Fretboard #1 fresh off the CNC!

Discussion in 'Tele Home Depot' started by Deed_Poll, Aug 19, 2019.

  1. Deed_Poll

    Deed_Poll Tele-Meister

    Posts:
    418
    Joined:
    Feb 8, 2013
    Location:
    Brighton, UK
    Hello all,

    I've been absent from these parts for a while now due to illness, but it's good to be back! Lengthy pre-amble warning ;)

    I've been working on designing a good process for building necks on my 2.5 axis CNC machine. I can go into more detail with the neck design but it's basically a Gibson style set neck but with a Fender type heel adjust single truss rod in an arched channel and a concealed skunk stripe which serves to align the fretboard with the shaft for gluing.

    The headstock is a non-tiltback Fender style design, but with the headstock face and back rounded over so that the front spills over onto the back, if that makes sense.

    My plan is to paint the headstock face, edges and back in black and that will turn into a kind of "stinger" at the back where it meets the neck profile with a fairly sharp volute.

    I designed the bodies so that you can access the rod adjustment without removing the strings, but the neck pickup will have to be taken out.

    The guitar design is like a cross between an early SG Jr, a Firebird and a Jazzmaster, with some Rickenbacker-like features in there for good measure. Here's a render of the design (let me know if you want more angles):

    [​IMG]

    I found some unusual looking Pau Ferro just over a year ago at my local instrument wood supplier. I quite like the even straight grain you typically find in Pau Ferro, but these boards had some wacky knots and shape to the grain that I thought was unusual, and it had some inky black streaks that reminded me of Dalmatian ebony or even Rio rosewood. The boards were going cheap, so I got all the interesting ones that were there - I think there were three or four, and I got them at a bargain price.

    When it came to testing my fretboard toolpath, knowing there were likely to be some issues (I have had to burn three neck shafts for that side of the project before I got it sorted!) I was unsure what to use. I wanted to use something that would behave properly like a fretboard wood, so I decided against practicing on plywood or something because there would be limits to what it might tell me. Besides, I was going to glue in some dots and a plywood fretboard seemed like a waste!

    So I dug out one of the old PF boards and had a go. Signs were not good when I planed it down to 7mm and had some pretty nasty tear-out along the knotty section and some chips and snipe at the ends. But my process involves crowning off about 1mm of thickness in the radius cut, so I put the dodgy face 'up' and threw caution to the wind, figuring I will at least learn something.

    The cut actually went better than expected! I think this one will actually be usable :)

    [​IMG]

    [​IMG]

    [​IMG]

    Sorry for the focus, but this is the only picture where you can see the iridescence of the MOP.

    You can see the tool marks from the way I cut the 12" radius with a 1/2" end mill

    [​IMG]

    [​IMG]

    I wish I'd taken more pictures throughout the process to show how I reference top to back, flip it over 4 times to cut certain features in the right order etc. to prevent tear-out, but I predicted it would be a disaster so I didn't! But I will do that next time now that I know the process works.

    Thanks for reading! I welcome your comments and questions on the whole process, the neck design, or guitar design in general.

    Dan
     
  2. backporchmusic

    backporchmusic Friend of Leo's

    Posts:
    4,129
    Joined:
    Nov 28, 2006
    Location:
    USA
    Cool design!
     
    crazydave911 and Deed_Poll like this.
  3. Deed_Poll

    Deed_Poll Tele-Meister

    Posts:
    418
    Joined:
    Feb 8, 2013
    Location:
    Brighton, UK
    Cheers! I've made a couple of bodies as bolt-ons before. I figured it would work better as a set neck though so I've redesigned some of the continuity and heel contours since to improve access and give it a more set-neck appearance.

    [​IMG]
    [​IMG]
    The bevelled edges of the body expand into the back of the neck at the joint, helping contour the heel; whilst the bevels around the top surface shrink accordingly so that the body edge has continuity with the side of the neck.

    [​IMG]
    [​IMG]
    [​IMG]
    [​IMG]

    Here's a bolt on body I made from korina that shows the contours better. I sculpted the forearm contour out from the top surface and kept the bevel the same thickness on the front, but on the back I made the belly contour by actually extending out the bevel surface in that area.
     
