One approach to modeling a neck for the cnc machine

Apparently there isn't too much out there yet on neck modeling of a gibson style neck, so I guess I'll share what I've done per a request I got from a new forum member. If you are attempting a set neck or even a tele neck, the steps apply to that as well. I'll go step by step and you can follow along at a moderately slow pace if you want. You need a drawing program that does 3D. I like Rhino and will be using that. My program is on my other computer which is not attached to the internet, so I have to switch back and forth. I'm a self taught Rhino user and so there may be an easier way to do things, but this works for me. A couple thanks are necessary for some help from the guys at the OLF forum, in the cnc section who helped me out to do this in the beginning when I was struggling. I still struggle...but not as much. :-).

Here is the Gibson scale length and fret calculations that I use from the stew mac fret calculator:
http://i152.photobucket.com/albums/s162/mmgguitars/gibsonscalefromstewmac.png

We will start with the fretboard portion of the model. I'll pick a nut width of 1 and 11/16 ( 1.6875") and the end of the fretboard will have a width of 2.25". The thickness of the fretboard will be .25". I'll use the standard gibson 12" fretboard radius too.

When I draw in Rhino, I like to keep the centerline of the part right on the 0,0 point and draw the length in the positive Y direction. A lot of times you can " mirror" the lines to generate symmetry.

If you are going to follow along and draw, I like to go out to 3 places with my measurements and work in inches....:-).

I open up Rhino and click on the Top view quadrant of the screen. My grid spacing is like 1/4"graph paper, a line every .25" and a dark line every inch.

I start by drawing a line from 0,0 the length of the fretboard. I will use fret 23 as the length of the fretboard although there won't be a fret there. That measurement is 18.056".

I click on my polyline tool and start the line at 0,0 and then type in 18.056 for the end point of the line. That is the blue line. Then at the bottom of the line, I draw a second line 1.125" long to the left. At the top of the line I draw a line .844" ( half of 1.6875) I connect those two lines with a 3rd line that shows the taper of the fretboard from top to bottom.


http://i152.photobucket.com/albums/s162/mmgguitars/model2.jpg

Then I click on those 3 black lines and go to transform/Mirror and make a duplicate on the right side of the centerline. You can just draw them if you are in a regular drawing program too.
http://i152.photobucket.com/albums/s162/mmgguitars/model3.jpg

So that is the outline of the fretboard.

I now will put a surface on those lines. In Rhino, lines are called curves...even straight ones..... and planar curves are all in the same plane. I click on Surface and drag down to planar curves. This creates a surface from those lines. They remain highlighted and I delete the curves.

This is the result:
http://i152.photobucket.com/albums/s162/mmgguitars/model4.jpg

And here is a Rhino screen shot showing the views.
http://i152.photobucket.com/albums/s162/mmgguitars/model5.jpg

Now the fingerboard radius. I placed 3 points at the ends and middle of the end line that represents the width of the fret board. The middle line is a line I drew up .25".

http://i152.photobucket.com/albums/s162/mmgguitars/model75.jpg
I went to curve/arc/ start,end, "radius". I clicked on the right and left points and entered the number 12 for the radius. That produced the curve in red.

http://i152.photobucket.com/albums/s162/mmgguitars/model9.jpg
Then I moved the curve upward .25" which is the thickness of the fretboard.

http://i152.photobucket.com/albums/s162/mmgguitars/model10.jpg

Then I copied that arc and placed it at the other end of the "fretboard" and drew lines up from the corners to meet the arc.

http://i152.photobucket.com/albums/s162/mmgguitars/model11.jpg
And finally I trimmed the arc to the nut width at those lines.

http://i152.photobucket.com/albums/s162/mmgguitars/model12.jpg

Guitarbuilder:

Depending on the orientation of the CNC bed, I probably would have started horizontally, going in negative directions towards heel. (Not that it matters yet, but makes it more like we're used to seeing the drawings posted every where.) Then when ready to mill make the 0,0 (home) at the nut center-line, and rotate 180 degrees (?).

