Official Luthiers Forum!

Do you know the circumstances? Did they warp just due to string tension at normal room temperatures, or were they left in a car on a summer day?

A CF support rod is definitely not loaded to any noticeable extent in shear. Depending on it's position in the neck, the loading will be primarily in the form of tension and compression. The shear loads are calculated using the longitudinal cross section, which is much larger than the lateral cross section...

The PRS acoustics also don't have adjustable truss rods. As to whether they will creep or not...I don't know....ask me in 10 to 20 years.

Also, keep in mind that all CF is not the same. You can change properties a good bit by changing the ratio of the volume of the matrix (i.e. the epoxy) to the fiber volume as well as using any number of different fiber and matrix materials that would all fall under the category of "CF".

I personally like how Collings sets their necks up - they use steel reinforcement bars in addition to an adjustable truss rod. I don't think I would ever build one without an adjustable truss rod. I think you could probably get away with building a bunch and having most of them work out OK over the long haul...but I would have to believe you'd run into trouble one way or another on them eventually.

Trev

_________________
http://www.PeakeGuitars.com

Chris, as Al pointed out carbon fiber resists tension or compression much better than it resist a shear load. When trying to bend any beam (in our case a neck) the outside of the beam is placed intension and the inside of the bend on the beam has a compressive load. For a neck, the string tension is placing a compressive load on the fingerboard and there is tension on the back of the neck.

So some are saying for the carbon fiber to add teh most stiffness to the neck, it should be placed where there is the maximum tension/compression load.......which means it would either be place near the surface of the fretboard (maximum compression) or near the back of the neck (mazimum tension).

For those that considering placing the CF at the back of the neck to resist tension, two methods are being debated. Here are the two ways of installing the CF deep in the back of the neck:



Either would work fairly well. I would have to have the exact size CF and location of teh CF to say which would be stronger.....and you would need to calculate the moment of inertia to determine which resists bending more. I personally like the idea of placing the CF parallel to the fingerboard (the top drawing).

_________________
Formerly known as Adaboy.......

I'm pretty sure the creep in wood happens in shear.

Wood, after all, is a material designed (so to speak) for the loads the tree puts on it. If it tended to creep in compression then the trunk of the tree would tend to get shorter and shorter over time, and it doesn't do that (except, maybe, in really big trees, where the compression load exceeds the strength of the wood). Wood in a tree trunk is normally only in tension on the upwind side. It turns out that wood is stronger in tension than in compression, but that has more to do with the nature of the cell structure, I think: wet wood fails on the compression side as the thin cell walls buckle. At any rate, a tree with a straight trunk that does not 'see' a strong prevailing wind doesn't have to do anything special to remain standing upright and tall.

Branches, stuck out the side, are a different story. Without some special adaptations, they would droop over time. The adaptation the tree uses is 'reaction wood'; cells with a lot of built-in stress that form interlocked grain. That's why you don't use branch wood for building: it tends to move around every time you cut it due to the change in stress, and it's impossible to get a good finish on it. It's interesting to note that in softwoods the reaction wood is on the top of the branch, and on hardwoods it's on the bottom.

You can easilly check this out. Just clamp a piece of wood to the bench top so that it overhangs the edge and leave it for a while. Depending on the size of the wood it might take a few months, but you'll find when you unclamp it that it has taken a set. More load will speed things up, of course. My wooden go-bars take a noticable set in only an hour or two.

This, by itself, does not prove that the creep happens in shear: to do that you'd need to show that the upper and lower surfaces have not changed in length. Since the actual difference will be small for any normal curvature, this might be tough to do. Still, my point is that you don't see creep unless there is some shearing stress involved, so I suspect that's what's heppening.

Alan, it seem to me that what you are describing is a situation of bending rather than shear.

I am not quite sure why the concept of shear has been introduced in this thread.

To me the main point to consider is where to place the CF reinforcement so that it acts in tension or compression as part of a wood and CF composite beam rather than placing it at the neutral axis where it only adds it's own bending strength to the unit.

It's not a common problem on the McPherson. I've only seen it a couple times, but I'm just saying carbon fiber is not the bulletproof answer. McPherson also uses African Mahogany (the not mahogany, mahogany) on all the necks.

