Curious about wood floor expansion in practice

[Kaymo]

Ok, so I know that wood floors, and to a slightly lesser extent, engineered wood floors need to have an expansion gap. I also understand that this is because wood contracts and expands depending on temp/humidity and this is particularly evident across the grain. The expansion gap is sometimes specified by the manufacturer depending on the area of the floor, but its usually no less than 10mm.

The thing is, if a floor was to actually expand by 10mm at the edge in practice, would it not break or pull out the screws/nails, or shear the adhesive if glued down?

It also seems like a contradiction, that you secure the boards in place so that they don't move and create gaps yet leave a gap for movement.

This is just my curiosity really. Has anyone seen this amount of expansion in practice, and if so, what did this do to whatever was securing the boards at the edges?

 

[John D v2.0]

great question! yeah it buckles up, basically if the fixing isn't 100% on all the boards you won't likely notice, but if it is jammed against a solid edge you would notice straight away as it literally makes a hump in the floor

 

[Koolpc]

Floating wooden floors need this gap. I have seen what happens when no gap is provided! Not good at all..

 

[Kaymo]

Thanks both, but I understand what would happen to the floor if it expanded and had no gap, it would obviously have to move upwards in places if it can't expand to the sides.

Maybe I am not explaining myself well. What I am curious about is what happens to the fixings at on the boards around the edges if it was to really expand by as much as the recommended gap which is usually 10mm or more.

The few floors that I have fitted in the last few years in my own home have been floating, so the floor is free to expand and contract as there is nothing fixing it to the subfloor. I'm going to be fitting another floor soon and I am planning to use tongue-tite screws, and it just made me wonder about how the floor can expand at the edges if it is fixed to a subfloor. I mean, I can see a nail/screw or flexible adhesive allowing for a few mm of movement, but 10mm is a lot and I wouldn't imagine fixings would have that much give.

Is it maybe the case that they don't normally expand by as much as 10mm, and 10mm is just a worst case scenario, or by saying 10mm they can be sure that the fitters will at least leave a few mm of a gap.

 

[Koolpc]

If it cant expand at the edges it will expand at its weakest point...

 

[cdbe]

My question has always been, what happens if you lay it without a gap on the hottest day of the year?

 

[Kaymo]

My question has always been, what happens if you lay it without a gap on the hottest day of the year?

I've had a similar thought, but isn't it humidity more than heat that causes expansion. I have a wooden outhouse door that I fitted in summer and it expands and sticks against the door frame in the winter, so I presume its the water content in the wood. On the other hand, the RH inside is probably higher in the summer and lower in the winter.

Maybe fill the room with steam then fit it hard up against the wall?

I suspect that in most domestic situations there probably isn't much expansion or contraction, or maybe no more than a few mm. Other than kitchens and bathrooms the temperature and humidity don't vary by a huge amount.

 

[Deleted member 174758]

The thing is, if a floor was to actually expand by 10mm at the edge in practice, would it not break or pull out the screws/nails, or shear the adhesive if glued down?

Not really. In my experience the glues specified have a degree of flexibility in them whilst nails or pins aren't that rigid and can low for limited movement. If using screws on solid hardwood flooring it is normal to drill a larger hole than the screw shank (e.g. a 5mm hole for 4mm shank screws) and sit the timber either on rubber grommets or dabs of a flexible glue (e.g. Stik). If you screw softwood planking to joists you always risk the timber splitting or swelling - which is onevreason why joiners nail standard softwood sub- flooring

This is just my curiosity really. Has anyone seen this amount of expansion in practice, and if so, what did this do to whatever was securing the boards at the edges?