I am fiddling with the design scenario as I write, and the Booleans for the top fretboard arced surface have a bit of complexity. I had to create a cylinder with diameter of 20" and make it longer than the fretboard, put it tangent to the origin, then do a boolean difference, but it failed. So I created a curve from intersection of the two solids, and made it only 4 sides, and created a surface from it to join to the hollow fretboard form. Fun, eh.?131208

Lots of ways to skin the old CAD CAT.

Guitarbuilder:

Depending on the orientation of the CNC bed, I probably would have started horizontally, going in negative directions towards heel. (Not that it matters yet, but makes it more like we're used to seeing the drawings posted every where.) Then when ready to mill make the 0,0 (home) at the nut center-line, and rotate 180 degrees (?).

I am fiddling with the design scenario as I write, and the Booleans for the top fretboard arced surface have a bit of complexity. I had to create a cylinder with diameter of 20" and make it longer than the fretboard, put it tangent to the origin, then do a boolean difference, but it failed. So I created a curve from intersection of the two solids, and made it only 4 sides, and created a surface from it to join to the hollow fretboard form. Fun, eh.?

My long axis on the CNC is my Yaxis. I usually draw in the direction that I expect to see it on the machine. I just like the consistancy of it. You have to do what works for you....:-). I've found that since I don't know as much about Rhino as I'd like that a stick built approach is easily understandable and produces a nice result. It's more linear and elementary I guess. This came about at failed booleans more than I'd care to admit.

Subscribed and watching. I want to see how you tackle the carved portions of the neck. Organic shapes like those are tough. Ask me how I know. :wink:

This type of Boolean works better when the geometry you're using as a cutter is outside the geometry you want to keep...a subtraction Boolean is used. Instead of using a cylinder as a cutter which would leave the fretboard trapped inside the cutter, use an object that looks like a long radius block which would leave the fretboard outside the cutter. Should work every time. Merging points after Boolean and Mirror is also a good practice to ensure you don't have multiple points occupying the same location.

Here I connected the corners of the arcs together to make a frame and then

I surfaced the top radius. To do this I clicked on Surface/Curve network.

Then selected the lines that make up the surface and hit enter. You can still see the red lines of the framework. It is important in Rhino to delete these as you go along. You can always go back and create the edges of the surface if you need to by using another command.



http://i152.photobucket.com/albums/s162/mmgguitars/model17.jpg

I did the remaining sides the same way. Essentially this is a hollow fretboard.



http://i152.photobucket.com/albums/s162/mmgguitars/model18.jpg


http://i152.photobucket.com/albums/s162/mmgguitars/model19.jpg


You could create the rectangle, extrude and cut away the sections of the fretboard and radius.... The end result is the same to the CNC machine which only sees the surfaces.

My long axis on the CNC is my Yaxis. I usually draw in the direction that I expect to see it on the machine. I just like the consistancy of it. You have to do what works for you....:-). I've found that since I don't know as much about Rhino as I'd like that a stick built approach is easily understandable and produces a nice result. It's more linear and elementary I guess. This came about at failed booleans more than I'd care to admit.

Looking forward to this. I can tell already that your approach differs a lot from mine so I'm sure I will learn from this. But in general I subscribe to your idea of breaking the design down into elementary shapes.

One thing I do differently is have my fretboard plane overhang all the fretboard edges by 1/4" or so. That serves two purposes - if I make the neck design wider, longer, etc., I don't need to rebuild the fretboard toolpath. It also allows the ballnose cutter to start out in thin air, and work gently into the side of the board (toolpath parallel to the length of the neck) and work completely off all ends of the board, so you don't end up with a tiny lip at the edges that you can sometimes get when your cutting plane is contained within the part.

Looking forward to this. I can tell already that your approach differs a lot from mine so I'm sure I will learn from this. But in general I subscribe to your idea of breaking the design down into elementary shapes.

