Official Luthiers Forum!

Thinking to get a 3018 Pro CNC with router and laser.
Any experiences out there?
I've been thinking for Logo inlay with wood, maybe MOP, or epoxy/crushed stone/glitter on headstock and Logo/model no/serial number on marriage strip for back, rosette segments maybe and personalizing FB inlay at 12th fret.

_________________
The name catgut is confusing. There are two explanations for the mix up.

Catgut is an abbreviation of the word cattle gut. Gut strings are made from sheep or goat intestines, in the past even from horse, mule or donkey intestines.

Otherwise it could be from the word kitgut or kitstring. Kit meant fiddle, not kitten.

I've also been contemplating buying one of these machines. They seem like a cheap way to get your feet wet with CNC. With the right fixturing you might be able to cut rosette channels and sound holes in soundboards and also cut the inlay cavities in fretboards in addition to what you mentioned. If there is a way to precisely reregister the work, perhaps also slot fingerboards.
I still don't know about the resolution the machine has and since it uses stepper motors, if it would drop steps as some of the earlier machines were prone to do. I would want the machine to do the work as well as I can do it by hand (or even a little better).

As long as you have realistic expectations, these small machines can be a lot of fun. They are definitely going to lack rigidity, but if you go slow you can do some pretty decent work with them. In my experience, stepper motors lose steps because they are being pushed too hard. I started out with a used base model cnc shark. I think I payed about $1200 for the machine, computer, and software. It was predominantly made out of hdpe so crashing it wasn’t usually catastrophic, it would just bend... 18”x 24” is kind of the minimum useful size for luthiery in my opinion.

M

My sense is that the 3018 is a toy not suitable for anything you may want to do. The bed is not large enough to hold down a fingerboard. The spindle only turns at 7,000rpm, which will not cut shell or frankly anything but wood. I wouldn't want to use a .0.23" bit at that rpm even in wood, so even if you could find a way of fixturing the fingerboard you wouldn't be able to cut fret slots. The little lightweight laser module will char wood and paper to make designs, but little else. As a combo laser/cnc I believe it sets you back about $450. But I also believe from other forums that people tend to spend nearly that much in upgrades.

I think it is still the case that the minimum CNC set up that makes sense is something like a ShapeOKO or; if you like tinkering, build your own MPCNC or a kit from OpenBuilds. Yes, this will set you back three times as much money as the 3018 Pro with laser, but you will have a lightweight tool rather than a toy.

Whoops.. I didn’t actually read any of the specs...

I've been using a CNC router for a couple of years now (5 guitars, a lute, & 2 harp guitars) and encourage you to get the MOST RIGID machine that you can afford. The advertised precision/accuracy of a machine is irrelevant because it's determined when cutting air. What matters is how accurately the machine cuts wood (or MOP or ???), and that is almost entirely dependent on the speed of cutting and the rigidity of the machine (assuming a sharp bit). I started with a MillRight Carve King (which is substantially more rigid than a 3018 unit) and have moved on to a MillRight Mega V. If you want to rout interesting rosettes and the soundboard channels for them, and then expect a good fit, get a sturdy machine.

Note: cutting fret slots doesn't require a particularly sturdy machine as the major forces deflecting the router bit are not transverse to the motion of the router. Forces due to bit friction and material cutting do include transverse forces, but they are consistent for all frets so don't distort the pattern. It does, however, require 20,000 rpm or more to spin the 0,023" diameter bit adequately for a decent cutting speed. Cutting almost anything else in instrument building will benefit from rigidity.

It’s certainly got an attractive price point.
I’d be most interested in the laser function as it’s the only thing I contract out. Since my guy is not always around it’s a vulnerability.

The Shakeoko looks much more like a real machine, and the CNC crash course would be very useful, for me at least. Didn’t see the z axis in the specs, know what it is?

This goes quickly into the conversation of quality vs price. There is such a huge spectrum that we dealing with; that coupled with my empathy for anyone who doesn't want to (or can't) spend thousands of dollars on the "right" machine the first time, I find it difficult to tell anyone not to pull the trigger.

The price for admission into the CNC industry can be very steep for a lot of people. This seems like a decent little machine as long as you stick to your expectations and goals. I would say all the points that the fine people here have mentioned have merit. A lot of what was said has workarounds and adaptations to minimize the issue, but they are still problems that you will begin to understand them in time.

There is so much to learn in the CNC world, and even a special subset of things in the luthier world. I expect you will want a different machine a couple years down the road, but that's normal, we all do. Lol

_________________
-Marcus
http://www.bailieguitars.com
Instagram | YouTube

40mm

_________________
-Marcus
http://www.bailieguitars.com
Instagram | YouTube

As rlrhett stated the 7krpm spindle isn’t gonna get you very far. The shell cutters I use from Precisebits.com have a recommended rpm of 24k. Another consideration with the spindle is runout. Think about the fatigue on a .010” bit when your spindle has .001” of runout, especially @ $25 a piece for the bit...

