Thread: WIP Chest of Drawers

Wings (for want of a better term) fastened to leg with floating tenons (joint clamped with screws) and shaped.

IMG20220825121556.jpg

Originally Posted by John Samuel

Wings (for want of a better term) fastened to leg with floating tenons (joint clamped with screws) and shaped.

IMG20220825121556.jpg

I reluctantly am going to offer an observation as I don't want to be a buzz kill. The shoulders you've added to the legs are 90 degrees off. Their grain direction needs to be the same as the legs. They will fight with the legs as humidity increases and decrease over the seasons and most likely cracks will form and ultimately the shoulder(s) will come loose. In traditional furniture where a cabriole leg is used the should grain is never in an opposed direction to the leg. At least at such an early stage, if you should choose to, can adjust...

Originally Posted by The Spin Doctor

The shoulders you've added to the legs are 90 degrees off. Their grain direction needs to be the same as the legs. They will fight with the legs as humidity increases and decrease over the seasons and most likely cracks will form and ultimately the shoulder(s) will come loose.

Thanks, Spin Doctor. That's something I thought about when I made my first cabriole legs. What I really should have done was to ask the brains trust for input.


When I made my first carriole legs I looked at the three examples in my home that are made by cabinetmakers, and all are like mine. The grain in the shoulders is at right angles to the legs.


Then I thought about pieces like those below and guessed I'd be making matters worse by aligning the grain of the frame between the legs with the grain of the legs.

IMG20200407172019.jpgIMG20201110142611.jpg

The piece below is 13 years old and the same sin is committed here. So far, we have had no issues. However, the timber is Cedar, which is a low movement timber, and 13 years is not a long time it terms of the life of a piece of furniture.

Leg Detail.jpg

I'd be interested in any other perspectives out there. My thinking until now has been that the distance over which the forces of shrinking and expansion can act is quite small, and this may be helping. However, I'd still like to ask the brains trust and to get some more thoughts on the matter before I cut the shoulders off these legs.

Finally, I wonder if it matters whether these legs are screwed or glued into place. If the grain in the shoulders is aligned with the grain in the legs, shouldn't we avoid gluing them to the cabinet?

IMG20220826115206.jpg

EDIT: Something I should have mentioned but have not is that all joinery is floating tenons.

I think in the case the wood movement can be ignored because the glue-line is so short. I would compare this to the joint between a table leg and the rail or the rail and stile of a frame-and-panel construction.

The loose tenon John did should be enough to hold the 2 pieces together.

Originally Posted by John Samuel

I'd be interested in any other perspectives out there. My thinking until now has been that the distance over which the forces of shrinking and expansion can act is quite small, and this may be helping. However, I'd still like to ask the brains trust and to get some more thoughts on the matter before I cut the shoulders off these legs.

Finally, I wonder if it matters whether these legs are screwed or glued into place. If the grain in the shoulders is aligned with the grain in the legs, shouldn't we avoid gluing them to the cabinet?

The examples of cabriole legs made by the cabinetmakers are all what might be described as "chest on stand" construction

Although this image looks to be all one piece, construction wise, the "chest" has a base that is attached to the stand, either by a shallow rebate or possibly screws.




The "chest on stand" construction is particularly evident in this example.




In each case, the stand is rigid and the chest can expand and contract independently of the stand.


The chest of drawers you are building for your grand daughter, is significantly different.
Rather than a "chest on stand", it would be better described as a "chest with feet".



It is this detail that worries me the most in respect to severity of relative wood movement.
The base of the chest will expand and contract with humidity changes, flexing the joint between the chest and the legs.
To my mind it is this "pigeon-toed" movement as the leg flexes relative to the base of the chest that is the primary cause of concern, rather than the horizontal grain alignment of the "thigh" portion of the leg.

I'm not sure that using screws or glue will make a significant difference to the chest's longevity.


Possible "fixes".
Although the current design has final "royal approval", it might still be possible to modify it to convert the design to a "chest on stand" construction.
This would mean removing the current thighs from the legs and building a frame for the chest to sit on.

Disadvantage is that adding a separate stand will substantially change the visual look of your design. This is probably too much of a change for the customer.


However, I note that, as designed, your chest includes a separate base panel that is attached to the legs.
If this base panel could be veneered manufactured board -- the board could be edged with 5 mm thick solid timber using the same species as the veneer -- relative wood movement between the chest proper and the legs would be a non-issue.
The base panel could be attached to the base of the chest proper using fixed screws at the front and slotted screw holes at the rear.

Lovely work John, cant wait to see it finished

Hmm, it seems to me that whenever we apply "ears" to cabriole legs we face a dilemma. If you are joining the legs to a frame or skirt, the grain of the leg & the apron run at right angles, & it's inevitable that he ears will be "mismatched" to one or the other component. Given that the width of material involved is not great, the stresses caused by humidity changes are not great & the elasticity of the material usually allows it to cope just fine (think if your average mortise & tenon in which grain direction is always "mismatched").

