Mike's passive solar greenhouse design/build

Hello friends, I'm finally ready to present my greenhouse design for your input and suggestions.  I plan to finalize the design this winter and build it this summer.

My mission is to create a greenhouse that allows people in a cloudy and frigid environment to grow food year round.  I am imagining that this will be particularly suited to the northern Midwest and New England.  It may certainly work in many places but the design will be optimized for my sun angle and seasons.  Once it works (I'm trying not to say if) I want to share the design broadly.  I want to reduce the need for shipping produce across the country/hemisphere.

My goal is to build a greenhouse that is expandable, effective, automated, cheap, uses no fossil fuels for heat and is DIY.  I really want to grow tropical plants in it.  I know this sounds too good to be true and maybe it is.  But I'm going to try.  If I fail to get tropical temps, I can do citrus or Mediterranean.  If that fails, I still have a tremendous amount of season extension.  But I really want to have Wisconsin grown bananas  

Our climate is cloudy and cold in the early winter and sunny and frigid after New Years.  Typical December weather is 20 days of clouds and highs/lows of 30F and 10F.  Typical January weather is 10 cloudy days and highs/lows of 10F and -10F.  Worst case is several days of overcast with high/low temps of 10F/0F or clear and sunny but highs of -10F and lows of -25F.  Summer is more sunny than cloudy and highs in the 70s.

Summary of the design:
-20' by 40' by 16' high structure on a frost protected cement footing at grade.  
-Cement block pony wall 32" high on the footing to allow for snow shedding and elevation variation across the site.
-Curved web trusses made on a jig that will look kind of like Dan Huisjen's post here on Permies
-Heavily insulated North, East and West walls
-The walls won't form a gothic arch.  The South wall will overshoot the North one to increase glazing and get sunlight farther into the back of the greenhouse in the summer.
-Vents low on the South wall and at the peak for summertime ventilation
-Heat provided by a large compost bunker
-Thermal mass provided by a phase change material in a tank
-Heat collection provided by a solar and/or thermal collector at the peak that feeds into the thermal mass
-Heat distribution from the thermal mass by an undetermined method as of yet
-Nighttime automated moveable insulation to block heat loss through the glazing
-Perennial trees planted in native soil, arranged by proximity to compost hopper depending on their cold tolerance

Design details (if I should break any of these out into their own threads let me know):
-The general shape is a 20x40 rectangle with the long axis running E/W.  I will clip the NW corner due to vehicle access around that corner of the building.  The footing will be 12" wide by 7.5" high with two rebar.  If that isn't sufficient, please let me know. I'll run 2-4" of styrofoam down the outside of the footing and then horizontally 3'.  This will keep frost from getting under the footing and also allow the warmer deep soil temps to help heat the soil in the greenhouse.  The grade rises 6-12" from West to East so I'm thinking the top of the footing should be flush with the grade on the West side and let it drop below grade as it goes Eastward.

-On top of the footing I'll build a pony wall (I had to look up the proper term) consisting of 4 courses of 4"x16"x8" cement block.  I'd normally use 8x8x16 block but I can get this size really cheap.  I'll have J shaped rebar pieces in the footing to line up with the holes in the cement blocks.  Then I'll use additional rebar vertically through the holes in the block.  I'll line those rebar up with the center to center spacing on the trusses.  I'll probably fill all the voids in all the block with mortar/cement.  

-The curved web trusses will be cool.  They will be on 4' centers and sit on post bases so they can't wick moisture up from the block.  They will be untreated wood.  My challenge was figuring out how to make them longer than 16' without having weak spots from joints.  I think I'll buy 20' LVL beams and resaw them down to 1.75" by 1".  I have a bandsaw so hopefully that will work out.  The webs will be chunks of 2x4 about 8" long.  Since they have rounded corners I may trim them down to 3.25" so the glue surface is the full width.  Or I may use 2x6s and cut them down to 3.5" and use the excess for diagonal bracing.  Or should I use 2x6 web members?  I'm planning on gluing the joints and screwing with two screws per side per web piece.  I'll build a jig from plywood so I can clamp the truss together as I go.  If anyone's built these before and has some advice on my dimensions I'm all ears.

Some advantages to the curved truss is that it should be stronger and lighter than a straight rafter.  By being curved the glazing plastic will stay taut easier.  More headroom near the N and S walls.  And they look cool

- Per solar greenhouse design methods, I'll heavily insulate the E, W and N walls.  I debated leaving some glazing in the E and W walls but in the winter my sun is blocked by large trees in the earlier morning and later afternoon so I figure I'll lose 20x more heat for the little gain those windows would give.  I'll talk to the contractor that insulated my house to get his thoughts on materials.  I'm either thinking blown in cellulose or a couple layers of polyiso and styrofoam.  I think 6-8" of cellulose may be a waste since after R20 I'm chasing diminishing returns.  The web truss would allow cellulose to get between the webs so the only thermal bridging would be through the webs.  I may also have purlins outside the trusses (1x4's on the flat) which could allow for further thermal bridge reduction with a blown in insulation.  Then again, styrofoam outside the trusses would eliminate all the bridging. Should I consider any other insulation options?  The interior will hopefully be pine siding and the outside will hopefully be pine siding on the vertical E/W walls.  I will also have temporary insulation in the peak of the North wall in winter.  When it's removed in the summer the sun can almost reach the whole floor.

