[Reiner Schmidt] was tired of renting an expensive 5-axis CNC head for projects, so he decided to build his own. It’s still a work in progress, but he’s made remarkable progress so far. The project is called Bridge Boy, and it is designed to use a cheap DC rotary mill to cut soft materials like plastic, wood and the like. Most of it is 3D printed, and he has released the Autodesk 360 plans that would allow you to start building your own. His initial version uses an Arduino with stepper drivers, and is designed to fit onto the end of a 60mm arm of a standard 3-axis CNC, so technically it’s a 3+2 axis CNC. With the appropriate software, it should be able to work as a full 5-axis machine, though, and it should be possible to integrate it with a CNC that has a 5-axis driver board without too much effort.
This is definitely a project to watch, as his work so far is very nicely designed and doesn’t look like it would be too difficult to print and build once the design is a bit further along.
[Via Reddit]
This is freakin awesome. Definitely what 3d printers ought to be used for. I’m grabbing those plans for sure. These kinds of projects make machining accessable to many people who otherwise wouldn’t get to experiment, and that’s fantastic.
>3d printed
>milling
pick one
Rigidity is crucial. Can he get it rigid enough?
no
Depends on what he want to cut, how fast it need to be cut, and what precision is required.
Probably kinda OK for cutting foam.
Also might make sense if you want cheap hardware that you can use for to learn yourself 5 axis toolpaths. Or you can scale it up for education. Imagine classroom with 10 cheap 5 axis CNC mills (but then you have to hope, that it can survive lots of crashes).
That’s what *she* said!
Did none of you see the part where he says he has no intention of cutting metal with this? Machine snobs.
Apparently none of the machine snobs paid attention to the part where it is made clear that there is no intention to cut metal and equally apparent is their own lack of skill with a milling machine, you have to select a cut that the machine is capable of, look at the accuracy that can be achieved with floppy old human hands mounted on equally floppy arms.
if you’re milling foam it’s fine.
I think, this can be 3d printed and then lost-PLA casted from aluminum. That might add some rigidity. If you can make it precise enough.
This. ^^
Why bother casting it? If you’re making this to put on a 3 axis CNC mill that means you have a 3 axis CNC mill. I don’t see any parts that couldn’t be made on a 3 axis mill with some smart work piece clamping and tool paths. (You might need an indexing head to do the belt pulley.
This is exactly what I was thinking when I saw the project. I’d love to give it a try but I’m not an expert when it comes to CAD.
My understanding is that there is roughly a 2% shrinkage factor when doing lost-PLA to Aluminum. If anyone out there is interested in helping me figure this out I’d love to give it a shot. Yes, my interest is in building a unit that can take larger rectangular foam shapes and carve them. A platform link this would be way cheaper than anything I’ve seen online.
you know you can have both right?
Regarding the component you are worried about shearing when printed in a certain orientation:
I would split the part in two through the middle length wise, then glue them back together using a high performing glue like superglue. That way you don’t have the layers working against you.
That doesn’t fix the problem because the next layer then becomes the weak point. The correct way to avoid weak spots caused by layer orientation is to orient the part in printing such that the stress is never across layers. The parts he has made here are basically ‘L’ shaped. There are three ways to orient this for printing:
1) Vertically, so that as it is printing it forms an upright ‘L’. This will break at the corner.
2) Laying on its “back”, that is, where the shorter leg of the ‘L’ is sticking up. This will break at the corner.
3) Laying down flat, where both legs are flat on the print bed as it prints. The stresses do not pass through layer boundaries, and the part will not break at the corner.
Or, if you look closely, you can notice that this isn’t FDM printed, as all the 3D print comment seem to assume. It looks a lot like SLS Nylon.
It is SLS nylon, just like he says in the video, between sniffles…
Yes, he says in the video that it’s SLS, but he also says he’d like to know how it works out, if anybody builds it in ABS, indicating filament printing.
This is going to look really brilliant, right up until the tool bit touches the thing it’s supposed to machine.
At that point, it will explode, because you’re trying to make a machine tool *out of plastic*. Not even just plastic, but 3d printed plastic.
Unless you only ever want to machine styrofoam, it’s pointless.
Nothing wrong with milling Styrofoam, and I dare say designing it with plastic before getting the parts in metal is a great idea.
I’ve seen people successfully machine things with very weak contraptions. Is this yet another “I know it all” outburst or do you have enough experience and knowledge to actually know this to be true? IOW can you show that the forces when cutting anything stronger than styrofoam will make this fail? Given your use of the word explode I think not.
The problem that is obvious is that of precision: rigidity or rather the lack of it will make the cutting tool wander, smart software can compensate for this slightly (essentially calculating deflections and trying to counteract them).
But not everything need extreme precision.
Exactly correct Sir.
This is so predictable. Every time someone comes up with something resembling a mill or lathe FOR CUTTING SOFT WOOD AND PLASTIC, using any material other than steel or cast iron, weighing less than their father’s first automobile, all of the wannabe machinists on the Internet declare it useless. And then the people who build these things, cluelessly go on building them and making use of them anyway.
