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.
You’re describing a fancoil supplied with cool, regularly replaced, municipal water (normally this water would be a fully closed loop cooled with an air source or ground source heat pump). Your energy needs will just be a circulation pump. You’ll probably notice a little cooling but it depends on how cold the water is, the surface area of the radiator, and the flow rate of the water. It has the advantage of being low maintenance so give it a shot and perhaps build it in a way you can access the components and improve / experiment over time.
Look into an approach / methodology called Passive House. Passive House focuses on making buildings that have near zero heating and cooling load. If you get the math right / design from scratch with this in mind you can make a Passive House in nearly any climate. Common modern single-family-home building techniques are generally not at all closely aligned with building a Passive House.
When trying to keep a house cool, here are the things I would focus on (in order of priority):
-
Reduce solar heating impacts: either place shade trees or awnings to block direct sun on the entire structure (or the windows at a minimum).
-
Build a highly-insulating enclosure (~R30 walls and ~R50 roof at a minimum, but you could push that further). If you are set on building with lumber you still can, you could building an offset double-stud wall filled with insulation, and of course an appropriate amount of exterior insulation a well. The goal in addition to insulation quantity is to reduce thermal bridging. Consider a “simple” house layout. Avoid too many corners / details / flourishes that add construction complexity.
-
Utilize free-cooling first: as your first stage of cooling, open large windows close to the ground and open clearstory windows in the roof / top of a stairwell or similar, it really depends on the layout of the home (and ideally the layout is design around this concept). This allows the heat to be drawn out naturally via convection. Include ceiling fans for comfort. This approach will work until outdoor air temperatures get quite high. Once free-cooling will not longer work
Once free-cooling will not longer be effective you can transition to mechanical cooling. Close all windows and cool your space either a high-efficiency air-source heat pump (and / or your free-cooling municipal water fan coil).
-
Similar to the design methodology to encourage natural air / heat flow out clearstory windows or “solar chimneys”, also consider just having higher ceilings where heat can pool but you won’t feel it. Your exhaust should pull from these areas.
-
Dedicated outdoor air system (DOAS): don’t design your mechanical ventilation system to cool using air (aside from the free-cooling described earlier). It’s inefficient. Hydronic heating and cooling (moving heat with water) is much more efficient. That means heat pumps for heating as well as cooling. Mechanical ventilation rates should be the bare minimum, just enough for fresh air but not for temperature control. Perhaps look at flow rates included in ASHRAE 62.1 or a standard more focused on residential homes. Also, your supply air can be separately ducted to each room (not a shared trunk), each being much smaller than what you see in a “normal” house, this gives more control for every single room.
-
ERV: of course you’ll want to install an energy recovery ventilator to capture what heat / “cold” you’ve worked to produce before instead of throwing it away along with your exhaust air.
-
For heating your domestic water, get a heat pump hot water heater (with tank). Instead of making heat it takes heat from the surrounding room and puts it into your domestic water tank. That means it “outputs cold” into the surrounding room, the opposite of a gas or electric resistance water heater.
-
Earth tubes: to naturally pre-condition your supply air by running it through the ground first. Another form of free-cooling but useful when the house is “buttoned up” because outdoor air temperatures are too high. This is when you’re only supplying minimum ventilation air.
-
Limit the things in the house that make heat. Efficient refrigerators / freezers (see energy star website), computers that are no more powerful than what you need, etc. Place these things in areas where the heat won’t bug you as much.
Hope this helps.
Holy hell, how do you know all this?
I can’t thank you enough for taking the time to write all of this. I have come across most of these concepts and have been considering them with respect to the options I have.
I’m trying to design the house considering the natural air flow and sun angles. There is a garden with trees on one side of my plot and I intend to make use of it while also planning for other sides. Currently I’m making house plans on my own and discussing with some people I know. This is helping me understand what we want/need, or not, in our home. At some point I’ll definitely take it up to the professionals. There used to be a community of floorplaners and home builders on Reddit but it was not much active.