  4. Ducerro

    Ducerro Tele-Meister

    Age:
    63
    Posts:
    311
    Joined:
    Nov 20, 2017
    Location:
    Yoakum, Texas
    That neck is stunning!
     
    crazydave911 and Deed_Poll like this.
  5. Deed_Poll

    Deed_Poll Tele-Meister

    Posts:
    418
    Joined:
    Feb 8, 2013
    Location:
    Brighton, UK
    Thanks! That actual shaft had some things go wrong with it with the toolpath and the machine crashing, but I kept it hanging around to check the fit with the fretboard. Fingers crossed the next one I cut should have all the problems ironed out!

    Here is how that shaft looked straight off the machine, so lots of tool marks. The error happened with the roundover on the headstock, giving it a weird "German carve" look! But hopefully this will show what I'm talking about with the roundover etc.

    The back profile came out quite nicely and I later sanded it to 1500 just for giggles! I'm amazed how good of a gloss you can get on torrefied maple.

    [​IMG]
    [​IMG]
    Imagine the back of the headstock being black gloss, wrapping around the edges of the head onto the front and masking off the line where it meets the neck profile. That's the look I have in mind.
    [​IMG]
    This is the heel transition. I actually skipped a cut on this one as it was already ruined with the headstock going wrong, but it will be cut away at a 1.9° angle like this:
    [​IMG]
    But you can see there was an error and it took chunks out of the heel on this one! Anyway, that's the heel / neck pocket design so that the neck edges and body edges are continuous

    :)
     
    Last edited: Aug 19, 2019
  6. RodeoTex

    RodeoTex Poster Extraordinaire

    Posts:
    8,515
    Joined:
    Sep 14, 2005
    Location:
    Nueces Strip
    Really, really nice.
     
    Deed_Poll likes this.
  7. Sarde

    Sarde Tele-Meister

    Posts:
    156
    Joined:
    Apr 16, 2016
    Location:
    San Pedro, CA
    Thing of beauty!
     
    Deed_Poll likes this.
  8. Paul in Colorado

    Paul in Colorado Telefied Ad Free Member

    Posts:
    23,351
    Joined:
    Mar 17, 2003
    Location:
    Fort Collins, CO
    Nice stuff. I like the SG/Jaguar mash up vibe!
     
    Deed_Poll likes this.
  9. Jim_in_PA

    Jim_in_PA Tele-Meister Ad Free Member

    Posts:
    280
    Joined:
    May 31, 2019
    Location:
    SE PA near New Hope PA
    Totally awesome design!!

    It appears that you're running your finishing tool path (at least on the back of the neck) across the grain. Why in that direction as opposed to with the grain. (lengthwise) 'Just curious...
     
    Deed_Poll likes this.
  10. backporchmusic

    backporchmusic Friend of Leo's

    Posts:
    4,129
    Joined:
    Nov 28, 2006
    Location:
    USA
    Right up my alley. Unique yet familiar.
     
    rze99 and Deed_Poll like this.
  11. Deed_Poll

    Deed_Poll Tele-Meister

    Posts:
    418
    Joined:
    Feb 8, 2013
    Location:
    Brighton, UK
    Double post
     
  12. Deed_Poll

    Deed_Poll Tele-Meister

    Posts:
    418
    Joined:
    Feb 8, 2013
    Location:
    Brighton, UK
    A good question! There are several reasons, mostly unique to me and my setup. I'll start at the beginning.

    I'm using VCarve Pro as my CAM software, which is fairly decent and usable for most things but not really up to it and needs to be wrangled a bit to behave.

    Since it imports 3D models as a height map, it means the resolution of the cut (regardless of how good the mesh source is) drops away significantly the closer your surface gets to vertical.

    So the first thing I do is analyse my model's draft angle relative to the direction of cut in Rhino 3D. I set my bounds to basically draw a line along the surface between the region the angle exceeds 45° and the region it is below 45° to the top view.

    I then take this line and use that as the boundary of the cut, so that the "3D Cut" in VCarve isn't touching anything that's too steep.

    I take the other side of the line and turn it all into contours in .DXF format (2D vectors). Since this isn't going through VCarve's height map compression, these come out lovely and smooth and accurate, although I have to program them all in as separate profile toolpaths - and I have to base these contours on a surface offset of 6mm that is then sunken by 6mm, to account for the 12mm diameter ball end and tell it to follow the line, instead of going inside or outside (always a nightmare with open vectors anyway!)

