Voila, here’s my first CNC router project, a Thundercats emblem carved out of an Ikea cutting board!
I know, it’s not very glamorous or of any practical use, but it’s a start.
Advertisement
The Road to CNC Router Ownership
I wanted to buy a CNC router for quite a long time, and have been going back and forth for several years.
See Also: Should I Buy a CNC Router (2018)
I flip-flopped constantly over this decision. A CNC router would help me build certain projects, and jigs such as router templates for use with handheld tools.
There are cheap CNC routers, affordable CNC routers, and higher quality “how much will that cost?!!” machines. The cheap ones are usually junk, requiring users to invest a lot of time into their machines rather than using them. The affordable ones have some limitations. Higher quality machines are more complex and much more expensive.
I’ve been holding out for a higher quality machine with sturdy framing and strong capabilities. But there is greater complexity – special spindles, specific software you are locked into using, and smaller ecosystems that could result in a steeper learning curve where I’d be mostly on my own.
Choosing the Shapeoko 3
I ultimately settled on the Carbide 3D Shapeoko 3. I also ordered one of their smaller benchtop self-contained machines, the Nomad 883, because it more closely aligned with my needs and seemed more capable for higher precision work on smaller workpieces. I devised an elaborate editorial plan that justified it.
Advertisement
Here’s my rationale – the Nomad 883 could be an office machine, working out parts for ToolGuyd needs, and the Shapeoko 3 would live in the garage, doing all the larger workpiece 2D machine type of stuff.
I went with the Shapeko for two reasons – first, it seems more robust and thoughtfully designed than others on the market, and also because the inventor (now a part of Carbide 3D) left a very positive impression with me the few times we chatted via email a few years ago when I first inquired about an earlier generation machine.
I was offered an X-Carve machine for free a while back, but declined at the time due to lack of space. When I did have space, their PR changed and I was told they were out of samples. X-Carve vs. Shapeoko is a very common comparison when it comes to beginner CNC router machines, and Carbide 3D’s page on the matter sealed the deal for me. At the time of this posting, it doesn’t appear you can eve buy the smaller X-Carve 500 anymore, only their full-size model.
I opted for the smallest of their 3 sizes, figuring the smaller footprint and 16″ x 16″ working area would be enough. Maybe this would be all the CNC router I need. If not, I can upgrade to their XL or XXL kits in the future, as they do offer upgrade kits. Or, I could put that money towards a larger and more capable machine in the future.
Minor Setbacks
There were snags.
I started assembling the machine more than a year ago, several months after I purchased it, and found that a rail was deformed and a v-wheel defective. They sent replacement parts promptly, but I struggled to find the time to take everything apart and swap things out. So, my partially-assembled machine sat on top of a large-enough tool box, and for a long time.
I finally got to getting my Nomad 883 and Shapeoko 3 operational these past few days, following two weeks of steady work cleaning and reorganizing my garage workspace.
I hit another snag, literally. The Shapeoko 3 doesn’t have much in the way of cable management, unlike the Shapeoko 3 XL and XXL machines, which have drag chains. I suppose for a machine this small, and with the electronics box located right behind the Z-axis, it’s not needed. But during rapid movement as I was getting to start on my self-designed tutorial process, the machine snagged and ripped the wires off the y-axis limit switch.
So, out came a soldering iron, where I quickly patched things up. I’ll likely have to redo this repair in the future with a little more finesse. But, it works and got me back on track.
Everything was also delayed because it turned out I was missing mounting hardware and custom brackets from a 3rd party dust collection attachment I ordered (from a different company) at the same time as the machine. I should have been more diligent in checking this when I first received it. I did see a hardware packet, but it turns out that only included the small clamps I ordered at the same time. An email to the company, and the missing “Suckit” dust shoe hardware was on the way.
You can probably tell, but I like things to be *perfect.* And when they’re not, my hesitation tends to become a firm barrier.
About Assembling and Running the Machine
Things aren’t quite perfect now – I need to build an enclosure, tame the wires a bit better, and find a way to support a dust collection hose so that I’m not hand-holding it the entire time – but the entire process did go smoother than I expected.