I had a "toy" bandsaw for a couple of years (a 10" Craftsman). At the time, I thought it was OK, if not great. I sawed slowly and anything approaching a resaw was impractical. Only when I got a decent 14" bandsaw did I appreciate the limitations I'd been working with. I expect that there's a similar process in CNC routing.

If you're willing to consider a 3018 machine as a "learning vehicle," it should serve. But the learning curve for CNC is far more involved than the learning curve for bandsaws -- and the hardware is the easy part. I would encourage learning CNC "programming" on the software tools that you expect to endure. Then, when/if, you upgrade your hardware, your software learning (by far the most complex and challenging aspect of CNC) will apply almost directly.

While there are a number of free CAD/CAM software tools available, all that I know of are either buggy, very limited, or quite challenging to learn. As a result, I purchased a license for VCarve Desktop by Vectric. I've found that tool easy to learn and adequate for almost all of the luthier tasks for which I wanted CNC support. It costs $350. That tool, plus a 3018 machine, could be a reasonable "getting started" suite. But, if you like what CNC can do for you, the 3018 may not be your last machine.

My first CNC was a CNC Shark. It had a Bosch Colt for a spindle, and worked OK. It came with VCarve Pro software, which for 2 or 2 1/2 D projects like inlays, slotting fretboards, etc. is about as good as it comes. The Shark flexed a lot under load, so I ended up running it pretty slow. When I started modeling in 3D, I felt the need for something beefier and got a used X Zero Raptor with a Colombo spindle. Big step up in capacity and speed, and had very little flex. I sold the Shark for near what I paid for it (on sale), and kept the VCarve software, so I did OK, and it was a great learning machine.

So, yeah, I'd say that getting an inexpensive machine at first to see whether or not you're going to like it is a good idea. But, if you do get into it more, you'll probably upgrade more than once...

I still have a lot more money tied up in software than I do in hardware for CNC work. And still have and use that original VCarve Pro license.

Dave

I'm sure the genmitsu 3018 would have some limitations and may need some up grades to be a more practical machine. I have heard of people swapping out the spindle on some machines for a pencil grinder unit - some have even used the one from Harbor Freight:
https://www.harborfreight.com/pneumatic ... 60244.html

I'm sure there would be a learning curve with most of the software, even the expensive stuff. I think the freeware has gotten better over the years, and there is always G code to do simple 2 D stuff. Knowing the limitations of the machine ahead of time may create more reasonable expectations for what it can do. I did find this video to be encouraging - but he is someone who is familiar with CNC routers and their related software:
https://www.youtube.com/watch?v=Y5nyjvytlBk&t=11s

The machines are improving every year and costs are coming down. How long that trend will continue is hard to say.

I bought a 3018 direct from China around 12 months ago. I wanted to make headstock logos from wood and was not expecting much more than a beginner unit that would do the job but not much else. What I was not expecting was the problems. The Z axis did not work at first and the supplier was not helpful, but I did get a discount from them. I worked out that the problem was the controller chip had not been installed correctly. Once that was fixed and the thing was working I had trouble with the grub screws coming loose from the vibration of the "spindle" and the cut would drift after a while. Some locktite and that solved that problem. Next problem was the "spindle", it was not up to the job. Massive runout and it does not run fast enough for the small bits I needed. I then made an adapter to hold a Dremel instead and the so called "spindle" ended up in the bin. Next was I needed to get the software to do what I wanted it to do. Many hours later and much head scratching and several broken bits it was all working a treat. I made a batch of headstock logos and have not touched it since, but it is now time to get to get it out and make another batch. Should be a breeze next time. The first batch and it has paid for itself. Main problem is it needs to go slow or the bit will break (0.3mm) so takes around 3/4 hr to make each logo and the noise drives me insane. For anything else I would get a more robust unit with a proper spindle.

So, it is a toy with limited uses. If all you want to do is to do headstock inlays then it will do the job, but you might have to replace the "spindle".

Good to hear some direct reviews...

Indeed it is.

Dave

you get what you pay for . I just bought a cnc this year and here is some advice for what it is worth
A ditch any router and go with a water cooled RF spindle ( about $300 ) off aliexpress
B software is that makes it work so be sure you know what you are or are not getting
C fixture and dust collection Understand how to hold things in and how much dust collection you need ( spoilboard with hold downs or vacuum )
D Cutting tools Not all are the same as a tool maker I used a number of cutters so this was easy . I have lots of end mills and different materials need
different cutters
E collet you want something with ER 20 collets this lets you use up to 1/2 in shank cutters
F dust boots this helps to control the dust and will work and attach to the dust collector

I will be doing a facebook live this weekend on my FB page about 1 pm covering this subject

_________________
John Hall
blues creek guitars
Authorized CF Martin Repair
Co President of ASIA
You Don't know what you don't know until you know it

I agree on most counts (my spindle is air/fan cooled, not water cooled, but water cooled is definitely a good way to go. But the spindle is just way better than a router, and not that much more expensive if you shop around).

Fixtures are definitely key. I use mostly vacuum fixtures that I've machined on the CNC itself for specific purposes. I have a good dust collection boot, lots of collets and a good strong stepper based motion control system (Servos are better in many if not most regards but they're a lot more expensive and not as forgiving when things go south. I worked in motion control professionally for many years and have a good knowledge of how to work around some of the downsides of steppers).

