I built a Voron

2025-11-13

Related: [3D Printer upgrades]

This is something that should've been typed out as I was doing it, but alas, I had not the energy to do so thanks to my job and other obligations.

I suppose there is a statement to be made that even projects as complex as this can be done during what few hours i got during the time in-between my actual job and the time I must dedicate to rest.

So, for those reading who are out of the loop, a Voron is a 3D printer. Or rather, a single name to refer to a bunch of entirely open designs of a 3D printer. All made by an online community of frankly very talented designers.

To be more specific about the printer that I built, it is a Voron 2.4, shipped as a kit by formbot. There are a few companies out there who specialize in making a single-order package that contains everything one needs to build a Voron, as the default option, initially offered by the design team, is a plain and simple BOM that one specs out for a given printer model they intend to build. From what i've gathered, these companies are collaborating with the design team and are infact also vouched by the design team themselves as reputable vendors of kits, so it's all surprisingly antagonist-free.

The Build

I sorta had an Idea as to what I was getting myself into, but the visceral experience of actually receiving a huge box, opening it up, to find nothing but a loose assortment of parts and baubles sorta.. blew me away. The idea of "this set of random items would eventually become a working robotic system" was mind boggling.

I didn't feel too intimidated, however, I had a tidbit of experience in wiring a 3D printer up, and the rest looked simple enough, on top of that, one of the main jobs of the actual voron design team is documentation, the build guide is both continually updated and also of very high quality. The entire build felt a little like an assembly of a lego set but for adults.

I started out with getting the frame itself built. All it took was screwing some 2020 extrusions together, only caveat was that the guide called for a precisely flat working surface. If the frame was not built square, basically everything else would be outta whack, and my ass was building this in an apartment with a tool collection that i started with the express purpose of getting the printer built. Alas, needs must, and what I ended up using was my bathroom mirror. It worked well enough.

Also, I did not have a measuring tape at the time, and the two struts that would become the attachment points for the build plate had to be somewhat precisely located in relation to the center of the frame. I ended up using a math notebook as my tool of measurement, it actually worked in a pinch.

Once the frame was screwed together, the next item on the agenda was greasing and attachment of the linear rails for the Z axis. Nothing too complex, just had to be careful to not let the linear bearings slide off.

After that, came the building of the Z-drives, or to re-word it, the assembly of the motors which were responsible for moving the toolhead up and down. The specifics and even the motor count is dependant on the design, but the 2.4 relies on four for reasons i'll get into later.

This was my first foray into working with voron-designed 3D printed components, and oh my gourd they are a joy to work with. The Z-drives specifically were really enjoyable to assemble.

I won't type the rest of the build in detail here and only cover whatever caveats and issues I ran into, since there is far too much to cover.

The mechanical assembly did get rather confusing at times, requiring me to make my own calls at a few points, due to the fact that the formbot kit does ship with a few modifications to the base design.

I also recall running into trouble with the X axis gantry assembly, I had not positioned the linear rail at the center, which gave me trouble when trying to attach the printed brackets, one would fit well, while the other would leave a small gap between the 2020 extursion and itself.

I also made a common mistake, specifically with the A/B belt routing. It is really easy to mis-route the A/B belts for the back brackets, so that they would rub against the printed A/B motor mounts. Luckily the guide does tell you to double-check your work at times like this, and I spotted it before moving on, though i've heard stories of people only discovering that same issue after dozens of hours of runtime.

Next up was the assembly of the toolhead, specifically, the stealthburner toolhead that shipped with my kit. It was.. okay. Truth be told, this part of the build felt somewhat fiddly and frustrating, but I can't blame the designers for this, I had to assemble the hotend, the toolhead board, the fiddly LED strip setup, wire the fans, which did not provide enough slack in the wires for things to be comfortable. Also, the cover was hard to screw in due to the sheer amount of wires in the way.

Afterwards came the electronics. The closer I got to the end of the build, the more comfortable the build became. It was an entire evening/night of me just routing wires and plugging things together. No major mistakes made or anything, I feel a little proud of myself about that.

Flashing

One major issue I ran into with the electronics, however, was actually getting the control board to boot. For context, all Voron printers run Klipper, which relies on a combination of a full linux system working in combination with a microcontroller to manage the entire printer. The kit I bought came with a control board that did all of that on a single PCB with a compute module attachment point. Issue was that the thing was claiming an ability to be powered via USB-C for testing purposes but that was not the actual case. I felt stuck and decided to reach out in the Voron community discord about it and I got help. Of the very friendly kind, in fact. There is a whole swath of people who are willing and capable to help you out with a Voron build, if you ever decide to give it a shot.

