I am looking to fit underfloor heating in a room of around 40m2 in a new extension. The floor slab is down and have plenty of room for insulation, heating etc before we get to the finished floor level. Looking at the polyplumb website, they seem to do 2 main types which could be used on the ground floor on top of the slab...
the traditional solid floor type - ie the pipes are laid in the screed
a "floating floor" type - where you lay some boards on the slab, put some metal heat spreader plates on top and fit the pipes into the plates. Over this you lay timber boards followed by your finished floor.
I have been told that the "floating floor" type is better if you want to control the heat easily as it is far more responsive than the traditional screed type. I guess it is more like a radiator and will heat up./ cool down more quickly than the screed.
Having had storage heaters many years ago when the room was either boiling hot or freezing cold, I was keen not to go down that route again.
Does anyone have the "floating floor" type fitted on top of a concrete floor? Is it any good? Am I worrying about the standard type unduely?
From a cost perspective, the floating floor one is at least 3 times the price of the solid one.
Mine is solid, and yes, it takes hours to warm up.
Conversely, it takes hours to cool down.
This is a plus and a minus: in summer, the rooms stay very cool.
In winter, you need to at laest maintain the rooms a couple of degrees only lower than you need when you use them.
My biggest mistake was calculating on a steady-state basis how much pipe I needed,. I went for 50W/sq meter. In very cold conditions this is a VERY long warm up time.
In the corridoor, where a LOT of pipes run VERY close together, warm up times are much better. ~200W/sq meter estimated.
So if you really lay in huge amounts of pipe, and have a suitably high peak output boiler, you can get very fast warm up times indeed. BUT beware the overshoot..where temps rise above the stat and go on rising..leading to SWMBO opening the windows..;-)
There is something to be said for high thermal mass/steady temperature versus low thermal mass/fast heat response: its really down to the use the room will be put to. If its a living area in constant use, go for high mass. If its a bedroom or workroom, often not occupied, go lower.
Oh, and all you NEED from polyplumb is the pipe, pump and manifold stuff, and maybe some zone valves. In screed I laid a grid of rebar, and tie wrapped the pipes to that. If I were making eg. a wood floor up, I would screed, lay joists on top, insulation between, and use notched bits of timber or bits of insulation to hold the pipes in place. Celotex in partiucular is foil coated and reflects radiant heat.
These clever ways to lay pipes fast have application in a commercial fit where stupid monkeys don't cost peanuts, but in a DIY context, pipe laying is merely a matter of ensuring the pipes stay in a reasonable relation to each other.
You should b able to fit a solid wood floor over the above, and provided you stick to ot more tha about 40C water temp, have a very decent heatng system. In screed you can go a bit higher, since the screed itself 'spreads the heat' but with simple pipes in an air space, the air will circulate and do the spreading. Remember that at a given temp under the wood floor, your watts per square meter are governed by how hot the underside of the boards are, and what the thermal resistance of them is. (and of course your final room temperature). There is a limit to how far you can realistically go. I haven't calculated it myself..lemme see, typical wood door is a U value of about 3.5, so a typical floor board might be twice that..say 7..watts per square meter per degree C. So to get 100W/sq meter you need at least you need 14C drop..from a room at say 20C, thats 34C..and then you need to allow for carpets and rugs..
well its doable, but not exactly far from the limits. Which is why a tiled screeded floor can produce more heat output..
Dear Lee I echo all that the NP has said. I would add the following things a) I personally found for my ground floor living rooms that it is best to have high thermal capacity with a low input temperature (I use a heat pump at between 25 and 33 C max for my input temp) and live with the resultant slow heat up and slow down times which do not inconvenience me at all. If - as is the case - SWIMBO complains that it is too hot sometimes and too cool others (when it is exactly the same temp!) I merely say put on the thermostatically controlled electric heater for a quick blast. b) I would never in a month of Sundays use a wooden floor - it would have to be ridiculously thin for the heat to get through quickly and is a contradiction in terms laying what is effectively an insulant between the heat source and the room (OK before the Pedants all reply I am well aware that it can be done but it is clearly better to have a conductor as a floor surface). I used tiles and if it is a living room you can use slabs like York Stone or the like wiht the occassional thin rug c) I used Kestrin for my design and found them superb - material and service wise but if price is the be all and end all then you have little option but to DIY the design. Kestrin use Uponor products and larger pipes (22mm) at about 150 cc which we have found to be fine
We were somewhere around Barstow, on the edge of the desert, when the drugs began to take hold. I remember snipped-for-privacy@yahoo.co.uk saying something like:
I fitted one of those seven-ish years ago and it was cracking. Heated up fairly quickly and cooled down reasonably so, so you avoid the boiling storage heater effect you describe. Approx 60sq metres of flooring, heating a barn conversion. German in origin, iirc. You must rememember to stuff underneath the spreader plates with rockwool.
That's the rub - it's not cheap, but it's effective. On balance I'd deffo go for the floating floor heat spreader type as it's a one-off cost and you'll be glad you did.
We do exactly the same. The downstairs living areas are high thermal contant, but we have open fires for a quick blast. Huge ones.
