Speranza Posted Thursday at 20:57 Posted Thursday at 20:57 Hi, frequent lurker, first time poster... We're planning some retrofit work on the 1950s brick house we moved into a couple of years ago. Ten years ago the previous owners spent a lot on upgrades and fitted a quality external insulation system on the walls, along with various other work. The floors though, were left untouched. Current construction is timber floorboards on joists over a crawlspace with dwarf walls. All in good condition but uninsulated. We are installing an ASHP, and want to go with UFH for the downstairs. For various reasons involving material preferences and low stack heights, we are thinking along the following lines: floorboards removed, windtight membrane laid into the joists holding 100mm hemp batts, then a light OSB deck between the joists holding UFH clipped in biscuit screed, finished flush with the joist tops. Generally it looks like this will do the outputs we need, and we plan to lay structural solid timber back down over the top in most areas. The hallway however is marginal on achieving the required output without pushing flow temperatures higher then I'd like, and so we are thinking about floor tiles. But how to do the structural floor? I've looked at plywood, but it doesn't sit well with me to put a fairly insulating material between the screed and the tiles. I looked at cement boards but the conductivity doesn't seem a lot better. Mortar screed on the other hand is ten times better.... So I had a mad idea. How about laying industrial steel grating (this kind of stuff) over the joists as the structural member, then mortar screeding it flush to give a solid surface for tiling. Great conductivity, very rigid for successful long-lasting tiling. A bit pricy, but it's not a huge area to cover. Obviously I need to check the joists for the dead load increase, but our spans between dwarf walls are short so I think it is probably OK. Any other pros/cons I haven't thought of? Or an obviously better solution I'm missing? I couldn't find any evidence of anyone trying this, which is not necessarily a good sign.... Thanks! Richard
Conor Posted Thursday at 21:03 Posted Thursday at 21:03 How deep is the void under the floor? Is there an option to rip out all the timber, lay and compact hardcore, a load of EPS insulation, then a screed floor with UFH? UFH on a suspended floor with only 100mm wool type insulation is going to be expensive to run.
Speranza Posted Thursday at 21:15 Author Posted Thursday at 21:15 The void is deep. House is on a slope, with about 1.2m void at the front, dropping to maybe 0.4m at the rear. Also services and wiring run in the space, so not straightforward to fill it. I take it you're concerned that running pipes at say 40°C with only 100mm of insulation between them and the void is a recipe for too much heat heading downwards? There would be a possibility I suppose of adding more insulation below the joists. Alternative would be to stick with radiators I guess (lower loss through the floor as lower delta t across the insulation) but might then need a higher flow temp so lose that way instead.... We have found this such a minefield of tradeoffs once you get into the detail!
marshian Posted Thursday at 23:09 Posted Thursday at 23:09 Similar situation with my house suspended ground floor with decent sized void under it - 75mm kingspan PIR between the joists made a massive difference to energy bills - rads all rooms no UFH Running WC flow temps max 34 deg c at -2.5 Don't discount rads as an option
Nickfromwales Posted Thursday at 23:28 Posted Thursday at 23:28 Any help? https://lewismetaldeck.com/our-solutions/flooring/
Speranza Posted Friday at 11:11 Author Posted Friday at 11:11 12 hours ago, marshian said: Don't discount rads as an option Yeah, there are advantages I'd not really thought about with the reduced floor temp. Though freeing up the wallspace is a big incentive too. Hmmm.
Speranza Posted Friday at 11:12 Author Posted Friday at 11:12 11 hours ago, Nickfromwales said: Any help? https://lewismetaldeck.com/our-solutions/flooring/ Thanks, I had spotted that yes. I was trying to achieve the same effect but with a reduced thickness. The Lewis deck is at least 36mm I think, whereas with grating I think I could do it in 20mm.
Nickfromwales Posted Friday at 13:49 Posted Friday at 13:49 2 hours ago, Speranza said: Thanks, I had spotted that yes. I was trying to achieve the same effect but with a reduced thickness. The Lewis deck is at least 36mm I think, whereas with grating I think I could do it in 20mm. I doubt it will be constructional at 20mm.
Speranza Posted Saturday at 14:33 Author Posted Saturday at 14:33 (edited) On 20/06/2025 at 14:49, Nickfromwales said: On 20/06/2025 at 12:12, Speranza said: Thanks, I had spotted that yes. I was trying to achieve the same effect but with a reduced thickness. The Lewis deck is at least 36mm I think, whereas with grating I think I could do it in 20mm. I doubt it will be constructional at 20mm. I haven't come across any useful tables for SI sizes yet, but this US supplier has a 3/4" grating with tables that look like it would be fine (27kN/m2 UDL at 2' span with only 2mm deflection!). After all, typical spans in steel structures would be much bigger than usual timber joist spacing. Edited Saturday at 14:36 by Speranza
Nickfromwales Posted Saturday at 20:29 Posted Saturday at 20:29 You may have to get it signed off by a UK SE and then BCO if it doesn't have accreditation for use here / in this application etc.
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now