One thing I do differently is have my fretboard plane overhang all the fretboard edges by 1/4" or so. That serves two purposes - if I make the neck design wider, longer, etc., I don't need to rebuild the fretboard toolpath. It also allows the ballnose cutter to start out in thin air, and work gently into the side of the board (toolpath parallel to the length of the neck) and work completely off all ends of the board, so you don't end up with a tiny lip at the edges that you can sometimes get when your cutting plane is contained within the part.

My method is the results of trial and error...a lot of error. As I said, it works for me and won't be for everybody. Once I get something that works I stick with it for a long time.

My method is the results of trial and error...a lot of error. As I said, it works for me and won't be for everybody. Once I get something that works I stick with it for a long time.

I can definitely relate...people think the CNC is a shortcut. If they only knew what is involved when you are trying to figure this stuff out on your own...

Subscribed and watching. I want to see how you tackle the carved portions of the neck. Organic shapes like those are tough. Ask me how I know. :wink:

This type of Boolean works better when the geometry you're using as a cutter is outside the geometry you want to keep...a subtraction Boolean is used. Instead of using a cylinder as a cutter which would leave the fretboard trapped inside the cutter, use an object that looks like a long radius block which would leave the fretboard outside the cutter. Should work every time. Merging points after Boolean and Mirror is also a good practice to ensure you don't have multiple points occupying the same location.

I've found that in order to get boolean to work more often, the surfaces have to be distinctly overlapping each other. Surfaces with common points don't like to boolean for some reason. Also, you have to get rid of all points, curves, surfaces, and things not related to the boolean. I've had more luck just splitting objects and then deleting what I don't want.

Subscribed and watching. I want to see how you tackle the carved portions of the neck. Organic shapes like those are tough. Ask me how I know. :wink:

This type of Boolean works better when the geometry you're using as a cutter is outside the geometry you want to keep...a subtraction Boolean is used. Instead of using a cylinder as a cutter which would leave the fretboard trapped inside the cutter, use an object that looks like a long radius block which would leave the fretboard outside the cutter. Should work every time. Merging points after Boolean and Mirror is also a good practice to ensure you don't have multiple points occupying the same location.

AH - Now I see. Good point about the difference (subtraction) booleans. I'll have to remember that one.

BTW - I just started working as a CAD designer for a major appliance manufacturer, and I am training (mandatory for all new employees) on CATIA v5 and it is the big dog in auto, aero, and manufacturing industries. But Rhino has better surface modeling techniques and tools (at least what I've seen so far).

Here I connected the corners of the arcs together to make a frame and then

I surfaced the top radius. To do this I clicked on Surface/Curve network.

Then selected the lines that make up the surface and hit enter. You can still see the red lines of the framework. It is important in Rhino to delete these as you go along. You can always go back and create the edges of the surface if you need to by using another command.



http://i152.photobucket.com/albums/s162/mmgguitars/model17.jpg

I did the remaining sides the same way. Essentially this is a hollow fretboard.



http://i152.photobucket.com/albums/s162/mmgguitars/model18.jpg


http://i152.photobucket.com/albums/s162/mmgguitars/model19.jpg


You could create the rectangle, extrude and cut away the sections of the fretboard and radius.... The end result is the same to the CNC machine which only sees the surfaces.

Not that it matters for CNC router or milling, but do you check for naked edges? That really matters for RP (rapid prototyping - 3D printing or growing). I've had great success with CNC milling, but the same model would have multiple naked edges that refuse to be grown. Just curious.

After my surfaces are created( usually using the neighboring surface edges that are in common), I check for and delete everything else.

OK Let's get back to it today a bit. I asked the forum member what kind of LP neck he was interested in and I didn't get a response yet, so I will do the standard lp neck with the 16th fret joint.
In Rhino, using the centerline I drew a rectangle for the peghead material.
I think it is 4" x 8". I also drew a 1.5" wide by 3" long tenon.

