Sand heat store under a slab, worth the extra digging or just insulate and go?

I put a sand bed under my garage slab about six years ago, and my honest take is that it works but not the way the diagrams promise. The sand holds heat fine, the problem is that heat goes sideways too. I skimped on vertical insulation around the edges because I ran out of budget that fall, and I can trace a warm strip of ground about ten feet out from the slab on frost maps of my yard. That's heat I paid for and never got back inside.

If I did it again I'd spend the money on thick foam around the perimeter and under the whole pit, not just under the slab, and I'd put the pipe lower and closer togehter than feels neccessary. Mine runs on about two foot spacing and the top of the bed cycles warmer than the bottom, which tells me the sand itself doesn't move heat around as well as people assume. As for whether it paid off, the garage stays comfortable and the small heater barely runs, but I also oversized the solar on the roof, so I can't cleanly split the credit between the sand and the panels. My neighbor did plain pex in the slab over foam the same year and his shop is about as pleasant in January. His floor warms up faster on sunny days, mine holds longer into a cloudy stretch, which is exactly the trade you're describing.

Given you're allready excavating for a shop anyway, the extra dirt work might be the cheapest part. Just don't let the edge insulation be where you save money.

I went the plain route on my 24x30 shop slab five winters back, just 4 inch concrete over 2 inch foam with pex tied to the wire, and I have never once wished I had dug deeper. The slab alone rode through a February cold snap here in Minnesota where we did not see the sun for nine days and the shop never dropped below about 55 with no backup heat at all.

The thing that killed the sand idea for me was doing the math on the electric element. Cheap elements are just resistance heat, so every kilowatt you store in the sand is a kilowatt you bought. The solar input helps but a small panel array through a Wisconsin winter puts out a fraction of what people expect, and most of it shows up in December and January when you need it least.

If I were you I whould spend the excavation money on thicker foam under the slab and outside the footing instead. Insulation buys you the same smooth ride the sand promises, it never underperforms, and if you ever want more capacity later you can add a standalone tank or wood stove without touching concrete.

Not saying sand stores are bogus, they just seem to shine in bigger passive solar builds with real glass area south facing, not a little shop slab.

I'd push back on the 'closer together than feels necessary' advice. I helped a buddy with one of these beds and we spaced his loops tight because some forum said so, and all it got us was a boatload more pipe, way more pump head, and an air purging nightmare every time we drained anything. Sand conducts heat slowly no matter how many loops you cram in there, so once the immediate area around a pipe warms up the extra loops just sit there doing almost nothing. Wider spacing with longer run times would've been the smarter play.

Did a rock bed heat store on a house back in the early 80s when all the solar books were pushing that stuff, and the thing nobody warned us about was water. Sand wicks moisture, and damp sand moves heat out of the pit a lot faster than dry sand does. Mine sits high and dry so it behaved, but if your site has any water table to speak of, go look at your neighbor's well test or a soil map before you commit. A wet sand bed is a heat leak you can't fix. Also, on the charging side, if you're putting an element in there anyway, put it on a timer or thermostat tied to the warmest part of the day so it isn't fighting the peak rates.

Mine's under a greenhouse addition rather than a shop, and the thing that surprised me was how slow the whole thing turns around. You can't dump a warm day's sun in and pull it back out that same night, it behaves more like a seasonal flywheel. Mine finally got good and warm by March, then took till early summer to cool back down. If you do it, bury a few cheap temp sensors on a string at diferent depths before you pour. Watching those numbers taught me more than any of my calculations did. How deep are you figuring on digging? Frost goes down a ways up there.