My fan-coil active thermal mass greenhouse is finally operating!

After many, many threads on here kicking ideas around for greenhouse plans that never came to anything, I've now got a 9' by 16' partially buried hoop house in the back yard, with 420 gallons of water in flat wooden tanks lined with EPDM pond liner. The water is pumped through an old car radiator which is connected to ductwork and a fan. The idea is that the water will absorb heat during the day and then release it at night, damping out all the temperature swings and pulling the house back toward the average temperature. I'll be raising vegetables and some Mediterranean perennials in it; right now I've got cool weather vegetable starts going.

So, the above is the idea. Reality is sometimes a little different. Yesterday a cold front moved in, and it has been almost twenty-four hours since the greenhouse had any thermal gain, and the thermal mass is not fully charged. Outside the temperature is 32 F. Inside it is 42 F. I'd prefer if it stayed at 50 F. What will happen in the next few days when the temperature drops to 20 F outside, still with no thermal gain?

I'm hoping that at a certain point the tanks of water and the radiator put a "floor" under the drop in temperature. If this does not happen, one of two things may be wrong. Either there isn't enough heat storage, or that the fan is inadequate to move the heat out of the tanks fast enough. In the one case, I need a bigger fan, in the other, more tanks. I suspect that the fan may be undersized. Of course, I could solve either problem with more insulation. Also, the thermal mass is not fully charged yet.  One thing that hampers me in figuring out what the system is doing is that I don't have a floating thermometer in the tank.

The fan is this one: https://suncourt.com/products/inductor-corded-in-line-duct-fan-db6gtc?_pos=1&_sid=2c50db266&_ss=r I had it laying around.

The north side of the hoop house has bubble wrap insulation installed between the plastic and the hoops. The whole structure is dug into a slope. Pictures coming shortly to help you understand the system.

Here you can see the fan and radiator sitting on top of the tanks, which are up on blocks to isolate them thermally from the ground. You can also see the plastic bubble wrap to the left of the picture.

And here is the pump, with a valve to control how much water it puts out. It pulls water out of a bucket full of holes wrapped in landscape fabric to filter out any floating junk.

Nice set up Gilbert!  How does that fan duct connect to the radiator?  I'm suspecting that the fan is undersized compared to the radiator.  I mean, imagine how much air a vehicle's radiator fan pushes and compare it to that duct booster fan.  

The tank may still be too small or big relative to the radiator as well.  I'm thinking if you get the fan and radiator matched, then it's just a matter of monitoring water temps and seeing how quickly the water temp bottoms out on a cold night.  If it drops near to room temperature before dawn, the tank probably needs to be bigger.

Where'd you get that pump?

There is a box of sheet metal connected to the radiator with screws and sealed with aluminum metal tape, with a connection to the flexible metal tube.

How many cubic feet a minute do you think I'd need? How many times an hour should I turn over the greenhouse air? All I can find about my current fan is this: "180 CFM free air to Max Boosted Air 350 CFM." Would that mean it is 170 CFM?

The pump was got for twenty dollars used; it is a Pan World Magnet Pump.

https://www.amazon.com/Pan-World-150PS-Magnetic-Water/dp/B001EHGE58

So the air duct goes into the top of the box and then turns 90 degrees and goes through the radiator... right?   I think that fan puts out 180 cfm in the open and can spin fast enough when in a forced air duct to help up to 350 cfm.  So I'd guess you have 180 to work with.  Ridges in the spiral duct probably take a bit off of that total.  Going through the radiator takes a lot.  90 degree bend into a big box takes a few bits.  Curiously, air going into a duct without a tuba shaped flare on the inlet also reduces its capacity a bit.  My guess is that you're effectively getting 60-80 cfm from that installation.

To test a bigger fan, how about taking a standard box fan and sticking that on the radiator to push air through it.  Then you can play with three speeds.  I think the test for fan speed (and pump gph) is to see how much heat you are taking out of the water as it goes through the radiator.  

For instance, let's say the room is 40 and the water in the tank is 60 degrees.  If you're running the pump and the water comes out of the radiator at 55 degrees, then either the fan is too small or the pump is too big or the radiator is too small (doubtful).  If the water comes out at 43 degrees, then the fan and radiator are taking all the heat out of the water and the pump could be a bit bigger.  If you have a box fan on there and at low speed the water comes out at 56, at medium it's 54 and high it's 51, then you either need a huge fan or the pump is a bit too strong.  