_________________
John Mayes
http://www.mayesluthier.com

I think you can see an example of creep in tonewoods if you check out a top that has "pulled up". Most of the loads in a top are bending loads and therefore manifest themselves as tensile and compressive stresses in the top.

Also, the term shear by itself does not denote any particular grain orientation. You can have shear going parallel to the grain or perpindicular to the grain...likewise with tensile and compressive stresses. Creep should be evident both across the grain and with the grain.....in tensile stress, compressive stress, and shear stress. You'll probably have different values for all the different load cases, but creep will be there in some amount.

As to where to put the CF to get the most bang for the buck, my calcs indicate that the neutral axis of a neck is pretty close to the fingerboard. As such, you get more benefit by placing the reinforcement lower in the neck, closer to the carve surface.

Sorry, I'll try not to nerd out too much in the future...

_________________
http://www.PeakeGuitars.com

Understood. But if it failed because of abuse, i.e. being left in a hot car, that's a different story than if it warped over time without heat.

If it bent because of heat, it's something that can be avoided with proper care, plus it can probably be heat-bent back. And it just means that anyone who doesn't want to worry about protecting their guitar from heat should make sure it's got an adjustable metal truss rod.

But if CF is prone to cold creep, then it means that in the long term it's not a good choice for neck reinforcement for anyone.

That's not necessarily true...if it only crept .05% of the initial deflection, then that's nothing.

I don't believe there are standards for CF such as there are for metals. I know when I buy 4140 Alloy steel what I'm getting...I can look up all the material properties I'm interested in...but with CF it is not that exact of a science. They basically bake this stuff as part of the process of making it and from the folks I've talked to everyone kind of does things a bit differently...and therefore product A from one vendor may not equal product A from another vendor.

I picked up a smattering of knowledge on this stuff working on a project to design composite power transmission components for high speed drives like turbines. I got the chance to work with some experts and it was really cool stuff to learn about..!

_________________
http://www.PeakeGuitars.com

The owners of the guitars, to my knowedge, never admitted neglect. Doesn't mean it didn't happen but I also was told by an employee there that he had seen the necks warp before despite the huge CF rod. No one that I knew there, beside matt mcpherson, liked the design. It had some possible advantages but most felt like it was a drawback in the grand scheme. But I suppose that's not relavant for this discussion.

So I don't belive it was a neglect issue.

_________________
John Mayes
http://www.mayesluthier.com

The problem that we all face is the CUSTOMER. He has preconcieved notions and traditions that outweigh all our research and practice.

I love CF in 12 fret necks, and build that way if left undisturbed, until a customer says "it has to have an adjustable rod" then it's either his way or the highway.

If a customer wants a unadjustable 14 fret neck, I'd better load that thing with CF, against my own better judgement.

Carbon fiber is a great material, I'm glad we have it available.

_________________
https://www.facebook.com/profile.php?id=100008907949110

For the record, I offer solid necks as an option on guitars, std being adjustable; mandolins are all solid. All my personal instruments are solid, though.

CF is the wrong choice for solid neck reinforcement, no matter where you put it or how much you use.

I thought this was going to be a good thread to read, based on the title, but y'all moved away on this other tangent. Again. Shame, really.

Have fun; argue amongst yourselves.

grumpy, would you care to share how you build a stable neck with no truss rod?

BTW, I've heard so many experienced members here mention how well built, how easy to play, and how great your guitars sound. You are a real asset to anyone learning or refining their craft. I wish I could hear from you much more on various lutherie topics.

_________________
Formerly known as Adaboy.......

I think Alan is saying that, unlike the normal bending that occurs when the load is first introduced (like the image on the left, with compressive fibers compressed and/or tensile fibers elongated), cold creep makes wood bend like the image on the right, with no fibers changing length.

Usually shear failure means one layer of fibers separating radically from an adjacent layer so the material splits, but I guess the thing he's describing would be a type of shear failure also. The compressive fibers are not being compressed and the tensile fibers are not being elongated, but rather each layer of fibers is shifting position slightly with respect to the adjacent layers of fibers above and below, i.e. they're sliding rather than splitting. Very interesting.

