Non-plywood subfloor options?

What subfloor options are there for solid wood or other natural materials?

I am planning an addition to my house. Somewhere in the range of 12x30ft. Everything I have ever done either used plywood or already had very old 150+ year old boards on it. Most of the 1x lumber I see these days doesn't seem strong enough for a subfloor. I supposed I could just line it with 2x4 or whatever dimension was cheapest. Are there options specifically for this?

What options have you heard about? What options are available near you?

Hey Mat.
So I’m guessing you have beams as a base?
The necessary thickness would depend on the distance of the beams. Also some insulation would be beneficial depending on your climate and the construction.

Here (dry, warm, southern Spain) we often build on the earth. In that case we like to use gravel, then a slab of limecrete with fiber, then whatever finished floor. Tiles or wooden floor.

Yes, my climate is way too wet and cold to build directly on the ground like that. It will be up off the ground, so I completely agree with adding insulation. The floor support will most likely be 2x10 floor joists spaced 16" on center.

Timber frame construction, which generally has wider spacing between supports like rafters/joists, typically uses 2x6 tongue and groove for decking on roofs, sometimes floors (especially lofts). Pricier than ply, but maybe worth it for certain advantages.

Will you be adding the insulation afterwards from underneath? Or will it become part of the floor?

I could imagine spanning the 16” with some type of rougher board (not planed, we get stuff here meant for cement casing) that might be cheaper to get. And then add your finished flooring on top.

maybe find local sawmill rough cut white oak. I guess I'm lucky to have a nearby sawmill that sells bundles of cull.  the boards range between 8 and 16 foot long from 4 to 12 inch wide . bundles that are 5 foot wide and 4 to 5 foot tall for less than $200. you can get either oak or poplar bundles

bruce , much appreciated, you invariably add knowledge to every conversation.
Bought rough sawn white oak , about 1.25 inch thick,  for decking years ago.  Amazing.   Local sawmill also, I had to go 4 miles into a thick forest. So sturdy, and with extra nailing it could be better than plywood?
Thanks for the tip bruce,  much obliged.  Hmmmm,  will be filling the tank in my truck, should be in your driveway by 9 am. tomorrow, ready to  visit your friend's sawmill............
Rico

here is the sawmill with pictures of stacks of culls too. $200 price may have gone up since 2022 but cant imagine all that much considering how much hardwood you get.
https://www.facebook.com/p/Clinch-River-Hardwoods-100046536247734/

My 1950's house has floor joists on 16" centers, 1x6 t&g subfloor boards running 45 degrees across the joists, and the flooring was on top of that.  Kinda squeaky but it's solid.

That was probably old growth wood though.  If I were replicating it today, I'd look for 1" rough cut boards (pine or similar), plane them down a little bit if they're too variable in thickness, maybe t&g them, and install them perpendicular to the joists.  

Maybe do a mock-up of some joists and the boards you find to see how beefy they are.  14" centers on the joists would really stiffen it up, if needed.

Mr. Chickadee (Josh Stevens) made a Korean style floor a few years ago, documented in 2 parts:




Worth a watch, in my estimation.  Very traditional, no plywood in sight.  But, it requires a timber framed floor structure.  My current (slowly being completed) project used very non-permie 3/4" CDX over short span 2X6 joists.  One of those "do as I say, not as I do" things, I guess.  "Not as though I had already attained... but I press on toward the mark..." (pace St. Paul).

One of my timber framing books discussed using rough sawn 2" thick oak planks, grooved on both edges, with a spline fitted between planks in the grooves.  It may be in the Fine Homebuilding book, in the chapter on the Rhode Island Stone Ender (ISBN-13:  978-1561580477), but I am not able to double check at the moment.  This would be over fairly widely spaced joists in modern timber framed construction, though older timber framed structures would likely have more closely spaced joists than the spacing for floor joists for modern stick built construction.  The top surface of the plank floor can then be adzed/planed smooth after installation.  An ox (two operator) plane, or at least a decent scrub plane, would be helpful here, I'd think.  The splines will accommodate some movement due to seasonal changes in moisture content, will prevent sight lines/light leaks between floors, and keeps floor sweepings above from dribbling down into the storey below.  One advantage of the splined method over tongue and groove is that you don't lose width of your heavy planks to the rabbet cuts forming the tongues.

In Hermann Phleps' book "The Craft of Log Building" (ISBN-13: 978-0969101918), he discusses using heavy planks for floors which protrude through the wall to the exterior.  At least one pair of planks has matching fine tapers, cut on one of the edges of each, as means of tightening the floor.  If I recall correctly, this was in an alpine European context.  The folding or fox wedge planks could be drive in from the exterior of the building to force the rest of the floor planks outward to tighten up any gaps.  But, there aren't quite enough details in Phleps for me to completely understand the system, and I am a bit leery of exposed end grain of the floor planks as regards rot.

