Official Luthiers Forum!

I make my own Bloodwood Bindings. I glue up the purfs before I cut the bindings, so I don't have the hassle of gluing up the purfs to the cut bindings. I also bend on a pipe.

The problem I have is that Bloodwood is so hard to bend - requires a lot of heat, that the purfs in the waist end up either delaminating or crinkling due to the curvature.

How in the heck do you keep that from happening? I tried to take pictures, but my cheapo camera doesn't zoom in that well, but I'm sure you all have had similar experiences. I really don't want to bend just the bloodwood then glue on the purfs, but that may be my only option for bloodwood.

Steve

You can usually fix the wrinkles with a spritz of water and a clothes iron.

I have not made binding before, but I have worked with making many laminated things. If you make your own bindings, why not glue them up on a rough radius? It does not have to be perfect at all, but if you glue it up so it's say 50 to 80% of where it needs to be in the final state, there should be no delaminating or crinkling in bending, since you will not be pushing dimensional changes of inner and outer lengths to extremes.

For a template simply bend a piece of aluminium flat bar to the desired rough form.

One other technical observation: Traditional bending irons actually put the heat in the wrong place for bending. As we know, wood is much weaker in compression. Defects from bending are almost always due to wood failure on the compressed side. This is exacerbated by weakening the compressed side with greater heat. You might want to figure out a way to get the heat on the tension side. This leaves the compression side cooler, and thus more able to resist failure, while putting the heat on the tension side allows the requisite dimensional change to be mostly relaxation and stretching of the tension side. Infrared heaters work well, since they deliver heat of a frequency that does not have a propensity to scorch. Tempco has some nice elements.

Interesting.

I don't think I would have near as much problems with a bender than with a pipe, and I think you just explained why. Maybe it's time to invest in a heating blanket.

Steve

Sure. I think any extra heat first delivered to the tension side would help in any bending process. With a bending iron alone, that outer face has no time to accept heat, and the full strength of the tension side works to create compression failure.

I don't know this. And my bending failures are almost 100% on the side in tension.

As for bending bloodwood bindings, the key thing is to have close to zero runout. They don't bend easily, but they do bend OK. However they will split easily along a grain line.

Also, I don't laminate purflings before bending bindings, so no comment on the delamination issue.

_________________
Howard Klepper
http://www.klepperguitars.com

When all else fails, clean the shop.

It's not merely my observation. Every wood science and technology book I have read (at least three well regarded texts) state wood is much stronger in tension, and bending failure is almost always in compression. A bending iron applies heat to the face already most apt to fail, while leaving the tension side cooler, and comparatively brittle.

I think there is an explanation for why it might often appear otherwise.

In practice, when wood fails, it necessarily fails in both tension and compression (at least in the thicknesses of things like ribs, purfling, etc.) but it's the primary cell collapse in compression that facilitates the failure on the tension side. It also seems that the compression failure may be rendered inobvious (compared to the tension failure) because the solid bending iron wall -with it's wood softening heat and less give than a backing strap- better constrains the failure. The compression failure may be uniform and thus not as visible on the macroscopic level, since there may be no obvious signs, like protruding splinters and separation seen on the tension side. This is my ancillary theory anyway.

Hmmmmm.... all the failures I've had in bending wood have been on the tension side while the compressed side remained intact.

Pat

_________________
There are three kinds of people:

Those that make things happen,
those that watch things happen,
and those that wondered what happened.

"If bending involves severe deformation, most of the bending should be forced to take place as compression. Wood softened with moisture and heat or by plasticizing with chemicals can be compressed considerably but stretched very little. In bending therefore, the wood must be compressed lengthwise while restraining it from stretching on the convex side."
--U.S. Forest Products Laboratory

At face value, that paragraph appears to contradict what I said, however, the mechanics of failure in wood bending are certainly complex, and involve many parameters and variables not mentioned in that paragraph (amount of heat, distribution of heat, thickness of material, radius of bend, backing strap or no backing strap, wood type, moisture content, etc.) The idea that most of the bending should be forced to take place as compression is actually not in contradiction with the statement that wood failure from bending occurs primarily in compression, although I am not quite prepared to offer a lengthy explanation right now as to why.

