GOAT logo

Join the conversation! The forum activity is now at GOATeach.org!  We are working to cross pollinate our conversations. Document and share tools at farm hack and talk at GOAT!  Also join GOAT riot and introduce yourself and your projects!

Cool room for spring storage

Topic Type: 

A while ago, I proposed a strawbale cold house, split in half, with one half filled with water barrels that were allowed to freeze, and the other half for cold storage.  A fan would move air through the insulated partition wall as needed.  The barrels would be frozen by running a fan from outside air during the winter.

However, according to one source even a partially filled barrel will bulge and fail.  I'm not impressed by the thought of three high stacks of round bottom barrels either. 

Not freezing the barrels (using brine) isn't a good option.  The vast bulk of the coolth comes from melting the ice.


Round 2.  Instead of a storing barrels, freeze a pool.

Dig a hole X feet deep.  Line with plastic.  Fill with water.  Build the shed around the pond.  

Inside you essentially have a deck over the pond.  Again, small fans assist the air exchange between outside during the winter, and between the pond and room during the spring.

Numbers:  Start with a 24 x 25 foot building.  Rubble trench foundation -- can be fairly shallow, I think as the subsoil here is very sandy.  Assume bales to be R30.  Assume ceiling to be R30 (In winter heat rises, so the ceiling should be better insulated.  In a cool house, you don't care.)  8 foot walls.

So our 5 wall envelope has 24*25 + 8(48+50)  = 600 + 800 = 1400  square feet.

At R30 it will take 46 BTU/F/hr.  Round up to 50.

Spring thaw here starts in March.  Average temp by the end of March is 5C.  By the end of June it's 25. Last frost is typically 3rd week in May.  Nightime temps are about 10 C lower.

So we will approximate it as a linear increase from 0 to 20 spanning 1 April to 30 June.

91 days (round to 90) average temp of 10 C = 50 F.  Want to keep the inside at 2 C = 35 F

Average delta T of 15 F. times 50 btu/hour = 750 btu per hour.  Round up to 1000.  That's 24,000 BTU per day.  Now a kg of Ice takes about 340 BTU to melt.  Warming up the ice from whatever it got down to is small potatoes compared to this.  (About 100 btu from -30)  70 kg of ice per day. Or about 6300 kg of ice for the season.  So if we had 8 cubic meters of ice we'd have enough, and about 25% over.

During spring breakup, I don't recall ice pans thicker than about 1.5 feet.  Without snow, they could grow thicker.  But lets assume 1/2 meter.  So 16 square meters, half a meter deep, gives us enough ice.  4 meters x 4 meters -- about 13 x13 feet.

The interior space, assuming a 18" wide bale+coating is 22x21 feet   We can put the edge of the pit 4 feet from the wall  

This allows us to bridge the pit with 16' doubled 2x10's on 2 foot spacing.  The floor above would be modular consisting of glued OSB 1" styrofoam panels I *think* this is enough insulation to keep ice at -30 from freezing the room early in the season.  (There is no convection.  Cold air will sit there. So all heat transfer is going to be by radiation.  Foil face worth it?)  Paint the bottom side white?

(2 foot spacing:  This allows reasonable working access to the pit for liner replacement, floor leveling, inspection...)

Whoops:  Our floor needs to several inches at least above ground level.  Better make that 10 foot ceilings.  That will add 200 more square feet to the envelope.  and increase the  the envelope heat loss by about 1/7.  Ok.  Make the pond 3 feet deep.  Outer dimensions 13x13, slope of 1, so bottom is 10x10

Styrofoam perimeter insulation.  Inside?  Or outside?  Inside we have frost several feet into the ground.  Since there is no snow insulation inside, I expect the frost depth to be close to twice the depth.  Further:  I expect the frost depth to increase from one year to the next.  (This spot will be at freezing long after other places have warmed up.  So the next year the ground doesn't need to cool as much.  I recall references to PAHS systems that take about 5 years to reach equilibrium.

Insulate below the pond and deck?  Would speed up freezing, but I'm no longer using the thermal mass of the ground.


One person suggested pop bottles.  That's 10,000 pop bottles to fill and stack.  No thanks.

Not sure how deep a pond will freeze.  Preliminary estimates and inquiries say that lakes freeze no deeper than about 18" here.  That is borderline for my use.  I could run a thousand feet of 1/2 line in the pond (divided up in 100 foot chunks in parallel) and similar amount of pipe on the roof, and circulate antifreeze.


One problem I haven't found a soluiton for:  Insulating the ceiling.  Originally I was going to put in a drywall ceiling and put a foot of celluose on top.  But the cold face is on the inside.  There is no practical way to put a vapour barrier in the trusses.  Is there an alternative to 4" of foam?