Segmented work using curves

[42Olds]

I'm thinking about making some bowls with a segmented center ring that has a repeating pattern that includes curves - an example of what I want would be a celtic rope like at http://www.earlham.edu/~peters/buziak2.gif. I probably should look at doing this with chip carving, but that has problems of it's own - the biggest being that I've never done that!

I thought that curves with-in a segment would be very hard to do, but have seen some on the web...

Any hints on doing this would be appreciated - the bowl will be in the 8 to 12 inch range.

Thanks,

Stephen

 

[Joe Greiner]

It would be quite tedious at the small scale you apparently seek, but two pieces of wood (i.e. adjacent segments) can be closely matched by attaching them temporarily to a backing of some sort, and running a saw along the mis-matched joint. Works best with straight lines (usually used to get perfect miters in picture frames etc.), but should be feasible on curves of limited arc length using a scroll saw. (Might need a spiral blade.) The idea is that the sides of the kerf are parallel/concentric so as to make up any mismatch. The saw helps to maintain a constant kerf width at a varying angle. Note the angle can't be very large; otherwise parallax (for want of a better word) would throw it off.

An option you might consider is to inset an inlay strip made up of veneer as in marquetry. Grizzly has a limited variety of these in their guitar supplies section. I expect that others such as Woodcraft, Rockler, etc. would have these, and probably in more variety.

Joe

 

[n7bsn]

I would love to figure out how to do a true celtic knotwork segmented setup. Not the offset rings (they really aren't knotwork).
The closest I've been able to puzzle out would be pure rectangular, which still isn't perfectly right.

 

[GeorgeH]

Let's take a look at the design. The inter-laced segments consist of only two different shaped pieces; one is straight, the other arched. The arched piece is reversed at every other intersection to create the alternating direction in the pattern. The other piece is relatively straight. If we created a length of material in the shape of the attachment, and cut it into strips, we'd have the alternating segments. Then all we have to do is cut the straight pieces that fit between these. That should give us an open weave. Now all we have to do is cut the pieces to fit into the openings (again only two shapes) to fill in the gaps. Labor intensive? Sure it is. But the beauty of the results might be worth the effort.

 

[42Olds]

Let's take a look at the design. The inter-laced segments consist of only two different shaped pieces; one is straight, the other arched. The arched piece is reversed at every other intersection to create the alternating direction in the pattern. The other piece is relatively straight. If we created a length of material in the shape of the attachment, and cut it into strips, we'd have the alternating segments. Then all we have to do is cut the straight pieces that fit between these. That should give us an open weave. Now all we have to do is cut the pieces to fit into the openings (again only two shapes) to fill in the gaps. Labor intensive? Sure it is. But the beauty of the results might be worth the effort.

Yeah, filling in the gaps of the curved portion would be a very interesting project...as Joe suggests, it might be done using a really precise jigsaw, assuming you have a really precise jigsaw. I can see that as a way if you do every segment separately, but not as a "stick" that is glued together then cut into segments.

I am wondering if you could do the curved parts using 2 small radius router bits - one for the inside curve, the other for the outside curve. Only questions would be if such small bits are available, and if the radius of the work would match up enough to form a solid joint.

I've thought about the inlay route - but haven't done any close knife work like that and am not sure how hard it would be to make the ends match up. I suppose I could use border pieces to isolate 2 or 4 bands of inlay.

I agree, it would be quite beautiful to see.

Stephen

 

[Joe Greiner]

See

http://n-fl-woodturners.org/ShowNTell.asp?meetingdateid=73

Scroll down to "Joe Greiner" for an example of a celtic knot executed in the round. The blank must be "perfectly" square. Make four successive cuts and glue-ups at 45 degrees. (i.e. cut, glue, cut, glue, ...) Also, the saw kerf must be the same width as the veneer thickness, else misalignment will grow with successive cuts. I don't know how this procedure could be adjusted to make smaller segments for assembly in a circle. Maybe separate pieces for the crossing tangent portions, but the arcs still wouldn't work quite right.

Scroll further down to "Bill Robbins" for an example of the inlay strip. I'm pretty sure the inlay strip was "store-bought," not DIY. To get the ends to mate, you need to turn the recess gradually, and test fit frequently. A preliminary practice turning would be advisable to establish an allowance for the glue thickness. I suppose you could also make a generous allowance for the glue, and accept a thicker glue joint than normal - let the butting ends establish the outside diameter and hence the glue layer thickness.

[Not quite what you're driving at, but I just noticed there's a picture of a burnt celtic knot pattern on page 38 of the Spring 2007 AAW journal. You could even dye some of the portions. (Same issue, page 30) How comfortable are you with lying about it? ]

Joe

 

[John Abt]

The Chesapeake Woodturners currently has a juried show at Montpilier Mansion in Laurel, Maryland. The "Best in Show" award went to John Noffsinger who had used pyrography to do a Celtic knot around the perimeter of a hollow vessel.

Mike Vore took pictures and are available at http://mike-in-columbia,smugmug.com.

Glen Bohursch also wrote an article "Intertwined Ring Vases" that appeared in the Fall 2006 issue of Woodturning Design.