carbon fiber instead of truss rod

[first string]

Maybe someday Dudenbostel and Nugget mandolins will use CF instead of an adjustable truss rod.

A lot of the builders who have weighed in on the side of CF in this discussion are right up there with those guys. They might not command quite the same prices, but that is largely because they have dared to part with tradition. The sad truth is that the bluegrass world (which is the primary market for mandolins) tends to be a little stodgy. Building a Loar replica is probably the most monetarily rewarding use of a luthier's time. Thankfully some have tried to at least subtly push the envelope sonically, and aesthetically, despite the inevitable resistance. You may prefer a traditional instrument. More power to you. That doesn't invalidate anything that has been said about the properties of CF.

[papawhisky]

I never expected my question to generate such a great thread, and I have learned a lot. No question that F models are about tradition, and truss rods are part of that. As I understand it, they have a function. The need for that function may be increased by the wood removed to accomodate them. But that was an innovation a hundred years ago. I suspect that when the Gibson company filed for the truss rod patent nearly a century ago that there were critics of the new fangled device.
My vote is for innovation where it fits in the tradition. My CF reinforced mandolin should arrive next week.

Papawhisky

[mandozilla]

If a person's mandolin has CF and no truss rod but wanted to maintain that traditional look, I suppose he/she could attach a truss rod cover to the headstock and no one would be the wiser.

[Nick Triesch]

The best mandolins I have ever played had adjustable truss rods. Who knows, maybe in a few years CF will be the rod of choice in a mandolin. But I think there will always be those who want a Loar type instrument that can be adjusted. I still think most players are old school . I owned a Martin 000 21 that the company
said never needed adjustment. But I could shoot arrows off it if I wanted to. Thats how warped it was. It needed hundreds of dollars to make it right. A CF neck mandolin may never need an adsjustrment...but if it ever did, you are just out of luck. With a Weber or Collings for instance, you just pop it right back into place. Takes just a couple of minutes. We all know how great mandolins with truss rods can sound, and how easy it is to adjust them, why would you not want a mandolin built with a quality adjustable truss rod? Why would you want to take away a way to fine tune adjust an expensive mandolin? Nick

[David Houchens]

I finally decided to give cf a try. I just threw a wild chery A model in the next batch of mandolins. I decided to try a few new things on this one. Try and keep it for a while and see how they work.I glued in two 1/8 x 3/8 cf rods in the neck blank. I'm also gonna give the Simminoff neck joint a try. Never done anything but dovetail joints before. I'll keep an eye on it.

[Dave Cohen]

NTriesch, at the 2006 CMSA convention in Louisville, two mandolins with CF reinforced necks and no truss rods took first and second place in a blind listening test against numerous other mandolins, all with truss rods. The test was jointly conducted by the CMSA and First Quality Musical Supply. Does that mean that mandolins with CF reinforced necks sound better than mandolins with truss rod necks? Absolutely not. The test cited is only an anecdote. Same is true for all of the anecdotes you have cited in your posts in this thread. Unless you have conducted experiments with quantitative measurements and controls, you can make all the claims you want about truss rods and they will amount to nothing. The listening test which I cited above also did not not involve any quantitative measurements, and there were no controls, so it, too, is nothing more than an anecdote. I cited it to make this point.

So (a) how many high quality mandolins with CF reinforcement/no truss rod necks have you listened to? And (b) How many mandolin necks with truss rods (beside your own mandolins) have you adjusted?

My point is not that mandolins with CF-reinforced necks sound better than mandolins with truss rod necks. I am not claiming to know the answer to that. My point is that you are not making your case. To the contrary, you are probably damaging your own case.

Bryce, if you put the 1/8"x3/8" CF rods directly under the fingerboard, you may not have enough reinforcement. The rods need to be set deeper in the neck, and get deeper as the neck gets deeper. I used the two parallel 1/8"x3/8" rods oriented vertically early on. Most of those necks did remain straight, but a few moved more than I liked.

http://www.Cohenmando.com

[labraid]

Re: placement: has anyone mathematically determined where the neutral axis is in a mando or guitar neck? Anyone got any diagrams? Numbers?

[barry k]

Hey, I have a thought, might end all discussion. Get a carbon fiber block big enough to obtain a neck blank, then walla....solved the warping , weight, adjustablity, filler and tone problem all in one swoop. And didnt have to chop down a tree either. Might just work ???

