I dont see the batten approach as a problem, the heat will still go through , and pipe sizing /flow temperatures /flow rates can all be optimised

I would not like the wood to see any high moisture levels from underneath -- fail safe design for me would be dpm, sorry - ft the wood gets damp from under it will fail

If the house is warmer than the ground then moisture vapour will move from the house into the ground , floods, spills and leaks are all risks

I'd think the higher temperatures in the slab needed to get the same heat flow with the extra thermal resistance of the air gap will result in more heat losses downwards into the ground. How much difference this will make will depend a lot on the characteristics of the ground, particularly how wet it is and, if it is wet, how much the water moves.

yes, min 300mm eps/foamed glass

Posted By: tony-- fail safe design for me would be dpm,

Thanks Tony, but... similar to others on the site, we've bought in (rightly or wrongly!) to the idea that a (non-breathable) concrete slab drives ground moisture to the sides and up the wall. Whether this actually happens or not has been questioned by some, but equally, from my research many experts seem to believe it does. Either way, we needed a (new) heating system and have opted for a Ground Source Heat Pump with UFH, so had to break up the uninsulated concrete slab and replace it with something; the Geocell (foamed glass) + limecrete solution ticked boxes on a number of fronts: environmentally friendly, sustainable, overcomes the "rising damp in walls" issue (if it existed). So whilst I accept that a DPM would be the safe bet from the wood's perpsective, it would also, at least partially, defeat the purpose of the breathable limecrete.

djh and Ed: yes, I was also thinking that putting the UFH pipes closer together might be one way of countering the reduction in upward heat due to the air gap if we went with the battened solution. One to discuss with the heating engineers I guess!

Of course the thing we don't know is how damp (if at all) the surface of the limecrete might get. As the Geocell is non-capilliary it shouldn't draw up moisture from below; so once thoroughly dried out and heated it should, arguably, be pretty dry.

Does anyone have real world experience of either (i) wood rotting or (ii) everything being fine when engineered wood has been floated on limecrete? This is certainly my preferred solution (for heat transfer reasons) but if I go down this route I'd like to have some level of confidence that I'm not going to have replace the floor again in 10 years time!

Posted By: tonyWater is more dangerous, floods leaks and spills.

But that also moves downwards, unless there's a DPM, and ceases to be a danger. And outwards through the doors in the case of floods, with the assistance of a brush.

The Geocell is a capillary break so the limecrete should indeed dry out: but very, very slowly. Maybe a year or so? If it's possible to loose-lay the floor and live with it like that for a while, that's what I'd do. Then you can check for problems, and adjust the plank spacing, before making final fixings.

For UFH that does not reject its heat into the floor effectively, you can end up with the supply and return temps at the HP being very similar, so the HP shuts itself down before sufficient energy has been emitted to the rooms. I've seen this a couple of times on projects I've been asked to visit after the "experts" couldn't get the heating to work, blaming the HP as being at fault. One was exactly as you described with an air gap from pipes to a wooden floor.

I'm wondering if the premis of high vapour movement and timber floors is not at odds? Thin timber boards will suffer a certain amount of vapour. Is that tolerance level above or below the vapour transfer levels you have in mind. No idea how you could ever know what those numbers might be, as there are so many varying factors on either side of the floor boards.

GreenPaddy and Ed: thanks for your comments. I'm not entirely sure I follow them completely but my comments back are:-

Posted By: Ed Davies

Posted By: GreenPaddyI'm wondering if the premis of high vapour movement and timber floors is not at odds? Thin timber boards will suffer a certain amount of vapour. Is that tolerance level above or below the vapour transfer levels you have in mind.

1) Yes, I'm using engineered wood which I believe should be tolerant of some level of vapour movement - but I have no idea what or how I would ever measure it, so that seems a bit academic.

Exactly, it seems weird to put down a sub-floor with the intention that it'll transmit water upwards then lay a finish floor across the top which might or might not be damaged by that.

I've no idea what would be a good vapour-open finish floor, though.

2) well, yes, elsewhere in the house I'm laying limestone flags which obviously won't suffer any issues from damp (that I'm aware of - someone might tell me otherwise!), but lots of people have, I'm sure, used wood over limecrete successfully (otherwise where are all the people saying "NO - don't do it" ?!). I think I'm going to take my chances with the underlay - though will discuss further with the heating engineers later today...

Posted By: owlmanHad you looked at Schluter Ditra or similar, de-coupling mats. I know they are intended for tiling over indeed I've used them frequently but they MAY?? allow the engineered floor to be floated over them.

I think Graham's idea is to have a breathable (vapour permeable) construction, not another vapour barrier? Hence the Geocell and limecrete plus engineered timber and limestone flags.

Posted By: djhI think Graham's idea is to have a breathable (vapour permeable) construction, not another vapour barrier? Hence the Geocell and limecrete plus engineered timber and limestone flags.

Thanks for the suggestion owlman but djh is right on both counts: we want a breathable floor and Ditra appears to be a waterproof membrane, so unfortunately a non-starter for us.