Not all of us are wannabes. Getting paid as a machinist does technically make me a professional, doesn’t it?
That being said, this project seems like a lot of fun. Its all about what kind of material you intend to cut, how many you want to make, and what tolerances you intend to keep. It gets really expensive really fast at the limits.
Well, I never worked in a machine shop, but I did a lot of machine work in my friend’s machine shops. I’d say this is a pretty good prototype, which can be digitized for a 3 axis CNC. For that matter it would work to make a drawing of it and hand crank it out of steel. I’m in the process of converting a Rockford 2 1/2 into a ball screw CNC. That’s the 1st stop, 3 axis. This old Rockford is also horizontal which makes it a pretty good stub lathe. So once it runs 3 axis, it will make everything else, which will include a top drive somewhat like this thing.
Could always mill Styrofoam and then use it to do lost-foam casting to make metal parts.
https://photos.app.goo.gl/SXI6ZJbiZzUVEMoJ3
Was milled on a cnc router, made with 3d printed parts and electrical emt. Folks at MRRF this year probably saw it. So yea, you can cut metal with a 3d printed mill, just go slow.
I’ve been making round stuff (& some Square) on a 6″ Craftsman lathe for 40 years. The secret? Go slow, take small cuts.
Machine the same parts in foam or wax and then cast them …
As my other comment says, this looks like SLS Nylon, not some ABS/PLA FDM print. Your tool will have a hard time cutting it, it won’t explode like PLA can do.
I think now that he has a working model, it would seem that the G code from the printer could be used for a tool path to machine this thing out of metal, or as I said before digitize it with a probe.
“With the appropriate software…”
That could be the killer. Anyone know of any affordable 5-axis CAM software? I’m guessing that if the OP has been renting a head then it’s not an issue in his use case.
Fusion 360, CNC Toolkit, or LinuxCNC
Does Linux CNC do cam (beyond 2.5d) now?
“let me just turn this off” . Bro, just turning the power off like that on a RPI is the best way to corrupt your uSD card. It’s not like Arduino where you can just turn the power off. Think of it like a computer, you just cant unplug it.
New RPIs aren’t nearly as bad as the old ones for corrupting on power loss. I still try to turn them off, but I’ve also just pulled the plug plenty of times and they’ve never ate the SD card like the old ones would do every single time.
Most likely they cut out the CPU before the SD card then. SD cards are extremely prone to corrupting when they have a power loss during a write. And, as you corrupt the internal administration, the SD card becomes useless afterwards for anything reliable.
So, if you cut out the CPU and keep the SD card powered for a bit longer, it can finish it’s write action and corruption be much less likely.
It occurs to me to wonder if FDM printing could take advantage of additional axes, to print objects with challenging geometries, without the need for supporting structures. Of course, the nozzle always has to point approximately down, but this can be managed by mounting the hot end on an X-Z carriage as is typical, and putting more degrees of freedom on the build plate, some geometries could be accomodated.
Or, how about a 3D printing “lathe”? This would work by first printing a core on a base oriented in the typical way, then turning the base 90 degrees so that core is horizontal. The base is then rotated for the remainder of the print, building outward from the core.
What may be more cost effective for threaded inserts.
PCB standoffs. Maybe file them into a wedge, rough up the surface or even just add a little material to the print, so they stay in place.
Slightly chew on the standoff with a cutter or something, will make nice serrations on the metal standoff , then insert in the plastic
What a cool project! Good luck with it and keep us posted. The belt cover idea is a good one as swarf will go everywhere. I’d also be interested to see how hot the spindle motor gets even with the fan on top. Hopefully there is enough moving air to keep the nylon from becoming too flexible. Great project!
If it is anything like the one that came with my CNC3020, the accuracy of those cheap spindles is greatly improved by preloading the bearings. Ideally with some shim stock, but just a wave washer works well enough for me.
As for the durability of the 3d printed parts, maybe once it is finished the most critical parts can be cut out of solid plastic.
I think the fact that the guy modeled the pieces with fusion 360, went ahead and got the parts printed and assembled it into a working machine is just fantastic. The amount of time and thought that went into this is substantial. After messing around with Fusion 360 myself and making a few simple models, I can tell you Reiner is on a level I can only dream to be on someday…. Anybody making negative comments about quality content such as this are just trolling and they themselves know it.
Great project. I have been into CNC as a hobby for many years and would like to try using 5 axis. The barrier to me is the cost of 5 axis cam software. Is there 5 axis CAM software that is cheep or open source that a hobbyist could use?
I’ve been working with 5 Axis CNC machines for over 20 years and they still fascinate me.
I need to upgrade my garage so I can biuld one of these bad boys.
Thanks for sharing such informative post with us. It will definitely help to maintain CNC machine. Looking forward to some advantages over 3-axis machining.
hello great job.
Could you put a list of the materials that are needed?
Thank you
Parts list would be nice as i see a lot of McMaster-Carr boxes.
If you changed it to accept and aluminum plate epoxied to and screwed into place you could get some crazy rigidity out of it. Laser cutting flat parts are cheap.