[Nick Triesch]

Dave, I don't think I am damaging my case at all. What case? Most of us mandolin players have mandolins with adjustable truss rods. That's the case. If you few guys want to build expensive mandolins with CF that you cannot adjust any warp or bow out then have at it. I have not done any experiments on mandolins or guitars. I have adjusted several guitars and mandolins in my life with great results. I grew up with terrible, cheap instruments because we didn't have a lot of cash. So now in my life I have some pretty cool instruments and I demand that they are set up (by me) to play like glass. The adjustable truss rod and good set up combined helps me do this. Thats all. I myself (and many others) will never buy a mandolin without an adjustable truss rod. How bout this....what ever floats your boat? Nick

[labraid]

Nick, I think your point is made. We're jazzing CF now, not who sells more mandolins for the most and why. It's immaterial.

[MnRoss]

I have a friend up here in Mn. with a "High Buck" CF truss rod mando built by a famous builder. Minnesota has pretty extream climate swings both in temp and humidty. There are times during the year his mando is unplayable. Sure in the perfect climate of his house it's fine but gigs up here in the winter can really, how should I say this "test the best of fine acoustic instruments". I adjust my truss rod twice a year and keep a good action, he switches to guitar for gigs when his mando is unplayable and just leaves the mando at home. It's not that the neck bends it's just the fact that the whole instrument moves around is what I think. I can compinsate and adj but he can't although this year he is going to build a new lower saddle and try that during the winter he says. It's going to be silly low and almost nothing at all so should be fun to see what happens.

I'm just saying that in some conditions it's nice to have that adj option and here in Minnesota I'm glad I do. My house I keep just fine temp and humidity but I play a lot of gigs that tend not to have those conditions. Nothing worse than having your e string bottom out on the frets in the 1st set because the room/hall you are playing in is so dry. I like to have the adjustment and after seeing and playing my buddies "5 figure" cf truss mando in conditions like we live in it would be hard for me to own something that I couldn't adjust. This is just my opinion and it has to do with where I live and playing gigs all the time in places that I can't control the humidity and temp during the winter.

[Michael Lewis]

Well, that is a common example of misinformation regarding 'adjustable" necks. The adjustability (truss rod) is to set the RELIEF, not the action. The action is set by raising or lowering the saddle. The truss rod does effect the action, but it is not there to do anything but set the relief.

[Dave Cohen]

Barry, a solid CF composite neck would be very heavy. The CF composites have a density greater than ebony. The all CF composite instruments made thus far have had a molded hollow neck.

What hasn't been mentioned so far is that it is possible for a CF/reinforced neck to bend if the reinforcement is not sufficient or is not done properly. The tension from the strings acts on the overall stiffness of the neck. So if a particularly soft piece of maple is used for a neck, just throwing some CF at it may not effect enough stiffening to resist the string tension. The shape of a neck + fingerboard puts the neutral axis pretty close to the glue seam between the neck and fingerboard. But just placing the CF or truss rod immediately below the neutral axis is not enough. The maximum stiffening in the case of CF, or the maximum adjustability in the case of a truss rod, is obtained by getting the rod or reinforcement (whichever is the case) as far below the neutral axis as is possible.

To amplify Michael's point, the greatest amount of movement in an instrument occurs in the plates. When the relative humidity is low, arched plates will sink a bit, hence the motivation for an adjustable bridge. The dimensional change comes largely from cross-grain shrinking. The shrinking of woods along the grain is minimal. If a neck moves, it is usually in response to string tension, as the shrinking along the grain is minimal. Left that way long enough, the deformation in the neck becomes permanent, as the wood cells "creep" past each other.

Sometimes truss rods are able to do their intended job, and at other times, they aren't. That's why Stew-Mac sells a truss rod repair kit. Truss rods can get maxxed out, the threads can strip, and other problem can (and do) occur. Adjusting a few rods on your own instruments is not enough experience. When repair is part of your living, you become less confident about the adjustment capabilities of truss rods. As I have already said several times, sometimes they work well, and sometimes they don't.

http://www.Cohenmando.com

[labraid]

I still have a beef with the neutral axis assumptions... I may be wrong, but that it is solely a function of profile seems a bit off. The compression forces down are so much greater than the vector pulling perpendicular from the nut, this would shift the neutral axis sufficiently far back in the neck to put most of the neck in compression, and only a small portion in tension. Any takers? Any math?