When I tried 3D printing for the first time, the machine I bought couldn’t print the test print, and I gave up. Here, despite the minor setbacks, the Shapeoko 3 assembly and setup was about as easy as I could have hoped for.
The first tutorial involved zip-tying a marker to the Dewalt router I clamped into the machine, and it worked perfectly.
Wow, was it really that easy??
Side note – Ben’s working on a Benchtop CNC Router Basics post, and so I’m deliberately avoiding going into too much detail here. He has one of the larger Shapeoko CNC router machines and far more experience with it. If you have questions, please ask away, and we’ll make sure they’re addressed here or in that post.
My First Project
What to do next?
Getting the hardware up and running is Step 1. Next, I have to learn how to use the machine. There’s only so much you can learn by reading. I’ve read all about CNC router capabilities and shortcomings over the years, but this – actually starting a project – was new territory for me.
Oh, I have a lot of “this would be great for the CNC router!” projects in various stages of planning, but first I need to learn how to use the machine, as different factors can affect the 2D and 3D modeling processes and design approach.
Carbide 3D has their own software for modeling and machine control, called Carbide Create and Carbide Motion, respectively.
Rather than try to design something simple, I decided to find a logo to carve into an HDPE cutting board.
Actually, I started to work out my own simple 2D design first, involving concentric circles, but I quickly hit a roadblock when trying to work out the intent. All of a sudden, my circles turned into a bit holder and my desire to learn the machine started to require my also learning the design software instead. So why not start with something decorative?
I conducted a Google Image Search for a Thundercats logo/emblem. If you click “advanced search,” you can opt for svg files. SVG files are 2D vector graphics, meaning they can be scaled and easily translated to machine code.
I could/should have simply done a similar carving with the ToolGuyd logo, but that would have required retrieving files from a different computer, rather than being able to do everything from the laptop I brought with me to the garage. If I went back in to another computer, I would surely have been distracted, pushing the project to another day. Future designs will likely be worked up and converted on a different computer, but for this tutorial-in-nature project, I went with the first “what I can do here and right now” idea.
You can create SVG files with free or paid software as well. Simply conducting a Google image search or general search for “keyword” and “svg” resulted in a mess of etsy listings and other marketplaces where they are selling such graphics. Opting for the Google advanced search where I selected svg as the file type brought up results where someone already did the hard work of tracing an image file into a vector graphic.
I did similar work in the past, when learning either Inkscape (free software) or Adobe Illustrator (paid software), and have all the files on a different computer somewhere. It’s not terribly complicated – you convert a jpg or other image file and then tweak nodes slightly if or where needed. But for this, I wanted an image that was ready to go.
Okay, so I had an svg file, imported it into Carbide Create, and scaled it down to fit in a 4″ x 4″ space.
The file had three shapes – the outer circle, the “cat” head, and the eye.
It then took some work with Carbide Create on how to figure out the toolpath commands. Ultimately, I realized selecting the cat head outline and the eye allowed me to pocket around the eye.
It took a few tries…
I also learned that I set my Z-axis off a little with the probe, resulting in a Thundercats symbol being engraved into my machine’s MDF support board.
The Ikea HDPE cutting board I was using (I have a lot of these, purchased for cheap jig, workshop, and project materials) is maybe 5/16″ thick.
I ultimately set the stock thickness to 0.300″. I set the inner components to be cut out at 0.200″. and the other circle at 0.2800″.
I ran into trouble when I output everything as a single file, as there was no prompt for re-probing for Z-height zeroing between tooling changes. That could be a Carbide Motion limitation, or something I need to program in via G-code, which is output via Carbide Create for Carbide Motion to follow.
So, I used Carbide Create to output two files, Part 1 and Part 2.
Part 1:
- Route outside the eye using a 1/16″ end mill
- Route inside the cat’s head using a 1/16″ end mill
This was done to create the shape for the eye, and for finer details for the cat head outline.
