Fixing T&G floorboards - where to gap?

[RogerS]

Everything I read about laying floorboards talks about having an expansion gap around the perimeter of the room but to my mind that is just plain daft. How is it going to work if, as is normally the case, each board is screwed down through the tongue with Tongue-Tite screws or nailed ? Surely if the board wants to move then it can't because of the screws/nails. I can see the argument is valid for engineered type clip'n'clik type boards as they float.

So surely leaving a gap between each board makes far more sense although the purist in me wants to see them tight together.

But then you read about using floorboard cramps to get those boards nice and tight together.

 

[Sgian Dubh]

Kilned joinery grade wood is dried to 20% MC ±2%. For the most part there will be further shrinkage rather than expansion if the floor is laid in a habitable building, so it makes sense to fit each board tight. In the case of outbuildings, unheated or only intermittently heated, there's likely to be less shrinkage. A small gap at the outer edge makes sense, but any such gap will normally be covered by the thickness of the skirting board. Slainte.

 

[Jacob]

They need to be as tight as you can get them as in all likelihood they will have some more shrinking to do.
I recently did approx 100sq meters 6"x 20mm redwood. Kept them in stick for as long as possible after delivery. Laid them without nails, covered with Protec sheet and plywood to allow other work to go on. Cramped tight and nailed at the last opportunity. Result - at the moment they are fairly tight , a mm here or there. Mid summer the gap was wider and the floor a bit creaky. As good as you get really.
Gap around the edge is easier than close fitting and covered by skirting anyway.
NB Protec is really good value as it is re-usable and very tough. It's similar to Tyvek and other roofing felt.

 

[RobinBHM]

I think softwood floor boards that will be fixed directly to joists and used as a sub floor probably want to be fixed tight with no gaps. As above, floor boards will be about 20% mc and in a heated home will shrink.

Its probably easiest to leave a gap around the perimeter anyway and would prevent any issues if the house is left empty and unheated for a period.

As you say, floating floors in solid timber, absolutely require expansion gaps.

 

[RogerS]

No-one has explained how if each board is nailed or screwed down that an expansion gap around the perimeter is of any value. Each screw is going to limit the movement surely. It's not cumulative.

 

[Jacob]

No-one has explained how if each board is nailed or screwed down that an expansion gap around the perimeter is of any value. Each screw is going to limit the movement surely. It's not cumulative.

It doesn't have any value except it's a lot easier to fit with a random gap (to be covered by the skirting). Different with chipboard where there is more seasonal movement (I'm told) and a gap is probably necessary.

 

[RobinBHM]

I think the theory is that if boards are nailed / screwed in place then each board will shrink and expand between its fixing points, so each board acts independantly. However this assumes the boards are new when fitted and will shrink leaving a permanent gap in between, as their in-service maximum dimension is always less than the original size when laid. Im not sure that the fixing points would stop boards expanding if the moisture content was raised significantly -I expect somebody that has been flooded would know!

Certainly timber expansion has massive power; a floating solid timber laid with no gap has the power to push out a brick course or buckle up.

 

[blackrodd]

No-one has explained how if each board is nailed or screwed down that an expansion gap around the perimeter is of any value. Each screw is going to limit the movement surely. It's not cumulative.

In the days Before chipboard, T&G was cramped up as each board would shrink a small amount and as it should not prone to any expansion owing to the lower moisture content.
I have been into an upstairs bathroom where the 7"X 1" T&G had shrunk as much as the tongue, which was hardly in the adjacent board, I would imagine the timber must have been mighty wet!
when T&G chipboard first became popular, laying and fixing was demonstrated on site and we were told that the end joins need not be on a joist, anywhere in between was OK without noggins and supplied us with glue for this occasion.
Well, we now know that was a load of old cobblers!
Working on a block of flats, I have also seen a 45 gallon plasterers water barrel that was on a chipboard floor, on 2"x 2", laid on a concrete Bison Beam floor beams.
Someone had apparently forgotten to turn off the tap, and the floor, under the full barrel had expanded and lifted the barrel up about 10" or 12", was amazing to see, so don't forget the expansion joint!
Regards Rodders

 

[RogerS]

Thanks guys for all the input. In my case the floorboards are KD oak. When they came out of the plastic wrappers fresh from the factory they measured just over 4% with a Wagner meter. A few days later they were up around 7-8% and have remained stable at that.

I can see the logic that they may shrink a bit more but TBH at that moisture content I think not. So the question arises as to whether they will expand and if so by how much (and assuming I don't leave an over-flowing water barrel on them!). And whether to have an expansion gap around the perimeter.

Here's my take. The expansion gap around the perimeter for boards that are individually nailed or screwed down is an old wives tale...passed down over the years without any thought as to relevance. My reasoning twofold. First, take my boards, for example. 200mm wide. Let's assume they each expand by 0.5mm. I've got 13 boards so it's also reasonable to assume that they all expand the same amount. So that's 6.5mm to accommodate. Let's start at the middle and work out towards the walls. The ones on either side of the central one need to budge up a bit more to accommodate the expansion needed by the preceding ones. So by the time you get to the each wall we are expecting that board to have moved over by 3.5mm. Which means that nail or screw is also going to need to move it's position by 3.5mm. Is that going to happen? I don't think so. And if it did, then what? That screw is going to be ripped out, the boards will never go back down flat again and so that expansion gap was pointless !!

Second, if this perimeter expansion gap was going to be of any use then at least somewhere down the line someone would have said 'For a room with width x metres you need a gap of 10mm, for a room of xxx metres you need a bigger expansion gap'. But they don't so QED.

The answer to my own question has, I'm embarrassed to say been staring me in the face. I have oak floorboards on the landing and in another bedroom that were laid 10 years or more ago. They were all fitted as tight to each other as I could get. None of shrunk. None have moved. Certainly no cupping of individual boards. Tight it is and only a perimeter gap where it helps laying the final board down.

 

[Mar_mite]

I've been laying squash court floors recently. Apparently the correct way is to leave a 1mm gap every 5 boards if it's beech. Or a 1mm gap every 3 boards if it's maple. I wonder if it's because they are left totally untreated. No finish on then at all. I'm laying the floor on existing batterns which are only in 1.8m lengths with a gap if about 15mm between them (end to end). Why this is I don't know.

 

[busy builder]

When on new build up here with timber framed houses, the outside walls go up, then the roof, then the floors, then the inside walls. So the floors have internal walls fastened down on top of them, no gap.

 

[RogerS]

.... I'm laying the floor on existing batterns which are only in 1.8m lengths with a gap if about 15mm between them (end to end). Why this is I don't know.

That's interesting as they've not been laid according to the latest Squash Court specs which state butt-ended.

Your comment re a gap every so often makes all the sense in the world to me.