
85
Project Caretaker. Part 6. Modeling and print
This article is part of a series
- Project Caretaker. Part 1. Beginning
- Project Caretaker. Part 2. Design
- Project Caretaker. Part 3. Photoresist
- Project Caretaker. Part 4. Chassic
- Project Caretaker. Part 5. Tracks
- Project Caretaker. Part 6. Modeling and print
Modeling is also one of the fields that interest me. But unlike many others, I have very little practice in it - as my girlfriend says, barely a cat's tear worth.
There were several attempts to approach the field through 3D Max - I drew an apartment project and even rendered it. For example, back in 2016, I designed a desk with a cabinet for my office:
I also made a bathroom design project, but there the main modeling was done in the simpler SketchUp, with 3D Max acting as the rendering engine, which allowed for a more realistic result (this was already in 2022).
I also installed Blender and did something with it, but most likely didn't get beyond simple tutorials, so I'll have to learn everything again.
Fortunately, we already have a model (if you can call it that, with the mesh quality that 3D generators provide):
I've already adjusted the approximate scale to fit the dimensions of the contents based on this model in Fusion 360:
The task is to retopologize the model with fewer polygons.
This article isn't a Blender tutorial, so I'll only briefly describe the process:
- Import the ready-made model
- Set up cursor snapping to the model's geometry
- Add a mirror modifier (if we want a symmetrical model)
- Slowly, tediously, and painstakingly "trace" our curved model with polygons
- Make small bevels where we want to keep rougher forms
- Smooth the model to an acceptable level
The most difficult part is properly forming the polygon mesh to be even, uniform, with polygons of approximately the same size. This will be important later when smoothing the model.
Some areas didn't turn out very well for me (I hope it won't make your eyes bleed):
As you can see, I drew in too much detail even in places that shouldn't be visible - this will come back to haunt me. Now we apply smoothing:
And here's a comparison of the retopology with the original model:
We export to STL - import into Fusion 360 as a mesh - convert the mesh to a solid body - and use boolean operations to cut out our brackets. Don't forget to make a gap on all planes (I got burned on this and had to redo it twice). And we get holes in some places that are completely unusable..
We go back to Blender, cancel the smoothing, and manually move the polygon vertices to try to close these holes (quite a challenge to do this blindly).
Then we repeat all the steps again. This is a very slow process - converting such a complex model into a solid body and further working with it is quite time-consuming. After several iterations, we get a model that satisfies us.
Along the way, I also had to remodel the sides to make everything fit. I deleted polygons and drew new ones from scratch.
Then in Fusion 360, using regular sketches, extrusions, and chamfers, I freed up space inside the model for all components. It would have been easier to do this with the shell tool, but Fusion couldn't handle this operation on my model, and Blender, due to the crooked mesh in the chamfer areas, produced wild artifacts that I had no desire to fix manually.
I also divided the model into 2 parts - upper and lower.
I printed all this on an FDM printer.
The result was so far from finished and so problematic in terms of support removal that I decided to try SLA printing. The model needs to withstand some loads, and SLA is quite fragile due to the physical parameters of the resin. I was also sure that the resin would shrink significantly and unevenly, and was afraid of fitting problems. But contrary to my concerns, a test print of a calibration cube showed no shrinkage at all.
The lid wouldn't fit lengthwise on the build plate, so I had to rotate it. This is how it printed (something like 8 hours):
On my first attempt, I ran out of resin and ended up with something that looked like a Terminator - I even wanted to keep it:
But then I reprinted it anyway and kept this one for experiments. I printed with ABS-like resin, and the result surprised me with its strength, although one corner of the first version broke off - but that was likely due to insufficient clearances. I was more careful with the new versions.
Towards the end, I also designed all the necessary cutouts for electronics - eyes, LEDs, camera, external charging pins, and all that.
As you can see, the eyes and camera box are also separate parts.
There's no need to mention the quality - photopolymer printing provides almost final surface quality ready for painting.
-
Regional Championship in Competitive Programming
I'll diversify my posts about robots with a life story. I'll return to robot in… -
Project Caretaker. Part 6. Modeling and print
Modeling is also one of the fields that interest me. But unlike many others, I… -
Project Caretaker. Part 5. Tracks
We're designing and printing tracks on a 3D printer. We will print them using… -
Project Caretaker. Part 4. Chassic
Since we're printing - let's do it all at once. We'll be inventing the wheel. -
Project Caretaker. Part 3. Photoresist
Mastering new technologies. Not just Toner Transfer. -
Project Caretaker. Part 2. Design
Given my lack of advanced artistic skills, I'll turn to external help here. For…
TestUser 16.04.2025 13:31
bujkfX punidf AQrbHCS* * * Unlock Free Spins Today: http://louga.net/index.php?gxqare * * * hs=7283a718a93e7d0df4a7181b4f8ccafd* ххх* 17.04.2025 18:55
c86n8t* * * <a href="http://louga.net/index.php?gxqare">Claim Free iPhone 16</a> * * * hs=7283a718a93e7d0df4a7181b4f8ccafd* ххх* 17.04.2025 18:55
c86n8t