On edit, here's a puncheon floor (i.e. smoothed on one side, left in the round on the under side):

You can see a bit of adze smoothing of floor planks.

And, one more for reference, a traditional floor in a Finnish log cabin:

The floor build process is interrupted by a window installation.  These planks are edge doweled, rather than tongued and grooved.

our ground floor level is polished sealed earthen floor, but it sounds like that isn't an option.  On our upstairs we did local rough cut 1x for our sub-floor and on top of that we did 3/4" milled tongue and groove softwood.  We think this to be ideal so the kids and us can beat it up and then later we could put hardwood on top of that if we ever cared to have a nicer flooring.

I used  1x6 tongue and groove boards for the subfloor of a loft a couple of years ago. My 4x8 joists were on 24 inch centers so I'm sure 2x10s and 16" centers would be plenty sturdy.

Once I saw a subfloor that was done entirely with 2x4s, not situated flat on the 4" sides, but all situated resting on the 2" sides, and pressed together. Extremely strong and eliminates need for close joists underneath: you could span them prob at least 3 feet, I imagine, but don't quote me on that. Probably adds a fair amount of insulation too, though it would be heavy and expensive to build.

Lots of great ideas above. Just wanted to share another option, perhaps for those who do find themselves using plywood. AFM SafeCoat can be used to seal a variety of surfaces to stop off-gas. https://afmsafecoat.com I've used their product over the past 20 yrs to make indoor environments safer, and family members with severe environmental sensitivity have noted how well they work. The company has expanded their offerings and their AFM QuickSpec Worksheet will point you in the direction for what you're seeking.

I would go with 1x4's (hardwood would be better than pine, but both have been used historically) and lay them 45 degrees to the joists. My house has all rough cut 1x4's so they are truly 1" thick. Some of them have been there since 1890.

I watch a reality show based in Hawaii for renovating houses.  

A lot of homes in Hawaii were built without a subfloor.

The floorboards were nailed directly onto the floor joists. The finished interior floor you walk on and the ceiling of the crawlspace underneath are the same.

Matt McSpadden wrote:What subfloor options are there for solid wood or other natural materials?

I am planning an addition to my house. Somewhere in the range of 12x30ft. Everything I have ever done either used plywood or already had very old 150+ year old boards on it. Most of the 1x lumber I see these days doesn't seem strong enough for a subfloor. I supposed I could just line it with 2x4 or whatever dimension was cheapest. Are there options specifically for this?

What options have you heard about? What options are available near you?

I don't understand the concern about plywood, but 1" (nominal) boards is strong enough for a residential subfloor. If the species available to you seems too weak for joists on 16" centers, just reduce the spacing. Expect a squeaky floor. Tongue and groove subfloor plywood is not only much faster to install, but it's also quieter, especially if you include construction adhesive or cushioning on the joist tops. Perhaps cushioning could be used with a board subfloor, something like the flashing tape I used for "100 year joists" on my deck rebuild.

1x 4,5,6 or whatever size is cheap. Diagonal is a must. If you can acquire white oak it would go a long way toward making it a forever structure. White oak is second only to cypress for rot proof.

Mike Haasl wrote:My 1950's house has floor joists on 16" centers, 1x6 t&g subfloor boards running 45 degrees across the joists, and the flooring was on top of that.  Kinda squeaky but it's solid.

That was probably old growth wood though.  If I were replicating it today, I'd look for 1" rough cut boards (pine or similar), plane them down a little bit if they're too variable in thickness, maybe t&g them, and install them perpendicular to the joists.  

Maybe do a mock-up of some joists and the boards you find to see how beefy they are.  14" centers on the joists would really stiffen it up, if needed.

Is your current sub-floor solid wood?

What would the advantage be for perpendicular orientation? The cuts for diagonal don't seem like they would be difficult but I'd think diagonal would provide better support to the boards.

I have never heard of diagonal boards in the sense explained here.
In Australia we do not have sub floors, what is their purpose

Coydon Wallham wrote:
Is your current sub-floor solid wood?

What would the advantage be for perpendicular orientation? The cuts for diagonal don't seem like they would be difficult but I'd think diagonal would provide better support to the boards.

I am not sure the diagonal boards on a subfloor would provide the same sort of advantage as the wall. On the wall, the diagonal supports help with racking strength. I'm not sure there is the same need on the subfloor. All the ones I have seen are perpendicular to the supports.

