A few months back a customer reached out asking if I could build a version of my Low Cherry Tansu as well replicas of the accompanying smaller tabi-dansu. They more or less wanted the same thing as what’s pictured here on my site, including the same hidden mitered dovetails on the low cherry tansu, as well as pegged box joints for the two tabi-dansu. One request which is a departure from the original is the scale. They asked me to make the cabinets nearly 2x larger. After some further discussion and working out pricing, it looks like the project is going to happen.
(Here’s my original set of modular tansu, including the Low Cherry Tansu which I designed and built, along with two antique tabi-dansu sitting on top):
As a woodworker, you’d think that any job that comes your way would be cause for celebration, and deep down it is exciting and I’m thrilled anytime I get a request to build something special. But new orders also come with a reality check that now the pressure is on to make something that lives up to, or better yet, exceeds the customers expectations. That means envisioning and working out any problems that could arrise either with the construction of the piece or down the road once it’s in the customers hands. Not to mention just thinking long term about the life of the piece and it’s functionality.
In the case of these cabinets the fact that they will be nearly 2x larger than the originals brings up a few particular challenges related to wood movement and functionality that I’m currently trying to work out.
One concern I have with increasing the scale of the cabinets is the likelihood that the lower cabinet could sag from the weight of the upper cabinets, especially once they are loaded up with stuff. The overall length of the scaled up lower cabinet will be nearly 2 m long or approx. 78″.
Sagging of the lower cabinet is particualry an issue when it comes to the functionality of the drawers and door. The door in particular is a Japanese style ken-don door, which sits in a set of grooves cut into the top and bottom panels of the main case. Removing the door involves sliding the door up into the deeper upper groove and swinging the door out from the bottom. Here’s a quick video showing how the door opens:
I like this style of door a lot, but you can imagine how any movement/sagging of the cabinet would negatively affect the functionality. So how to design and build the cabinet in a way that prevents sagging while also maintaining the overall lines and proportions of the parts is the big challenge. The back of the cabinet is one area where I can add a ton of strength and resistance to sagging. But another area I’ve been considering is the base.
Probably the simplest way to reduce sagging would be to incorporate a leg or two in the middle of the base, but clearly that would also have a huge affect on the aesthetics of the cabinet and that option was pretty quickly set aside. Another option which I’ve been thinking about is reinforcing the long support rails of the base with a stronger material, and brass is one material that comes to mind. My initial thought was to sandwich a piece of solid brass between a couple pieces of cherry forming a laminate of wood and brass to form the long support rails.
I have to imagine the brass would really stiffen up the support rails, and it would be a nice little special detail for anyone who cared to get down low and admire the ends of the base. I still have to work out how certain aspects of the base’s joinery would work if I go this route, but overall I think this has potential.
Another big challenge with scaling up all three of these cabinets is the door construction. Specifically how to build the doors such that they aren’t too heavy, and so that they stay nice and flat. With the ken-don style construction and the need to fully remove the doors to open up the cabinets, the lighter the door the better if you ask me. The original antique tabi-dansu are made from paulownia (Japanese: kiri 桐) which is extremely lightweight material, almost like balsa wood. Given that the new cabinets will be made from hardwood, plus the increase in overal scale, the doors are definitely going to weigh more, but my goal is to keep the weight as low as possible. That means figuring out a construction method that allows me to keep the door parts are thin as possible.
I see two option for constructing these doors, namely the solid wood approach or a veneered panel approach. I’m definitely partial to the solid wood approach so that’s where I’ve been focusing most the past few days.
Looking at the inside of the Long Low Tansu’s door here is an initial concept for reinforcing a solid wood panel. The orientation of the panel is in keeping with the original antique tabi-dansu and their doors with vertical straight grain, but it’s definitely not the easiest or pehaps smartest way to build a door. My thought is to reinforce the panel with battens that run across the grain, using unglued sliding dovetails to join the battens and panel. The battens could then be joined to the vertical frame members of the door with mortise and tenon joints.
Here’s version 2.0 of that concept, this time reducing the horizontal battens down to just one piece, and incorporating a pair of vertical supports to reinforce the horizontal batten. The horizontal and vertical supports are all much thicker than the door panel and frame parts in an effort to maintain as much stiffness as possible, hence the ends of the supports which taper down to cleanly meet with the surrounding frame. I came upon this idea after reading through Chris Hall’s blog recently, but now that I think of it, this style of construction is also really reminiscent of the inside of an acoustic guitar.
Here’s the frame work with the panel removed.
And a close up.
