Official Luthiers Forum!

I've been lurking on here for a few weeks, and there's a lot of good info to be had. I'm considering getting into CNC as a small business, and was wondering if any of you guys that use CNC professionally would ever consider using a DIY CNC kit rather than buying a plug and play unit? My budget to get everything I need to get started cutting would be 10k. I would like to do everything that I can with it. From detailed inlay to sign work. The more my possibilities for cutting, the better my possibilities for being able to pay for the machine, and eventually make a living doing it.

_________________
Jason McGowan
M&S Guitars
_____________________________
United States of America, looks like another silent night
As we're sung to sleep by philosophies that save the trees and kill the children....
Casting Crowns

The machine is much less important than the programming. Do you have CAD/CAM skills or do you have someone you can work with who does? Until you can answer "yes" to that, a CNC machine will be a dust collector in your shop.

But...for $10,000.00 you can get something useful.

As I see it, there are categories of machines:
1) Basic hobby machine using a standard router
2) A machine with a VFD spindle
3) #2 with a good vacuum table
4) #3 with automatic tool change

You can just about get to level 2 or 3 for 10K. ATC bumps you quite a bit higher.

Thanks for the response. I'm actually a 3D drafter in AutoCAD professionally, so I have the 3D modeling part down. I also used to do some programming, so I have the ability to learn CAM. Hopefully quickly.

I was hoping to get to your level 3 with this setup, with the possibility of a ATC later. Is it logical to try and get all of this, in a professional level with a DIY though?

_________________
Jason McGowan
M&S Guitars
_____________________________
United States of America, looks like another silent night
As we're sung to sleep by philosophies that save the trees and kill the children....
Casting Crowns

They guys who've built machines will have a different opinion. I've sunk way too many years into trying to modify and upgrade a sloppy, limp machine to do accurate work. This was money spent and time lost, instead of generating income. That's OK if you don't have orders to fill, but it really held back production for us.

K2 makes a machine specific to guitar making. It can be spec'd with either steppers or servos. It's gantry is a little on the light side but I know of at least two builders that use them.

If you are making a real pro stab at this, consider upping your budget and then lease/purchase the machine.

To me, the biggest drawbacks of the cheaper machines is having to have a screaming router going full time; for me it was bad enough when I had a Bosch table top pin router, but the machine run time was shorter with that. You'll actually have a CNC running a lot more than you would a shaper or router. Real spindles are where real CNC starts as far as I can tell.

If you're an AutoCad guy, you're way ahead of the game. Now it's figuring out sequence of operations, holding fixtures, and the CAM which really isn't that difficult.

As a guy who's making a go at it full time, I can say that for $10k you can make a go at the kit rout and make money with your machine. However, keep in the back of your head that this is simply your first machine and begin planning on the next one. You can get your business off the ground with it but it probably won't be able to get you to a sustainable salary. If you're starting out part time, it's probably a decent cost effective way to go.

As Rick said though, you absolutely have to get a spindle out of the gates as a router is extremely light duty compared to a spindle. The good news is that you can get a ubiquitous CCS (Cheap Chinese Spindle) and VFD for well under $1k and you can even get an Italian spindle+VFD for around $1k if you look hard enough.

I use an Xzero Raptor for my business and although it's one of the roughest kits out there (basically a box of parts), from everything I've seen it's the best one out there bang for your buck wise and performance wise. The setup I've got out performs K2 machines that cost more than double what I paid for mine. It's also going to take the most work from you to get it dialed in and working reliably where as a K2 you plug in and go. For $10k, it's going to get you closest to a pro machine.

Something to consider though is that although you have access to AutoCAD at work, you're also going to need some CAM software to go along with it. I went with VisualMill pro for SolidWorks which is a $4k package. There are others that will work but, if you want to go pro you'll need the options provided by the higher end packages. At $4k, VM is actually on the really cheap side when it comes to CAM software.

As far as ATC goes, it's probably essential to get into production work. It's also is going to cost about $10k to get into it. Yes, you could add an ATC spindle for as little as $3 or $4k but don't forget that cheap tool holders are $100 each (on the cheap) plus compressor, line driers etc. etc….

I'm already planning my next machine and am torn on whether to go with a turn key setup or make mine from scratch or something in between. I can tell you for sure that it's got to be accurate, precise, and have an ATC. I'm torn because I know that I can make something top notch for about 1/4 the price of an industrial setup but it will also take 6 months to design and complete.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

I don't see this particular machine on their website. Is the ViperXZ comparable?

