chill the water, pump it through the radiator, and power the fan, is lower than that necessary to run a conventional AC, and

2) most of the chilling required will be done during daylight hours with electricity that would otherwise be dumped, and

3) it may take a few days from initial setup to chill the solution, but I'll be able to maintain the chilled thermal mass through the evening & regenerate daily because it'll pick up less heat during "use" than refrigeration pups out when the sun is up, providing for the dump load.

What I don't know is how to do the math to determine whether this will be feasible. I've seen ice storage systems, but they seem to focus on cheaper night-time generation and daytime chilling. I'm planning on the opposite - daytime freezing for A/C at night, and doing so with consumer-grade components.

Is someone willing to either assist, or confirm that I am indeed bonkers, kookoo, and operating with a sunburned brain?

 

[FilterGuy]

I don't think I can help.... but I am going to follow the thread.....

 

[Supervstech]

Chiller systems are used in commercial large systems due to the efficiency, and power can be sized for all the refrigeration equipment in the mechanical room. So, only piping, and small amp blowers are needed in the locations being cooled.

I think a small setup could be designed to do what you propose, but the coolant energy requirements are going to be the same as an air exchange system. Just the time of energy transfer is shifted...

An 8x8x8 heavily insulated cube could be chilled, then reversed during the night... I wouldn’t use sailing, as it is fairly corrosive, unless you plan on stainless tubing and coils... a rather poor heat conductor... glycol is commonly used for very cold storage fluid.
Research pool Heatpump’s, and geothermal equipment for ideas...

 

[boondox]

Ahh, an eutectic system! I was thinking about making on for a van a while back, even collected up some of the stuff. The idea was the same, I would cool the system while driving and use the cooled system for food storage at night. In the end I decided to stick with a cooler. I have somewhat related system at my place, I have a device I made that sets the temperature setting of my chest freezer lower when there is surplus solar power. It saves a little battery juice at night time. Not all that much but every bit counts.

But I digress. In your situation I think it is going to be tough to really crunch the numbers well enough to get a clear idea of whether it will work. But you could get a ball park idea, at least to know if you are stark raving mad. If you decide on the solution and volume you would use, find out how many BTU's it would take to cool it. That will tell you if you have anywhere near enough power to chill the system. That will also tell you how many BTU's you could pull out. Take a look at your room value, assign an R value. There are AC calculators to size systems, so use one. If the numbers are anywhere near each other then it is worth an experiment. Or, you could go in reverse, take a SWAG (scientific wild ass guess) as to how much cooling you would require to keep your room cool and go from there. Or go sideways and figure out how much spare power you have, then how much you could cool the liquid and hence how much you have to cool the room.

 

[mvas]

My assumptions are that

1) the electricity required to chill the water, pump it through the radiator, and power the fan, is lower than that necessary to run a conventional AC, and

2) most of the chilling required will be done during daylight hours with electricity that would otherwise be dumped, and

3) it may take a few days from initial setup to chill the solution, but I'll be able to maintain the chilled thermal mass through the evening & regenerate daily because it'll pick up less heat during "use" than refrigeration pups out when the sun is up, providing for the dump load.

What I don't know is how to do the math to determine whether this will be feasible. I've seen ice storage systems, but they seem to focus on cheaper night-time generation and daytime chilling. I'm planning on the opposite - daytime freezing for A/C at night, and doing so with consumer-grade components.

Is someone willing to either assist, or confirm that I am indeed bonkers, kookoo, and operating with a sunburned brain?

How did you calculate that your DIY Chilled Water Cooler will be more efficient than a very High Efficiency Inverter style Window A/C Unit?
Can your system achieve an EER rating of 14.7 or higher ( where higher = better ) ?
How many BTU's do you need to cool your 8'x 8' room?
What container are you going to store your many gallons of Cold Water in, that does not sweat?

