Curious about wood floor expansion in practice

[Kaymo]

Thanks JobAndKnock for your responses. My apologies if you think I am criticising your advice, that's not my intention. You clearly have a lot of experience in this area and your posts are very knowledgable.

At no point have I suggested that wood doesn't expand or contract, or questioned the degree to which is does under different circumstances. In fact I thought I had made that clear by giving the example of my outside door which expands and contracts every year. I'm also not having this discussion in the context of what you or other professionals should do, or what should be done in other houses, I am talking about what I do in my own home, and like I always do, I am trying to understand a little deeper why I am doing things, rather than blindly following standard guidelines that are intended for worst possible scenarios. I may not be a joiner but I am an engineer, so I understand a fair bit about materials and structures in general so your suggestion that I am ignorant of the issues is way off.

When you do a job you have to consider the fact that you don't know whether the customer will have an heat recovery unit that keeps the house at a constant humidity and temp, or if they will paint all the windows shut and keep the doors open when having a shower. I get that. I'm not challenging that. I am trying to think through the issues as they apply specifically to my own home. There are a lot of things you can get away with when working on your own house, or make choices on, that would be risky for a professional. Likewise a manufacturer has to specify in ways that reduce their liability to comeback. I could make a choice to fit wood flooring in a bathroom because I know we don't tend to use the bath in that room and there is always low humidity, you probably couldn't recommend that because you don't know that the customer won't be splashing water all over the room.

In any case, the point of the original post was not really about the degrees to which wood expands or how much of an expansion gap to leave, it was about the fixing at the edges where the expansion is greatest. If the outside board moves to take up the whole of the expansion gap of 15mm then the fixing also needs to move or bend by 15mm. You are right that nails are probably more forgiving in this aspect than screws, but its actually because I was considering using tongue-tite screws on a wood floor I am going to fit sometime in the next few weeks that made me think of this. I've mostly fitted floating floors in the past and with a floating floor there is nothing to resist expansion at the edges. With fixings there is, and the two most obvious scenarios, are that either the fixing holds and the floor buckles anyway, or the fixing breaks. Third scenario is that the fixing flexes by 15mm and maybe with nail will just pull out by that much, but I can't see how screws can move by this much and although I don't really know about the type of adhesive normally used, it seems a lot for it to flex without shearing. This is the nub of the question, can typical fixings really move by this much or is it more of a case that if the floor was to expand by this much the fixings would be sacrificial. If this is the case, maybe screws are a bad idea, as they are less likely to give or pull out.

 

[Kaymo]

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

Sorry for using too many words.

No, I was trying to politely explain that you either hadn't read the full post or didn't understand what I was asking. If the issues was only that "individual sections will expand by a fraction of a millimetre" then you wouldn't need a gap of 15mm would you?

 

[Deleted member 174758]

I still think you are predicating this on the basis that wood will behave like metal or plastic and expand or contract in a uniform manner - an engineers's approach, surely? Unfortunately it doesn't work that way with wood. Individual trees vary, and the manner of cutting (whether the timber is rift sawn or quarter sawn, for example) will also have an effect so the amount an individual piece will swell or shrin as well as cup with changes in the RH of the atmosphere (quarter sawn is a lot more stable, will tend to swell and shrink less and will rarely cup). If you are screwing a floor down what you are doing is constraining the ability of the individual planks to move laterally - so should it swell (as it might well if the house were left unheated over a wet winter) then the individual planks will potentially cup, whereas if you install during a wet winter and then we have a particularly dry summer being screwed rigidly to a sub-floor the chances of your timber splitting during that.summer are very much increased.

 

[foxhole]

Sorry for using too many words.

No, I was trying to politely explain that you either hadn't read the full post or didn't understand what I was asking. If the issues was only that "individual sections will expand by a fraction of a millimetre" then you wouldn't need a gap of 15mm would you?

Yes you would , as the collective effect of every board moving would require 10mm expansion gap . The boards don’t expand in isolation .