    Because the contours are all equidistant in the Z axis (1mm apart), they get further apart from the point of view of the tool as they go up the model and the profile starts to flatten out. They also taper slightly as the neck tapers in width.

    After the contours are cut, you're left with a sort of mesa or "flat top" hairstyle - it's done the skirt around the edges so now it just needs to do the <45° section on top. But of course, we now have square edges all around the top, so tear-out becomes a concern.

    If you tell it to do the path longitudinally, it will be fine right up until the bit that's left gets towards the thin side. Let's say the 12mm diameter tool has 15mm more to cut. Each time it goes up and down the model, that "wall" that's left to cut gets thinner but it stays the same height. It will eventually snap off in dramatic fashion, and that long tough shard might get caught in the extraction or actually tug on the neck itself and pull it loose from the vacuum. This is what caused the "German carve" on the headstock where the released edge got sucked vertically into the extractor, got stuck, and wrenched the neck around ruining the piece.

    Due to the fact the back of the headstock and the heel of the neck are at "sea level", it's actually safer, when faced with the "mesa" or "flat top" or whatever you want to call it, to start at one end and bridge over to the other. It won't leave you with a "wall" to be snapped off, it will just remove a manageable small slice every time it ticks along. By the time it gets to the other end, the height of the slice will *gradually* be brought to zero.

    Additionally, because you are cutting a convex curve (neck) with another convex curve (tool) it is more efficient for the toolpath to follow the tightest convex radius. It will miss more wood for the same length of toolpath to do it longitudinally. If you imagine a round ended tool trying to cut a flat surface, it will look like a child's drawing of water. If you make that surface concave, it gets more accurate / smoother, up until the point the radius matches the tool at which point it's perfect. If you make it convex, however, the opposite happens - you need more and more passes (more stepover) to maintain the same smoothness as on a flat. By machining it against the grain, the surface I'm modelling is basically flat - it's up and down the neck. If you do it along the grain, your surface is quite convex (the neck profile). The machine can follow that surface perfectly with its movements and disregard it.

    I hope that is in some way understandable! I don't think I did the best job explaining.
     
  13. Guitar MD

    Guitar MD TDPRI Member

    Age:
    66
    Posts:
    35
    Joined:
    Nov 26, 2016
    Location:
    Brevard, NC
    Beautiful neck.
    Multiple personality trait guitar seems a detour for someone with these talents. Paint your own masterpiece.


    Sent from my iPad using Tapatalk
     
    Deed_Poll likes this.
  14. Deed_Poll

    Deed_Poll Tele-Meister

    Posts:
    418
    Joined:
    Feb 8, 2013
    Location:
    Brighton, UK
    Many thanks for the compliment! If I am understanding you correctly, you're interested in something a little more eclectic or unusual... I design that sort of thing as well but most people think they're ugly!

    [​IMG]

    It's just as well I get a lot of enjoyment focusing on the small details and using the wealth of history and meaning in the heritage of the electric guitar as well. I don't think I could ever make replicas, but I enjoy playing with the visual language and coming up with a novel take on an era, even if that isn't what you might call "revolutionary"!
     
  15. I_build_my_own

    I_build_my_own Friend of Leo's

    Posts:
    2,843
    Joined:
    Mar 9, 2012
    Location:
    New York
    Just Wow!!!! I admire your CNC skills!!
     
    Deed_Poll likes this.
  16. jvin248

    jvin248 Poster Extraordinaire

    Posts:
    7,591
    Joined:
    Apr 18, 2014
    Location:
    Lions & Tigers oh Mi !
    .

    You mentioned you flipped the part(s) four times to get all the details. How are you locating the parts for the flips? If drilling pins, how are you locating to the pins in the waste board consistently for multiple guitars or are you only doing one-off designs?

    Instead of removing the pickup to adjust the truss rod, make a window or notch in the end of the fretboard and drill cross holes in the truss rod nut (several 60s MIJ guitars did that) and is what I'm doing. Or buy a wheel style truss rod, which requires a wider window.

    .
     