What ended up resolving the problem was temporarily rigging up the 24V power supply to power the board and re-flashing the OS image for the compute board. I suppose the thing was just not receiving enough power to function properly and that I also flashed the wrong OS image or something.

Also worth noting was that I had a toolhead board that came with my kit. The idea being to reduce the amount of wires needed to go between the control board and the hotend. Old systems had to have like.. 6 to 8 pairs of wires routed for the hotend, it's temperature sensor, fans, a touch probe and an endstop switch. The kit I got ends up using two pairs -- one for 24V power and one pair for data. Data would be sent over CAN. That was a bit of a hassle to set up, and it does require setting special resistior jumpers for the ends of the network, due to how CAN is designed, but, thanks to my general experience in working with embedded systems, the given guides were mostly unnessecary for me. I ended up flashing katapult on everything, which would allow me to arbitrarily reflash both the main control board MCU and the toolhead MCU as nessecary over the already established connetions, wheras the default would require me to use the provided USB ports on for the toolhead and i don't even know what on the main control board.

At this point, everything was ready for a smoke test.

Configuration/Tuning

This was a rather fun portion of the build. Klipper gives the end user a lot of freedom as to how they want to set things up, as it's a one-install solution for many different hardware configurations.

I started off by ensuring that all of the kinematics were correct. Namely, I had to ensure that all motors were moving in the right direction. That was easy enough, and it allowed me to run the printer's first ever leveling routine.

For context, this is specific to printers which have a flying gantry and four motors for their Z axis. The configuraiton allows the printer to perform bed leveling by itself, pretty much. What it does is that it measures Z height at every corner and then makes incremental adjustments for each corner until it measures all corners to be within a certain configured tolerance.

I had to bump said tolerance up on my printer as the routine would fail consistently due to the noise of said measurements. Not by much, however. I think that the current tolerance is still set to be within one hundreth of a milimeter.

Next up from the kinematics is PID tuning of all heating elements. This is important as even a few degrees of difference in the temperature of a hotend or the heated build plate matter and can affect a print. Luckily, klipper provides automatic routines for this, and this process was as simple as running a command in the console for both the hotend and the build plate.

Lastly came input shaping / resonance testing. This is a somewhat recent development of 3D printers that provides a major benefit to the process. Essentially, it's a firmware-enabled feature that allows the printer toolhead to actively compensate for vibrations within the printer itself. This is most visible in cases where the printer is running at high speeds and is making a lot of changes in direction quickly. In terms of getting this set up, all one needs is a gyroscope attached to the toolhead. Some setups offer a temporary attachment point, while others, like mine, have a chip built into the toolhead board. The actual configuration is also as simple as running a command in klipper's console.

First prints

At this point, The printer was well capable of handling basic prints in lower temperature filaments.

Unfortunately, the formbot kit decides to throw a wrench into your plans. See, in order to print with the recommended material, ABS, one needs a printer with an enclosure. ABS is sensitive to temperature shifts, drafts and breezes, so it is pretty much nessecary to have a high temperature environment in order to print without issue.

On the other hand, if you order the 'nessecary parts only' 3d printed part kit from formbot, they do not ship you all of the panel clips nessecary to attach the provided acryllic panels which would from an enclosure.

A chicken and egg problem.

I solved it by running the parts nessecary with PETG which can be printed in open air, building out the enclosure, and then reprinting said parts in ABS.

I also kept consistently running into build plate adhesion problems with ABS and i had difficulties getting my initial Z offset right, but the more I printed, the more I got familiar with how the thing worked and the more reliable my prints became.

Conclusion

not that this is a review or anything, more of a 'where the fuck have I been' kind of post. Also see [this] if you were wondering about that. Overall, i'm not sure if I could reccomend this for someone who just wants a printer. If your goal is to just have something that will work, then by all means go invest in a Bambu printer, they will print reliably, they will probably print better and they will be a far nicer experience.

What a Voron offers is repairability, gained skills, community support and upgradeability. There is a huge aftermarket for voron parts. I can choose any hotend, any control board, any motor, any extruder, and an entire myriad of other upgrades.

The experience in it of itself is hugely rewarding. I would highly reccommend a voron if you are just looking for a project to dive into for a few weeks.