I drew in the nut of 3/16"wide by 1 11/16" long.

I measured down the distance to the 16th fret. It rounded up to 14.82 on the dimension itself, but was the distance from the stew mac fret calculation.

I also added a line under the fretboard edges, since I will remove the fretboard from the drawing in the next step. You can see the tenon lies within the confines of the fretboard. This is the more modern tenon, if you wanted the long tenon which extends into the neck pickup cavity, you can draw it longer if you wish. The steps are the same. All these lines are projected to the cutting plane, so they are all at the same level in the other views.

http://i152.photobucket.com/albums/s162/mmgguitars/shaft1.jpg

I took all the individual lines and made them "closed curves" by connecting them together into rectangles. I clicked on each individual rectangle and hit the surface "planar curves" button. That creates a surface within the rectangle. I deleted the rectangles leaving just the surfaces... I probably will go back and add a few lines later here and there.
http://i152.photobucket.com/albums/s162/mmgguitars/shaft2.jpg

Next I tilt the peghead 17 degrees. I click on the peghead surface and then go to Transform/Rotate. The center of rotation is the peghead edge of the nut and I place the end of rotation at the far end of the peghead surface. I enter -17 and it rotates the surface to look like this.
http://i152.photobucket.com/albums/s162/mmgguitars/shaft3.jpg

I want the peghead to be 9/16" thick... why...why not.... I click on the surface and go to "Solid"...extrude surface....straight...and enter -.562. The surface is extruded into a solid and I delete the original surface.
http://i152.photobucket.com/albums/s162/mmgguitars/shaft4.jpg

I did the same thing with the surface for the tenon and extruded it 1.5" down.
http://i152.photobucket.com/albums/s162/mmgguitars/shaft5.jpg

Now I am drawing some lines in to form a skeletal framework for the transition from neck shaft to tenon. I drew lines down from the shaft corners 1 -5/8" long on both sides and then toward the nut, two more lines that are 3/4" long. I connected those line ends together with a line. I rotated the isometric drawing so you can see the lines in red better.
http://i152.photobucket.com/albums/s162/mmgguitars/shaft6.jpg

Here I am placing 3 points in which to draw the curve of the heel. I think the position is self explanatory from the picture.

http://i152.photobucket.com/albums/s162/mmgguitars/Shaft7.jpg
I use the curve/arc/start- point- end command to draw in the arc and a line connecting the end of the arc points.

http://i152.photobucket.com/albums/s162/mmgguitars/shaft8.jpg
I bisect the arc from its center back to that line.

http://i152.photobucket.com/albums/s162/mmgguitars/shaft9.jpg

I delete the red lines leaving the heel radius. If you wanted a different shape for say a '59 burst clone, the steps are the same but the heel curve would look different. I'm not trying to make a copy.

http://i152.photobucket.com/albums/s162/mmgguitars/shaft10.jpg

I drew a line down at the centerline where it meets the nut edge a distance of 5/8". This is in white.
I drew a line down at the centerline where it meets the 16th fret a distance of 3/4". This is also in white. These are arbitrary numbers and you can decide how thick you want the neck to be and use different numbers. I added a little short extention to them which will be deleted later.

http://i152.photobucket.com/albums/s162/mmgguitars/shaft11.jpg
I connect the bottom of the lines and this produces a tapered line.

http://i152.photobucket.com/albums/s162/mmgguitars/shaft12.jpg
Here I add in transition curves at the heel and the the peghead from that tapered line.
http://i152.photobucket.com/albums/s162/mmgguitars/shaft13.jpg
I delete the parts of the tapered line showing the curve of the neck.
http://i152.photobucket.com/albums/s162/mmgguitars/shaft14.jpg
Lastly I placed two points at the end of the straight line. This is where I will draw my cross sectional curves for the general shape and feel of the neck.
http://i152.photobucket.com/albums/s162/mmgguitars/shaft16.jpg

guitarbuilder, thanks for ALL your contributions to the forum. I don't own a CNC machine, but this is interesting and informative nonetheless!