The Wood Manual from the U.S. Forest Product Laboratory was in fact one of the books I read. The others were two titles by Bruce Hoadley, a well known wood expert, McGraw Hill's Textbook of Wood Technology, and yet others. In spite of that one paragraph you posted, I am going to largely stand by my remarks. If I get ambitious enough to track through my books, I will try to cite some of the things that corroborate the core concepts in my post.

Re-reading my original post, there is a glaring error. I said "... putting the heat on the tension side allows the requisite dimensional change to be mostly relaxation and stretching of the tension side."

The deformation in bending is primarily and obviously compression and not stretching, regardless of heat distribution. However, there is a distinction between the stretching quantities of very little, and none. At certain bending extremes, having just a little more room to move may well be the difference between success and failure, and this little bit of extra movement can only be afforded by supplementary heat to the tension side. Mind you, one could also make a further and fair argument that that by adding heat to the tension side, one is actually putting more heat toward the compression zone closer to the neutral axis, thus improving plasticity through and through.

I'd like to say I knew this perfectly well, but miswrote in haste. One can perform work successfully and almost automatically based on ideas absorbed long ago. It had been a while since I thought about bending in specifics.

Having said that, some quick reading reconfirmed that wood is indeed stronger in tension, as I indicated, but one does have to distinguish between axial, radial, and tangential axes. Of course all bending done by luthiers is axial, where wood is stronger in tension.

Here is a quote from the book The Testing of Engineering Materials: Davis, Trauxel, Hauck - McGraw Hill 1982, that indicates basically what I said earlier

Therefore, the first visible signs of failure may be in the tensile face even though the wood is stronger in tension than in compression.

Reference to those texts, please. The failure strength of the wood beam under a load from cell rupture is not the same thing as it's failure when heat bending. And it's not a reason to apply heat to the side less likely to fail. On the contrary, I'd suggest that the side more likely to fail under a cold load is the one that is more in need of heat to facilitate the cells slipping instead of rupturing.

_________________
Howard Klepper
http://www.klepperguitars.com

When all else fails, clean the shop.

How do you know it is different? Or rather, indicate precisely what the difference is. and I would be similarly grateful for references of any forthcoming answer. Heat bending is indeed bending under load, with heat. And a long plate is nothing more than a lateral collection of connected bars or beams.

Regardless, if I am going to cite something, I need to know exactly what point or points of mine you want me to elaborate or validate by citing. Please be more specific.

Note: you edited more quickly than I could respond to : -) I will need some time to digest your points. But I will quickly say that I cited a source in reference to how failure can be misleading in appearance, just above your post.

Even if you were correct (which I am still ruminating on) there is a big difference between more heat and inordinately more heat. My contention is that conventional bending allows significantly less heat to conduct to the tension side compared to the compression side. In addition, what heat does reach it, is easily conducted away by the backing strip, which being metal, conducts well. I believe in many bending instances, the temperature differential is too great.

Edit: O.K. Howard, I concur. The side more likely to fail under cold load (the compression side) does indeed need more heat.

However, this is hardly proof that the tension side does not benefit from a certain quantity of heat during bending, which may be less than the compression side, but more than that sometimes provided by traditional bending irons and processes.

I enjoy an oblique and tasteful insult as much as anyone, but they cannot be so oblique as to be indecipherable.

I've bent Bloodwood on a pipe, and I'd say that all of the breakage issues I've had have been related to runout, due to the angle of the breaks. However, it repairs like a dream with CA glue. When I bend, I have the best result using little blocks on top of the bend to keep tight contact with the pipe. I usually slide them apart as I sense that the wood is hot enough, and put pressure. This keeps tight pressure on the bend part of the wood. I bend in small increments, take my time, and sometimes it breaks and sometimes not. I generally don't even attempt to bend it at less than 350.

_________________
Waddy

Photobucket Build Album Library

Sound Clips of most of my guitars

And what is the typical thickness of bindings, like those mentioned by the OP, and to which all my remarks were directed?

Steve, I'm sorry but I have no experience to help with your problem. If you want help, please ignore the rest of my post.



OK, I'll bite.

This is just my interpretation.

I interpret it as real world personal experience specific to the issue at hand as opposed to siting academic research (with all of it assumptions and simplifications) that merely resembles the issue.