Part 2:
- Pocket the cat’s head using a 1/8″ end mill
- Cut out the circle using a 1/8″ end mill
I felt that a 1/8″ end mill would allow for faster cutting, and allow for finer details to remain compared to if a 1/4″ end mill were used.
Much More to Learn
As with any tool, there is much more for me to learn.
But hey – this is a good start, right?
There is also much to do regarding optimizing the machine environment, such as adding in better user controls (such as a physical emergency stop), limiting dust, and lessening the noise output, but now that I have my first taste of what it can do, I’m not letting myself get hung up on small obstacles anymore.
Paul
Shine it in the sky. Thundercats will come running. ?
Leonard
It’s a great start.
I started down the path of owning a CNC mill years ago. I was partially employed and I would get side jobs to build prototypes. I would do everything on a small manual mill, drill press and band saw. Then the inevitable, they would say “Can you build 5 more?”
So I bought a CNC mill and set it up. There is sat for 4 years because I got great full time job that week.
Fast forward. I finally decided that I needed to learn this. I’m already very familiar with CAD programs and I model in 3D for a living. I wasn’t familiar with CAM programs. It’s been a multiyear process. I have a great relationship with both my CNC and CAM software companies. Always there to answer questions.
Here’s some of the things that I have had to learn, or bolster my knowledge.
Speed and feeds. You may have an easier time since your not cutting metal.
Workholding.
Chip containment or dust collection.
Coolant and chip clearing. I don’t think you need this.
Path design. Making sure that the bit doesn’t hit the clamps or fixtures during travel.
g-code.
It’s fun and rewarding.
My first project was a Logan’s Run flame pistol prop.
Leonard
Oh yeah…
Finding the X,Y,Z. Accurately and quickly.
Stuart
That’s one of the benefits with starting with a system like this than say the “CNC Router Parts” benchtop CNC that I’ve been drooling over. Their controller has moved from Mach 3 to Mach 4 control software, which I’ve heard a lot of issues about, but my understanding is that Mach 3/4 requires much more user input when it comes to feed and speeds.
With Carbide’s software, they apparently have a library for the bits they sell, which eases the learning curve a bit. When designating the tool paths, I select the material and then the end mill or bit, and it comes with default settings.
One problem is that woodworking router motors can’t go as low as they recommend, and so that’s something I’ll need to sort out in the future.
There’s a lot to learn, and I’m sure more mistakes to make, but right now I’m happy to seemingly be past the initial roadblocks.
Lots of experimentation to come.
Leonard
I have the exact opposite problem. My mill wouldn’t get to the RPM that all the website tooling recommends. I upgraded my VFD from the manufacturer and was able go from 2000 RPM max to 3000 RPM max. That’s been a huge difference in the quality of finish.
I can feed at the proper rates but that have left me with a pile of snapped bits.
Still so fun…
Clinton Caraway
Adding a HDZ (Heavy Duty Z-Axis) and 1.5 or 2.2 kw spindle is a expensive but absolute game changing accessory. The Shapeoko is a fantastic tool but the addition of a linear bearing/ lead screw driven Z-Axis along with a real spindle gives you the ability to control the speed of spindle from 1-24000 RPM…..plus the extra rigidity to use 1/2″ dia end mills.
Stuart
Thanks!
I know there’s also the Z-Plus, which is in between the stock z-axis and the HDZ. A spindle will cost quite a bit more, and I might need a dedicated outlet in case the VFD doesn’t play nicely with my GFCI outlets.
At this time I’m not thinking about any upgrades, lest I fall into the “ooh I can make this even better” hole. Once I put much more time into the machine, then I’ll be in a better position to think about these things. But also, hundreds of dollars for upgrades might not make sense if the money can be put aside to go towards a bigger and better future machine.
James
We need a picture of the finished Logan’s Run flame pistol. It sounds awesome.
audiovideofreak
It’s a shame the making of the movie wasn’t as much of a success.
RCWARD
Have to admit none of that sounds like “fun” to me. I’ll pass
Stuart
For me, there are things I want to make that I can’t do with the tools I have, at least not very easily.
