turindev
Commercial User
- Jul 1, 2005
- 514
- 1,594
- Disclosures
- Builder, m7bassworks.com
+1 on fusion360. That's what I use and it has both CAD and CAM built in .
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.
Any amateur builders using their own CNC?
- Thread starter Jeff Siddall
- Start date
I’ve never built or used a CNBC but I would imagine that overall rigidity, as well as a securely clamped workpiece, would be paramount for success. Also, I imagine a typical router has runout and slop in the bearings and collets etc. Depending on the quality and previous use.
in my work experience a .023 diameter endmill isn't that small. you need ones with a highly abrasion resistant coating. TiCn is a good choice or "Diamondlike". One of the keys to using tiny tools is a steady stream of air right at the tool for chip evacuation. I also suggest a ramping tool path where you just don't plunge down vertically into the cut but feed to your pass depth on a slope across a longer area.
i'd really like to see someone regrind the ends of a micro miniature carbide endmill correctly for a reasonable price. in nearly 40 years of high precision manufacturing i've never seen this on anything but twist drills and rarely on the smaller diameters. with the right equipment it can be done, but it's generally considered a waste of time on a consumable tool at any price other than when you absolutely need one and have a microscope, a lap, and a steady hand for a hail-mary, no-choice effort.
i'd really like to see someone regrind the ends of a micro miniature carbide endmill correctly for a reasonable price. in nearly 40 years of high precision manufacturing i've never seen this on anything but twist drills and rarely on the smaller diameters. with the right equipment it can be done, but it's generally considered a waste of time on a consumable tool at any price other than when you absolutely need one and have a microscope, a lap, and a steady hand for a hail-mary, no-choice effort.
Last edited:
My son looked at a 3D printed CNC machine. I used a combination of 3D printed parts and metal conduit or all-thread. He said that it could be however big you needed because you just bought the conduit to the length you needed. The 3D printed parts would hold all of that together. I watched a couple of videos on it and it seemed very plausible. If you already had the 3D printer, the concept was fairly cheap to make.
BnB
BnB
Fusion 360 is the way to go. A bit of a learning curve, sometimes glitchy but a very powerful software and will handle practically anything you throw at it. Especially for FREE... or "education license"
Yeah, looked into it. We don't have anything as good as Makers Place, and there is no CNC at the places we have.
Yes, and you can see in some of the demos that there is some "fuzzyness" in some of the routes due to slop/runout. But it appears to be small fractions of a millimeter so should be adequate for fretwork. One concern I have about the design of the small CNC machines is flex in the linear rails. The machines I am looking at use 8 mm rods and I can imagine there will be some deflection there. Again, probably good enough for fretboards since it can likely be sanded out with a beam.
Thanks for the chip evacuation tip. In my mind I envision doing the slotting last and don't expect to "plunge" at all. Just take consecutively deeper passes over the already radiused board.
Agreed. I am sure this is done at the factory.
I have no doubt Fusion 360 is good, but it doesn't appear they support Linux. I got fed up with Windoze being so bad so I have stopped using it. Building a new machine just for this one app isn't likely to happen unless I can't use FreeCAD for some reason, but I will cross that bridge when I come to it.
8mm rails seem a little skimpy to me in the overall scheme of things. And as I’m sure you realize, the smaller the bit the more fragile they are, and the less deflection they will take before breaking. On the plus side, you can take your time with the passes only taking off a couple thou at a time. Another thing you could look at is to have the bit start out each slot accurately and finish to depth with a fretting handsaw
Yeah, I dug out some deflection formulas and there could be over 1 mm of deflection using 8 mm rods. Re-ran the calculation with 20 mm rods and the deflection came out at 0.05 mm. Thinking I will upgrade the rods. They aren't really that expensive.
Regarding hand sawing, my plan is to use blind slots so hand sawing is not an option.
Have you looked into a shared space?
Faculty of Engineering | University of Ottawa
This might be a good place to make a few cuts and learn about CNCs before you invest. If you network with the community you might find someone with one in their garage already you can share with.