There are diminishing returns.  No need to get the water down to 40 when it comes out of the radiator.  In fact, I'd be happy if it came out at 50 because then the air that's heating the greenhouse is 10 degrees warmer than the existing air and you'd actually make a difference.

Once that part of the system is balanced, then the only remaining variable is the size of the water tank.

Hopefully this makes some sense, it's late and I should be in bed already....

Thanks for the advice on figuring out how to size things. The pump has lots of spare capacity, so I could easily increase that variable, but I don't think I will need to. Using a box fan would be a good idea, except that getting the ductwork attached to the radiator was difficult, and I'd rather not pull it apart. I might just get a bigger duct fan. The interior area of the greenhouse is approximately 800 cubic feet. How many times an hour should all the air go through the radiator?

Overnight the outside temperature dropped to 27 degrees. Inside it dropped to 34 degrees. Interestingly, without any solar gain so far, and before the outside temperature rose, the inside temperature went up to 38 degrees. There was a heavy snow and I left it banked up around the greenhouse, which may have added insulation and allowed the fan-coil to catch up.

I don't know if there's an ideal air exchange rate for the radiator and greenhouse.  There are for houses but that's a whole different ball of wax.  I think first I'd try to maximize the amount of heat you're getting out of the water by tweaking the fan.  Then evaluate if you have enough water and/or if the heat you're putting into the air isn't enough.  

If you run out of heat before morning, you'd need more water.  If you still have heat in the water in the morning but the greenhouse is too cold, you need to increase your heat delivery.  Thus more gpm and more cfm and hopefully not a bigger radiator.

Or at least I think that's a logical progression.....

Side note, as a test you may be able to put a box fan on the outlet of the radiator and suck air through instead of pushing it.  But you'd still need to remove the inlet box since that would restrict it significantly...

Great build! I've been thinking about a system like this for our greenhouse (propane heat).

You've got two different goals: Excess heat to water storage, and stored heat back to the air.
If you think about just the excess heat to storage first, if the fan is underperforming the excess heat will build in the greenhouse as the fan struggles to keep up and/or the pump fails to supply cooler water to absorb it.
Ideally, I think, if thermostatically operated, the system would cycle on/off as excess heat built up and was successfully removed.

Another thing to think about is managing the water temperature(s), and you said you have tanks (looks like two? maybe the one under radiator spills over into the one with pump).

Keeping the hot/warm/cool waters separated, improves the efficiency of the heat exchange, by having the biggest Delta T.
Ideally:
When storing heat, you take the coolest water to the radiator, and return it (heated) to the hot tank, which overflows to a warm tank, and again to the cool tank.
When using heat, you take the hottest water to the radiator, and return it (cooled) to the coolest tank, which overflows to a warm tank, and again to the hot tank.

A home water heater tank does something similar by stratifying the temperatures vertically with fresh/cold water entering the bottom, while hot water exits the top.
This is a case where the system is maximized for one direction, heating domestic hot water. (there actually is no reverse...)

You look like you are doing great with yours, since you have it (if I am seeing it clearly) maximized for heat storage.

Mike Haasl wrote:If you run out of heat before morning, you'd need more water.  If you still have heat in the water in the morning but the greenhouse is too cold, you need to increase your heat delivery.  Thus more gpm and more cfm and hopefully not a bigger radiator.

Or at least I think that's a logical progression.....

Side note, as a test you may be able to put a box fan on the outlet of the radiator and suck air through instead of pushing it.  But you'd still need to remove the inlet box since that would restrict it significantly...

I like the box fan idea, Mike, but the inlet box is necessary to pull from the hottest air at the peak. It could be reworked. The 90* elbow = 10' of straight pipe, flow would improve greatly without that.
Could use a big cardboard box (appliance?) to make the inlet box reach UP to the peak, eliminate the duct all together...

Thanks for the suggestions everyone!

Mike, if I attached the box fan pulling air through the front of the radiator, and left the little duct fan in place, I would boost the airflow considerably, right? That way I could test more airflow without pulling everything apart.

Kenneth, I think I'm doing what you suggest. I have three tanks; one under the radiator at the end of the greenhouse, two lined up along the wall. The pump pulls water out of the farthest tank from the radiator, pushes it through the radiator, and the radiator dumps into the tank underneath it. Then it flows back into the other tanks.

The cardboard box would work except that the greenhouse is really wet from condensation! Maybe a sheetmetal ductwork?

Here is a picture of the tanks under construction, showing how they flow one to another. Also some flats of soil blocks.