Kevin Looker

_________________
I'm not a luthier.
I'm just a guy who builds guitars in his basement.
It's better than playing golf.

That helps. Just for clarification, I have done my share of wood bending, not necessarily guitar bindings, but then I made that plain in my first post. In addition, I have experience bending wood in some unconventional ways not likely attempted by many guitar makers. And it is a mistake to assume technical books are written by people with no practical experience in the domain about which they write. Maybe the odd time, but not universally.

Well Todd, that's thinner than I imagined when I posted my response to the OP. I provided some comments I thought would be helpful to him, like ensuring that sufficient heat reaches both faces. I'd write it just slightly differently now, but what can you do.

That wood is stronger in tension, and usually fails in compression, are not remotely my ideas, but are the observations of wood engineers and material scientists. The full theory of wood bending and failure and every possible exception are obviously way beyond anyones scope to elaborate in a forum, even if one did know it all.

Thanks for all the replies guys. I love the exchange and the wit .

I made the bindings out of dimensional Bloodwood with the purflings glued on with TB prior to slicing and thicknessing to .095. The I hit the old trusty BBQ Pipe for the bend. At that thickness, the pipe internal temp needs to reach about 450-500 degrees before the wood "lets go". Otherwise I'm sitting there for an hour waiting for the slow relax. I've not had any Bloodwood bindings break on me, although, I've seen some potential run-out wannabees trying hard.



Well, at that temp, I get the expected delam; and at the waist, I get the unfortunate binding cramp/crinkle where the surface area of the bindings does not take kindly to being squeezed. Maybe, I need to lam AFTER I bend - but that seems like such a pain in the ....


Questions, for those using pipes or even benders - I assume the same applies - reducing the surface area of the laminations seems to me it would still want to resist reduction and its only option is to crinkle/crimp.

Thanks again for the help. I LOVE Bloodwood bindings. Maybe I need to find a wood that relaxes at lower temps.
Steve

I am still confused as to why one must glue up the binding and purfling straight. What exactly would prevent one from making a mold and gluing up in some rough pre-radius that is approximates the waist? I am positive that doing this would mitigate delamination. Perhaps there is some technical reason I am not aware of, but I would like to know precisely what it is. I mean at first glance, it might appear as more steps or labour, but if it prevented delamination...

Just read this on a second read:

Well, at that temp, I get the expected delam; and at the waist, I get the unfortunate binding cramp/crinkle where the surface area of the bindings does not take kindly to being squeezed. Maybe, I need to lam AFTER I bend - but that seems like such a pain in the ....

I have done exactly this, i.e., pre-bend wood laminations (although not for bindings) before gluing, precisely to avoid delamination in later more precise bending. You do not have to be perfect. Any rough bend built in will make tweaking later under heat vastly simpler. Then again, tweaking might not even be necessary in the case of bindings.

The question is, what is more of a pain in the ..., setting up a nice repeatable system (once) that delivers excellent results every time, or gambling every time? I really don't know in your case or anyone else's, but I will go out on a limb again anyway and say the latter, and by no small measure. It's a problem I find interesting.

Actually, de-laming IMO isn't such a big deal, as it all gets re-lamed as you tape the aggregate into the channels. I've quasi bent purflings with a clothes iron on a flat board, but of course they don't like going sideways which is the problem in the first place. They are many times thicker than tall in the direction of the bend.
So I guess to OP, as long as you can get the crinkle flat, either by using an iron or the vice/heat gun method, it doesn't matter if it's de-lamed a bit. It will go back to where you need it.

Now if I didn't know better, I'd say that's another subtle swipe at me. If the "luthiers" were a perfect help, maybe he wouldn't still be having problems with delamination, or "the unfortunate binding cramp/crinkle where the surface area of the bindings does not take kindly to being squeezed" as he just indicated ; -)

I was curious, so I just tried bending a .080" thick piece of Bloodwood on a maxed out Ibex with a surface temperature of 452 F. It bent very easily with a thin backing strap to a 1.375" radius (much tighter than a typical guitar waist) with no remote sign of failure. Neglecting the delamination issue -which would easily be solved with pre-bending before gluing- I am curious about your heater Steve. Maybe it has inadequate mass, and thus gives up heat too readily to complete bends. How is it heated? What is the material? I am not so sure you could trust that gauge, unless you have a wired thermocouple directly under the zone you normally bend over.