Joint Wars: Half-Lap vs Bridle and House or not?

[smorgasbord]

My current design for my slab-top desk's base is this:
[attachimg=1]

The two bottom joints could be half-laps or bridle joints.

In a recent video, Jon Peters decided to use half laps, just to keep it simple:=2xMoWQ-7U9_LDSR5

Looks like this:
[attachimg=2]

One of his tricks is to cut the joint so that it's a tad too tight, so when he finish-sands the joint will be good.

I'm thinking a bridle joint is stronger:
1) Adds structural integrity.
2) Eliminates the sanding of one of the pieces having an effect on joint tightness.

Here's an example of a bridle joint:
[attachimg=3]

From:=zlMoI3uOW46Rev3E

These two bridle joints are pretty much all that will stop my desk from racking, as the connection of the top of the legs to the bottom of the top won't be reinforced with a top stretcher (others have incorporated a top stretcher). I could do that, too, but I think I'd rather just use DF700 connectors so the desk can come apart for moving.

Now, these bridle joints are inherently strong. Nakashima used them for his Conoid chairs to attach the feet to the legs (of which there are only two, so there's a lot of racking forces on them). I think he did something similar for attaching the seat to the two legs, which has even more torque on that joint.

Another thing I could do is "house" or "haunce" the bridle joint. Instead of notching the bottom stretcher for the full width of the vertical leg, I'd notch it narrower. That way the leg could be sanded without affecting the joint tightness/appearance.

[attachimg=4]

Here's a cross-section of the bridle joint on the left and the housed bridle joint on the right:
[attachimg=5]

For the regular bridle joint, on the left, I can sand the bottom stretcher (yellow) without affecting the joint tightness or appearance. For the housed bridle joint, on the right, I can sand the vertical legs (brown) without affecting the joint tightness or appearance.

I'm pretty sold on the bridle joint over the half lap, even over the housed/haunched half-lap, but am willing to entertain opinions and theories otherwise. But, if I stick with a bridle joint, does it matter if I house/haunch it or not?

For the housed bridle joint I could sand the vertical leg (dark brown) as much as I wanted without affecting the tightness of the joint, as long as I didn't sand in the joint areas, which is pretty simple to do. For the

 

[jeffinsgf]

The bridle joint has significantly more glue surface, and I would think equal or better mechanical strength. I wouldn't trust myself to pull off the haunching or housing. I would go for full, through bridle joints and focus on execution.

But, I'm a hack.

 

[mino]

My current design for my slab-top desk's base is this:
...

I will avoid recommending a specific joint, far from enough an expert for that. I will instead advise to NOT rely on Youtube or even publications as guidance, use them only as a "what is possible" information source only.

The issue with /any/ publication of a complex topic is that, by necessity, it has to cover all the aspects one-by-one and depends on the reader to put it all together in his mind. This is made worse with woodworking as most publications, more so with Youtube, even those I seen for schools(!), are designed for a "mathematics/physics-unaware" audience so are intentionally dumbed-down even if the authors knew ..

Unless it is in a form of a recipe - i.e. "This is how one (can) make a particular piece." - be wary of applicability of any generalist statements made by pretty much anyone anywhere.

on topic:

Strength is important, but it is more important in which direction a joint is strong. This includes effects of any material removal from the pieces being joined.

In your case, you are completely relying on a single-joint strength. Hence the joint needs to be strong on rotation (i.e. not a simple pull). A completely different situation compared to the picture where he has two joints far apart, hence does not need to concern himself with rotational loading ..

Then, when you think aesthetics, think first about wood movement. I.e. how the wood will contract/expand and how that will affect visible gaps and joint strength. Many (most) of the more complex joints consciously trade outright strength for contract/expand capability, aesthetics, or both.

Lastly, "how do I sand/make it" IMO should not play much for one-off pieces as long as one has the ability/tooling to achieve something. The workflow savings do matter with series production. With a one-off piece? Not so much.

 

[cpw]

Since you have outward force on that leg, maybe you could incorporate a top stretcher, but inset it into the bottom of your work surface so that it is not visible?  Just thinking about it, it seems like a short bit of 1/8" flat bar screwed into the top of the legs could add something that would work well in tension to keep the top of the legs together without relying just on the bridle joint.