John C Daley wrote:I have never heard of diagonal boards in the sense explained here.
In Australia we do not have sub floors, what is their purpose

Here in the US, it is common to put something to span across the supports to make a walkable floor. However this floor is generally rough in one way or another. That floor is generally covered in something else (rug, hardwood, tile, etc) which becomes the floor. And what is underneath is relegated to "subfloor".

Overall based on what is available around here, I think either using the tongue and groove that is used with timber framing or else rough cut 1x lumber is going to be my best bet.

John C Daley wrote:I have never heard of diagonal boards in the sense explained here.
In Australia we do not have sub floors, what is their purpose

John -

It was typical, in US homes with a higher degree of fit and finish, to use a sub-floor as the structure over the joists, with the finished flooring material over that.  The finished flooring could be ceramic tile, laid in a mortar bed, or a more durable hardwood flooring, or (more recently) carpet, vinyl tile or snap together laminate.  Our 1890s vintage house, with additions made as late as the 1920s, has board subfloor with tongue and groove maple flooring over that, in the ground floor.  The upper storey has tongue and groove pine flooring only, as both subfloor and finished floor all in one.  Rougher structures would have had planks or log puncheons or some other sort of sturdy flooring, with the top surface more or less well finished, depending on the time and tools available, and the purposes to which the building would be put.  This is for suspended floors, of course, rather than floors built directly on the soil.

Laying the subfloor boards on the diagonal provides a stiffer floor - at least theoretically.  I would think that installing bridging between joists would actually do more.  For wall sheathing?  Sure, diagonal board sheathing would provide a structure stiffer and more resistant to wind racking than boards nailed on square to the studs, but I am less convinced that it would make a whole pile of difference for flooring.  Nevertheless, my grandfather would only lay board sheathing and subfloor on the diagonal.  He also believed in let-in bracing for stick framed walls until his dying breath, even if sheathed with plywood.

But, he was old school; his father built wooden boats from trees, and my grandfather wasn't far removed from that, even though he did paying work until the early 1980s.  I still have his claw hammer, and a portable saw vise that I believe was his, which could be clamped to a 1" thick rough sawn board, and would hold a hand saw steady on the job site, so one could give a saw a quick touch up with a triangular file at coffee or lunch break time.

I may be able to find a reference to Townsend's "Carpentry", which was my grandfather's favorite carpentry book (it went through multiple editions).  My brother has my grandfather's copy, new enough to discuss plywood, but I think I may have a PDF scan of one edition or another.

On edit: I have attached a screen grab from the 1949 edition of Gilbert Townsend's "Carpentry", showing diagonally laid board subfloor and diagonal exterior wall sheathing.  I could not see any place withing the text proper where Townsend addressed laying the subfloor on the diagonal, though he does discuss diagonal sheathing as a means of bracing the frame.  I could only spot one illustration in this 1949 edition which shows board subfloor not laid on the diagonal, though illustrations abound showing the subfloor laid on the bias, as seen here.  Curious...

On further edit:  Townsend's "Carpentry" is an interesting window into the transition from heavier timber framing to lighter (and faster) stick built construction.  He still details heavy posts for corners, with girts gained into them by mortise and tenon; shows tusk tenons for housing the ends of main beams on which dimensioned lumber joists will rest (but also shows built-up beams, made from multiple pieces of dimensioned lumber); details how to properly house beams into a masonry wall, so that, in the event of collapse of the floor, due to overload or fire, the masonry wall remains intact; etc.  Here's a link to the 1949 edition on Open Library (there was also a 1936, and a 1953 edition, at least, and maybe others):
https://openlibrary.org/books/OL16333977M/Carpentry

Thanks Matt, I wonder how your flooring system was created in the first place?
I found this which helps a bit.

I just checked an unfinished part of my basement and learned a couple things.  The sub floor is actually 1x8 t&g and the joints are cut at an angle and meet on the floor joists (as in the pictures above).

I believe the main reasons are for racking prevention on floors above grade, and to keep the t&g joint from accidentally lining up with a nailed hardwood floor joint above.  If the nails all went into a t&g joint below, it would probably squeak like crazy.

 A diagonal subfloor will help stiffen the floor just as plywood has differentiated layers. If both the subfloor and the finish floor were oriented identically then there would be more chance of buckling and less weight bearing ability. The different direction of the grain adds stability.
 In a wall the diagonal will solve racking problems if each half of the wall is oriented 90 degrees from the other half. If all of the boards are oriented the same across the complete wall it will greatly aggravate the problem.

