Preparing Steve for printing ended up being the most difficult part of this project.
By the end of our idea session, we knew exactly what we wanted our object to be. However, all we had was two twisted objects overlapping in Rhino that simply would not join or boolean.
In the final few minutes of class we tried a few simple fixes together. We tried splitting the faces and deleting the inside bits, but that didn't really work. We tried remaking the object in various ways, but none of them would join. On the off chance we did get something to join, it would be filled with intersecting faces. We ran out of time and decided that we'd both take the file home and see what we could do.
I tried a few simple things myself, such as changing the objects into solid forms and removing inner faces manually. I didn't get very far, though. I had trouble dividing the long, twisting surfaces into smaller pieces to work with, and couldn't find a reliable method to make it happen.
Finally, my partner messaged me that she had figured it out! She said she manually deleted all of the inner faces by splitting the object into an inner and outer shell. Kudos to her, that sounds extremely time consuming!
She sent me the file with a few different options for finishing the inner layer without supports. She even made notes all over the file so I could follow along with her process!
We ended up deciding to simply finish the inside with a cone and print our object upside down from what we originally envisioned. We also chose to remove the horizontal pieces between the holes on the side, worrying that they wouldn't print without supports. I added the little cone, then got to work exporting the file to Cura to prepare it for our print.
The object was not completely sealed as a solid, but we wanted to keep it open to be able to see the insides. Cura would automatically cap the holes in our object when it was imported. I tried going back to Rhino and using an offset to make a hollow but solid object, but the offset failed every time. After a few attempts, I went back to Cura to investigate. I ended up simply setting the top and bottom thickness to 0 to keep each side open and uncapped.
I chose to do a quick test print to make sure everything was working correctly before my partner arrived at my house to make the final print. I am very glad I did! The test print was done at full scale with a 0.4 layer height for a crude but quick test. I chose to use my rainbow PLA filament since that's what we were going to use for the final object, and I had never used it before.
The first attempts were a total disaster. They kept coming off my bed no matter how many times I leveled it. I ended up giving my bed an extra half turn or so to raise it closer to the nozzle for better adhesion, making the first layer a smushed mess, but giving it more area to adhere.
The test print was crude and ugly, but it did its job. Some areas didn't adhere to the previous layer due to the high layer height, so there were gaps and holes all over. The inside was filled with strings. However, I felt that all of these issues would likely be resolved simply by printing with a normal layer height, so I left it.
I made some adjustments in Cura such as adding extra retraction distance to prevent stringing. I also discovered "surface mode" which treated my open polysurface as a solid model of its own, removing the need for having the top and bottom thickness set to 0. Perfect!
The next day, my partner arrived at my house for the final print. I introduced her to Cura, then we went right to my printer to get things going.
The final print ended up great. We just plugged in the card, waited for the printer to heat up, and watched it go. In the meantime I taught my partner about how my printer differs from the school printer, and we chatted about different filaments and printing techniques. She hung around to watch the print for a bit, then left me to babysit it for the rest of the print.
The first print came out so good, I decided I wanted to make one for myself. I mean, I have a printer, why not? This part ended up being more of a personal side project, but I figured I'd include it here anyway!
I decided to play with the orientation this time and see if I could print it the right way up. It... Didn't go well. The center cone had only a small surface area attaching it to the base and it snapped off, leaving my printer pushing it around and printing in mid air.
Back to the drawing board. I also wanted to try a different technique this time. The first print was done with a single perimeter, using the spiral vase setting. The second go I decided to print it using two perimeters. I also wanted to cap one end to use it as a vase of sorts - this became a bigger issue than I thought.
Using surface mode in Cura would not allow a cap on either end, even if the model had one. I tried everything - it just wouldn't work. I even reworked the model in Rhino, getting rid of extra faces and filling holes as I went.
Finally, I gave up and switched it back to solid mode in Cura. This automatically filled any open holes for me. I set the bottom thickness to 0 so it would keep the bottom open, but cap the top. This seemed to work just fine and did exactly what I wanted! I also decided to use the model that included the horizontal beams on the inside, just to see if it would work. I was fully expecting it to fail spectacularly.
From the outside, the second print turned out great. Because of the double wall it used filament faster, which made for a faster color change. I almost got three colors on this one!
However, the inside was string hell. Strings everywhere. Stringageddon.
I still have no idea why that happened after the first print was so nice. Maybe I accidentally reset the retraction settings when setting it up? This filament is extremely stringy to begin with, so I honestly wasn't too surprised by this. In any case, some loving attention from my jewelry files took care of most of this.
By the end of our idea session, we knew exactly what we wanted our object to be. However, all we had was two twisted objects overlapping in Rhino that simply would not join or boolean.
In the final few minutes of class we tried a few simple fixes together. We tried splitting the faces and deleting the inside bits, but that didn't really work. We tried remaking the object in various ways, but none of them would join. On the off chance we did get something to join, it would be filled with intersecting faces. We ran out of time and decided that we'd both take the file home and see what we could do.