 

[smorgasbord]

In your case, you are completely relying on a single-joint strength. Hence the joint needs to be strong on rotation (i.e. not a simple pull). A completely different situation compared to the picture where he has two joints far apart, hence does not need to concern himself with rotational loading ..

I'm confused because I think it's exactly the same situation. I have two bridle joints on the bottom stretcher, spaced apart, for the two legs, as well.

I even have the advantage of the two vertical members not being parallel - a trapazoid should be more stable than a rectangle.

Since you have outward force on that leg, maybe you could incorporate a top stretcher, but inset it into the bottom of your work surface so that it is not visible? 

In order to actually add any strength, the joints to the top stretcher would also have to be bridle joints, which means having it have some decent width (vertically) so that I could use bridle joints up there, too. The top is 48mm thick, so I don't need a stretcher for support purposes - Desk is only 5' wide total.

 

[jeffinsgf]

Stating only my aesthetic opinion, I think your base is upside down. I see myself kicking that stretcher constantly. I would use the central beam stretcher as you've shown, but run it down the center of the top and have the wide leg and narrow leg both splay outward. And in either orientation, I would prototype it in something cheap to make sure the skinny leg end isn't tippy.

 

[smorgasbord]

Stating only my aesthetic opinion, I think your base is upside down. I see myself kicking that stretcher constantly. I would use the central beam stretcher as you've shown, but run it down the center of the top and have the wide leg and narrow leg both splay outward.

Yeah, I get that. I'm copying a Nakashima design. This design is also stronger since the legs are attached at both ends.

And in either orientation, I would prototype it in something cheap to make sure the skinny leg end isn't tippy.

Great minds think alike:
[attachimg=1]

 

[Packard]

I am looking at your design and I have a couple of observations:

(No drafting program, just a pencil, straight edge and graph paper).

Based on the geometry alone, your design will have great strength when pressured in the direction “Vector A”.

“Vector B” will depend entirely on the quality and integrity of your joinery. 

Adding any one of the pieces show as dotted lines will greatly increase the strength of your table in the direction of Vector B.

Alternatively, a think steel cable running diagonally the opposite direction and under tension will do nearly the same and might be less visible. 

With strong enough joinery your design may very well stand up to daily use just as designed, but vector B will always be the weak point and loading weight on the right side of the table will also be the weak point. 

On the other hand, I have seen mid-century modern chairs that look like they would fall apart the first time they were sat upon, and they seem to defy physics.  (This Z-chair in particular.).

I don’t know how comfortable they are, but they apparently don’t fall apart.

 

[jeffinsgf]

I see your 2100 photo-bombing your prototype.

Nice bench. What brand is that?

 

[smorgasbord]

On the other hand, I have seen mid-century modern chairs that look like they would fall apart the first time they were sat upon, and they seem to defy physics.

Here's a Nakashima Conoid chair, with just two legs and bridle joints:
[attachimg=1]

Here's the desk I'm copying:
[attachimg=2]

I do think there's a top "stretcher" that is there for attachment purposes at the wide leg end (doesn't go across the whole desk), but also gives some additional racking resistance.

I see your 2100 photo-bombing your prototype.
Nice bench. What brand is that?

Yeah, my trusty one-owner Inca 2100. I love so many things about it (12" blade, pretty deep top, nice throat insert, vertically rising blade, etc), although I added the Wixey DRO and replaced the original aluminum extrusion with a taller one. I do wish it had a cast iron top, but only to be able to use the many magnet things available now. I believe Inca made a 2200 tablesaw with both standard miter slots and cast iron top, but mine's in such great shape and I doubt I could find a 2200 in as good a shape.

As for the bench, that's a rare one-of-a-kind Smorgasbord brand. Solid maple top, with a piece of bird's eye in the front. Trestle base with wedged through tenons, so I can knock it down when I finally move. Lee Valley's Tucker vise at the front left, self-made tail vise out of Narra at the right, and an embedded moving dog hole block for the second row of dog holes down the middle. I used Ulmia's large bench dogs, but I only have 4 and wish I had bought more because I can't find them anywhere. I'll probably make some more out of wood with my CNC - any suggestions on best species?