Ken Barston wrote:  A diagonal subfloor will help stiffen the floor just as plywood has differentiated layers. If both the subfloor and the finish floor were oriented identically then there would be more chance of buckling and less weight bearing ability. The different direction of the grain adds stability.

I had not thought about the different directions helping to stiffen up the floor. I learned something new today.

Ken Barston wrote:In a wall the diagonal will solve racking problems if each half of the wall is oriented 90 degrees from the other half. If all of the boards are oriented the same across the complete wall it will greatly aggravate the problem.

Oh, for sure, and it's good to keep that point in mind.  Similar to knee (and elbow) braces in timber framing, both diagonals should be present within a wall to stiffen the structure against racking.  The modern scheme is to use plywood (or, heaven forefend, OSB) as the shear diaphragm to prevent racking.

But, I am more skeptical about floors racking, in general.  Not that it couldn't happen (mud flow/lahar/snow slide or dump truck mishap or something), but racking of a floor out of square in the plan view isn't very likely under ordinary circumstances, and forces that would tend to cause racking seem most likely to involve catastrophic occurrences.  However, if the additional floor material needed, and the extra time and effort to lay it, aren't impediments, then I'd say go for it.  I don't think there's much harm (other than the foregoing, and slightly longer spans between the joists), and it might be helpful.

I'd be curious just when laying subfloor on the diagonal came into vogue.  Any older structures with wooden suspended floors that I've seen or seen documented have the floor boards laid parallel (or perpendicular) to a wall - generally perpendicular to the smallest cross section floor beams/joists, though R. J. Brown in "Timber-framed Buildings of England" does show floor boards laid between, and let into, closely spaced floor beams.  I may be able to find an illustration of this approach.  But, I'm not anywhere close to enough of an architectural historian to be able to point to when people began laying diagonal subfloor.  On edit: I've attached a PDF of a page from Brown's book, showing historic methods of suspended flooring on the other side of the water from me.

Kevin Olson wrote: Laying the subfloor boards on the diagonal provides a stiffer floor

From the 1997 edition of Canadian Wood-Frame House Construction (p. 74):
Diagonal subflooring permits finish strip flooring to be laid either parallel to or across the joists.

The same source recommends 19 mm (nominal 1") thick boards, although 17 mm (11/16") can be used with 16" o.c. joists. Another book, Modern Carpentry, from 1969, says subfloor boards should not be more than 8" wide, which is understandable when you consider cross-grain stability.

I used 1x6" (nominal) fir for a cabin floor in 1971, but after that it was always 5/8" T&G plywood for my buildings.

David Wieland wrote:

Kevin Olson wrote: Laying the subfloor boards on the diagonal provides a stiffer floor

From the 1997 edition of Canadian Wood-Frame House Construction (p. 74):
Diagonal subflooring permits finish strip flooring to be laid either parallel to or across the joists.

Well, that makes good sense.  I wonder if that's the original reason for so doing, or if that's a convenient side effect?

Kevin Olson wrote:

David Wieland wrote:

Kevin Olson wrote: Laying the subfloor boards on the diagonal provides a stiffer floor

From the 1997 edition of Canadian Wood-Frame House Construction (p. 74):
Diagonal subflooring permits finish strip flooring to be laid either parallel to or across the joists.

Well, that makes good sense.  I wonder if that's the original reason for so doing, or if that's a convenient side effect?

I can't think of any other reason, because a subfloor doesn't need the bracing that diagonal cladding can give a wall.

Benjamin Dinkel wrote:Hey Mat.
Also some insulation would be beneficial depending on your climate and the construction.

I would suggest looking into encapsulating the crawlspace so that your insulation will be in the walls from the ground up verses insulating the floor joist bays. I don't think the overall benefits can be beat by doing this from my experience.

Kevin Olson wrote:In Hermann Phleps' book "The Craft of Log Building" (ISBN-13: 978-0969101918), he discusses using heavy planks for floors which protrude through the wall to the exterior.  At least one pair of planks has matching fine tapers, cut on one of the edges of each, as means of tightening the floor.  If I recall correctly, this was in an alpine European context.  The folding or fox wedge planks could be drive in from the exterior of the building to force the rest of the floor planks outward to tighten up any gaps.  But, there aren't quite enough details in Phleps for me to completely understand the system, and I am a bit leery of exposed end grain of the floor planks as regards rot.

On edit, here's a puncheon floor (i.e. smoothed on one side, left in the round on the under side):
...
You can see a bit of adze smoothing of floor planks.

And, one more for reference, a traditional floor in a Finnish log cabin:
...
The floor build process is interrupted by a window installation.  These planks are edge doweled, rather than tongued and grooved.