_________________
Jason McGowan
M&S Guitars
_____________________________
United States of America, looks like another silent night
As we're sung to sleep by philosophies that save the trees and kill the children....
Casting Crowns

Check the forum area of his website, I think he's got a few raptors left.

the Viper is an unsupported rail machine where the Raptor is a profile rail machine. You won't be able to push a Viper as hard as you can push a Raptor.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Gotcha! I contacted him, and he only has a 30" x 36" remaining. Have you found a use for a travel that's larger than that? I want to remain as versatile as possible. I would hate to have to turn down work because my cnc travel is 1" smaller than I need.

_________________
Jason McGowan
M&S Guitars
_____________________________
United States of America, looks like another silent night
As we're sung to sleep by philosophies that save the trees and kill the children....
Casting Crowns

30x36 is ok but 30x48 would be better. Remember that the cutting area is smaller than the dimensions. I have a 24x36 and really wish I had gone with a 30x48 but that option wasn't available at the time.

The 36" long axis isn't terrible but it's a little too small for thru necks. 30" short axis would be nice as you get a full 25" of work area. I only have about 20" of width.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

He got back with me and said that he can make it a 48", he would just have to order a ball screw. So if I get that, with the motor and driver kit, all I would need is a vacuum table (or a way to clamp work), a spindle with VFD, and a computer with CAM?

_________________
Jason McGowan
M&S Guitars
_____________________________
United States of America, looks like another silent night
As we're sung to sleep by philosophies that save the trees and kill the children....
Casting Crowns

I would highly recommend going over to CNCzone.com and spend at least several days reading through the various DIY build threads on there: http://www.cnczone.com/forums/diy_cnc_r ... _machines/

Check out what questions people are asking and what types of responses they get. You'll also quickly learn who gives good answers and who doesn't!

There's a million little details in DIY CNC and you'll probably drop $1000 in nuts, bolts, proximity sensors, angle aluminum, wiring, etc. etc. etc. You also left off a controller computer on your list. Controller software too (e.g. Mach3). Possibly a controller card such as smooth stepper or Kflop or who knows.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

I have large Haas machines and can't offer advice on the cnc routers.

I'll just say that having mastered modeling you're a good ways down the path...that's good. Mastering cad/cam/cnc is best approached in that order...learn to draw, learn to make tool paths, and start with a basic machine to learn the process...and then spend smarter money down the road on a machine more suitable for your type of work. So, think of the first machine as a learning curve and if the idea of getting the "perfect" machine promotes anxiety...console yourself by knowing it will always be useful for certain tasks after you've purchased the next one.

Don't skimp on the cam software. I've always felt that the one place you could draw the line between truly a useful cam package and one that will eventually frustrate you is the ability of the software to "project" tool paths onto a surface. That kind of separates the wheat from the chaff in my opinion. That module has been a staple of every high end package I've ever looked at (I use MasterCam X5) and frankly, I'd have a hard time replicating the work I do without it. It allows precise control over tool approach and direction (along with reverse spindle rotation)...which is quite handy in a number of situations when cutting non-homogeneous materials....like wood.

...and don't underestimate the value of a tool changer. This won't likely will fit into your budget so I'd definitely prioritize the cam software initially...but you'll want to think hard on it for the next machine.

_________________
I read Emerson on the can. A foolish consistency is the hobgoblin of little minds...true...but a consistent reading of Emerson has its uses nevertheless.

StuMusic

I'm on it! I read so much last night before bed that's it all that I dreamt about. Now that I think about it, its all that I've dreamt about for the last week.

_________________
Jason McGowan
M&S Guitars
_____________________________
United States of America, looks like another silent night
As we're sung to sleep by philosophies that save the trees and kill the children....
Casting Crowns

I definitely don't want to skimp on the software, that's a no no in any industry IMO, but my word! The prices are incredible on this stuff. Are there any packages that are affordable that do a good job for a small business? I have to figure a way to pay off everything that I purchase, and make a living at some point before its time to retire.

_________________
Jason McGowan
M&S Guitars
_____________________________
United States of America, looks like another silent night
As we're sung to sleep by philosophies that save the trees and kill the children....
Casting Crowns

That was my strategy with the Shopbot. I rationalized that I'd go broke supporting larger lease payments while I learned the machine. In retrospect that likely would have been the case.



This is maybe better saved for another thread but could you elaborate a little more on this concept Stuart? Are you talking about pencil tracing in 3D, pocketing with a 3D floor?