 

[Kaymo]

I still think you are predicating this on the basis that wood will behave like metal or plastic and expand or contract in a uniform manner - an engineers's approach, surely? Unfortunately it doesn't work that way with wood. Individual trees vary, and the manner of cutting (whether the timber is rift sawn or quarter sawn, for example) will also have an effect so the amount an individual piece will swell or shrin as well as cup with changes in the RH of the atmosphere (quarter sawn is a lot more stable, will tend to swell and shrink less and will rarely cup). If you are screwing a floor down what you are doing is constraining the ability of the individual planks to move laterally - so should it swell (as it might well if the house were left unheated over a wet winter) then the individual planks will potentially cup, whereas if you install during a wet winter and then we have a particularly dry summer being screwed rigidly to a sub-floor the chances of your timber splitting during that.summer are very much increased.

Not at all. I understand that wood is much less predictable and variable, than metal screws, although the detail of your explanation in relation to wood is excellent. My question here is not really about the wood, its about what happens to the fixings in the event that the wood was to expand to the full extent of the 15mm expansion gap.

I'm not disputing the fact that wood could expand, or contract, by this amount, depending on all of the factors you mention, i'm curious about what would happen to the fixings in this case and from my experience of those, I struggle to see them allowing for 15mm of movement. It sounds like you are confirming this above, at least in the case of screws, that they might constrain the ability of wood planks to move, which is partly what I was curious about.

 

[Kaymo]

Yes you would , as the collective effect of every board moving would require 10mm expansion gap . The boards don’t expand in isolation .

Which is exactly what I have been saying, but you didn't say in your original post. You said that the fixings could cope with fractions of mm, I am talking about the fixings at the extremes, on the edges, which would have to cope with up to 15mm of collective movement.

 

[Deleted member 174758]

My question here is not really about the wood, its about what happens to the fixings in the event that the wood was to expand to the full extent of the 15mm expansion gap.

In swelling or shrinking timber glues, because you tend to use flexible glues, stretch - nails and pins bend and can sometimes pull put or pull through - screws eventually get ripped out or pull through, but they do a lot more damage. This is, however, only in extremis, however with shrinkage screws are more prone to split timber than nails IMHO

I occasionally install solid hardwood flooring, although I mostly just deal with the sub-floors and leave the finished floors to the floorers (the main exception to that being stair floorings, especially if details like cut decorative stringers, shadow gap details, etc are required). I don't like boomerang jobs (jobs where we need to go back time and again), so on concrete floors I always check the floor moisture content with a Tramex meter, I check the MC of materials coming in, and so forth. If I am in the least bit concerned about a (concrete) floor being damp I'll put down an epoxy resin DPM, and so on. Better to be cautious at installation time, I say. This is one of mine - solid T&G oak onto concrete using a combination of elastic adhesives, rubber rings and screws:

 

[Deleted member 174758]

An afterthought: OP, were you thinking about using screws to fix finished flooring onto a sub-floor? I'm not sure I'd want to use a rigid fixing such as screws to hold any.form of finished floor in place unless the sub-floor was really rigid, like the concrete/screed sub-floor we had in the example above (BTW that was installed in 2017 and was still good last summer)

 

[Kaymo]

The subfloor that I was considering the screws for is original timber floorboards, with hardboard on top. I'd imagine its fairly stable as its been there for a long long time, and its upstairs, so no dampness issues.

 

[Kaymo]

I should have clarified that I meant that I was considering using screws to secure new wood or engineered wood flooring to a subfloor of original floorboards. Above looks like the original floorboards would be the finished floor.

So it would be original floorboards as the subfloor, which already has hardboard covering it (underneath current floor covering which is carpet) and new finished wood/eng wood on top. The reason I was attracted to the tongue-tite screws was 1. that nailers are expensive for a one off job, not sure about floating in this context, but mainly because if I need to get access for wiring/plumbing etc, screws are probably the best option for lifting the floor without damage.

 

[Deleted member 174758]

I should have clarified that I meant that I was considering using screws to secure new wood or engineered wood flooring to a subfloor of original floorboards.

