11/1/2025 - Intial Design Ideas, Frame, Gantry Components

So this is a full redesign of my highway printer:
https://github.com/Renovic/Cuprite-OLD-

There are a couple of things that happened that made me want to redesign:

• Sponsors: I reached out to a couple of companies and I got some sponsors which means I needed to change a couple of parts. The companies are: LDO, Sunlu, JLCPCB(OSHWLABS), BST AUTOMATION, Filastruder
• Makerspace access: Now I have access to a makerspace and a source for sheetmetal so I have access to more manufacturing capabilities
• Better Design Ideas: My original design was rushed and wasn't ideal. For example, the toolhead COM wasn't next to the carriage, so it would lead to worse results.
• Better Aesthetics: I've gotten a lot of new ideas for improved aesthetics

Day 1- Full Redesign Concept (Oct 1st - 3rd)

Over these few days, I consolidated my ideas on what I wanted to redesign. I also started sourcing parts for the build.

So here are all of the changes from the old version I plan:

Frame:

2020 + 2040 - > 4040 verticals (JLCMC)

• This will imporve rigidity greatly and allow for better mounting

Insulation/Panels:

No Insulation -> 3/4" PIR Insulation

• Fits inside of the frame, which is why 4040 verticals are important

Gantry:

9mm belts -> 12mm EPDM (partialy sponsored by Filastruder)
2WD -> 4WD (Sponserd by LDO)
1/8" Alu sheetmetal -> CNC aluminum (JLCCNC)
2020 extrusion x-gantry -> 2030 aluminum tube

• Should allow for more rigidity
• ~30% Input Shaper gain from 9mm -> 12mm belts
• ~100% Input Shaper gain from 2WD -> AWD
• More Input Shaper gain from extra frame, and gantry rigidty

Toolhead:

Peopoly Lancer -> Custom aluminum hotend (JLC3DP)
Fully SLM/CNC toolhead (JLC3DP & JLCNC)
G2 Extruder (LDO)

• Better extrusion from G2
• More rigidity from bottom supported hotend
• Centered COM from redesign
• higher flow rate

Cooling:

WS7040 -> 130k RPM Ducted Fan
No flap -> Stepper motor flap

• Cheaper
• Higher max cfm

Z Axis

In chamber motors -> bottom-mounted motors
BMG Gear Reduction -> G2Z
3dp joints -> slm joints (JLC3DP)
No spring tension -> spring holding down bed frame to joints
Extrusion bed frame -> Laser cut aluminum frame

• More rigidity
• More travel from bottom-mounted z motors
• Better z motion consistency, and as a result better layer stacking

Electronics

Fystec Spider -> Leviathan + Expansion board (sponsored by LDO)
tmc2209 XY motor drivers -> tmc5160 XY motor drivers (Built in to Leviathan)
24V XY -> 48V XY (Partially sponsored by Filastruder)
0 mainboard cooling -> 4x 6020 fans (JLCMC)

• Higher stepper motor speeds
• Better step control
• Can cool stepper motors

Chamber Heater

None -> 100W Chamber Heater

• Cheap
• Allows for good chamber temps ### Aesthetic Improvements Rear backpack -> top and bottom electronics enclosures Exposed panels -> printed covers Door hinges integrated into panel covers
• Displays Leviathan and good looking electronics on top
• Hides PSUs underneath
• Nicer visuals

Even with all of these improvments, the cost of new components should be cheaper than the original.

Hours Spent: 12
Total: 12

Day 2 - SLM Toolhead Redesign Concept (Oct 13th)

I started by designing a hotend similar to the hotend in the monolith toolhead, or DK's slm bmg toolhead. This should get high flow rates and have mounting for a bottom hotend brace.
I spent the rest of the time messing around with different extruder options, and hotend locations to find anything that could get me a good com and nice design.
The main issue was with bmg gears the gear covers the bolts mounting the main toolhead body into the mgn12h carriage.
I also wanted to try using a g2e kit because it should give me very precise extrusion without any wood grain affect thats normally caused by double gear extruders. Although I didn't get as much as I wanted done, I was able to figure out the location of the hotend and extruder that could give me a decent center of mass, and I got some ideas of the rest of the design.

Hours Spent: 6
Total: 18

Day 3 - CNC Front Motor Mounts (Oct 17th)

Main Goal for this was component was to incorperate the front extrusion brace and rigid part.
This is using the monolith belt path like before, but now for 12mm belts.
Also I made sure to insert every fastener required so I can make sure I source all of the components and don't have any intersections. But this made it take a bit longer than how I was designed previously.
During this day, I also found sources for the hardware required, most of the hardware like bearings, standoffs, and bolts were from JLCMC.

