6. 3D Scanning and printing¶
This week’s Group assignment was to test the design rules of our 3d Printer, and explain its limits. The individual assignment was to design and 3d print a small object that could not be made easily subtractively. We also had to 3d scan and object and prepare it for printing.
Group project:¶
The group project for this week required that we all characterize the design rules of our specific printers. There are some vague guidelines for 3d printing, but every printer is different. We needed to determine our printers capacity for the following:
- Overhangs
- Bridges
- Clearance
- Dimensions
- Surface finish
- Infill
- Anisotropy
- Wall thickness
All the files we used to test these features can be found on the fab academy page, and on our Group Project page. I specifically helped with the research on these terms, so our group would understand what we were defining, and helped prepare the files and printers from 3d printing. I then helped measure the results and document our analysis.
Individual Design:¶
For my individual design I decided I wanted to make a small headphones case. My headphones are perennially tangled, so I wanted to make something simple I could wrap them around. I tried a couple of different designed but settled on a triangular one I found here. Then opened Fusion 360 (3d design tool) to design my own version.
I started by making a straight line that is the height of my case (2 inches) and used straid lines and a conic curve ending halfway down the height to make the outside of my case. I think my case will have a circular face instead of a square.
I then used the mirror tool to create a symmetrical curve for the other fourth of the case.
I then drew a triangle 2 x 2 x 2 inches with the midpoint of one side over the side of the first sketch.
I then used the sweep tool so this body would follow the outline of the path.
I edited the sketch a little so the dimensions would be more similar to what I had in mind. In fusion, design changes on a sketch will propagate through to bodies made directly with a sketch.
At this point, the entire body had a hole going through it, however I needed the bottom to be closed. To do this I split a small section of the body off the bottom, and used a press pull to close the opening. I then regroupsed the body.
Besides this general shape, I need 3 places for the ends of the headphone wires to rest. I found this website claiming the thickness of the cable was 2.76mm (0.1086614 inches) and confirmed this approximate through my own measurements with a ruler (I thought it was .11 inches). I extruded these rectangles to make a cut.
I then extruded a new body below this with a taper angle of 52.5 and dimensions of -.05, and subtracted it from the bigger body. I then rounded out all the corners.
I then rounded out the edges. The top will allow the cord to barely pass through, then the bottom will let the cord have some room to shift around (instead of being under pressure during all times while being used. I then rounded out the corners so there would be even less pressure when putting the cord in.
I rejoined the edges, and rounded more edges with fillet
I went to print it but I realized my file was too large and would take too long to print, so I decided to hollow it out.
I did this by “splitting the bodies” into multiple pieces so I could “shell” the insides, and use a “press pull” to close the insides. After this was done I regrouped the bodies.
To Shell- select the two sides because they are different dimensions) and shell. Wall thickness = .1inches (which is a bit bigger than what our printer can handle.)
I then added a small cylinder attached to a ball, and a little slit next to this cylinder so I will later be able to shake it and get the sphere to fall off. I was conscientious of the sphere’s area and the area of the slit so that the sphere will not fall out of the slit.
I had to remove the headphone slots in order to do the shell, so I put them back here. This time I set user parameters for chc(wire thickness) and co (½ wire thickness) just to make things go a little faster.
I decided to go back and give the bottom of the higher extrusion a taper angle(3 degrees), that will make it easier for the wires to slide through. I then chamfered the top side (that does not have those 3 degrees to help it slide through.
I then tried to 3D print it, but before printing I discovered that my 3d print was far too big. At first I tried to fix this by simply scaling my entire object down, but I needed the slots for the iPhone wires to slide through. I had made my spaces for the wire parametric in the beginning with the hopes that scaling did not change the parameters dimensions. However, it appears that scale effects parameters as well. So instead of re-designing the entire thing, I decided to cut out a vast portion of the outside of my earbuds holder.
When I did this, I also cut out the cylinder, so I had to redesign this. I followed the same steps as above on the bottom.
I also removed the other sides that were covering the inside. I wanted to minimize the amount of material I was using and the print time. I did this by splitting the bodies, then splitting the bodies in a triangle shape, so the interior and exterior would be seperated. I then made the exterior parts invisible, then regrouped the ones I wanted to keep into one body.
It ended up like this:
(separated, hollow parts visible).
(all grouped together)
Using the 3d printer:¶
We use the Lulzbot 3d printer and the software Cura. I made sure the bodies I wanted to print were visible (and the bodies I didn’t want were not visible) in fusion, then exported my file as.obj file, and opened it in our 3d print software. There are two major screens in Cura, prepare and monitor.
In the prepare screen, I placed my object and scaled it. I forgot to change my Fusion file to mm(which cura operates in) but I was able to do a little bit of math to figure out what the dimensions of my file would be in mm.
I switched to the monitor screen and preheated the print bed to 220 degrees and preheated the nozzle to 60 degrees. Preheating instructions can be found on the side of the material you are using (I’m using PLA).
After a few minutes, I made sure the bed and nozzle were warm. To test that the nozzle was working and the PLA was coming out of the nozzle correctly, I pressed “extrude,” which simply pushes a little bit of filament out of the nozzle. I then removed it with a pair of tweezers, let it run the “wipe nozzle” setting to remove any excess filament. When this was done, I hit “print” to print my headphone case.