Another note on the construction; the greenhouse hoops are chain link fence top rail. I tore down an unneeded fence and bent them into arches by driving stakes into the ground and feeding the pipe through a gap between two of them a few inches at a time, bending it little by little. They are quite strong and have stood up to a lot of snowfall. In the above picture you can also see the automatic vent arm.

Gilbert Fritz wrote:Mike, if I attached the box fan pulling air through the front of the radiator, and left the little duct fan in place, I would boost the airflow considerably, right? That way I could test more airflow without pulling everything apart.

Yes, I think that would be worth trying (with both fans running).  Try to gauge how much air is going in now and then see if it's appreciably different once you rig up the box fan.

Gilbert, Nice to see the construction shot! It wasn't obvious that there were three tanks, that's great!
Maybe just a duct made from wood strapping and scrap poly sheeting? or thin plywood? Or even a big contractor garbage bag with a hole at the bottom/side?
Agreed the cardboard won't survive long, but maybe long enough to determine if it is the right approach, before investing time and materials... just a short detour on its way to recycling, or mulch...

Ideally that auto-vent won't open if your system captures all the excess heat!

Right, if the system works the auto-vent is redundant. They don't really slow down the temperature rise that much in any case.

So I went and checked the temperatures on the air and tanks. Last night the outside temperature dropped to 17F, with a clear sky and snow covered ground. The greenhouse indoor low was 30F, and by noon had come back up to 56F. The inlet/ cool tank temperature was 42F. The outlet temperature from the radiator was about 46F (Difficult to take exactly I ran the outfall over the thermometer, which could lead to a lower than accurate temperature, and I ran some water into a bucket and put the thermometer into it, which might lead to a high temperature, and I averaged them. Anyway it is between 45 and 48.)

So what would that mean for my system?

I actually think it is working pretty well with the tank temperature so low and so little thermal gain. Would more air flowing through actually gain me that much? The air was twelve degrees warmer than the water, and the water gained about four degrees. I'll try to take more mesurements later today when it should be warmer inside.

I just checked again. Greenhouse air temperature is 62F, outdoor air 60F, intake temperature 47F, outfall temperature 50F. Mostly overcast right now.

I'm wondering if I actually have a fairly efficient, maybe even overkill, system for the size of my greenhouse, but the thermal mass is still cold from the tap. (Tap water is 46F right now.) I put the water in about two weeks ago, but the system has only been circulating for a few days.

Thanks for the pics. Great setup! I appreciate the amount of work involved.

Duct fan ratings are very misleading IME. I think it's the fan design, which is intended to operate in an already-powerful air flow. By themselves, they barely move enough air to blow out the candles on a cake. Bathroom exhaust fans, on the other hand, move a lot of air, and if Energy Star rated don't suck a lot of juice.

Would it make sense to put plant seedlings on racks right above the tanks when it's cold outside? That's probably the warmest spot, right over the thermal mass. With a mini-hoop setup over top, using plastic or frost cloth, I think you'd concentrate the available heat. You could even direct the blower through the hoop tunnels.

A final thought: It occurs to me that the ultimate no-tech radiator in a greenhouse would be a wide, slow waterfall, which can be run by a very low-powered pump. Also nice to listen to.

Hi Douglas,

You are probably right about the warmest place in the greenhouse being right above the tanks, at least a night. I plan to deck the tanks over and put flats of plants on them. On the other hand, during the day all the mass of cool water might actually slow things down. I'm using an improvised inner layer right now, putting bubble wrap over the seedlings to keep in ground heat.

And thanks for the advice on the duct fan. I'll see if I can find a bathroom fan.

I wonder how efficiently the waterfall idea would work to transfer heat. It would be an interesting thing to measure.

A caveat on air temperature readings; my thermometer is sitting on a sheet of insulation in the shade of the Southern wall, probably the coolest spot in the greenhouse and out of the sun. The air intake for the fan is at the peak of the roof in the sun. I'd guess the intake air is considerably warmer, which means my system is less efficient than the numbers above might indicate.

So when the greenhouse is clearly warmer than the water, the system increases the water temperature by three degrees across the radiator?  I'm thinking it isn't harvesting as much heat as it could.  

As I think about it, collecting heat during the sunny hours (5-8 hours?) is as important as delivering it during the cold hours (12+ hours?).  So you can test it both during the day and at night to see how it's performing (as you are).

I think where you're sitting now is that either the fan is not moving enough hot air through the radiator (my suspicion) or the pump is moving the water too quickly through the radiator.  

