Finger Jointing Plywood: Discussion Thread

[ancient kayaker]

I have seen finger joints for plywood in CNC cut ply boat kits and would like to try that at home, but with simpler equipment.

I could use a router with a flush cutting bit, but I would need a matching pair of perfectly-fitting guides. Such guides are probably available, but they are probably hard to get and expensive so I would like to make my own. Also, that would allow me to vary the pattern, which might be interesting.

Idea 1: cut a wavy slot across a piece of scrap ply using a 1/8 dia bit, yielding 2 guides that are 1/16 offset from the required cut. Then cut the planks using a flush cutting bit, with the guide roller upsized by 1/8. This will work in theory, the problem is, guide roller thickness changes with diameter so the new roller is too thick to fit.

Idea 2: cut a wavy slot across a piece of scrap ply using a jigsaw, yielding 2 guides, one of which is slightly offset from the finished cut. Paint the edge of the "good" guide - several coats, sanded to get a smooth finish, wax it and "glue" the guides back together with epoxy. Once set and sanded flush, the "joint" can be broken yielding 2 perfectly fitting guides. Then the plank pieces can be cut using a regular flush cutting bit.

Any comments? Any other ideas out there you have heard of or tried out? I haven't tried either one out yet as I am still thinking about it and shamelessly trolling for ideas.

[wardd]

http://www.duckworksmagazine.com/06/...zzle/index.htm

[ancient kayaker]

Thanks Wardd: yes I've seen that one. The puzzle joint shape consists of semi-circles joined by straight lines at right angles to the joint. To me it looks more like a series of halved joints than a true finger joint, and Duckworks seems to think that the joint needs to be taped to have adequate strength.

I think the joint shape needs be more of a zigzag with the radius of the circular arcs - where the irection changes - kept to a minimum for maximum strength. I don't think these joints should need taping to be as strong as the ply, like the finger joints in paintable wood moldings.

[wardd]

the thinner the fingers will give you a greater length of glued joint so it would seem to me stronger, like a fine thread is stronger than a course thread

but it could be carried too far so tests would be in order

i have a hunch the min width should be at least 2 x thickness

[ancient kayaker]

Originally Posted by wardd

the thinner the fingers will give you a greater length of glued joint so it would seem to me stronger, like a fine thread is stronger than a course thread

but it could be carried too far so tests would be in order

i have a hunch the min width should be at least 2 x thickness

The parallel parts of the joint don't do much IMO, except if stressed in tension parallel with the face grain. For bending forces only the glue at the end of each straight run is working; that is where I expect failure will start as they are only butt joints, so the length of the joint is key.

I have thought some more and realised a few things:

A zigzag consisting of straight lines only with sharp ends is stronger than a zigzag with rounded ends. The purpose of the rounded ends is to allow the cut to be made using a router bit. I think you are intuitively correct about the length of the joint, which should be at least twice the plank width: I have tested butt joints in plywood and they have a bit more than 50% of the strength of the plywood.

A zigzag consisting of straight lines can probably be cut precisely using a flush trim saw to avoid damaging the guide. See attached. The 30 deg angle provides joint length 2x plank width and allows use of a standard 3-sided file for cleaning out the ends of the cuts.

Note: this is not an interlocking joint so it will require clamping.

[wardd]

Originally Posted by ancient kayaker

The parallel parts of the joint don't do much IMO, except if stressed in tension parallel with the face grain. For bending forces only the glue at the end of each straight run is working; that is where I expect failure will start as they are only butt joints, so the length of the joint is key.

I have thought some more and realised a few things:

A zigzag consisting of straight lines only with sharp ends is stronger than a zigzag with rounded ends. The purpose of the rounded ends is to allow the cut to be made using a router bit. I think you are intuitively correct about the length of the joint, which should be at least twice the plank width: I have tested butt joints in plywood and they have a bit more than 50% of the strength of the plywood.

A zigzag consisting of straight lines can probably be cut precisely using a flush trim saw to avoid damaging the guide. See attached. The 30 deg angle provides joint length 2x plank width and allows use of a standard 3-sided file for cleaning out the ends of the cuts.

Note: this is not an interlocking joint so it will require clamping.

even with the zigzag the tips and troughs can be radioused with no ill effect and still use a router

[ancient kayaker]

Originally Posted by wardd

even with the zigzag the tips and troughs can be radioused with no ill effect and still use a router

My concern is, the radiused parts of the joint are effectively butted and are therefore weak spots. This suggests to me that the radius should be as small as possible. The "pointy" zigzag therefore seems the logical choice.

