I'm pretty new myself -- nothing like these old hats here. :) However, I can let you know what I've gathered so far, plus a couple insights I came into while writing a greenhouse heat balance simulator.

Insulation: You may be surprised to learn that, if your greenhouse has any sort of insulation at all, your biggest losses will be radiative, not conductive. I discovered this while ramping up the R-value to obscene levels. Heat loss through radiation is proportional to the temperature to the *fourth* power. If it's -10C outside, and it's 10C inside, the difference between incoming radiation and outgoing radiation is about 92 watts per square meter (assuming a perfect blackbody). That's significant energy.

As a consequence, while you want R-value, it's not the most important thing. The most important things are 1) reflective insulation on the north wall (which will also give your plants more sunlight) and a good transparent IR-absorbing inner layer on the south side to recapture your radiative losses. Insulating the side walls is a good question; not too much light comes in through the sides in the winter, so I insulated mine.

Of course, the risk of a reflective north wall is that, if you don't have a lot of plants on the wall, you risk reflecting your daylight right back out the south side. So, keep that in mind. :)

Snow: My greenhouse's skin is pretty thin, but with a 45 degree slope on the top, I've never had problems with too much accumulation. Of course, with that low of an angle, and a skin that flexes, I do have to sweep it off unless enough accumulates to have it slide down.

Heating: This is going to be my first year with winter heating. There are advantages and disadvantages to both.

Electric: Needs wiring. Less thermally efficient than propane (your wall electricity comes from burning fuel, losing over half the energy in the conversion, losing energy during transmission, then you (with further loss) turn the energy back into heat). Electric heaters are cheaper, although if you don't have the wiring, you have to factor that into the cost. No risk of carbon monoxide, propane leaks, or other propane risk factors. No need to lay venting to help prevent these problems. Does not provide carbon dioxide (a well-sealed greenhouse in the winter may have a shortage. A cubic meter of air only has about half a gram of carbon dioxide in it. Of that, less than a third of that CO2 is carbon. Since most of the carbon plants incorporate into their structure comes from CO2, a complete lack of incoming lack of new CO2 means that the plants' growth will slow to a crawl). Electric heaters are often weaker than propane, although you can get sufficient heating in either type pretty easily.

Propane: Just the opposite. Needs tanks changed, although not as often as you might think, from what I've read.

How much heat you need is heavily dependent on how quickly your greenhouse loses heat and how much heat stores there are in the greenhouse. Picture it this way: if your greenhouse was simply inflated, and had no plants in it, it would lose heat almost instantly; there's only going to be a dozen kilograms of air in there, and with their low specific heat, it will hold almost no energy. On the other hand, factor in a frame, plants, and most importantly, water (especially sealed black buckets full of it), and you'll be storing up lots of energy in the daytime. The air will lose energy to the skin (and objects will radiate through the skin), but everything else will re-heat the air.

If you've insulated well, just look at how much area the heater is rated for. They usually will state how many cubic feet of air they're rated to heat.

Other notes: One of my favorite energy storage designs, which I plan to use on my next greenhouse, comes from Japan. You use a small solar powered fan with a battery pack so that it can run through the night. You hook it up to a maze of PVC pipe twisting through the ground, reaching down 1-2 feet into the soil (I was thinking of renting an auger and drilling slants that intersect to get a zigzag pattern through the ground without having to "dig"). Optimally, you have insulation running into the ground as well, straight down from the sides of your greenhouse.

What will this do? Picture how much heat capacity several cubic meters of soil (especially damp soil) has. It's a tremendous amount. During the day, when your greenhouse is prone to overheating, you're dumping the excess heat into the ground. In the evening, when it gets cold, you're drawing up the heat from down there.

It's a pretty clever design, I think. I'll be sure to post about it once I build it a few years from now. ;)

Also, keep your solar angles in mind. Here are the formulae to roughly plot the sun's altitude (height above the horizon) and azimuth (angle) at different times and different days. This will be assuming that you work in radians, not degrees:

Declination = 23.45 * -cos((DaysFromWinterSolstice / 365.24) * (2 * Pi))
TimeAngle = (TimeInSeconds / 86400) * (2 * Pi)
Altitude = asin(sin(Latitude) * sin(Declination) + cos(Latitude) * cos(Declination) * cos(TimeAngle))
Azimuth = asin(-cos(Declination) * sin(TimeAngle) / cos(Altitude))

Note that the azimuth is subject to errors from the fact that there are multiple solutions to asin. However, the solutions that your calculator generally picks are correct for during the day; it's the nighttime solutions that are typically wrong, and then you don't care about the angle that the sun is at.