Stone is the cheapest construction material in my area but it practically converts the house into a furnace unless the walls are very wide. I plan to avoid it at all costs. On my radar is a technology called EPS (Extended PolyStyrene) Core panels which are nothing but high density PolyStyrene with a galvanized steel mesh in it. It is supposed to be load bearing and good enough to build the roofs along with walls. It has been used in some countries at various levels of success though not in this exact form (with wire mesh). It is supposed to be highly non conductive in terms of heat and so good for maintaining inside temperatures with minimal cooling. Most of the material available online are either research papers or companies selling it, both of which only have the incentive to talk good things about it. I intend to go and meet people who have actually built using this technique and take their feedback before I commit to it.
As regards this pet idea of mine, I think the consensus in this discussion is that it will not cause much difference. But since it does not involve much effort/investment, I’m going to try it anyway.
Another cooling technology that has my attention is desiccant based cooling. That too has some research papers and a nice set of videos by Tech Ingredients (https://www.youtube.com/watch?v=5zW9_ztTiw8). It looks more daunting and more suited for humid regions/months.
I keep dreaming about a setup with evaporative coolers for dry spells and desiccant based thing for humid ones, assisted with this pet tech of mine.
I hope I’ll have the courage to implement these ideas finally.
Thanks again.
-
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!
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
I’ll definitely take your advice regarding getting a quote from the geothermal system guys, if there are any around me.
I’ll also try and learn the science/maths to see if you, or the other guy who has suggested that this might not be worth the effort, are right.
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.
No, unless you are leveraging evaporative cooling, that amount of circulation isn’t going to get you much.
Just get a real geothermal hvac system if you have the opportunity. Incredibly efficient.
Back of the napkin conversion: 20btu/sqft recommended cooling capacity. 1btu = 252 calories (small)
A 60k btu cooling needs
15120000 gram degrees C of water. Assuming you have perfect heat exchanger on both ends, that’s 15120 liters-degrees circulated per hour.
Pumping that much water alone is going to be quite a bit of energy.
Then you have the problem of heat exchanger. There are lots of sizing mostly based on the deltaT temperature difference.
Realistically, without some agent evaporating and recondensing, you’ll have a massive water to air heat exchanger that’s not practical at all.
If you want to do more research yourself, heat exchanger sizing can be found in mechanical engineering and chemical engineering handbooks.
Thank you for your reply. It has given me the factors to consider and the terms to look for. I’m going to cross verify the values you have quoted and if it is indeed as bleak, I’ll just temper my expectations.
But I’m not giving up on it totally yet. What I’m still not sure is how geothermal cooling can provide more heat exchange as compared to the system I’m proposing. At most geothermal systems also only have a few tens on meters of piping underground.
You should look for cross flow heat exchanger formulas, water / air. Sorry I can’t be of more help but it’s been a while since I studied this.
Thanks for mentioning the terms that I should look for. I never had the opportunity to study these and having these as pointers helps a lot.
Sounds similar to the concept of a swamp cooler to me; maybe look up how they calculate efficiency?
Also, swamp coolers are only effective below a certain temp, so you’d need to think about bypassing (cutting the fan and turning on air con) above a certain threshold
I think it is different from swamp coolers since swamp coolers constantly add moisture to the air in the room which is okay for arid conditions but does not work as the air increasingly becomes saturated with moisture. We use swamp coolers when we can in the initial part of the summers. But they become useless once the rains arrive and the air turns humid.
In the contraption I’m suggesting, does not expose water to the air inside the room. It will work purely on heat exchange principle.
You should look for cross flow heat exchanger formulas, water / air. Sorry I can’t be of more help but it’s been a while since I’ve seen any of this.
This will work, in theory, and if you’re willing to use a lot of water. It’s probably a bad idea.
Heating one kilogram of water by one degree Celsius without phase transitions (freezing/melting, evaporating/condensing) takes 1 kilocalorie of energy. That’s roughly 4 kilojoules aka kilowattseconds, or 0.0012 kWh.
Thus, to get 1.2 kW of cooling, which is about half of what those tiny portable air conditioners promise, at a 10 degree temperature difference, you’d need 100 liters of water per hour. If water costs $0.40 per 100 liters, and electricity cost $0.40 per kWh, an air conditioner (using about 0.4 kW of electricity to pump 1.2 kW of heat) will be a lot cheaper, and that’s ignoring the power you might need to run the pumps and fan on your solution (all of which you get back as heat!)
