Roth Panels on an uninsulated on-grade slab

fortja

First time poster, have been searching this forum for thoughts and ideas.
Currently I am gutting / renovating a 1700 SF home in Vermont which is sitting on a 4" uninsulated concrete slab. I am planning on weather sealing and insulating all the exterior walls (2x6 construction) to R30 with closed cell spray. I will be trenching around the exterior foundation to 2' and adding 2" EPS to the exterior of the frost wall to just below the clapboard siding, and installing new windows and doors. I will be doing a Manual J to properly size all components. My current plan is to use Chiltrix Air to Water Heat pumps with their V18 backup unit to service the Roth panels. Roth documentation says their EPS is rated at 90 psi. and in my discussions with them, they say they don't have any studies showing the advisability of adding more insulation below their panels. They are concerned with deflection /deformation of the panels which may cause problems with whatever flooring is used.

Has anyone here any experience with adding more insulation under the Roth panels?

Thanks.

fortja

Any thoughts, comments on the pros and cons of what I am proposing above?

Jamie Hall

No experience, sorry. But I have to ask, before I start figuring from basic principles, what is the complete stack going to be like? That is, what materials and thicknesses between the existing concrete and the panels? And then what materials and thicknesses over that? There are two loading conditions to contemplate here: overall floor loading (usually 100 pounds per square foot, uniformly distributed) and point loading (the worst case being spiked heels, oddly enough -- the loading from a diminutive person wearing spiked heels is astonishingly high -- transport flooring is designed for a point load of one ton per square inch)

hot_rod

I think it would depend on how flat the slab is? The first foam layer would need to be stable, no voids below.

Most EXP foam board is 15 or 25 psi, 90 sounds high for that type of foam?

If I were to try it I would clean the slab well, maybe even a weak acid clean. Glue down the first layer with approved foam adhesive, applied with a notched trowel. Then glue the Roth on that lower layer, running the opposite direction, span any seams. I think if the two layers were bonded together and onto the slab it would work fine.

I'd use a plank type engineered "floating" flooring and have it span across the Roth perpendicular to the panel length.

I have some foam and Roth panels here, not sure what the test would be?

TAG

Am I to understand that Roth does not recommend installing the panel over anything -- only direct to the slab? Maybe VB first?

Question -- how cold is the slab as you move across the room?

I built a large single story addition to a suburban NJ house about 15 years ago. Cathedral ceiling closed cell foam 2x6 construction ... block foundation/ slab on grade. I wanted to do a mono pour .. but my code official was not familiar -- same with contractors. So we went conventional -- I had to do a foundation as well. We insulted the slab edge and the block wall outside with foam board. No insulation under the slab. I would do it a bit different today. As the project went forward the space was heated by flow through from the main structure and I was surprised that the slab was actually not really cold. It was around the edges ....colder ... not cold. My plan was always to take 2x PT lumber on the flats and fasten it to the slab. Install Warmboard S (they did not have the R at that time) on top and fill all the spaces between the 2x with 1.5" foam. 3/4 finished hardwood to match the rest of the house. This makes for a thick assembly and we had to get the slab perfect. 1 1/2" + 1 1/8 + 3/4. This addition was to a house with other radiant floors .... pex in concrete slabs w/ tile over it. Wood subfloor Pex / overlay concrete w/ tile. Pex with extruded plates under wood subfloor.. It worked very well and it had the next to lowest water needed. I also just did a small basement area using a VB/ Homasote and Warmborad R .. it's also working well.

When doing a radiant slab you typically use 2" under them .... I have been in well insulated buildings where people have used less and actually one where it was only used in the perimeter and not the center. Reviewing with the owners and seeing the fuel uses .... I wonder where the sweet spot is. I would never do a new slab w/o 2" .... but, I wonder what it really takes to isolate the slab from the ground. That's the goal in my mind -- don't heat the ground ... obviously, the higher the output needed from the floor = hotter water =higher delta ..this will have more heat loss potential to a poor insulted slab. My basement did not require very hot water ... so the small separation the Homasote type product provided was enough IMO.

It's like the old track houses built with slabs and no insulation -- they work with enough input. Why not look at the Warmboard site and do some research on the outputs of the R over Homasote or other like products. My guess is the Roth must be around =. Roth is plywood and WB R is OSB. The WB S is plywood.