Polycarbonate reflects thermal IR, so radiative losses aren't that significant. Polyethylene doesn't, and can get very cold on a clear night.

Water is more than 3 times as efficient for storing heat by volume and 7 times more efficient by mass. If you want to store heat then keep a few thousand litres in a tank in your greenhouse.

An easier way to think about the sun is to note that mid winter the sun is at and angle of (your latitude + 23), mid summer it is your latitude - 23. (This is stated in your formula, but might be a bit hard for most people to see.

They do make polyethylene with IR additives, which helps some. According to this chart, double layer IR polyethylene has a thermal transmittance close to double polycarb (though, admittedly, "SB

Here is a link that might be useful: Glazing material comparison

Molly, with regard to film, according to the link above, IR poly reduces the heat losses by 15-20% and has a favorable payback.

Any number of GH heater calculators are out there, I've included a link below. Using this calculator, assuming a double poly GH and the temps you give above, will require 34K BTU heater. Adjust this downward for a good kneewall, insulated north wall, reflective north wall surface, etc, etc.

Here is a link that might be useful: Heating calculator

Thanks all,

We decided to go with the HFGH. Everything we read indicated that it was the best way to go for our zone.

karenrei, thank you for your recommendations. I gathered that you might be abroad (perhaps the UK?). My soil here freezes to a minimum of 3ft below ground so it's not practical to insulate straight down or even to lay pipes below the GH.
Wiring the GH is the easiest part and something I would do myself. No expense there other than for materials and I have some of those around. Our winter temps get to -10F on average but have been known to hit -20F. Our average snowfall is 5ft over the winter. There are of course years of above average snowfall and our roof would need to be able to withstand that load. It's not practical to think that we're going to go out in the middle of a blizzard and scrape snow off the GH!
I've planned on the following modifications to the HFGH. North wall: removing all the polycarbonate panels (saving those for later repairs as needed) and replacing them with 1/2" plywood. Studding that wall out and filling with insulation. Covered with reflective insulation. This is where my seeds will be sprouted as I'm assuming it will be the warmest part of my GH. My door will be on this end wall leading into a storage shed that will double as an air lock. It too will be insulated. That should provide plenty of extra protection on that side.
From November to early April my roof, west and east sides will have a solar pool cover over them. South wall will remain uncovered throught the winter. All panels will have silicone caulking on inside and out. Roof vent panels will be secured as they will probably not need opening till April by which time I will have installed solar openers.
The floor will have a thick layer of sand topped by patio blocks. I understand these can help in retaining heat. I won't be growing anything in the ground so the entire floor will be covered. Plastic film (6mm) will be laid under the sand. Have considered also adding rigid insulation there too.
Heat,for this winter at least, will be two small electric heaters from Walmart. I found online thermostatic outlets that will turn the power on at 35F and off at 45F. For my plants this would be optional. Electric lights will be added as I find the need for them. Have to see how the plants respond to the GH environment first.
We plan to reinforce the walls as others have done to prevent blowouts.
I've wondered about adding extra film inside the GH ceiling to keep heat down lower given the HFGH's 10' ceiling. Would this impair the function of the GH or make the light levels inside too low?
Have I overlooked anything?

Thanks to all,

MollyD

[quote]karenrei, thank you for your recommendations. I gathered that you might be abroad (perhaps the UK?). My soil here freezes to a minimum of 3ft below ground so it's not practical to insulate straight down or even to lay pipes below the GH.[/quote]

No, I'm an American; I just like metric. It's easier to do math in metric. ;) My partner agrees, so we use metric in our everyday lives.

As for the depth: That's the beauty of the thing. You're not trying to sap energy from a yearlong heat store like you do with a home heat pump. In a well-insulated greenhouse, you'll have surplus heat, even in the winter, on sunny days. By pumping that heat into the ground, you're warming the ground. At night, by pumping your cool air into the warm ground, you warm the air.

In short, you're basically extending your greenhouse underground, with everything that's underground playing the role that water barrels normally play in a greenhouse. Of course, even in normal circumstances, the ground under a greenhouse will never freeze. The average temperature over that soil is typically well above freezing, unlike the frigid ground outside. The only thing sapping heat away, apart from the greenhouse at night is the outside soil. However, soil is an insulator (not an incredible one, but an insulator nonetheless); thus, the rate of heat flow is typically slow, and can be essentially brought to a stop by installing a bit of insulation into the soil, about a foot deep, around the edges of the greenhouse. For heat to flow to the outside, then, it would either need to go through some soil plus insulation, or go two feet (or more) around the insulation. Neither heat flow happens at a relevant rate.