Hours Spent: 4
Total: 22

Day 4 - New Frame & Panels (Oct 30th)

I redesigned the Frame to use 4040 verticals.
I also was able to find 3/8" aluminum panels, so I designed the aluminum panels to be mounted onto the frame.
Another thing I realized I missed on the first design is the rubber strips to reduce heat and vibration transfers from the frame to panels, so I designed those in.
Finally I added insulation to the internal faces of the panels.
The overall width of the printer is 50mm wider than the orignal printer, but it includes insulation.

Hours Spent: 2.5
*Total: 24.5

Day 5 - X Axis (Oct 31st)

Today I redesigned the X axis and XY joints to use 12mm belts and 2030 Aluminum Tube.
The XY joints are 2 part that clamp onto the mgn9h carriage and bolt into the carriage.
Then I made the 2030 aluminum tube and mounting hardware. I also needed to add 3d printed crush blocks into the tube so I can mount the rail to the tube without worrying about bending the tube inwards.
I also found the sources for aluminum tube, hardware and compression springs for correct spacing on the live idler joint.

Then I wanted to see how the gantry looked, so I made a gantry assembly and put it into the frame.

Hours Spent: 5.5
Total: 31

11/2/2025 - Designing Rear Motor Mounts

Day 6 - Rear Motor Mounts (Nov 1st)

I was really confused about how to design the CNC belt tensioner for a while. But I finally had an idea to use slots to guide the tensioner, have a countersunk M3 screw to mount the tensioner bearing stack pin, and M5 bolts to clamp the tensioner in place.

These took way longer than I expected because of the amount of hardware required for this part.

Hours Spent: 6
Total: 37

11/3/2025 - Bed Assembly, Front Z drives

Day 7 - Bed Assembly, Front Z drives (Nov 2nd)

I started by designing the bed assembly. It uses a sheet metal carriage and some CNC components to mount the bed and 8mm rods that will be used for kinematic coupling.

Then I designed the front Z drives that use G2Z planetary gearboxes for precise Z motion. Part of the design was making sure the part is entirely sealed so air from the chamber can't leak into the bottom electronics area.

Hours Spent: 6.5
Total: 43.5

11/5/2025 - Day 8 - Back Z drives, Tweaks to Z axis, More research

Day 8 - (Nov 4th)

I started with designing the rear Z drive, and this was pretty similar to the front Z drives, but mounted differently.

Then I started thinking about how I was going to do the Z belt tensioners. Although I originally just planned on designing a belt tensioner similar to the Annex K3 z-axis tensioners, I wanted to see if there was something more compact. So I did some research and found Hex 0 had some very compact belt tensioners.

These tensioners pressed the belt against the linear rail carriage and had one part slide to adjust tension. The Hex 0 z drives used 6mm belts and mgn7 rails, so I was a bit concerned about whether this would work. But thankfully, I found that the mgn12 rails had enough width for 9mm belts. I had to adjust the front z drive pulley location to make the belt line up with the edge of the carriage. I also added some channels for wires to come down the vertical 4040 extrusions, and go to the stepper.

Then I went to make some tweaks to the bed carriage. I was a little concerened about the weight because I was using 3/8" thick aluminum and my original design was very beefy, so I checked and it was 400 grams. I decided to redesign it to be a little simpler and much lighter, and it ended up being 250 grams.

Finally, I started thinking about how I would get the most travel possible. So I measured out the toolhead length, z joint length and bed height and found that I only had 104mm of travel. The 200mm rail I have has a full travel of 150mm, so Im losing out on a lot. My two ideas so far are to raise the rod mounts on the bed assembly, to effectively drop the bed down, or redesign the z joints with the bearing further down.

For reference: my original z joints (I have to redesign these for slm printing, different tensioner, and 9mm belts)

** Hours Spent: 5.5 **
** Total Hours: 49 **

11/6/2025 - Day 9 - Front Z Joint

Day 9 - (Nov 5th)

My main goal was to finish the z joints, but this one z joint took a lot longer than I was expecting.
Yesterday, I found that the bed travel would be significantly limited if I went with my previous idea of the 8mm rods resting on top of the z joint rails, so I was trying to come up with a way to move the 8mm rod mount further down to increase Z travel. Eventually, I came up with a decent idea and designed it.