I have done quite a few projects with the WB S ... I stick with what I know.

How much height do you have to work with ?

hot_rod

Is Warmboard R rated for direct over concrete application. I think most building codes want treated lumber when in contact with concrete? Or a rim seal foam product.

What I like about Roth is the 6" spacing. I'm a proponent of tight spacing for quick response and lowest possible SWT operation.
3/8" B pex is the most "workable" tube.

TAG

They say with VB and no moisture .... I used with the Homasote for the thermal separation. Both ways are given.

Having used the Warmboard in a bunch of projects -- You stick with who brought you to the party. The thick AL really transfers the heat and it has worked very well with hardwood floors. The S product makes for really solid floors. IMO -- the hardwood worry is overblown.

I'm doing 8/10/12 ... and a couple of king board 14" in my new project. Nice white oak .. Have done wide prior with warmboard with no issues. Red oak ... wide cherry. My last house was all 8" beach -- at the shore. Humidity is IMO a bigger problem.

It's also great under tile

hot_rod

I did some infrared testing on the Warmboard back 20 years ago. No doubt it does an excellent job with heat spread,
Now think of that same aluminum gauge, tight tube grip with a 6” tube space, that is what I like about the Roth, virtually no stripping and increased output with twice the tube moving the energy. Which also allows lower SWT for the same btu output. There are applications for both products 

TAG

Can't argue about closer spacing. Always do tighter than 12 in my slabs. But, interestingly the outputs and water temps between the few premium Warmboard like products out in the market are basically the same.

The Roth panel looks to be ideal for the OP's situation -- especially using a floating floor product on top. Modern glues are fantastic -- Glue and mechanical fasten Roth where necessary. Float the floor ....

The Warmboard S subfloor product has no real equal in the market. The Warmboard R is nice on a large remodel project where both may be used -- allows for same skill set.

hot_rod

Interesting how quickly and definitively the heat transfer stops where I routered across the aluminum layer on Warmboard to do a short loop for a demo panel. It clearly shows the power of a good conductor. Same SWT and flow rate.
The pex in ThermoFin is 8" on center, so basically you get a higher average surface temperature= higher output.

Here is an example of output difference in a slab with different OC spacing. I don't know if anyone has modeled 12 vs 8 or 6" on a dry system.

Cost vs benefit I suppose, I doubt closer than 6" changes as much. It is a bit tricky trying to get three runs of transfer plates in a joist bay.

If tighter OC spacing allows a mod con boiler to stay in condensing mode longer or constantly, the extra tube cost may be worth it.

TAG

Thermofin is a nice product and I guess if you wanted to go to the trouble of placing it above the subfloor ... it may indeed excel. I always use the extruded plates -- from the manufacturer. But, always under the subfloor as retrofit.

My experience: When you turn on the system ... the Warmboard is faster to heat and it radiates better given the same water temp even after running .. constant circulation. The Warmboard type products must have hot spots .. but, when you feel the panels they are warm all over

I used the thicker version plates this time -- trying to get a bit more output with lower water so I could match the Warmboard and not be way too hot for the slab in the lowest level of the project.

The OP should get about the same output -- the Roth is a tiny bit thinner AL plate and you do lose the ends. I always use ALPEX in the Warmboard

hot_rod

Agreed, so many options and they all have pros and cons. You can always find one or a combination that matches the application, and budget perfectly.

Of all the various methods and products I have played with, copper tube in ThermFin seems to be the very best transfer, two great conductors with a tight tube fit.

TAG

HR -- I think they have that on the web site. Plates w/ copper. That would be expensive ... both material and labor.


What are the other two staple up systems -- is it different pipe types only?

I hope if the OP uses the Roth he comes back with pictures. I have a studio to rebuild on this project and 1/2 of it is an old slab that I'm either going to take up or go over and finish somehow.

hot_rod

Under floor dry systems, lots of different plate types, wide, narrow, thick, thin.
I have used UltraFin successfully. If you have a high temperature fin tube and just want to add some underfloor warmth it works fine.

Plain suspended pex was promoted by Wirsbo for years, it too can work for low load areas. I consider suspended pex and UltraFin as more of a joist bay heating method, convection heats the air in the space basically.