I have seen two main methods used to install engineered flooring - nailed and glued (generally floating), although some manufacturers do a click lock system these days, rather like laminate flooring (again a floating floor solution). I might be wrong, but I cannot recall ever seeing a manufacturer recommending screwing engineered flooring down. In fact I have a feeling that if you install with screws and subsequently have issues many manufacturers would use improper installation as their get out from paying for any defect correction

The reason I was attracted to the tongue-tite screws was 1. that nailers are expensive for a one off job, not sure about floating in this context, but mainly because if I need to get access for wiring/plumbing etc, screws are probably the best option for lifting the floor without damage.

Against that:
1. I think Tongue-Tite screws are primarily aimed at small volume T&G sub-flooring projects - they are too expensive and too slow to use (because they are not available collated) for larger scale projects
2. You can hire nailers such as Porta-nailers - no need to purchase and a lot cheaper than buying. Sa.e goes for flooring staplers, etc
3. If you really need to get access you will need to lift your finished floor, cut out a section of hardboard then cut out sections of the original floorboards (which your engineered flooring should be laid at right angles to to avoid telegraphing). This is not like taking the wings off a Land-Rover, it's potentially a lot more work. In fact, if you think it will be necessary to ever dig through your floor to access services you would be better off lifting the original floor now and sorting out any potential problems under there, that and/or install access hatches at appropriate points so that if you have an issue you don't need to lift anything - just open an access hatch. This is actually done on some (a few) commercial projects, but not many. And just to point out one thing - it is generally a lot cheaper and faster to cut a small section of plasterboard out of the ceiling below to carry out such repairs in a hurry, then make good afterwards, than to lift installed flooring, sub-flooring, etc

 

[Kaymo]

I have seen two main methods used to install engineered flooring - nailed and glued (generally floating),

Yes, floating (glued at T and G) is the only method I have done before. Only reason I started to think about nailing, is that we have been struggling to get eng flooring, seems to be stock issues right now, and it looked like real wood was available, although that fell through too! As I need to do the upstairs floor in stages for various reasons, rather than hiring a nailer multiple times, the screws seemed like a good alternative.

I cannot recall ever seeing a manufacturer recommending screwing engineered flooring down. In fact I have a feeling that if you install with screws and subsequently have issues many manufacturers would use improper installation as their get out from paying for any defect correction

Good point. I have certainly seem some suppliers recommend them https://www.factory-direct-flooring.co.uk/tongue-tite-screws-ewa10 and doing a quick look around that site, one of the brands Fyfield included tongue-tite screws as a recommended method, but its probably worth checking with the specific manufacturer before buying, if using screws.

Edit: The above advice to use tongue-tite screws appears to apply to all solid and engineered wood flooring sold by the company above. Whether the actual manufacturers agree is unclear, but the fact that this is one of the fixing options in their installation guidelines PDF the seller would have a hard job refusing to honour the warranty if issues arose. I had a look around and some of the other big online suppliers sell these type of screws for this purpose, and at least one other mention screws as an installation method. For the reasons you have already given, this might not be the best method, but it does appear to have widespread acceptance from the suppliers.

better off lifting the original floor now and sorting out any potential problems under there, that and/or install access hatches at appropriate points

A hatch is not much use in this context as its upstairs, so a hatch would only get you into one joist run, but I wish someone had done this in our downstairs when they installed real wood before we got here, although to be fair, a hatch is a tricky decision as there isn't always an obvious place that won't be visible if furniture/rugs are moved. There aren't any potential problems that I know of at the moment, but it can be sods law that the moment you finish that something could come up. One advantage of carpet is that its easy to lift and you don't need to worry about the aesthetics of the boards underneath when cutting for access.

And just to point out one thing - it is generally a lot cheaper and faster to cut a small section of plasterboard out of the ceiling below to carry out such repairs in a hurry, then make good afterwards, than to lift installed flooring, sub-flooring, etc

Another really good point. I didn't even think of this, but yes plastering up a hole and painting might is probably easier depending on how many boards have to be lifted to get where you need to go. As I have lath and plaster ceilings which are rubble filled, its a bit more messy in my case, but probably still better in a lot of cases.