On my original Low Cherry Tansu the door measures around 708 mm wide x 275 mm tall x 15 mm thick (27 7/8″ W x 10 3/4″ T x 5/8″ D) and the panel itself is close to 630 mm wide (24 3/4″), and made up of a glue up of quartersawn strips of solid cherry. The panel itself is 10 mm (~3/8″) thick and there are no internal supports or cross bracing. Building the door this way was defnintely a bit of an experiment, but so far over the past couple of years it’s been just fine. Japan’s humidity level is fairly stable, but even so the door has stayed flat and the panel only moves a small amount relative to its width. In winter when we are using our heater and the air gets drier the panel shrinks opening up a 2 mm gap on each side of the door, and during the hot humid summer those gaps all but close as the panel expands. All that is to say that so far my experience using a door of this style with a quartersawn cherry panel has been enlightening and it’s nice to have an idea how much the wood will move.
But again, in scaling up the design I think it definitely makes sense to reinforce the panel with some battens, especially since the cabinet will be heading to a much drier climate than Japan.
An alternative to the solid wood door approach that I’m considering is to make a veneered panel. A veneered panel should in theory be far more stable and less likely to cup/warp/shrink excessively. However my experience with veneering is somewhat limited, and here in Japan things like vacuum presses seem to be virtually unused. Vacuum pressing and veneering is pretty common in all sorts of work back in the US but that seems to be not the case here in Japan. I can’t even find a vacuum press kit or supplies here. Instead what most people seem to use are massive clamping presses like this one:
These presses seem to come in a huge variety of sizes and styles, and I could see how they could be useful if you have the space. But I definitely don’t have the space, and I don’t know of anyone who has one of these, so I’ve been searching for alternative options instead.
One option I’m interested in are the vacuum press kits sold by ROAROCKIT. I have no experience with their products but from appearances I feel like a simple vacuum kit with hand pump like theirs could work for a couple simple veneered panels. If anyone has any experience with these I’m all ears. The main benefit with a kit like this is that I could order and have one shipped to Japan relatively easily, compared to a more full blow vacuum pump setup.
So that’s where things stand at the moment. Ultimately the wood choices for these cabinets has yet to be officially decided but once that decision is made, I’ll be sourcing some material and begin work on the smaller cabinets.
More on this project next time.
Thanks for visiting.
Maybe you could veneer with a hammer? It’s somewhat messy and you’ll have to fashion a veneering hammer, but for one off piece seems to be the least troublesome way. Or what about the clothing iron method? Unconventional, but it works
Interesting idea! I hadn’t thought of hammer veneering. I’m all for learning new techniques and this would be an interesting one. My main concern would be the possibility of the thin hide glue soaking through the veneer, but perhaps this wouldn’t be a problem with nice straight quartersawn veneer? I’m planning to use a liquid glass finish on the cabinet and I have a feeling that if the hide glue soaks through it may be noticeable since liquid glass has no film build of any kind. This is definitely an approach worthy of some testing. Thanks for the comment!
Have you looked on the site https://www.joewoodworker.com/, for a vacuum kit?
Regarding veneer presses, you could probably jerry-rig your own out of couple of car jacks and a bunch of 2x4s. I’ve seen people using similar contraptions for building longboards.
Not sure if building one would be worth the shop time in a professional setting.
Yeah this is the route I’ve taken in the past on the few occasions where I decided to veneer some small panels. I wasn’t super happy with my clamping setup last time but I think with a little more care I could come up with a solid clamping rig, and this approach would then be a fine option.
Thanks for the comment.
or if making your own veneers, make them 1/16″ thick and use slightly thickened epoxy and a caul. Make a balanced panel (veneer both faces equal thickness and species veneer)and if acceptable use a plywood core like 1/4″ baltic birch or better, whatever you can find that is flat enough. That creates a balanced panel that is very stable, and wouldn’t require any battens on the back. the epoxy is much more forgiving than other glues and doesn’t require much pressure for a good bond. Press not required, just a flat work bench and a few cauls. preferably slightly cambered to distribute pressure to the middle of the panel. If accessible 3/4″ MDF makes a good caul, with plastic film between to keep from sticking to panel. Very easy once you’ve done it or seen it done. We use balanced panels like these all the time on high end yachts that are subject to extreme humidity variations. If stuck using paper back veneers epoxy will still work even the super thin 1/40″ veneers unless open grained and then they might bleed through. The trick is to use the right amount of epoxy, about like applying latex paint with a roller, even coat the substrate only usually is fine, depending on porosity of substrate and veneer. Do a few tests.
unless this is too modern and not traditional enough. Save your energy for grain and color, craftsmanship and design and make a well engineered stable panel. If the panel is flat once you make it then it will pretty much stay that way unless you leave one side open to a dry air and other side down against a bench or similar. sooner sealed the better.
Thanks for the input and comment. If I go the veneered panel route I’m definitley planning to cut my own veneer from the same stock that I build the rest of the cabinet for the sake of keeping the grain as uniform as possible. And epoxy is a good thought as well.