A very basic example: why do joiners always install skirting with scribed inside corners? After all, the skirting is installed with glue and nails or screws so surely it can't move? Yet it can, and does, which is why you don't mitre inside corners (because if the skirting shrinks , the mitres will open)

Now a real life, and very expensive, example from my own experience: about 10 years ago I had an idiot site manager on one job who insisted installing an 18mm plywood sub-floor onto what I (and others) regarded as a potentially iffy concrete floor, without any form of DPM, and without any attempt to ascertain how damp the concrete actually was (a simple task with a Tramex floor meter). It was also impressed on him that any plywood needed to have gaps between sheets to allow for expansion, given that this was a very warm dry July, and that in the winter the same site (in the north west of England) was going to be a lot damper. He didn't listen and got his guys to install the ply butted tightly together. Believe it or not, this moron had once been a joiner (just not a very good one). So we got him to sign a waiver that the sub-floor wasn't to our set standard and he'd carry the risk of buckling

We then installed the solid oak flooring (about 3000 square feet on the ground floor alone) using Portanailers, which fix the finished floor to the plywood beneath using barbed nails. We allowed the usual 1 to 2 washer gap between planks every 4th or 5th row of planks to allow for expansion. When installing the oak the washer spacers always have to be removed after about an hour to avoid them getting trapped as the oak expands quite quickly to fill the gaps (it comes in with a MC of about 8 to 10%, so it always expands).

That winter the concrete got damp (the DPM was probably flaky or non-existant and as the premises are at the side of a body of water, the water table is already high). This resulted in the plywood swelling as well as the oak flooring, which subsequently 'popped', bowing upwards off the ply by up to 30mm or more and ripping out all the barbed fixing nails over more than 800 square feet of the oak flooring. The affected part of the floor then had to be lifted and scrapped, together with the underlying plywood, an epoxy DPM installed and the oak replaced. I believe on the insistance of the QS they actually installed over 1500 square feet of new oak flooring, plywood and resin DPM in the end. Total cost to the company was over £30k as the work needed to be done piecemeal on night shifts because the client, obviously, wasn't prepared to close the restaurant for the duration, nor to suffer the noise and dust of the renovation work. I believe the idiot was sacked for this particular screw up...

...because wood is a natural material and invariably moves with changes in atmospheric moisture content

 

[Deleted member 174758]

My question has always been, what happens if you lay it without a gap on the hottest day of the year?

See my earlier response!...

I've had a similar thought, but isn't it humidity more than heat that causes expansion

Yes. The MC (moisture content) of wood varies dependent on the RH (relative humidity) of the surrounding atmosphere, combined with the finish on the timber. This is why you always try to allow materials to acclimatise (i.e. reach equilibrium with the environment in which they are to be installed) before installation. It is also why you need to protect (paint or lacquer) unfinished woodwork such as doors, door casings, etc promptly after installing them (and why I dislike oil and wax finishes in many cases - they are not sufficientlt moisture resistant). People are often unaware that timber expands and contracts across the grain much more than along the grain (if you think about it the polar opposite of what heat does to metals and plastics). This phenomenon explains why a frame and panel door that becomes sticky generally needs the width reducing, not the height, even though the height is more than twice the width

I suspect that in most domestic situations there probably isn't much expansion or contraction, or maybe no more than a few mm. Other than kitchens and bathrooms the temperature and humidity don't vary by a huge amount.

There isn't, really, but it doesn't mean that you can just ignore the possibility of it happening. In new builds, for example, a big issue in terms of sticking doors, etc is when the heating goes on, often just after the decorators have gone through and slapped loads of water based paint all over everything. So moist, then dry.

On the other hand In the sort of commercial stuff I used to do, where we could be laying £30k or more of flooring, reaching equilubrium and/or dealing with expansion issues are much more serious problems

 

[Kaymo]

First off, thanks for this response JobAndKnock, interesting.

Not really. In my experience the glues specified have a degree of flexibility in them whilst nails or pins aren't that rigid and can low for limited movement. If using screws on solid hardwood flooring it is normal to drill a larger hole than the screw shank (e.g. a 5mm hole for 4mm shank screws) and sit the timber either on rubber grommets or dabs of a flexible glue (e.g. Stik). If you screw softwood planking to joists you always risk the timber splitting or swelling - which is onevreason why joiners nail standard softwood sub- flooring

I've no argument at all about expansion and contraction of wood, I see it every year with my wooden outhouse door which expands every winter and sticks when I try to open. However, I don't see the same issue with interior doors, and I don't notice any significant expansion and contraction with the wood floors in the house either.