With this design I should be able to get 135mm of travel, which I think is acceptable. The only way I can increase z travel after this is by decreasing the bed assembly thickness (currently 30mm), or increasing the height of the overall printer, which I don't think is worth it for an extra 15mm of travel.

I also had to adjust the bed carriage again slightly to match the new position of the 8mm rod mounts.

The problem with this design is that I don't have a spring keeping the 8mm rod pressed against the bearings. The bed's weight will hold it down, but it is only a 160mm bed so it isn't very heavy, so I need some springs to hold it down. I wasn't sure how to implement it while designing the main part of the Z joint, so I'll try to add it next time.

Hours Spent: 3
Total Hours: 52

11/12/2025 - Day 11 - Front Z Idlers

Day 11 - (Nov 12th)

For some reason, designing the front z idlers was a lot more challenging than I expected.
When I first started designing them, I thought they would be really easy, just a pulley and some bearings, but I didn't realize how tight a packaging I needed them in, and how they would mount next to the front motor mounts in the gantry.

Originally, I was using a 5mm shaft with bearings on either end of a pulley, which required at least two components to mount. I was thinking about tapping holes into the front motor mounts to mount them, but it was interfering with other components.

After a couple of attempts at designing this part, I decided to switch to using an 8mm shaft and having the pulley cantilevered. Because it's an 8mm shaft, there weren't any worries of the shaft not being able to support the tension of the belt and the load of the bed. And because of the cantilever, I was able to keep both bearings out of the way of the front motor mounts and give the most Z travel possible.

Hours Spent: 3
Total Hours: 55

11/15/2025 - Day 12 - Kumiko Panels Start

Day 12 - Kumiko Panels Start - (Nov 14)
I really wanted to work on something other than the motion components of the printer, so I started working on the panels. I've had this idea for a while, since Paper View made a YouTube video about Kumiko for wall art. I thought it would look great if I made these panels to cover up the slightly ugly structural panels. Basically, it would consist of 3d main components, the corner covers, the frame, and the inserts. The corner covers would be rounded and bolted into the frame, and cover up the M5 bolts that the structural panels use to mount. Then the frame is the triangle grid, which will connect to the corners. And finally, the inserts will be able to be swapped around to create unique designs.
When I first had the idea when designing the old version of Cuprite I made some mock designs using a online kumiko designer.

(These are larger than Cuprite currently, because this was before I properly calculated everything)

I also wanted to make similar panels for my other 3d printer, an Ender 5+ that's been modified significantly. So I needed to make it parametric to scale up to cover the much larger Ender 5 frame. This made it significantly harder to design as I needed to make everything parametric and ensure it scales up and back down reliably. When I first started planning this out I also thought about releasing as a standalone design for other people to use, as there are people trying to figure out ways to cover up structural panels on their own printer.

I started by figuring out what size pitch (triangle side length) would work between cuprite and the e5, so I made a Desmos to try out different sizes to find one that was reasonable. I ended up landing on 55mm as it looked good at both sizes (6:9 for Cuprite, 9:14 for Ender 5). When I figured out what dimensions work, I made all of the parameters in Fusion.

I only got to making the frame and corners parameteric, but didn't end up making the mounting or dovetail connections.

Also, while I was doing this, I took the time to learn about fusion configurations, which is useful to modify many parameters for scaling between sizes quickly.

Hours Spent: 7
Total Hours: 62

12/3/2025 - Day 13 - Kumiko Panels Partially Done

Day 13 - (Nov 15 - Dec 2nd)
So this took a lot more time than I was expecting (a couple of days). The main problem was the number of things that needed to change between configurations, and how difficult the fusion internal body identification system is. So a major aspect of how the panels are designed is having dovetails to lock the panel sections together and lock the panels into the corner and top pieces, so there isn't any glue or tons of hardware required to assemble them.
This is how I designed the panels' dovetails:


The main consideration was how I would print them so as not to require any supports, and make sure the dovetails weren't visible. This part wasn't that difficult as long as I added them before splitting the panels into multiple pieces.

The next main component was splitting the panels into 4 parts to fit on the build plates. This portion was inspired by Paper View's printed panel, where he split the panel along the center between triangles. However, one thing I didn't like much about his design was using a seperate insert to cover up the gaps between panels.

After thinking about some ideas and sketching stuff out, I found a solution I was happy with: adding a chamfer to one piece and a lip to the other that covers the lower piece.

