You're right in that one of the nicest things about parametric modeling is the ability to establish relationships between things. I made a formula for scale length to do frets and for bridge width to do neck taper. As for the body, I find its very nice for adjusting pickup routes, and establishing neck pocket dimensions and bridge locations that'll adjust with scale length and number of frets. Also can setup control cavity that'll adjust to body thickness, upper horn that'll always maintain its relation to the 12th fret as you adjust its shape, angle, and scale length changes.
Enjoy the benefits of a premium membership!
Upgrade Account
-
TalkBass has been independent since 1998. Add your voice.
Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
Join free Log in
Want zero display ads or expanded classifieds tools? Compare plans.
Free software for 3D model to feed CNC?
- Thread starter StuartV
- Start date
That kind of capability does seem like it would be nice to have!
I'm moving on from trying out eMS to trying out BRL CAD. I'm hopeful, because it has NURBS. But, if it ends up not comparing well to Blender, then I think I'm just going to go with Blender. It really looks awesome. A steep learning curve that, I think, will pay off down the road.
I'm moving on from trying out eMS to trying out BRL CAD. I'm hopeful, because it has NURBS. But, if it ends up not comparing well to Blender, then I think I'm just going to go with Blender. It really looks awesome. A steep learning curve that, I think, will pay off down the road.
Just a thought, but if you are serious about learning a CAD program, why not take an introductory course at your local community college. The CCs in my area are helpful in letting students come in and use their equipment on off hours. There may be other like minded students that could help you.
Also, back to the parametric vs direct topic, the big difference is history. I personally prefer having history as it gives me the ability to go and look at my original sketches and edit them directly. Also you can move your construction planes later down the line and the objects derived from them will move a well. This is nice for angled planes for example such as neck and headstock angles.
I don't have a lot of experience with direct modeling but one of the key differences is the lack of history. You don't edit models from their sketches or construction planes/vectors, but rather would directly pull our edit its faces. This can be really nice on the artistic side of things, but less so on the geometric/assembly side of things. It's not that things can't be constrained, changed in a similar manner, it's just a different process to do it.
Direct does have the nice benefit in parts with a lot of complex operations in there history. With parametric, if you updated the first sketch in the series, out would then have to regenerate all associated operations that happened after, which could take a minute. This can get a little annoying I if you're doing repetitive changes to see the outcome on the final effect. Direct is nice in this situation since it does maintain a history and therefore there is no order of operations. If you make a change to the model characteristic that was defined in the original sketch, it only has to update the operations that are directly connected to that change since it makes no association to order of operations, just the constraints on the end results. This allows for quicker processing of changes which is nice when you're trying to tweak things in.
Hope that all makes sense and feel free to ask for clarification if it doesn't
I don't have a lot of experience with direct modeling but one of the key differences is the lack of history. You don't edit models from their sketches or construction planes/vectors, but rather would directly pull our edit its faces. This can be really nice on the artistic side of things, but less so on the geometric/assembly side of things. It's not that things can't be constrained, changed in a similar manner, it's just a different process to do it.
Direct does have the nice benefit in parts with a lot of complex operations in there history. With parametric, if you updated the first sketch in the series, out would then have to regenerate all associated operations that happened after, which could take a minute. This can get a little annoying I if you're doing repetitive changes to see the outcome on the final effect. Direct is nice in this situation since it does maintain a history and therefore there is no order of operations. If you make a change to the model characteristic that was defined in the original sketch, it only has to update the operations that are directly connected to that change since it makes no association to order of operations, just the constraints on the end results. This allows for quicker processing of changes which is nice when you're trying to tweak things in.
Hope that all makes sense and feel free to ask for clarification if it doesn't
Ok, I'll chime in as well. 
First of all, the software list in the first post is a bit of a mix. Don't confuse CAD and CAM software. What you want is a CAD solution. Leave the CAM part to the machine shop.
I'm currently using sketchup, TurboCAD and Cut2D. Sketchup is nice to do a 3D model, just to figure out how you want the end product to be. For the precise drawing of parts I use TurboCAD. From there I export a DXF file to Cut2D where I make the Gcode for the CNC router.
I did try out several CADs. I did some research and read reviews, but all I learned that CAD software needs to be tested to figure out what fits your own needs.
As for the actual build I would recommend that you go for 2D CAD and some very basic hand tools. Cut the parts in a CNC router and do the carving by hand. In my opinion carving is something that require feel. It doesn't matter how good you 3D drawing is. If the neck profile feels wrong you still need to do some manual carving, or start all over.
What you could do is to make a 2D drawing from a 3D model where you cut the profiles and pockets. You could even make a stepped pattern on the carved parts that you later smooth out by hand. A file combination file and some sand paper will go a long way.
In my opinion your approach is ineffective. If you want to build cheap, doing things manually will get you further for less money. CNC machining the entire 3D profile makes sense if you want to make several identical basses, but not on a one off.
Having that said, I fully understand the urge to do this to learn new techniques. I'm actually building my own CNC router at the moment.
If you want details I can get back to you when I'm a bit more awake.
First of all, the software list in the first post is a bit of a mix. Don't confuse CAD and CAM software. What you want is a CAD solution. Leave the CAM part to the machine shop.
I'm currently using sketchup, TurboCAD and Cut2D. Sketchup is nice to do a 3D model, just to figure out how you want the end product to be. For the precise drawing of parts I use TurboCAD. From there I export a DXF file to Cut2D where I make the Gcode for the CNC router.
