PDA

View Full Version : Question on segmented turnings



Steve Bearden
05-10-2010, 8:48 PM
Hey guys, i have a quick question to all the guru's. When i am cutting my segments how do i know the angle on them is correct without cutting a bunch to see if the circle closes up tight. I have a digital protractor and find that my jig is set up to within a tenth of a degree and i am using positive stops for length control. When i get twelve cut i glue half and half with a dowel in between the two halves to take out some of the error. When it's time to glue the two halves they are what i would consider horribly off. when i cut enough off to get the angle right now i have an ellipse. I am being as precise as i can and still no luck. now for the question how do i figure if the angle on the jig is too small or too long with out wasting a bunch of segments. Sorry for the rambling just a little frustrated. It should not be this difficult. Thanks in advance for the info you guys rock.

Ryan Baker
05-10-2010, 9:22 PM
You do exactly that ... cut a bunch of segments and test fit them. You can use something like MDF to avoid wasting a bunch of good material.

It's also useful to make a number of fixed sleds/jigs for cutting the segments at different angles. Once you get the angles set right, you never move them again so you don't have to do so much testing later.

Fred Belknap
05-10-2010, 9:24 PM
I may be wrong but I think if you divide the number of pieces you want to use in to 360 and divide that by 2 should give the angle to cut.

Malcolm Tibbetts
05-10-2010, 9:40 PM
Steve, as Ryan said, to really "dial in" the perfect angle, it requires the cutting of test pieces. If your largest segments are 2" wide (inside to outside), rip a piece of MDF of cheap wood 2" to 3” wide, and then cut 6 segments. Place them perfectly aligned against a straight edge (like a jointer fence) and see if they form a perfect half-ring. Adjust your miter angle until you can form a perfect half-ring. Once achieved, you can assembly a dry fit of 12 segments and further confirm the accuracy. By using wide material, any error will show itself more readily. Dialing in that perfect angle can be a hassle, but as you've figured out, you don't want to try stacking oval rings.

Richard Madison
05-10-2010, 9:43 PM
Steve, When you match the two halves, if the gap is on the I.D. the angle is too much. If the gap is at the O.D., not enough angle. Even when your cut angle is "perfect", some half rings will require light sanding to match perfectly. Consider a 16 segment ring in which each cut is "off" by only 0.1 degrees. That's a total of 3.2 degrees error and you have a gap. One workaround is to mark the joints where the half rings were glued together, giving you slightly oval rings. When building your piece keep these joints together (obviously offsetting joints half a segment left and right). This will help maintain correct joint alignment all the way around the piece. Hope that made sense. Many words to attempt to describe very simple technique.

Edit: Beat me to it Malcolm. But don't you sometimes do half rings and lightly sand? Thought I saw that on "Seggy".

Jim Underwood
05-10-2010, 9:58 PM
Trial and Error (and then adjustment) seems to be the way to ensure the perfect cut. Don Russell showed us this one night with one additional tip. He was using a miter saw, and after getting the angle close, used a rubber mallet to "adjust" the saw, because loosening the lock let the angle change too much.

Quite fiddly, this segmentation stuff, yes?

Bill Bulloch
05-10-2010, 10:13 PM
The hardest thing for me was getting the angles right. I used a four inch wide piece of MDF and spent hours cutting and adjusting until I found that perfect setting to give me a nice tight ring. I now use a Incra 1000 Miter Gauge on the Incra Miter Express Sled and I use it for all the angles, so every time I changed it I had to start over with the test pieces to find that perfect setting again. Finally, I took a picture of each of my angle setting, now I can duplicate the cut each time with out a problem.

Here is the picture I use to set the 15 degree angle for my table saw sled using the Incra 1000.

Malcolm Tibbetts
05-10-2010, 11:46 PM
Bill, I like your "take a picture" trick.

Richard, even though I strive for perfect whole ring dry fits, I still always glue them in half-rings first.

Another tip for those that sand their segment ends before gluing. Don't sand the four ends that create the gap during the half-ring gluing. You'll always have to lightly touch up those ends after the halves are glued and if you sand them first, then you end up sanding them twice.

Richard Madison
05-10-2010, 11:52 PM
Thanks Malcolm.

Steve Bearden
05-11-2010, 8:42 PM
I really appreciate all the help with this. I found 1 problem today is that my digital protractor is not right by anywhere between .1 to .3 of a degree. This has been killing me. I brought home a mechanical starret gage today to test. It was alot closer but still not what i would consider perfect. I had no idea that you had to tweak the angles a little to get it right. I figured that math dont lie 15 degrees times 12 segments should be perfect. OH NO it aint. Now that i know you guys tweak it a little i can get this crazy thing beat. This is probably the most i have used my damn brain in a long time. To all you guys that do such great segmenting i give you all the praise in the world. Trust me folks this is not just cut up some scraps and glue it together. I will get there and you guys will be the first to know. again thanks for all the help.:D

Kurt Barker
05-11-2010, 9:22 PM
I am such a newbie here that I'm reluctant to chime in, but the possibility of helping another trumps the possibility of embarrassing myself.

I elected to skip the protractor completely and use a pair of inexpensive drafting triangles. The triangles come in 2 basic shapes: 45/45/90 degrees and 30/60/90 degrees. I've always found it easy and quick to build table saw sleds with a fence that is precisely 90 degrees from the blade. By laying the 45 degree triangle and the 30 degree triangle against the fence, I get 75 degrees, which is conveniently 15 degrees from the blade. I attached (screwed) a temporary fence on my sled base by using the triangles to determine the position.

When I finished cutting the 12 pieces, the collective error in the cuts was about 1/16th of an inch. I suspect that that's about as close as one can get, but I'd love to hear other's thoughts on this.

Since the small error I had in the cuts caused a gap on the inside of the joint, I was able to glue all but two opposing joints and clamp the ring with a large hose clamp, then use a bar clamp across the unglued direction of the ring to 'collect' the error into the two unglued joints thereby making the glued joints tight.

I then used a pencil mark and disk sander to get the two halves 'perfect'. (Several of these techniques are not original thoughts - I learned them from Kevin Neelly at turned wood dot com.)

Robert McGowen
05-11-2010, 9:54 PM
I figured that math dont lie 15 degrees times 12 segments should be perfect. OH NO it aint.

Actually it is. The tweaking is to get to 15 degrees exactly. :)

Ralph Lindberg
05-12-2010, 9:35 AM
... Trust me folks this is not just cut up some scraps and glue it together. I will get there and you guys will be the first to know. again thanks for all the help.:D

Steve, if you don't have Malcom's book get it. It will save your hours of learning on your own

I have yet to buy his DVD's, but I suspect they are as good as his book

David Warkentin
05-12-2010, 9:52 AM
I agree. Get Malcolm''s book. I have it and Ron Hampton's and Ray Key's and Malcolm's work is very good. It will help you alot. David

Jim Underwood
05-12-2010, 10:06 AM
When I finished cutting the 12 pieces, the collective error in the cuts was about 1/16th of an inch. I suspect that that's about as close as one can get, but I'd love to hear other's thoughts on this.


You should be able to get it dead on, with no gaps. Certainly not any gaps as much as 1/16th. It should be so close that you can just use a strip of tape and roll up the pieces in it to check the fit of a quarter of the pieces against a right angle surface, then check the fit of half the pieces against a flat surface, and finally check the fit of all of them together.

In order to get this kind of accuracy you have to use a good sharp blade, and adjust the angle ever so minutely - hence the rubber mallet tap tappity tapping...

This is one reason I've not tried it yet. It's very fiddly...