A lot of people treat the ground as their enemy. With a little insulation around the sides of the greenhouse in the soil, it can be a very good friend, helping average out those problematic day-night temperature swings. ;)

Good luck!

Karenrei, do you have a link for this idea? Months (maybe 1-2 years) ago, we discussed this in a thread. There was a link to a web page that used this idea as the primary source of heat/cooling for the greenhouse. I've been Googling to try to find that link unsuccessfully. It took miles (figuratively) of 4" black pipe to make it work. The black pipes were brought to centrally located 55 gallon drums fitted with a fan. The pipes were buried under the greenhouse or in raised beds in the GH. I'll keep looking for the site. SB

My dislike was that you ended up using lots of energy to make a very large hole, with all the associated soil compoaction issues to make a giant mould making machine. The actual heat transfer rate into soil is poor, and the heat tends to preferentially rise through soil, meaning that unless you insulate the top of the soil too, most of your heat will come out of the ground very quickly.

We should all be storing water anyway, as climate models predict that places where people are living are going to dry out, with less rainfall, and less predictable rainfall. Here in Australia it has already happened, with a 15 year drought, and fires burning Sydney in early spring, rather than late summer.

It's much easier to move heat in and out of water - my 4kL tank is already 20C when the outside temperature has been averaging 10C (that's a 20F difference). It's even easier to cool water down - just evaporate a little.

Stressbaby: Unfortunately, I can't find the link right now either. I read about it something like 1-2 years ago too, long before I joined this forum. :)

Nathanhurst: I don't propose making a big hole. Rather, I propose using an auger to drill slants that intersect, and running the pipe that way. It won't take that much PVC. If you space your pipes one foot apart and use 45 degree angles, your pipe usage would be around length * width * 1.414. So, for a greenhouse that's 3x4 meters, that would take 17 meters of PVC pipe. Say PVC costs 1$ per meter (I haven't priced it recently, but that sounds about right if you buy in bulk), that's only 17$ of pipe. Double that to factor in joints, throw in a dozen or two dollars for hooking it to a duct, another $20-30 for a duct fan, and it's still quite affordable.

I'd imagine that you could drill all of the needed holes for that greenhouse on a gallon of gas or so. That would certainly have energy-payback.

As for heat preferentially rising through the soil, what's the proposed physics for that? Or is it just a "truism" because the ground above in winter (in places without a greenhouse over it ;) ) tends to be colder? I can't picture that there's much fluid flow in soil, so you're left mostly with conduction, which shouldn't be direction-preferable.

The heat transfer rate shouldn't be a significant issue. You have all day to get the heat down there, and PVC pipe doesn't have that much of an R value (nor does soil, except over long distances). If one is concerned, they could use metal pipe.

Oh, and about water: you'd actually drink the water that you've been storing in your greenhouse? And you're keeping your containers open (to allow for evaporation)? Don't you have problems with algae, dead insects, dirt, mosquito breeding, etc? Uck. I had open water containers for a while, and well, never again. Now they're nicely sealed black buckets. ;)

Also, where I live (Iowa), water isn't a problem. There is no worldwide freshwater shortage. The world has far more freshwater than humans can use, and it is replenished faster than humans can consume. There are, unfortunately, many "regional" problems. Just because, say, there's superfluous water in the Amazon doesn't mean that a person in Saudi Arabia is any better off.

Here in Iowa, we're roughly water-neutral. Even with all of the farming here, we get about as much in rainfall as we consume.

Er, darn it, that's what I get for mixing units. You'd actually need about 50 meters of PVC for my example greenhouse. Still, it's not that extreme of an amount.

Whoa, some of ya'll are getting wayyyyy to serious for me! (And I say that with a smile.)

Di

Do the calculations and tell me the answer... I can't be bothered today. To get you started: you have a surface area of roughly 16m^2, an R-value of at least 0.2 for the boundary layer (giving a heat transfer rate of about 80W/K), roughly sinh shaped heat distribution around your pipes, soil thermal capacity of about 1J/K.g. Tell me how much you can get in and out in 6 hours. How fast does the air need to move through the pipes?

