Melody Maker build

[kwerk]

Originally Posted by flatfive

I worked out the error you'd accumulate if you measured
from fret to fret.

Roughly:

If, on average, 30% of the time your measurement
error is .01" or more, then 30% of the time your 22nd fret
location will be off by .05" or more.

(.01" is about a quarter of a mm)

Feeling lucky, punk?

Actually, that's not too bad. But the calculation assumes
that you're as likely to err on one side as the other.
So errors often cancel out. In real life I think we often err
more on one side than another.

The little bit of data I got suggests that it might be possible
to mark a fret position to less than .01" error 70% of the time.

If you want to use a number other than .01", just multiply it by
4.7 to get the second number (I rounded .047" to .05").

---- math stuff follows ----

This was derived by assuming errors are independent
and normally distributed. Assume the normal distribution
for each measurement has mean 0 and a standard
deviation of x. Then summing 22 independent random
variables over this distribution, you get a random variable
with mean 0 and a variance of 22 x^2, or a standard
deviation of 22^(1/2) x, which is about 4.7 x.
The 30% appears because samples from a normal
distribution lie outside one standard distribution about 32%
of the time.

I think I did that right...

Holy mackerel, Glenn!!




I gotta sit with Nick on this.

Ever since I saw one of these:



I feel a lot easier about marking with a mechanical pencil off a combo of calipers and steel rules. If accurate fretting means a different fret position on each string, there's no amount of accuracy that can get me perfect position from straight frets.

[Nick JD]

Originally Posted by flatfive

Does your router have a guide?



I use it to route trussrod slots in angled headstock necks.

[flatfive]

Hey there kwerk!

That is a cool neck, but you've got to ask yourself,
is it worth suffering vertigo to play in tune?

I figured there were a bunch of statistics experts
around here who were going to straighten me out on
my little endeavor, but I guess they don't get their
kicks from guitar making.

I think a combo of calipers and steel rules sounds a
pretty sound approach.

Originally Posted by kwerk

Holy mackerel, Glenn!!




I gotta sit with Nick on this.

Ever since I saw one of these:



I feel a lot easier about marking with a mechanical pencil off a combo of calipers and steel rules. If accurate fretting means a different fret position on each string, there's no amount of accuracy that can get me perfect position from straight frets.

[flatfive]

I don't have a guide for my router. It's a PC 690, so they
must be available. Perhaps I'll learn the value of router
guides soon, just as I learned the value of fences this year.
So thanks for pointing that out.

On the other hand, I already have my jig and it does make
routing the slot totally foolproof.

Originally Posted by Nick JD

Does your router have a guide?
...
I use it to route trussrod slots in angled headstock necks.

[flatfive]

We've got a man-vs-nature scenario here, where I'm
trying to keep working (on the front porch) as the
weather gets nastier and nastier.

To try to protect the wood from temperature changes,
I'm getting stuff ready, taking the wood outside,
and then rushing it back in after it's been operated
on.

Anyway, after getting the truss rod jig sorted out
yesterday, it turned out not to be suitable.
The problem was that the neck isn't uniformly
thick, so the truss rod slot was sloped at the bottom.

I ended up using the router table with a fence.



I marked the back of the neck and the fence so I'd know the
location of the slot end.



After slotting:



Then screwed the neck template on and routed the sides.



Did the same thing with the headstock template:



You can see a little rough spot at the side of the neck
where the headstock and neck routs met. I'll clean this
up after shaping the back of the neck.

With care and a sharp router bit the end grain at the top
of the headstock was smooth.



(It might look like a little tear occurred but that's just a
reflection.)

Things are going okay. Not sure what to do next -- either
figure out and make the body routs, or do the drilling of the
truss rod access hole.

[Mojotron]

Originally Posted by flatfive

...
Very roughly: the error grows as the square root of the
number of measurements.
...

Depends on the Beer involved

[guitarbuilder]

My filler strips tend to be thin near the nut too.

