Yes, it sounds like sapwood vs the heartwood.

Sleeper floor installation has been around forever. Works well

OK, thanks.

My mind is now at ease with the wood I've bought; now, I'm just thinking through this changed-plan for installation, and have a bunch of detailed questions:

1) With 2cm thick, 10cm wide T/G Cumaru solid flooring, to be nailed directly to sleepers [no intermediate plywood], is 40cm (16") sleeper spacing adequate? Or is 30cm (12") more appropriate? In the long term, what difference does it make? [FWIW, sleepers themselves are also made of Cumaru, 3cm x 4cm]

2) What is an appropriate fastener schedule for attaching the sleepers to the slab? I'd thought 24" - but maybe 36" could also work? Also: what kind of fasteners are best here? [Local practice is countersunk wood-screws with plastic masonry anchors - so fairly tedious to install]

3) Between each sleeper: is it a sound suggestion to fill the voids with foam-board? Or am I better off leaving them hollow? [In case of any liquid spills, for example, which approach would leave me better off?]

4) Between the sleepers and the hardwood - the suggestion was to glue each floorboard as I go. Didn't see much written about gluing to sleepers in addition to cleats... is this really a good practice? It presumably would mean I'd need to skip the Aquabar B - is the advantage of gluing (whatever that is) worth the trade-off?

5) Since I'll be installing in the driest moment of the year, I've thought it might be wise to use a thin spacer as I nail... something miniscule, like a doubled-over (or maybe trebled) sheet of copy paper, just to leave a bit of play for when the humidity returns. Is this a crazy idea? Would it be useful? Harmful?

Thanks in advance for any insights you can give me on the above. I may well be overthinking things; but since I only want to install one wood floor in my life, I want to do it right.

Here ya go buddy. Have fun.

http://northcal.com/PDF/hfim.pdf

Sleepers fastened to the slab should be fastened well. The flooring may lift them off the ground with moisture increase. I use treated sleepers that are on rubber cushions, and not fastened. Sleepers no more than 4 feet long, off set so they can "overlap about 6 inches, in rows 12 inches apart. You can still install a foam between them, but is trouble. If the floor does get flooded it would be better to have space between the sleeper rows so it can dry out quicker. The plastic membrane goes over the treated sleepers. There is a 10 mil membrane under the slab?
If the flooring is not acclimated you may figure how much it will expand, on average, during the wet part of the year and have a spacer installed between rows, but it will be more practical to figure on a spacer every four rows and have a little gap at the beginning that will fill in over time, or every two feet, depending on the spacer. I use 3 mm washers depending on how much the flooring will expand according to the engineering table the expansion co-efficient of the flooring species. Washers will be easy to remove and will be re-used.

Stephen, thanks for that PDF resource, and Pete A., thanks for your thoughts.

I've been furiously studying all this, and am a bit more confident than I was before. But still, some outstanding issues:

1) At the very start of this project, before the slab was poured, I allowed myself to go along with local Brazilian practices, instead of insisting on following the North American norms I was more familiar with. Since nobody in our region uses any kind of vapor-barrier beneath their slabs [or had ever heard of anyone doing this], foolishly and regrettably, I acquiesced. Result: slab was poured directly on the earth. To remedy this folly, several coats of waterproofing membrane (paintable polymeric cement) have been applied atop the slab. This said, I don't want the future performance of my floor to rely solely on that barrier, so I'm inclined to roll out 10-mil visqueen underneath the sleepers. [I also have stock of Aquabar B, if for this application, it makes more sense to use that instead].

2) Pete A: I'm intrigued by your use of sleepers that aren't fastened to the slab. Can you tell me more about this? I'd originally intended a floating plywood subfloor, so floating sleepers make some intuitive sense to me. [It would avoid perforating the visqueen, and would save lots of work; but with solid-wood less stable than plywood, maybe its a recipe for disaster...] If I do end up fixing the sleepers in place, since I can't guarantee in the long-run how well the cementous waterproofing will adhere to the slab, I'll need to mechanically fasten them - probably screws/anchors every 24" or so. In any case, all clear on the need for 12" spacing between the sleepers, with 6" overlap rather than simple butt-joints.

