Official Luthiers Forum!

Yeah, he missed a couple

[QUOTE=fmorelli] That's 15 rebuilds

Filippo[/QUOTE]

Yeah...after about the 10th time I might start getting a little grumpy myself!

_________________
JJ
Napa, CA
http://www.DonohueGuitars.com

I should add that repeatable accuracy and, as Mario suggests, the fact that fretboard is the main touchy feely bit in relation to guitars playability is the reason I built my jig. But JJ makes a good point as well, big chips don't clog your lungs like fine dust, sure you can wear a mask, but the most dangerous stuff is still there in your shop.

Cheers

Kim

Here's a fairly simple jig for radiusing fretboards. It will do cylindrical or conical surfaces. You just have to make an appropriate template for each end. I don't know how well it works but the author of the thread says it's perfect. You just keep sliding the board from end to end until the sander doesn't take anything off.

It's from a French forum for amateur luthiers hosted by the website of professional luthier Benoit de Bretagne, who makes some beautiful instruments. You'll find it about half way down this very long page.
http://www.benoit-de-bretagne.com/phpBB2/viewtopic.php?t=210 1&start=0

Making a neck and fretboard that play best for a given player is a matter not just of technology but also of art. This makes repeatability less important than it is in the production of things like machine parts. It makes the adaptability of the production method to the individual instrument and player important. If the best boards and necks were all the same, then there would be no way to improve on the playability of a neck made by robots in a place such as the Taylor factory. I believe that necks and boards machined to a factory standard can usually be improved for the purposes of a player.

A perfectly straight board with a perfectly even radius (even one that is a conic surface rather than a cylindrical one) is the starting point or reference for a fretboard. It may or may not be the end point, depending on how the guitar is to be played. One size does not fit all. I would say that the fact that the board is the main interface of the player with the instrument argues in favor of human hands and eyes being its final judge.

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

When all else fails, clean the shop.

[QUOTE=Howard Klepper] Making a neck and fretboard that play best for a given player is a matter not just of technology but also of art. This makes repeatability less important than it is in the production of things like machine parts. It makes the adaptability of the production method to the individual instrument and player important. If the best boards and necks were all the same, then there would be no way to improve on the playability of a neck made by robots in a place such as the Taylor factory. I believe that necks and boards machined to a factory standard can usually be improved for the purposes of a player.

A perfectly straight board with a perfectly even radius (even one that is a conic surface rather than a cylindrical one) is the starting point or reference for a fretboard. It may or may not be the end point, depending on how the guitar is to be played. One size does not fit all. I would say that the fact that the board is the main interface of the player with the instrument argues in favor of human hands and eyes being its final judge.

[/QUOTE]

You won't get any argument from me on that front Howard, but the jig I made can take me very close to where I want to be and any finiting can be done from there real quick.

I guess it all depends on what YOU as a builder are setting up to do. If for instance your building 50 guitars a year, best build a jig to assist you to keep up with production. If on the other hand you only build 5 PA, best build a jig to give you an accurate datum cause you may not be doing this enough to keep your eye in and some of this FB wood can be pricey.

I guess if your somewhere in between the 5 and 50 mark, you should make your own choices but the dust is a real factor that should not be fobbed off IMHO.

Cheers

Kim

Howard, I don't "get" your last post. 

what part of the radius can be shaped to the player? None. Yes, we can change the radius to suit some players, but once a radius is decided upon, there's no changing it part ways up the fretboard. and having that fretboard surface true and perfect is how I'm able to fret with SS and not touch but a couple 'spots' on the frets. Spots, as in maybe 2-3 1/4' or shorter spots.

If our fretboard isn't true, that's where we get to filing and shaping the frets so much. The frets index off the fretboard; a more perfect fretboard, makes for a more perfect fret job, and this is an area where we should try to machine to a .001" tolerance.....

The rest of the neck, shape, size, width, etc.. can certainly be changed, and is why I can't go CNC, not even CNC'd fretboards.

I'm also quite sure that hand-shaping fretboards is what lead to my issue with PauFerro and Bois D'Rose. Sanding vigorously in close proximity and breathing deeply couldn't have helped at all. As JJ said, router chips aren't near as bad.

