Technical Joinery

[wtnhighlander]

Can anyone recommend a good joint design, housed like M&T, that connects the end of one vertical part into the face of a horizontal part, and can withstand a reasonable amount of load suspended on the vertical part?
The idea is for a half-length shelf in a tall sitting bench. The half-shelf will attach to the end panel or legs at one end, with the opposite end of the shelf being suspended below the seat. The 'load' will consist of a bowl or tray to hold a gentleman's pocket contents, so not very heavy. I'm looking for a no-fastener solution, but so far a foxed tenon is all I can think of.

Suggestions?

[Wood Basher]

Sliding dovetail? Stopped (at the front) and possibly tapered. From the front it would appear to be as housed as M&T but not from the rear where the vertical piece would be inserted into the horizontal.

[wtnhighlander]

Sliding DT is a good option for one of the possible style choices, where a molding offers an easy way to cover the entry point. This is probably the way I will go, but another design option has a cleaner, almost MCM style, where interrupting the edge with a DT entry would be less desirable.

[Chestnut]

A double or triple tenon making sure to get as much long grain to long grain glue area would be my bet. If the horizontal piece is thin and the top of it is hidden I'd make the tenons through tenons to get as much glue area as possible. Otherwise a mechanical fastener sounds like a good bet as a dado will only provide end grain glue area.

[Bmac]

I'm having some trouble envisioning what this shelf looks like on the piece you are describing. Approx how long is the shelf, how far "out" does it suspend? Is the shelf grain the same direction as the panel grain, or is that connection a cross grain orientation? I'm assuming it's a short tray that is suspended off a side panel that has a cross grain relationship.

I really like screws with plugs for this but not in your plan. 

Have you thought about under the shelf supports? Two supports that attach to the side panel with a simple long grain long grain glue joint, the shelf could then be fastened any number of ways to the panel and the supports. You could design the supports in a way to match the feel and design of the piece.

 

 

[Askland09]

Is it within the realms of the design to drawbore it and match the dowel grain as best as possible? Or even stop it just before it protrudes out the front?  

[gee-dub]

Wedged tenon in a flared mortise?

[Immortan D]

I've been successfully using dowels for years and they can withstand a good amount of load, for example these shelves I made 3 years ago for my daughter out of pine wood:

Old pic, but they're still as good as new!

But I know everyone here hates dowels ha!

[wtnhighlander]

As a visual aid, the joint just below the mouse pointer in this quick concept sketch is what I'm talking about. 

Wondering how best to achieve that without showing through the top, which would be twice as thick on the real thing.

[Chestnut]

I'd do a dado with stub tenons. Cremona covered this well in his tool cabinet video.... would it be wrong to post a screenshot?

[Askland09]

I definitely pictured something different when I threw out the drawbore idea (which clearly wouldn't work for this ).  The dado w/ stub tenon or a stopped sliding dovetail are significantly more practical. 

[bradpotts]

I like the stopped dovetail. 

[Bmac]

Ok, now I see. That is a tough one, as I would be concerned over time with that joint. Not so much from the weight of the contents on the shelf, but more so just from the weight of gravity on that joint. Any tenon or dado will not have the best surfaces oriented for gluing. No matter what you do there will be end grain involved with each mating surface on the glue joint (except at the very ends of the stopped dada or tenon), no long grain to long grain mating surfaces. Now I understand the thought behind the foxed tenon. So I think we all would agree a stopped sliding dovetail would work great, but I'm assuming you don't want to see the joint on the one side that is not stopped. 

I really like the idea of a few smaller through tenons wedged and finished flush with the top, that would work very well and would be pleasing visually, but it also seems like you might be against that also.

A domino would shine here for this joint, it's really what most guys would use that are making furniture for a living and I think it would do very well. But with that said, if you don't have the domino..... 

My other question is how set are you on that design? If the vertical support went from top to bottom, and wasn't suspended, then you've eliminated the problem, the vertical piece would be supported and you could do pretty much anything like a stopped dado for that joint and get the clean look you are after

[Ronn W]

IMO the best joint would be a good fitting sliding dovetail.

[wtnhighlander]

All great thoughts everyone, and thanks for the input. Let me answer some questions I see.

First, the suspended shelf was specifically requested by the client.

Second, wedged through tenons would be my first choice, but an unbroken top surface was also requested.

Third, I like the sliding stopped dovetail, and will probably go with that, as the design has evolved to include some trim that can disguise the open end, and execution is relatively simple.

Third, @Chestnut has pointed out that multiple stub tenons can provide considerable long grain glue surface, which will hold quite well, even when the grain is perpendicular.

Last, the foxed, a.k.a. blind wedged tenon idea would certainly hold without glue, but presents a level of difficulty I prefer to practice on prior to using it in a paying job.