3) I totally take your point on the foam-board, and will leave the voids empty.

4) Both Pete's comments and the NFWA Installation manual both suggest it's not crazy to leave an expansion gap between boards when conditions require it. I've been reading a lot about this. Cumaru, it seems, has a dimensional change coefficeint of .00364, so for maximum seasonal flux between 8% and 13% moisture content, plain-sawn 4" planks like mine can be expected to expand at most 5/64" [1,8mm]. Over the maximum width of the floor (18 feet), this would require a total of around 4". After discounting the 5/8" [15mm] expansion gap I'll leave along the walls, this still leaves nearly 3" of expansion space to be provided, or nearly 1,5mm per board. [And looking now at an 80 year old Ipe floor, that's about how much gap is present between each plank]. What to do with this information? If the wood acclimates to 9% MC or so, before setting each cleat, my idea is to use a putty knife [blade thickness = 3/64", 1,2mm] as a spacer, to ensure a uniform gap along the length of each plank. This should be sufficient, I hope, to accommodate whatever expansion may occur during the humid season that starts in October.

5) Also, to further prevent moisture transfer to the wood, I'd drape Aquabar B over the sleepers before installing the Cumuaru, and would abstain from gluing.

6) If I do proceed with leaving a slight, uniform gap between each plank, another question that comes up is how this will alter, if at all, the expected consumption of sealer & varnish? Because it's prohibitively expensive here in Brazil [$400/gallon], I'll very soon be bringing Bona NaturalSeal and Traffic HD from the US in my suitcases. To ensure I'll have enough for the job: in your experience, does gaps in the flooring make much difference?

6) Lastly - there are two areas in the house [downstairs kitchen, and underneath a claw-foot bathtub in our bedroom] where splashes of water from above may from occur. Obviously, we'll dry any spills as soon as they happen, but still - I wonder: in these higher-risk areas, is there anything special I can/should do to the flooring to provide extra protection? Apply varnish to the underside of the planks before installing? Some other product? Nothing at all?


Thank you so so much if you've read this far. I know I'm bombarding you with questions... can't be more grateful for whatever further answers/insights you can offer. [And if you think any of what I'm thinking is totally wrongheaded/foolish - please tell me!]

Cheers,

Glen

Did you see page 10 in the PDF about using washer rows? You do not use washers in every row, 4 to 10 feet depending on the width of the room. Plus you can start in the middle and spline it so the wood expands both directions.

Will definitely be using a spline and installing from the center, precisely for the reason you mentioned.

As for washer rows: I did see them mentioned in the PDF - but that was in a section on gymnasiums, which also called for 2" expansion gaps around all edges, which caused me to disregard that entire portion of the text.

This said, Pete A also mentioned washers every few rows - but I understood that this was more for convenience than anything else, and that spacing between all rows would work just as well (and look more uniform), if an appropriately sized spacer could be used. (Page 7 of the PDF suggested a putty-knife for this purpose, which I thought was a good idea.)

Anyway - wood will be delivered on Saturday [about which I'm super excited]. Will sticker-stack it on the ground-floor (with some visqueen covering the slab below, just in case), away from direct sunlight, where it can acclimate for at least a month, perhaps two, depending on exactly how dry it is when it arrives, and how close that is to equilibrium moisture content for current temperature/humidity.

While it's acclimating, I'll install the sleepers, first upstairs, then downstairs (except where the sticker-stacked wood is in the way!). First major decision is what to put, if anything, between the sleepers and the slab. (I've got plenty of 10mil visqueen and Aquabar B on hand). Anything special I should consider?

(I also remain curious about the floating sleepers - but haven't seen so much written about it, so hesitate to proceed along that line without more info).