You have to walk before you run.
The example above limited the number of things I would have to learn concurrently.
I will probably have to go through a couple more tutorial-like projects before I can get to the stuff I had in mind when I purchased the machine and drooled over others in the past.
I’m finding it’s a lot different than say cutting a sheet of plywood down with a table saw.
Clinton Caraway
Welcome to the Shapeoko community.
If you haven’t discovered the Facebook group then you are missing out on the best part of owning a C3D product. It is amazing how the staff of Carbide 3D interact with the customers via social media. There are also the users of the Shapeoko and Nomad that help newbies work through the learning curve. It is a amazing experience!!
Ben V
Hey Clinton, I follow the Carbide3D forum, it’s pretty active. What’s the name of the FB one?
Ben
William Adams
It is the “Shapeoko and Nomad Users Group (Unofficial)”:
https://www.facebook.com/groups/unofficialshapeoko/
William Adams
Also, your project was mentioned there:
https://www.facebook.com/groups/unofficialshapeoko/permalink/851208401953787/
(but you’ll need to be an approved member to see it since it’s a private group)
Chris Freitag
As a coach, I’ve been teaching high schoolers to use a Shapeoko 3 for a while now. We cut a lot of acrylic and polycarbonate, some 1/8” aluminum. Best upgrades: Fusion360 for CAM, and CNCjs for sending g-codes. Used foam pipe insulation as a quick and dirty cable guide. Future upgrades: drip-oil for aluminum, dust collector.
Plaingrain
I see these CNC machines rapidly evolving in the coming years. And the same money buying getting you much higher tech. They sure interest me though.
Stuart
CNC routers and mills are fairly mature technology.
I’ve been watching the hobby scene for a while, and there are a lot more entry-level options now and with fewer people building their own.
Controllers have come down in price and there are more options.
But as with 3D printers, companies that started at entry-level pricing have upgraded their machines to be better, which also means pricier.
There’s a limit as to how inexpensive these machines can get without significant sacrifices to even introductory quality.
JoeM
*Looks Down at His Custom Made Thundercats T-Shirt* …Stuart… Don’t take this the wrong way… But I Love You… Your choice of project fills me with joy…
Jp
I want to learn how to use these, but every time I consider it, I read and I am overwhelmed. I was never good with 3-d visuals in my head, and I am allergic to math. That removes ratios, scaling, and damn near anything else someone may need to learn. Mostly everyone says that the programs are hard to learn unless you spend on a machine which is too limited to be useful for anything other than trinkets/baubles.
JoeM
Gonna sound weird… I totally understand if you don’t feel comfortable with this.
Try your local Public Library. In the past… I’d say 10 years? Across the US, Canada, and the UK, Public Libraries have had 3D printers, CNC machines, and the odd Laser Cutter donated to them for public use (and some revenue stream as well.) and usually straight off the factory line from the company that makes it.
As a result… Some libraries have Summer or Seasonal lessons for individuals (for free) tutoring them in how to use the programs involved, and how to use the machines. Also, if you have mastered the design already, you can send them the file to be made, and they’ll run your final project through their own machine for a dirt-cheap cost. This may well give you a nice bridge for a learning curve to get you into your own machines. AFTER you’ve made use of the experience gained working with the Library’s machines.
Hope that helps a bit.
Stuart
Here’s my plan:
Steps 1, 2, 3, 4 – find simple projects where variables are constrained.
Steps 5+ – work out simple practical projects
Step 10+ be able to design parts I can use, in ways I can make them.
Carbide 3D’s Create software looks to be free. Give it a try. https://carbide3d.com/carbidecreate/
Inkscape is free – give that a try. https://inkscape.org/
2D is easier than 3D, same as with other tools.
I have some prior experience with 2D and 3D modeling. You can use Sketchup for learning basic 3D modeling.
Software such as Fushion 360 is very powerful but also has a steep and long learning curve.
The best way to learn software – model what you need to make.
2D and 3D modeling is required for CNC work, 3D printing, and other digital fabrication methods. But it’s also very helpful for all kinds of project work.
