1. Please take 30 seconds to register your free account to remove most ads, post topics, make friends, earn reward points at our store, and more!  
    TalkBass.com has been uniting the low end since 1998.  Join us! :)

Trussrod Dilemna

Discussion in 'Luthier's Corner' started by Hambone, Dec 8, 2004.

  1. In true personal fashion, I've reinvented the wheel again and come up with my own trussrod system. I'm to the point of installing an aluminum channel over my trussrod for the full length of the neck. The aluminum channel is flanged at each end to catch in slots on the sides of the trussrod channel. This makes it stay put and isolates it from the trussrod itself.

    Now I'm stumped on whether I should also glue in the channel with epoxy or put a fingerprint full of silicone along the sides or just leave it and glue the fretboard on. To those neckbuilders out there, what would you do?
  2. budman

    budman Commercial User

    Oct 7, 2004
    Houston, TX
    Formerly the owner/builder of LeCompte Electric Bass
    Is there is a possibility of something rattling? Perhaps a little silicon.
  3. pilotjones

    pilotjones Supporting Member

    Nov 8, 2001
    Without knowing what your goal or method is in having the channel there in the first place, it's hard to say. Is it more than just a plain flanged channel, perhaps with an integrated nut? If so, that could have some bearing on the matter, depending on what action is required. Is it acting simply as a reinforcement to bear the tensile or compressive stress (such as is produced by the truss rod) that would otherwise fall on wood? In that case, the choice would depend on whether you want the stress to bear on the channel only (use silicone, to allow slip) or in the channel+wood (epoxy).

    Is your channel flanged outwards at the ends, creating two "ears", with the ears sitting in thin slots in the neck shaft? If so, you are creating a reduced cross-sectional area in the wood at that point; the narrow slot shape adds even further to its functioning as a stress riser, probably at the thinnest and narrowest point of the neck, by the headstock. You could alleviate this somewhat by filling and bonding the tabs into the slots a strong and stiff epoxy. This would be all the more a consideration if for example the tabs were pulling inwards on the wood (towards the middle of the length of the neck), and then an exernal force were applied to the headstock to spread the slot in the opposite direction… not good.
  4. andvari7


    Aug 28, 2004
    Can we see what it is you're trying to do? I'm really interested in knowing, since I'm gathering information about my truss rod (the rod itself is a standard, Gibson-style rod, but the channel is what's unique; I'm borrowing this technique from Carl Thompson).
  5. Without having to produce a drawing, here's a little more on the design...

    The aluminum channel is 3/8" wide x 3/8" deep and sits in a square bottom routed slot over the TR with the web up against the fretboard. The fit is friction tight. The TR is NOT connected to the channel. This is by design. The ends of the aluminum channel have been made with very short lateral flanges - around 3/32" - that seat in short vertical slots alongside the TR channel. The flanged ends will be epoxied into their final position. That keeps the aluminum from moving up or down the TR slot. The TR itself is a ¼" dia. rod covered in heat shrink tubing so there's no prob with rattle.

    The purpose of the channel is simply as a stiffener. So, as before, my question is should I also glue the sides of the aluminum channel to the neck wood or let it stand? Also, would I put adhesive on the top of the channel under the fretboard or leave it clean during glue-up?

    It seems to me that if I were to use epoxy on the sides of the aluminum channel that it would increase the stiffness of the entire assembly with the pair of extra glue surfaces this would provide.
  6. Interesting. A take on the Martin u-channel trussrod...the only thing, with the rod free-floating, is it a conflict of interest? You're adding a third factor against the stress of the string, sorta like the rods in a Warmoth neck, but more localized along the same path of the trussrod... I'd be apt to think they wouldn't work well together as they're not peened to each other, and could result in a bit of punching thru the wood eventually...but in practice, it's just a variation on a theme that's worked for a long time....

    Not to say it won't work. I'd add nothing more than a bit of silicone to keep the rod from moving...I'd be more concerned with the trussrod rattling inside the tube. (edit: ah, heat shrink tube...nevermind)

    Good luck with that, I'd love to know how it turns out.
  7. bwbass


    May 6, 2002
    If you want the aluminum channel to add stiffness, you should glue it to the wood in the neck, preferably with slow-cure epoxy. However, it probably won't add much stiffness because of its placement in the neck. The steel bars we use in our necks (just under the fingerboard) add more mass loading than stiffness, reducing dead spots by lowering the resonant frequency of the neck and increasing sustain.

