The Truss Rod
In todays steel string acoustic guitars, the truss rod is as common as the soundhole. But what does it do? Is it really necessary? And what would happen if it were removed and/or replaced with something else, such as carbon fiber? Truss rods are available from several sources and come in more than one shape and size. They all share one thing in common: they have been engineered to stabilize, and compensate for the potential movement of, the neck of the steel string guitar. This need arises as a result of the tremendous force exerted upon the neck by the strings, and can be compounded by environmental factors.
A six-string set of D’Addario EXP11 Light Gauge strings applies 159 lb (72 kg) of tension on a 25.5″ scale length guitar. A light gauge twelve-string set from the same supplier (EXP36) is rated at nearly 242 lb (110 kg) of tension! Comparatively, non-steel string guitars exert much less tension on the neck, typically ranging from 75 to 100 lb (34 - 45 kg). Subsequently, truss rods are rarely found in such instruments as the necks are inherently stiff enough to resist the pull of the strings.
Truss rods are set in place during neck construction, embedded into a channel in the neck just below the fingerboard. Contrasted with a perfectly flat fingerboard, an ever-so-slight forward bow in the neck introducing what is known as relief can be desirable for reducing string buzz, especially for low action setups or aggressive strummers. With nothing but the wood of the neck to resist the pull of the strings, a neck having no truss rod and under high steel string tension may simply continue to bow forward, eventually making the instrument un-playable, as the strings continue to elevate farther away from the fingerboard.
Truss Rod History
Early renditions of truss rods were simply fixed stiffeners. Extra-stiff wooden bars or metal square tubes, T-bars, I-beams, U-channels, etc, went a long way toward prolonging the inevitable forward bow but, as they possessed no inherent adjustability, they eventually fell out of favor as innovations were introduced. Wooden trusses are still, well, wood. To be rigid enough to be effective they are, by necessity, heavy. And they are subject to the same environmental factors the rest of the neck is subject to. By contrast, to be light enough to be usable, fixed steel trusses are still subject to bending and, once bent, may form a rather permanent forward bow. I believe the fault of the early fixed stiffeners was one of material availability, more so than adjustability. More on this topic in a moment...
A single-action truss rod, or compression rod, is typically secured at one end, compressed into a slight concave bow and held in place via a filler strip, and the opposite, threaded end is exposed through a recess (typically routed into the headstock on acoustics and vice-versa on electrics). Tightening the nut on the threaded end attempts to shorten, and subsequently straighten, the embedded bowed rod, compressing the back of the neck and lengthening the fingerboard to counteract the string pull. To demonstrate it’s functionality: with no strings attached, tightening this rod would result in a slight back bow of the neck, and loosening it would (should) return it to dead flat again. Since the strings are attempting to pull the neck into a forward bow, the result is a balance between two opposing forces. The compression rod has proven to be quite effective, but is not without faults. The embedded rod is held in a pre-determined curve based on an educated guess regarding the future movement of the neck. Over time, the compensation effected by the tightening of the compression rod may not be sufficient, as the rod can only move from it’s original slightly bowed position toward straight (at best). Many a compression rod has had additional threading applied in order to apply more pressure. Over-tightening often results in a snapped rod, a very undesirable condition. Due to the one-way adjustability of the design, it is often difficult (if not impossible, without special jigs) to return the neck to the straight, dead-flat condition needed for fretboard/fret work, setups, etc.
Dual-Action Truss Rod
Enter the clever, if not ingenious, dual (double, twin) -action truss rod. It’s self-contained, two-way adjustability overcomes the limitations of the uni-directional compression rod. Turn clockwise to induce a forward bow, and counter-clockwise to bow the neck backwards. Along with the ease of installation, requiring only a simple, straight channel, it is understandable how the design skyrocketed it to truss rod super-stardom. Does the dual-action truss rod have a downside? More neck material must be removed to accommodate it and, unlike the compression rod, the channel it occupies must be left open perpetually requiring greater care when attaching the fingerboard. It is typically heavier than it’s single-action cousin, has been known to introduce the dreaded intermittent rattle (Ugh!) and can also fail (welds, stripped threads, snapped rods, etc). Despite it’s shortcomings, it is ubiquitous today.
Whether universally quantifiable or not, a growing contingent of builders and players are convinced that these steel machines inside their instrument necks can be/are detrimental to both the playability and the sound of their guitars. It is believed that the weight of the added steel does not make for an optimally balanced neck, and the combination of the quasi-hollow channel and metal fixture affect the tone. The need to counter the pull of the strings remains. What is a luthier to do?