My own approach depends on the situation, to some extent. My strongest impulse if the wood has any prominent grain patterns is to try & maintain continuity of grain pattern so that the ears look like they are part of the leg. If the wood you are using doesn't have prominent grain features, it doesn't matter so much, & in that case, I would align the grain of the ears so that the longest dimension "matches" with either apron or leg. I would say that in most cases. the length of ear attached to leg & that attached to skirt is not that much different that it matters, tbh.

Looking at one of my own pieces with cabriole legs, I can see that the area of "cross-grain" glued to the to skirt is a good deal smaller than the skirt tenon inside the leg: Cabriole leg.jpg

- both parts have remained solid for over 30 years, so if they just hang on for as long again, it'll be somebody else's worry if one or t'other lets go....

So John, I wouldn't get too anxious about what you've done - the fact that your previous attempts are holding together should bring some comfort....

Cheers,
Ian

That

is a nice example of what I would term a cabriole leg fitted to the stand with a chest placed on top.


However, what John has designed is what I would term as cabriole legs attached to a thin flat board.
Look at the thin flat panel sitting immediately below the chest John proposes to build.


it is movement in this thin horizontal panel that concerns me.

I do love this forum.

Many thanks to the brains trust for the input. Over the years several people here have been excellent teachers and advisers. Three of them responded here; Wongo, Ian and IanW. Their contribution to my development as a woodworker is very significant. I am grateful that such capable folk give of their time to help the remainder of us.

Don't we all wish that wood movement was a simple subject with easy answers for each situation?

Ian ... that is not a panel under the cabinet (chest of nine drawers). It is a bit like a web frame ... about 60 mm wide and about 20 mm thick. The legs/frame are screwed to it. To date I have not been concerned about movement because of the short distance (about 60 mm) where movement can occur. However, that thinking is about the legs/frame interacting with the routed frame. The cabinet is another matter. It can move considerably, so the routed web frame is screwed to it through slots to allow the cabinet to expand and shrink independently of the routed web frame.

I introduced that "routed web frame" component to thrust the legs outward, past the line of the cabinet, to a greater extent than seen for the red cedar bathroom cabinet, which always disappointed me.

IMG20200407172019.jpg



Leg Detail.jpg
I am tempted to pull the legs/frame off it and to remake it like the chest of drawers. If I do, I'll do a proper job on the cabriole legs.

Interestingly, it IS a panel that appears between the legs/frane and the top of the camphor laurel occasional table with a sliding top, seen below. This panel is made as Ian recommended. It is an MDF panel lipped with timber.

IMG20201110142611.jpg

Again, thanks to everyone who contributed, especially Spin Doctor, IanW, Ian and Wongo. There will be visitors to this thread who will leave wiser and more capable than when they arrived.

Ian, the pic I posted is the corner of a coffee table: Coff tble 1.jpg

But it does have a feature in common with many stands for chests or chests in that it's a framed construction where the legs are continuous with the corner posts & the rails are tenoned into them. I prefer this way of incorporating cabriole legs on a piece (where possible) rather than adding them to a sub-frame, where there's only room for a stub tenon or dowels. I've seen too many poorly-conceived (commercially produced, I would add), cabriole legs that have parted ways with their main structure. "Little goat" legs are more about appearance than sound mechanics, we have to accept; they flout truly "sound" construction in a number of ways, one being that the shape reduces the load-bearing capacity, not only by thinning the material, but by introducing cross-grain in the curved bottom part of the 'calf' & 'ankle' & very much so if you get carried away with the "sweep" in the lower part. I've seen a couple that broke at or just above the ankle where the grain gets really short. Despite that, I've also seen plenty of old pieces like chairs & chests with perfectly sound legs & "ears" despite many years of not-always gentle use.

Woodworking is beset with compromises. We have "rules" like not glueing pieces "cross-grain", but we flout that particular rule all the time out of necessity, a prime example being mortise & tenon joints as I've mentioned, yet M&T clearly works (most of the time!). It's about scale - over the relatively small area of a furniture-sized M&T joint, the power generated by typical annual moisture cycles usually isn't enough to cause failure; the stresses that do occur can be contained by the elasticity of the wood and/or the glue. If you use hide glue, which has virtually no elasticity or plasticity once dry, you are relying totally on some elasticity in the the wood, whereas PVA-based glues have a degree of plasticity (particularly when warm) that can give & take a little more. You can use that property to advantage in a joint that will experience some movement but little loading, but it's not good where heavy loading is likely (a principle that was driven home to me rather dramatically when I glued up some Windsor chair seats with PVA glue!).