 

[jeffinsgf]

Lovely work on the bench. I thought it might be owner-built, but the square dog holes made me think manufactured.

I'd be tempted to use Lignum Vitae for the dogs. Tough enough to hold up. Oily enough to work smoothly. Almost as dense as your steel dogs.

You have any beauty photos of the bench posted anywhere?

 

[Packard]

The Cassina defies-gravity-chair (AKA “Zig-Zag chair”, has joinery that would be difficult to replicate in a home shop.  They show it in this excellent (and brief) video.

 

[BarneyD]

I built my bench and dogs 35 years ago. I used Padauk for the dogs (because I had some scraps leftover). They all have a fair share of battle scars, but are still holding up OK.
View attachment 1

Cheers.

 

[rvieceli]

[member=77266]smorgasbord[/member] that Minguren desk has an upper section/brace on the slab leg that appears to be at least as long as the leg is wide. In addition, the skinny leg has an upper brace the extends to almost the entire width at that end of the top.

I also feel that a lot of the design appeal of Minguren desk series with the wide leg and the skinny leg is that the slab narrows rather significantly over the skinny leg.

The tables that are relatively even width-wise end to end use two wide legs. You might want to do a small mock up model with slab pattern you are going to use to see if it looks right.

Ron

 

[smorgasbord]

[member=77266]smorgasbord[/member] that Minguren desk has an upper section/brace on the slab leg

Yeah, I found this picture that shows the upper brace:
[attachimg=1]

And then this variant with a wider top and a drawer:
[attachimg=4]
[attachimg=5]

For a wider desk, there's this design, too:
[attachimg=2]
[attachimg=3]

Note that my top at 48mm thick is probably more than twice as thick as the Nakashima designs.

 

[Packard]

My issue with those gravity defying table and chair designs is that we only see the final finished product.

I’ve got a feeling that that Zig Zag chair was built initially and fell apart when tested.  Then they added the finger joints and it fell apart.  Then they added that little wedge and it fell apart, and then they added all those dowels and gravity was defied.

We are looking at a table that might have been the final of five or six tries. And we are trying to come up with a construction method that works on the first try.

For all we know, there might be a solid hardened .750” diameter steel rod inserted in each joint to combat joint flex.  Antithetical to the woodworking craftsman aesthetic we love in these forums, but perhaps a viable after-assembly-lord-save-this-project-hail-Mary-pass.

 

[smorgasbord]

Hey, I just wanted to say that I really appreciate everyone's comments and suggestions here. I've gone back to the drawing board on the base, even thinking about whether I really can live with just one shallow drawer, too.

Here's another Nakashima desk with a full set of drawers and still a slab leg:
[attachimg=1]

I don't really need the big hanging folder drawer on the bottom, but having a couple/three shallower drawers might be good.

 

[Packard]

For the hanging folder drawer, build an independent low rolling unit with one small drawer, one hanging drawer, and perhaps a tray at the top for pencils, etc.

Use nice polished wheels and match the style to the desk and it is an “ensemble”.  And quite possibly handier than an attached drawer cabinet.

I like these wheels, but you could also design it so that the wheels are largely hidden from sight.
https://paxtonhardware.com/products/brass-furniture-casters-small

Save your scrap. You can hollow out the tops of pieces to make pencil trays, hold post-its, paper clips, etc. 

For the scrap wood, trays would be handy.  Glue felt to the bottoms for a rich effect, or just get felt sticky discs.

I’ve made cove cuts on the table saw for trays, but I cannot picture doing the same with a track saw.  In any case, a similar effect can be done with the router.

 

[smorgasbord]

You have any beauty photos of the bench posted anywhere?

No, but here are some shots I took this morning:
[attachimg=1]
[attachimg=2]
[attachimg=3]
[attachimg=4]

 

[jeffinsgf]

That my friend, is a thing of beauty. Love everything about it. Love tusk tenons for both their portability and strength. Here's the miter saw station I built for the Woodpeckers shop.

Got so excited when it all fit, that I mounted the saw immediately instead of breaking it down and sanding out the layout lines and breaking the edges. I'll get around to it...someday.