- The curved trusses will pass by one another at the peak and the South one will continue a few feet into the air.  There won't be a ridge beam.  I'm planning on bolting the mating trusses together with carriage bolts.  I think four 3/8" bolts would do the trick.  At the top of the South wall there will be a board that connects them all together.  Diagonal bracing and purlins will hold the trusses in place.  Plus the North roofing will give a lot of shear strength.

- There will be a vertical section that connects the top of the South wall to the North wall which will double as a vent.  Automatic vent openers will operate those vents.  They'll be the full 40' length of the greenhouse.  At the bottom of the South wall there will also be a 40' vent to let cooler air in.  The lower vent will be manual and probably stay open all summer since the surrounding area is protected from critters.

- Heat via compost:

This will be kind of make or break.  If it works, it will be awesome.  In the NE corner of the greenhouse I'll build an 8' by 8' by 6' chamber.  It will be accessible from a door on the North side of the building and a hatch inside the greenhouse.  That way I can load feedstock in from outside in the fall and remove compost in the spring either from outside or inside the greenhouse.  The bunker will probably be made from cedar unless I find a better material.  From what I can tell, the Jean Pain method uses wood chips from green wood (leaves included) and it gives off heat for 18 months.  I only need 4 months of heat and I really don't want heat in the summer.  So my plan is to use a combination of wood chips and some high N materials to skew the mix a bit away from the highly C based Jean Pain system.  But I don't want to use manure for that N due to ammonia and offgassing.  I'm currently thinking coffee grounds and greenery.  

Don't compost piles need to be turned you say?  Well, kind of. They need oxygen.  I read the The Compost-Powered Water Heater by Gealan Brown and it drove most of this design.  If you introduce air under the pile with a perforated drain tile and exhaust it from the top of the pile with a fan, you create the aeration you need.  The feedstock needs to allow air to pass through it so I think the coffee grounds and wood chips would work well, not sure about big chunks of green leaves.  So the plan would be to have a fan draw air out of the top of the chamber on a timer.

How to get heat out of the compost bin?  I have a few ideas.  One would be to duct that hot aeration air down the middle of the South grow bed (underground).  It will give off its heat and then vent the air either into the greenhouse out outside if the compost gasses pose a risk to the health of the greenhouse inhabitants.  Another heat distribution plan is to plumb 4" pvc channels through the compost pile that enter near the ground and exit near the top.  They would be "airtight" so the coolest greenhouse air enters at the bottom and rises through the pipe on its own and shoots out into the room.  The third distribution method is natural convection from the surface of the bunker.  I could circulate water lines through the compost and around the greenhouse but I'm hoping I don't need to do that.  By avoiding extra pipes in the compost it would be easier to fill and empty the bunker.

-Thermal mass via phase change materials.  Changing the temperature of a pound of water by 1F takes 1 BTU.  Making that same pound of water go from 32 and liquid to 32 and frozen (phase change) takes about 80 BTUs.  If only there was another material that freezes and thaws at a more useful temperature.  But wait, there is.  Coconut oil comes to mind (melts at 76F) as does glycerin (melts at 64F).  So the plan is to get a couple 55 gallon drums of glycerin to make a heat battery.  One challenge is that it's hard enough to get heat in and out of 55 gallon drums of water.  Their size works out pretty well to give thermal tempering over the space of a few days (from what I read).  But if that same drum is holding 80x the energy it would be like a Tesla battery hooked up to a solar calculator.  It would take forever to heat it and forever to drain it.  

- Heat delivered actively to the phase change battery via a coil of piping inside the drums.  Water (without anti-freeze) would run from the coil in the drums up to the peak of the roof (in the triangle area above the North wall) to a redneck solar collector.  I'm imagining a series of  ten 3/4" irrigation pipes that run E/W in the peak.  The water would be pumped to a manifold at one end and then down the 10 pipes before they rejoin at another manifold and then return to the battery.  I'd need a solar thermal pump controller for this so that it only pumps water when the temp in the triangle is hot enough.  This would rely mainly on solar gain to heat the water and slightly on the warmth of the air in the triangle to heat it.  At dusk the water would "drain back" down to a holding tank.  

- Heat removed from the phase change battery is tricky.  How do you distribute "low grade" heat?  The water in the coil should be around 65F.  The greenhouse air would likely be 40F.  Blowing air through a radiator would be one way.  Running black irrigation lines around on the soil surface or slightly underground and then pumping the 65F water though it should also work.  Any thoughts on other options?

- Glazing is 85% of the heat loss.  My plan is to attach 1x4 boards to the bottom of the South wall trusses to act as small ledges.  Then a 4' wide roll of Reflectix (or equivalent) insulation will be attached up at the peak of the roof.  It will roll down on the ledges until it ends up at the bottom of the wall.  Effectively sealing off each truss bay.  If the Reflectix wants to sag due to the 4' span I can simply run 50 lb fishing line E/W across the bottoms of the trusses before I attach the ledges.  The insulation rolls will act like the sort of window blinds in the attached picture.  They'll raise and lower via strings that attach to a pipe at the peak.  Rotating the shaft will raise and lower the rolls.  Initially I'll control them by a bike chain and crank on the wall.  Once I get the kinks out of the system I'll hook up a motor to it.  I believe the Reflectix will add R3 to the glazing which will be huge.

