3D Printing And Scanning

This week's assignment is to characterize the design rules for 3D printing (in-group); to design and 3D print an object that could not be made subtractively (individual); to 3D scan an object (individual).

* All G-CODE, STL, and original editable files can be downloaded here.

* This week's group assignment is to test the design rules for the 3D printers available at Fab Lab Barcelona.

* Besides, we are required to individually design and print something with a standard FDM printer and 3D scan an object - in my case, I chose to scan myself.

* I also attempted to go through another 3D printing technique which is stereolithography with an SLA printer.

3D scanning

Individual assignment - 3D scan myself
Scanning with Kinect and Skanect

I used a Kinect Xbox 360 V1 to scan myself since I knew it's a common practice by looking at previous alumni's websites. Below are the detailed specs of the device:

The software paired with Kinect that we used to process the scan is Skanect. Here we can find a step by step instruction on using Skanect. Choosing myself as the scanned object means I required further supports from others: someone held and tilted the Kinect vertically (Santi - our instructor), someone rotated me while I was kneeling on the chair (my classmate Hala), and someone documented the process (Zoi - an MDEF student). The reason why I was not sitting on the chair is to avoid having to process the chair from the exported mesh later.

Steps to scan:

And here is the outcome imported to Blender (the size of the file is too big, therefore I just posted a screenshot here):

Processing the mesh with Blender

Let's make myself a bit prettier by 3D retouching the outcome! We had a class by Victor Barberan regarding fixing the mesh using Blender. I simply followed his instructions to work on my mess, oh I meant, mesh. Some steps followed (IMPORTANT: following the order can be life-saving. I started sculpting BEFORE switching to Edit Mode and I hit my head to the keyboard soon after):

And let me introduced you to the 3D Tue (not in a perfect state)!

I studied sculpture in college before, so the Sculpt Mode was really tempting to try. I indeed played around a bit by keeping the same amount of triangle faces from the scanning outcome, and fine-tuned the model with other sculpting tools, such as Draw, Blob, Crease, Grab, and Pinch. Of course, the model was way visually prettier and detailed. However, the size of the file was unnecessarily large, and I could not 3D-print it properly since there were so many non-manifold faces and messy layers. Cleaning them up could be a real pain, therefore I decided not to continue with it.

I didn't proceed to print myself, because I would like to allocate more time trying SLA 3D printing technique. In the end, I couldn't make it on time for the additional exploration, but everything is in process ;)

3D printing

First of all, what is 3D printing, or should I say, additive CAM? 3D printing is a technique of adding successive layers of material until an object is created, while subtractive CAM removes material from a larger piece (by cutting, drilling, milling, grinding or turning) to create objects. 3D printing has significant advantages: avoiding the waste of materials and allowing us to create models that would be tricky to be made using subtractive techniques (having undercuts, overhangs, nested parts or complex hollow parts).

Depending on the technology, the 3D printer deposits material (FDM), selectively melts and fuses powder (PBF), or cures liquid photopolymer materials (SLA) to create parts based on the CAM data.

Group assignment - Characterize the 3D printers

We proceeded with our tests using the most common printing technology: FDM. Fused Deposition Modeling (FDM) belongs to the material extrusion family of 3D printing techniques. In FDM, the materials used are thermoplastic polymers and come in a filament form. Once the nozzle has reached the desired temperature, the filament is fed to the extrusion head and in the nozzle where it melts. The extrusion head is attached to a 3-axis system that allows it to move in the X, Y and Z directions. The melted material is extruded in thin strands and is deposited layer-by-layer in predetermined locations, where it cools and solidifies. When a layer is finished, the extrusion head moves up and a new layer is deposited. This process is repeated until the part is complete.

Before participating in the group test, I took a look at this useful article in order to understand what we were going to test.

First test - General design rules

I teamed up with David Prieto and Antoine Jaunard for the first test. We used the Creality CR-10 S5 printer to print this test file using 1.75mm PLA. The detailed specs of the machine:

We used Ultimaker Cura in order to slice the model, to modify some important settings for the 3D printing process, and to generate a .gcode file. G-code is the language through which we can communicate with computer-controlled machine tools (in this case, the 3D printers) and give them instructions on what to do. The Cura apps installed in Fab Lab Barcelona's computers have the profiles of all available printers with tons of proper settings. However, we had to, and should only, modify some settings to manipulate the printing time:

Here you go the video recording the printing process:

The printing was done in 1 hour 27 minutes:

As per the images above, the quality of the print was quite good, many details were respected. Here are the test results after I measured carefully the outcome with the caliper:

Second test - Tolerance and joint clearance

For the second test, I teamed up with Roger Anguera, Bruno Molteni, David Prieto, and Lynn Dika. We used the Anycubic Kossel Plus printer, and our Fab Lab Manager Mikel Llobera guided us through the process. The detailed specs of the machine:

The estimated printing time was too long, hence, we had to scale-down the object to 70%. We then used the same settings as the first test. The printing was done in 1 hour 35 minutes:

Generally, the quality of the print was poorer compared to the one printed by the Creality printer. Here are the detailed test results measured by the caliper:

What I personally learned

Individual assignment - Design and print using FDM printer and PLA filament
Designing with OpenSCAD

I utilized a flower-shaped vase I designed in the 2nd week since it can be considered as an object that cannot be made subtractively (having overhangs and undercuts). My vase was designed in OpenSCAD, and I basically modeled a thin cube, then continuously cloned, rotated, enlarged and moved it up x 180 times to generate the base module shape(). 3 shape() modules were combined and subtracted to form the final model. In order to make sure of the ability to stand still of the vase as well as to reduce the total size, I modified it a bit by lessening the additional size of the cloned cubes.