So suspended tube as the least expensive maybe weakest transfer, UF a step up, any direct contact method would be better yet. Then, of course better but more $$ metal tube in plates. That is my good, better, best thought.
Floor as my first choice, walls next, ceiling as #3. You do get some stratification with radiant ceilings, all things being equal the walls offer a tad better output.

I have thought of EMT tube, or any steel tube in plates, easy to bend the return loops but the $$ vs the 10' lengths, labor wise may not be worth the hassle.

I bought a load of smooth style ThermaFin years ago, basically the copper tube used in solar collectors, copper sheets forge welded onto copper tube. I've been hoarding it for a special ocassion. Thermafin is part of AET solar.

TAG

Too many of these products sound alike !!

One is also afraid of the unknown ..... sadly there is much. All the parts needed for some systems were a turn off and worry. Would they be noisy. Maybe misguided .... but, I stayed away. Few people do radiant and of those that do ....fewer live with it. It's the same with high efficiency HVAC especially zoning .... few do it and fewer have it personally.

Way back in the 90's I used an extruded flat faced baseboard with twin press fitted copper tubes ...made by Embassy. It looked just like a baseboard wrapping the room ... did a walk out basement/ office. Worked great. I think they must be out of business.

hot_rod

There were a few brands of the flat aluminum baseboard, there were issues with the brass fittings in the aluminum turning int a dielectric corrosion issue. The aluminum would deteriorate and spring a leak. probably some non barrier pex tubing sped up the process

archibald tuttle

little reboot because this thread is perfectly titled. just what i'm thinking of doing since i can't justify jackhammering the slab and starting over.

@hot_rod "Most EXP foam board is 15 or 25 psi, 90 sounds high for that type of foam?"

I thought roth panels were EPS . maybe more than one kind. i'd prefer XPS. there are various chinese purveyors of foil faced pre channeled XPS panels in 2'x4' size (well 600x1200mm, close enough for government work).

pricing is really attractive but getting them here is not. is there someone who is importing these or making them here? xps has better density, better thermal properties and better vapor resistance.

most of these panels seem to be around an inch which seems underdone to me since 5/8" is lost at the channels. I would prefer something that is 1.5" thick but beggars won't be chosers. you don't need lots. I'm not asking for 3" thick. 1.5" is good because that is about the best value thickness vs. btu loss according to spray jones theory on this. this guy rocks. spend a lot less on spray foam now than I used to after I started listenting to him (which is to say I use a lot more spray foam because I can actually afford it at the thicknesses he recommends!. now this isn't a job for spray foam and I would only have like 7/8" under the channels is I could use 1.5". or I could put a layer of 3/4 xps under the warm floor etched panel.

to complicate things further, i'd prefer to deal with 3/8 pex but i'm a dinosaur in that department so I'm kind of resigned to having to this with 1/2 pex but stating the preference if channeled board for 3/8 existings. one advantage I have is I don't plan to turn the pipe in the floor. going to make thick hollow baseboard around the rooms so the runs will be all one direction with the pipe coming into the baseboard for turning or manifolding.

the other question I have is these often envision gravity holding them down. or maybe adhesive? I tend to think of putting hardibacker over the system with fasteners. i've done this before with PT 2x4 sleepers fastened to the slab with 1.5" xps between the sleepers and then a warmboard or insulated board with channels laid such that hardibacker as last layer on top can be fastened to the sleepers without fear of hitting the pipes. this is a faux slab kind of thing and gives a surface to adhere floor to. depending on the flooring I have sometimes gone with 5/8 underlayment ply and then hardibacker to try and strengthen that layer against movement if i'm going to tile. if it is something floated its unnecessary.

actually i have made my own xps panels for 3/8 tubing using a 1/2" box core router bit and industrial aluminum foil (like 4 times as thick as the stuff you buy in the market). but this job is big enough i'm looking at alternatives to DIY.

thanks for thoughts,

brian

hot_rod

Roth panels are either 3/8 or 1/2 as I recall
I thought they were made in NY?

you are correct there is not much foam under the tube

You could put 1” foam down first, then the Roth

I used foam adhesive on the jobs  I did

I would buy a cordless caulk gun if I were to do another large job

Unless the floor is perfectly flat I would use adhesive to eliminate squeaks

Between the layers also, maybe find a liquid adhesive that you could roll on?