-I think I'll start my trees in pots so I can evacuate them to the house if I need to.  Once I'm comfortable with the greenhouse I'll put most of them in the ground.  The tropical fruits will be closer to the compost heater.  Citrus and other cool tolerant trees would go on the West end.  If needed I can subdivide the greenhouse into two halves to concentrate heat in the East half.

I've sketched 38 geometries for the greenhouse over the past two winters.  Hopefully I have it now.  I wanted to optimize the glazing for sun angles in late Nov so that the maximum solar gain is late Oct through late Feb.  Optimizing the shape to cover my 20x40 footprint and have a steep enough South wall and yet have some sun on the floor in the summer was tricky.  But I think I did it.  In the sketch, South is to the left.  The sun angles at the solstices and equinoxes are indicated by the black line segments.  The permanent North side insulation is shown in brown.  Temporary winter insulation in the peak is the brown spots.

Well, that's it for now.  Sorry for the long winded post without a lot of pictures.  Let me know if you have thoughts or ideas of how to make it better.  Let's make the world a better place!

Nice to see more … unconventional architecture appearing here!

My thoughts on the parts that I understand so far:

Curved trusses of 16' (~5m): green wood of that length can be obtained and bent by pulling the two ends together with a strap. The force required is quadratic with the thickness of the wood.
The tension between the two ends will need to be maintained, which gives additional load carrying, but also requires to add it into the design.
I don't trust wood glue in an environment that isn't stable in temperature and humidity.

On the thermal battery: Intuitively I would distribute the barrels on the floor (or lowest part) of the greenhouse and hope they will transfer enough heat to the cold air, as well as radiate some.
A fairly big pot of water that was boiling in the evening, is only slightly warm in the morning, so there is quite a bit of heat transfer going on. I don't think it will drain it completely in one night, but that doesn't have to be a disadvantage.
If the barrels are placed partially in the soil itself, the most cold sensitive plants can be placed surrounding it.

I've been thinking about the same goal for a few years and one thing I'm considering is shou sugi ban treatment of the wood to help protect it from the humidity of the greenhouse.  An added benefit is that I think it could look beautiful.

Mike, I'll be following this closely since we are dealing with practically the same climate.  Your ideas are pretty much parallel to mine with the exception of your use compost.  My plan is still to use earth tubes for heat.  I doubt I will get mine built this summer.  We are building our (pretty large) shop and our house this spring.  That on top of the land purchase means I'm going to be working on a lot of projects I can do for free for awhile :)  That may work out better for me, because you can share your "lessons learned" if there is anything you would do differently, or hopefully, everything will work exactly as planned and then I can just copy :)

As a side note, I was wondering about using a char-treatment of wood.  I didn't know it was called "shou sugi ban"! cool!  Sounds good to me.  I figure it would work on any type of wood, and this is very good since not everyone has rot-resistant species on hand, especially in dimensional lumber.  Yeah, a greenhouse is kind of a damp, rot-prone location.  I strongly advocate some sort of wood treatment, even if it's just stain or paint.  If you use untreated wood, plan on treating it and/or replacing it in a few years depending on how damp it gets from leaking rainwater on the roof and so on...  I figure the charring process could be done over a hot campfire/coals.  Energy-wise it might be more efficient in a (horizontal) tube style charcoal maker; although process-wise, I don't know if that would cook hot enough or evenly enough.

Anyway to get back on topic...
Is the roof glazed or unglazed?  Will you use a double or triple glazing of plastic film?  Will you attach the film to the outside and inside of the frame;  how?  What type of film?

How will the curtain move?  The IR reflecting curtain supposedly prevents freezing (in the Chinese design).  

I calculated the heat storage of 2 drums of glycerin at about 100kBTU.  That seems in the ballpark for a daily amount.  

I'm confident that air-convection is the bottleneck for heat transfer.  The heat-transfer coefficient for natural convection is amazingly low at only around 20 W/(m^2*K), (indeed even for the walls of a pot of boiling water).
To moderate the temperature under the heating rate of the summer sun, I believe a heat exchanger or two of the aluminum-fin type used in an car radiator with fan or a window-box air conditioner is in order.  Even folks that use these still see temperature swings of 40F (22C).  The car type might be hardwired to a "12 volt" solar cell.  The Dutch style greenhouse uses a heat pump to overcome temperature drops across the heat exchanger, so an a/c unit could be handy for that.  

Growing tropical plants seems like more of a challenge.  Are you prepared to use a fuel heat source as backup on the coldest days?  Any artificial lighting?  reflectors?

I'm reading this book at the moment:
http://ia800306.us.archive.org/30/items/fe_The_Solar_Greenhouse_Book/The_Solar_Greenhouse_Book.pdf

I was thinking that for the thermal mass I could just start with the heat collection system and coil and hope that the barrels (if spread around the greenhouse) would give off their heat convectively.  If that fails I could tap into the heat collection system to add a heat distribution system.  I hadn't thought of partially burying them (thanks Sebastian!).  Usually I'd be afraid that the ground would suck the heat out of them.  But maybe if they are operating at the 65F range and the soil might naturally be at 55F, the heat loss would be minimal and could allow for planting heat loving plants nearby.  Maybe a phase change battery could be the center of a banana circle?