Re upwards heat transfer, I'm sure google can tell you more, but the basic mode of heat transfer in soils is by evaporation and condensation between the grains. Water vapour is lighter than air, so it tends to move upwards.

Re open water, I keep fish in the water and use it for irrigation and toilet flushing (besides thermal regulation, humidity regulation and fish home). The fish address all of your complaints :) And I get to eat the fish when they're big enough (they sell for $30/kg filletted here). Also, fish water is high in nitrates, which are excellent plant foods.

Open topped containers can collect say 10MJ/m^2 day from direct irradiation to the black liner alone.

No, I don't drink the water, that would be silly when I have treated rainforest-filtered water on tap.

Just because you get enough water by rainfall does not mean you shouldn't conserve water - if you are not collecting rain and instead using water out of rivers, that is less water available further downstream. Google says "Even in the United States, groundwater is being used at a rate 25 percent greater than its replenishment rate, the report said.". The prediction that populous areas are going to dry out presumably includes Iowa. I wasn't talking about the Amazon.

You haven't addressed my query about mould generation. Also, how do you get water out of the bottom of your tubes (as it will invariably collect there by condensation)?

It's a neat idea, but every implementation of it I've read about has been lacklustre and expensive. If you can solve all of the problems, I'm all ears! I'm not knocking you, only trying to get a reality cheque for the problem.

Well, I just googled "heat transfer" and "soil". The first page I find is:

http://www.wtamu.edu/~crobinson/SoilTemp/index.html

"Conduction is the primary mode of heat transfer in soil."

Do you have a better link? Conduction was what I assumed was the primary heat transfer method when I wrote my simulator, so if it's not, I'll need to modify my simulator a bit.

As for heat balance. Let's assume that the PVC's diameter is 0.1 meters. That's a circumference of ~0.31 meters. Times 50, that's 15.7 square meters. Assume a metric R-value of 0.2 (American R=1.14). Conduction is (Temp1 - Temp2) * Area / Insul_R, so Delta-T * 78.5W. 10 degrees average temperature difference will unload 785W into the soil. I would anticipate, as well, that with soil's low R-value, heat will flow much faster through the soil than from the PVC to the soil, thus equalizing temperatures (I'd have to write a simulator to check this out). My heat balance simulator shows, for such a greenhouse (simply adjusting my greenhouse to be the proper length and width; your geometry will be important), a peak input heat of 1458W added to the air. This happens early in the day, when the sun is bright but the greenhouse is still quite cold. By high noon, it's down to 1237W, and by 3:00 PM, it's actually cooling because of the less optimal sun angle. I can get you a breakdown of the calculated heat inputs and outputs if you'd like.

By these numbers, at peak heating time, the PVC will be sapping half of the greenhouse's incoming energy. At non-peak heating times, it will be reducing the heat input even more. Again, I could do a more detailed simulation if you thought it necessary, but a rough look at it appears that this system would do a very good job at heat storage.

As for heat capacity (the other issue), my research indicates that soil specific heat ranges from 800 to 1500 (lets say 1100) J/kg*K, and soil density is typically around 2000 kg/m^3 (although highly organic soil can be much less). Assuming a pipe depth of 0.5 meters, we're looking at 6 cubic meters of soil. This translates to 12,000kg of soil, storing 13.2 MJ of energy per degree, I.e., 3.6kWh/degree, meaning that the soil would only change less than 1 1/2 degrees in 6 hours. Now, this a very rough calculation for a number of reasons. It assumes even soil heating, which is false (this works against soil energy storage). However, it also assumes that no soil will be heated below the pipes, which is also false (this works in favor of soil energy storage). Assuming that the factors roughly cancel out, the soil temperature should fluctuate very little between day and night. I already know, by my pipeless-greenhouse simulation, that this indeed is the case. About 1 foot down, I show the temperature shifting only about one degree between day and night. Of course, since you have some complaints about the conduction-dominated soil heat transfer model, I'm open to changes in this.

As for water usage, the areas downriver from us are rainer than we are (east Texas, Louisiana, etc). The US's water problems are mostly in the rockies and especially the desert southwest. Phoenix is subsiding from the overtapping of its resevoir. LA has basically terraformed the Owens Valley by the amount of water it's diverted. Etc. However, we have little to no effect on them out here. As I stated, the world doesn't have a water shortage; rather, there are regional water shortages. Conserving water in an area that doesn't have a water problem doesn't help with water problems in areas that do.