Where you are most likely to strike trouble is if you make wide cross-grain joints. LOML has a small Hoop-pine COD which is an attractively-proportioned little thing, but of rather cheap 'n' cheerful construction - I think of it as the IKEA furniture of the 1920s. The carcase is as basic as it gets, held together with hide glue & nails (lots of nails!) & the thin (~9mm) sides, have drawer-runners ~25mm thick, glued & nailed squarely across the grain of the sides. At some time in its life (quite early on, I suspect), the inevitable happened to both sides: 3.jpg

It's a contest between relative strengths of components & glues and the power developed by humidity changes - a thick, wide hardwood bench top can develop a very significant amount of power, but a wide piece of thin Hoop can develop enough to self-destruct, obviously.

Of course it's best to try to observe the 'rules' around gross-grain construction where we can, but where I can't I just try to minimise the area involved & cross my fingers......

Ian

I started woodworking with a lot of enthusiasm but not much knowledge.

When I was making the cabinet in the pic below Wongo and IanW alerted me to the fact that I was breaking all the wood movement rules. The sides were Surian Cedar boards and the nine web frames were tenoned and glued directly into the sides. There is no allowance for wood movement.

Cabinet 3.jpg

I made the same mistake on a bunch of pieces. Amazingly, all have survived so far, and some of them are 20 years old. I think the use of low movement and/or low power wood has saved my bacon ... so far ... but watch this space.

I had an old benchtop made of laminated Sydney blue gum. It weighs a ton. Many years ago, before Wongo and IanW attended to my education about wood movement, I made a long coffee table with this laminated timber which was screwed directly to the frame. It moved, of course, and it broke the frame. When a dense eucalypt decides it's going to move, no power on earth will stop it.

Made up the lipped panels for the sides of the cabinet and sanded the veneers that were cut on the bandsaw yesterday. One panel is shown in the pic along with the NG Rosewood veneers that will soon be glued into place.

Sides.jpg

I always use 20-25 mm lips as a minimum, even when the edges are not going to be routed. I used 10 mm wide strips of timber (as is often recommended) in my first attempt at lipping and was cured for life. There was far too much fuss clamping such a thin piece (it wanted to bend) and I wanted more "meat" to hold screws in hinges, tenons, dowells and etc. Lips at 20-25 mm are easy to clamp into position and they give plenty of strength.

I aim for 2 mm finished thickness in the veneers. This thickness, along with the lipped edges, makes the panels tough and very repairable, especially when compared with the 0.6 mm veneers used commercially. I made an occasional table a few years ago. It had a veneered top. The day I put it in the TV room my youngest grandson picked up a stone drinks coaster and flogged the top of the cabinet with it, leaving significant depressions. It was back in the workshop the same day. I sanded the dings out (between 0.5 and 1 mm) and relacquered it. It was as good as new.

46 years ago I bought a veneered table. My kids dinged it at short order. I did repair it, but the dings went all the way through the veneer and were always visible after the repairs. Shop cut 2 mm veneers over lipped panels is a time consuming approach, but we get a strong, easily repaired cabinet (perhaps with book matched veneers) that is not subject to the problems of wood movement. It is cheaper in materials than solid wood, but requires more labour.

When I build veneered cabinets like the one under construction I use web frames. The end of each web frame will have about five floating tenons holding it to the sides of the cabinet. We should not do this with solid wood cabinets for wood movement reasons. This is a bit of over-engineering, but it creates a very strong, rigid cabinet.

Originally Posted by John Samuel

I started woodworking with a lot of enthusiasm but not much knowledge.....

That's probably the right way to start anything, as long as you're open to learning on the job....

I doubt there aren't too many of us amateurs who haven't made at least some 'classic' blunders. I will freely admit to making my share (perhaps more than my share?), mostly in my earlier days, but I saved plenty for later, too! I think one of my best was the first set of frame & panel doors I made, where I glued the panel in the frame - reeal good!

Like John's "mistakes" that one never bit me because I only kept the cabinet a few years and gave it away when moving house. By then I had read/learnt enough to know those doors probably had a limited lifespan, so the cabinet went to the new owner with a clear warning that the doors might do something funny down the track....

Cheers,

Originally Posted by John Samuel

Thanks, Spin Doctor. That's something I thought about when I made my first cabriole legs. What I really should have done was to ask the brains trust for input.


SNIP

In those examples the construction is using a rail and there isn't anything else you can do. Where as the cabinet you're going to make is more traditional construction where the bottom rail of the carcase sits on top of the knee blocks, therefore the grain direction of the knee blocks should be inline with the leg. Overall you project won't fall apart, but because it's a hardwood the joints will probably open up enough over the years to crack the finish, but if that doesn't bother then carry on!