For the curved trusses, I was thinking of using a homemade sealer that I read about in the Bioshelter Market Garden book by Darell Frey.  Otherwise I may use white paint to increase reflection inside the greenhouse.  Any surface that isn't deliberately collecting heat should be reflecting it to bounce it around inside the greenhouse.  While I'd love to have pretty wood trusses, I may have to settle for pretty white wood trusses.

I was also thinking more about the benefits of using sytrofoam/polyiso for the North insulation.  It would allow the trusses to be exposed on that side which gives more headroom for trees and people.  It also allows more room to run utilities and it may look cool to have the trusses mirror each other on the inside.  I'll need to compare costs to see if that makes the decision for me.

Looks like we were typing at the same time Mike

I'll definitely treat the exposed wood.  I don't think I want to experiment with shou sugi ban though.  I will probably put a layer of something (duct tape, teflon tape, etc) on the outer surface of the south trusses so that the wood isn't directly touching the glazing.

Glazing!  I forgot to mention my glazing plans.  Originally my plan was to use a Solar Pool Cover as the glazing.  It's basically really thick bubble wrap.  I've seen it used to insulate greenhouses in the winter and some folks use it year round.  They're designed for wet environments and to not degrade in the sun.  I called a manufacturer and the guy guessed it would have a R3 but he didn't really know since that's not what it's designed for.  But I worry that I'm trying too many experiments at the same time.  

I figure that with the moveable insulation, the R value of the glazing isn't as important and I should concentrate on light permeability.  So now I'm leaning towards two layers of poly film with a blower.  I think that gives R2 and if I use 92% transmission film it will let the most light through in the most affordable manner.  I'll attach it to the outside of the South trusses in a way that I can replace it easily or upgrade to the pool cover.

I do wonder if I should make allowances to someday install twin wall polycarbonate sheets.  I think they need their supports to be at a 24" spacing and the 4' sheets may require a support at 4' 1/4" spacing instead of exactly 4'.  I'm not sure but those details would complicate things.

Thanks for the heat storage math!  I did the same math and got 99kBTU (yay, I did something right).  I agree that I think convection will not do the job to get heat out of the barrels.  What is the time unit on your convection heat transfer coefficient?  I calculate a drum as having 2.3 square meters of surface.  So for a delta T of 20F I get a transfer amount of 506 watts per drum.  Is that per hour?

In the summer I'm counting on ventilation to moderate the temperatures.  I may find a way to use the phase change battery to help a bit but I don't expect to use it much once winter is over.

I may initially have a propane backup for the greenhouse but I really hope to not need it (of course).  I could do a wood fired back-up but I don't want to build that sort of infrastructure if I don't know I need it.  And if I'm on vacation and the greenhouse needs heat I can't really start a fire.

I have given thought to artificial lights.  I'm thinking I could install strip LED lights on the underside of the trusses.  That way they'd not really be obvious and the light would come from all over.  I will have grid power to the greenhouse to run pumps, lights, etc.  

The only reflection I'm planning on is the internal white surfaces, especially the North roof.  Based on my winter sun angles a good portion of the sunlight entering the greenhouse will hit that wall.  I have noodled on having external reflectors to bounce additional light in but I can't really see how I could make them beefy enough to handle wind and snow.  

Thanks for the book, I'll read through it.  I can't remember if I read that one already.  They're all starting to blend together on me

Mike J,

A blue pool cover might be an interesting roof/top material.   I suppose the south side should be clear. It's anyone's guess what it does with near and far infrared.  I'm fairly certain the bubbles would be R-2 and non-bubbles R-1, assuming equal area then on average a single layer is probably R-1.5 , and two would be R-3.

Blowing air between 2 layers of poly film probably reduces it from R-2 to near R-1.  Better I think just to uses spacers to make the gap and save on heat and electricity.

>What is the time unit on your convection heat transfer coefficient?  I calculate a drum as having 2.3 square meters of surface.  So for a delta T of 20F I get a transfer amount of 506 watts per drum.  Is that per hour?

506 watts = 506 Joules/s = 0.482 BTU/s = 1735 BTU/hr --> x2 drums = 3470 BTU/hr

Okay maybe that isn't quite as low as I thought but if it was me I'd want it a lot higher, like 3 to 10x higher.

> In the summer I'm counting on ventilation to moderate the temperatures.  I may find a way to use the phase change battery to help a bit but I don't expect to use it much once winter is over.

Okay.  I'm studying the possibility of year-round ground storage.  This seems necessary to keep it warm in zone 5 or 6 or less with no extra heat.  Zone 4?!!? Oh man!
Although I don't know how big your compost pile is or how much heat that puts out.  I didn't factor that in.

Here's what I found for banana tree temperatures.

deg.F
--------
100+ stops growing
90 optimal
<80 slows
53 stops, for all but the hardiest varieties

That's a 20 deg.F swing.  A 40 deg swing isn't gonna grow bananas.  
I figure this is pretty tough to do at our latitude with no gas/wood/oil.  I think it can be done but I figure one has to go all out.  You might want to either aerate the compost for more heat or make a digester to burn the methane.