[Dave Cohen]

The location of the neutral axis is not solely a function of profile. It is influenced by the profile, but also depends on ther elastic modulus of the neck material itself, the modulus of the fingerboard, and the "wedging" action of the frets. That kind of calculation can be done, but to me, it is not worth the time. I think you would find that much of the neck below the fingerboard is in tension, but the tension is a lot greater much deeper in the neck than it is just below the fingerboard. It is enough for me to know that I get the greatest stiffening by placing the CF composite bars as far below the neutral axis as possible, and hence as deep in the neck as possible. As I already said, the same applies to the capacity of a truss rod to adjust the neck's curvature. The deeper the rod is located in the neck, the more easily it will adjust. Look, if the part of the neck where the truss rod is located were not in tension, the truss rod would not work. What the truss rod does in effect is bring back the tension that is lost as the wood cells move past each other when the neck begins to take a "permanent" set.

http://www.Cohenmando.com

[John Bertotti]

So then isn't the neck join also a function of the amount of tension the neck will feel? I would think the dovetail would be more secure placing a bit more tension in the deep part of the neck but a mortise and tenon maybe might give a bit? I'm not math man so I am just guessing but in my reality at the moment it makes sense!

[labraid]

My argument is as follows:



where the red are force vectors and the blue is the neutral axis in each case.

> The deeper the rod is located in the neck, the more easily it will adjust.

This is true, but I believe the neutral axis can change in an adjusted versus an unadjusted neck, and this is why a compression rod works. In effect, by its use you eliminate the neutral axis by keeping the rear of the neck in as much compression as the fret side. Simply neutralizing or equalizing forces. The further away from the median zone (not a true neutral axis, but a relative, inferred one), the more this compression force torques around that central point, making it more efficient..

> if the part of the neck where the truss rod is located were not in tension, the truss rod would not work.

Respectfully, I do not think that is necessarily true. One area/zone may appear to be in relative tension, but in effect it only needs to be in lesser compression to do that: you could bend a solid rod, for example, by simply compressing one edge more than the other with a perfectly in-line force, but in this example the whole rod will be under compression (though varying amounts, thus!) and that bend will still occur.

[Dave Cohen]

Brian, that just ain't right. The neutral axis of a given beam relative to its own dimensions doesn't change with a change in tension. Nor does it change with a change in the location of the forces acting on it The amount of shear changes, but the neutral axis doesn't. The location of the neutral axis is a function of the cross section, moduli, and homogeneity of the beam itself.

John, first, a dovetail joint - pin plus tail - is a kind of mortise and tenon. Second, regarding a straight mortise and tenon "giving": if it does, then your neck joint is in trouble. If a straight mortise and tenon joint is secure and the glue doesn't creep, then it is not going to result in any change in tension. Btw, when old dovetail joints fail, it is sometimes because of shrinking of the wood in the joint. I recently reset an old Guild guitar neck that needed about 0.020" of shims total to make the joint tight again. The self-locking aspect of a dovetail joint was just about gone. The part of the tail (tenon) nearest the back plate was just about coming out of the mortise by itself when the guitar was brought to me. With the shims in place, the joint was once again self-locking.

http://www.Cohenmando.com

[markishandsome]

"Usually after the adjustment you will not need to touch it for several years. And if you need to it will only be a very tiny tweek. I want to make the tweek for $ 3000 to $10,000 bucks!"

I don't know, if i spent 10 grand on an instrument I don't think it would bother me to have to throw down an additional hundred IF it needed to be tweaked by a professional several years later. Frequently played instruments need fretwork every several years anyway. Even the finest instruments require periodic maintenance. Would you refuse to buy a new car if it didn't have an transmission you could personally disassemble with an allen wrench?

[labraid]

Compressing a piece of foam in my shop vise proved my point to be true. A 2"x2" cross-section of homogenous foam 150mm long (~6") was cinched at a lower extreme edge (compression face), and the vise tightened. The 150mm tension-face mark was measured at the beginning, during, and at the end of compressing the block of foam over 1/2" in the vise. At first it appeared you were correct, the 150mm dimension became 151mm. Still, a mighty small gain, but there was an outward bow in that dimension, which when flattened out by pressing with a straightedge, put the dimension back at 150mm. One side compressed 1/2", the other not at all. Illustrating precisely my drawing at center.

I did something similar with a taller piece of foam putting all of my weight on one edge. That edge compressed the other bowed. When the bow was pushed back into place though, simulating actual rigidity in the perpendicular direction as a mando neck has for all practical purposes, the edge with my weight was compressed, and the other would not flex/expand/what-have you. This was easily visible without measurement, and I encourage others to try it.

