ECN Forum Forums General Topics of the Trade Non-US Electrical Systems & Trades Power Boost to Floor Heating

James,
According to AS/NZS 3000 this sort of a transformer set-up would not comply.

Also bear in mind that connecting a device like this up, is Prescribed Electrical work (NZ Elec. Reg 17) and would need to be tested and certified before connection to a supply.

[This message has been edited by Trumpy (edited 08-06-2005).]

"DO NOT DO WHAT YOU PROPOSE!!"

____________________________________________

Good advice

Thanks Trumpy and others. Installing a heater ( to the code ) now rather than start again. We are still puzzled though why the installation would not perform ! Only thought is that the element run itself was spaced slightly too far apart ?
Kiwi2

Underfloor heating's generally not a great idea to be quite honest. I wouldn't use it at all if avoidable.

The heat output entirely depends on the make up of the floor and the floor covering. For example, you could be loosing most of the heat into the foundation of the house if the concrete isn't holding much heat.

Also, as suggested the tile adhesive or the tiles may be insluating the floor too making the heat transfer pretty poor. (particularly if the adhesive is put on very thin behind the tiles as you effectively create an air cavity insulation chamber!

Other reasons for not installing underfloor heating:

(1) Under wooden floors it creates a lot of problems as it tends to dry out and warp the wood. It can shrink, get brittle, change colour etc over time.

(2) In bathrooms, where it's usually used, it can create horrible smells. If the bathroom floor is in any way not 100% clean it can really create a whiff. Particularly where a toilet is present. i.e. any splashes will be evaporated.

(3) it's woefully ineffective in many cases and very unreliable.

(4) if it goes wrong you need to dig up the floor!

Personally, I think just heat your bathroom conventionally and don't rely exclusively on underfloor heating. It's a nice gimic to create warm tiles but, that's about it.

On a purely technical level, your idea of boosting the mains with a step-down transformer may not work too well anyway, due to the primary to secondary phase relationship which depends on load.

djk; Oooer! Well, I hope my bloody system works then, I'm £2500 ($4000) into the build for materials alone already! You are obviously somewhat disappointed with your system, but to answer some of your points:
Wood floors, absolutely granted, some shrinkage will occur, but I had my cedar boards kilned locally to 8%MC at a mill, after 4 years air drying, was able to do this as I bought the tree after a storm and had it milled. This should give them a c.2% MC region of non-movement, by some quirk of the kiln process. This also seems to have made the wood more brittle for some reason, a PITA on machining. I have deliberately down-rated the 2nd floor water temperature to 35c to ameliorate this problem. My back up position is the boards are t&ged, but screwed via the faces like a ships deck, and plugged, so it's possible to recramp, but not something I want to entertain!
Smells. Blimey! I hadn't considered this! I'll have to improve my aim, and cork tiles may have to be ceramic instead. Good point.

Ineffective/unreliable. Not the same thing. Any ineffectiveness must stem from poor design; ie. inadequate house insulation, poor design of the pipe runs, -(DO; run the first length of pipe from a feed manifold near the walls, tails to the middle of rooms; minimised spacing (8" diameter bend on 15mm pipe) maximises power; all loops in zones are 100m long, giving intrinsic balance from manifolds without excessive throttling, ball-jacks (valves) for full flow when open, automatic air-bleeds on all horizontal runs, a pump for each zone, a programmer for each zone, thermostat(s) for each zone ). Plus, inadequate design of the ventilation is another failure cause. All of which mean the power output of 100W/m2 max. can't cope. This problem is accentuated in bathrooms where you feel colder, there's a lot of vapor to dispel, and a small net floor area. I am addressing this with heated towel rails.
Unreliability comes in two areas; wrong sort of pipe, ( copper in cement will leak by corrosion within 10 years), and poor circuit design, components or materials. I don't see any difference to conventional systems here. I have a neighbor with a system built 8 years ago, (plastic pipe, oil-fired) and it performs beautifully- it was what sold me on the idea initially- his house beng so well heated in winter. I must ask his wife if her bathroom stinks!
Digging up the floor I have no fittings in the pipe runs in the concrete, and just 4 unavoidable in the wood-floor area to the manifolds. The pex-crosslinked plastic is incredibly tough. One fitting failed on air-test, & it turned out I damaged the 'O' ring on assembly by snagging it on the copper lining bush. Provided I don't drill any holes in daft places, I might be ok! Presently [this afternoon] it had held 1.5 bar (22psi) for the previous 36 hours on compressed air. I will be filling 2nd floor pipe run next week with 50% anti-freeze and corrosion inhibitor. I'm not looking forward to this, as the 1st floor pipe run was a sob to get all the air bubbles out!