-Mike P

Another element helping the pool cover is the thickness of the plastic.  It's much heavier than poly sheeting so that may help it insulate better.  I was scheming on putting the bubbles outward facing and then adding one layer of greenhouse poly on top.  Then it would probably add another R1 and the cheap poly would take the brunt of the UV.  But I figure the transmission of light would drop even more.  I'm guessing the pool cover is down to 70% already.

Many greenhouses blow air between the layers and I assumed it provided more than the mere benefit of separating the layers.  If that isn't the case, yes I could easily space out the two layers to avoid the fan.

Thanks for doing more math for me.  Duh, watts are timeless  That heat loss of 3.5 kBTUs per hour isn't anything to sneeze at, is it...  That would deliver 83 kBTU in 24 hours.  From tanks that were holding 100 kBTU.  Normally we'd have to do calculus to figure out the transfer as the tanks cool down but due to the phase change they kind of stay at 64F

Ground storage could be cool but I figure that takes away my "cheap and expandable" goals.  And many (but not all) of those systems seem to do summer storage (cooling) but also rely on a lot of winter sunlight to keep recharging the ground.  I think I'm too cloudy for that...

The compost pile is an 8' octagon that is 6' high.  I figure that out at around 12.4 cubic yards.  In Gaelan's book I thought he gave an approximate heat yield from a yard of compost at 1,000,000 BTU/month.  But now that I read my notes I see another reference that gives a lower number that works out to 180,000 BTU/month.  That second reference is for an outside pile with water pipes circulating through it to heat a nearby structure.  So I should get somewhere between 74 kBTU and 413 kBTU per day.  Maybe I need to check out his book again and reread it....

Aerating the compost will be handled with the fan I mentioned in the original post.  It will draw air out of the sealed compost chamber which draws fresh air in under the pile via perforated drain tile.  If the pile needs additional aeration I figure I can have access holes in the top and make a corkscrew that is 6' long that I can screw down into the pile and then turn with a cheater bar.

Yup, bananas will be tricky and just a novelty.  Maybe they'll just go dormant for a few months and then spring to life in March...

I like the synergy where the compost is a source of heat, CO2 and soil, and the greenhouse takes all that up and produces compostable material.  Plus there aren't any safety or labor issues with trying to have animals or open flames inside.  I guess one can even use fans to regulate how much heat moves from the compost pile into the rest of the greenhouse.  


I'm kind of intrigued by this idea of movable insulation based on soap bubbles.  http://www.tdc.ca/bubbleinsulation.jpg  
I like how there are very few moving parts, and has high total R-value, and is potentially reliable.  Maybe with a clever design it might even be affordable, although I suppose it could take a lot of R&D to make it really good.  Supposedly a startup company or two had a go at it but in the end the business didn't succeed for whatever reason.  I don't know about the business prospects but engineering-wise it seems like a good idea.  

http://www.tdc.ca/soapbubbles.jpg

http://www.tdc.ca/bubblegreenhouse.htm

Thanks Mike, I like those synergies too.  I'm also planning on having a platform above the compost for hot yoga or just laying there.  At one time I had thoughts of a chamber inside the compost to act as a little sauna.  

I liked the soap bubble idea too.  If someone had that working I'd be all over it.  With my curved trusses there's plenty of space to put bubbles.  But that's a level of experimentation I don't want to attempt along with all the other innovation I have going on.  

Thanks for the ideas and input!

This video explains it well.
http://www.youtube.com/watch?v=JTBGE5IZa2Q
They claim the retractable curtain is cheaper and simpler.  That could just be because it's an older technology.  

I mean how hard can it be if this young kid can make a foam machine in his driveway for free by pointing a mini shop vac blower into soapy water?
http://www.youtube.com/watch?v=0BBd3KXg4dQ

I know you're not going to do this, I'm just discussing design ideas and making conversation because we said: "make the world a better place".

Mike Phillipps wrote:I'm reading this book at the moment:
http://ia800306.us.archive.org/30/items/fe_The_Solar_Greenhouse_Book/The_Solar_Greenhouse_Book.pdf

That is a good read!  I just got through the part starting on page 78 about 55 gallon barrels.  The author seems to prefer small 1 to 5 gallon storage systems due to stratification in the barrels and the radiation losses off the back side of the barrels.  I've read elsewhere that 1 gallon jugs lose their heat in hours, 5 gallon ones do in a day and 55 gallon last a couple days.  So that source suggested smaller jugs for reliably sunny locations and 55 gallon drums for periodically cloudy areas.

I found a source of glycerin in plastic 55 gallon drums.  Mike, do you know if plastic would provide the same 20 W/(m^2*K) as steel?  Or does the material matter?  Thanks!

Mike J,

Amazingly, plastic is the same.  Convection is the limit, not conduction.

Did you find a local biodiesel producer, or did you have to pay a lot to have glycerin freighted?  

Mike P.  

