Fruity tangent type thought - floorboards

Jan 11, 2011 28 Replies

The recent discussion on floorboards led me to one of my many off the wall thoughts:



Why do we use T&G floorboards?



Why not a half-overlap joint which would allow boards to be lifted without destroying the joint (let's assume we screw them down)?


I was going to use square edged boards upstairs (they *must* be readily liftable) but it got me thinking - wouldn't be hard with a router to pop a short overlap joint on both edges - maybe including a slight round or tiny bevel on the edge too for visual interest.



It would achieve the stated aim of not alloing draughts up or crud down and would have some of the mechanical interlocking of T&G for rigidity (half).



???



It keeps the boards in-line when distortion occurs, the method you suggest would mean that any bowing etc would not be controlled along the length of the board by the T&G. IYSWIM

I think your slight round or tiny bevel would collect dust.

Cheers Richard

because those would bow and cup and create ridges. And let draughts through.

T & G pine is the poor man's alternative to chip[board.

see above.

Why make em liftable? cant you do a solder joint or a wiring job that will last 30 years?

Wrong on both counts.

Because only the board with the lower overlap will support the adjacent one. T_G supports on both sides. While better than plain boards, T+G would be superior in terms of load distribution.

That said, I'm right with you, I detest working with T+G, it's a right pain to get right again.

With a half lap joint (or "R-flap" if you're a cockney), you'd have to fix the upper half somehow, whereas T&G needs no visible fixings.

The Natural Philosopher ( snipped-for-privacy@invalid.invalid) wibbled on Tuesday 11 January

2011 15:08:

Yes - but that's not the point.

Everytime I fancy pulling a new comms cable (be it aerial, sattelite, heating control or Cat-6 I want it to be easy - And I could see that happening once every 1-3 years.

Not to mention access to motorised valves and the odd cable inspection (eg mice).

That's why it's an absolute design criteria that the floor must be liftable.

I was thinking that shiplap on sheds seems to manage quite well. Agreed, T&G is better in that respect, but a half-lap must be better than square edge, even if not perfect?

The Natural Philosopher ( snipped-for-privacy@invalid.invalid) wibbled on Tuesday 11 January

2011 15:08:

I suppose that could be a risk. I was only thinking of a 5mm overlap to allow for normal dimensional cycling. And two screws per rafter per board.

And "The wife wants a light over there and the wiring's over here" situation. And moving radiators. And so on.

Shiplap on sheds is bowed and warped and lets in draughts and is never level

Agreed, T&G

cant see it makes any difference really. Boards can still lift with respect to each other.

Oversize skirting with a duct behind :-)

Or a false ceiling...

You've been watching Gilliam's "Brazil" too much.

You employ mice to inspect your cables? Marvellous idea.

Cheers Richard

The Natural Philosopher ( snipped-for-privacy@invalid.invalid) wibbled on Tuesday 11 January

2011 16:33:

With due respect, you're rather missing the point.

The point being: "floor must be liftable". That much is not debateable.

Everything else is moot, except as a general idea for others.

On the skirting cavity: good idea, but can run foul of the "zones" rules if containing mains cables.

False ceiling - not sure SWMBO wants the kitchen to look like an office and we have only 2.4m anyway.

I wouldn't mind ceiling cornice trunking, but the trouble is it tends to not be big enough and it looks like plastic from the odd ones I've seen on the web.

The Natural Philosopher ( snipped-for-privacy@invalid.invalid) wibbled on Tuesday 11 January

2011 16:26:

OK - explain this (because I can't see it):

****** A *** B T&G slot side ****** C ****** E *** F Half lap

I can seeing the issue regarding E cupping upwards (if grain wants). But surely, A could just as easily cup by a similar amount?

A B C

******ooooooooooo***** ***ooooooooooo********

3 boards, A, B and C. If you step on A, B takes some load. If you step on B, C (but not A) takes some load. T&G of course would load transfer both ways, PAR no ways. So a half lap joint must be stiffer than a square edged board but not as stiff as T&G boards.

What am I not seeing?

Cheers

Tim

Why two screws? Would it not be enough to put a screw through each board near the edge at which it overlaps on top of its neighbour, but not near the other edge where it overlaps underneath its neighbour?

In the diagram below, I'd say board 2 needs a screw at S (which has the side-effect of fixing the right-hand edge of board 1, while the right hand edge of board 2 is held down by board 3 and its equivalent screw.

Your scheme does really need a screw at each board/rafter intersection, though, while T&G can probably get away with a quarter as many.

---------+ +----------S----------------------T----+ +--------------- | | S T | | board 3 | +-----+ S T | +-----+ +-----+ | S board 2 T +-----+ | board 1 | | S T | |

---------------+ +----S----------------------T----------+ +---------

----------------------S----------------------T------------------------- S T S T rafter S T

-----------------------------------------------------------------------

Criterion. Criteria is the plural.

But yes, I've not understood why planks are not screwed down with nice brass screws. In this house, in the old part, the floor doesn't move as I walk on it. In the new part, it does. Is this down to cheese-paring with the joists on builders' part these days.

Even worse in our old house. Not only did the floor move, in some of the rooms where it was chipboard it squeaked too. I think this was because the joists were not deep enough (not enough headroom to make them deeper). How does one stop this? Would putting polythene sheets over the joists and then screwing the chipboard down through that make a difference?

Aargh. Having put screw T into the diagram, I then didn't refer to it in the text to say it isn't needed. Corrected version:

In the diagram below, I'd say board 2 needs a screw at S (which has the side-effect of fixing the right-hand edge of board 1), while the right hand edge of board 2 is held down by board 3 and its equivalent screw, and therefore there is no need for a screw at T.

Ronald Raygun ( snipped-for-privacy@localhost.localdomain) wibbled on Tuesday 11 January

2011 17:27:

This is a very good point.

Two screws only because that is the norm with PAR.

We (by that I mean I) probably need fixing at every joist due to slight unevenesses of joist levels - will attack with a plane but I don't expect perfection to be acheivable.

But one screwing down one side makes sense if the joint is accurately routed so the joint maintains downpressure.

It would be zero risk to try - extra screws are trivially added later if needs be.

I see. Par for the course, eh?

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required