Hello everyone! As I mentioned in last week's Reddit update, I finally completed all of the metal parts. The side panels are finished, tested and painted. After completely messing up the first attempt because of incorrect bend radii, the second version turned out exactly as intended.
With the sheet metal work behind me, it was finally time to continue the assembly. A handful of parts still needed some customization, but I was fairly confident that everything would fit without requiring any major changes. Well ... mostly.
Assembly has now progressed all the way up to the toolhead. Along the way I fully tested both the back assembly and the heatbed assembly and the CORE One Mini model page on Printables has been updated with all files required for Step 3 (Back Assembly) and Step 4 (Heatbed Assembly).
I'm happy to say that these sections turned out exactly as hoped. The parts are stable, easy to assemble and didn't require any redesigns after the first physical build.
For the back assembly I ultimately decided to go with the more "expensive" solution. Instead of using an angled C14 to C13 adapter, I designed a custom extension bracket that allows the power connector to remain at the back while keeping the WiFi antenna in its original location. This is entirely optional - if you don't use WiFi, a simple angled adapter is perfectly sufficient. If you do want WiFi, the antenna simply needs to be relocated using a small extension cable.
One of my favorite little details turned out to be the cable management inside the base: The routing came out surprisingly clean and as a bonus, these printed cable guides can also be installed on a completely stock CORE One if you're simply looking for a neater alternative to zip ties.
The redesign also includes a new holder for the stepper splitter. It is slightly more involved to assemble than Prusa's original design with the two plastic retaining pins, but it simplifies things in the long run. Potential future revisions will allow the lower front profile to be identical to the upper rear profile, eliminating yet another "special" part from the BOM.
(And yes ... before anyone points it out: the cables in the photos are actually plugged into the splitter the wrong way around. The picture was only taken to show the routing, i was in a hurry)
Speaking of simplification, one thing still bothers me: Because the rear profiles are modified original parts, they currently aren't symmetrical. If custom-made profiles are used instead, this can easily be fixed, allowing both rear profiles to become identical parts rather than mirrored versions. Another small step towards reducing the number of unique components.
Some of you may also have noticed that something looked odd around the rear fan grid and thermistor: That isn't a mistake.
Besides the custom fan grid - which is necessary so the part cooling fan doesn't collide with it - the cable routing is intentionally flipped upside down. This wedges the fan and thermistor wires neatly between the rear fans while also shortening the cable length. It's one of those tiny details that took far longer to figure out than I'd like to admit, but the final result is much cleaner.
The heatbed assembly is largely straightforward: Mechanically it follows the original CORE One design, although everything is mirrored, scaled down and simplified a bit.
The LED strip requires one small modification: either fold the cable back onto itself or solder the connector on facing the opposite direction. After making a few small cuts with a hobby knife, even the original LED diffuser can be reused.
Looking ahead, I already have an idea for a future revision: Instead of reusing the original diffuser and sheet metal parts, the diffuser, spacers and LED cover could all become a single 3D printed assembly - either as two single-color parts or one dual-color print using black and transparent polycarbonate. Replacing two sheet metal parts, one laser-cut spacer and one custom component with a single printed part sounds like a pretty good trade-off.
The bed cable covers are similar to the originals as well. The only real change (besides downsizing) is switching from hex nuts to square nuts. They're much easier to insert during assembly and simply feel like a more modern solution.
Otherwise ... business as usual. Just with a much smaller print bed.
With the mechanical assemblies largely completed, I could finally test mound the side panels to see if they would fit. Getting to this point wasn't quite as straightforward as I had hoped.
The first set of panels ended up as "expensive practice pieces" after I underestimated the required bend radii that need to be bent in an odd location with non professional equipement. There is nothing wrong in CAD, but once the parts came off the brake it became obvious that the bends weren't going to work as intended.
For the second iteration, relief cuts were added to the bends to not distort the holes. If you don't have access to a professional brake, scoring the parts before bending would also be advised. The bends in the final part are perfect: they panels fit exactly as intended, the gaps are nice and even and after sandplasting and painting, they again look like factory made parts
With all the clickbait in the title: well, the problems are waiting further up the assembly.
The CoreXY section still needs a few CAD updates to properly accommodate the filament sensor and the current cable routing for the rear fans and thermistor is ... let's call it functional. It works, but while assembling everything I found a significantly cleaner solution that I'd like to implement before releasing those files.
Those are relatively small issues. Then I manually raised the heatbed all the way to the top: a very loud "F-word" echoed through the workshop ... and probably half the neighborhood. For a moment I genuinely thought I had made a catastrophic design mistake.
Luckily, it isn't nearly as bad as it first appeared, fhe first collision is relatively simple:
The screws holding the trapezoidal nuts collide with the rear guide because I completely forgot to include those screws in the CAD model while designing the holder. On top of that, the bed carriage itself touches the angled section of the guide.
Ironically, I had already changed that sheet metal part later in CAD specifically to avoid this issue - but never updated the actual manufactured part before painting it.
So I'm currently a few millimeters short in Z and roughly one millimeter too long in Y.
Annoying? Absolutely. Difficult to fix? Not really. The guide simply needs another redesign to create some additional clearance.
The second issue is more interesting: the toolhead currently collides with both the bed alignment pins at the rear and the bed cable cover.
Why? The alignment pins are slightly longer than on the original CORE One and, because of the reduced space available, they're also positioned slightly further forward.
To make things even more interesting, I'm currently using the CORE One L fan duct, which has a flatter lower profile because that machine doesn't have a bed cable cover in the same location.
That leaves me with two choices at the moment:
Use one version that collides with the rear fan assembly and the alignment pins.
Use the other version that collides with the print bed in two different places.
Neither option is particularly attractive. Fortunately, this is exactly the sort of problem CAD is meant to solve.
I've already printed a crude scaled-down mock-up of a redesigned fan duct. Functionally it works, but ... let's just say it won't win any beauty contests. So that's the next task on the list, make a Fan duct that does not collide and look good at the same time.
The good news is that every issue I've found so far is completely solvable. This is exactly why I build and test every single assembly before releasing the files - it's much better that I discover these little surprises now than probably thousands (yeah, I'm dreaming) of builders finding them later.
So ... back to CAD we go.
Before wrapping up, again I'd like to say a sincere thank you to everyone following this project.
The comments, ideas and words of encouragement genuinely keep this project moving forward. That's exactly what I was hoping for when I started this project.
It's also been amazing to see people already preparing parts and gathering materials while the design is still evolving. Knowing that there are builders waiting to assemble a CORE One Mini makes every redesign and every "back to CAD" moment feel worthwhile - and hopefully enough of a push to make Prusa help us out with an affordable hardware package for the DIY community.
As always, if you spot something odd, have suggestions for improvements or simply want to share your thoughts, don't hesitate to leave a comment.
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.
Comments
█
█
█
█
█
█
█
█
█
█