Even in what you are saying above (5mm hole for 4mm screws), that's more like the level of movement up to a few mm that I would expect flexible glue or other fixings to handle, but 10mm? I'm not saying it doesn't happen, but as a metal screw isn't also going to expand by 10mm, something else would have to give, either pulling the screw partly out or damaging either the wood floor or subfloor.

The example you give of the concrete subfloor is not really a normal situation, and it sounds like you are being generous with the term 'moron' and no amount of gap is going to remedy that disaster. I installed eng wood floor on a concrete subfloor in my house a couple of years ago, and even as a DIY'er, I knew to that I needed a barrier, so I don't know how someone gets to be a site manager without figuring that out!!! Expensive lesson though!

There isn't, really, but it doesn't mean that you can just ignore the possibility of it happening. In new builds, for example, a big issue in terms of sticking doors, etc is when the heating goes on, often just after the decorators have gone through and slapped loads of water based paint all over everything. So moist, then dry.

On the other hand In the sort of commercial stuff I used to do, where we could be laying £30k or more of flooring, reaching equilubrium and/or dealing with expansion issues are much more serious problems

I see what you mean. Manufacturers obviously have to give gap requirements for the worst possible scenario, floor has been baking for weeks in a warehouse in the hottest July on record, arrives on site and is installed in same dry conditions, then gets installed in a new build with paint and plaster still drying out and winter approaching.

It sounds like its safe to say, that someone installing in their own home with boards that are well acclimatised, might be able to skimp a little more on the gap, at their own risk of course.

 

[foxhole]

The gap is to allow for expansion of the entire floor, individual sections will expand by a fraction of a millimetre .The fixing method can cope with that .

 

[Kaymo]

The gap is to allow for expansion of the entire floor, individual sections will expand by a fraction of a millimetre .The fixing method can cope with that .

Those fractions of a millimetre sum to millimetres by the time you get to the outside edge of the floor, which is why the expansion gap at the edge is supposed to be 10-15mm. If the floor was to actually expand by this much in practice, the boards in the middle will only move by a fraction of a mm, but the ones on the edge would need to move by 10-15mm, and that seems unlikely if they are fixed with screws for example.

 

[Deleted member 174758]

However, I don't see the same issue with interior doors, and I don't notice any significant expansion and contraction with the wood floors in the house either.

When you install doors and floors in a house the materials are supposed to be stored inside the building to acclimatise, to reach equilibrium with the surroundings. This potentially takes several weeks, which often simply isn't available in the real world. Hence the need to make allowances for expansion.

Once timber attains equilibrium it is obviously going to be more stable, especially if they are properly sealed (in the case of woodwork, by painting) and in a stable environment where there are only minor changes in RH, such as a centrally heated building.

If you don't see any movement in your house, then I'd hazard a guess that your house is centrally heated, that the woodwork has been there several years, that it is adequately sealed (lacquered or painted) and that the RH of the house rarely varies much.

Other people live in different types of houses and as a joiner you don't know what situation you are installing into - so you need to make allowances. There are also situations where a house may be shut up and left, such as when a house is up for sale, or the owner is working away for several months - and if the house isn't heated, the RH of the air in the house can go up - then floors which have been fitted too tightly, with little or no expansion provision, will swell and bow upwards. I have seen this with laminate floors of houses which have been on the market for a while

As to doors not moving - I have spent a day and a half in the last week adjusting fire doors in one building, all internal doors, in a building which was ostensibly complete in October and.which were allegedly gap checked back then. In this instance I was dealing with shrinkage, which is nonetheless critical in items such as fire doors.

The fact that you haven't experienced wood movement doesn't mean it doesn't happen. Living with potential movement in a living material like wood is part and parcel of my daily job. Dealing with it is a combination of training, experience, manufacturers instructions and judgement

Even in what you are saying above (5mm hole for 4mm screws), that's more like the level of movement up to a few mm that I would expect flexible glue or other fixings to handle, but 10mm? I'm not saying it doesn't happen, but as a metal screw isn't also going to expand by 10mm, something else would have to give, either pulling the screw partly out or damaging either the wood floor or subfloor.