Carbon Fiber Stiffeners
Carbon fiber is a strong, stiff, lightweight composite material that boasts exceptional stiffness-to-weight and strength-to-weight ratios. Fibers comprised of long chains of carbon atoms are embedded in a matrix of epoxy resin, resulting in material having properties similar to steel with the weight of a plastic. Steel and plastic are said to be homogeneous and isotropic, meaning their properties (mechanical, electrical, thermal, etc) are equal at all points, in all directions, throughout the material. Carbon fiber, on the other hand, generally exhibits strength along the axis of the fibers only. A carbon fiber part would need to be specifically constructed to approach isotropic (called quasi-isotropic) properties.
The use of carbon fiber for stringed instrument neck reinforcement is now several decades old, and it is easy to understand why. The fact that carbon fiber often weighs less than the wood it replaces and is certainly many times stiffer and stronger can be reason enough to incorporate it. Some builders (me, included) have chosen to supplement an adjustable truss rod with parallel bars of carbon fiber running the length of the neck.
The decision to use carbon fiber as a complete truss rod replacement hinges on a given builder’s / player’s belief regarding neck relief. If I believe that a general, or proper, neck relief measurement exists, one that can be applied to all, or if I am willing to shape a relief into the fretboard if needed, then I am free to consider replacing the truss rod completely using carbon fiber. If I am convinced, however, that neck relief should remain a subjective factor, a target measurement will never apply to all (or even most), and frequent adjustments will be necessary, then a truss rod will be required.
Most luthiers and guitar technicians acknowledge that neck relief, while related, is to be treated independently of string height (action). Assuming we are starting with a neck having an adjustable truss rod, and that neck is dead flat (planing and radiusing are perfect, frets are perfect, string height at the nut is perfect, etc.), when we correctly set up the guitar for playability we first set the relief (completely independent of action), then address the string height at the saddle. Sadly, many tweak and fiddle with the truss rod to address action, which only exacerbates the issue.
If I believe the two-way adjustable steel truss rod is a necessary component in the construction of the guitar neck, then the notion of completely replacing it with a fixed carbon fiber stiffener seems like a tall order. First I need to settle on a target relief height, as it won't be adjustable, later. Long before the advent of carbon fiber, guitars were constructed using fixed necks, having pre-set neck relief, so that is not a new thing. Don't different gauge string sets exert different amounts of tension and won't that require adjustability to compensate for? Wouldn't it stand to reason that a set of heavy gauge strings in standard tuning cause a slightly greater deflection in the neck than a set of light gauge strings tuned down a step or two. Is a carbon fiber structural stiffener suited to the task of accommodating differing string tensions? But then, how many players are constantly changing string gauges? And what about flex? If a carbon fiber stiffened neck remains perfectly flat once the tension of the strings is applied, then the concern over flex is rendered moot. If there is any movement, what will it be and how will we calculate it ahead of time in order to build in relief? Are we going backwards by removing the truss rod? This article is a work in progress intended to address those very questions.
Along with my own subjective experience, I have obtained sufficient anecdotal evidence regarding neck stiffness and carbon fiber to purge a neck of a truss rod altogether. Rather than continue to experiment with various carbon fiber products, I will be using the Dragonplate D-Tube Neck Beam, a $75 carbon fiber structural component specifically designed for the task. This 16″ hollow beam weighs a mere 1.8 oz (0.05 kg). My trusty Martin-style adjustable truss rod weighs more than three times as much at a whopping 6.06 oz (0.17 kg). Comparatively, weight-wise, the D-Tube is practically non-existent! As of this writing, there are four (4) versions of this product available:
- Straight D_Tube 0.5″ wide x 0.45″ high x 16″ long
- Straight D_Tube 0.75″ wide x 0.5″ high x 16″ long
- Straight D_Tube 0.75″ wide x 0.5″ high x 20″ long
- Tapered D_Tube 0.75″ wide x 0.5″ high x 20″ long
I opted to begin with the second item in the list, a straight tube 3/4″ wide x 1/2″ high x 16″ long. The groove required to seat this particular item, being the same depth throughout, would be straightforward to rout.