Unfortunately, the first time I tried to print it I switched a decimal place and my dimensions were wrong. Once I noticed this I selected “Abort Print” and the print stopped. I scraped off the filament on the bottom. The second time the filament didn’t stick and I had to restart. The third time it worked!
Final product:¶
(you can hear the ball inside the interior if you listen closely!)
(side views):
3d Scanning:¶
The second part of our individual assignments was to 3d scan an item. We have a 3d scanner at my school, however this 3d scanner is notorious for taking a long time to scan objects. I do know there are a number of applications on phones and computers using “photogrammetry” (the science of taking measurements from photographs) to 3d scan, so I decided to experiment with one of these while simultaneously trying out our school’s 3d scanner. I ended up liking QLone a lot, because even though it requires more active participation, it goes a lot faster and is a lot more convenient.
The photogrammetry app I used is Qlone, you can download it on an iphone.
I started by trying to 3d scan a Bobblehead of Wonder Woman I had in my room. I did try a couple of different objects found in the lab before I brought this Bobblehead in, and photos were taken with the object being used as I discovered new features, so some of the images are not of this figurine.
The real object, for reference:
The app works by scanning on objects placed on a printable grid (found on their app). This does limit the size of the object you can scan, however our 3d printers aren’t that much bigger than the size of the mat, and you can scale the object in the software.
Once entering the app you press the plus button on the lower right hand corner to start a new scan. You then raise the camera over the object on the scan mat, wait for the semicircle to generate over the object, and then you rotate the map slowly with the camera at different angles. You can also rotate the camera and keep the mat steady, however I found some of the upper layers to come out better when moving the camera. Granted, this may also be the effect of me getting better at using the software and keeping the camera steady.
It then brings you to a processing area, where it takes the scan and renders it.
The top has a lot of distortions at the top that does not reflect the actual body. I tried to fix it with the “Flatten body” tool on the right side, however I discovered that this object only works with truly flat objects.
There are other ways to maximize the accuracy after the initial scan is taken. One of the most helpful ones was the “merge profile” tool, which allows you to scan another profile of the object, then merging the two by selecting identical points. You may notice that the top of my bobblehead’s head had a triangular shaped extrusion, that does not reflect the actual object. This is quite common in this app around where the top of the semicircle used to scan is. At first I tried to fix it by using the “flatten top” tool, but this was designed for much flatter objects and removed too much of my actual scan. Through some research, I discovered that I needed to use this merge profile tool, so the scan would see another face and be able to tell that there was no material to add above that original point.
I misunderstood this feature originally, and simply rescanned the same profile I had done originally. So the wonder woman scan didn’t turn out perfectly. By the time I returned to scanning, I had restored my bobblehead to my house, so I started experimenting with 3d scanning of a stamp that another student had made. However, the effects of this merge tool are evident in my other scans. Original body of stamp:
I tried scanning and going through the merge process for two different profiles. It made the end result a lot more accurate but there were still flaws
After scans:
I started to experiment with the post-processing features. The options are as follows:
Texture:
Texture allows you to edit top layer features by adjusting the saturation of the color, changing the brightness, allowing you to switch between “normal resolution” and “sharp” resolution. Color pick allows you to tap an area of your scan and select that color to paint with. Paint will color any areas you touch with white, unless you have previously selected a color from color pick. Blur blends the colors together. I changed the brightness and saturation a bit, built mostly left this alone
Resize/Shape:
Exactly what it sounds like, allows you to change the length, width and height of your object, and also do something similar to “compression,” the number of pixels (resolution) will go down when you use the simplify tool.
Sculpt:
The most useful tool here is “amplitude.” after selecting an area with the lasso tool (simple drag your finger in a circle/whatever shape) you can pull entire sections in and out (It’s almost like extruding). You can then use soften to make the edges created with amplitude blend into the rest of the object more. Push and pull will choose the direction of the amplitude, pulling will bring the section out and pushing will push the section into the object.
Clean:
I used this part to smooth edges. If you select a part with a lasso, smoothing it appears to push everything a little bit closer to an “average” position.
Using it:
I used the sculpt the most for my scan. I first used it to push some of the jagged edges in, then used a combination of the smooth tool and the soften tool to make these edges blend with the rest of the object. I also used smooth all over the object (this was done in multiple pieces) to make all jagged edges blend. This was especially helpful for relatively flat surfaces.
There was also one area where there was a hole in the bottom. I used the amplitude tool to extrude parts of the body to cover the hole. It came out quite jagged, so I used the smooth tool to make it blend and I thought it turned out quite well.
The biggest problem I ran into was user error. I discovered the most important part of generating an accurate scan to be keeping your hand steady, turning the mat at a constant rate, and not moving anything too quickly. My scans improved significantly the more I used the app.
Final:¶
Files:¶
To download something from Qlone I would have to upgrade to their premium service. I do not need to use the scan as of now, but if I ever did, I would upgrade to premium and export it as a .OBJ file and import it into the Cura software to print.
