It is probable that many of us have noticed a variety of very cheap CNC mills in the pages of Chinese tech websites and been sorely tempted. On paper or as pixels on your screen they look great, but certainly with the more inexpensive models there soon emerges a gap between the promise and the reality.
[Brandon Piner] hopes to address this problem, with his CNC Mod Pack, a series of upgrades to a cheap mill designed to make it into a much more useful tool. In particular he’s created a revised 3D-printed tool holder and a set of end stop switches. The tool holder boasts swappable mounts on a dovetail fitting with versions for both a laser diode and a rotary tool, allowing much better tool positioning. Meanwhile the end stops are a necessary addition that protects both tool and machine from mishaps.
The same arguments play out in the world of small CNC mills as do in that of inexpensive 3D printers, namely that the economy of buying the super-cheap machine that is nominally the same as the expensive one starts to take a knock when you consider the level of work and expense needed to make your purchase usable. But with projects like this one the barrier to achieving a quality result from an unpromising start is lowered, and the enticing prospect is raised of a decent CNC machine for not a lot.
Bringing up the machine to a much better level take much better parts:
1.the spindle motor
2.the bits/drills.
The spindle motor 11.000 rpm
https://wegstr.com/image/cache/catalog/ZBO%C5%BD%C3%8D%20WEGSTR/Nahradni%20dily/vreteno/1-1920×1080.jpg
The bits: https://wegstr.com/30%C2%B0-spiral-carbide-engraving-bit
https://wegstr.com/30%C2%B0-spiral-carbide-engraving-bit
WatchiImpressive videos on quality made pcb milling machine and perfectly made cuts.
https://www.youtube.com/channel/UCu7mD9RQqvZgw_qK-ET82VQ/videos?disable_polymer=1
#1 critical thing lacking on most CNC builds is machine rigidity.
Frame design is 90% off what makes the thing work properly it doesn’t seem like it should when you look at it but that is where most of these fail in good cutting.
#2 it’s definitely motion effector. If you can get a hold of ball screws and linear rails scrapped from something small it will go a long way
#3 is spindle that’s balanced at high speed, and good collets
#4 is well selected cutting tools
What is the purpose of taking a crap machine and changing the majority of it to make something half decent? Why not just save for a little longer, support a company that makes solid products and come out with something useable that will last?
If the point is to build your own, it’s even more pointless to buy a cheap machine.
I will never understand it.
Because people like playing. If you have a job to do and you need it done right than buying a good tool makes sense. However if you want to play around with something, something cheap and imperfect is great. Some people like spending time in their workspaces tinkering. Some people like devising a design and send it to a reliable tool that will work. Both are fine ways of doing things, they are just different mindsets.
Why not start with quality electronics and just build your own machine from scratch then? Between this and that other post about an auto cutoff to keep trash 3D printers from catching fire it seems ignorant to continue to support the people turning out very low quality and dangerous control hardware.
Because just the motors for a higher quality machine are going to cost more then this kit, add a better spindle and the frame and you have easily doubled the price, ball screws instead of t-scews and you can triple it, and you still need the electronics which unless your using an arduino is gonna again cost much more for what many of us is a trial run hobby or tinkering project. Getting the 3018 to be able to do PCB’s is really just a challenge to attain, there are easier and cheaper ways to make a PCB.
Now that I have had my 3018 about 7 months I have figured out that I can’t justify spending the money for something bigger/better in this area right now, which means I’m glad I went this route since the better option was pretty much 2k and that means I still have 1700 to blow on other projects. I will probably build a MPCNC this summer sometime with my trash 3D printer that has never caught fire in over a year of use, but again that will just be a personal challenge really
I politely refer you to the title of this blog.
There always seems to be two kinds of comments:
* “This is not a hack”
* “Why hack when you can buy”
I politely refer you to read my post. I didn’t say don’t hack and buy (but rather to use quality electronics if you’re going to build your own machine), but this trend of buying very low quality hardware seems like a waste of time. If you’re going to rework the electronics why not just make your own?
Any comment that doesn’t fall in line with the love for cheap hardware seems to always be met with a comment like yours. I’m sorry we see things differently, but that’s okay.
I feel the pain, bro. But it will never be cnczone > had , so we need to bear with crappy cnc setups in exchange for interesting electronic designs.
I have a smaller version of the same machine and two upgrades that really had me liking the machine more was to add a raspberry pi zero w as the controller using serial port json server and chillippr. I was in constant fear that the machine controller would stuff up a usb device on my computer, I looked into a usb isolator and the chip alone was more expensive than the $15 I spent on the pi.
The other great upgrade I got was just using a er11 collet on the machine. Now it does come with some downsides, my z height is basically 10mm but for a PCB mill that’s not super important.
I just finished adding a webcam to the pi zero and streaming using mmjpg streamer and that’s been a fun little side project. I like to think of these machines as like the anet a8 of CNC machines. You need to do alot of stuff to them to make them safe and useable but once you do they will get you where you need to go.
I think many people buy these machines and are severely disappointed by them.
Some of them are nothing more than a Dremel type tool strapped to an aluminium frame. They should probably more accurately be called engraving machines. They will work quite well for engraving your name name on a sheet of soft brass or cutting some plastic but that is all you can expect.
Often cheap DIY tools can be adequate for the needs of an amateur but in the case of a CNC machine it is usually not the case. The problems are much more than not having end stop switches.
The screws are not usually ball screws so there will be severe backlash, meaning that circles will not come out round and that repeatability is lost, with each cutting pass there will be slight movements in the position making it impossible to get good dimensional accuracy.
The beds and rails are not flat, to some extent this can be worked around by placing a sheet of material on the bed to act as working surface and ‘face’ it by running it under the cutter to compensate for variations in the bed & rails.
The machines are not rigid enough for hard materials, you can cut modelling wax softwoods and plastics quite successfully but most people expect to be able to cut metals and it just isn’t going to happen.
The ‘spindle’ especially if it is one of those Dremel type machines really does not have either the power or the accuracy to do a decent job. The bearings do not keep the cutting tool straight and balanced causing again causing straight edges to not be as straight as you hoped.
If all you want is something to play with, making yourself a few name plates and plastic objects where the accuracy is not that important then something like this will work ok. but if you want to do something that requires accuracy or able to cut metals don’t start here. You will need to take a big step to a much more expensive machine that includes a proper spindle, ballscrews, linear rails and a good rigid design.
KT data mill 700, vertical 3 axis has #50 taper 30 HP motor,96inch x axis this machine is old but will hog titanium. All day long year after year machine can handle crashes all day long too. Add a fanuc 15m control and learn your trigonometry and you are ready to rock and roll.
I’ve got one of these, though it came with a better spindle and chuck.
I was amazed how good it performed, especially at the price.
However, the lack of limit switches is a clear issue. Saw this project and was really excited until I realised they’re 3D-printed parts, not CNC’d. Doh!
Now designing my own CNC’d parts that the machine can make itself.