You’re going to eventually have to learn the math, but a little at a time. For this project, I set the scale to 4″, which didn’t requite any math. I came up with cutting depth, but it didn’t require math – I measured my workpiece and came up with a smaller value.
I mentioned Inkscape because there are so many different tutorials and how-tos out there, and also physical books.
2D and 3D modeling and vector graphics require some learning, regardless of the machine you need.
As I’m learning, toolpaths and what-not have an added level of things to learn.
William Adams
For the tool change issue, one has to output separate files per tool as noted at: https://docs.carbide3d.com/tutorials/tool-change/ unless one has a BitRunner.
Knowing or learning vector drawing techniques will go a long way towards working with Carbide Create. In addition to the video tutorials on our docs pages and a static web page, there is a keyboard shortcut PDF: http://community.carbide3d.com/t/cheat-sheet-for-carbide-create-and-motion/7839 and I wrote up a bit at: https://willadams.gitbook.io/design-into-3d/2d-drawing
Naturally if anyone has any difficulties or issues or queries on using the machine please write in to [email protected]
Stuart
Thank you!
I unfortunately realized that the hard way, albeit it’s a lesson I only had to learn once. I skipped on most of the online tutorials, lest I be distracted and end up with 20 different tabs of information I would forget once I stepped back in front of the machine.
I got around the limitation by branching my one file into two separate ones, but then realized I could maintain a single design file and disable certain toolpaths in order to output different G-code files.
Did you mean the BitSetter? That might be in my future, but in the meantime the process of splitting a project into two separate components isn’t enough of a hassle to warrant it. Maybe it’ll be an upgrade once I’m more experienced with the machine.
William Adams
Yes, BitRunner, my apologies for the confusion.
Ferreri
Oops again! BitSetter
William Adams
Yes, BitSetter — third time is the charm.
William Adams
For the wiring management on the Shapeoko 3 — yes, the smaller machine has efficient enough wiring that drag chain isn’t necessary and it usually doesn’t get snagged.
FWIW, I put spiral wrap on my wiring back when my machine was SO3 sized (before upgrading to XL).
Nathan
at first look I thought the picture was a Holden logo and I got excited.
meanwhile I have questions about the shapeko and I notice there isn’t much discussion on price. Is that a dewalt router in the machine – will it take any or does it come with that.
does the router come out for other use
Perhaps I missed it but how long did your final test take to run? Looks like a fun product
Stuart
The Shapeoko does not come with a router motor or spindle.
You can buy their compact router motor for $80, a Dewalt (or Porter Cable?) compact router, or a Makita compact router.
If you can figure out easy temporary cable routing, the motor pops in and out with relative ease, and so you can use it normally.
I also bought a 1/8″ bit holder when I ordered my machine from Carbide3D, and it’s Dewalt-specific. There are also 3rd party collets you can buy.
The Shapeoko compact router looks to comes with both 1/4″ and 1/8″ collets.
If I recall correctly, the final test run was maybe 14 minutes or so – around 4-5 minutes for the outlining and then 9 minutes and change for the pocketing operation. That’s why I went with a 1/16″ bit for the outline details and 1/8″ for faster material removal, otherwise it would have taken a lot longer. The model was scaled to 4″.
William Adams
The Porter Cable 450 while a fine router (and the same diameter as the DeWalt DWP611) lacks variable speed control — it’s on/off — so is only appropriate if planning on implementing a VFD/Super PID.
Just for the record here — Carbide 3D no longer carries the 1/8″ precision collet for the DeWalt.
Also, there is a 3rd party which has these (and other sizes): http://elairecorp.com/dewaltroutercollets.html#dwpanchor
They also have precision collets for the Makita in a variety of sizes (which also fit the Carbide Compact Router): https://elairecorp.com/makitaroutercollets.html
Matt
Almost bought a Shapeoco but then came across the Onefinity. No belts or limit switches, just a few minutes to setup and seems to be a very built. Anyone have experience with the Onefinity?