I see drawbacks as; no open fireplace, (I rather fancied a big hearth with firedogs), the need, as with storage radiators, to foresee tomorrow's weather conditions,- you can't heat 12 tons of cement instantly on demand! - so once winter starts, it's on for 5 months! I have added a detail where in the night the 1st floor 'sets-back' to 10C on a thermostat. Wood floors heat up much faster. You also need really good insulation- 8" of glass wool or equivalent. Ventilation needs careful assessment to avoid excessive losses- I'm fitting centralised ducted venting which will run 24/24, using an 'S&P Akor' air-air heat exchanger and twin centrifugal fans with auto balancing, to recuperate about 50% air-heat or better, at the rate of 1/2 ACH ( = 250M3 per hour), run on a programmer for boost/run/trickle. Cooker hood will exhaust direct out, (grease problems!) Also going in is a 20,000 Btu reversible Fujitsu AC, which will allow 4kw of instant heat to be obtained in cold snaps if needed, and if I dun me sums right help keep it a bit cooler in summer. Plus, the plan is to fit 2 humidifiers (1 on each floor level) and H'stats to keep air levels between 30-50% RH, (still looking for suitable models), to improve the 'dry-throat' sydrome. All this is only feasible due to the insulated floors (2" of ext ps under the concrete), and 8" of Rockwool behind the sheetrock, and Pikington 'e' double-glazing. The theoretical heat demand, including venting, is just 3280Btu, less than 1Kw in a -1C frost and 21C (70F) inside. So, wish me luck chaps! I'll be posting a few pics shortly, I'm sure you could all do with a good laugh at my crap wiring!

Alan
PS. Air-spaces under ceramic tiles. These are easy to find. The tile cracks under the first person to stand on it. My French tiler has a good system. He lays tiles on 1.5" of mortar and beats them into submission with a rubber mallet. Thinset? Never heard of it!

[This message has been edited by Alan Belson (edited 08-07-2005).]

[This message has been edited by Alan Belson (edited 08-07-2005).]

I recently went through the "certification" for one manufacturer of under-floor heating.

Oddly enough, he focuses almost exclusively on the "bathroom" market.

His particular system uses a fabric-like mat, custom made for a particular floor- hole for the toilet and all. The material is also primarily intended for placing under a ceramic tile floor, though there are methods suggested for placement on other floor systems.

Of greatest interest were the extensive testing required at different times during the installation. The testing was centered upon megger tests, with documentation at each step. As you might guess, this is, in part, to find problems before you have to tear the whole thing up.

Alan : ... Oh Boy what a setup . I will be most interested with the results of your new heating system. From what you say I think you have more than ' dun your homework '. I look forward to the photos.Incidently, I think maybe my tiler was like your french man,I would imagine 1.5" of grout would certainly have negated a decent heat output from my bathroom installation ??
Kiwi2

I've installed a few Underfloor heating systems as well as Under-tile systems.
These have invariably been of the electric element type.
Some examples:
Warm Up NZ
And the one I've used a bit:
Devi-Flex

Some pictures of the Devi-flex system, before the concrete slab is poured:

The cold tails and sensor tube:



The element wires correctly fitted to the re-bar:



John (Reno),Is quite right about the testing required before, during and after installation of this sort of system.
It should also be stated that the main cause of failure of an electric underfloor heating system here in New Zealand, is by the failure of the installing electrician to fit Surge Diverters at the switchboard to protect the UF Heat system.
First lightning strike on the mains nearby, will cause the elements to irrepairably fail.

jamesnz,
Thickness of concrete/ mortar/ tiling over the heat source does affect output. Here is a table published by BSRIA. http://www.bsria.co.uk
The correction factor is applied to the design water temperature to achieve the surface temperature. eg. a 4% cf = increase water temperature from 40C to 42C.
Use of electric elements would have the same effect, ie reducing output because power is fixed.

concrete.....correction
over pipe.......factor
25mm, 1".......... 0.92
30mm, 1.25"........0.96
35mm, 1.4".........1.00
40mm, 1.5".........1.04
45mm, 1.75"........1.08
50mm, 2"...........1.12
55mm, 2.2".........1.16
60mm, 2.4".........1.20

So 1.5" of mortar, as opposed to thinset, will degrade output by 4%. It will also increase the inertia of the floor, increasing heat up time.
So I don't think thickness of screed was your problem.
Heat output from a floor depends on the finish material, for instance-
Up to 10mm ceramic tile- no cf needed
Stone, 20mm, cf = 106%
plastic tile, 5mm, cf = 115%
berber carpet, 8mm, cf = 142%
parquet, 10mm, cf = 130%
As you can see, a bathmat will have more effect than a few more tons of cement!

Mike, that looks a good way to support the elements/pipe, off the rebar. I didn't use rebar, just a floated 4" slab, but sectioned the floor off into 5 metre lengths with expansion joints. Pipe is fixed to the insulation boards with clips. Never thought about the lightning problem- worrying for the electric heating owner!

Alan