Chris Schwarz's Blog

Cross About Cross-grain Construction

Today is “Blasphemy Friday.”

I’m preparing the panels to start assembling this Gustav Stickley No. 802 sideboard and wondering if my project is going to self-destruct after a couple years. Here’s the problem: The sides of this piece have the grain running vertically. Yet the grain in the stretchers near the floor runs horizontally.

It sounds like a recipe for disaster: The side panel will shrink with the seasons, and the tenon shoulder facing the floor will open up. Then the panel will expand in the wet months, and the tenon shoulder facing the ceiling will open up.

One way around this problem would be to rotate the side panels so the grain runs horizontally. But that would look funny to me , not like the original piece at all (shown above). Another solution would be to steal a trick from the Hall Brothers, the craftsmen who built the furniture for Charles and Henry Greene. They would build pieces with a two-step mortise , so the shoulder of the tenon would be buried 1/8″ in the leg.

This seems fussy and excessive to me, and I’ve never seen any evidence of this on any piece of the hundreds of Arts & Crafts piece I’ve examined (I used to collect the stuff back when it was all flea-market grade).

So Senior Editor Bob Lang and I put our heads together and figured up how much a 13″-wide black cherry (Prunus serotina) panel would change in width during a 3-percent change in moisture content: about 1/16″ to 7/64″. Then I looked at the bending characteristics of cherry. The legs will flex a bit between the stretcher and the panel.

And then I remembered another project I’d built like this in 2002: A Harvey Ellis-designed magazine stand that’s in my home. It has exactly the same problem (perhaps Ellis didn’t believe in wood movement). The stretcher in this project is as tight as the day it was made. No gaps. And the mahogany I used for that project moves about the same amount as cherry.

So I’m going to build the No. 802 with the cross-grain construction and see if I get spanked.

(By the way, the smaller photo of the sideboard is of the other Stickley No. 802, the one owned by my photographer friend. Note how the curved stretcher and double-tapered legs shown in the  picture at the top change this piece for the better.)

- Christopher Schwarz

17 thoughts on “Cross About Cross-grain Construction

  1. Christopher Schwarz

    Chris,

    I didn’t calculate the MC change based on the relative humidity. I based it off of 10 years of tracking MC in stabilized hardwoods and softwoods in our climate-controlled shop with a moisture meter. Perhaps that’s the wrong way to go about it, but it has proven to be accurate.

  2. Chris Somers

    Hey Chris, Robert,

    Why did you use "just" 3% for a seasonal EMC swing? I have figured (and tracked readings) in my house that during the dry summer months here 15-20% RH sustained is easy; and since our house is not "tight" it’s easy to have a 55% RH indoors during our wet/rainy winters. Based on the Hoadly chart, that’s roughly a 5% EMC range.

    So, just curious as to how you figured your number.

    -Chris

  3. Christopher Schwarz

    Mike,

    This is a cherry sideboard, so fuming it isn’t in the cards. Senior Editor Robert Lang has done a fair amount of fuming here, and I actually prefer some of the alternative processes he’s come up with that look the same.

    They’re safer, more consistent, faster. We covered one of these processes in the last issue of Woodworking Magazine.

  4. Bob Lang

    I think you’re right that part of the discussion is design, and part is engineering. In my mind they need to work together and if there is too much emphasis on one over the other the piece as a whole suffers.

    I spent the bulk of my former career trying to engineer wacky designs (some of them my own and some from architects and other designers) so they could be come useful and decorative objects in the real world. It’s a lot of fun to figure out how to make something that looks impossible at first glance, and some of my favorite projects are the ones I swore couldn’t be built at first glance. It’s also nice to solve an engineering problem in a graceful, elegant way.

    I think one of the things that appeals to me about the work of Stickley/Ellis is that these pieces work both as good designs and a solid functional objects. They really had both of them working at the same time, even when there is some risky engineering or a bold design.

  5. Christopher Schwarz

    Tom,

    In my mind, I’ve never been able to separate the form from the engineering. When both are in harmony, that is the best of all possible worlds. However, more often than not, I am forced into a corner with a piece like this sideboard. Or, more precisely, I force myself into a corner.

    I think that sometimes we value the engineering over the form because it is the easiest, emperical path. I know I do that all the time. I stay away from constructions that please my eye but are a wood-movement management nightmare. To wit: anything with large miters.

    Well, that’s the way I see it.

    Thanks for the thought-provoking comment.