 

[svetz]

The problem you'll most likely run into is a lack of thermal mass. A general rule of thumb is 20 BTUs/hr per sqft, but that's really squishy as it depends on how hot it is, insulation quality, and humidity. So, your 8x8 room for 10 hours would need 20x8x8x10 = 12,800 BTUs

How the Math Works
So, let's talk theory. The definition of a BTU is the amount of heat required to raise the temperature of one pound of water by one degree. So, if you stored 20 lbs of pure water at 33 degrees and used it to cool a room to 83 degrees that's (83-33) x 20 or 1,000 BTUs. That's not much cooling over all. The energy change in a material without a phase change (e.g., a phase change is like ice melting to water) is called sensible heat.

Most cooling cycles involve a phase change as it takes more energy. For example it takes 144 BTUs to melt 1 lb of ice, 970 BTUs to boil off 1 lb of water. The energy to change from a liquid to a gas is called the latent heat of vaporization. In an air conditioner with R22 it's 100 BTU/lb of R-22 (ref).

Now, let's say you freeze that 20 lbs of water during the day and run an antifreeze solution through a coil in the ice and one in your room. Now you get the latent heat of melting + sensible heat, so 20x144 + 1000 = 3880 BTUs. Still probably not enough cooling to be worth the bother.

How much ice would you need to keep the 8x8 room at 78° F for 10 hours if you froze it to 10° F? Well, for 1 lb of ice it would be a sensible heat change of 78-10=68° which is 68 BTUs or sensible heat and the latent heat of melting is 144, so 68+144=212 BTUs/lb. We estimated earlier you'd need 12,800 BTUs for 10 hours, so 12,800 BTU / 212 BTU/lb = 60 lbs of ice.

In practice, the larger the temperature difference between the inside and the outside the quicker heat will enter. A block of ice might chill the room down to a low temperature, causing more heat to enter the room, making it need more than the estimated 20 BTUs/hr/sqft (which is for AC units that turn on/off to maintain a constant room temperature). The shape of the 60 lbs of ice (and whether or not it's partially insulated) controls how quickly it will absorb heat and what the room temperature is. As it melts and raises temperature heat will flow into it more slowly allowing the room temperature to rise, so it might be difficult to control the room temperature.

Swamp Coolers
By now you're thinking Swamp Cooler. If you let 20 lbs of room temperature water evaporate per hour, that's roughly (20x970) 20,000 BTUs of cooling per hour. That's some serious air conditioning!

Why would the water evaporate? How well it works depends on the temperature and amount of moisture already in the air. If you live in low humidity conditions (desert) they work great. If you live in high humidity conditions (e.g., near a swamp, Florida), then the water won't evaporate as fast or at all. There's also the problem that as the water evaporates, the humidity increases, so you need a two stage system so outside air is humidified and exhausted while the cooling is transferred inside. Or just occasionally open the door and let the humidity out. I found a couple of cheap swamp-cooler box fans in HomeDepot years ago for cheap, but you can make them too.

In your case, you could also make ice cubes during the day and dump them into your swamp cooler to get the two phase changes plus sensible heat.

 

[svetz]

... People used to do that ...

There's towing icebergs too!

Mid 1800s: According to the Encyclopedia of Antartica, small icebergs were towed from southern Chile up to Valparaiso as part of the brewery supply chain. A Chilean researcher said, "The icebergs were towed by ships of the conventional type. Sometimes the icebergs were supplied with sails to utilize the prevailing winds. The ice was used for refrigerating purposes in the breweries and was generally substituted for artificial ice." Apparently, the business continued until about the turn of the century.

... Air conditioners mostly cool by pulling moisture out of the air.

Not so!
Air conditioners work by pumping heat, condensation is a side effect and nothing to do with making it cooler. However, drier air generally feels more comfortable. Also, condensation actually makes air conditioners less efficient as you have to pay the energy cost for condensation (it's the latent heat of vaporization of water, but in reverse).

... But if you bring in a big cold mass in a radiator, aren't you going to have ... condensed water to deal with?

Yes. But as to how much there's only so much water in the air in a closed to room to condense...after that it's how much humidity is leaking back into the room and if you brought in ice, of course there's the melt from the ice to deal with.