Software is still the biggest part of the picture. I know that a lot of the folks out there who do this professionally use Solid Works to model, but AutoDesk's free to hobbyists online CAD/CAM is certainly good and Solid Works isn't cheap. I use Rhino CAD and MadCAM, as well as Vectric's VCarve Pro for modeling (since I've used them for years and could afford them). They work great for me. On the machine control side, I had used Mach 3, but switched over to UCCNC a couple of years back and have been quite pleased. A definite improvement over Mach, and again, it didn't break the bank. It does what I want it to do and does it quite well.

I've built entire instruments via CNC in the past when I was doing some semi-production banjo stuff, and it was fun. 3D modeling is challenging and fun. Doing the rough out on archtop plates with CNC instead of router jigs like I used to use is much better. These days in retirement, I prefer to do most everything with hand tools and maybe a router for bindings and such, but the CNC is sure great for slotting fingerboards, cutting symmetrical inlays like diamonds or stars that are quite hard to get right by hand, and other stuff like that. I've also been CNCing metal parts for pedal steels. CNC is a good place for computer geek/instrument maker guys like me to spend some time.

But if you don't have some kind of background in machine shop work/coding/CAD design, what's really fun and fulfilling for me could easily be a lot of money down the tube for those who don't have the same background.

Dave

Dave wrote:
"But if you don't have some kind of background in machine shop work/coding/CAD design, what's really fun and fulfilling for me could easily be a lot of money down the tube for those who don't have the same background"

I guess that is why a cheap CNC is attractive. Many years ago I did wrote some simple programs using G code to cut out cabinet parts. Back then the machines and CAD/CAM programs were much less available and most were quite expensive. Now children are getting 3D printers instead of erector sets for Christmas. The technology is much more accessible than it used to be. Probably in the not too distant future - instead of buying a set of printed plans to build a guitar, you will buy the CAD/CAM files to cut the parts out on your CNC.
There will always be luddites who eschew anything but human powered tooling, and technogeeks that prefer to sit back and watch the machines do all the work. Then there are the rest of us, who want to take advantage of the technology that is available, but not let it control the final outcome of our work.
For those of us who do this as hobbyists in our "golden years" (or perhaps more accurately "weather worn tin years" ) making a large initial investment in tooling we may not find as useful as we anticipate it will be, doesn't make sense. "Real CNCs" and related tooling can be quite expensive - some of the diamond tipped bits where I used to work cost more than the price John quoted for his spindle. Finding the balance between "toy" and "tool" is what I am trying to do in my search for an affordable CNC.
My "takeaway" from this discussion is that the 3018 has some limitations and that at a - minimum - requires replacing the spindle to make it useful for any of our work. Thanks to Peter Coombe and to all who have contributed to this discussion and to Colin for bringing it up in the first place.

I agree. A cheap Harbor Freight mini die grinder for a spindle would run fast enough to cut pearl and is light weight enough that small steppers should be able to handle the load, and once you dial in the feeds and speeds to suit what the machine is capable of, it should at the very least be a good proving ground for CNC in your shop. In my case, CNC proved to be somewhat addictive and the cost escalated fairly quickly, but that was when I still had a job and could still afford it! Living on Social Security has limited my tool buying, but that's OK. I stocked up early...


Dave

I glue my pearl to a small board the feed and cut rate are key in understanding how this works.
For a first timer set up to cut air so you can see the travel and watch how the machine moves. In my
machinist days this was a rule it is better than the machine doing something harmful. It will do what you tell it
even if it is wrong
On pearl cut about .010 at about 75 in per min. when done soak it off in a tub of warm water

_________________
John Hall
blues creek guitars
Authorized CF Martin Repair
Co President of ASIA
You Don't know what you don't know until you know it

Yes, cutting air is my first step on any new operation. Makes sure the G Code is OK, but also gives a good look at what to expect in terms of the live cut.

I cut a little slower than John does, but not by much. Depth is the same, and I do the same "glue it down, soak it off later" method, though I've been thinking about using the masking tape/super glue method. I could probably cut faster, but I'm chicken.
Dave

I glue my shell to a piece of wood using CA and accelerator. Once I CNC cut the part, I run it through the bandsaw, leaving about a 1/16” of wood still glued to the shell. Then using a sharpie, I mark the orientation followed by a couple minutes in boiling water. The shell releases leaving zero CA residue on the part. The main benefits of using CA is being able to cut the part immediately, and you get a stronger hold when cutting truly tiny parts (1/4 grain of rice).

Here is the feeds and speeds chart I use. It’s from precisebits.com

That's the same chart I've used, and I've been happy with the results. I've been able to feed faster using a higher speed dental handpiece as a spindle, running in the six digit RPMs, but using dental burs instead of the precisebits shell cutting mills. I can't say that one has been better than the other in terms of final results. Dental burs are a lot cheaper though.

Dave

Where do you get your dental bits, Dave?