  6. Tom McMahon

    As an ex-design teacher, maybe I am being a bit picky, but it seems to me that in many of the conversations like this one we are actually talking about two different things at the same time. In my opinion there is a difference between the visual DESIGN of a piece ,how it looks, and the engeneering of a piece, how it is made. In any one piece there can be an excellent design but poor engineering or excellent engineering and a poor design. The definition I use for design is . The organization of visual space using the elements of design [line, value, color, texture and time] thru the principles of design [repetition, variation, opposition, transition and balance] to achieve an outcome. Am I the only one that sees this distinction or maybe this distinction is only important to an old design teacher.

  7. Bob Lang

    The confusion over which design elements are authentic Stickley and which "don’t look right" is the main reason I wrote my "Shop Drawings for Craftsman Furniture" books. Way too many magazines and authors have presented projects as authentic reproductions when in reality they have been far from that. I believe it does a real disservice to history and the study of design when someone slaps corbels and spindles on some hideous piece of furniture and presents it to the world as a representation of the style. One of the things Chris and I try to do with the projects we publish is to be clear about whether it is a reproduction or an adaptation. When it’s a reproduction we give some of the story behind the design as well as the "how to do it".

    What Chris is taking on here is a reproduction, and the challenge is to make the joinery work without screwing up the design. Grain direction is an important design element and changing that would, in my opinion, just be wrong. This is one of several sideboards and servers that were either designed directly by Harvey Ellis, or heavily influenced by him. Interestingly, some of the other pieces have the grain in the side panel running horizontally, and some don’t have the lower shelf.

    Had Harvey Ellis not died young due to the effects of alcoholism, he would likely be remembered today along with other great designers and architects of the period. Today we know him for his furniture and inlay designs from his brief tenure with Gus Stickley, but in his obituaries published at the time, this work was not even mentioned. His pen and ink and watercolor work is extraordinary and his sense of proportion and design flawless.

    There isn’t a lot of good published material about Ellis. Most writers either give him too much credit for his influence on Gus Stickley, or gloss over his many talents. The best resources I know of are Ellis’ own work published in Stickley’s "The Craftsman" magazine, and my book "Shop Drawings for Craftsman Inlays and Hardware."

    Bob

  8. Christopher Schwarz

    Running the grain of the stretchers vertically is another option, but I know that it would look odd. I’ve seen a couple antique pieces that did this and it leaped out at me as just wrong.

    Plus, because of the narrow dimensions of the stretcher, this would weaken the side assemblies. A swift kick to the stretcher could crack everything.

    It’s definitely and thought worth airing – thanks.

  9. P. M.

    And I wonder what if you run the grain on the stretchers vertically? Would it be to weird? Am I suggesting to break any design golden rule?
    If I may push the envelope a little more, you could make these stretchers from the side panel even, if you plan carefully enough, the grain lines could match the ones on the top.
    I kindda know that mechanically this is not the way to make a stretcher but, in the other hand the side panel is doing some (more) work than merely decorative.

    P.M.

  10. Christopher Schwarz

    Mattias,

    Perhaps Bob Lang will chime in here with a comment on Ellis, who is one of his heros. The line of case furniture that Ellis designed had the tapers on the outside throughout. Why? I don’t know.

    Chris

  11. Christopher Schwarz

    William,

    Harvey Ellis was designing furniture for Stickley for just a short period. Then things migrated back to square – no tapers no curves. I’ve seen a couple of these pieces in the flesh (the cellulose?). They are fantastic.

    We’ll see how it looks in a week or so.

    Chris

  12. William Claspy

    Now that you describe it, I can see the tapers in the photo. Still don’t think it seems- or looks- quite right though :-) Stickley needs solid, full square legs. Of course, a Schwarz can have whatever it wants, even cross grain construction :-)

    Best wishes from the other end of the state,
    Wm

  13. Christopher Schwarz

    William,

    Perhaps "double tapered" is not a good term here. The legs taper only in the elevation. However, they taper both at the top of the leg and at the bottom.

    So the legs are 2-1/4" wide at their widest point (which is 15-9/16" from the top of the leg). From that point, the outside of the leg tapers to 2" at the top of the leg and tapers to 1-5/8" at the foot. The thickness of the leg (1-3/4") remains constant throughout.

    So there are two tapers, but not what we traditionally call "double-tapered" – my mistake.

    Hope this clears it up.

    Chris

  14. William Claspy

    Chris-

    The curved stretcher looks great (I used the same on the oak A&C table I’m sitting at right now), but I’m scratching my head about the double tapered legs. Doesn’t seem right, and I sure can’t remember seeing them on the original Stickleys.

COMMENT