PDA

View Full Version : Bent Miter operation



dp
09-04-2011, 10:49 PM
How would you cut this using a machine? The dimensions don't matter - the assembled part should explain it all. It is a 90 quarter round at one end of the gap, square at the other, the left side wraps around the quarter round so the gap is Circumference/4 wide.

To get you started assume a radius of 1 and go from there. Any flexible but non-elastic material that can be made to bend in this fashion.

This is a common construction method in the PNWet that some of you locals may recognize. It is a bent wood box corner.

http://metalworkingathome.com/images/bentmiter.png

aostling
09-04-2011, 11:07 PM
I've seen those bent-wood boxes in the museum on the UW campus, made by the Haida, I think, or some related tribe. I'll be interested in the answer to your question.

Do you know how the Indians made the cut without a machine?

Chris S.
09-04-2011, 11:09 PM
A Scroll Saw.

Chris S.
09-04-2011, 11:17 PM
I've seen those bent-wood boxes in the museum on the UW campus, made by the Haida, I think, or some related tribe. I'll be interested in the answer to your question.

Do you know how the Indians made the cut without a machine?

Not sure. Did Indians have Coping Saws?

Really though, they were using stone cutting tools when the Europeans arrived, or are you referring to India Indians? :D

Chris

dp
09-04-2011, 11:36 PM
I've seen those bent-wood boxes in the museum on the UW campus, made by the Haida, I think, or some related tribe. I'll be interested in the answer to your question.

Do you know how the Indians made the cut without a machine?

They also make simple 45 bent miters, but this one from a museum in Victoria caught my eye as I'd not seen the style before. The Haida had access to volcanic glass and there is quite a bit of flint on the northwest coast. An original pre-contact bentwood box is worth a load of money though if I had one I'd give it back. They were used for all manner of things. I thought it would be an interesting project to try this particular corner method. The final corner is stitched. The bottoms are either stitched or inset and the box wrapped around it.

dp
09-04-2011, 11:56 PM
A Scroll Saw.

I think a series of form cutters in a shaper can do it. A chisel can do it in wood, of course, as can a plane, but to do this in steel or aluminum requires a machine.

2ManyHobbies
09-05-2011, 09:09 AM
In wood and some plastics, that would be two passes with a router. I'd imagine metal to be similar but the radiused tool would be a bit more rare.

wierdscience
09-05-2011, 09:29 AM
Plow a groove with an endmill first,then grind a custom form mill to mill the radius and undercut.Could probably start with a dovetail cutter for the blank and grind the radius in it,

Forrest Addy
09-05-2011, 09:58 AM
I'm not sure what that joint is called. As was said you see it in Haida and Kwakiutl storage chests.

Can't make that joint with a rotary cutter. It has to be shaped. Do the math. A rotary cutter has to be twice as wide as the profile it cuts and add for the shank. The width from the tangent to the shoulder is r * pi / 4. The gap is 0.2146 * r. Steam to bend.

It makes a strong joint. Handsome too: the grain wraps right around the corner. I've tried them on narrow stock and a bandsaw with a skinny kerf works well but the interior angle has to be incised if the joint is to be invisible. Back in the day the native craftsman split and joined 30" wide boards from cedar downed 100 years earlier. No such material exists today.

I've seen these boxes in the Bagley Write Museum and the BC Museum in Victoria and they are superbly made. Before the white man and steel came along the Haida and PNW Coast Indians (yeah, I know Native Americans or First Nation but the local Suquamish glare at you when you refer thus to them over the fireworks counter or the salmon stand.) Anyway, it's amazing the quality of the craftsmanshi attainable with shell, bone, and stone tools.

It aint the tools. Tools make it quicker not better.

Chris S.
09-05-2011, 10:14 AM
<snip>

In wood and some plastics, that would be two passes with a router.

A rotary tool would leave too wide a kerf and produce ugly "gaposis". :D

Chris

vpt
09-05-2011, 10:19 AM
I'd use the router and make two passes. One pass with a strait shaft to open up the groove, then the second pass with an inverted radius cutter to do the undercut. Use a fence with clamps on the end to make the cuts strait.

Chris S.
09-05-2011, 10:37 AM
My best guess is that after the cuts were made, the artesian soaked the wood in water or steamed it, then stretched it tight when bending the joints. Roy Underhill demonstrated this on one of his shows. He steamed the wood. With modern yellow glue the joint will swell on its own accord. Since the glue penetrates the wood fibers it will not shrink when it dries. Either way, wet or steamed wood is best for bending.

FYI, a very old wood workers trick for removing a dent is to lay a wet cloth on the work and heat it with a hot iron. When it dries it does not shrink back.

Chris

Chris S.
09-05-2011, 10:51 AM
I'm not sure what that joint is called. As was said you see it in Haida and Kwakiutl storage chests.


I'm fairly certain it's called a "Bent Miter".

