Digital Wood Joints
Nice. But the link is to a clickbait site with dark patterns for the privacy policy. The real article is here.[1]
Somebody should run these through a finite-element package to compute their strength. Some of those look fragile.
[1] http://winterdienst.info/50-digital-wood-joints-by-jochen-gr...
Charles Hayward [1] has a great book on the topic:
Woodwork Joints https://archive.org/details/woodworkjointski0000hayw_k7x4
Hayward discusses not only the joints, but techniques for cutting them with hand tools.
There are plenty of used copies out there. Hayward has many other useful books.
This is phenomenal. I wish I had it to share with a student I worked with last year. Even without this, he came to similar realizations and explored similar territory with significant results.[0] He began by comparing and contrasting traditional Japanese joinery with CNC milled joints. He realized that the critical difference was that Japanese carpentry joints are almost always entirely squared off (due to the flat saw and chisels). Yet the inside corners created with CNC were always rounded. So he let this critical difference be the animating concept of the project.
So if you enjoy this, you might enjoy Luke's work too!
[0]https://issuu.com/lukemurrayarc/docs/portfolio_2023 starting at page 26-27.
Very cool graphic. However emphasizing the juxtaposition between pictures of wood joints on the internet and actual real-life wood joints with the phrase "digital" is a bit perplexing.
Based only on the pics in the “poster”, it looks like a lot of the joints in the second half have visible gaps. I think I would have expected more emphasis on joints that can be cut with the CNC’s round cutter, but which also come together cleanly.
Also, an over abundance of lengthening / scarf type joints, more than the typical woodworker would use, IMO.
I wish I could find the clip now, but with the advent of practical router bits with pantographs, there was a fashion in the late 1800s to replace dovetail joints with dowel and C type joints. They were really hard to make fast by hand, but easy to do with machines.
They could be made to higher tolerances, and had more surface area, and I assume were more reliable to make/hold draws together.
These are very similar to the idea here.
EDIT: thankyou to mikey_p and WillAdams who pointed out the joint I was looking for was the Knapp Joint: https://www.finewoodworking.com/2018/09/26/history-cove-pin-...
I'd like to think we're at a place where CNC machines are cheap and accessible enough that clever tricks or new tools are just around the corner to overcome some of the inherent issues of making basic wood joints on a CNC router.
To me, these look like a stepping stone to something better. They're all intended to be used with a flat-shaped endmill, and all probably have strength and aesthetic issues. One solution is using different router bit shapes in combination with traditional techniques, but this is more labor-intensive.
Interesting idea, using the strengths of the tooling to expand how joinery can look. Some of these are photos; looks like they made actual wood examples of some of these joints.
What is really missing for me here is complete pieces, utilizing some or many of these new joints throughout. Are they practical? Do they look good as part of a larger whole? Even 3D renders would be a good start, if not an actual physical piece.
Most of these are basically impossible to cut without CNC. Once you’ve embraced CNC as your joint method, why wood is the next question.
If you want to learn wood joints, learn mortise and tenon, dovetails, dowels or floating tenons, and nailed/screwed joints. Some extra love for biscuit miter joints but really that is all the main ones people use.
Offenbach also hosts a relatively large Japan festival (in Germany). Wood joinery, at least in my mind, is a very Japanese craft. I wonder if this is by accident.
Off-topic, but that's a giant fuck you of a cookie popup... it's got 3 options of "high/medium/custom data privacy", a duration of how long I would like to keep those settings (just like those nagware popups of "Try OneDrive/Teams/[whatever fucking bullshit] now!" with no option for "Get off my OS!", only "Remind me later"), a customize toggler that makes it even larger to have more settings. Someone programmed this to frustrate the user with "
And the description for the 3 options is contradictory. Above them it says, "Select a Data Access Level", but selecting "High", the description is "Highest level of privacy"... So what does it do, give the advertisers a high level of access, or give me a high level of privacy?
Meh, Matthias Wandel's still going to beat you on speed (and quite possibly strength) with his homemade pantorouter or box joint jig.
As much as I like nice joint design, I think a lot of that aesthetic beauty comes from when it's not machinable (or not traditionally anyway, iirc the pantorouter can make some pretty tight dovetails - because of the orientation vs. trying to route them 'normally'). It's not how fancy and curvey can it look, it's how intricate and fine the detail. Chunky dovetails don't look better than box joints in my opinion.
this is an excellent reference - not only for computer-controlled wood processing, but also for digital fabrication. When trying to create pieces made of plastics, you will sometimes need a strong joint, and wood-working is the perfect place to look
the images are great, but need more info about strengths & weaknesses of each type
I've been working on this sort of thing for a while.
For a Japanese spin on this see Tsugite:
http://ma-la.com/Tsugite_UIST20.pdf
which I worked through at:
https://community.carbide3d.com/t/a-study-of-joinery/28492
Traditional joints (box, dovetails, or obscure variations such as Knapp (cove and pin)) require a vertical fixture and 3 setups (at a minimum) --- cut parts to length and machine internal features, mount four board and cut joints in 2 corners, flip boards (with correct orientation) and cut other two corners.
Rabbet joints are simpler --- so simple that they were covered in a video as "The Simple Box":
https://www.youtube.com/watch?v=V93xDM3lXsM
(ob. discl., I work for Carbide 3D)
There have been a number of programs developed for joinery. A current commercial option is:
http://www.g-forcecnc.com/jointcam.html
(but it requires a vertical fixture)
One commercial option became freely available:
https://fabrikisto.com/tailmaker-software/
and ingeniously has an option where a 30 degree V endmill is used, but to cut boards held at a 15 degree angle, affording a 90 degree cut with a great deal of control and flexibility --- this can multiply setups to 9.
A variation I've been experimenting with is full-blind box joints:
https://community.carbide3d.com/t/full-blind-box-joints-in-c...
They're reasonably easily drawn up, though they do have some rather specific tooling requirements (a narrow 90 degree V endmill, a square tool of that or smaller diameter, and to make things easier, a large V endmill)
One test project was so tight that after putting it together for a dry-fit before gluing I was unable to get it apart:
https://cutrocket.com/p/63781eaf9822f/
I've been working on a programming system to make this sort of thing a bit easier:
https://github.com/WillAdams/gcodepreview
and have some sketched out joints which I've not been able to make using existing CAM tools which I hope I'll be able to do using this system (if anyone could recommend books on conic sections, I'd be grateful --- that's where I got bogged down last time).
think the site suffered a hug, not opening for me ;(
[dead]
These are all designed for CNC cutting, and appear to be intended as one-sided jobs worked with the face of the board down. That is generally the simplest way to hold a board on a CNC.
I would caution anyone thinking of doing joinery this way to consider whether it's actually suitable for the application. I was at a coffee shop once where absolutely every chair in the place was starting to fall apart. The reason was that the joinery was all some variation of half-lap, which doesn't constrain the movement of the pieces in all of the directions that matter. Once the glue failed, the chairs started coming apart.
I would also add that the corner joints meant to replace dovetails or a box/beehive joint are unsightly with all the required dogboning and will not improve aesthetically with the addition of glue. I would further point out that there are already quick ways to cut dovetails or box joints with a router quickly and efficiently. It would be hard to convince me that there's a truly useful role for cutting an uglier version of a box joint on a CNC.
Source: am a furniture maker who does some CNC work.