Slicing with Cura

After exporting the .scad file to .stl, I started slicing the object in Cura. Since my design was more about a nice shape rather than an object with too many details, I used these settings:

The estimated printing time was more than 3 hours, therefore I had to reduce the size of the model by 50% directly in Cura, using the Snap tool. Ready to print!

Printing with PLA filament

I used the same Anycubic Kossel Plus printer to print my vase since we can use it without reservation. Josep guided me through some essential steps:

I then had some problems with the printing process. Although I reduced the overhang angle from the bottom of the vase to its top, the overhang angles between the twisty parts were obviously too big and the printer started to print in the air. In order to both reduce the size and maintain a nice shape, I modified the cube in my design to a hexagon cylinder by utilizing the module used in the 3rd week.

Then I faced the 2nd and 3rd failures. For the first print, I only selected the Skirt option in Plate Adhesion and the adhesion was fine. However, for the next print, the model started moving while being printed. I tried the second time by selecting the Brim option, but things didn't go well either. In the end, I had to print the 4th time with the option Raft selected.

Finally, I had the vase perfectly printed!

Final result

I brought the vase to the landlady where I'm staying, and here you go the hero shots of the final result:

Additional test - Stereolithography using SLA printer and UV resin

Stereolithography (SLA) belongs to the Vat Photopolymerization family of 3D printing techniques. In SLA, an object is created by selectively curing a polymer resin layer-by-layer using an ultraviolet (UV) laser beam. The materials used in SLA are photosensitive thermoset polymers that come in a liquid form. The build platform is first positioned in the tank of liquid photopolymer (UV resin), at a distance of one layer height for the surface of the liquid. Then a UV laser creates the next layer by selectively curing and solidifying the resin. The laser beam is focused on the predetermined path using a set of mirrors, called galvos. The whole cross-sectional area of the model is scanned, so the produced part is fully solid. When a layer is finished, the platform moves at a safe distance and the sweeper blade re-coats the surface. The process then repeats until the part is complete.

At Fab Lab Barcelona, we have some Formlabs SLA printers and a new Anycubic Photon S. The Formlabs ones were being calibrated, therefore I asked Mikel to let me join him in testing the exposure time of the Photon S printer. Some basic specs of the machine:

We tried to use this file to test the printer. R_E_R_F (Resin Exposure Range Finder) is a feature provided by Anycubic with which we can find the correct exposure time for any given resin. When any sliced file with the name R_E_R_F.pws is printed on a Photon S, it divides the platform into 8 equally sized sections and exposes each section for a progressively longer time. If we place 8 copies of the same model in those sections, we can compare different exposure times, and thereby decide which time gives the best results.

However, the estimated printing time was much longer than the FDM printers, and the people at Fab Lab Barcelona were about to clean up the 3D printing room and calibrate all the printers. Since it's better to do the test under the supervision of Mikel, we had to delay the test until we can find a better time slot and the machines are ready.

Conclusion
Additive vs. Subtractive CAM

Things I like about 3D printing:

Things I wish 3D printing could match subtractive techniques:

Design rules for FDM printers

Machine Settings Dimension accuracy Minimum clearance (Press-fit) Minimum clearance (Free-fit) Minimum hole size Minimum wall thickness Minimum distance between walls Maximum overhang angle Maximum horizontal span
Creality CR-10 S5 Layer height: 0.2mm - Infill: 10% + 0.08mm (x,y) and + 0.1mm (z) 0.3mm 0.6mm 0.6mm 0.1mm 0.6mm Up to 55° without support, preferably 35° 16mm
Anycubic Kossel Plus Layer height: 0.2mm - Infill: 10% + 0.15 - 0.2mm (x,y,z) 0.5mm 1mm 0.7mm 0.1mm 0.3 mm 45° -

3D printing techniques assessment

Technique Materials Printers Dimension accuracy Surface finish Supports required Lead time Post-processing techniques Fragility When to use
FDM PLA, ABS, Nylon, etc. Anycubic Kossel, Creality, Prusa i3 ± 0.08 - 0.2mm - Lower dimension accuracy Lower quality of surface finish Only required when there are joints or large overhang angles Relatively shorter Not always required - sanding, priming, cold welding, vapor smoothing, epoxy coating, etc. More rigid Rapid, low-cost prototypes and precision is not crucial
SLA Photopolymer resins (thermosets): clear, flexible and castable Formlabs, Anycubic Photon ± 0.01 - 0.1mm - Very high dimensional accuracy and intricate details Very smooth surface finish Always required Relatively longer Always required - sanding, spray coating, mineral oil More brittle Visual prototypes with fine details and strength/durability is not crucial

That's it, enough testing with 3D printers. The next step is to design the physical parts of my Final Project and to see how can I assemble parts made with subtractive and additive techniques. Before doing that, I still need to make it clear about how I can connect the modules physically, electronically and informatively. Hence, I will delay the task a bit and update later on the Final Project page.

Previous Assignment Next Assignment