On to the peghead. I drew a curve to approximate the transition from the shaft to the peghead. I mirrored it on the other side.

http://i152.photobucket.com/albums/s162/mmgguitars/shaft17.jpg
I extruded the curve in both directions to make a surface through the peghead rectangle.
http://i152.photobucket.com/albums/s162/mmgguitars/shaft18.jpg
I split the peghead from those two surfaces.
http://i152.photobucket.com/albums/s162/mmgguitars/shaft19.jpg
I deleted the " waste" of the peghead.
]http://i152.photobucket.com/albums/s162/mmgguitars/shaft20.jpg
Then I exploded the peghead surface into its individual pieces and deleted the bottom or back side of the peghead.

http://i152.photobucket.com/albums/s162/mmgguitars/shaft21.jpg

Now to start working on the surfaces of the neck shaft, heel and peghead. I first rotated the neck upside down at the 0,0 location. I also extended the white centerline up to the top of the peghead.


http://i152.photobucket.com/albums/s162/mmgguitars/rotated-1.jpg

I added a curve to both sides of the heel to help with the surface there. Rhino has more difficulty with really complex curves and my limited knowlege, so I've found that it is easier to simplify some tasks...many tasks by making the surfaces easier for the software to create.

http://i152.photobucket.com/albums/s162/mmgguitars/surface5.jpg

To surface I click on curve network and click on the curves or surface edges I want to include in that surface. One important thing I learned, perhaps I think from somebody here...(Mgdesigns?) was that Rhino will have an easier time creating surfaces if the curves you are using have the same numbers of points. So rebuild curves and use the same number of points and you will have a much easier time with this. Some of the curves I split in two by the centerlines. Some surfaces got mirrored to create the other sides. I just did it section by section.


http://i152.photobucket.com/albums/s162/mmgguitars/surface7.jpg

http://i152.photobucket.com/albums/s162/mmgguitars/surface7.jpg

http://i152.photobucket.com/albums/s162/mmgguitars/surface10.jpg

http://i152.photobucket.com/albums/s162/mmgguitars/surface11.jpg


http://i152.photobucket.com/albums/s162/mmgguitars/surface12.jpg
For the peghead surface, I added a line across the transitions to again make it easier to create the surface. I also deleted the fretboard surface that existed for a while.



http://i152.photobucket.com/albums/s162/mmgguitars/surface13.jpg

http://i152.photobucket.com/albums/s162/mmgguitars/surface14.jpg

I lastly deleted any curves and points leaving just surfaces and here is the neck shaded in.

Things I didn't do here.... I didn't include a 4 degree angle in the tenon area to make the neck pitch and I didn't create the Gibson peghead. That of course would be copyright infringement.



http://i152.photobucket.com/albums/s162/mmgguitars/surface15.jpg

So there you have it, one approach to modeling a set neck for a les paul style guitar. You can manipulate any of the variables you want. I'm sure that there is an easier way to do some of these things...but when you are self taught and have no manual, then trial and error problem solving methods that produce results work for me....

Thanks for putting this together, I have access to Solidworks but not well versed in it yet. This tutorial should help with my struggles with Solidworks, but may have to look into Rhino as well.

THANKS

Nice!

Man, Soooo Cool, but that gave me a headache!

You want a headache, try learning Rhino on your own....LOL.

http://i152.photobucket.com/albums/s162/mmgguitars/surface11.jpg


http://i152.photobucket.com/albums/s162/mmgguitars/surface12.jpg
For the peghead surface, I added a line across the transitions to again make it easier to create the surface. I also deleted the fretboard surface that existed for a while.



http://i152.photobucket.com/albums/s162/mmgguitars/surface13.jpg

http://i152.photobucket.com/albums/s162/mmgguitars/surface14.jpg

I lastly deleted any curves and points leaving just surfaces and here is the neck shaded in.