If the zigs and zags were nice and large with the curves joined with long more-or-less straight lines, radiused roots and tips should not have too much of an effect. But for me, with the very small boats that I make, such large joints might spoil the appearance, unless it turned out to be virtually invisible, and there is also the question of waste material - like using unecessarily wide scarfs.

I think the flush trimming saw will be nice to use: it cuts at least as fast as a small router bit can, and the cut needs less cleaning up. When I route ply (okoume and luan) it leaves a lot of fuzz o the edge and it is easy to burn the wood, whereas the saws cuts very smoothly and is as accurate as the surface guiding it. Now the idea needs to be tested!

Looking at router bits intended for finger jointing of solid wood, I see that the joint has long more-or-less straight sections joining tip and root radii of comparatively small diameter, so I think I am on the right track.

Returning to method 1 in my first post, Freud make rabbeting bits with a set of guide bearings for different rabbet depths.

[wardd]

ok , one possible problem with the zigzag is that you are butting the zigzag

if you don't get the zigzag closely matched you'll have gaps

[ancient kayaker]

Originally Posted by wardd

ok , one possible problem with the zigzag is that you are butting the zigzag

if you don't get the zigzag closely matched you'll have gaps

If the patterns mesh perfectly then the planks should do so as well, so a little time spent "off-line" from boat building making good patterns should pay for itself later. At this point accuracy is my only real concern, I am convinced it's doable - just need to find some time! If it's not possible to make it as strong and better looking than a scarf joint then I will probably not use it.

[wardd]

also at the bottom of the v there may be stress concentration due to flexing

not sure on this but something to consider

[ancient kayaker]

Originally Posted by wardd

also at the bottom of the v there may be stress concentration due to flexing

not sure on this but something to consider

It's a good point although if the joint is bonded well, it should be minimal. I think only a force across the width of the plank would cause that, which is probably the least stressed direction for a typical boat plank. If it does happen though it should still be less than for the rounded joint.

[Ilan Voyager]

I'm a bit slow and old. But what's the interest and advantage of using finger joints instead of scarfs if the panels are structural?

[ancient kayaker]

Originally Posted by Ilan Voyager

I'm a bit slow and old. But what's the interest and advantage of using finger joints instead of scarfs if the panels are structural?

Like I said -

Originally Posted by ancient kayaker

... If it's not possible to make it as strong and better looking than a scarf joint then I will probably not use it.

- and of course, being me, it's also about trying out something I haven't done before!

[PAR]

My understanding of these types of joints (finger) is they should be in the thickness of the material, not the breadth for best loading results. I've experimented with stepped scarfs, which are very easy to machine with precision. Testing suggested these were just as strong as a conventional scarf.

I also once saw a 7 layer veneer of 3/4" plywood, where every other veneer was knocked out of one piece for a few inches and the exact opposite was done to the matting piece. The result was pretty and tough, though I didn't like the natural stress risers it offered. For my comfort level, it needed more fingers to be acceptable.

All of these joints I've described are in the thickness dimension of the material (the end grain) as is a regular scarf.

[Ilan Voyager]

For transmitting the stresses continuously for one panel to the other, no steps or changes of direction are allowed if you want to insure a resistance of the joint equal at least 95% of the panels resistance. There is no other solution than the scarf able to offer such percentage...

It must be insured that the joint is not starved in glue. With finger joint is impossible to control that. Finger joint ask also for a very precise machining, easily done on a small section of massive wood, almost impossible on a plywood panel. The glue lines of the plywood will kill also the router bit in a short time, and these bits are pretty expensive.

With epoxy, gluing scarfs is a breeze as the EP fills all the minor defects.
A good scarfing table with a circular saw running on slides makes cutting scarfs more than easy for those who are not masters of the planer. While building 40 feet pro fishing boats in plywood epoxy, we had hundreds of scarfs to make until 10 feet long (panels 10*5 6mm thick), so I made a vacuum scarfing table with 2 saws. One 10*5 panel was scarfed on the 4 sides in less than 5 minutes by the aprentice...

I do not know if the Gougeon bros sell anymore the Scarffer. The cost was low, and allowed to scarf until 1/4 of inch or at least to make the most delicate part, the feather side of thicker panels. Easy to finish the remaining part. Good blades ares less expensive than special finger joint bits.

On 3 mm (1/8) okoume panels a belt sander on guides does the job in minutes, if you have a good straight table.

Scarfs are easy to machine (yes, there are easy with the right tooling), easy to glue and reliable. Furthermore scarfs follow all the good rules of naval engineering. Why to go in complications?