CYA, I could be wrong.;)

Chris

lynnl
09-05-2011, 02:29 PM
Can't make that joint with a rotary cutter. It has to be shaped. Do the math. A rotary cutter has to be twice as wide as the profile it cuts and add for the shank. The width from the tangent to the shoulder is r * pi / 4. Steam to bend.



If it's going to be bent anyway, what's to say one couldn't bend the squared shoulder section back out of the way, whilst milling the radiused portion?

If I were doing it in wood I'd probably do something like that. Or just produce the radiused end on a thinned down piece, then laminate the square shoulder section, with a thin extension, to build up to final thickness at the same time the bend was to be finalized. Paying attention to grain matching of course.

Or... what about just doing it with some careful band saw work?

There's no way it's going to be an invisible joint, at least when viewed on edge, due to long grain crossing end grain.

dp
09-05-2011, 02:36 PM
If it's going to be bent anyway, what's to say one couldn't bend the squared shoulder section back out of the way, whilst milling the radiused portion?

If I were doing it in wood I'd probably do something like that. Or just produce the radiused end on a thinned down piece, then laminate the square shoulder section, with a thin extension, to build up to final thickness at the same time the bend was to be finalized. Paying attention to grain matching of course.

Or... what about just doing it with some careful band saw work?

There's no way it's going to be an invisible joint, at least when viewed on edge, due to long grain crossing end grain.

Making this joint in wood is rather trivial but esthetically interesting. Doing it in metal, which is actually my objective, is mechanically challenging. It cannot be done with a conventional rotating cutter. It will require a broach or other linear cutter.

aostling
09-05-2011, 03:20 PM
Doing it in metal, which is actually my objective, is mechanically challenging. It cannot be done with a conventional rotating cutter. It will require a broach or other linear cutter.

I'm mystified as to why you would want to make this joint in metal.

You could stamp the shape in a thin plate, then stack dozens of plates to get the depth of box you need. But I'm guessing that is not what you are on about.

dp
09-05-2011, 03:30 PM
I'm mystified as to why you would want to make this joint in metal.

You could stamp the shape in a thin plate, then stack dozens of plates to get the depth of box you need. But I'm guessing that is not what you are on about.

Just to see if I can.

winchman
09-05-2011, 03:30 PM
Could you make it with an oscillating tool with a form cutter?

Smokedaddy
09-06-2011, 12:27 AM
DP,

Is this what you are looking for?

http://www.finewoodworking.com/item/1800/a-deco-box-with-kerf-bent-corners

-SD:

dp
09-06-2011, 01:13 AM
DP,

Is this what you are looking for?

http://www.finewoodworking.com/item/1800/a-deco-box-with-kerf-bent-corners

-SD:

Nope - that is how I bend plywood, though. Bent wood boxes as made by the PNWet tribes are steam heated and bent as the simple drawing shows. And they use hand tools to create the cut. That I have done using a "V" rabbet to notch a cedar plank. Once well steamed cedar bends easily. The V rabbet is the more common bent miter joint and very easy to cut out with hand tools, even when made from stone or shell.

I'm surprised at how difficult it is to locate good pictures of bent wood box details, but this one shows the beyond bookmatching that happens with the grain:

http://www.flickr.com/photos/kostaspagiamtzis/3780332750/

Cedar also planks very well using wedges, and original growth cedar trees 10' in diameter were planked to make walls for huts and longhouses. Splitting doesn't raise much dust which in the case of cedar can be quite nasty. Bugs don't like it, either.

However - it is my intention to reproduce that joint in aluminum and steel, partly as an exercise and also as an art piece.

Forrest Addy
09-06-2011, 03:43 AM
You guys who would make the joint illustrated in the OP of this thread should lay it out full size before you make any rash statements of how-to. Better yet, make one. Let up know which router bits work best.

As I said I made a couple of sample joints of 1 x 4 (19mm x 89mm) using a 0.015" (0.36mm) band saw blade. It worked pretty well but wide boards 12" nominal (300mm) where the band is easily steered by the grain irregularities pose limitations; chief of which is the cutting height Vs secion of the bandsaw itself.

If I was going to make a couple of boxes using this joint I would lay it out, saw kerf down the the "tongue" thickness to form the butt edge and then, parallel to it, saw the width of the gap. Slick out the waste and finish the depth of the gap. Saw at about 45 degreesd to come almost taangent to the radius. Slick out the waste and clean up with the router place to depth. After that I'll have to get creative making some special tools one to shape the radius cross grain and one to clean up the inside corner where the radius washes out to the tongue flat. The last tool may be a modified router plane cutter. I have to study on the first tool before I'd commit to metal but I think it might look like a "drag iron".

Very interesting problem if the joint is to roll up tight to the radius, the gap wall seats tight, the joint intersection is a neat right angle, and the steamed radius is smooth, unsegmented, and free of cross-grain fractures.

Black Forest
09-06-2011, 04:02 AM
metal seems to be easier to make this joint than wood. Make first pass with endmill. Heat the bottom thin area and bend opposite direction of final bend. Use a form cutter to make the radius. heat again and bend up to final position.