As you can see in the previous images, I also needed to implement the dovetails for locking the panel sections together. The primary issue with this was that fusion would change the internal body IDs between configurations if there were any differences in the features between configurations, which would break links to bodies. I had to ensure there weren't any features that affected a body in one configuration and not another. This took by far the longest out of part of designing this printer, as whenever I added something, it would break in a new and unexpected way. But eventually, I found workarounds for all the problems, and I was able to get the configuration switching working with all of the dovetails.

Here are the panels sized to Cuprite:

And here are the panels sized to my E5+ corexy conversion:

The only other thing I didn't mention so far is how the panels are attached. I couldn't figure out a way to do concealed screws that weren't a pain to remove, so I decided on using countersunk M3 screws. If I use black oxide screws, it will blend in with the black ABS I plan on printing this out of and won't be too distracting.

And here are the panels on the printer so far:

Now that the panels are done, I still need to make all the colorful inserts and skirts to go with them.

For the inserts, I planned ahead a little and added these lips to contain them inside the panels. I want to make sure no part of the panels disassembles itself while the printer is shaking from printing quickly.

I doubt the inserts will take long, as they don't require any configurations, just a couple of parameters.

The main hurdle is the skirts, which will have to be configurable just like the panels, and need to support fans, power inlets, and a screen mount.

Also, here's how many parameters the panels took so far:

I'm hoping to release this as a standalone mod for people to add onto their printers. Many people with structural panels don't currently have anything to cover them up, and they sometimes take away from the look of a printer. That's why I have all of the parts of these kumiko panels parametric to fit most people's printers.

** Hours Spent: 14 **
** Total Hours: 76 **

12/10/2025 - Day 14 - Kumiko Panels - Door & Some Inserts

Day 14 - (Dec 11)

I wanted to finish up the panels before moving on to the skirts, so I planned out how the door would fit into the front. The hard part was making a door that wasn't too visually distracting from the smooth corners and panels.

My main constraints for the door were:

• sit flush with the rest of the panels
• Support enough weight in order to scale well
• Removable without a lot of disassembly
• Fit in aesthetically

So I came up with this hinge idea where there is a shaft within bushings to rotate freely and support weight, and be in two parts to be removable.


The two bolts clamping the shaft are removable with the door open, but are hidden otherwise, and the bolts holding the hinge to the door are always hidden.

Within the corner pieces of the panels, I added bushings for smooth rotation

Heres the door:

And the door on the printer:

I also started on making the inserts. They were just recreating common kumiko patters with my parametric values, and adding a chamfer so they sit inside the panels. I also needed to do the half variants due to the edges and corners.

I only got a couple of inserts done, so I'll keep adding more as I go on. There are over 60 inserts listed on kumikogenerator.com, so i have a lot more to add. I likely won't add all of them, but probably 25-30.

Hours Spent: 5
Total Hours: 81

12/14/2025 - Day 15 - Handle & Starting skirts

Day 15 - (Dec 14)

Heres a render of the full printer design so far:

I made the handle latch into the extrusion and rotates on bushings. It also has some magnets to automatically latch.

I then started thinking about ideas for skirts, a lot of designs were pretty ugly, but I knew at the very least I needed some sort of border to separate the panels from the skirts, which would also provide the mounting holes for the skirts.

This design also gives +40mm of electronics space because the skirts are outside of the frame, rather than directly on the frame. I'm not sure about the rigidity differences, but I doubt it's that much of a problem.

Finally, I made more inserts and colored them to demonstrate how the panels will look with inserts.

Hours spent: 6
Total Hours: 87

1/20/2026 - Day 16 - Skirts

Day 16 - Jan 19

So, Its been a month since my last update. For most of that time, I wasn't working on it; I was working on my other 3D printer, but for the last few days, I started working on Cuprite again.
I was looking to get a lot of progress done, but I got stuck on the panel configurations. Previously, It was working entirely fine, I could switch between configurations and everything would update well. However, at version 19, it stopped working. I somehow didn't notice as I designed the door and accent pieces. So I worked on finding the issue that's causing fusion to lose references to bodies. I looked through around 80 features comparing the two versions to ensure they were identical. I spent many hours trying to find the solution, but I just wasn't able to. I ended up giving up on fixing this for now, and just focusing on getting the design complete for Cuprite.

Here is the skirt design:

It has 3 pieces per side, and a corner piece, so that it's printable on small build plates. There's a top acrylic panel bolted into all of the pieces, and fan mounts on the sides.

I think the most interesting part of designing the skirts was figuring out the angles of the top and bottom rows of triangles. I wanted them to be angled in a way that the intersection at the corner lines up with the triangle sides.

