The Ship Foundry Project (Part 2)

So in my last blog I posted some of the printing fails from my first couple of print jobs. While my last blog was a tragedy, this blog will be a comedy of errors. So after the Comet job I was discouraged by the failure on the winglets, but encouraged that the level of detail was even more intricate than I could have dreamed. These models were mostly printed on medium settings because I'm an impatient fellow and wanted my ships now!

So I thought if I took on the Vexor as a project, I could eliminate some of the problems I had with detail since the ship was simply bigger. I decided to try to print the Vexor from a different angle. After some debate as to what direction I should split it, I chose to tip the Vexor onto its end and split it where the forward engine pylon made a convenient base to print from. Unfortunately, this turned out to be a bad idea for two reasons:

1. Unbeknownst to me, rotating the model in SketchUp broke it and made it not manifold.
2. I didn't generate any support material, which made the pylon very unstable.

The problem began when Slic3r broke the manifold model. While the model was still "manifold" according to the software, it lacked any closed edges, which meant there were parts of the model that weren't rendered in the G-Code for the Vexor. When Slic3r runs into this problem, it decides that instead of doing something, it should do nothing. So there was several layers where nothing was printed on that part. Then suddenly there was G-Code again when there was suddenly G-Code to print over that part. The printer began printing in thin air. This caused a giant blob of plastic to form on the extruder head. This blob then knocked over the nascent pylon. Dauntless, the printer continued and by the time I came to check up on the print, there was a giant mess. The result is a hull bit that I can use as a test bed, or a Vexor wreck.

For the second attempt, I tried a split that didn't require a rotation, and just split it right down the middle. This print turned out much better, but resulted in support material being needed to get the hoses under the ship rendered properly. You can see the printer working on the right. This build looked much better. Once the job was done, I took some time to remove the support bits, using a razor blade, a screw driver, and some minor flesh wounds.


To join the pieces, I used a mixture of acetone and the same plastic I used to print the Vexors. This 'Plactitone' is a really useful tool because it basically welds your ship together so that the pieces are functionally one piece. Then you can use it to fill in the imperfect seams. Something I've noticed about 3D printers is they make little variations each time you run a print. It's not a perfect technology, and you can sometimes end up with variances of up to a millimeter. I chalk these up to how the process is not entirely identical each time. The heat from the external air, and variances in the plate and extruder can occasionally lead to poor alignment between parts. The result is the Vexor you see here, with a seamless join between two parts.

So onto the cool looking part: painting! I went down to the local hobby shop and let them know about my project. They gave me the following instructions:

1. Sand with fine grade sandpaper. I used 320 grain.
2. Clean with soap and warm water. Don't use hot. Remember that its extruded at 240 degrees Fahrenheit.
3. Dry thoroughly with a towel, or air dry for 30 min.
4. Prime with Solvent based Primer. I used Armory brand primer, black, for your warhammer-like miniatures.
5. Paint with Acrylic paints. I bought some Citadel brand paints from the hobby shop.

I got some additional advice care of Matto Voltaro of Aideron Robotics:

"A couple of things I've picked up painting minis over the last 20 years or so:

Always wash your hands before handling the mini, fingerprints don't take paint well and can cause an orange peel effect as paint drys over them.

I always use a liquid dish detergent like Dawn. Just a couple of drops in warm water. Too much and it's a real chore to get out of the detail. A quick soak of 4 or 5 minutes. Then a quick gentle brush with a soft toothbrush to get rid of any sandpaper grit or dust that's stuck in the finer detail. I allow to air dry overnight. I've found the $0.99 cans of spray primer from Walmart work just as well as the expensive private label brands from Games Workshop or Armory for a base coat. Thin coats, don't stop moving the can and don't spray in high humidity.

If you haven't painted mini's before, don't cake your color on in one coat. Build it up in thin layers. I prefer to have my paint around the consistency of milk. Colors like whites, reds, and yellows may be hard to build on a black undercoat. A coat of medium gray applied first on areas can help bring out the whites, reds, and yellows."

And here is the end result!

I decided to make my own Aideron ship paint job in lieu of CCP actually giving me a ship painting system. PLEASE MAKE THIS HAPPEN CCP!

The Ship Foundry Project (Part 1)

Hi, this is Blastil sounding off with my new Eve-Online themed blog! For this post I'm going to show off my latest side project that happens to be Eve related.

So a couple of months ago my boss bought the company a 3d printer for rapid prototyping. Fortunately we don't use this printer 24/7, so I asked my boss if I could take a crack at 3d printing. He enthusiastically agreed, and I started to plan my 3d printing projects.

So I asked some people in our Aideron Slack what I should print. Resoundingly the response was "Eve ships!" So armed with a laptop, Slic3r, Sketchup Maker, Blender, and the determination only afforded to space ship captains, I departed on a journey into 3d printing. Before I begin, I have to give a shout out to FATTOX4189 of www.thingiverse.com who has done most of the hard work for me and already exported the models from EVE and did a lot of the initial clean up work. These models are of Excellent quality, and stand up pretty well to printing, and cut up pretty easy in Slic3r with a few exceptions.

So a few notes about the process of 3d printing for those who are not familiar with the technology. first of all, 3d models in the form of .stl or .obj files are exported from your favorite 3d editing software (like Blender or Sketchup). These files are then imported into a software desgined to generate the G-code to send to a printer. Gcode is an industry standard code for any 3d design or manufacturing equipment like CNC mills or lathes. After that you take that Gcode to a 3d printer. There are several technologies for this, but our printer is an extrusion deposition printer which means it works like an inkjet except it uses plastic and not ink. For the sake of simplicity, when I talk about printing i'm only going to talk about extrusion deposition printers, since these are the only kind I have experience with.

Part of why I wanted to undertake this project is that I want to push this 3d printing technology into a different direction than we're doing at work. At work this printer is being used to make large models to attach to jeeps to pre-fit bits to each other before we make them in steel and aluminum. I wanted to see how well this thing applies to a level of detail that we simply don't do normally at work.

So the first ship I decided to try was the federation navy comet. After a lot of figuring out 'what the hell am I doing here' and learning a few very important lessons about how to even turn the printer on and get it to temperature I discovered that the best thing to do is find a symmetry point on the ship and split it there. then you can take that model and print it. This was the result.

Looks pretty good right?!

Awwwh shit. Apparently the printer doesn't do good with those thin little winglets. In fact the other winglet broke off this model later. I haven't gotten a chance to re-approach this ship yet, but I have learned a few things from when I printed Rainbow Dash about small printing small winglets that I want to try.

As you can see, I chose to do a vertical split on rainbow, and the wings wound up being extruded slowly but with much more substance to them. There are a few errors on her, but nothing the sword of EXACT-Zero won't fix.

Rotating the models comes with its own risks though. I've tried to rotate a few of the models and the results have been rather mixed. I'll show you why when I detail the Vexor project in another blog.