Thanks again for your ideas. That is one thing I can always cound on here!

[Wood Basher]

6 hours ago, wtnhighlander said:

Third, I like the sliding stopped dovetail, and will probably go with that, as the design has evolved to include some trim that can disguise the open end, and execution is relatively simple.

I may be misinterpretting what you mean here. I envision a stopped sliding dovetail where the entry is from the rear and the joint stops short of the front. But if that is true, why would you add trim to disguise the open end when the open end is at the back, up against the wall?

[wtnhighlander]

3 hours ago, Wood Basher said:

I may be misinterpretting what you mean here. I envision a stopped sliding dovetail where the entry is from the rear and the joint stops short of the front. But if that is true, why would you add trim to disguise the open end when the open end is at the back, up against the wall?

A bench for sitting will not necessarily have one side against the wall.

[Chestnut]

That could make a really cool space divider that offers seating. I like that idea, too bad i don't have any spaces like that in my house. I'm interested to see how it turns out.

12 hours ago, wtnhighlander said:

Third, I like the sliding stopped dovetail, and will probably go with that, as the design has evolved to include some trim that can disguise the open end, and execution is relatively simple.

This is probably better than my idea as assembly could get dicey trying to seat tenons in mortises that are on perpendicular planes.

13 hours ago, Bmac said:

Not so much from the weight of the contents on the shelf, but more so just from the weight of gravity on that joint. Any tenon or dado will not have the best surfaces oriented for gluing. No matter what you do there will be end grain involved with each mating surface on the glue joint (except at the very ends of the stopped dada or tenon), no long grain to long grain mating surfaces.

You have to rotate your frame of reference 90 degrees (think blind box joint except in the middle of the board) and there is a decent amount of long grain glue area. Also keep in mind that multiple stub tenons  with a shelf in that orientation would work as a system. Gravity is goign to pull the shelf down but the left side is resisting that downward force. The shelf is then trying to rotate around that left connection point effectively wedging the shelf in place. This is a similar mechanic to how hold fasts work and they hold amazingly well without glue.

[Mick S]

I'm thinking about a dovetail also, just not a sliding dovetail. hard to describe, but doable.

What if you mortised out an area on the underside of the top immediately to the right of the vertical member (VM). The mortise would be about an inch shorter than the width of the VM and the same thickness. Depth would be ½ the overall thickness of the top. Where the VM will meet the top, cut dovetail pins. Cut dovetail tails into the end of the VM to match, slide the VM into place via the mortise and plug the mortise. Sort of like a stopped half blind dovetail.

[Immortan D]

Floating shelf hardware may work.

[wtnhighlander]

That is some outside the box thinking, Mick!

Might be a bit tedious to execude, but less risky than a blind wedged tenon, and would certainly be very strong.

[wtnhighlander]

@Immortan D, a floating / cantilevered shelf was my first suggestion, but clients wants it boxed in. I suppose the supports would still do the trick, though. With hardware like that supporting the shelf, the vertical member would be mostly decorative, not requiring such a strong joint.

[Immortan D]

2 minutes ago, wtnhighlander said:

@Immortan D, a floating / cantilevered shelf was my first suggestion, but clients wants it boxed in. I suppose the supports would still do the trick, though. With hardware like that supporting the shelf, the vertical member would be mostly decorative, not requiring such a strong joint.

I meant using the hardware to support the vertical member, so the poles will be pointing down. I don't know if it is going to be strong enough with just a screw, maybe epoxy on the poles could help. No guarantees though.

[Bmac]

4 hours ago, Chestnut said:

You have to rotate your frame of reference 90 degrees (think blind box joint except in the middle of the board) and there is a decent amount of long grain glue area. Also keep in mind that multiple stub tenons  with a shelf in that orientation would work as a system. Gravity is goign to pull the shelf down but the left side is resisting that downward force. The shelf is then trying to rotate around that left connection point effectively wedging the shelf in place. This is a similar mechanic to how hold fasts work and they hold amazingly well without glue.

Very true, I had not considered the support from the horizontal piece. And yes, now I can envision what you mean by long grain glue area when you used the box joint reference. Sometimes it takes me a while. So a true dado joint, if blind on both ends would just have long grain gluing surfaces where the dado ends on the front and back, but the multiple stub tenons increase these long grain gluing surfaces, now I get it.

[derekcohen]

Assuming the material is solid wood, my choice would be a tapered and stopped sliding dovetail. Be mindful that solid wood needs to move. There is a cross grain join. Dominos and biscuits will not work well.

The taper will allow the joint to tighten up strongly. This is important since it is in tension rather than compression. 

The stop would hide it. 

Regards from Perth

Derek