It doesn't matter if you disbelieve it. Wood moves, nails and screws bend, sub-floor timber fibres get crushed. You will also note that few joiners screw finished floors in place - better to nsil because nailed structures withstand movement far better (e.g. timber roof and floor structures which are invariably nailed NOT screwed)

The example you give of the concrete subfloor is not really a normal situation...

You asked for an example. I gave you two, one specific - one general (the skirting scribe instance), including the most extreme example I have ever personally experienced. I could equally have mentioned the bathrooms where the architect has specified laminate flooring, which even though it was installed with extra expansion allowance, it still expanded and buckled. I could also mention the parquet flooring which has bulged and buckled when the building was left unheated for months. And so on and so forth

It sounds like its safe to say, that someone installing in their own home with boards that are well acclimatised, might be able to skimp a little more on the gap, at their own risk of course.

Don't you believe it. People don't always follow installation instructions, not every house has a consistent environment, materials are't always stored or transported properly, stuff often isn't properly acclimatised or sealed, etc. You may think people ignorant of the issues can get away with it, and that my professional opinion is of no value - as someone who has installed more than a few floors over the years I flatly disagree with you

 

[Koolpc]

When you install doors and floors in a house the materials are supposed to be stored inside the building to acclimatise, to reach equilibrium with the surroundings. This potentially takes several weeks, which often simply isn't available in the real world. Hence the need to make allowances for expansion.

Once timber attains equilibrium it is obviously going to be more stable, especially if they are properly sealed (in the case of woodwork, by painting) and in a stable environment where there are only minor changes in RH, such as a centrally heated building.

If you don't see any movement in your house, then I'd hazard a guess that your house is centrally heated, that the woodwork has been there several years, that it is adequately sealed (lacquered or painted) and that the RH of the house rarely varies much.

Other people live in different types of houses and as a joiner you don't know what situation you are installing into - so you need to make allowances. There are also situations where a house may be shut up and left, such as when a house is up for sale, or the owner is working away for several months - and if the house isn't heated, the RH of the air in the house can go up - then floors which have been fitted too tightly, with little or no expansion provision, will swell and bow upwards. I have seen this with laminate floors of houses which have been on the market for a while

As to doors not moving - I have spent a day and a half in the last week adjusting fire doors in one building, all internal doors, in a building which was ostensibly complete in October and.which were allegedly gap checked back then. In this instance I was dealing with shrinkage, which is nonetheless critical in items such as fire doors.

The fact that you haven't experienced wood movement doesn't mean it doesn't happen. Living with potential movement in a living material like wood is part and parcel of my daily job. Dealing with it is a combination of training, experience, manufacturers instructions and judgement


It doesn't matter if you disbelieve it. Wood moves, nails and screws bend, sub-floor timber fibres get crushed. You will also note that few joiners screw finished floors in place - better to nsil because nailed structures withstand movement far better (e.g. timber roof and floor structures which are invariably nailed NOT screwed)


You asked for an example. I gave you two, one specific - one general (the skirting scribe instance), including the most extreme example I have ever personally experienced. I could equally have mentioned the bathrooms where the architect has specified laminate flooring, which even though it was installed with extra expansion allowance, it still expanded and buckled. I could also mention the parquet flooring which has bulged and buckled when the building was left unheated for months. And so on and so forth


Don't you believe it. People don't always follow installation instructions, not every house has a consistent environment, materials are't always stored or transported properly, stuff often isn't properly acclimatised or sealed, etc. You may think people ignorant of the issues can get away with it, and that my professionsl opinion is of no value - as someone who has installed more than a few floors over the years I flatly disagree with you

Great, informative post...

 

[foxhole]

Those fractions of a millimetre sum to millimetres by the time you get to the outside edge of the floor, which is why the expansion gap at the edge is supposed to be 10-15mm. If the floor was to actually expand by this much in practice, the boards in the middle will only move by a fraction of a mm, but the ones on the edge would need to move by 10-15mm, and that seems unlikely if they are fixed with screws for example.

You seems to have repeated my post but just using twice as many words .?