>My friend John Bartlet does quite a bit of segmented turning. He almost always segments the bottom of the bowls he turns too. Recently he put bowls in a shop. 5 of 18 had the bottoms crack. Note that this was in Wisconsin in the winter, where the humidity indoors is drier than Arizona. The bottoms segments are cut with the grain running across the segment, so the bottom should expand radially. They all crack across the center of the bottom with the bottom edges still looking good. He uses titebond glue. (not titebond II) I think that the wood shrinks radially and since the out sides are glued to the bowl sides the wood shrinks away from the center. With endgrain glueing the crack follows the glueline. Is this anywhere close to correct?? Can you suggest a way to prevent it from happening?? TIA
I think you answered the question yourself "bottom should expand radially". Given the orientation of the wood, the radius of the piece will try to shrink as the wood dries out. Only one of two things can happen:
- the diameter will get smaller - i.e. the wood shrinks towards the center. As each segment is in essence a wedge with the grain running across the wedge (i.e. the wedge isn't going to get much thinner) the outside is not capable of moving towards the center.
- the diameter stays roughly the same and the wood pulls away from the center towards the rim.
The latter is what happened in the three pics you showed.
As to how to prevent it! The easy answer is, don't do that. ;-) I don't have much experience with segmented turnings and am always amazed that so many of the segmented pieces don't suffer more from wood movement. I bought a maple/ebony segmented bowl a number of years ago and some small splits open up each winter.
>Bear in mind this is coming from someone who has absolutely no experience in segmented turning. It appears, from the photo, that while the sides of the bowl are segmented, the bottom is not. Therein lies your answer. The sides of the bowl move radially, while the bottom does not.
You are dead on as to what the problem is. As for the cure I'm not sure. I had to stop building segmented work because my shop was so moist I had too much wood movement.
There are 2 things that worked for me but I only made a few pieces after this so I don't know how effective it is. One method was to make my own plywood bottom. I took multiple layers of veneer and stacked them like plywood with the layers alternating in the way the grain ran.
The ohter method was to make my bowls different shaped so the bottom was simply smaller. thus less wood movement.
I also know from some experiments that cone shaped bottoms don't crack. I've been experimenting with leaving the pith in but making the pieces very cone shaped. The wood shrinks but the outer portion is free to move so they don't crack. I've also tried this with one segmented piece and it was still together after 1 year when I gave it away.
You have stepped into one to the great debates that rages (?) within the segmented turning community - Segmented Bottom or Solid Bottom.
There are good/great segmented turners in either school. I know segmenter extraordinare Malcolm Tibbetts is in the segmented bottom school. Some of the real masters of segmenting like Bud Latvin, Curt Theobold and the late great Ray Allen are in the solid bottom school.
Not that I compare myself with any of the above, but I am in the solid bottom school. I turn here in the humid deep South, and have sent many segmented bowls to family members in dry Phoenix, Arizona and to date, none have cracked at all. They were finished in all different manor - Oil, Lacquer, etc, so that is not a consideration.
I have had bowls crack, but only from breaking a cardinal rule (before I knew it): Gluing side grain to end grain (at 90 deg angle) to each other in design rings, etc. Now, all grain then the bowl runs round the bowl and I have not had problems (knock on wood).
I should mention that most, but not all, of my bowls have 1-3 veneer layers between the solid bottom and the segmented layers (for looks). Like John Lucas mentioned, this may work as a plywood.
>Greg - I have a little experience with segmented turning, so here's my $0.02 worth.
One way to avoid this, which I sometimes do, is to glue the base as your friend does (although I never use that many segments, usually 6, 8, or 12) and then turn a recess in one side of it. The bottom of the recess has to be dead flat and 1/4 inch deep. Then cut or turn a 1/4 inch thick plywood disc (good quality plywood) to fit EXACTLY in the recess and glue that in with high quality slow curing epoxy. I use System Three, available from Woodcraft. The plywood is stable and the epoxy is strong. If I want the wedge fit of the segments -the place where the points come together - to be prominent in a vase, I put the plywood disc side on the inside so the wedge fit is on the outside. In a bowl, I would probably put the plywood disc side on the "outside", or the bottom, and have the wedge fit be prominent in the inside bottom of the bowl.
But the plywood disc is not essential. Good preparation IS essential - my shop is typically at about 40 - 50 % relative humidity, but the wood isn't stored there, it's in the garage. So before I put anything together I (a) let the wood sit in the shop for a week or so, and (b) I measure the moisture content with a pin-type moisture meter. I don't start gluing anything until the MC of the wood is the same as some reference samples that stay in the shop. ONce the piece is done and finish applied, I put a good coat of wax on it and caution the customer to avoid exposing it to severe humidity fluctuations. I've made a couple of hundred segmented pieces this way. Yeah, I've had glue joints come apart, but not for awhile.
Some people seem to think it's a law of physics that if there is a force on something(such as stresses induced by moisture content fluctuations in wood), then there has to be movement. This is not true. Think about the overhead lights hanging in your shop - are they exerting a force on the ceiling? Yes. Will the ceiling eventually move and come down on your head because of that force? No. So the problem is not how to avoid the forces so much as it is how to constrain them. Along those lines in this case, the quality of the glue joint is VERY important. How are the segments cut to fit? Are they sanded? If so, with what grit and at what speed, i.e., are the surfaces glazed? IF they are, the joint will probably fail since it's endgrain gluing. Before they're glued, is there COMPLETE surface contact between every pair of segments? Clamp the ring without glue and hold it up to the light. IF you can see any light between any part of any two segments -it's not good enough. An old patternmaker I know, who has probably made upwards of a few thousand things like this in his 80-some years, recommends a light coat of glue on each surface, let it dry for a few minutses (tack, really), then apply more glue and clamp. I have looked closely at pieces he made twenty years ago and none of the joints are coming apart. He uses regular Titebond. And speaking of that, make sure the glue isn't out of date. Titebond puts a date code on the bottle and they'll be happy to tell you how to decipher it if you call their 800 number. I never use it if it's over a year old, at least not for stuff like this.
Segmented turners have lots of different ways of doing things. No one method is "right", what's right is what works for you. There's a little bit about my assembly process on my website. Take a look and let me know if you have any questions. I'll help if I can.