I’ve used the Roarocket kit for years and it does a great job on veneering flat panels, I’ve even used it for a few bent laminations, the little hand pump can pull a very tight vacuum, it’s amazing.
I’ve since got a vacuum pump, as with the hand pump you do need to check your pressing every now and then, as if you mis-seal the bag you could lose your vacuum. So the pump is like extra insurance things will work out and allows me to not need to check my pressing.
That’s great to hear, thanks for sharing! I do like the simplicity of the Roarocket a lot, especially since I’m likely to only need it every once in a while. Glad to hear that you’ve had a good experience using it.
I know someone who once made a bench for outdoor use from flat-sawn curly hickory, and they used laminated pieces of 1/8″ wood to make the legs and top piece. It was still all solid hickory, so there was no need for any veneering, and yet the cross-grain laminations were rock solid. Since the edges were covered with a frame, you couldn’t even tell it was laminated.
Making a project this big with the constraints you’re dealing with will be tough. I can’t wait to see what ingenious solutions you come up with for this one.
Hi Jon, your blog is such a good issue for every woodworker. I really enjoy your posts about this kind of problem solving and designing process. Our work is full of toughts of this type, and reading someone else solutions and ideas will for shure come good for other future works.
Thank you for the comment, and glad to hear you’re enjoying these posts.
A bit unconventional, but the last comment on this thread mentions using vacuum seal bags for clothing as a press which might be something more available in your area! https://myssmc.groups.io/g/main/topic/vacuum_veneer_press/104058449?p=,,,20,0,0,0::recentpostdate/sticky,,,20,2,0,104058449,previd%3D1706728402066227126,nextid%3D1700082805961148303&previd=1706728402066227126&nextid=1700082805961148303
Thanks for the comment and suggestion. I think that sounds like an interesting idea, and it got me searching around for others talking about using clothing bags. Unfortunately I didn’t find many people talking about using them with much success. I saw several references to people saying they leak even with clothing. I suppose for a one off project it may be an option, but it also seems a touch risky as to whether it will be successful or not.
https://www.woodmagazine.com/woodworking-how-to/veneer-inlay/inexpensive-vacuum-bag-options
Hi Jon, with your knowledge and skill, I say go for solid wood. You can’t handplane veneers : )
Thanks for the comment! I was wondering when someone would chime in with a vote for the solid wood route. I’m pretty partial to using solid wood and I’m definitely strongly considering it as an option still.
With your handplaning skills, it would be a shame not to. I’ve watched your “making koushi-do” video numerous times.
Do you use an oil finish to seal the doors/cabinet? I’ve always worried that an oil finish on the backside of a door would smell rancid over time, but leaving one side unfinished may be problematic too. It would be nice to leave your cabinet with a handplaned finish, but I wonder if it would dry up too much. I know you’re just in the beginning of this project, but would love to hear your thoughts on this later in your build.
Brendan
Thanks Brendan, Right now I’m thinking of using Hassui Ceramic “Liquid Glass” for the finish, but it will ultimately depend on the clients preference. The glass finish does a really good job of preserving a handplaned finish which is one of the reasons I really like it. It’s also alcohol based and any smell tends to dissipate relatively quickly as the alcohol evaporates, so I think it would work well even on the interior of a cabinet like this one. I’ve used it on a couple of tea boxes and it left no noticeable smell when you open the lid.
I have done some hammer veneering. It is fun and messy and works best in a very warm room. I will say that a piece that large is not one I’d choose for a first project. It definitely takes some practice on smaller pieces first. The upside is that if you make a mistake you can steam it off and try again. Generally speaking hide glue should not interfere with subsequent finishes. But that is easy to test.
The hardwood panel itself would be the roughly the same weight as a plywood/veneer panel, no? If so then the extra weight would be in the battens. Could you make the battens of kiri?
Thanks for the input on hammer veneering. I’m probably going to skip it for this project, but it would be an interesting skill to learn someday. My main concern with the hide glue and finish is simply in terms of the final sheen. I’ve used hide glue a fair bit, and I’m aware that it’s highly compatible with most finishes, however if I use a liquid glass finish, which I’m hoping to for this project, theres no surface film build of any kind so I have a feeling any glue left of the surface would show up as a shiny spot relative to the surrounding wood. But that’s all speculation on my part, I’ve yet to do any tests yet.
As far as a hardwood panel vs veneered panel goes, I’m not sure how much different they will be in terms of final weight, though the solid wood route with the addition of battens will definitely add a fair amount of weight I imagine. Especially since the battens will need to be fairly large to be strong enough to adequately support the panel. It’d be great if I could keep the battens light using a wood like kiri, but kiri is overall a pretty weak wood so I think it’d be better to stick to either the same wood as the rest of the door/panel or perhaps something even stiffer.