    Reinforcements glued into the neck add more stiffness exponentially the farther they get from the neck's neutral axis, where the wood fibers are being neither being compressed nor stretched as the neck acts against string tension. This is usually just below the fingerboard. The same amount of aluminum in tall, thin bars, as deep as you dare to go into the neck, would make the neck much stiffer.

    In a maple/maple Jbass neck, your aluminum channel would make the neck about 8.5% stiffer than if you used no reinforcement at all. Two 1/8" x 1/2" tall aluminum bars would make that figure about 16% and weigh about the same. 'Course that would mess up your truss rod design...

    Just wanted to put my 2c in on my favorite topic!
  8. pilotjones

    pilotjones Supporting Member

    Nov 8, 2001
    As you surmised and Brian also said, if the aim is to increase stiffness, the channel should be unified with the neck along the entire length, and on all contact surfaces. This follows the general principle of long members sharing a stress. Think of it this way: take four slats of wood, 1/4 x 1 x 36, and suspend them, stacked on top of each other, between two tables. put a 10 lb. weight in the middle of the span. Now try the same thing with the four slots glued up as a solid body. You get the idea here. And the same effect is evident if you stood the slats vertically, and put a weight on the top - in the first instance, they will buckle; in the second, they have no problem supporting the weight.

    Brian has another point here, too. When a beam bends, the concave surface is under compression; the convex surface is under tension; and there is a neutral surface (comparable to the neutral axis present in any given cross-section) that is under no tensile or compressive stress. The farther into the high-stress zones you put any reinforcing material, the more good it does. So, while the web of your channel is just under the fingerboard, which is as good as you can get without routing the fingerboard itself, the neutral axis is (due to the neck shape and the position of the string-applied force) not too far under the fretboard. So, the flanges of the channel are probably straddling the neutral axis, and not doing much good. Deeper in the neck would be more effective for any reinforcement.

    Also, I'm a little wary of aluminum for temperature reasons. For another thread, I looked up and found that oak (the first wood I could find the proper numbers for) had a thermal expansion coefficient that is 1/2 to 1/3 that of steel, IIRC. But aluminum has an expansion that is - maybe double? I'll look it up tom'w at work - that of steel. So, with any temperature change, you'd be setting up maybe double the "normal expected" stresses, which would affect glue joints, and neck warp.

    Off to bed.
  9. I might have confused things by not explaining that even though this is a single TR setup, it is also a double action setup. So, like Mon observed, I'm using the components of the Martin (or Gotoh) Dual action TR's but as seperate pieces. I no longer rely on the aluminum channel to bend against the fretboard as the means of straightening the neck. The channel is just for stiffening. I also have 2 additional strips of "hardwood" sunk into the back of the neck as additional stiffening. If it seems like overkill, that's my diabolical plan.

    Do any of you think that turning the channel over and putting the web at the bottom of the TR slot would be better? As I understand the concept, this would place the surface with the most tensile stress further from the center axis and should utilize whatever stiffening properties it offers to the utmost.

    Pilot, the thermal expansion thing doesn't bother me much. It's my contention that the instruments are going to be in fairly controlled conditions most of the time and any temp fluctuations are likely to be brief, minimal, and aren't going to leave a lasting effect. Besides, my TR design is extremely adjustable so a quick tweak won't be out of the question if things begin to go wobbly as you suggest.

    So, basically, I'll be keeping my fingers crossed. :)
  10. pilotjones

    pilotjones Supporting Member

    Nov 8, 2001
    When using a rod in a non-curved channel, any flexing of the neck that it achieves to counter or add to the string pull bend is due to it's being off-center to the neutral axis. A rod lying on the neutral axis will do nothing. And that is what I'm concerned about here.

    If you start with the idea of a homogeneous rectangular beam, the neutral axis or plane runs down the middle. Now, a neck is shaped like an approximate D in cross-section; this brings the n.a. towards the fingerboard side. The fingerboard is typically a stiffer wood than the rest of the neck shaft (unless it's a maple-maple or wenge-wenge neck), and it's also the widest part of the neck. These two factors causes the n. a. to go further towards the fingerboard. Now, you add stiffening members towards the back of the neck. This will shift the n. a. towards the back. Now, if you add a channel with the the web towards the back, assuming the n.a. had been just under the fretboard, you now pull the n.a. a little further towards the back. What I'm afraid of is that you may put the n.a. right at the rod, disabling it from doing anything.