    Deed_Poll likes this.
  17. Jim_in_PA

    Jim_in_PA Tele-Meister Ad Free Member

    Posts:
    280
    Joined:
    May 31, 2019
    Location:
    SE PA near New Hope PA
    Good response. I can appreciate the small limitations you mention relative to VCarve's finishing paths and it's good you have been able to develop some ways to get beyond them. I upgraded from VCarve Pro to Aspire in January to support a well-paying project for a client, but the tool path creation is the same...it's the same software. I've only cut a few necks, but with the expectation of some hand work afterward...which I'm fine with.
     
    Deed_Poll likes this.
  18. rze99

    rze99 Poster Extraordinaire

    Posts:
    7,004
    Joined:
    Feb 26, 2014
    Location:
    South London UK
    Terrific work!
     
    Deed_Poll likes this.
  19. Deed_Poll

    Deed_Poll Tele-Meister

    Posts:
    418
    Joined:
    Feb 8, 2013
    Location:
    Brighton, UK
    Good question!

    First answer is that I designed the bed to handle multiple different necks - all different scales as well, but only up to 26" scale so no bass necks. In my professional life I have been a parametric designer, mostly in packaging. That's where you design basically a "stretchy" model where you can feed in a few parameters and the design and its components will morph seamlessly to the new settings.

    I was actually too ambitious. Though I got the parametric neck model to work (so you can feed in scale length, fret count, back contour depth and taper, U-V shape % and taper, nut width, etc.) and the computer model updates, it has been such a pain switching to different software to have a macro take that model, extract the relevant contours and surfaces, save them in a proper format and import them on the right layers in toolpathing and so on, that my immediate plans are scaled down a bit.

    As a consequence, I could design the system more efficiently if I was making it again. For instance, the kind of "wall" surrounding the suction that is somewhat neck-shaped is actually as small as I predicted I would ever reasonably make a neck, in case I wanted to make some very short scales. You can make a bigger neck on a smaller one of these (with some overhang), but to make a smaller neck on a bigger one you would be cutting through the "wall".

    I will probably do three back contours on two or three neck scales and leave it at that for now until I get some more sophisticated toolpathing software or learn to use what I have to better effect.

    The locating pins are 2mm x 8mm steel, and my zero point shared with any neck I could throw into it is under the nut. This is a convenient location, because it will always be covered up. You can see the steel pin in the nut slot in the pictures.

    So there are 7 location pins in total, including the nut, always the same absolute coordinates from zero.

    There are a couple stretched out along the bass side, and these are mirrored on the treble side. Then there is a pair which straddle the truss rod route. The latter pair only feature on the back of the fretboard or the fretboard-side of the neck shaft.

    I make the fretboard and neck separately almost in their entirety before gluing them together. The three remaining pin holes would come in handy for this, but I designed my truss rod "concealed skunk stripe" (no idea what this should actually be called) to interface the two and align them, so that I don't have to entomb metal pins in the neck.

    As for the truss rod, do you mean an actual sort of viewing window in the fretboard itself? It's an interesting idea and would probably be more practical, but I like something about the vintage Fender style nuts. I was wondering about some kind of ratchet wrench system if I could conceal it under the guard and have a magnetic removable piece of pickguard material there, but haven't devoted much time to it yet :) one of the reasons that appealed is that it would prevent the need for me to chisel or file the inside corners of the pickguard square where it meets the neck.

    A couple of pictures of the set-up

    [​IMG]

    [​IMG]

    Hopefully you can see some of the 2mm peg holes in the picture above.

    [​IMG]

    This is a jig I made to mount necks in vertically, heel at the top, finely adjust their angle to the vertical and clamp them in place. I'll use it to cut the truss rod nut cavity at about 3.5°.

    [​IMG]
    I designed this wedge to interface at 1.9° with both the pin holes on the bed and on the piece to cut the neck pocket at the right angle.
    [​IMG]

    [​IMG]

    I use a bracing bar machined at the right angle and threaded inserts in the bed to clamp it down while I cut the neck pocket. The pins keep it in place side to side.

    Cheers
     
    Jim_in_PA likes this.
  20. Deed_Poll

    Deed_Poll Tele-Meister

    Posts:
    418
    Joined:
    Feb 8, 2013
    Location:
    Brighton, UK
    Yes, a lot of this stuff comes down to patience and familiarity! A lot of the time you're better off using what you've got. There are work-arounds for most things if you are devious! But it's still a more frustrating process than many would imagine - a lot of people have this view of CNC where you just press a button and it works! It couldn't be further from the truth! :D
     
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.


  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.