But even if that's true, putting the CF at the bottom of the neck still won't prevent cold creep. A CF strip down there doesn't connect one layer of wood fibers to another, so it won't prevent them from sliding against each other under load. It'll only reduce the portion of the load they carry, reducing or retarding the slide a little.

On the other hand, assuming CF is in fact much more resistant to cold creep than wood is, then a bar of it on edge anywhere in the neck will start to take all the bending load as the wood starts to fail due to creep. If the CF bar is strong enough, there'll be no bending.

It's true, as someone suggested, that an I-beam shape of CF would be a much more efficient use of the material, but the top and bottom flanges of the I-beam have to be connected to its web. Using the wood as the web of the I-beam will work against initial loading, but it won't prevent long-term bending if the wood is subject to creep, regardless of whether the creep is a shear, compressive or tensile failure.

So John's report of a solid CF neck warping is troubling. It suggests that CF is no more resistant to cold creep than wood is.

To the point of the original topic, of those of you who have built with a solid neck (no truss rod or truss rod slot), did you note an increase in volume or sustain? I ask wondering if John's increased volume was due to a solid neck.

_________________
Formerly known as Adaboy.......

Thanks qrwtz.

Any time you bend something there are shearing forces introduced in it. The fibers on one side get compressed, and the ones on the other are in tension, and both sides resist changing length. The piece 'wants' to go to the sort of shape qwntz shows on the right side, and it's only the shear modulus of the material that keeps it from doing that. In the end, which of his two drawings the part most resembles will depend on the relationship between the shear modulus and the young's modulus of the material, and, over the long term, whether there is any creep involved.

The structural part of wood is composed about 50% of cellulose, with the remainder being about equal parts of lignin (glue) and hemicellulose (filler). Of the three the lignin is the stuff that is thermoplastic. When you bend wood, you're heating it up to the softening point of the lignin, so that when you bend it the fibers will slide past each other in shear, Then, when it cools off, the lignin hardend back up and holds the new shape.

Like a lot of thermoplastics, lignin will cold creep. That is, there is no lower limit on the force which will cause it to bend over time. In theory, if you were patient enough, you could bend your sides cold. Heat just speeds up the process.

Carbon fiber itself does not cold creep. If there's a problem along those lines, it's in the epoxy (or whatever) matrix that the fibers are in. My limited understanding of this sort of thing is that the heat cured epoxies are much more creep resistant than room-temperature cure ones. They do use heat cured composite components in high performance aircraft, in very highly stress parts, so there must be matrices that will do the job.

So, Mario, why is CF a bad reinforcing material for necks? And why don't you steer the thread in a more interesting direction, or start a better one?

Does anyone manufacture carbon fiber reinforced bars with phenol-formaldehyde as the matix? That might make a bar that's highly resistant both to cold creep and to heat bending. You could even leave it in the car.

Of course, there'd still be epoxy joints around the bar, bonding it to the wood, and those joints could slide. But that would be true even with Grumpy's metal tubes. I've used solid steel and aluminum bars embedded in the neck with epoxy, and they've been stable, about 30 years for one and 20 for the other.

John (kinnaird) - What kind of truss rod was it? Did the CF that was installed in place of the removed rod pass thru the neck block too?...Was the CF glued into place in the neck block too?

Guys I am the last one to know something about this but my mind is reeling with questions about the concept.

_________________
There is no difference between the man that thinks he can....and the man that thinks he cannot.

Chris
I had installed an LMI double action truss rod. The channel was 1/2 inch deep which accomodated the rod and a wood spline covering it. My necks bolt on so the rod did not extend past the heel tenon. On the peghead end, I normally let the cf rods on either side of the truss rod cavity run into the peghead and then cover their exposed ends with the overlay. However there was a closure to the tr pocket that prevented the replacement cf from going past the nut.
In my opinion this neck is overengineered. I am confident that it will not creep and fail over the long run. As for twisting I can't be sure, but then an adjustable truss rod would not help correct a twist anyway

I'm convinced that neck construction, stiffness, weight, ect can, and will effect the tone, timbre, and.....personality of a guitar. For that reason I prefer to keep my necks consistant. I feel that I have more control of the finished product by leaving the neck alone, and (trying to) control the voice with the "box". I had one customer who didn't want an adjustable truss rod. I told him since he knew so much about it maybe he should build his own guitar. He took delivery of his guitar (with an adjustable truss) for last Fall and loves it. I'm not saying that adjustable truss rods are good, necessary or anything. It's just the way I do it.