My 18 year-old son, Elias, has been working in SolidWorks on our instrument parts, and I have another part time guy, Jason Lindsey, who has been working primarily in Rhino CAD and CAM to get tooling done and then make parts. The advantage of SolidWorks seems to be that it's easy to plug in new dimensions and work parametrically with designs...in our case, for instance, Eli was able to do a 12 fret tenor uke neck and then quickly morph it into a concert uke neck, a 14 fret tenor neck, a 5 string neck (wider nut). Rhino does not allow that, but it is better for the complex surfaces where the barrel of a neck transitions into the peghead. Since all our earlier work was done in Rhino and that's the CAM system we have, we now just bounce back and forth between the two programs.

These young guys have it all over me in the CAD department, but luckily, I'm really good with production sequence and fixture design, and now, especially since my machine is paid for, I'm really happy with my choice of the bit Techno with the automatic tool changer and massive vacuum pump. I can see a smaller machine in our future, though...maybe something dedicated to inlay work where I don't necessarily need an ATC.

I have a DIY machine and am very happy with it. It will do everything of what I need except heavy aluminum work and accuracy is very good. That being said, accuracy was marginal when I started the build, I soon learned why I needed good ball screws.

My only mistake was in the footprint of it. Mine is a 2x4 which equates to 20" x40" cutting area with 6" Z.

One thing about a DIY is you will know your machine intimately, that has been a blessing when something goes wrong. Also maintenance is much easier, since I know the machine and it's weak points. Other than that, I see no benefit over some of the new offerings out there other than costs.

Software is something different, Get the best you can afford, period. Make sure though that you are comfortable with its usage though by testing the trial versions. I use Rhino5, have added Tsplines for Rhino from AutoDesk as well. My main software for 2D and CAM/Toolpath gen is Vectric Aspire. It also will project 2D on to 3D so fret slots are easy. Aspire has and is improving their 3D modeling capabilities. For a simple fretboard including compound radius it is hard to beat the speed and layout using it. Other things are easier done in Rhino and exported to gen the toolpaths in Aspire.

I use Mach3 as the controller software and have been very pleased with the results. It has some features that if used correctly or applied outside of the box so to speak can enhance your cut times and finish as well.

Just my 0.02cents worth.
MK

_________________
I'm outside looking in, just farther from the window than most.

Hey Mike,

Would you have been happy with a travel of 25x42x8?

BTW, I see that we're in the same neck of the woods. I would love to see your DIY cnc run some time. Are you a part of L.I.N.T?

_________________
Jason McGowan
M&S Guitars
_____________________________
United States of America, looks like another silent night
As we're sung to sleep by philosophies that save the trees and kill the children....
Casting Crowns

Just to chime in on a few points:

As far as work area goes, you want at least 25" of useable area as there's often stuff that you want 24" wide (e.g. radius dish). With a work area of 25" that gives you enough width to clear your bit.

42" long is just barely enough for a through neck guitar. If you want to tackle thru neck bass stuff, you'll need even more length. As length and width increases, it becomes a lot harder to maintain accuracy and rigidity (i.e. it costs more)

As Rick said, SolidWorks is a full parametric modeler with the capability to handle assemblies which is why I chose it. It's about 4x the cost of Rhino. It's easier to do complicated organic looking shapes in Rhino but it my opinion, especially if you want to go pro, going with a full parametric modeler is the way to go because it's so much easier to re-use old designs to make new ones. It definitely has a steeper learning curve because if you don't build the model in the right order, it's much more difficult or even impossible to re-build using new dimensions. Just as an example, with my fretboard model, I use a spreadsheet to enter the parameters and just push a button to make a new fretboard with the new dimensions and scale length. That said, Rhino is a very good product and lots of people are using it for guitar work.

For CAM software, I agree with Stu. You don't necessarily need MasterCAM, but you need something with comparable features. From my research, VisualMill Pro (I use the SW plugin) is the best bang for your buck as far as that goes. Pencil tracing isn't that important but 3D offset pocketing and 3D engraving are essential. Multiple work coordinate sets are also necessary for multiple sided parts as well.

In most cases, there are work arounds if you're doing it with cheaper software but it's obviously more work. E.g., with VM standard, you can do 3D engraving and in conjunction with SolidWorks, you could trick it into doing 3D offset pocketing. You can also do multiple setups by creating separate parts. As they say, necessity is the mother of invention and you can usually figure out a way to get something done if necessary. I don't even have my VFD wired up to run in reverse as it's too big a hassle to change tools for just a few left hand cuts but I've figured out ways to avoid tear out by tightly controlling the order of my cuts and pre-cutting certain things.

