Small CNC milling engraver

In my review of a 3D printer, I mentioned small milling machines that for some reason are more often marketed as engravers. Since this is a pretty interesting topic, and readers later asked me about this kind of machine, I decided to review one.
So, we're putting together another puzzle-like kit, and there's a chance we'll end up with something interesting.

Machines of all kinds are a fascinating subject, and not so much for the results as for the knowledge I plan to apply to something on a larger scale. But it's best to start small.
A milling engraver is essentially very similar to a 3D printer: the same three axes, but instead of feeding plastic, you switch on the spindle motor. The difference is that a printer builds the object itself, while a milling machine leaves the object untouched and cuts away everything around it that isn't needed.
Also, you can't get away with a simple design like a printer's. A printer's print head isn't subjected to any loads, but with an engraver, structural rigidity is crucial because cutting involves pressure on the bit as well as rotation.
That's why it's hard to make a milling machine large and cheap, unlike a printer: the larger the workpiece and the greater the required accuracy and rigidity, the harder it gets.

This machine comes as a kit that still needs to be assembled. I ran into a few issues during assembly, one of which was that there were no assembly instructions in the box. So the research started before the assembly did :)

The machine is a 1310 model, meaning the workpiece can be up to 130 x 100 mm, with a height of up to 40 mm.

All the fasteners were sorted into separate bags. Some bags contained several types of fasteners, others just one, but I didn't have to hunt around for the right ones like I did when assembling the 3D printer. Everything was quite convenient.


Now we can get started with the assembly.
I assembled everything without any instructions, mostly by intuition. Later, I did manage to find both an assembly video and a manual, but they only confirmed that I'd done everything right.
What's more, I initially assembled it as shown in the instructions (as I found out later), ran into a minor snag, took part of the frame apart, and reassembled it my way. The result was even better than the original design.

Just in case, here's an assembly video. It's in Chinese, but the music is nice :)


The entire assembly process will be broken down into steps. Since most of this procedure applies to many devices in this series, I think it may be useful.


Then I screwed on the remaining parts.
At this stage, the main thing is to keep the frame square. I had to take it apart again later anyway, but it's still best to try to get everything as straight as possible from the start.
In general, at every stage of assembly, pay close attention to keeping the structure square and rigid. The screws need to be tightened quite firmly, because there's a fair amount of vibration during operation.



This is what the finished assembly should look like.


This view makes the mechanism described above—and the whole assembly in general—easier to understand.



The result. Technically, the table and Y-axis movement assembly are ready.









The assembled unit looks a bit monstrous, but it’s actually simpler than it looks :)





The whole setup assembled and ready for experiments :)


The engraver is all well and good, but without milling bits it can't cut anything, so the kit includes a “starter set” of four identical bits.
At first I tried to make sense of the markings and match them to my measurements, but I was genuinely stumped by the fact that all three bits had the same dimensions but different markings. It later turned out they were made from broken bits of different sizes, which is why the markings differ :)
I'm thinking of adding spindle reversal for working with bits like these. The bit is symmetrical and has no preferred direction of rotation, so it will cut either way. Reversing the spindle would let both sides wear evenly and could double the bit's service life.
The bit has a very sharp tip that tapers to a point. During testing, I chipped the very tip slightly, and the result was even better. After that, I even used this bit for finishing work.


Install the bit in the collet, tighten the screws, and check for runout.




For something a little more ambitious, I decided to start by engraving a small picture of a giraffe.
I used the included bit, even for the finishing pass. Although using a bit that tapers to a point for this kind of work is the wrong choice. For finishing, you should use a bit with a ball end, not a needle point. But since my bit was slightly chipped, it actually did a pretty good job on the finishing pass.

The spots at the top appeared by accident. I should have set a preliminary plunge depth, so the engraving would start below the surface by 0.5-1 mm rather than at zero, to compensate for the workpiece being uneven.
Dimensions: 26x28 mm; cutting depth about 1.5 mm; finishing step-over: 0.1 mm.


A clip showing the engraving process for this model. Sorry about the video quality; I filmed it on a vacuum cleaner camera.


I ordered some extra bits for the machine, but I'll only show the ones I could fit in this collet.


Next, I used a larger model. It measures 50x35 mm, the model is 5.5 mm deep, and the total milling depth is 6 mm.
Roughing was done with cutter number 4 from the list above. I made 6 passes, each 0.9 or 0.95 mm deep (I don't remember exactly).
This is what it looked like afterward.


