CNC version of Knapp joint

[Ronn W]

While visiting relatives last month I was admiring a slant front secretary desk and noticed amazing drawer joinery where I expected to see dovetails.  I have since learned that the piece was made using Knapp Joints and that these date to the late 1800s' and that  Knapp invented the machine to make them.  My question is,  "Does any one use these joints today?"  Has anyone seen a jig that would allow these joints to be made without the Knapp machine or CNC machine?  No big deal. Just curious.

[Brendon_t]

I've never seen them but I'm sure if you're dedicated enough,  a few hours of trial and error could yeild a set of templates for router cutting. 

[wdwerker]

Templates and  a lot of patience. 

[bradpotts]

MLCS sells jigs that make them and other fancy joints like that. http://www.mlcswoodworking.com/shopsite_sc/store/html/smarthtml/pages/catalog.html

[Ronn W]

Thanks bradpotts,

Expensive toy but very cool.

[weithman5]

i could see imitating it at least the one half by drilling the holes with forstner bits then a smaller diameter hole in the middle of those to put dowels. I haven't figured a good way to cut the piece that flips over those pins yet.... but i am pondering it

[Ronn W]

Thanks for your responses.  I have continued to look on line and, while I have found similar joints and jigs for similar joints, none have the scalloped look of a real Knapp Joint, where the wood between the semi circles meet in a point. See pic of antique drawer attached.

[wtnhighlander]

To me, that looks like a forstner drill and dowels. The mating side would be coped to match the overlapped forster pockets, then the centers drilled and dowels glued in. The same fence / jig used to position the drawer front for the overlapping pockets could be used to allow marking the side scallops by lightly scoring with the same forstner bit. That marks center for the dowels, too. Assemble, then plane or sand out the score marks.

I think it is purely decorative, though. The scallops add no real strength to the joint.

[Ronn W]

Here is another pic that I found.  The pins appear to be part of the drawer front, not separate dowels.   I will try to attached a pic that show this and also a pic of the original Knapp machine.

[Tpt life]

I think a shaper type cutter can cut the side by cutting from the end of the board. The forstner type bit for the face is not a bad thought. The bit would just need a hollow core for the first inch or so. 

[Ronn W]

I think I see what you are thinking but that would be a huge bit.  I have seen bits that do beaded table edges but this one would have to be the depth of the drawer.  I think you have identified the toughest part of duplicating this joint - the points where the curves meet on the drawer sides - very sharp interior corners.  No wonder he had to invent a machine to do it.

[Tpt life]

11 hours ago, Ronn W said:

I think I see what you are thinking but that would be a huge bit.  I have seen bits that do beaded table edges but this one would have to be the depth of the drawer.  I think you have identified the toughest part of duplicating this joint - the points where the curves meet on the drawer sides - very sharp interior corners.  No wonder he had to invent a machine to do it.

Huge is a relationship. The spindle would be long, but the width would only need to be the spindle plus cut depth. If it turned while supported from both ends, it would resemble a lathe operation. I am just guessing what my approach might be if I need to put this into production. If the bit spun down into a table, chip out could be countered with scrificial backing. 

[Ronn W]

Aha, I see where you are going.  Horizontal mounted spindle at least as long as the drawer is deep.  The spindle, in addition to spinning down would have to actually move up and down by the thickness of the material (or the materail would have to raised and lower.  I bet it's the latter - probably a foot pedal (this was pre-electricity).  I just did a brief online search and could not find any photos or drawings or detailed description of how Knapp's machine was built. There is only one drawing of the overall machine which appears one several sites.  I will look again when I have some time.  If we make this tread long enough we will have it designed. 

[Slick218]

Hand tools ... That's all

[WillAdams]

I worked up an OpenSCAD file for making such joints --- it's on Thingiverse and should easily be findable. Wrote up the design process on the Carbide 3D Community Forums --- let me know if you have questions.

[Robby W]

I believe the original machine used a bit with a hollow center, sort of like some plug cutters, to cut the drawer front. Plunge it in and you get the circular recess with the pin left in the middle. Index to the next location and repeat.

The side would be a combination of a drilled holes and a cutter that left a half round shape. 

Not hard to do with a dedicated machine.

[wtnhighlander]

 

[lewisc]

@wtnhighlander Pask Makes is one of my favourite Youtubers. He's got a great channel that's quite entertaining.

[Ronn W]

Thanks for the video.

[wtnhighlander]

Here's a version using Matthias Wandell's pantarouter:

 

[pkinneb]

I actually prefer the crispness of the homemade version. Thanks for sharing!!

[Mark J]

I thought both videos were fun to watch and neat techniques.  I was interested to see Wandell throw sawdust on the glue squeeze out of the first joint he put together.  I've never seen that done before.  What was the purpose?

[wtnhighlander]

1 hour ago, Mark J said:

I thought both videos were fun to watch and neat techniques.  I was interested to see Wandell throw sawdust on the glue squeeze out of the first joint he put together.  I've never seen that done before.  What was the purpose?

That's a quick way to 'clean up' the squeeze-out. Sawdust absorbs the glue, and you just rub it away.