There's also BobCAM which has a horrible reputation for their sales department but it is really cheap, relatively speaking. It does have a full feature set but it is a little quirky from what I've heard. A buddy who also has a Raptor went with BobCAM and he's able to accomplish everything he needs with it.

For inlay, I use Vcarve Pro as it blows everything else away for that. Just a few mouse clicks and you've got G-code for the pocket and the inlay.

For machine control, Mach 3 is quite ubiquitous but I'm not entirely satisfied with its performance (which is contrary to most folks on this board). On the plus side, it's really inexpensive, has a large user base and tons of free or cheap macros but it has a lot of negatives too. There are two big minuses for me - first is the lack of S-curve acceleration and second, probably more importantly is that there's a bug in the CV mode that can cause a perfectly reliable stepper machine to stall and ruin a part (which incidentally happened using code from BobCAM - pretty consistently) and that can cost serious money. In fact, CV mode on Mach3 is barely passable but with enough fussing, you can get the job done with it…..most of the time.

There are a number of other bugs and so forth in Mach 3 but most of them are a nuisance rather than a show stopper and you'll notice them and get used to them as you use your machine.

For now, I'm sticking with Mach 3 because I've got my workflow down but I'm investigating other options as well. When I free up some cash, I'll most likely go with a Kflop first running the Mach 3 plug in as it's the fastest to get working but eventually I'd like to run the Kmotion CNC software as it's got some pretty advanced trajectory planning that Mach 3 can't touch (and from what I've heard from Beta testers of Mach 4, won't even be on the implementation list for at least 2 to 3 years).

I'm not trying to push you away from Mach 3 too hard as it's an inexpensive way to get a machine running with a very large feature set but, it's definitely pretty far down on the list as far as machine performance goes. The harder you push your machine, the more bugs and quirks will show up.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

One of the things that Eli does in Solid Works is to basically draft in very close to the same steps that we'll then go through to make the part. He does a lot of the work subtractively starting with a 3D block on the screen that represents the wood blank. This helps us really think through the whole process of making parts, and that sequence of operations is as important as having the part drawn nicely.

That's an interesting approach Rick. One that I'd not thought of.

I typically use the "Master Model" or "Master Sketches" approach where you begin with simple sketches that define all of your critical parameters and planes first at the start of the design tree. Features further down the design tree reference only the original basic sketches so you don't accidentally break something when changing parameters. In an assembly, the individual parts only reference the original surfaces etc. to allow for simple modifications. good stuff.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Actually I would prefer about 60x32x8
PM me for contact info, I'm in Farmers Branch.
Yes, I am a member of LINT but not active.
Mike

Also I have a friend that is dying of cancer, I received an email from him asking if I knew anyone who would like to purchase his ASPIRE software, His is 3.5 he wants $1000 for it, the Vectric transfer fee is $75, and the upgrade to the latest 4.0 would be $400, That's a saving of $500. Just an FYI for anyone.

_________________
I'm outside looking in, just farther from the window than most.

Andy, it's been pretty amazing to have Eli working with me. He started with SolidWorks when he was 15, but he's been around my workshops all his life, and so he has a very good understanding of the sequence of operations in busting down large planks of wood into blanks and then processing them into real parts.

I would consider a DIY machine for my next one. An old friend gave me a bunch of pretty hefty servo motors and a couple of very nice ball screw assemblies. I picked up some Thompson linear bushings and rails quite a while ago. And...my oldest son is a master metal worker...welding, full CNC machine shop, etc. So I'm kind of more than 1/2 the way there, and having an excellent machine to begin with has given me the understanding of what I'll need in a next machine, though for my situation, a laser cutter may be more important than a 2nd CNC router. I'm already subbing out some laser work...magnetic pickup bobbins...and we could cut tops, backs, and sides for our acoustic and semi-hollow instruments as well as doing some nice marquetry with a laser.

This is exactly what I do. To me it just makes sense. I know that if I go thru it as though I was actually building it by hand, and the process works virtually, I should be able to duplicate it physically.



Sent from my iPhone using Tapatalk

_________________
Jason McGowan
M&S Guitars
_____________________________
United States of America, looks like another silent night
As we're sung to sleep by philosophies that save the trees and kill the children....
Casting Crowns