Now back to the first example: when the foam was held straight up in the vise, and I pushed perpendicularly against it, it was easy to see how much it moved, 100mm became 102mm easily. It is the bowing action, the flexing which creates the neutral axis you're describing, just like in my left-most drawing. That neutral axis never moves if the force remains purely perpendicular, and it factors in with component vectors, but it is not the be-all, end all. Take any material which is restrained from bowing (for practical purposes this is the mandolin neck in question), compress it at an extreme edge, one edge gets shorter, the other doesn't move. If I had another force vector in the drawing above, this time pressing at actual center of the neck and not at an end, where would the neutral axis be then, the axis which you've say never changes position within a cross section? It would be "undefined", because there is none possible.

[David Houchens]

Dave, I didn't cut the groove for the cf parallel to the back. I guess I should have. I did however cut the groove 1/2" deep and used a cover strip as I usually do for a 3/16" adjustable rod. So its not quite at the fingerboard. BOUT AN 1/8" DOWN. My neck is only .640" at first fret(not counting the fingerboard), with a V shape so I wouldn't want to lower that end much more. it only leaves me with .140" of wood in the thickest place. Is there a tearout factor if you get the cf too close to the back. How close too the back would you recommend as being safe? I have seen older banjo necks that had split under 2nd-3rd frets. Don't know if too much tension on the adjustable rod caused this or not. If I do another, I do believe it would make sense to lower the heel end to be parallel with the back of the neck. Its about .740" at the 8th fret, so I could lower that end about .100" or so.
Also, I placed the two 1/8"x3/8" bars side by side. Would they be any stronger spaced apart a little?

[Rick Turner]

Brian, are you trying to say that a compression rod eliminates the neutral axis?

In a bending structure there is always a neutral axis as the force on one side of the beam, be it compression or tension, tries to overcome the force on the other. If you put too much compression on the beam, it will rupture and buckle, but you've still got a neutral axis in there 'til you simply crush it.

Dave quite rightly has pointed out that a material like CF, which is at it's very best in resisting tension, does the most good ounce for ounce as far away from the string tension as possible. That's why in my first patent (1978) for a graphite neck, I made the neck a hollow CF shell...a tube. CF also happens to be pretty good in compression, and can be improved by adding carbon nanotubes to the epoxy that sticks the whole mess together.

In practice, we often use CF in less than perfect ways which are, perhaps more practical. The stuff is so good and so versatile that using CF rods as we do, even when part of the rods are in the neutral axis, the result is really good. I like the combination of CF and wood because I like the feel of a wood neck, and I like being able to change widths, depths, and shapes to suit various customers. With a fully molded CF neck you are locked into a particular shape, and tooling is prohibitively expensive for custom instruments. Combine the best of both worlds, and you've really got something. And you can put a truss rod in a neck with CF if you want to...

[Nick Triesch]

Mark....I think it could cost a lot more than a hundred bucks if your non truss rod neck needed an adjustment. But forget all that....I really like Ricks idea of using both....CF with an adjustable truss rod. If you have to have the CF installed. Nick

[labraid]

My physics may not be up to PhD snuff, but my understanding of the neutral axis as I learned it some 12 years ago was that it was the plane within a solid where the wood was neither in distortional compression nor in distortional tension, rather it is a plane separating these two distortioning forces. If a some rigid material were pressed in a vise until it compressed 1/16", there would in fact be no neutral axis -- it would be under pure compression -- there is no tension in the wood, because tension is a distortional force and requires displacement. Tension of wood net compressed in a vise is negative, therefore there is no divide between compression and tension, therefore no neutral axis.

I know a guy who once overlooked gluing in his CF rod, rod which was directly beneath the fretboard. He went and strung up his new octave and heard these creaking sounds from within the neck. The neck was bowed up into exagerrated relief. Now tell me, why was the CF inside the neck creaking past the neck wood? Was it creaking one way on the top of the rod and one way on the bottom of the rod?? because that is where the CF was, directly on the "accepted" neutral axis.. No, the wood that gave way and caused that "big relief" was wood missing from the CF slot -- it was compressing around it and the CF wasn't taking up the fight. The neck wood was shrinking right over that CF rod...

[Rick Turner]

Brian, I'm not going to get into a pissing contest over engineering semantics here. You build 'em your way, I'll build 'em mine. You understand 'em your way, I'll understand 'em mine. But I have yet to see a neck that is entirely in compression...and I don't expect I ever will unless it's got strings on the back of the neck as well as the front.