Funny thing about the size is that it's currently 416 x 416 x 625mm, while a P1S is 389 x 389 x 458mm, even though my printer's build volume is 100mm smaller in all dimensions.

The skirts are almost complete, only the foot mounts and the power inlet are left to design. This has turned out to be a much longer side quest than I expected, but it should make the printer much more unique and nice to look at.

Hours Spent: 15
Total Hours: 102

1/21/2026 - Day 17 - Skirts Complete

Day 17 - Jan 20

First, I worked on the feet mounts.

They are integrated into the corner pieces, but also somewhat hidden to make it look nicer.

Then I worked on the power inlet, which I think turned out a lot better than the old skirts. It nicely matches the corner pieces.

I then worked on the bottom electronics bay, which has two PSus a 350W-24V and a 200W-48V. I also have two SSRs, one for the bed, which I am going to use JLCPCB's flex heater service for, and another for a 500W chamber heater. Originally, I was planning on using my 200W 24V bed heater, but that would be slow, and take a ton of power supply wattage. I only have 350W total, and I am using a 130W fan and 160W of hotend heating, so I don't have the wattage to fit a 200W bed.

I used the 2010 extrusions here to make it configurable in the future, in case I need to add more stuff. I plan on keeping the top electronics fairly clean and would rather have the mess happen below.

Hours Spent: 7
Total Hours: 109

1/22/2026 - Day 18 - Bottom electronics + Z almost done

Day 18 - Jan 21

First, I added extension springs to my z joints, so the rods are always pressed onto the bearings. I had similar Z drives for my e5+ conversion, which were just the original version of Cuprite Z joints but with leadscrews instead of belted Z. I had to remove the bed anytime I wanted to tilt the printer, which I really don't want to do with Cuprite.

I finally finished up the rear Z joints next, which I put off doing for a while, since the belt orientation was different than the front Z drives. I ended up redoing the rear Z drive so the belt orientation and spacing matched the front ones relative to the carriage. This allowed me to reuse the front Z joints, just by changing the angle of the bearing mounts, and adding the extension spring mount.

Next up is the rear Z idler. I hope to design it as dual purpose, both the idler and a connecting bracket between the rear vertical extrusion and the gantry extrusion. After this, I'll be done with the full Z assembly.

Hours Spent: 4
Total Hours: 113

1/23/2026 - Day 19 - Z axis done

Day 19 - 1/22

I first designed the rear Z joint, which ended up being a lot simpler than I was expecting.

Then I modified the Rear Z joint slightly to strengthen the spring mount.

And with that, the whole Z axis is done.

I then started working on making a new wiring diagram.
I started by listing out all of the components and connections I needed, and I realized I was going to use almost all of the ports on the Leviathan + Extension. The only ones I didn't plan on using was z endstop, probe, filament sensor, and neopixel port. I'm considering whether it's feasible to add specific wire channels/ guides for the wires, since there is a fixed number of wires, and I know exactly where all of the wires go to from the top ebay. I'll probably include the rest of the ports in the wiring guides, so its a easy upgrade if I need to attach anything. I know it's a bit risky to lock down the wiring by cadding it out, but I think there will be no changes in the future other than replacing components.

Another thing I worked on today was the initial idea for aux fans. This is a quick 2d sketch of what the duct could look like.

The fans are 8038s which have 124 CFM, and pull 42 W. My bed size is 160mm, so I'll be able to fit two on each side for a total of 4 fans, 496CFM, and 168 W. The main issue is tapering the 80mm fan size down to a 5-6mm outlet without losing a ton of airflow. If I get the duct design right, I shouldn't have to worry about min layer times ever, and I'll be able to heatsoak the chamber a lot faster with this extra 500 CFM of airflow. I'm not entirely sure whether this idea will pan out, because the ducts are in a pretty confined space, 88x160mmx160mm. The main reason I am choosing these over more conventional blower fans, is that they are 3 dollars each from filastruder.

Hours Spent: 6
Total Hours: 119

1/24/2026 - Day 20 - Aux Fan Ducts

Day 20 - 1/23

Today I tried designing aux fan ducts based on the sketches I made yesterday. Here is my first attempt:

Because I barely know how to design fan ducts, I decided to simulate it, in order to test how flat the outgoing airflow is, and if the bends are too tight.