Would it ultimately matter if the pump moved water too fast? Wouldn't five gallons of water raised four degrees come to the same thing as ten gallons raised two? (Since either volume will be mixing with a much larger volume of water; wouldn't they carry the same amount of heat? )

I think you're right in that respect.  I think I was trying to come up with reasons why it's only picking up 3 degrees of heat and those were the two.  

Assuming the pump speed doesn't matter for total heat transferred, it's still a variable that can confound our analysis.  So leaving it alone and playing with the fan is a good way to go.

Don't get me wrong, it's great that you're storing and delivering heat!  Now let's figure out how to store and release so much heat that you either have to get a bigger tank or the greenhouse never goes below 50

Could I overdo it and get a greenhouse that never goes above sixty?

I'm looking for a heavier duty fan.

I doubt you could get enough thermal mass in there to accidentally keep it that cold year round.  If your system actually got so efficient it was sucking all the heat up during the day (when sunny), you could just run it for a few hours instead of the full daytime period.

Last tank measurement for the day; the intake was 52F.  So I raised 420 gallons of water ten degrees, or roughly 33,000 BTUs stored.  Obviously not enough to keep the temperature at 50F when the outdoors is 10F!

Sweet, that's great data to have!

If I added another inline fan to the current setup (in addition to the fan I've now got), how many of its rated CFM should I expect to actually get? Most of them?

I'm not sure but I suspect two together would move about 20% more air than one.  Those fans aren't all that powerful and then if they're put in-line with one another, the turbulence leaving the first fan would affect the second one.  And the restrictions of the entry to the box by the radiator plus the radiator fin resistance is just a lot to overcome.  I think...

Gilbert Fritz wrote:After many, many threads on here kicking ideas around for greenhouse plans that never came to anything, I've now got a 9' by 16' partially buried hoop house in the back yard, with 420 gallons of water in flat wooden tanks lined with EPDM pond liner. The water is pumped through an old car radiator which is connected to ductwork and a fan. The idea is that the water will absorb heat during the day and then release it at night, damping out all the temperature swings and pulling the house back toward the average temperature. I'll be raising vegetables and some Mediterranean perennials in it; right now I've got cool weather vegetable starts going.

So, the above is the idea. Reality is sometimes a little different. Yesterday a cold front moved in, and it has been almost twenty-four hours since the greenhouse had any thermal gain, and the thermal mass is not fully charged. Outside the temperature is 32 F. Inside it is 42 F. I'd prefer if it stayed at 50 F. What will happen in the next few days when the temperature drops to 20 F outside, still with no thermal gain?

I'm hoping that at a certain point the tanks of water and the radiator put a "floor" under the drop in temperature. If this does not happen, one of two things may be wrong. Either there isn't enough heat storage, or that the fan is inadequate to move the heat out of the tanks fast enough. In the one case, I need a bigger fan, in the other, more tanks. I suspect that the fan may be undersized. Of course, I could solve either problem with more insulation. Also, the thermal mass is not fully charged yet.  One thing that hampers me in figuring out what the system is doing is that I don't have a floating thermometer in the tank.

The fan is this one: https://suncourt.com/products/inductor-corded-in-line-duct-fan-db6gtc?_pos=1&_sid=2c50db266&_ss=r I had it laying around.

The north side of the hoop house has bubble wrap insulation installed between the plastic and the hoops. The whole structure is dug into a slope. Pictures coming shortly to help you understand the system.

Gilbert, Great idea!!  

(but was wondering if you are any relation to Red Green? TV show.)  ;-)

Great idea and a very interesting post. I just put in a much smaller greenhouse with a stone thermal mass at the bottom.
I also installed a small hose to keep water spillage down, but it leaked.
I think the extra evaporation was 'cooling' the stones while they were wet.
How much do your open tanks cool the air from evaporation?
I don't know the physics, just curious about what cooling affect open water sources have.
Thanks for sharing your trip!

Dennis, I don't think I am . . . but I'm very, very fond of duct tape! We had to remediate our house for mold and water damage, and we used duct tape to build containments and tubes out of plastic for the negative air containment and blower. And an earlier greenhouse I built was held together with duct tape. I've since found that plumber's tape, or plastic strapping, is almost as handy, at least for outdoor projects. This greenhouse is partially held together with that stuff.


Mark, I don't know how much effect the open water sources have; I can't separate it from the effect of the fan-coil. It is a good question, though. Once it is summer and I don't have to worry about freezing the greenhouse, I'll probably do some experiments where I shut the system down to measure conditions over a few days with and without it running.

