[pshupe]

Hello - total newb here. I model a lot in AutoCAD but mainly for architectural stuff which doesn't have a lot of curves.

My dad has recently purchased a CNC machine and I want to model up some guitar objects and try and machine them. I'm starting with a fretboard. I have drawn a planar curve that is the profile of the large end and extruded the planar curve to a solid. I have tapered the edges by drawing plines and extruding and then using the intersection tool to remove the tapered portion.

Now I am trying to use the same boolean function - intersection to subtract a solid from the fretboard to remove the fret slots. I have made one fret solid to be subtracted from the fretboard but it gives me a boolean error.

The funny thing is I have made the solid the same as I usually do, and used the boolean intersection command to remove the bottom portion so that the bottom of the fret is curved to the save radius as the top of the fret board. Then I move it down the necessary distance and use the boolean - intersection command again, and no dice. It gives me an error. Here is the file showing what I want to do. Thanks for any tips

Is the boolean command finicky? In Autocad it is straight forward and I can do this with one command and very quickly.

Cheers Peter.

PS - it's funny you cannot upload a 3dm file????

[EHNisja]

There's something going on with the fret board. I would rebuild it and try again. I ran the intersect command on it and then exploded it. It would trim with the curve but not join back into a good polysurface.

[Reyad]

there is some problem in connection in the big object
so you need to explode it then re join it it will work will. i tried it

[U2fletch]

Since you are making a fretboard have you considered the idea of starting with a conical shape? A guitar fretboard will have a compound radius if the strings truly follow the bridge and nut profile and all the frets in between. The strings will lie on the surface of a cone. The effective radius at the nut will be smaller than at the bridge.

Other than that however, Boolean subtraction can sometimes be finicky. Have you tried using the Trim tool instead? Trim seems to be less particular, although it does require a little more effort deleting the pieces you don't want.

[pshupe]

Thanks guys -

reyad - thanks for the tips I'll take a look at that. I am very new to Rhino so I can see I will have some challenges.

U2fletch - the fretboard I am making is for a Les Paul replica which does not have a conical radius. It is 12" radius consistently and just tapers each side.

Thanks again guys I will check the large fretboard solid and see if I can straighten it out.

Regards Peter.

[U2fletch]

Peter,

I can also confirm that if you just explode the piece and then rejoin, it will work just fine. Don't know why, but something internal gets reset and it's happy again.

Thanks for the clarification about the Les Paul, funny how some builders build flat, and others go for the conical. I never thought about it much until I tried to model a fretboard and started thinking about what is going on with the nut vs. bridge width and string height.

Have fun with the project!

Jeff

[pshupe]

Thanks Jeff - I tried exploding and joining it back again and it did not fix the problem?? Weird. I was spoiled by autocad, I guess. If the solid was made it was a solid. You cannot explode a solid in autocad and put back together.

Could you upload the fixed file?? Thanks a bunch.

Cheers Peter.

[U2fletch]

I took your file. Clicked on the fretboard. Clicked Explode. Then clicked Join. Then did the boolean subtract and it worked perfectly.

Here you go!

Jeff

[BurrMan]

Are you starting Rhino with a template?

[pshupe]

Thanks Jeff - It worked the second time I tried it, just following your instruction. What I really need to do is a couple of step process. I need to make a solid (fret slot size) then subtract that from the fret board. It seemed to work fine when I did it with the box, subtracting from the fret board, but this gave me a flat bottomed slot. I want a slot that has the same curve as the top of the fret board. So I subtracted the fret board from my box solid, and that seemed to work OK. Then I moved that object down the depth of the slot and tried to subtract that from the fret board and "boolean difference failed". I tried explode and join for both objects again with no benefit. Not sure why booleans are so difficult in Rhino. I did this in Autocad in about 10mins.

Burrman - no template used - just opened rhino used the small millimeters and started drawing.

Cheers Peter.

[U2fletch]

Here is a good description of why this does not work all the time.

rhino:booleanfaq · McNeel Wiki

[BurrMan]

You can fix the issue for yourself in the future by recognizing this... The length of the neck has a very "shallow" angle of curvature, in both directions. Your cutting object to get those edges were constructed of minimal density.

(This would be the equivelent of drawing a very long line of very shallow curvature, with only 3 points)

The "seperate and join" is hit and miss here. I made it work by deleting the side surfaces and rebuilding them, either with a network, or a sweep. The same rebuild works for the top and bottom. Either way, it's just getting the good trim curve definition for the part.

So, the original cutting object for this situation could be a more dense object to avoid the boolean failure.

[pshupe]

Thanks Jeff I'll take a look at that.