[flatfive]

Originally Posted by Mojotron

Depends on the Beer involved

Hi Mojo. You thinking of that related statistics problem,
the random walk?

[flatfive]

Thanks for mentioning this. I'm still trying to figure out
how to work the area around the nut and access hole.

In some builds I've seen a fillet about 2" long installed around
the nut area, and then a hole drilled from the headstock
for the adjustment nut.

Why not instead drill the hole first and then install a single
fillet strip along the length of the truss rod slot? Is it
so the access hole will be neatly round all around the
adjustment nut?

Originally Posted by guitarbuilder

My filler strips tend to be thin near the nut too.

[Mojotron]

Originally Posted by flatfive

Hi Mojo. You thinking of that related statistics problem,
the random walk?

Yes - sorry it was Saturday night and brewsky was on my mind

I like the thoughts on successive approximation - it's worth putting a lot of thought into - anyway laying out measurements is done it's always been my experience that one can never be careful enough if one wants to play extended chords above the 12th fret to have that 'ring' that really well made guitars seem to have.

I'm learning from this thread and taking notes...

[guitarbuilder]

If you look at the typical gibson neck, there is about 3/16-1/4" of neck wood/filler strip behind the nut before the cavity flat end starts ( where the washer sits). This allows the small rod nut cover screw someplace to go.

I've done the cavity a few different ways, using a roundnose bit right after cutting the truss rod slot while the neck is centered on the fence.

I've done it by hand after everything is glued on with a chisel and xacto knife.

I even made a boring jig for my hand drill where a forstner bit mounted in my portable drill ran on a slot into the peghead. None was better than the other really.

I've recently just did a couple necks where I took a 1/2 D forstner bit and drilled down some in front of the rod. Then I took a chisel and made a V shaped cavity up to the end. Then I have this curved carving tool from xacto that is really sharp, and I just clean everything up. A sharp utility style knife cuts the fibers down by the washer.

Probably the best thing for me now, would be to use the jig I made to drill tele 3 degree adj. rod holes, which I made after the gibson necks were done. For that jig, I have a 3 degree wedge that I can remove so I can try gibson adj rod cavities.

Regarding the thin filler strip... that means you have more wood under the nut which is a good thing for a gibson style neck :-).

[flatfive]

Hi Mojo. Yeah, the thing that I still don't have a feeling for
is: how much do small differences in fret placement make in
the sound and feeling of a guitar?

One thing I realized is that some guitars continue to sound
good after significant fret wear, and fret wear can change
the effective fret location by ... what, maybe about 1/2
of the fret width, or about .04".

Originally Posted by Mojotron

...
I like the thoughts on successive approximation - it's worth putting a lot of thought into - anyway laying out measurements is done it's always been my experience that one can never be careful enough if one wants to play extended chords above the 12th fret to have that 'ring' that really well made guitars seem to have.

Hope you're learning more than I'm a greenhorn.

Originally Posted by Mojotron

...I'm learning from this thread and taking notes...

[flatfive]

Thanks a lot for taking the time to explain this. Do you have
any build threads here or elsewhere showing some of these
steps? Maybe I should spend more time looking at threads
on Luthier's Corner on mylespaul.com.

Originally Posted by guitarbuilder

If you look at the typical gibson neck, there is about 3/16-1/4" of neck wood/filler strip behind the nut before the cavity flat end starts ( where the washer sits). This allows the small rod nut cover screw someplace to go.

I've done the cavity a few different ways, using a roundnose bit right after cutting the truss rod slot while the neck is centered on the fence.

I've done it by hand after everything is glued on with a chisel and xacto knife.

I even made a boring jig for my hand drill where a forstner bit mounted in my portable drill ran on a slot into the peghead. None was better than the other really.