As for mold, that's certainly an issue to consider; however, I'm not concerned. Sealed PVC will only have the moisture from your greenhouse affecting it. As mold doesn't grow very quickly on the walls of a PVC greenhouse, I see no reason to expect quick mold growth in the underground PVC pipes. If mold does become a problem, with the "auger-slant" approach, what's the difficulty? All pipe segments come up to the surface, so you can clean them out, or even potentially replace them (if your soil is compact enough that the hole won't collapse) without any digging.

As for water collection, the PVC will only be colder than the inside air during the daytime. At night, it will be hotter. As the PVC will have air actively blown past it, if there's any water in the system, the air should reach the saturation point with ease.

If, by some accident, water accumulated, I could take a few joints off and blow each row out with a leaf blower (I doubt one could blow water out of the whole system at once, but a single row should be doable. If not, a half row, a third row, a quarter row, or whatnot).

I'm not knocking you, only trying to get a reality cheque for the problem

Oh, by all means, do! I'm actually enjoying this discussion. Some things I actually hadn't considered before, such as mold. Naturally, I'd start out with just a single row of PVC and see how it works out. And while I certainly don't see anything close to a showstopper here, who knows what the future will bring.

By the way, your water system sounds nice -- making use of your heat-sink water as a fish pond. I'll have to consider that for the future. ;) All of my simulations show that the water tends to stay at a nice comfortable temperature even when if air temperature gets uncomfortable, so fish should be in good shape.

I think I need to get hold of your simulator. Should we polish it up and put it on sf.net or something?

The problem with any thermal store is getting the heat in and out, and I do worry about your assumption(?) of a ten degree difference. This mean your temperature is going to be swinging +/- 10C, which seems a little large for most people (actually, a large swing is useful in some cases). Also, this is the peak transfer, or have you integrated the whole system? Depends on the goal really.

I can't remember where I read about soil heat transfer, might have been some project to store house heat in the ground. The term 'conduction' might be used rather loosely on that site. In any case, did you consider the heat 'soaking' rate as it passes through the soil? soil is normally considered to have R1/foot in US units, which means that a cylinder of 40cm diameter around the pipe will have a dT of 52W/K (where it then tails off to around 50W/K)

12000kg of soil is about the same as 3300kg of water. But I'm still not convinced you're going to get your heat in and out. (Mind you, I'm not convinced that I'm going to fare any better with water. The big advantage of water is that you can move the water around as well as the air)

One idea I've been thinking about is extracting the heat at the peak of the greenhouse, where it gets the hottest. Something like a fin-tube or a large spiral of PE pipe with a ceiling fan.

I haven't implemented this yet, mainly because I am unconvinced I'll transfer enough heat to make it worthwhile. Forced air movement reduces the boundary layer (from memory it is either linear or reciprocal) so it may be more efficient to have a smaller heat exchanger with faster moving air. The trick would be to combine this with the HAF.

OMG, two peas in a pod. When you guys get it all worked out, will you translate for the rest of us? LOL

stressbaby,

Yeah this is a great thread... to watch. Ya need at least a 12 pack inside of you before you can start to understand it and then have a couple more beers before you post!

karenrei,

The only part so far I got is the word "Iowa"

Now, If you live in the same Iowa that I do and you are planning to build a solar greenhouse, and you have room, then the first thing you should do is go to a farm auction and buy one or more of those big poly tanks. You know, the ones you see farmers carting around in spring.

Then bribe any 16yo son (or daughter) of a farmer to borrow a tractor with a back hoe and invite them to a hole digging competition and hog roast one sunday afternoon. (kids are so gullable, they love to rip up dirt with heavy machinery and when they find out the "hog roast" is a pack of oscars finest they won't complain much!)

As soon as I get my paws on a 3000 gal tank it is going underground.

OK now back to the math!

P.S. For those who speak metric 3000 US Gallons is 11356.235352 Litres give or take a teaspoon or two.

stressbaby, hopefully better, we'll have a robust simulator to model how each greenhouse improvement will affect the long term performance. Then you'll be able to answer your questions by twiddling parameters and looking at pretty graphs :)

chris: that's a lot of digging.

I'm sitting over here next to Chris, watching; I've got to admire you folks for having both the knowledge and the determination to work all this out.

Just a little tangential, that hole-digging competition idea just might work. During some home renovation we needed to remove a wall (not a load-bearing wall) ... invited a neighbor's curious and somewhat hyper pre-adolescent to have a go and handed him sledge hammer and pry bar ... he had a ball and the wall was gone in no time.