I was looking at "Byygnadsteknik och Timmermanskonst" (translated to English in the front matter summary as "Building Techniques and the Art of the Carpenter"), Peter Sjomar's 1988 doctoral dissertation at Chalmers University of Technology in Goteborg, Sweden.  The scan I have was made and posted by the University's library.  The subtitle of this work is translated as "a survey with examples from medieval timbered churches and vernacular buildings constructed on the corner-joint principle", so broadly treating historic log buildings, but with some timber framing relevant to roof structures included , as well.

Though I was looking at other things (corner notch variants, and hewing procedures inferred from tool marks), I did spot the attached plate relevant to this thread's subject, showing a method of wedging floors, whether planks or puncheons, without having exposed ends (though one version does have the plank/puncheon ends exposed to weather).  Closing the gap with a covering board like this would interrupt the floor surface, so not suitable for all applications.  The method alluded to by Phleps in "The Craft of Log Building", with full length tapered planks rather than the short folding wedges seen here, would permit a plane floor surface without interruption.  Also shown is a method of constructing an opening in a puncheon floor for access via ladder or stair from the storey below.

The right hand plate shows a puncheon floor doweled together, similar to the plank doweling used in the Finnish cabin video, linked above.

Traditional subfloors were made of 1 x 6 pine.

Then the narrow hardwood floor of maple, oak or chestnut was laid over that.

Kitchens were usually linoleum.

Anne Miller wrote:Kitchens were usually linoleum.

A few years ago, we renovated our kitchen.  It had last been done some time in the 1950s, as best I could tell.  It had very minimal kitchen cabinets (nicely made from high quality plywood, by some local craftsman - I kept them to use in my eventual though as-yet theoretical shop), and the floor had numerous layers (3, anyway) of 12" square self-adhesive dollar store vinyl tiles over decrepit sheet linoleum.  We recycled (re-jiggered) used cabinets which had solid quarter sawn oak face frames and doors, eventually installed soapstone counters (gym-sealed MDO form ply for a while, which remained as a substrate for the stone), and replaced the old flooring with ceramic tile.

After reviewing numerous historic analogous, roughly contemporaneous with our house (which was built in stages, with several expansions), we chose to use a ~320X320mm, 13mm thick Spanish tile from Gres Aragon, rated for exterior use on patios and pool decks.  It's not the typical hex tiles often used in service spaces (kitchens, sculleries, pantries and bathrooms) from the late 1800s up to the beginning of the Second World War here in the US, but it is still somewhat traditional looking and is reasonably non-skid while still being easy to wipe up.  It was also on sale from Wayfair at the time; when the included delivery (no additional charge) was factored in, Wayfair was far and away the best price I could find on this tile among the usual suspects (Lowe's, Home Depot, etc.).  This tile does not appear in the most recent online catalog from Gres Aragon, however.  This tile isn't really period correct for the house, but it's at least a modern interpretation with a nod to tradition.

I scraped the post-WWII linoleum and its tar paper and mastic backing off the plywood subfloor as best I could, screwed down 1/4" nominal thickness cement backer board (Hardiebacker), then had a friend who specializes in tile work lay the square Spanish tiles.  I grouted and sealed the joints myself (not as tricky as getting the tiles laid flat-ish!).  We used the backer board rather than a mortar bed (the traditional way of laying tile) due to several consideration, but chief among them: firstly, no one local to me lays tile in a mortar bed anymore, out here in the hinterlands, and I wasn't about to try to learn on this project; secondly, the available vertical space was tightly constrained due to existing steam baseboard (Base-Ray) radiator/convectors and the kitchen's exterior door threshold height.    All of the latter could have been adjusted, with  more or less tear-up and cost, but there was no easy solution to the former.  So, Hardiebacker it was.

Anyway, ceramic tile (usually hex or square) seems to have also been fairly common in service areas, including kitchens, based on my investigations prior to our kitchen reno project, even though our house had sheet linoleum, as you say.

I was puzzling over the rationale for diagonal subflooring. Racking prevention is conceivable for upper floors, but not for the floor directly on the foundation. The ability to lay finish flooring in either direction, as well as joints and ends never lining up with subfloor joints, seems like a strong impetus for the practice.

Glenn Herbert wrote:Racking prevention is conceivable for upper floors, but not for the floor directly on the foundation.

Not directly on a monolithic slab, but for something on piers or even perhaps cinderblock or other mortared walls...?

True, though the buildings that have diagonal subflooring are generally on masonry foundations where racking is not likely to be an issue.