Unless the water in the loop is below the dew point, you also won’t get any dehumidification. This is actually more important than cooling, and a big reason why air conditioned rooms feel so much better (sitting in the shade in 40° C dry weather would be unpleasant but fine, at 100% humidity, it would be reliably fatal regardless of fitness).
If you’re building new, look into:
- proper insulation
- insulation and windows that optimize for the right thing for your climate (in countries like Germany, I suspect windows are optimized to let as much heat in and as little out as possible, which saves heating costs in winter and turns apartments into hellholes in summer)
- passive cooling paint and panels - I don’t know if they’re commercially available and in a practically usable state yet.
- solar to power the AC
- swamp coolers aka evaporative cooling (the split kind that evaporates water outside). Downside is they use water (which actually is lost - evaporated), so if you’re in a drought prone area where water is restricted or expensive they might not be the best choice. Also, it has to be actually dry (low humidity) when it’s hot. Get actual, local climate data, not gut feeling. Check if there are commonly used commercial solutions, possibly combined with actual A/C (very common for industrial scale setup, not sure if common for home setups).
- regular air conditioning. I’m assuming you’re trying to build a house to live in, not an art. Economies of scale mean that going with suboptimal but standard solutions almost always beats custom hacks. If you have the same brand of AC as everyone around you, the repairman will know how to repair it, will have spares, will know how to design it so it is sufficient for your house, etc. - if you build something yourself, you will be the only one who can maintain it.
- ceiling panels - these cool the room by running cold water (generated using normal A/C heat pumps) through pipes/panels under the ceiling. The upside is that they also remove radiant heat, the room feels about two degrees colder than it is thanks to this (look up “wet bulb globe temperature” for a rabbit hole). The downside is that they can’t dehumidify and actually stop working in high humidity when you’d need them the most: if you run water colder than the dew point through them, it’d condense and start dripping all over your stuff, so it shuts down or limits how cold the water can be (and thus how much it cools). Consider them as an addition only if they’re common and installers are familiar with them.
In the end, you’re building a new building, so you now have a chance to do everything right using modern but already proven technology. I wouldn’t DIY anything critical and hard to change like this. Remember, you’re trying to find the best (likely: cheapest in the long term while meeting your reliability requirements) solution that will solve your problem. There’s a very high chance that’s simply “add more A/C and solar according to what’s locally available”. And that’s fine. There’s nothing bad about that.
I wouldn’t, for example, try to build with different materials than locally common, even if those were “better” by some metric. That often doesn’t give you a better house, that gives you a unique house, and unique can be a nightmare.
Wow! A lot of sane advice coming from an account claiming to be a shitposting account :D.
Your points are well taken and I’m slowly coming to the same realisation. Thank you for taking the time to reply.
Someone with more experience can correct me, but I doubt you’ll get enough energy transfer via the radiator in the house.
I think you’d get a lot more mileage out of an attic fan. Keeping the attic cooler during the day will slow the heating of the house.
You may also want to take a look at better roofing materials. There are roofing materials designed to reflect more of the sun’s energy as well as radiate what heat does get absorbed better than others.
I have been considering newer construction materials and the trouble is that most of the material available online is either research papers or companies selling it: both of which only talk good things about it.
I liked one such tech and intend to seek and meet people who are living in houses made using it. That should definitely assuage my doubts.
You should look for cross flow heat exchanger formulas, water / air. Sorry I can’t be of more help but it’s been a while since I studied this.
I’ve read about a system where excess energy from solar panels is stored as energy in an underground reservoir.
I’ve tried finding an article, so to see if it included your heart pump idea. I think it may be feasible, but I’m not sure how efficient the system would be.
So I’m quite interested, but cannot be of more use, hope someone gives you more of an answer.
I believe this is what you are referring to
What is happening in this process is a cavern is pressurized with air and the air pressure is later used to run a turbine.
Sort of, it wasn’t in bedrock, but in a cement resevoir.
But cheers for the source!
In one of the iterations of my idea, I thought of using solar energy to compress air and keep it stored in an underground tank. But it felt more hazardous and prone to failing. I’ll explore the link in the other reply to your comment and reconsider it. Thanks.
To be honest, that pressure storage is from the link the other guy came up with. The idea I’ve read about was about storing thermal energy in underground water, more akin to your original plan.
But I could not find the link to where I read about that, unfortunately.