Thanks Mike, that's great news.  I did find a "local" biodiesel producer (within 4 hour drive).  I haven't contacted him/her yet and I don't know the price but I'm thinking it will be pretty reasonable.  Depending on the reasonability, I could entertain the idea of more than two barrels.

I did read in your book that some exploration into phase change materials (salt hydrates and waxes) was being undertaken and one downside to some of them was that the phase change broke down after a number of cycles and the materials just stayed liquid.  I don't see how that would happen but it must have.  Could you ever see water getting tired of phase changing and just staying liquid at 20F?  I may have to see how hard it is to  donate or dispose of exhausted glycerin.  

I couldn't sleep (obviously, it's 2am) and was thinking more about the triangle at the peak of the greenhouse.  I think that will be a multi function space.

In the summer it will be open to the greenhouse below to allow for massive venting out the North side of it via automatic vents.  And it allows for more sun to reach deeper into the greenhouse when the sun is high in the sky.

In the winter it will be isolated from the rest of the greenhouse by removable insulation (2-4" of styrofoam) to hold in heat better.

I'll run solar thermal collection pipes in that space in the winter to harvest the solar energy that is entering the triangle and pump it down to the phase change drums (or greenhouse).  So I'll get the effect of preserving the heat in the lower greenhouse while adding in some of the solar gain from that upper triangle.  Or in other words it's a seasonally usable built-in solar thermal collector.

If that triangle gets toasty hot on sunny winter days (which I'm pretty sure it will), I can set up an intake pipe at either end to draw hot air out of the triangle.  I could terminate those pipes down low in the greenhouse, or bury perforated drain tiles a foot deep in the greenhouse to dump that heat into the soil.  They'd then exhaust back up into the greenhouse.  This could also be a cooling system in the summer.  Very much like the climate battery systems, just less extensive.

For the above scenario I'd have to leave out a section of the removeable insulation to allow replacement air back up into the triangle.  That would also allow for some limited automatic venting if that area overheats.

Yes sometimes phase-change materials can lose their effectiveness over time, particularly if they have impurities.  
This is not really an issue with water but it is with sodium sulfate for example, but I think can be remedied by purifying it either before or after.  http://en.wikipedia.org/wiki/Sodium_sulfate#Thermal_storage

Glycerin is more difficult to purify, but there are probably ways to do it.  Maybe run it through a water filter that has activated charcoal and ion exchange resins.  Another way might be to periodically drain off the liquid that doesn't solidify.  In general this tends to purify a substance.  This works with water for example (not that it's needed, except when making drinking water).

> I may have to see how hard it is to donate or dispose of exhausted glycerin.  

Good point!  But if you have to burn it, the biodiesel producers were probably going to burn it off anyway.

-------------

I'm of the strong opinion that manufactures of radiators and air conditioners would not put zillions of aluminum fins on there if it wasn't necessary.  I believe it is absolutely necessary for it to work.  I strongly believe a fully-closed, unvented, unshaded, sunny greenhouse will get too hot without a high-surface-area radiator, for the same reason that a car or air conditioner will get to hot without a high-surface-area radiator.  Roof vents and shade clothes work great, but then that heat is gone and can't be used later.  Now if you have tons and tons of compost, maybe you don't need to store solar heat and maybe you don't need radiators.  You will need a large source of compost though, especially in autumn, but I imagine the post-harvest ag-residues are available if you have the equipment to move it / deliver it.

Do what you want.  If it was me I'd probably put more aluminum fins on the thermal mass, and use less compost.  But you might use more compost and less aluminum fins.  That's fine so long as you have enough compost.  In that case it might be possible to not use any glycerin, and just use venting for cooling, and compost for heating.  The rate of heating from compost is a bit difficult to quantify exactly but you already have the numbers for the range of how much is required.  

Thanks Mike, I agree that fins on a radiator make them much more effective.  My hunch is that a car radiator in warm air would heat up a circulating fluid fairly efficiently.  My hunch is that a 200' run of 3/4" black irrigation pipe would also heat up a circulating fluid fairly efficiently.  If you're blowing air across the device, a radiator would be much much more effective than a bunch of pipe.  If you're also collecting direct solar radiation gain, the large surface area of the pipe exposed to the sun may make it much more effective.

I suspect that if I need to heat the greenhouse air from the thermal mass I'd try a radiator and fan.

You're right, if the compost works at the upper range of the numbers I listed, I shouldn't need any thermal storage.  Here's to hoping for that outcome

Mike Phillipps wrote:Mike J,


Blowing air between 2 layers of poly film probably reduces it from R-2 to near R-1.  


-Mike P

Could you expound on this?

Mike Jay wrote:
My hunch is that a 200' run of 3/4" black irrigation pipe would also heat up a circulating fluid fairly efficiently.  

Not really.  It has 1.1 m^2 of area.  So as a convective heat exchanger, the thermal resistance is around 1/(20*1.1) = 0.044 deg.C / Watt.  
It can only move 200 watts (0.2 BTU/s ~= 720 BTU/hr) with a temperature drop of about 9 deg C ( 16 deg.F).  That's only enough to cool about 2 ft^2 of greenhouse floor area at high noon.  Woefully inadequate.  With that much tubing there wouldn't be any room for plants!  haha.  Seriously.  So 200 ft of pipe would be a token gesture and the cooling would be mostly venting.  Almost might as well not even bother with the pipe just to cool 2 ft^2, lol.  