A few of us got to watch David Collins work on a neck last week.  David is a natural born teacher and explains things very well as he moves along.

He mentioned to us that for some players he will level the frets to differing degrees in different areas, effectively creating localized relief again to match a specific player's playing style.  This may include creating sweeping areas of more relief on the bass side say from the 3rd fret through the 9th fret.

I did not fully understand what David said at the time but I have been thinking about it ever since  Now I think that I understand and I think that Howard is speaking of the same thing here but taking the customization to the fret board level - again to match a specific playing style and desired end result.  Howard am I close here?

[QUOTE=Howard Klepper]A perfectly straight board with a perfectly even
radius (even one that is a conic surface rather than a cylindrical one)
is the starting point or reference for a fretboard. It may or may not
be the end point, depending on how the guitar is to be played.
[/QUOTE]



Word.



What Howard is referring to, and what (I gather) David was also talking
about in Ann Arbor, is the playing style of a particular player and
optimizing the fret tops for that player. 



A fingerpicker with a light touch who is all over the neck would be
able to get away with very low action and almost no relief (dip)
between #3 and #10, except for some fallaway at the highest frets
(above #12). 



A flatpicker who strums hard with a heavy right hand and is always
between the nut and #5 will need more mid-neck relief between #3 and
#10, and more fallaway above #12 than the fingerpicker to avoid
buzzing.  Some of this relief can be had with truss rod tweeks,
but the relief should be a little more pronounced on the bass side
since the amplitude of string vibration is larger there (thus requires
work on the fret tops, as does the fallaway).



Guys who like low action often tend to like low frets, but some guys
prefer the ease of string bending that comes with jumbo frets. 
And on it goes.  I agree with how Howard described it, the
fretboard is the reference surface and the fret tops are where you fit
the instrument to the player.



(and since the nut and bridge string spacings are not the same, the
strings by definition describe a surface closer to a conical section
than a cylindrical one, something to keep in mind when thinking about
action, relief and fallaway)

Well as I understood Howard's last paragraph, to which my last reply was directed in the main, he was suggesting that one size does not fit all in terms of degree of compound and that the craftsman's eye and sense of feel should be the final judge of that transition. I recon that's a fair enough statement.

I maintain however that at the very least, a good jig will offer an accurate starting point from which to begin applying those skills. If you were after a mild compound, say 16" to 20", and you used a jig to cut the whole board to 20", that will give you a better reference point from which to shape in the compound by hand than starting out with a flat board.

That said, Mario makes a good point in suggesting that using your hand and eye skills can introduce human error which could then lead to chasing your tail when leveling the frets.

At this stage however I don't see much need for a compound radius fretboard on an acoustic guitar. I never say never so that could change in time but for now, what's the point?. Anyhow, at this stage in my journey I am really happy with a jig that gives me an accurate single radii of my choice that is easy to set up and does not create fine dust.

Cheers

Kim

There are a few different issues going on here; I'll try to separate them a little, but I can't say as I have my thoughts fully organized.

1) Accuracy: I'm all for accuracy, and I can understand that if you are installing pre-arched SS frets, the arch of the board has to be a match. But a lot of accuracy is possible sanding the shape with blocks and checking with straightedges and radius gauges. I may be a couple of thousandths of an inch off in places, but I may be a couple of thousandths off in tapping the frets down, and I'm going to file (lightly if things have gone well) anyway.   

IIRC, Mario does a single radius board. If so, the taper of the board and spreading of the strings from nut to saddle makes his board an inaccurate match to the strings right from the start--the fret tops will not be in a straight line below the path of any string, and they will be more off toward the edges of the board. The usual reply to this is that it's just a couple of thousandths of an inch and doesn't matter. Maybe it doesn't, but this is inconsistent with saying that you require machine shop accuracy for the player interface.

Also, if you use a fixture or jig, are going to sand the board anyway to get out the router marks and smooth the board. If you use a sanding platten, you still have to sand through the grits to polish the board. So the possibility of inaccurate sanding is there for those who use jigs and fixtures, too. We all have to watch our sanding technique. I guess you could sand through the grits with your belt sander and platten but that introduces a lot of opportunity to mess up; as Mario said, that system takes a lot of tweaking and care in how it is used.