I do not possess a CNC or Milling Machine, and instead rely on hand tools, a router table and jigs/fixtures with a handheld router. Adjust your own techniques, accordingly. I will need a few supplies: neck blank, template, the Dragonplate carbon fiber stiffener, router bit and epoxy.
Neck Construction and Preparation
For this particular application, I am using a glued dovetail neck design. It could just as easily be a bolt-on, mortise-and-tenon neck or even an integrated, Spanish heel. Mahogany stock is laminated (vertically) from five (5) pieces. This results in an attractive centered stripe that runs the length of the neck, all the way from the tip of the headstock to the heel. More importantly, it aids in stabilizing the neck against environmental factors, reducing the likelihood of twisting and warping. The mahogany block I am beginning with has been sized to accommodate two (2) of these necks by reversing the template and flipping it vertically. The necks are separated at the bandsaw.
My roughed-out necks need to be trued up prior to cutting any grooves, so they go the the bench to be hand-planed. I am mindful of the 15° angle of my headstock as well as the intersection of the headstock at the fretboard (a single pass with a hand plane will alter the dimensions).
Installing the D-Tube
Up to this point, these steps would be required for any conventional acoustic guitar neck built this way, with or without a truss rod. The only difference I am going to make in this neck that would differentiate from my standard necks is to rout a single groove into the surface of the neck and epoxy in the D-Tube, as opposed to routing three slots, one for the truss rod and the remaining two for carbon fiber stiffeners.Cost-wise, I have approximately $50 invested in my truss rod plus 2 carbon fiber stiffeners. The Dragonplate D-Tube replaces those three items.
I use a 3/4″ wide Whiteside 1411 3/8″ radius round nose router bit that perfectly matches the half-round profile of the carbon fiber beam.
A trip to the router table provides me with a couple of advantages over using my handheld router with, say, a centering base. Dust collection is a given at my router table and can be harder to control with a handheld tool. Additionally, the fence allows me to cut grooves quickly, safely and accurately. With a handheld router, I need to be much more vigilant regarding the proximity of the router base to the neck. If I would have to use a router this way (if I didn't have a router table), I would employ a jig to both cradle the neck and to guide the router along.
A 3/4″ wide by 1/2″ deep groove / channel is routed down the center of the face of the neck (the side to which the fingerboard will be attached). I intend to keep the D-Tube flush with the face of the neck blank, as opposed to recessing it. At 1/2″ deep, this material will sit very close to the back of my finished neck, so I want all the clearance that can be afforded me. For my first D-Tube installation I have opted to rout a through groove from the headstock to the heel block, as opposed to a stop dado, which I accomplished in three (3) deepening passes over the router bit.
The groove is cut ever-so-slightly deeper than the height of the D-Tube to accommodate a thin layer of adhesive. Once the epoxy has cured and the surface is trued-up again, the D-Tube will (should) be perfectly flush. I want 100% contact of neck material to carbon fiber at all points. I fashion a plug that will fill the void (the portion of the groove not filled by the carbon fiber), butting up squarely against the end of the D-Tube, a process which takes just a few extra minutes at the bandsaw and the disc sander. Alternatively, I could have attempted to shape the end of the D-Tube to conform to the compound radius left by the stop dado, or simply have plugged the end of the D-Tube and filled the small void with epoxy.
As an alternative to stopping the 16″ D-Tube short of the dovetail tenon end of the neck, I could use a longer 20″ D-Tube that runs the full length of the neck and projects out onto the tenon. Note that I have run the carbon fiber up into the headstock, rather than terminate the groove short of the point where the fingerboard ends at the nut, as I would with a truss rod slot. It has been a long time since I fashioned a neck without using some form of carbon fiber stiffener, and I have always extended those stiffeners into the headstock area. Having witnessed a few cracked headstocks from other builders, I believe that crucial juncture of the neck benefits from the extra attention. As others have rightly pointed out, beware of the dimensions of the D-Tube, the slot in which it resides, and the target thickness of your finished neck.
The groove gets coated with a thin layer epoxy. Although the D-Tube is hollow, I am not flooding the application with adhesive, so there is no need to plug the ends. I am using the same 3M Scotch-Weld 2216 2-part epoxy that I use for all my significant carbon fiber-to-wood bondings. Both the carbon fiber D-Tube and my custom wooden plug are pressed into place. I lay a piece of parchment paper over the neck and *lightly* apply a few small clamps, merely to prevent any shifting during cure. Epoxy, unlike PVA or AR glue, does not benefit from high-pressure clamping; in fact, heavy clamping is advised against. The neck is left to dry overnight.