I'm lucky enough to have 24/7 access to a big old CNC at my makerspace CNC Router – Ace Monster Toys Makerspace and now I couldn't imagine nibbling through something with a tiny hobby machine. I used it for all my fretting and radiusing now. It has no problems doing a compound radius or multi scales if I wanted it, all that difficulty is in the design.
I like getting the collared bits so when one breaks I don't have to reindex the Z height.
Yeah, I think that to do this complicated job 100% bang on it will take a commercial level machine to do so. Maybe some kind soul here will offer to take on the job as long as there’s no rosewood being shipped across borders (not trying to put you specifically on the spot).
Yeah, I looked, but around here it seems like everyone has inexpensive 3D printers but almost no one has CNCs, and the ones that do exist are pricy.
Quite honestly the idea of having my own CNC, even a slow one, is appealing. Having to drive across town and pay by the hour sounds tedious and expensive, especially since fretwork has to be done slowly, no matter how big the CNC.
I would be happy to make the model for you. I'm at that stage where I just love modeling things. And if you want to come out to the San Francisco area we can cut it. If you do get a small machine I could even set up the gcode. The CNC makes things like compound radius or getting an even slot depth simple.
I feel very lucky to be able to go in at midnight and not watch the clock. I've had access for about 2 years now and find that a 15 minute job might keep me there for 4 hours as I struggle with little annoyances. A radius and fret slotting should take less than an hour including setup and clean up afterwards. Funny how it always takes more. So having a CNC at home would be great.
Thanks for the offer. Next time I am in bay area I may take you up on that, but it could be many years from now!
Somewhat surprisingly the modeling part seems pretty easy. Last night I drew a model of a nut, complete with conical arced slots, and also created a compound radius fretboard with the FreeCAD macro I found. I already have the FretFind multiscale DXF for my project but have not run that through PyCAM yet. I did run a cut simulation on another much smaller example project and with a 3 mm ball end and roughly 1 mm path separation it took about 2 hours. Anything done in less than an hour sounds like it would just be a rough pass. I guess it depends on how clean a surface you want the CNC to leave. With my own machine time would be on my side instead of against me.
Yes, Maslows are cool in their own way (I recall watching a video of a guy who made a plywood guitar body out of one) but I need something about 50 times more accurate!
Last edited:
It sounds like you are on your way to success. I use a 5/8 ball end so even with the same 1mm stepover it will leave a smoother finish as the cusps will be flatter.
My time for the radius operation on a 29 fret bass neck with 1mm stepover is 17 minutes 21 seconds. It still needs a bit of sanding but the shape is done in that time.
My time for the radius operation on a 29 fret bass neck with 1mm stepover is 17 minutes 21 seconds. It still needs a bit of sanding but the shape is done in that time.
Yeah, a bigger ball end will definitely do a nicer job of a fretboard. I just didn't figure you could make about 700 passes in 1000 seconds, but thinking about it again I am guessing the tool path was along the length of the fretboard. I suppose for a single radius board that would make the most sense as you would do the entire length at the same Z-depth. For a compound radius it probably doesn't matter much as the every pass has changes in 2 axes.
One more question: finding quality 0.6 mm (~0.023") fluted end mills with enough cutter length to make flat bottom fret slots on a small radius board seems trickier than I thought. I figure I will need about 3 mm (about 0.115") depth in the middle of the board but many cutters have just 0.079" depth of flutes.
So, has anyone used fishtail cutters for fret slots?
They are cheaper and have longer cut depth.
Alternatively, I could make a 3D radiused slots but that is a lot more work than just dumping the fretfind output into the CAM software -- especially on a multiscale board.
So, has anyone used fishtail cutters for fret slots?
They are cheaper and have longer cut depth.
Alternatively, I could make a 3D radiused slots but that is a lot more work than just dumping the fretfind output into the CAM software -- especially on a multiscale board.