2) Choice of radius or compound radius: I guess you could have a set of templates or plattens (but doing a compound radius with a sanding platten looks very hard--perhaps practically impossible). So your fixture can get the radius the player prefers. So far so good. But (before I get to relief and drop off), there are a lot more shapes possible for a fretboard than a cylindrical or conic surface, where the board at any position is a portion of a circle. Have folks seen those twisted necks that are supposedly more ergonomic? That's an extreme, but it can be done with the fret tops in a straight line under the path of each string. Even with a conventional looking neck and board, it is not hard to see that there could be a lot of shapes (an infinite number) other than circular for the curvature of the board. As long as the same curve is repeated all the way down the board it could be parabolic, spline, catenary, or whatever shape you like, and run in a straight line under each string, with equal distance from the fret for each string at every position.

One way this plays out is in choosing more radical compounding. The basic compound board would be one in which the taper of the board and spread of strings determine the radii. If you have the radius at the nut and the taper, AND you wish to maintain the height of the edge (where the binding goes) along the whole length of the board, AND have the fret tops in a straight line under each string, there is a single compound conic surface that is possible (I'll call this the "ideal" compound board, although it may not be ideal for a player. It's the simplest way of compounding just to compensate for board taper and string spread). It took me a while to get this visualized, and being able to use a 3-D drawing program would help a lot, but here goes:

The thickness of the board is a variable that allows room to play with the "ideal" conic surface. For example, take the single radius cylindrical surface board, with constant thickness; i.e., the basic factory made board. If you take that shape and cut a tapered board, it's not too hard to see that the height of the binding edge is going to diminish as you go up to where the board is wider. You could keep the binding edge even in height, though, by having the board get thicker in its center as it goes wider.

Going the other direction from the "ideal" compound board, you could have more radical compounding, say 10" radius at the nut going to 20" at the 20th fret, and have the board be a conic surface, but you would have the binding edge getting taller toward the wider part of the board. Again, you could compensate by having the board get thinner in its center where it is wider if you wanted to keep the binding height constant.

In practice, what I (and I would guess nearly everyone else) do is start with a flat board of constant thickness. So if I want to have the board be a conic surface, and deviate from the "ideal", for example by a greater range of radii (I often go from 12" at the nut to 20" at the high end), the height of the binding could vary. But I am also making the board thinner in its center as it goes toward the high frets, which compensates for the variation in binding height. In the end, it is not noticeable unless one is looking for it.

This so far is just for conic surface boards with a circular curve at each position. Similar variables hold for other shapes. In practice, I sometimes round the edge of the board down a little under the low E, so that at a given fret the curvature is circular in the center, with the radius of curvature diminishing a bit toward the bass side edge. This does not mean that fret tops can't lie in a straight line under all the strings; they can if the same curve is repeated (with scaling for the variation in board width) at each fret. And it doesn't require a CNC machine to do that scaling and repeat the curve; I get it just by sanding the board lengthwise with a block, and checking it with a straightedge under the line of each string.

3) Relief and dropoff:

A lot has been said here and elsewhere about these. I'm not a fast typist and I need to get to work. So let's just leave it at these being variables that will change a board from describing a straight line under each string, which vary depending on playing style, and which are not going to get built into the board by any of the fixtures or jig that have been described so far. Maybe Bob Taylor can get a CNC machine set up to add relief and dropoff in varying amounts for each player. Those of us without six figures to spend on a fretboard machine have to do it by hand.

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

When all else fails, clean the shop.

I'd add that using a greater than "ideal" compounding together with diminishing board thickness is a way of getting drop off.

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

When all else fails, clean the shop.

IIRC, Mario does a single radius board. If so, the taper of the board
and spreading of the strings from nut to saddle makes his board an
inaccurate match to the strings right from the start

Wrong again. I've been a big proponent of correct compound radii for years. And getting it "right" is tricky enough to do once, but repeatedly requires accuracy. Or a phone call to Bob Garrish <bg>