Thanks Burrman. - actually the length of the neck does not have any curvature. It is completely flat in the "x" direction, also the radius is consistently 12". It is like cutting the top from a long cylinder, then tapering it to be slimmer at one end. That is actually kind of how I modeled it. circle, depth from quad, then width and trim to make a planar curve, then extrude the length needed and then I tapered using booleans.

The curve (fret slot object) I used to subtract is the same curve that I'm trying to remove just moved down a fraction of an inch. One would think the defining points of these two curves would be the same this the Boolean operation should be simple?? I guess I have to, obviously, learn more about how Rhino works. Thanks again guys.

Cheers Peter.

[BurrMan]

Thanks Burrman. - actually the length of the neck does not have any curvature. It is completely flat in the "x" direction, also the radius is consistently 12". .

So the curvature with a minimalist definition I was speaking of is this one here along the side.



However that was cut there, the defining object should have been more dense. With such a length and shallow curvature, the surfaces are not well defined enough at the trim boundry join.

So if you extract the side surface of the fretboard and view it's points, you will see a point structure like the 4 red circles. The curvature is being defined with the 2 outermost points offset from tangent. If you delete those side surfaces and rebuild them with say, a network, you will see "more points" created to define that inner curvature (at the very least, the green)

I guess more points ls not really a good description. They have a more dense placement with regard to the curve.

So anyway, that was more the area I was talking about.

[pshupe]

Thanks Burrman - I had to re-read your post a couple of times. When you look straight down at the fret board, Plan View, it is made up of 4 straight lines. There is no curve on the top and bottom of the board, but I may still be not understanding your "curvature" assumption. The only curvature is the top of the board and it is a consistent 12" radius.

I am assuming the plane that makes up the side of the board could be described with 4 lines as it is really just a rectangle.

The next step for me is to subtract the fret slots, which I want to be curved like the top of the board. I made a planar curve and extruded it to a solid for the correct thickness. Now exploding and joining the board worked when I subtracted it from the solid, which was good. Then I moved the solid down the depth of the slot for the frets and tried to subtact again, and no dice. "Failed Boolean" again. I tried to explode and rejoin both, without any success. Not sure what is going on, it seems to be a fairly straight forward procedure here. Oh well I will try experimenting some more and see if I can figure it out.

Regards Peter.

[BurrMan]

Thanks Burrman - I had to re-read your post a couple of times. When you look straight down at the fret board, Plan View, it is made up of 4 straight lines. There is no curve on the top and bottom of the board, but I may still be not understanding your "curvature" assumption. The only curvature is the top of the board and it is a consistent 12" radius.

I am assuming the plane that makes up the side of the board could be described with 4 lines as it is really just a rectangle.
Regards Peter.

Hey Peter,
I made a quick video to show the curvature I was refering to.

Fretboard curvature - YouTube

Those sides are NOT rectangles, but very shallow 3d curved surfaces, with not enough definition to have the joins and booleans perform well. If you look in your original file in the top view and then shade that view, you will see those sidwalls. If you cranck up the mesh display settings and shade that top view and look down on those walls, you will see the bad trims too.

If you rebuild your solid fretboard in this way, you should not have an issue with your fret subtractions. Hope this helps to make it clear.

[pshupe]

Thanks Burrman - that completely clears it up. I really appreciate the video. I guess I'm a little dense! I have no idea how I ended up with that small curvature. It "should" be a straight line and hence a rectangle. I used a boolean to cut that piece out of there. I drew a line from intersection Thanks again I really appreciate the effort in taking the time to do that video.

This is supposed to be the easy part of the guitar.
I still have to figure out the neck transition piece and the carved top. I am really enjoying rhino. I find it a good combination of ACAD and MAX. Thanks again.

Cheers Peter.

[BurrMan]

I have no idea how I ended up with that small curvature. It "should" be a straight line and hence a rectangle. I used a boolean to cut that piece out of there. I drew a line from intersection

Remeber it's ok to have the shallow curve. It just needs more "definition" on the innards of the curve.

This is supposed to be the easy part of the guitar.
I still have to figure out the neck transition piece and the carved top. I am really enjoying rhino. I find it a good combination of ACAD and MAX. Thanks again.

Cheers Peter

LOL. Well, a guitar will be a good, more complex design, so when your done, you should be pretty good with it. Dont rush past any part that seems "sticky". Figure that part out.

Good luck.

[pshupe]

No - it has to be straight, is supposed to be straight. I'm not sure what happened there. Will start over and see how it goes.

Yeah - that is the plan - figure it all out. I, thought, I was pretty good with 3d. I am liking rhino it is much more precise than MAX for modeling.

Thanks a lot.

Cheers Peter.

[BurrMan]

sweep the neck - YouTube

I had some time to re-sweep the neck out...