I’m planning to construct a home sometime in the near future in hot, arid part of the country. Obviously, keeping the home cool is a major concern and I’ve been considering all available options.
One of the recurrent ideas discussed online is using geothermal cooling. But I think I don’t have enough land to implement it. I have a related idea though.
Water supplied during set hours of the day by the municipality is fairly cool even during the summer months and it is a common practice around here to first let it collect in an underground tank and then pump it to the overhead tank as needed.
What if I create a closed loop system of circulating water with two car radiators: one in the underground tank, submerged in cool water and another in the living area of the home with a fan blowing behind it.
Do you think it’ll cause perceptible change in room temperature if there is, say, a temperature difference of 10 degrees Celsius between water and ambiance? I intend it to only reduce the load on the air conditioners and/or just delay the use of AC.
What kind of math/physics will be needed to assess if this is a feasibility? I tried looking hard but could not find anyone else discussing this idea, so I’m reaching out to you guys.
Thank you for your time.
I’ll preface with my qualifications, so if a more qualified person comes along you can disregard me. I’m an engineer who has taken a few thermodynamics courses and has worked as an engineer for a hvac manufacturing plant. I’ve never done anything strictly related to geothermal, but I’ve read a decent bit about it (and watched Technology Connections’ video on the subject, it’s a good entry point)
You may want to call up a company who does geothermal cooling and see what options you have, they’ve gotten pretty creative on how to bury the cooling lines. (See the video mentioned before)
Going the route of just sticking a large water tank underground probably won’t do a ton. I expect that you will have a poor surface to volume ratio, which means poor heat transfer, which means you’ll saturate your thermal mass fairly quickly. What this may allow you to do is run your HVAC system during the night/morning when it’s much more efficient, and ‘charge’ your thermal mass for the hottest part of the day.
Assuming you use 300kg of water in a day, and you can get a 10°C delta, my very rough back of the napkin math says you’re only going to have about 3 kWh of cooling from just the cold water, which is a decent bit, but it’s not a ton. Best case scenario you cut your cooling needs by around 10-20%.
I’m too lazy to do the math of the heat exhange with the ground, but my bet would be you’re better off spending any money you have set aside for this on better insulation techniques and/or a proper geothermal cooling system.
I do like your creative idea though
Thanks, hoping that your back of the napkin math is in the correct ballpark, 10-20% lowering of cooling bills sounds very lucrative to me.
And I owe a clarification after reading your analysis since I’m not hoping to achieve any geothermal gains by heat transfer between water tank and ground. I’m assuming that the water will heat up as it cools the room down. But since fresh water is supplied periodically and this fresh water is relatively cool, it will keep the cooling cycle running. I will try to build the tank in shade and isolated from direct heat of the sun as far as possible.
As regards your point on the budget, i don’t think that two car radiators, some insulated tube and a small water pump will cost much, more so if I can run the pump off a small solar panel in the day time.
All said and done, if there is nothing in theory that advises me against trying this, i would like to give it a shot and document it so that someone else might learn from it.
Thanks again!
Engineers describe heat transfer with a “heat transfer coefficient”, and the rate of heat transfer is this coefficient multiplied by the temperature difference. So you can calculate what the heat transfer coefficient must be by measuring room air temperature initially, water temperature initially, and then running your system for a little while and measuring the room temperature again. The smaller room area you can cool the more accurate this will be. You will need to look up heat capacity and density of air (easy to find), and the temperature change of the air with the volume of the room and the temperature change will together give you an amount of heat you removed from the air to the water. Simple!
While I have a background in science, this is not the specialization i chose and so almost all of it is almost Spanish (won’t say Greek, since at least I’ve heard of the words) to me.
I thank you for summarising the science of it though. It gives me enough keywords to start learning.
Well there is plenty of reason to not do it, but I’m assuming you’ve thought about what tampering with your water supply means (and that car radiators are not food safe, and could contain lead or other nasty metals). I think it goes without saying you’re also running the risk of leaks, a high water bill, and mornings without water if your system has issues.
I’d also like to cover my ass a little and do the typical engineer thing and remind you that an idealized number like this is never realized. You will have to account for losses due to inadvertent and incomplete heat transfer. But you may also get a higher reduction due to the ground heat exhange, which I am still too lazy to work out. ¯\_(ツ)_/¯
All of that said, I really hope you succeed at getting cheaper cooling and have fun doing it. There are houses which use water piped through concrete flooring to regulate temperature, so you’re not very far in uncharted territory. Doing big projects like this are also a really good way to learn a lot very quickly.
I also still highly recommend getting a quote for a reputable geothermal cooling system. If nothing else but to brag about how much you saved by DIYing