I've recently just did a couple necks where I took a 1/2 D forstner bit and drilled down some in front of the rod. Then I took a chisel and made a V shaped cavity up to the end. Then I have this curved carving tool from xacto that is really sharp, and I just clean everything up. A sharp utility style knife cuts the fibers down by the washer.

Probably the best thing for me now, would be to use the jig I made to drill tele 3 degree adj. rod holes, which I made after the gibson necks were done. For that jig, I have a 3 degree wedge that I can remove so I can try gibson adj rod cavities.

Regarding the thin filler strip... that means you have more wood under the nut which is a good thing for a gibson style neck :-).

[flatfive]

I made some progress today but also made a blunder.

First I shaped the notch on the top of the headstock.



For that I used a small triangular file, a small piece of MDF
with some sandpaper glued on one surface, and other little
bits of sandpaper.

Decided to do the control cavity while thinking about the
neck stuff. First step was to cut out the cavity area of
the plan and to trace the shape on the body and on a piece
of 1/2" MDF.



I made the MDF template, then taped it to the body, aligned
with the tracing made earlier. I used Duck double-sided
tape -- that stuff holds super tight.

Routed the cavity after hogging out material with a forstner bit.



Oh - earlier I mentioned a blunder. After hogging out material
with the forstner bit, I realized that I was working with a symmetrical
body and hadn't considered which was the better side to be the
top.

I flipped the body over and saw the two holes that I'd drilled to
hold the body routing template. They were placed so that
the pickup routs would remove them.

This is crummy -- the wood is really nice. I'm determined to do
a very good job of filling those holes.

Last step today was to prepare for drilling the tuner holes. I cut
the headstock part of the plan out and was planning to tape it
to the headstock and drill.

Before doing this I got the tuners out and recalled that
the tuners on each side are on a "stick". The tuner post
diameter is 1/4" -- there is no way that I can drill 3 holes,
about an inch apart, such that the 1/4" posts will perfectly
align with the holes I drilled.



I guess I'll make a drilling guide for this step. But also, I guess
I'll want to use slightly oversized tuner post holes, either
17/64" or even 9/32".

Any thoughts on this?

[kwerk]

Originally Posted by flatfive

I made some progress today but also made a blunder.

First I shaped the notch on the top of the headstock.



For that I used a small triangular file, a small piece of MDF
with some sandpaper glued on one surface, and other little
bits of sandpaper.

Beautiful job, Glenn. That's some precision work.

[guitarbuilder]

Do you haveany build threads here or elsewhere showing some of these
steps?

No, it's a tele forum...LOL. I just have a few pix of the necks here and there occasionally in other threads. Your triangular file and sandpaper step is exactly how I do mine as well. IT is a very nice looking moustache there.

Regarding the holes for the tuners. I clamp a fence on the drill press table for the distance from the center of the bit to the edge of the peghead. Using an accurate scale and square, layout the distance between holes and from the edge of the peghead. It should be 2 -3/4 from outer peg to outer peg, and 1 - 3/8 to the middle if I recall correctly. I use a scribe to make a dent in the spot for drilling and use a brad point bit to drill through into scrap underneath. The scrap is for breakout protection. The fence keeps the holes in line. I always make the holes 1/64 larger than the diameter of the posts,and once in a while I'll open them up one more size if they are still tight. You should have ferrules for the top side right?
Try it on scrap first.

[crazydave911]

Not to mention, IIRC, those tuners have the little pressed sheetmetal bushings on the top similar to the Gibson L-O (they even look like the same kind of tuners)

[guitarbuilder]

pressed sheetmetal bushings aka ferrules? :-)......bushings sound more mechanical....I like that.

[crazydave911]

Originally Posted by guitarbuilder

pressed sheetmetal bushings aka ferrules? :-)......bushings sound more mechanical....I like that.

, yup

[flatfive]

Originally Posted by kwerk

Beautiful job, Glenn. That's some precision work.

Thanks, kwerk. My ability to do things like this improved a
lot after getting these.