And there is definetly a lot of issues. There is a lot of airflow from the left side, while the right is quite slow, indicating that there is too much restriction in the fan duct. Also, the output shows the air blowing in all different directions, and definitely not flat, which likely indicates that my output area is way too small. So I will need to adjust the fan design significantly. I likely will need to remove the insulation next to the fan ducts, so there is an extra 20mm of space. Hopefully, I'll be able to fit it, and won't have to switch to smaller 6025 fans, since they have only 33 cfm, a quarter of the 8038 fans. I'm also considering just using 1 8038 fan per side, so the output area is doubled.

It took a while to figure out how to use Simscale, but it was definitely useful, as I'll be able to use it for my toolhead cpap ducts, instead of just guessing how good it will be.

Hours Spent: 6
Total Hours: 125

1/27/2026 2 PM - Day 21 - Aux Fan Ducts pt.2

Day 21 - 1/25

So I kept working on the fan ducts. I really wanted to get them functioning well as they would be the most powerful sheet cooling system I've seen.

The first change was updating the output area to 160mm width instead of 80mm width. This cuts down the number of fans to just 1 per side, but I think it's more than enough cooling. The main reason I did this was that the input area of the fan size was significantly larger than the output area, resulting in too much restriction. I also changed the geometry a little to make it more favorable to this 160mm width.

I then did the sim on this, and I realized I missed inputting a setting properly last time. For setting up the input vs output parameters, I forgot to set the pressure for the input area, which may have caused some of the issues with the simulation.

This run looks a lot better than the last one; the output is mostly a flat plane. However, there still seems to be the issue of too much restriction near the output. Also, there's this weird slight waviness in the sheet cooling, which I don't understand.

I then asked for input on the duct design and got a lot of valuable feedback:

• Don't reduce duct cross section while turning
• Add a bit of a straight before the final exit since air has momentum and won't fly out exactly straight
• Increase the taper length significantly

The issue is that to implement any of the suggestions, I need a lot of space. However, because this is such a small printer, I didn't really have that much space. I realized I measured some things wrong last time, which gave me extra vertical height, but less horizontal space. So, I decided to remove some of the insulation right next to the fan, which gives me an extra 25.4 mm of space to work with.

Here's the new sketch/layout of the new duct:

The taper section will be a loft between the circle and the rectangular output + a few mm in width. The curved section will be almost purely for turning the air, but there is a slight reduction since I didn't want to make the taper section too sharp. The curved section also features a longer straight part at the end to ensure the output stream of air is flat.

A large benefit of having a separate taper and curve section is that it makes it easier to split. I didn't really consider printability much when designing the original ducts, and any duct requires some split in order not use supports. Now I have a clean area where I can cut and add the features for bolting the two pieces together.

Once I get this new duct design done, I'll test 80mm width vs 160mm width in sim to see how much worse it is. I doubt it will be that useful anyway to have 2 fans per side, as most cooling needs to happen towards the center of the build plate.

Hours spent: 4
Total Hours: 129

1/27/2026 8 PM - Day 22 - Aux fan ducts pt.3

Day 22 - 1/26

Based on the sketches, I made the new fan ducts. This time, it seems like the output flow is more consistent, and there isn't as large a jump in fan velocity in the curve.

I think it can still be optimized more, but I'm pretty happy with the current design. It is flat enough to reach across the bed without much deviation, and it is faster and has more air volume than any aux fans that I've seen. I will try getting more feedback on this design to improve it to maximize the potential of the 8038 fans, before finalizing the fan duct with the mounting.

Hours spent: 2
Total Hours: 131

1/28/2026 - Day 23 - Toolhead pt.1

Day 23 - 1/27

I had an interesting idea for toolhead design, where I could place the motor behind the mgn12 carriage to bring it down a few mm and back by 10mm. I did a quick mockup to test the com using DK's toolhead ducts, an old version of a slm heatblock, and a ldo extruder motor. Also, I cut the toolhead ducts because this design will be around 10-15mm shorter than DK's toolhead.

It looks pretty promising, but it might get shifted upwards with more extruder hardware.

I then worked out the location of all of the gears and the hotend location. I looked at a lot of different gear sets and wanted something with a larger filament grip wheel than bmg gears, but without sacrificing the effective gear ratio. I originally wanted to use LGX or HGX gears, but I ended up finding the K1 gear set, which seems to have larger bearings. The only issue is that the stock there is only one bearing in the filament drive gear, but with an extra bearing inserted, it will be able to hold up better.

As for hotend design, I looked at the hotend in DK's toolhead design and the Tricorn hotend. Based on the monolith toolhead 80W is sometimes too little for an uninsulated hotend with a nozzle brace. However, higher wattage heaters are around $20-30 each. So I decided to make a dual heater hotend, since triangle lab 80W supervolcano heaters can be $3. I also found a small heatbrake which lets me push the hotend closer to the extruder for an overall shorter toolhead design.