Sorry but if I get tired or frustrated repeating this explanation, please refer back to these messages.  Convection sucks!  (tip of the hat to Paul's "Greenhouse suck factor" thread)

Mike Jay wrote:
If you're blowing air across the device, a radiator would be much much more effective than a bunch of pipe.  

If you're also collecting direct solar radiation gain, the large surface area of the pipe exposed to the sun may make it much more effective.

I suspect that if I need to heat the greenhouse air from the thermal mass I'd try a radiator and fan.

You're right, if the compost works at the upper range of the numbers I listed, I shouldn't need any thermal storage.  Here's to hoping for that outcome

Yes.  I agree.  Solar radiation is a much more "efficient" mode of heat transfer than convection.  The heat exchanger you describe is a fairly economical solar hot water heater.  

thanks,
-Mike

Mike, have you thought about using Solexx 5mm, it diffuses the light and gives you a R=2.3

Thanks Laurie, I had looked into them a while back and I couldn't remember why I hadn't kept them in my brain.  I just checked again and now I remember.  To do my greenhouse would cost over $1500.  A double poly layer would be closer to $250 and still give a R2 (give or take).  

If they were in 4' widths instead of 49 inches, I could retrofit in the future.  But since they're a bit too wide, I doubt I'd be able to try them in the future.  Unless I'm missing some other great benefits of the Solexx?  Thanks!!!

The reason for 49.5" is so that you can overlap panels and attach directly to frame.  I am just concerned for you using polyfilm.  Just from my own past experience, you lose power, you loose your insulation factor, same if the outer layer gets punctured.  

Anyway, I was ready you post and wish you the very best.  I have actually seen this design using Solexx on the web and it was pretty amazing.

Ahhh, that makes sense now.  I thought they butted together and had a T strip to cover them or something.  

Regarding the power for the blower...  If no one chimes in saying that the blower part is needed, I'll just put on one sheet of poly, then a 1" spacer, then another sheet.  

I'm still not understanding the blower part.  My understanding, and it may be wrong, is that going from 1 sheet of plastic to 2 increases R value from R-1ish to R-2ish.  Using a blower does two things:  keeps tension on the plastic so that there is less chance of tearing, and keeps the sheets apart so the R-value stays at 2.  If you just stick 1 inch pieces of wood between the pieces of plastic I would think that a) there is more possibility that the plastic can stretch slightly or it won't be tight enough and will sag until the pieces of plastic touch, and/or b) the pieces of wood you attach to create a thermal bridge that is at best R-1 again.  That may not be a huge concern, but it may be if you are trying to grow bananas in WI.  Is there a flaw in the way I'm looking at this?  I don't understand how using blower can make the R-value go down?

The whole point of insulation is to *prevent* air from moving.  Try putting a blower under your parka in the wintertime and tell me if that's warm.  
-Mike

Hmm, I think we need input from someone who's done the blower trick.  I think the blower only fills up the air gap.  Once the proper pressure has been reached a check valve holds the air in and there's no air movement.  If you don't have the check valve type and the blower blows the whole time it still wouldn't be going anywhere (just building/holding pressure) so I don't think it would be like a blower under your parka.  Maybe like a blower in your parka if you have a bungee cord holding the parka tight around your waist, wrists and neck to trap the air...

Mike Jay wrote:Hmm, I think we need input from someone who's done the blower trick.  I think the blower only fills up the air gap.  Once the proper pressure has been reached a check valve holds the air in and there's no air movement.  If you don't have the check valve type and the blower blows the whole time it still wouldn't be going anywhere (just building/holding pressure) so I don't think it would be like a blower under your parka.  Maybe like a blower in your parka if you have a bungee cord holding the parka tight around your waist, wrists and neck to trap the air...

That is my understanding as well.  

I've done some quick digging to see if blowers are needed and most of the recent posts that the search engines provided just said that when you have two layers you inflate them.  Primary reasons given were to keep the layers from touching (and losing insulation value) and to stay taught and flexible when it's windy outside.  Here's the most Official Looking One I found.   I did find one abstract of a study done in the olden days that mentioned needing a better system that two layers with battens in between them.  I don't have access to the full article but here's the Abstract.  

I do see that people recommend inflating them with outside air to keep from causing condensation between the layers.  I did find a link to a diy plan to add a check valve to the blower so it doesn't have to run nearly as much.  But the link was broken

I'm sure the blower inflates it and reduces condensation, but I think it's a dubious claim that it increases insulation value, unless maybe the average air flow is very low and the plastic is tightly sealed.  In the videos I've seen, the plastic didn't appear to be well sealed.  In order to get R-2 insulation and not just R-1, the air temperature in between the plastic needs to be about half way in between the temp inside and outside.  If you have to blow a lot of inflation air, then heat-wise, the outer layer might as well not even be there.  So it's R-1.  I think the forced-convection is going to pull away more heat then natural convection, even if the plastic layers are touching in a few places, unless maybe the air flow is very low, like lower than that of the natural convective cooling.  