_________________
I didn't mean to say it, but I meant what I said.
http://www.brackettinstruments.com/

I had an experience with a '79 D-28 that makes me know that I would never want a guitar without an adjustable truss rod..

I was living in Santa Cruz (rather high humidity all of the time) and the neck had actually back bowed which caused buzzing behind where the strings were being fretted...very annoying...had the guitar had an adjustable truss rod (preferably double action, though I don't think those were around then) it would have been easily 'fixable'....

from an engineering standpoint I can see where anything hollow in the neck would cause a loss of energy (and a truss rod can be considered semi-hollow, neh?) but from my viewpoint said loss of energy is a necessary evil where the means justify the end...a nice tight fit and proper use of epoxy should limit the loss I would think....

John, I find your observation very interesting. I think, though, that what some others are suggesting is that your experiment doesn't necessarily lead to any conclusions that can be applied to other guitars. On THIS guitar, the added CF had that result, but that could be due to some sort of different coupling/interaction of resonant frequencies that happened on this particular guitar, given its unique formulation of resonances. On another guitar (or even on this same one, for that matter), one might achieve a similar improvement, or a different kind of change perceived as improvement, in the sound by making the neck slightly more flexible rather than stiffer, depending on how that change interacted with all the ways that particular instrument vibrates. Do you follow me? Anyway, that's my interpretation of why some are questioning the conclusiveness of your result. Still, I'd venture to say that the vast majority of us appreciate your sharing your experience, and hope that you continue to do so as you experiment further with these ideas.

_________________
Todd Rose
Ithaca, NY

https://www.dreamingrosesecobnb.com/todds-art-music

https://www.facebook.com/ToddRoseGuitars/

The conversation could be way over-engineered…
I mean, it's just a neck…

To corroborate John's findings, a couple of years ago I exchanged necks on two guitars and found that different necks alter the tone, and not subtly.
A heavier material tends to increase sustain, and CF rods change the tone besides stiffening the neck.
I can't really express it, but I feel it adds a bit of "airiness", or very high partials. Some do not like it, Dana Bourgeois told me he does not like the tone of a neck with CF.
Maybe 2 guitars are not enough, but I could bet anything that any guitar will benefit from a stiffer, heavier neck for those who are looking for sustain, attack and clarity. Again, this may not be seen as a desirable quality for a particular guitar. I sometimes like to use the lightest mahogany, or Spanish cedar, I can find on some guitars.

Another way to really stiffen a neck is to use compression fretting, or even better, bar frets. Then, just an ebony rod may suffice.

As an experiment I built a neck 2 years ago with the hardest mahogany I had on hand, just two 1/8" x 3/8" CF rods, no truss-rod, T-bar or square tube. .250" thick ebony fretboard.
It failed, the guitar sounds nice, but can only handle extra-light strings before the neck starts to bow. I'll have to take it apart when I have time…
BTW using .080" or .090" CF as binding material on the fretboard sounds like it could add a lot of stiffness where it's needed, especially at the neck/body junction.

_________________
Laurent Brondel
West Paris, Maine - USA
http://www.laurentbrondel.com/

I hate to sound too much like an engineer, but...

It would seem that the primary change that would result from adding Carbon fiber in a truss rod slot would be in increase in stiffness of the neck. The most apparent result of this increase in stiffness would be a significant increase in resonant frequencies of at the first few modes.

i.e. tap it and it will ring with less amplitude and at a higher frequency.

-jd

Well, if you really want to sound less like an engineer, just mention a bunch of fat people eating ice cream. Worked for me!

Dennis

_________________
Dennis Leahy
Duluth, MN, USA
7th Sense Multimedia