Things I didn't do here.... I didn't include a 4 degree angle in the tenon area to make the neck pitch and I didn't create the Gibson peghead. That of course would be copyright infringement.



http://i152.photobucket.com/albums/s162/mmgguitars/surface15.jpg

Marty: You've done a really informative tutorial of the Rhino Neck designing experience. Lots of ways to skin a CAD Cat, and yours looks like the best. I have done a few necks, but the transitions on yours are great. Thanks. I will have to go back over my Airline neck, and follow your technique to get the transitions correct. The hardest part for me is checking the watertight analysis (a must if you ever have to grow something - like stereolithography or 3D printing). For the CNC mills, that's not a problem as long as the tolerances of the cracks are much smaller than the cutter being used. Really looks great.

And Rhino is one of the most versatile CAD programs around. I've been using CATIA for 3 months now, and it is tough to learn. Solidworks was a piece of cake in comparison.

Mark,
One of the drawbacks of not having a manual in front of me or any classroom training (and I'm not willing to print it out on my inkjet) is that I end up doing" trial and error" to get it to do what I want. This limits Rhino's power to some degree because I know I'm missing out on some important stuff. If I wanted to model a more realistic LP top for example, I'd be pulling my limited amount of hair out. Googling youtube vids helps once in a while, but really we need a Rhino for Dummies book. One of he best things I learned was when you told me to rebuild the curves with the same points to create surfaces. That was an immense help.

One thing that I never figured out and you might know, is how to keep things inline when you are copying and pasting. It always wants to move up or down in other views. Do you hold the control or shift key or some other key while you are pasting to keep it in place? I hope that question makes some sense.

Sometimes I think a CNC might complicate things. But if you could download the 3D model, it'd be easier.

http://www.cnclutherie.com/

Mark,
One of the drawbacks of not having a manual in front of me or any classroom training (and I'm not willing to print it out on my inkjet) is that I end up doing" trial and error" to get it to do what I want. This limits Rhino's power to some degree because I know I'm missing out on some important stuff. If I wanted to model a more realistic LP top for example, I'd be pulling my limited amount of hair out. Googling youtube vids helps once in a while, but really we need a Rhino for Dummies book. One of he best things I learned was when you told me to rebuild the curves with the same points to create surfaces. That was an immense help.

One thing that I never figured out and you might know, is how to keep things inline when you are copying and pasting. It always wants to move up or down in other views. Do you hold the control or shift key or some other key while you are pasting to keep it in place? I hope that question makes some sense.

Marty: You are doing great. I just tried to re-create the Airline Coronado Neck, and I am bumfuzzled. As for creating indexable paste ins - create a block and save it as a file with an origin point, then paste in (insert block) and specify the point you wish to insert and angle to insert. Very similar to WBlock in AutoCad. This allows greater repeatability with fixes. I work out individual components the insert into the whole file. If I want to correct something, I go back to the sub-assembly file, fix it, and then back to the assembly file and insert from file. As soon as you hit enter the old BLOCK gets re-associated to the new, and you delete the new one and keep on going from there. Remember, if you design in one viewport as the plan, stick with it or you'll be rotating the object in 3D space to get it re-oriented. I hope this helps.

My Airline build up is done at a component level, then pasted into the whole assembly. That way I can edit or add new components as needed to make it better.

I don't know if you are understanding what I'm asking. If I copy something in the top view and paste it in the front view, the X measurement may be correct, but it pastes maybe higher or lower in the Z. Then I have to go to the other view and drag it back to the position I want it.

Another way to look at it is if I copy and paste something in the top view and drag the copy back a certain measurement, in the Front view, it isn't in line, but has dropped down some amount of measurement and I have to drag it back up. Sometimes there is too much manipulation in my opinion to get it into its spot. I'm thinking that there must be a simple key command to hold down while doing this to keep things lined up during copying but I don't know what it is yet :-).