From my understanding, the blower pressurizes the foil and thus stiffens it, which protects it from flapping in the wind (and self destructing).
So if you manage to find another way to stiffen the foil (using a double curvature and tension, for example), I would expect it to work just as well.

Good points everyone.  My assumption and plan would be for the poly to be very well sealed.  Especially since it's just one face of the greenhouse that I have to seal.  Maybe that's hopelessly optimistic?  I can see how there'd be more leaks if the poly wrapped around all the sides.

FYI, I'll be away from my computer till Monday so feel free to keep the discussion going.  I'll join in when I get back

That all depends on what your goals are.  Some people seem to not want condensation for whatever reason.  I don't know if that's a valid assumption but I'll go with it for now.  (By the way if there is a lot of condensation, and it's supposedly being "ventilated" to "reduce" condensation, these things partly indicate there isn't a good seal.)  I'd guess that when the air and plastic cools, dew forms.  If warm air is pumped in, then this may help dry it out temporarily, but the warm air might also add water vapor to the system that will just re-condense later.  If someone really wanted it dry, they might consider folding the plastic over and heat-sealing the edges to make an enclosed pouch.  One could also seal down most of the middle to make a "C" shaped channel that returns to the starting point.  Then one could use a blower to continuously circulate air through the channel.  One could also put a desiccant in series with this to remove water.  To recharge the desiccant, swap it with another container of desiccant in a solar oven.

In theory one could use a single blower to serve both functions of circulating air and pressurizing the system, and the blower might even run at different speeds depending on which function it was carrying-out at the time.  The amount of air inside could be adjusted with some sort of inlet valve connecting to the vacuum side of the blower.  (If an inlet check-valve needs to be left open all the time this is proof and an admission that the system leaks.  This also provides a way to measure how much it leaks.)  Anyway, assuming no pressure-drop across the inlet valve, and using atmospheric pressure as the reference pressure, the blower will increase this pressure by some amount.  There will be some pressure loss as the air flows through the channel, increasing with the air flow rate and the length and restriction of the channel.  Basically the system will reach an equilibrium where the blower pressure equals the pressure drop through the channel.  It may be prudent to have some means of throttling the air that is returning back to the vacuum side of the blower, perhaps with the use of an adjustable valve.  By partially closing this valve and increasing the throttling restriction, the blower could operate at a lower flow rate, thereby saving energy since hopefully all that the blower is doing then is circulating air for drying purposes.  
When air needs to be added to the system, the throttling valve could be opened up and/or the blower power could be increased, thereby increasing the pressure generated by the blower, essentially acting as a vacuum to pull outside air in through the inlet valve.  

-Mike

Toss my 2 cents in the ring

https://youtu.be/jBrc8e-Jqb8

Lots of information and more videos on utube this one has been up and running for a while now

http://www.lowtechmagazine.com/2015/12/reinventing-the-greenhouse.html#more

always love this site for very interesting articles

Love the arches, good luck

Ok team, I reviewed the design with some folks this weekend.  I've been assuming the "triangle" at the top will collect a lot of heat in the winter.  I forgot to realize that the glazing on it is at a relatively flat angle so more light may glance off of it than I originally thought.  The other advantages to having a triangle up there is that in the summer (when the temporary insulation is removed) the sunlight will reach farther into the greenhouse to keep the Northernmost plants getting light.  A second advantage to it was that it allowed the vent to be vertical and easier to seal against rain.

If I get rid of the triangle all together and make the building symmetrical, can I still get those benefits?  I can still have removable insulation to allow light in through that part of the North roof.  The vent can be the first 3-4' of the North roof and be transparent for summer use.  The rest of the roof can be opaque and insulated permanently.  Is there any reason to suspect that a vent at a 45 degree angle would be an excessive challenge?  I believe many greenhouses have angled vents so it must not be a problem.

If I get rid of the triangle, then any solar thermal collection via water in black pipes in the peak of the roof would be within the insulated envelope of the building.  So it wouldn't need to "drain back" and would be simpler.

If I get rid of the triangle, I could possibly butt the N and S trusses together at the peak and not have them bypass one another.  That avoids using carriage bolts in shear and is probably stronger.  I could use a 2x2 at the top and bottom of the trusses as mini ridge beams and they wouldn't cast too much shade.

So it seems like I should eliminate the triangle.  Any objections or other thoughts?  I'll sketch up a new layout and submit it as soon as I can.

Hi Ron, thanks for the links.  I had seen summaries of the Groundswell greenhouse but the excavation for it seemed a bit daunting for my project.  I also suspect that storing heat underground for removal in the summer is very well suited to sunny locations where they can "charge" the ground on sunny winter days.  It very well could work for me but I didn't want to do that much work and have it not work out.  I'm not sure it matters but I'm intending to plant trees in the soil of the greenhouse.  And I'm possibly aiming for higher temps than they require (50F+ on the inside).

If anyone can debunk my assumptions based on working examples in a cloudy and frigid area, please let me know!!!

The Chinese greenhouse was an inspiration for my design.  I can't really bury the North side (or build it up with a thick masonry wall) due to access needs on that side of the building.  I also don't think I could insulate the exterior of the glazing without the labor that they undoubtedly have available (when I'm gone there's no one here to maintain my stuff).  But that is what lead me to a curved truss and interior glazing insulation.