    At mimf someone had done a few cross section - neurtal axis plots. You might want to look at these, keeping in mind whether or not they accounted for the typical stiffer fingerboard materials when the plots were done.
  11. pilotjones

    pilotjones Supporting Member

    Nov 8, 2001
    Regarding expansion, I ran some numbers on how much the aluminum might want to expand w/r/t wood. (See image.) In a neck assembled at 70ºF and heated to 120º, with a 24" channel, the differential expansion is .012". Is that enough to cause cracking or glue separation some where? I don't know.
  12. andvari7


    Aug 28, 2004
    Okay, I now have a question for you, Mr. Jones:
    The technique that I will be using for my truss rod channel, which is the technique used by Carl Thompson (I acquired this technique indirectly from him, so it's accurate), is as follows:
    The rod itself is a standard, Gibson-style rod with a 1/2" diameter steel slug at the other end.
    The channel is a straight route, however, it is angled. Think of it this way: assume the neck is at a slight angle (like a Gibson Les Paul). The fingerboard is a mirror, if you will. The route will mirror the strings, which are obviously angled off of the fingerboard.
    There is a block that goes in between the rod and the board. That's just large enough to leave a little room for the rod to work.

    That said, is there any sort of problem with this approach?
  13. pilotjones

    pilotjones Supporting Member

    Nov 8, 2001
    First, analytically: If I understand correctly, the route is shallower by the headstock, and deeper towards the heel. The rod is then covered up by a wedge-shaped block before the fingerboard is mounted. Right?

    Sounds like for most if not all of its length, the rod is on the far side of the neutral axis (from the strings), so it should work fine-- the rod tension will counter the string tension on the opposite side of the n.a.

    Second, experientially: if the method works once (for CT), it will continue to work for you. If you put strings on a CT bass with this design, and the strings bow the neck forward a little too much (action too high), then you adjust the rod and it works properly to pull the neck back (lower the action), then it works-- regardless of any analysis. I would keep in mind though that dimensions could be important. For example, it you make the channel half as deep as CT does, or, if you make it the same depth but your neck is substantially thicker, it might not work properly.
  14. andvari7


    Aug 28, 2004
    Wow. That explains, in quite wonderful detail, why I'm going to do what I'm going to do. Thank you very much. Now I can spend more time on the buttered cat principle.
  15. pilotjones

    pilotjones Supporting Member

    Nov 8, 2001
    ?? :confused:
  16. Hey guys,

    I'm not ducking the thread here. It's just that now I'm in a total state of doubt and confusion over this design - a very uncharacteristic condition for me. I've read with great interest everything discussed here and I understand the principles and this has put me in a position where, either I scrap the design for another OR I modify what I've got to work better. I'm leaning towards the latter.

    Thing is, I've got 2 partially completed neck blanks here and one of them is to be the neck for my latest and greatest. Until this little hitch is straightened out, the instrument isn't getting finished. That's my motivation here... ;)
  17. Tim Barber

    Tim Barber Commercial User

    Apr 28, 2003
    Serenity Valley
    Owner: Barber Music
    Well, I wouldn't scrap the idea based on theoretical discussion. Like Pilot says, if it works, it works. Is the new bass a bolt-on? Make two otherwise identical necks, one with your U-channel idea and one with whatever you normally use. String it up for a month with each neck and take notes.
  18. A9X


    Dec 27, 2003
    Sinny, Oztraya
    My thoughts late in this discussion were similar to Pilot's, and my concern is for thermal expansion, after all didn't the Kramer al necks have issues with that? Moses sell a U shaped graphite channel for $20 or so iirc which would at least get around that problem whilst retaining the same concept.
  19. Dharma, the big difference in my experiment and the Kramer model is the amount of aluminum in the neck. The Kramers were mostly aluminum including all of the structural components while my own design contains probably less than .5% by weight. I also take into account that thermal expansion is a progressive thing. The increase doesn't all take place at once and it doesn't reach it's maximum at the temperatures the instrument will likely see. Since Pilot has already calculated that it will take a temperature swing of 50º from 70-120º F to affect a .012" change, I feel assured that I'm pretty safe in this regard.
  20. A9X


    Dec 27, 2003
    Sinny, Oztraya
    Weight isn't the issue, it's the longitudinal length of the component, which for two say 34" BO necks is going to be roughly equivalent, eg Kramer Al and yours.
    Not doubting Pilot's numbers, I just offered it as a suggestion and was more concerned for longevity as 120F is not uncommon in cars and any thermal cycling may cause cracking in the neck over time depending on how it's finally constructed.