EDIT: A debate exists regarding whether or not to add a filler strip of wood above the carbon fiber. The contingent adding the filler suggest that a (micro-thin) wood strip, applied over the carbon fiber and secured with epoxy, will somehow provide a superior glueing surface for the soundboard. Technically, the argument rests on the adhesion capacity of the glues and materials in use, in conjunction with the concerns over delamination. I have successfully bonded fingerboards to neck blanks containing carbon fiber stiffeners using protein-based glues (Fish and Hide Glue) for some time, now, with no adverse results. These fingerboards can be heated and removed. I have also used epoxy to attach fingerboards and, while fingerboard removal can be tougher, there is zero concern regarding obtaining a sufficient bond with the carbon fiber. Perhaps a greater concern for this specific application is the fact that the groove cut for the D-Tube is already 1/2″ deep. Be aware that increasing the depth of that groove in order to allow for a thin veneer on top of the carbon fiber could potentially compromise the overall thickness of the neck.
After releasing the clamps and lifting off the parchment paper, I removed any accumulated, cured epoxy using a scraper. My go to scraper these days was designed by luthier Al Carruth and is available through StewMac. When setting the depth of cut for the groove, I had taken the time to ensure the D-Tube would sit just shy of the surface of the neck blank. I was rewarded as the scraper just kissed the surface of the carbon fiber as all the squeeze out was removed.To protect neck geometry and to simplify cleanup, I would advise applying low tack tape to the surface of the neck blank prior to adding the epoxy. Two narrow strips, one down each side of the groove, or a single, wide strip whose center is removed with a razor knife will mitigate against having to aggressively remove any squeeze-out, later.
The section of D-Tube that extended out over the headstock was sanded back nearly flush at the disc sander. I could also saw the waste off and true it up with a sanding block. (Tip: Wear a mask when machining carbon fiber!). For final dressing of the face surfaces of the neck and headstock, I rely on strips of adhesive-backed sandpaper affixed to a dead-flat marble slab. A few careful strokes across that slab are all that are needed to complete this stage of the neck construction.
I have found that, while carbon fiber machines nicely, it can dull my tools rather rapidly. Having an effective (read: fast, accurate) means of keeping my hand tools honed makes the process much more enjoyable.
The neck blank is surprisingly light and stiff at this point compared to my standard necks. I am very curious to see what, if any, effect the strings have on moving this neck away from dead flat. I have now replaced the truss rod (or combination truss rod / carbon fiber stiffeners) with the Dragonplate D-Tube Neck Beam, and will continue completing the neck with the same steps I use for all my necks.
Cutting the Neck Angle
For both mortise-and-tenon bolt-on and dovetail glue-on necks, I use a bench-mounted router jig that simplifies cutting the desired neck angle at the heel. The fixture, manufactured by Chris Klumper of luthiertool.com, clamps a neck with a centered truss rod slot to a vertical plate, neck heel pointing up. Two (2) perfectly centered locating pins accurately position the neck, while a dial indicator, juxtaposed to an adjustable horizontal plate that will support a hand held router, lets me calculate and dial in the precise angle needed to mate the neck heel to the body, as well as cut the tenon, be it straight or dovetail. It accomplishes this by letting me compensate for the (desired) offset between the plane of the fretboard and the height of the bridge. As I am constantly altering guitar designs, not having to stop to think in terms of angles is of tremendous benefit to me.
The D-Tube eliminates the truss rod slot which my neck angle jig relies on for positioning. The purpose of my neck angle jig’s two (2) centering pins would therefore be nullified unless I were to precisely locate and rout slots (not just holes) into the carbon fiber D-Tube. An alternative approach, prior to installing the D-Tube, would be to rout an initial, narrow slot (as though I were installing a truss rod) in order to utilize the pins of the neck angle jig, cut the tenon, then re-rout the groove for the D-Tube.
With the D-Tube already glued in place, in an attempt to center the neck without the locating pins, I first opted to modify my jig to reference the edge (side) of the neck instead of the center. If you are interested, you can view that article here.
While my modification worked, there was a lot more fussing involved in ensuring the neck block was perfectly centered prior to clamping than I was happy with. I returned to my initial approach: precisely locate and rout small slots into the carbon fiber D-Tube.
Stay tuned for more as I complete the neck and test it on a guitar...