 

[LanduytG]

Webasto basically did the same thing for truck sleepers. It did not work well but then again sleepers are poorly insulated and day time solar load is tremendously large. Worked so so at night.


Greg

 

[Jeremiah]

Have you seen Will's video no his solar shed's Air conditioner? i'm not sure if it's the one I posted or another one but out there in Nevada in like 90 degree weather his shed is nice and cool and his AC has only been using like 330 watts.
Maybe check out a little unit like he's got, insulate the b'jebus out of your cube and enjoy.
330 watts should be doable and that's day time load you're talking night time so who knows maybe it'll go as low as 200 watts? I have no experience with this modern small efficient AC's but seems like the way to go and less trouble. Otherwise the icecube to swamp cooler thing is also what I was thinking if you are in a dry climate (just set it up blowing on you).
*** sorry, I just realized I did something I hate. You asked about a specific action and I proposed something else. I'll leave the post up b/c it may still be useful but just read it with an asterisk "my bad" ****

 

[Sparky]

Jeremiah,

No worries on the alt solution. I appreciate it, but from an expense standpoint alone, that bluetti solution is out. The freezer must be chosen wisely, for efficiency + longevity, but ice storage, if I can do it, will not degrade like lithium.

 

[Jeremiah]

Ok just brainstorming, what if you get a water bed, and run your copper coolant lines/radiator under the "mattress" so instead of depending on convection cooling through naked air (not a high specific heat capacity, but instead use the water in your waterbed (high specific heat capacity) to make your idea more efficient. You'll cool the water in your water bed where you'll sit/sleep, insulate the copper coolant line for the whole journy to your bed, then no insulation on it as it makes contact with the mattress / bed radiator.
I think this would be a doable way to maximize your idea since the whole problem is efficiency of heat transference.
You could also have a system where you pump the fluid right into the water bed mattress but you'll probably loose cooling energy too quickly this way and it makes things more complex. You could also consider having the deep freeze on a timer to run every few hours at night or if the liquid gets too warm only (better). That'll use less energy than an aircon. I bet you'd only have to run the compresser for 15 minutes or so.

Could also have a variant of this. A "WAter pillow" that contacts your lower back or head. That could get chilled by your system. Less water so your system will last longer before running out of cooling capacity ...

*** edit - Ok still brainstorming ***
You could also hook up just PVC to feed your "water bed" on one side the coolant directly, then on the far end another pvc pipe (both insulated) to return the coolant to the Deep freeze system. Cycle every x amount of time with low power pump (outside of your room). To make sure you don't get hypothermia, just pump say 20% of mattress volume in slowly (and have blankets between mattress and yourself ha). This could get dangerous depending on your temps so we'd have to address those things later (outside of brainstorming time).

 

[Sparky]

I like where you're going with this... Since I'm stationary, the weight of a water mattress isn't a deterrent.

 

[Jeremiah]

I like where you're going with this... Since I'm stationary, the weight of a water mattress isn't a deterrent.

I think we cross posted as I added something so may wanna reread in case it gives you more ideas.
I'll be very interested to see how this goes for you. If it works out share your designs please!

 

[Sparky]

yep. responding to your edit, this would have to incorporate a thermostat to selectively run the pump. It's no surprise that a waterbed usually has a heater, so instead of an electric heat pad,

I could run temperature controlled water through a coil under the water mattress, as you suggested, and the water could be controlled with a thermostatic mixing valve, selectively drawing water from my water heater or my water chiller (which I assume at this point would be a deep freezer)

 

[Frank in Thailand]

I've done a lot with cooling.

2 years ago I had mining farm, 18 KW 24/7 on heat produced.
And I used less then 1 kw to cool it all.

Evaporative Cooling Pad/ Hydropads / swamp cooling / hydro
cooling / dew point cooling / wet bulb cooling all names for the same thing.

The biggest problem with Evaporative Cooling is the moist air. That isn't nice to work or live in.

But it does cool the air effective.

