Dresser: … versus Humidity

Published Wednesday, May 26, 2021 by Bryan

It has been a little over 14 months since I finished my dresser. I've learned a few things that I'll consider during my next build, like how epoxy doesn't really stick to slate, and how heavy that much 4/4 cherry is. But the most surprising one has been the reaction to humidity.

It's not exactly hard to find free advice on the perils of ignoring seasonal wood movement. We all know that wood expands and contracts as humidity fluctuates. But many of us also get to nearly ignore it. Kiln-dried wood, cut and assembled in a climate-controlled shop, and used in a climate-controlled home, doesn’t actually move much. I built this dresser in San Jose. Even without climate control, the environmental humidity hardly changed 10% throughout the year.

But I haven't stayed in that environment. This past year, I moved to a home that experiences "real winter" (for two weeks in January, the outside temperature remained below zero Fahrenheit), uses force-air heating (without air conditioning), and also overlooks a lake. The air inside was so dry in January we bought moisturizing lotion at Costco. Spring has finally warmed up the place, and brought days and days of rain. I wouldn't compare it to the mugginess that the southeastern US experiences … but in comparison to winter, my dresser might.

Humidity has caused the drawer face (cherry, the darker-colored wood) to bow outward, pulling the front of the drawer box away from supporting the drawer bottom.

The front of every drawer is bowed out. It actually doesn't look terrible with all of the drawers in place. Since they're all made of aligned segments of the same pieces of wood, all of the bows match. I accounted for some expansion, so the drawers aren't jammed. But, the bottom panel of each drawer is no longer supported by the groove in the front, so they sag under the weight of clothing. The next-to-lowest drawer bottom sags low enough to catch on the front of the lowest drawer face.

Why did this happen? I used plain-sawn boards instead of quarter-sawn, mostly. But, I alternated the growth ring direction of each board, as is commonly advised. The theory behind that technique is that if each board swells away from its outer growth rings, alternating which side the outer rings are on will cause each board to cup in the opposite direction. The face still wouldn't be flat, but it would have a few smaller waves instead of one big bow.

With screws removed, the drawer box moves back into place. Alternating growth rings (highlighted in white) didn't correct the arch.

I didn't get waves. I got an arch. One, rainbow-like arch. Why? Probably two things, both to do with the face being screwed to the drawer box. First, each board probably didn't absorb humidity evenly. If they had, each board would have cupped its own way, producing the expected wave. But, the back side of each face, flush against the box of the drawer, likely absorbed much less than the face in the open air. So, the open-air face expanded, and the sealed face didn't. Second, the drawer box is multi-layer finished birch plywood. It didn't expand in the humidity at all. While I did oversize the screw holes to allow for some movement, I doubt it was enough to compensate for this much. So, the box front itself kept the back side of the drawer from expanding. I bet if I removed the drawer fronts from the dresser and let them stand free in the open air for a few days, the expected wave shape would show up.

But I'm not interested in getting a wave shape. I'm interested in continuing to use my dresser through the coming humid summer. There are a few ways to think about correcting the bow. The one I've decided to use focuses on the use problem, instead of the look problem. The real issue I have with the bowed drawer fronts is that they pull the drawer box front with them, which leaves the drawer bottom unsupported.

Stretchers force the drawer box front to stay against the drawer bottom, even when the tension of the face bow is reapplied.

What I've done is to make sure that the front can't be pulled too far from the back, by installing stretchers between them. This is something I considered putting in the long drawers (40 inches side-to-side) anyway, to add some structure and divide up the space. I used a sliding dovetail on either end of the stretcher, to give it a strong grip on the front and back. I also drove a pan-head screw through the front and back into the stretcher, too keep the stretcher from sliding along the groove as I pull clothes from under and around it.

With the stretchers clamping the box front and back on the square bottom, the screws holding the face to the front were able to easily pull the curve back out of the face. Maybe this is kicking the ball down the road, and I'll have to deal with a worse problem as humidity continues to soak into the drawer faces, or when it all leaves again next winter, but it seems like this is working so far.

Categories: Woodworking