Commercial Radiant

PhillipJones

So here's one for you commercial radiant gurus. We're doing 19,000 SF in-floor radiant for service bays for heavy equipment. Think 8" slab with #5 rebar on 12" spacings, and 3/4"x500' loops on 12" spacings. The dilemma is the slab. Due to compression / weight ratings, the design team has chosen to omit insulation altogether. Realize that efficiency and operating cost are not in the picture. There will be (5) 299,000 modcons for the heat source, allowing about 75 btus/SF. I'm planning to send 2 gpm per loop and use snowmelt manifolds and use a UPS26-150 pump for each of the 5 manifolds. My hesitancy is where I have to place my loops in the slab. The rebar mat claims the middle, and I don't want to be above, with only 3/4" clearance from control cuts. Therefore, I get to put my loops on wire mesh at the bottom of the slab, below the rebar mat. With no insulation, and 8" slab above, I know my output is going to be greatly impacted. I'm shooting for a 140* setpoint, and will have the btus to get it there, but I just don't like it. Has anyone got any success stories with no slab insulation, or is it all bad news? This project is 2 separate bays- see attached loop layouts.

Rumpke, Struble Rd, Bulletin #2-Drawing.PDF

hot_rod

did you run load calcs with and without the underslab insulation?

Maybe showing them the operating cost difference would make them reconsider

Dow makes high load foam in 60 and 80 psi ratings. Standard stuff is 15 or 25 psi

The lower the tube in the pour the higher the downward heat loss and the higher the SWT. Much longer ramp up also.

The conductivity of the subgrade material matters also.

Definitely want edge insulation, maybe a 4’ band of flat around the perimeter if they go without underslab

GGross

Yeah that wouldn't fly in my area, we've done similar with thicker slabs, actually just recently finished one. No issue insulating the slab, sounds like the owner or builder is just trying to cheap out on the install to stick the contractor with responsibility. You do want to raise the tubing up during the pour because typically the pex manufacturer has a max depth for concrete. Load calc first either way.

hot_rod

get a signed assumption report once you present both options

Most radiant design software has the report included

PhillipJones

Agreed. So we're working with a "basis of design", where the PE specs a very generic design. He picked the boilers and tubing size, which I have no problem with. He also omitted the insulation, which is troubling, but I hesitate to butt in due to the structural consequences. I have played with the heat loss reports, but based on all the unknowns with infiltration, and massive doors which I know get left open way too long, the best I can come up with is to fake out LoopCad so that I create a worst case scenario in which I provide a system that will carry away all the heat the boilers will provide. There will be perimeter foam at least. I've tried to steer this toward insulation, but no one wants to buck the basis of design. I do plan to have a sign-off on our layout, and the design stamp is not mine. There is supplemental heat via overhead tube heaters, so I'm not on the hook for quite everything. I've read posts previous that deal with omitting slab insulation except for a perimeter band; maybe I'll see if they'll at least consider that. @hot_rod , thanks for the pointer on Dow Hi Load product. I'd feel better with an 8' band of 60 psi around the perimeter.

tim smith

GroundUp

We did a very similar system in a public works building about 15 years ago and 5 or 6 more in manufacturing facilities, without insulation (per the engineer, not my choice). The public works building is still on my service regiment so I am there a couple times per year and as much as I hate being wrong, the system actually does very well even without insulation below. We tied the tubing to 1/4" mesh panels (NOT the cheap rolls of wire mesh) and suspended it all on 3" chairs, then the rebar was tied on top of the tubing. Efficient, no. But it'll do just fine.