Finishing was done with the tapered cutter that came with the machine (see photo above), whose tip I had snapped off.
The milling step was set to 1 mm, the feed rate to 2.5 mm/s, and the plunge feed rate to 2 mm/s (if I remember correctly).
The entire finishing process took 4 hours. I could have increased the feed rate, but that would have affected the quality. To maintain the quality, the machine's spindle should ideally run faster. For my calculations, I assumed 15000 rpm, but there are relatively small machines that run at 40-50k rpm.
I think that, if I wanted to, I could increase the machining speed on this machine by a factor of 2-4; I just didn't want to rush. The higher the speed, the worse the quality and the greater the chance of damaging the cutters.


This is the model I ended up with. The photo doesn't do it justice, so I'll give you a hint: it's Pegasus :)
The original model was considerably larger, but to get the job done faster, I had to scale it down to the specified 50x35 mm.




The level of detail is very high, but you can see marks left by the cutter. The base came out perfectly smooth, while the model itself has marks even on flat areas.
Overall, though, I'd say the result is excellent for this machine; it reproduced the design very well.


Along with the photos, I also recorded a video of the last part being milled. Since the process took about 6.5 hours in total and the video is only 6.5 minutes long, I had to cut it down considerably. I cleaned the chips off the model as I went, so it looks different from one shot to the next.
At the beginning, you can see that the roughing cutter is completely unsuitable for this job: instead of turning the wood into fine chips, it simply peels off a strip of wood.



Now we can draw some conclusions.
First, I'll go over the design issues with the machine and possible ways to address them, in order of importance.
Essential changes.
1. The plain bearings aren't very good—though they're probably just bushings, not bearings. I didn't pay much attention to them during assembly, but later realized I should have tapped them in with a hammer before assembly. That would have seated them deeper and helped them work better. In any case, I plan to replace them with linear ball bearings.
2. The metal sleeves connecting the stepper motor shafts to the lead screws (threaded rods). This is one of the trickier problems. In principle, you can align and secure them with almost no runout, but it's genuinely difficult. I had to wrap tape around one of the rods so it would fit snugly, and only then tighten it, but the problem turned out to be at the connection between the sleeve and the motor shaft.
I recommend replacing them with proper couplings; fortunately, they don't cost much.
3. Plastic nuts.
At first, I planned to put these at the top of the list, but in practice they turned out to work better than expected. I think replacing the sleeves with couplings would make the machine more reliable. The kit also included three spare nuts. You could replace them with metal ones, but that may have its downsides.
This design does have its advantages, though. This assembly has virtually no backlash, and it can be adjusted as it wears.
4. Chuck.
The chuck needs to be replaced. Not because it's bad, but because it's inconvenient. With this chuck, you can only use bits with a 3.175 mm shank, and I'm sure many people have bits and drills with other shank diameters.
5. Lead screws (threaded rods).
It's hard to say what would be better here. On the one hand, these are just threaded rods, and a proper worm drive would be more reliable (I'm not even considering a ball screw—the price is in a different league). But with the threaded rods, the machine has twice the thrust and better relative positioning accuracy. Replacing the rods with a proper worm-and-screw drive means replacing the whole set: both the screw and the nut.

Less essential changes
6. Thrust bearings. There simply aren't any here. Overall, everything works fine without them; I don't even know if they make sense in this setup, but if you can, it's better to install them.
7. Motor. It's more than adequate for this machine. A more powerful motor would make the Z-axis drive work harder, but increasing the speed of the existing motor is entirely feasible. I think you could safely raise the supply voltage to 27–30 V, especially since the power supply can handle it.
8. Remove the 12 V power supply from the design and replace it with a DC-DC converter. I don't see the point of having two power supplies when one can do the job just fine.
9. Add the ability to reverse the motor. This would extend the life of some types of bits.
10. Make all the changes listed above, experiment, put the machine on a shelf, and then
use what you've learned to build something of your own—bigger, more powerful, or more versatile.

What does this kit offer?
1. Good hands-on experience with small CNC machines.
2. An understanding of how this type of equipment works.
3. The ability to use it for practical applications, such as
engraving, drilling circuit boards, and making small parts from plastic or wood.
4. If you stay interested, you'll have the information (and some of the components) to
build your own version of a similar machine.

This router is a simplified version of the machine covered in this review, so much of what I wrote there also applies here. It costs less, but has a simpler design.

That's all. I hope you found it interesting.