I currently have a 12x12 insulated shed on slab on grade. I use it to store seedlings during the planting season. I pre-chill the room by opening it in winter and using a box fan to circulate air between inside and outside. In addition I have three quarter full barrels of water. There are 12 of these in the room taking up over 1/3 of the floor space.
This works. This year I had ice in the barrels until mid July.
I need to build a larger room. I'd like to go to at least 24x24.
My concern is that I am deliberately allow the foundation to freeze -- indeed, the whole point is to get the foundation to freeze. All the conventional wisdom is that buildings need to be supported on something that stays put, and that freeze thaw cycles don't guarantee this.
This hasn't been a problem so far, but perhaps I'm just lucky. The building footprint is small, so if the slab shifts as a whole, I may be getting away with it.
Yes: I know that conventionally I would put in a massive chiller and insulate the floor too. I only need cooling for 6 weeks a year. Conventional actions seem like overkill. This would be a hundred thousand dollars, and a new electrical service. (Our farm currently is on a 7 KVA service)
My current thinking is that I need to ensure that the foundation volume is well drained. Shifts are caused by the expansion of water into ice. If the water content is minimal, then the foundation will stay put. I'm leaning toward a rubble or rock filled trench for the foundation,E.g. Dig X feet, line trench with infiltration cloth, lay weeping tile with a sump, fill with rock, top with road crush, and set my foundation.
I thought of going the pole barn route:
Auger deep holes Y feet deep, and use poles as the support structure, and NOT putting down a concrete floor inside. The poles would need to be set with a slip layer (several layers of 12 mil poly?) on the part of the pole that was in the freeze thaw zone. Walls themselves would need to be designed with a potential sacrificial layer for the bottom to accommodate rising and falling ground levels. (Some form of compressible insulation?) Not sure what the size of Y should be for this application where I'm trying to get the ground to freeze.
Can someone point me to resources for designing foundations that are supposed to freeze and thaw?
Problems with conventional design
All conventional designs are based on the notion that the bottom of the foundation is below the frost line, and indeed, that enough more is below that frost movements near the surface don't grip the foundation hard enough to lift it.
In using the earth as part of my reservoir of coolth, I am deliberately trying to extend the frost line deeper into the earth. I don't understand frost resistant design well enough to figure out what changes are necessary.
Edit to answer @Freeman's comment:
I'm looking at either 20 x 30 or 24 x 36. Might go 30x30. Internally I want a COLD room and a not-so-cold room where I can pack.
The present system is vulnerable to early spring. While I had ice in mid July this year, barrels don't have enough surface area, and even with a 3000 cfm fan circulating the air, I was getting temps of 3.5-4 C. This year we had a late spring. with the creek opening up almost a month late. The barrels also take a lot of room. If I can put the ice under the floor, it doesn't get in the way. If I have a backhoe come in for the foundations a pit for an ice sump is cheap.
12x12 is close to the usual 'score the concrete every 10 feet so that the cracks occur in known locations' I suspect my present slab is of a size that if it moves, it moves as a unit.
My present building is on borrowed time. It has OSB internal sheathing, plywood external sheathing. I suspect this is turning into a sea of mould from constant 95% humidity. While I have the door open and a fan running from July onward, I doubt very much that I get the interior of the walls truly dry. Next round, I think has to be double foam layers, overlapping edges, glued and taped, with the foam interior to the building frame.