My primary goal for this toolhead is to make it as compact as possible without sacrificing hotend length or toolhead com, which would let me get a more rigid toolhead, less weight, and more z-axis travel.

This seems doable to design, but it will definitely be the most complicated part of the project.

Hours Spent: 5
Total Hours: 136

1/31/2026 - Day 24 - Toolhead pt.2

Day 24 - 1/30

As a part of the compact toolhead design, I had to design a custom CNC'd hotend. My primary inspirations for this are Tricorn hotend, DK's BMG toolhead, and the Monolith toolhead. I looked a lot at tricorn as it's the only one out of the 3 that's CNC'd instead of SLM printed. CNCing is more limited design-wise, but it lets me use traditional heatblock materials like copper.

The top part needs to be small enough to fit as close as possible to the carriage, while the bottom isn't as restricted.

The extra tapped holes at the bottom will be used for the slm titanium hotend support/heatshield.

Next up is designing the part that's used for belt clamping, hotend mounting, heatsinking, and extruder. It's probably going to be the most complicated single part, and its cnc'd.

Hours spent: 4
Total hours: 140

2/1/2026 - Day 25 - Toolhead pt.3

Day 25 - 1/31

I started working on the main body of the toolhead, which incorporates belt clamps, a hotend mount, and an extruder. I also added fins for more surface area to reduce clogging in the heatbreak.

I was originally planning on using a belt clamp similar to the monolith toolhead, but due to the hotend being so close to the carriage, I didn't have space for it. I'm now going to use the ducts to clamp the belts against the main body, which has teeth in it.

I added the hotend and extruder motor for a general idea of size.

And here's a comparison with a common toolhead, Stealtburner.

This isn't a direct comparison as I haven't added the ducts yet, but the general size won't increase that significantly.

Next up is designing the rest of the extruder; this will also be CNC'd in order to transfer heat away from the stepper and provide more surface area for conduction.

Hours Spent: 6
Total Hours: 146

2/7/2026 - Day 26 - Toolhead pt.4

Day 26 - 2/6

A few days ago, I saw Carve-off, a CNC adaptation of the takeoff-toolhead. What caught my eye was the cnc belt clamps which were simple to machine and held well.

So I started by redesigning the belt clamps.

Now it should be much easier to machine and hold just as well as my previous design.

I then worked on the rear extruder housing, which was somewhat complicated due to the sliding idler side.

Finally, I started working on the ducts. The main issue is the length below the belt clamps not allowing for significant ducting.

I'm really happy by how compact it is in y.

Here's the comparison again:

But the center of mass is really bad right now. It will improve a little with the hotend brace, duct bottoms, and probe, but im not entirely confident it will center it completely.

Hours Spent: 8
Total Hours: 154

2/8/2026 - Day 27 - Toolhead pt.5

Day 27 - 2/7

I first started working on the heatshield/brace for the hotend.

The goal is to shield the hotend block from the cpap airflow, and provide 3 mounting points for increased hotend rigidity.

I then redid some of the ducting.
At the inlet, I switched from 19mm to 15mm tubing because I found a source for somewhat high-temp 15mm cpap tubes. At the bottom, I found a decent angle and positioning of the ducts. The heat shield makes it much harder to getthe ducts to cool below the nozzle. I utilised cuts to bring the ducts closer without protruding out of the bottom or intersecting with the heatshield.

The top ducts are nice because it provides for an entirely straight filament path.

My previous toolhead designs required some bending of the filament path and ptfe, but with a straight filament path, it will be easier to insert Bowden tubes and filament.

As for com, it improved slightly due to the duct changes and heatshield.

With a placeholder probe and extra mass for the hotend brace mounting, it seems possible to bring the com down enough to work.

Hours Spent: 5
Total Hours: 159

2/13/2026 - Day 28 - Toolhead pt.6

Day 28 - 2/13

I designed to redo the hotend shield/bracing. The previous shield was so large that it didn't leave enough room for the ducts to get close enough to the nozzle.

So I made some minimal bracing supports to brace the hotend while limiting heat transfer.

I then made the ducts incorporate the hotend shielding. It should block airflow from hitting the nozzle or heatblock which are the primary heat losses.

}

It took a couple of iterations to get the duct and hotend bracing to not intersect while also directing the air enough. However, I think the opening at the bottom is still a little too large and doesn't direct the air as close to under the nozzle as I was hoping.