And I'm still planning to upgrade the fan.

Have you thought of or are planing to use the tanks as an aquaponics system with floating mats to hold the plants?
Could you run the hose from the pump to the radiator, through your earth mass on one side to sore heat in that and provide bottom heat for your plant starts?

Hi Hans,

I have thought about that. For now I'm just going to put some plywood over them to hold flats, since the current level of complexity is about as much as I can handle. But I'm hoping to go to aquaponics/hydroponics at some point. And yes, I've thought of doing other things with the pump and hose; for one thing, I might connect an external solar collector panel to gain more heat.

I've got a 500 CFM fan on the way (if it does not get stuck in a warehouse somewhere!)

Seedlings and salad greens are growing nicely.

Two equipment semi-failures to report. First, I'd been noting the daily high and low temperatures. Some days, I noticed that the nightly low was the same as the low outside; and that struck me as unlikely. The tanks were loosing ten degrees or more overnight. The sun had been heating the soil all day. There was a layer of plastic to trap air. Even if my fan was a total waste of time, the temperature should be a little higher . . . and some days, it was quite a bit higher! I think my max/min thermometer is a bit off. It is an analog, not a digital, and when I brought other thermometers in I got a range of temperatures covering about ten degrees. To further complicate matters, my outdoor thermometer is perhaps not quite reliable either . . . so I really don't have any way of figuring out what is happening. But, this morning, I went out before the sun had time to hit the greenhouse. It was a chilly morning. Inside it was noticeably warmer, with a soft, damp feel. So it is working somewhat, anyway; the overnight low is higher inside.

(This does not effect the tank measurements; since I use the same thermometer on the inflow/ outflow I can get accurate relative temperatures to see how efficient the fan-coil is, even if the thermometer is off.)

Second, even though I tried to wash out and off the radiator really well, it has contaminated the water with some sort of oily slick. I sort of expected that; I figure after it has run for a year or so I'll change all the water out, and in the meantime I won't be growing anything in it.

I had the same problem with my high/low thermometers. I finally had to gather them all together and chose to use the digital thermostat and a new digital as the standard. I was able to adjust the scale behind each column to get them consistent. I still find the low inside and out for air to be close to the same on clear nights with high radiant loss and high wind loss. The apparent warmth however is quite different. Soil and water barrels were still in mid to high 50's even though air reached freezing there would be frost outside but not inside.

Hans, that's good to know. Now that I think about it, the nights with temperatures nearer the same were on clear nights. The feel is certainly different, though.

One thing I'm looking into is a big insulated cover, something like an insulated pool cover, and some sort of system whereby I can roll the cover over the greenhouse on cold nights.

It will be interesting to see how the new fan changes things.

Last night, the outside temperature dropped to 20F after a cold, cloudy day with no thermal gain. The greenhouse only dropped to 34F (and if the max/min thermometer tracks 4 degrees too low, as I suspect, only 38F.) The tanks dropped to 44, from a high of about 68F two days ago. I think the superior performance during this cold snap compared to last time is due to a much higher starting temperature on the tanks. 
The better fan is on the way, we'll see how that changes things. One thing I am realizing is that these tanks are going to have HUGE daily temperature swings, which probably eliminates the idea of using them to raise fish, unless I add a lot more tanks to damp the swings on that side of the system. Water plants are still a possibility, though. 

If you're ready for a crazy idea...  Phase change materials.  They store/expend a ton of energy as they freeze/thaw.  That's quite useful if the material happens to freeze/thaw at a useful greenhouse temperature (say 50 degrees).

For reference, I think coconut oil's phase change point is around 76, glycerin is around 63 and water is around 32.  There are likely other ones in the 50 degree point.

How would this help you?  If you can get a bunch of this magical material and seal it up into a bunch of plastic or aluminum bottles, you can chuck them in your water tanks.  So as the tanks heat up, and they try to pass the thawing point of your 100 "ice cubes", the energy is soaked up by the phase change material until they all melt.  Then the tank continues to heat up.  In reverse it's the same way.

So the hundreds of bottles act as heat batteries that all act at one (hopefully useful) temperature.  And since they're just sitting in your tanks, the plumbing doesn't need to change.  The phase change energy of water to ice is about 100x the energy it takes to get water to go up a degree.  Other phase change materials may be closer to 60x.  So if you can put 30 gallons of glycerin bottles in your tanks, they'd have the effect of adding 1800 gallons of water tanks to your greenhouse.