Gluing for laminating wood for turning?

Sorry for the confusing title. Should have read:
[h=2]Gluing procedure for laminating wood for turning?[/h]Posted too early in the morning!

I have a project coming up that is best described as a tool handle-- like a wood lathe tool handle--only a bit larger in diameter. It has a re-enforcing rod running most of the lenght at the center of the handle. Each half has a channel routed out for the re-enforcing rod, then the re-enforcing rod and two halfs are glued together, then turned on a lathe. I built one of these about thirty years ago out of white oak and it held up great, but the glued up laminates were noticable in the feel/handling. Here is a end view drawing of the laminating before turning to help explain what I am trying to describe.


Each of the four sections are 1" x 1" by whatever lenght I'll build it this time. I have a planer so the wood will be flat and square before gluing. The question is: Is there a best way to orient the pieces for best lay-up? Would hickory or white ash be a strong alternative the white oak that I used previously? As I recall the oak was somewhat time consuming to turn.

Last time I used wood glue to laminate and epoxy to glue in the re-enforcing rod. This time I thought I'd use epoxy throughout. Good idea?

Thanks in advance for any advice given.

Using four pieces is part of the design?If not how bout just one seam to deal with by using 2 pieces 1"x2" and route the channel in both halves.Two pieces or four pieces how bout the System Three for the glue Buy System Three products at Woodcraft
I have never use it but know others that have and very happy with it,another choice would be Unibond 800 which I have use it with great results.Comes in a lite color shade like maple or in a dark color,lots of drying time
http://www.highlandwoodworking.com/unibond800laminatinggluewithmediumcatalystgallon.aspx-----kroll

LA:

I'd rotate each piece in the glue-up 180 degrees so the grain is oriented like it was in the tree. That will tend to minimize differential shrinkage forces on the glue joints. It should also turn more smoothly, with less tear out. Either Ash or Hickory would work well and may be easier to glue. Oak can be difficult to glue sometimes.

Regards,

DB

Thanks, guys, for replying. As advised, I think I'll go with just two pieces instead of four.

Cheers!

I disagree with DB's reasoning. You are better off having the rings the way you have drawn- opposite the way they would be in a tree. The reason is that boards tend to cup against the rings, so if you glue them like a tree they are pulling away at the edges which is more likely to smile. Gluing them the opposite way- heartwood out, makes them try to pull apart in the middle which is less likely to fail.

Although, I have no experience with his turnability argument, so that could be true and is perhaps the more important factor.

Pete

if the wood is dry, movement will be held by epoxy
original 4 pc orientation may look nice if grains are same size
if glued as 4 pcs,
glue 2 pcs together
plane,route groove for metal support
gluing 2 half assemblies and metal support as 3 pieces will be easier
Mike

I agree with DB.
Positioning it that way (as grown in tree) best puts the rings for the least expansion and contraction in thickness "all the way around". So less issue with failure at the joints. IMO

Looks like someone needs to do an experiment.
Or has the FPL already done so? (my 'net service is too slow to search)

smt

Stephen:

Reference this publication:

Wood as an adherend

At 240 pages it's a bit of a slog, but the best info on gluing wood I've found. They don't expressly discuss the grain orientation problem, however the material on the differences in shrinkage in the radial and tangential directions makes it pretty clear.

Regards,

DB

DB-

It's not clear whether the publication will successfully download on my slow 'net service.

But I can say I've been reading and thinking about the subject for over 40 years. Per the fundamental radial vs tangential shrinkage diagrams that everyone "gets" in wood 101, the squares will all shrink to diamonds, (long points radial) or expand to diamonds (long points circumferential) whether they are oriented as drawn, or as you and I propose. But the way we propose offers less shrinkage in thickness any way across & it is uniform. Also, contrary to that crazy guy's statement; more, or perhaps "heartier" heartwood is contained in the as grown orientation.

To my view, there is less differential stress along the glue joint with changes in MC. I *think* I see more of a roll-and-peel effect going on with the inverted (opposite of as-grown) drawing. But since any effect described is somewhat mirrored by both orientations, I have to be open to the possibility someone did test and and maybe the results are inconclusive or even (horrors!) my lifetime of experience fooled me on this one.

I am fairly convinced that if expansion is the major problem, the as-grown orientation would be better. But Pete is considering shrinkage as the major issue. In that case, I am less confident.
Finally, I am pretty sure that the as-grown orientation edge joints are better glue joints that the opposed reverse 45° - 90° grain orientation in the drawing.

So I say it still needs an experiment. And it has to be 4 pieces.

With pool cue shaft wood, some are arranging 6 to 8 pie slices with the grain all radial. I don't think it has been proven to be superior, given the vagaries of natural wood anyway. Even microlam shafts (laid up flat from thin veneers) are sometimes better, sometimes worse; and no doubt frustratingly both continue to warp. Perhaps a diversion from the glue question, though since in those small sections it is seldom an issue.

smt

I tend to think that the fact that skilled craftsmen on this forum haven't reached a consensus re: grain orientation might suggest that there isn't an obvious best way.

Re: which wood to use... I favor hickory where shock resistance is critical (hammer handle) or ash if bending strength is critical (peavey handle; straight grain is important!). Ash is typically more stable and should machine better. Hickory is heavier.

Either way, I second (third?) the "an experiment is in order" proposal and look forward to what you learn!

-Jeff

If I faced this problem, I would do as Steve describes for pool cues. 8 pie sections with radial grain. I would also use a stiff epoxy. The final result would change diameter slightly with moisture, but would be about as stable as a wood part can be. I doubt that there would be problems with feeling the glue joints either.

Here's my 2 cents - Either of the orientations are symmetrical, which should reduce or eliminate differential movement. And the parts are small enuf that movement should not be dramatic enuf to overcome a good glue joint.

I think most problems with glue joints have much more to do with preparation of surfaces and moisture content than grain orientation. I like a well hand-planed face, with a tiny bit of hollow. And the wood of course at equilibrium with its intended home.

I prefer titebond to epoxy for wood joints, tests I've done have convinced me it makes better joints on well prepared wood. Epoxy has much better gap filling qualities, moisture resistance, and way longer open time. If I need a really rigid glue line I'd use urea resin glue or hot hide glue. Most of us think of hot glue as obsolete, but it's still well worth using

Last, if you really want the glue lines to disappear, just take a 2" square, rip it on the bandsaw, hand plane, rout for the metal, and reassemble it. Or better yet, bore it from the end and eliminate the glue line entirely. If you want to use white oak, try to find some old, slow growth stuff with very close rings, it's much more stable, and entirely different animal from coarser fast growth material.