In my house in Thailand I have 5 * 9000 BTU air-conditioning.
Normally way to low capacity for the space it cools (3*8*2.7m)
They are 23.6 SEER, really energy efficient. (Energy safe mode 550 watt, boost 850 watt, 11500 BTU)
And I make them even more energy efficient to pre-cool the air with misting system (0.1mm nozzles)

The condenser until are outside and build to withstand some water / moist.

With temperature over 42 degrees Celsius this week, pre- cooling is almost a must.
It saves 200 watt per air-conditioning for little water.

The white tube is the misting system.

This is not new.
Many people do this already for years, mostly seen to "boost" older air-conditioning to be effective enough.

My Evaporative Cooling pads are 60*60*25 and would be more water efficient, but difficult to make for the air-conditioning condenser unit.

Heat radiates.
Cold doesn't (not that much)
Simply do cold water via the radiator is not effective.
Naturally, it will cool a little, but compared to the hot water radiator...
Like nothing.
Unless you use ventilator...
And then it's not so energy efficient anymore.

If you are grid connected, you can sell back to the grid, and buy your own energy at higher costs in the night.
Probably a few cents per KW.
That is many, many years before you earn this back!!

For a waterbed.., yes that can be energy efficient method.!

 

[Frank in Thailand]

To make it even more confusing..

Condensing units and condense are 2 totally different things.

Your refrigerator have condensing unit
A pressure pump for the coolant.

Yes, an air-conditioning system takes away the heat by vaporising the coolant in the room what will need energy to do so, and it extracts it from the air in the room.

Then this vaporised coolant need to be brought back in liquid shape, that happens outside the room (for split air-conditioning system) and there the heat from the room is released to the southside air.

By nature hot air can hold more moist then cool air.

Inside the room, during the vaporising process, the air is cold and the moist will condense on the heat/cold exchanger.

That an air-conditioning also reduce the humidity is normally a side effect.

Dryer air feels cooler as your body can more easily lose excessive heat by sweating, what vaporise more easy in low humidity....

Many buildings with air-conditioning have special system to increase the moist in the air as it is too dry.

Condensation will be a problem if you only use radiator.
An air-conditioning system have special condensed water line, and if you ever watched it...
It's condensing a lot!!
Liters per hour in moist environment.

I'm sure you don't want this in your floor

 

[Frank in Thailand]

OK, that was mostly incoherent. You're the main reason I'm leaving this forum and never coming back. Where were you when I wanted some help picking circuit breaker sizes?

You could have asked me

 

[Frank in Thailand]

yep. responding to your edit, this would have to incorporate a thermostat to selectively run the pump. It's no surprise that a waterbed usually has a heater, so instead of an electric heat pad,

I could run temperature controlled water through a coil under the water mattress, as you suggested, and the water could be controlled with a thermostatic mixing valve, selectively drawing water from my water heater or my water chiller (which I assume at this point would be a deep freezer)

Why do you want to mix?
The heating pads normally used also don't mix water, just heat up.

You could make PVC tubing like they do for floor heating and simple have the bed on top.

I assume you don't want to cool or heat fast as short time makes it energy intensive.

As it's being cooled 24/7 with the pre-cooled liquid (pre-cooled by the excessive energy from the solar panels) refrigerator or other systems, a temperature controlled pump is all it needs (having the coolant flow in slow or faster rate)

I can't help about thinking that an ice cube maker probably is the most efficient way of making the cool buffer.

Melting ice needs more heat (coming from the waterbed) then a cold mass would.

I have no Idea about their costs.

 

[OffGridIdaho]

My neighbor installed a radiator of pipes into the ground. Gives him heat in winter and cool in summer for free. His house is passive and has 12" insulation r100. He never has to use heat or ac in the house.

 

[Supervstech]

My neighbor installed a radiator of pipes into the ground. Gives him heat in winter and cool in summer for free. His house is passive and has 12" insulation r100. He never has to use heat or ac in the house.

Where does he live? What design of pipes? That sounds impressive