I tried a few simple things myself, such as changing the objects into solid forms and removing inner faces manually. I didn't get very far, though. I had trouble dividing the long, twisting surfaces into smaller pieces to work with, and couldn't find a reliable method to make it happen.
Finally, my partner messaged me that she had figured it out! She said she manually deleted all of the inner faces by splitting the object into an inner and outer shell. Kudos to her, that sounds extremely time consuming!
She sent me the file with a few different options for finishing the inner layer without supports. She even made notes all over the file so I could follow along with her process!
We ended up deciding to simply finish the inside with a cone and print our object upside down from what we originally envisioned. We also chose to remove the horizontal pieces between the holes on the side, worrying that they wouldn't print without supports. I added the little cone, then got to work exporting the file to Cura to prepare it for our print.
The object was not completely sealed as a solid, but we wanted to keep it open to be able to see the insides. Cura would automatically cap the holes in our object when it was imported. I tried going back to Rhino and using an offset to make a hollow but solid object, but the offset failed every time. After a few attempts, I went back to Cura to investigate. I ended up simply setting the top and bottom thickness to 0 to keep each side open and uncapped.
 |
| Cura's automatic caps on the preview |
I chose to do a quick test print to make sure everything was working correctly before my partner arrived at my house to make the final print. I am very glad I did! The test print was done at full scale with a 0.4 layer height for a crude but quick test. I chose to use my rainbow PLA filament since that's what we were going to use for the final object, and I had never used it before.
The first attempts were a total disaster. They kept coming off my bed no matter how many times I leveled it. I ended up giving my bed an extra half turn or so to raise it closer to the nozzle for better adhesion, making the first layer a smushed mess, but giving it more area to adhere.
The test print was crude and ugly, but it did its job. Some areas didn't adhere to the previous layer due to the high layer height, so there were gaps and holes all over. The inside was filled with strings. However, I felt that all of these issues would likely be resolved simply by printing with a normal layer height, so I left it.
I made some adjustments in Cura such as adding extra retraction distance to prevent stringing. I also discovered "surface mode" which treated my open polysurface as a solid model of its own, removing the need for having the top and bottom thickness set to 0. Perfect!
The next day, my partner arrived at my house for the final print. I introduced her to Cura, then we went right to my printer to get things going.
The final print ended up great. We just plugged in the card, waited for the printer to heat up, and watched it go. In the meantime I taught my partner about how my printer differs from the school printer, and we chatted about different filaments and printing techniques. She hung around to watch the print for a bit, then left me to babysit it for the rest of the print.
 |
| Ta-da! Steve. |
Print #2 - Steve 2.0
The first print came out so good, I decided I wanted to make one for myself. I mean, I have a printer, why not? This part ended up being more of a personal side project, but I figured I'd include it here anyway!Print #2 - Steve 2.0
I decided to play with the orientation this time and see if I could print it the right way up. It... Didn't go well. The center cone had only a small surface area attaching it to the base and it snapped off, leaving my printer pushing it around and printing in mid air.
 |
| Oops. |
Back to the drawing board. I also wanted to try a different technique this time. The first print was done with a single perimeter, using the spiral vase setting. The second go I decided to print it using two perimeters. I also wanted to cap one end to use it as a vase of sorts - this became a bigger issue than I thought.
Using surface mode in Cura would not allow a cap on either end, even if the model had one. I tried everything - it just wouldn't work. I even reworked the model in Rhino, getting rid of extra faces and filling holes as I went.
 |
| Where is the hole? WHERE IS IT? |
 |
| Maybe those weird sticks have something to do with it. |
 |
| Just gon' cut that off... |
 |
| Or not. |
 |
| I ended up just lofting new surfaces in there. But it didn't fix the hole issue. |
Finally, I gave up and switched it back to solid mode in Cura. This automatically filled any open holes for me. I set the bottom thickness to 0 so it would keep the bottom open, but cap the top. This seemed to work just fine and did exactly what I wanted! I also decided to use the model that included the horizontal beams on the inside, just to see if it would work. I was fully expecting it to fail spectacularly.
From the outside, the second print turned out great. Because of the double wall it used filament faster, which made for a faster color change. I almost got three colors on this one!
However, the inside was string hell. Strings everywhere. Stringageddon.
 |
| (send help) |
I still have no idea why that happened after the first print was so nice. Maybe I accidentally reset the retraction settings when setting it up? This filament is extremely stringy to begin with, so I honestly wasn't too surprised by this. In any case, some loving attention from my jewelry files took care of most of this.
The horizontal beams on the inside didn't totally crash and burn either, which was a pleasant surprise. You can actually see through the vase from one end to the other through them!
In the end, Steve 2.0 was a success. I think I might fill the bottom of it with sand or pellets and put pencils into it.