What am I missing? Of course why would you do it in metal!!!

Forrest Addy
09-06-2011, 08:32 AM
OK. Let's see if I got this right. Working it Metric for the German crowd.

metal seems to be easier to make this joint than wood.
Starting with 25 mm aluminum plate.

Make first pass with endmill.
Working the math and using a 20mm radius the gap slot would be 20mm x .2146 = 4.292 mm wide x 20mm deep. (roughly 5 to 1 depth ratio but readily do-able with a 4mm endmil. (Hold the 0.292mm stock for clean-up after re- bend)

Heat the bottom thin area and bend opposite direction of final bend.
OK, but we should have milled the slot with a ball nose cutter. Aluminum is pretty notch sensitive and may fracture if bent with a sharp inside corner nearby. But how far do we bend? Looks like damn near 90 degrees.

Use a form cutter to make the radius.
Now that's straight forward but the lower edge of the radius cutter has to be ground away to a vanishing thickness.

heat again and bend up to final position.
Well, no. You have to heat and restore the machined joint prep flat and doctor up details so the radius is in proper relationaship to the gap. THEN heat and fold up so the thin part lies tight to the radius and the vertical wall laps tight to form a right angle. Presumably without metallurgical damage to the aluminum alloy.

What am I missing? Of course why would you do it in metal!!!
My sentiments exactly. Wood is SO much easier.

john b
09-06-2011, 09:30 AM
20x2 =40
40x3.141=125.64
1250.64/4=31.41
31.41-20=11.41=gap
john b.

dp
09-06-2011, 10:03 AM
My sentiments exactly. Wood is SO much easier.

It truly is - but then this question would never have come up here :)

However - the question of why I would want to do it in metal should not be an issue. True, it is my decision, but what if I were being paid to do this by a wealthy state bureaucrat who wished to have a piece of fine art for his new boondoggle in down town Olympia? :)

dp
09-06-2011, 10:11 AM
You guys who would make the joint illustrated in the OP of this thread should lay it out full size before you make any rash statements of how-to. Better yet, make one. Let up know which router bits work best.

Exactly - this is also a bit of a puzzle once you start to really think it through. No spinning tool can do the job.


The last tool may be a modified router plane cutter. I have to study on the first tool before I'd commit to metal but I think it might look like a "drag iron".

Again, exactly right.


Very interesting problem if the joint is to roll up tight to the radius, the gap wall seats tight, the joint intersection is a neat right angle, and the steamed radius is smooth, unsegmented, and free of cross-grain fractures.

One person suggested cutting a rabbet and gluing a quarter round in place. He got credit for manufacturability and zero for craft. :)

It is perfectly possible to use a full round in this joint if that were the objective, and it would be an attractive, wrapped joint when seen from an edge, and a half-round router cutter could be used to create the slot.

Forestgnome
09-06-2011, 10:28 AM
I'm thinking a pull scraper.

kendall
09-06-2011, 12:34 PM
I'm thinking a pull scraper.

Could possibly mill the slot, then use a shaper with 2 or three different cutters on it to creep up on the actual profile needed. Though if you could rotate the tool holder on the vertical axis, it could possibly be done with one cutter.
(rotate from \ to _ )

Other wise mill slot, us dovetail to get close, and use progressively closer profiled scrapers to get the finished profile.

dp
09-06-2011, 12:43 PM
I had thought of trying a special cutter on my shaper. It requires locking the clapper, and the cutter has to be able to rotate slightly on the return stroke as if it were hinged. I'm also going to see if the head of my shaper will rotate 90. I'd never considered it before, but that would work, too. A return tensioner would also bee needed.

Weston Bye
09-06-2011, 12:57 PM
Design adjusted for manufacturing. Probably not as strong as the original design.
http://img.photobucket.com/albums/0803/Weston/MVC-011F.jpg

strokersix
09-06-2011, 12:59 PM
Tools make it quicker not better.

Yup. Kind of like the quality workmanship thread. It's an attitude.

Forrest Addy
09-06-2011, 01:55 PM
Hm, there is a version of this "partail penetration wrap joint" that's symmetrical and could be made with a router cutter. I just thunk of it which is a sure indicator that everyone knows about it.

Imagine the joint made so the radius is shared on 45 degrees of arc on each side of the joint and when laid flat the abutting faces form a truncated V having a 90 degree included angle. That doesn't make a very good picture in my head. I need to work on it.

Dang! I was writing this but had to leave quick. When I came back I posted it. Then I looked back over recent posts and see Weston bye beat me to the punch. Hate it when that happens.

dp
09-06-2011, 08:27 PM
Design adjusted for manufacturing. Probably not as strong as the original design.


I like that - doesn't solve the problem of racking which is a bit of a curse for bent wood boxes, but it should also be easy to key the miter to prevent that.

huntinguy
09-06-2011, 08:54 PM
In wood, I would think it is done with something a kin to a old woman's tooth.

In metal... I think you have just found the reason to own a shaper. :p

You don't have a wire EDM do you?