It wasn't practical to fit UFH upstairs really. I have fan convectors and teh odd radiator in bathrooms etc etc.
WE seldom heat them at all...the UFH ges upwards nicely. And ou dint need them to be that warm. Currently SWMBO hits the 'one hour boost' button a half hour before a shower..
I have screed with engineering floor over.Including a foam underlay. It ain't optimal, but it works.
A friend has uninsulated copper pipes under floorboards under tiles in a bathroom. Accidental UFH, but works very well indeed.
The issue with using non optimal materials between pipes and riom are to first make sure that the U value DOWNWARDS is at leats 1/10th of the U value upwards, to achieve 90% heat transfer into the room, and secondly that the actual U value upwards does not necessitate very high temperatures underneath. A combined U value of much less than 7 is probably never going to be satisfactory.
Ber in mind that much conventional thinking about flooring is predicated on the assumption that the floor, and floor level draughts , will be the coldest part of the room. Ergo shag pile fitted carpets, 'warm' wood, and the like are Good Things. With UFH the reverse is true and teh more conductive the floor, the better radiator it will make. Hence stone and tiles are preferred.
Ive just dug out my bookie wookie.
It gives a k value for timber as 0.14, so 19mm of wood has a u value of
7.3 appx.
My engineering floor is less than that..probably 10mm so will have a U-value of 14 or so, but the 1/8th of foam underlay - say 3mm at 0.03 or so has a U value of about 10, bringing the overall 'to the bare wood' value down to about 5.8, and then if you add the rugs..its no wonder I don't get rapid heating. At 50W/sq meter I am up around 10 C over ambient at the pipes..say 30C..or more..
Which shows overall that it isn't the wood floor that is the problem, so much as the foam underlay and the rugs and sofa on top!
Well I used polyplumb, and frankly the pipe size and material aren't a huge issue as long as they don't degrade in the floor! Its the manifolds and flow valves and temp reducer that was nice with poly.
For a suspended wood floor, you might argue that copper pipe gives better heat transfer..but the price!!
At the end of the day all a UFH system is , i a damned great radiator laid on its side and built into the house: it needs inuslation between it and the outside world, and good conductivity to the room its to heat, and to be built to lats.
The rest is just about how you control the temperature inside it to avoid breaking it.
Only if you have a perfect insulator underneath will every last bit get through eventually, otherwise some of it will be wasted going in the wrong direction. But we have underfloor heating and a wood floor downstairs (in most rooms), and it works. It works under carpet too.
The floating floor you describe is very similar to the situation you meet fitting UFCH upstairs where the pipes sit between the joists, directly below the floorboards. I used the PolyPlumb manifold, pump and mixer along with thier spreader plates but with speedfit tube. The plates really do spread the heat, they are very thin aluminium which provide a very heat- conductive path sideways. The system responds fairly quickly and no- one has complained about the rooms not heating up fast enough. I fitted Celotex under the speader plates (20mm in the parts between the tubes and 40mm under all of it). As others have said the important thing is to control the ratio of thermal conductivity downwards and upwards. The plates and the Celotex helped keep the tube in place while the boards were fitted too.
My next job is to fit my downstairs system which will be a high thermal mass "traditional" type. Once again the trick is to have lots of resistance downwards (lots of Celotex) and much less resistance upwards (in my case a tiled floor because that's the finish we want anyway). I won't be using PPs spacer system, I can see why it makes sense for professionals for whom time is money but my time is free at point of use and I can't see why (with a little planning and care) I can't clip the tube to the celotex - after all it only needs to be held until the screed sets.
As a complete OT aside I used the same principles to make an in-wall CH radiator behind the mirror which forms one wall of my shower cabinet. I fitted a 20mm slab of Celotex and groved it slightly to accept 10mm Speedfit tube which runs in a double serpentine. A sheet of ply on top and the mirror glued on to that (I got the glass company to do that bit - I was scared!) and I have a shower wall which gives off a little heat but more importantly which doesn't stay steamed up for very long. It's on an inside wall so any heat getting through the Celotex isn't wasted. You might notice there is no spreader plate involved and it's really easy to see where the pipes are as the condensation burns off.
I am thinking of making in wall panels myself for where the floors are not deep enough, due to exposed beams..
Even in screed, you can see - as I discovered whne I PVA'ed mine before levelling it, where the pipes run by the dry patches. Screed ain't a LOT better tan wood, conduction wise.
I must say that my aim wasn't to make a radiator in the sense that I wasn't looking to heat the room. I was looking for a way to keep a
2m2 mirror from steaming up. In-wall heating is becoming trendy in Germany (at least) apparently although I can't help but worry that the wall is an obvious target for a drill... I figured that I didn't need to worry about anyone ever drilling into mine as no-one's stupid enough to drill into a mirror... are they?
We were somewhere around Barstow, on the edge of the desert, when the drugs began to take hold. I remember snipped-for-privacy@chiark.greenend.org.uk (Alan Braggins) saying something like:
I just knew someone would pick up on that as soon as I hit 'send'.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.