I did try out several CADs. I did some research and read reviews, but all I learned that CAD software needs to be tested to figure out what fits your own needs.
As for the actual build I would recommend that you go for 2D CAD and some very basic hand tools. Cut the parts in a CNC router and do the carving by hand. In my opinion carving is something that require feel. It doesn't matter how good you 3D drawing is. If the neck profile feels wrong you still need to do some manual carving, or start all over.
What you could do is to make a 2D drawing from a 3D model where you cut the profiles and pockets. You could even make a stepped pattern on the carved parts that you later smooth out by hand. A file combination file and some sand paper will go a long way.
In my opinion your approach is ineffective. If you want to build cheap, doing things manually will get you further for less money. CNC machining the entire 3D profile makes sense if you want to make several identical basses, but not on a one off.
Having that said, I fully understand the urge to do this to learn new techniques. I'm actually building my own CNC router at the moment.
If you want details I can get back to you when I'm a bit more awake.
- Dec 20, 2004
- 4,267
- 2,481
- Disclosures
- Player-Builder-Founder: Honey Badger Pickups & Regenerate Guitar Works
An important clarification re 'parametric' modeling:
'Parametric' modeling is more accurately defined as 'relational design' due to the geometry having the ability to be driven by both parameters and also interactive relationships with other geometry.
Sound relational design practices involve both 2D and 3D workflows, and should have a well thought out 'flow' as to how the relationships are propagated thru both the specific part and also the larger assembly. These workflows also define what geometry is driven by or is driving other geometry. Poor relational design planning can be (typically is) the ugly demise of an otherwise viable geometric design.
Designing relationally requires forethought and planning - you must understand that you are designing relationships with a purpose, and decide early in the conceptual phase how you want to manage these parent/child relationships so that the relational flow is always in a well-defined direction. Failure to do so has the real potential to accidentally enable cyclical update loops. Sound geometric design is the byproduct of prudent relational planning.
In most every aspect of guitar making, 2.5D design is more than sufficient for what needs to be accomplished to define the part at the correct level of detail required to machine it and to validate that it conforms to realistically defined tolerances. There are a few instances where a Class A surface definition would be nice to have, but a lesser quality 3D surface will suffice since the part will be hand detailed after the machining tasks are completed.
all the best,
R
'Parametric' modeling is more accurately defined as 'relational design' due to the geometry having the ability to be driven by both parameters and also interactive relationships with other geometry.
Sound relational design practices involve both 2D and 3D workflows, and should have a well thought out 'flow' as to how the relationships are propagated thru both the specific part and also the larger assembly. These workflows also define what geometry is driven by or is driving other geometry. Poor relational design planning can be (typically is) the ugly demise of an otherwise viable geometric design.
Designing relationally requires forethought and planning - you must understand that you are designing relationships with a purpose, and decide early in the conceptual phase how you want to manage these parent/child relationships so that the relational flow is always in a well-defined direction. Failure to do so has the real potential to accidentally enable cyclical update loops. Sound geometric design is the byproduct of prudent relational planning.
In most every aspect of guitar making, 2.5D design is more than sufficient for what needs to be accomplished to define the part at the correct level of detail required to machine it and to validate that it conforms to realistically defined tolerances. There are a few instances where a Class A surface definition would be nice to have, but a lesser quality 3D surface will suffice since the part will be hand detailed after the machining tasks are completed.
all the best,
R
you're not going to get much of a CNC 'product' if you have to do a lot of contouring, etc by hand
- Dec 20, 2004
- 4,267
- 2,481
- Disclosures
- Player-Builder-Founder: Honey Badger Pickups & Regenerate Guitar Works
for woodworking you need to know where to stop machining and where to start sanding - you do not waste valuable machine time making final surfacing passes .001" apart when making final passes .045" apart works perfectly well to have a contoured surface face that will clean up with a minute or two of sanding with 80-grit paper.
go visit any guitar production shop and discover where they stop machining and pass the part over to the sanding team to start sanding the final details. those sanding nuances are what imparts the final surface contouring details into the pre-painted product.
likewise, you need to know where to stop on your machine cutting depths to account for sanding else you run the risk of having a part that is too small or a pocket that is too large. most professional players I've worked with can easily feel a .015" thickness difference in a neck contour down by the nut - knowing how far to offset the final cutting paths so that sanding hits the target dimensions comes with robust CAD/CAM tools and lots of hands-on design and manufacturing experience.
all the best,
R
- Dec 20, 2004
- 4,267
- 2,481
- Disclosures
- Player-Builder-Founder: Honey Badger Pickups & Regenerate Guitar Works
it's the definition of a surface with respect to its definition and reflective qualities
http://en.wikipedia.org/wiki/Class_A_surfaces
a visualization is often done utilizing a zebra line type reflection on the surface to visually see how light reflects across the contours. even the smallest inflections will cause ripples in the reflection - and these inflections also play havoc when generating clean tooling paths.
Invalid Link Removed
all the best,
R
I'm gonna revive this thread to ask @StuartV if you had any luck with your files, and if so, where did you get the files?
The sad fact is that I got as far as starting to go through tutorials and play with Blender and then life got in the way and this project has been back burnered ever since. It's not dead, just still on the back burner.
