CORE One Mini - Back Assembly: Almost There & Yet So Far

Time for another update on the CORE One Mini - and this one once again feels like a proper milestone. While I skipped ahead a bit earlier to work on the CoreXY assembly and the heatbed assembly, it’s now time to focus on the back assembly, which is mostly completed at this point.

It's starting to look and feel like an actual finished machine rather than just a collection of parts.

One of the bigger steps here was completing the bent sheet metal part - by far the most complex component (in terms of sheet metal manufacturing) in the entire build. Seeing this part installed really changes the overall perception of the project. It adds rigidity, improves consistency and gives a much better impression of what the final product will look like.

Corrections and Broaching Nuts

There are still quite a few details to refine, especially on the side of the 3D printed parts, but overall the majority of the structure is now in place - giving, in principle all the necessary parts for a working machine.

As expected with custom sheet metal work, not everything was perfect on the first try. I’ve been going through a full round of correcting hole positions, both physically on the parts and directly in CAD. Due to slight variances in bend radii, some of the holes intended for M3 threads did not align properly.

Since I anticipated this, I intentionally designed them with a diameter of 2.5 mm (the nominal core diameter for M3 threads). However, because 1 mm sheet metal is too thin to reliably cut threads, broaching nuts are required in these locations. The nominal diameter of 4.2 mm for the broaching nuts gave me enough tolerance to carefully adjust the hole positions with a file before drilling them out to the final required diameter.

The goal here is to ensure everything lines up cleanly for assembly and future revisions. Again, this means double-checking tolerances where multiple parts interact and making sure no stress points are created where parts would otherwise need to be forced together. This is one of those less visible steps, but it’s crucial for making the next iteration smoother and more reliable.

Speaking of broaching nuts: in most cases, on thin sheet metal with small thread diameters that can be accessed from both sides, broaching nuts are superior to rivet nuts. They are much faster to install and sit flush with one surface. Additionally, they do not create a burr that can cut you, unlike those formed by friction drilling. The best thing, the name is misleading. You don't need to broach them in, with a rivet nut tool or an arbor press. Especially on planar pieces with small M3 nuts, you can get away with just hammering them in. You just place them where they need to be and then give them a firm, controlled wack with a hammer - done.

Those are probably the reasons why the original CORE One parts use the same manufacturing technique, and why I’m also using it in this project. This results in a clean and consistent-looking assembly. It’s a small detail, but it makes a big difference in long-term usability and serviceability.

Electronics Fitment

I also worked on integrating key components into the back section. The PSU and xBuddy Box are now test-fitted - all other mounting points and cutouts fit properly. The general placement and accessibility have been verified,and the alignment with mounting points and surrounding parts is spot on.

Since there is significantly less wiggle room than on the CORE One, it’s a tight fit - but it does fit, more or less. As you might notice, there is no space for the WiFi module, which is a problem that still needs a solution. I currently have two different approaches in mind.

The first option would be to move the location of the WiFi module and simply add a 90° adapter for the C13 to C14 connection - this was originally the solution I was aiming for. I also have the hardware on hand to create custom extension cables or 90° adapters for the 2×4-pin Dupont connector used by the WiFi module. However, I’m not fully convinced yet, as I can’t really find a clean placement for the WiFi module.

The second option would be to move the position of the C14 socket. The most logical location would be the PSU cover. Since a regular 90° C13/C14 adapter won’t solve the clearance issue here, I’ve ordered a special low-profile adapter for ATX power supplies used in SFF builds. Unfortunately, it hasn’t arrived yet, so this will have to wait for another day.

Alongside that, I’ve been reviewing the hole and cutout positions for wiring. It works and follows the same design principles as on the CORE One, but I’m not entirely happy with the decision here. The cable paths are acceptable and not overly constrained, but it would have been far easier to route them through the bottom. Enlarging the cutouts wouldn’t significantly reduce stiffness and would have allowed much easier access to the 24 V wiring and the Power Panic cable - but that’s a problem for another day. I won’t redo this part, as my OCD kicks in here.  

Cooling & Custom Parts

Another small addition is a custom fan grill for the back assembly. It’s a minor visual and functional upgrade, but it helps keep the design consistent across the entire system.

The thermistor and fan wiring on the back panel need to sit very close to the rear of the assembly. Since the entire Y-axis has been moved about 18 mm further toward the back, there is very little available space in this area. As a result, the part cooling fan would otherwise collide with the chamber fans and the wiring.

The fan grill is essentially the original design, but with added cable-tie holes arranged in a 90° bend. Due to the lower wiring position, the fans and thermistor also need to be routed slightly differently to neatly tuck everything into place.

The same applies to the two belts, which would otherwise rub against the cables. The plan is to create a dedicated cable guide system, similar in concept to the C1L, to keep everything controlled and safely routed. This is one of the many 3D-printed parts that are still still missing.

Heatbed Assembly Improvements

Outside of the back assembly, I also revisited the heatbed assembly and fixed a few smaller issues that came up during testing.

Most importantly, I spent some time on the laser cutter again and replaced the temporary printed parts with proper metal brackets made from 3 mm steel instead of Prusament PCCF. This significantly improves stability and durability and brings it up to the stiffness level of the CORE One. Besides a proper paint job, the heatbed assembly is now basically done (I hope).

These kinds of upgrades are part of a slow transition from prototype to something much closer to a final design. At this stage, everything may seem done, but the Pareto Principle (the 80/20 rule) starts to become very noticeable - the remaining 20% of the project will likely take up 80% of the total time. 

Since the Prusa USS has now been released, I’m hoping a dimensionally accurate CAD model will become available soon or that I can get my hands on a physical unit. That would allow me to focus next on the right-side sheet metal panel, which is the last part that is not yet fully finalized in CAD due to uncertainty around the correct mounting and hole positions.

---

If you want to learn more about the project, make sure to checkout the main post "CORE One Mini - The Rocky Road to a Prototype." for the complete background and FAQ.

If you intend to buy a Prusa machine, you can use the referral code "@suit" at checkout in the Prusa online shop. This will give you some Prusameter points and 1 kg of free filament and i'll get some points aswell (which I can use for some free filament). Or you can just download, like and make some of my models here on Printables and even consider to become a member in my Printables club.

I've also set up a GoFundMe campaign to tackle the costs. Thanks to everybody who has donated a few euros already - that means a lot and helps me cover the expenses.