I am (still) making this frame & panel based book case for my son.
I'm pretty much done with the frame part of it. So far I've done a good job of screwing up all the mortise & tenon joints as well as ruining some 8/4 cherry.
Now it's time to start thinking about the panels and how I might screw up each panel.
There are 4 panels, two sizes. I guess they will be solid panels.
If I made them to fit exactly in the 3/8" grooves of the frames-
the two side panels would measure 11-1/2 wide by 53-3/4 long
the two back panels would measure 18-1/8 wide by 53-3/4 long.
I drove way out to a lumber place this morning South West of Cleveland. Not even sure where I was. Thanks to GPS I found the place - and my way home. I bought some (pretty much all flat sawn) soft maple for the panels. The lumber was in an unheated metal pole barn so I stickered it and will let it acclimate for a while. It's KD'd.
My question... to allow for movement, how wide should these panels be made?
I still need to do the M&T's for the the front upper and lower stiles. Dreading doing it at this point...
If you don't have a moisture meter, you could weigh accurately one of the pieces weekly to determine if the boards are still drying. Somewhere I read a guideline that (completely) air dried lumber still stored in a space not heated and air conditioned should be stored in a heated and air conditioned space for six months to dry the wood to typical interior moisture content. Almost certain I read that in Bruce Hoadley's Understanding Wood book. The rationale was that air drying would not dry wood below 10-12% moisture, but that wood in a heated and air conditioned space would stabilize at about 7-8% moisture content, and that the wood should be at that 7-8% before incorporating in furniture.
I've used a rule of thumb of 1/8" per foot of width. So, for your applications of 11-3/4" (I think that's what you wrote), I'd knock off 1/8" and make the panels 11-5/8". If that's the width of the opening between the stiles, then add tongues to both edges.
I occurs to me that the wood for the panel is wetter than it's final state and thus slightly swollen. Make the panels full size and they are likely to shrink and leave a space. I would make the slots for them at least 3/8" and preferably a 1/2" deep. Prefinish the panels so that when they shrink they don't expose a raw stripe up the edges. Either glue the top and bottom for about 2" at the center and/or pin it with micropins or tiny brads. Attaching the center assures you the shrinkage will be divided and even on both edges.
I feel your pain on the M&T joints. That's why I bought a Domino. Or maybe I just NEEDED one.
If it will live in a dry climate, the piece will stabilize at a lower MC than in a wet climate. Assuming it will be indoors and heated in the winter, and that it will be in an average climate zone, it's not likely to drop much below 7%-8% in winter nor expand to more than 10%-12% in summer. The coefficient of movement of flat-sawn soft maple across the range of 6%-14% is .00165" per inch for every 1% change in MC. So, taking your wider panels as an example, assuming a 5% MC swing per year, you'd expect .00165 x 18 x 5 inches of dimensional change from winter to summer MC, or 0.1485". If you plan for worst case scenarios (e.g., 6% to 14% swing), it works out to 0.2376".
The width you cut the panels to when you make the piece should be based on the current MC of the wood, and if you don't have a moisture meter, now might be a swell time to get one. If you find that the current MC is around 12% or more (likely if the wood was stored under cover in an unheated building in Cleveland area), the panel width isn't likely to ever be any greater than it is right now, and will probably stabilize at a narrower width. So, technically, if you make the panels now, they could almost bottom out in your side grooves without worry of blowing the case apart because of further expansion. They can bottom out in your top and bottom grooves anyway, since lumber doesn't move much at all in the length dimension.
Seasonal shrinkage from this dimension will be greatest in year one and finally average out after a couple years to a smaller annual MC range. (Think of a damped sine wave.) But, even if the MC drops to 6%, it shouldn't amount to a dimensional change of more than 3/16" to 1/4" over 18" (proportionately less for your narrower side panels). Personally, I've most often found these theoretical dimensional changes to be greater than the actual observed changes. So, assuming your grooves are at least 1/4" deep -- personally, I'd go a bit deeper -- the panels won't shrink out of the grooves, although it's always a good idea to pin the top and bottom tongues at the center point of the width, so contraction will occur equally on both side edges.
I get my coefficients from the Wood Handbook, published by the USDA Forest Products Laboratory and updated regularly. It's a treasure trove of useful wood information. You can view it online at https://www.fpl.fs.fed.us/documnts/fplgtr/fpl_gtr190.pdf
I have followed the width of a 25" wide piece of walnut since 2007. Humidity has ranged from 30 to 65%. The width of this piece has varied +/- 1/32" from a mean of 25 7/32". I follow some other test pieces with similar results.
There are good reasons why actual experience in wood changing size over the seasons is considerably less than the "Tables" predict that I have explained in the past.
These are flat panels, not raised?
Keeping the thin wide back panel flat will be the main concern. I would have used birch plywood. Well actually I would have used cherry because I have some 1/4" cherry.
The moisture content of wood depends solely on the humidity. Air dried wood in Death Valley could be as little as 2%. Temperature affects how quickly the wood responds to a humidity change. Temperature changes will result in humidity changes, which in turn will change the moisture content of wood.
I also pre-finish at least the inside edges of the rails and stiles. Not inside the groove, but the edges that face the panel. It's much easier to get the stain on properly (and wiped off evenly) without the panel in the way.