Next steps are cfd simulating the ducts and adding a probe mount.

Hours Spent: 5
Total Hours: 164

2/18/2026 - Day 29 - Toolhead pt.7

Day 29 - 2/17

I did a couple of tweaks to duct design, and cleaned up some other features of the toolhead.

I then ran cfd simulation to see if there were any issues with the ducts or if the air was directed too high or low relative to the nozzle.

I think these are pretty good, because it directs the airflow pretty close to the nozzle without blowing directly on it. There isn't much area cooling but that will be handeled by the 2 aux fans.

Hopefully, I don't need to do more duct design. They are the most complicated parts to design and take a while to run cfd tests and analyze the results.

Next step is adding probe mount and figuring out if I need to move thing to balance com. I'm a little worried that I'll need to drop the hotend down a few mm to move com down.

Hours Spent: 4
Total Hours: 168

2/20/2026 - Day 30 - X axis redesign

Day 30 - 2/19

First, I got a ton of useful feedback on my toolhead design. Here's the main things I need to change:

• Hotend braces are too short to be enough of a thermal barrier
• Hotend braces aren't rigid enough in this configuration (switch to flat ones)
• Ducts should be pointed front or back to allow the air to escape in a certain direction
• Top hotend brace
• Add more space for supervolcano heater wires (they could heat up the extruder area)
• Switch the braces to tap into aluminum instead of titanium
• Add gap for the extruder motor because not all mgn12 carriages are the same dimensions
• Possibly smooth out duct transition more

Some of these are pretty easy to implement, but the hotend braces might take a bit more effort to implement well.

I'm a bit tired of working on the toolhead (literally the past 7 journals have been about the toolhead design), so I reworked the x axis.

Because of the toolhead design I needed to make it compatible with a 15mm tall x tube, which required redesigning the xy joints partially. But I took it as an opportunity to completely redesign and save weight.

I took heavy inspiration from dderg's CNC XY joints, but had a lot of changes because of the 15mm tube and com balancing.

It's a lot lighter and looks nicer, so I think it's a pretty good improvement. An additional benefit is using 2 screws through the tube, which is more rigid than the 2 on top and bottom of the previous design.

I then modified the tube and inserts for the 15mm tube height.

And here's the completed x-axis:

Com is centered on the carriage. It doesn't matter nearly as much as the toolhead, but it doesn't take long to adjust, so I got it centered.

Hours Spent: 7
Total Hours: 175

2/21/2026 - Day 31 - Aux fan ducts pt.3 & Hotend Redesign

Day 31 - 2/20

After thinking about the toolhead design more, I decided to redesign the hotend completely. The main reason is that dual cartridge heaters take a lot more space than I expected, and there is no clean way of adding bottom-mounted bracing to the hotend.

An easy solution would be to switch to a single cartridge heater in the front, which would be a lot more compact, but would only be 80W unless I spend $25-30 on a single heater.

But a better solution, that was suggested to me, is to use mch heaters. These look the same as ptc heaters but have less wattage fall off at increased temps and are generally more powerful. And they are much smaller and lighter than catridge heaters.

I found a bundle of mch heater, clips and thermistor for $17 which is about the price I would have payed for just the two catridge heaters. The only downside is 120W vs 160W, but I don't think that's much of an issue if I have good nozzle shielding.

So heres the new design, its half as wide as the old one, so I have the ability to make the toolhead more compact, and it's half the weight.

But this comes with an unintended consequence, the com shifts up because of the reduced mass.

(without probe)

So I have two options: add more weight to the bottom or move the hotend down. Adding weight just reduces accelerations and is not great if it doesn't contribute to rigidity. However, I am already limited in Z travel and would like to not reduce it too much. I also want to shift the extruder motor up to support 20x20 tubes, as locking the toolhead to 15mm tall tubes with turn away anyone who was potentially interested in using it.

Anyway, I got a bit tired of working on the toolhead, so I switched to working on the aux fan ducts.

I first split the ducts to make them printable, then added bolting locations to assemble them.

I then added the mounts to the fan so I could bolt them to the bottom panel.

They are a bit oddly designed due to how close the fan is to the panel, and not wanting to restrict airflow.

I then mounted them in the printer.

I also realized that I needed to move the bed because of my toolhead, so I adjusted it to be within the travel of the toolhead.

There will be one more update to the aux fan ducts, changing the height to reflect the nozzle, and possibly bringing them off the panel a little to increase the intake area.

Hours Spent: 7
Total Hours: 182