Week 5 3D Scanning and Printing
I have designed a clip to hold a hard casing of a Iphone on a bicycle bar. It is printed on an Stratasys Dimension. The machine has issues with the walltickness around the axis. Instead of filling the walltickness, the machine left it open and printed only 2 lines of material.
The same problem with the PP3P Up. But the Up closes the gabs better than the Dimension
After a design change and increase of the walltickness by 0,5 mm the walltickness is printed closed.
I designed a similar part for the leafbot. Two parts printed at ones that should be able to turn. Just as with the other clip, I designed it with a play between the joints of 0,5 mm. This worked fine for the clip. The Up and Dimension support material was easy to move. I designed this new part to be printed on an Ultimaker 2
In this picture the axis is been printed. The gab of 0,5 mm is visible
The part was printed with support everywhere. But the strange thing happend that the Ultimaker didn't print support material between the parts. The result is poor. The axis is not printed well. Very worse printed I would say. The gab between the parts so big that they have alot of play. Much more than the 0,5 mm. Also stange is that the axis aren't printed the same way. One worse then the other.
Almost half of the axis is not printed
The gab in the middel section of the part should have bene filled with support material but nothing is there.
The bottom is poorly printed.
On tuesday 1 marz I had to work on the Rapidpro fair at Veldhoven the Netherlands. It is fair for 3D printing. I showed my poorly printed sample to the material supplier Colorfab. They advised me check the Ultimaker. The Ultimaker 2 has some issues with the quality of the PTFE COUPLER. I came home late, so I didn't checked the coupler yet if the coupler is worn. Other thing to be changed is the slicer software. I used the standard Cura. I was advised to use Simplify 3D because it can control the position of support material better then Cura. This is what I will discus with my fablab administrator to pursuit this sofware.
My interest was to investigate if the Ultimaker 2 has similar problems with thin walltickness as the Dimension and PP3P Up and what the minimal walltickness is what can be printed on an Ultimaker 2. So I designed an test part. Goal was to investigate the minimal sharp edge and the minimal printable walltickness. I also design samples to test the minimal printable draft angle without support material.
I startet to draw a corner with a 60 degree angle and reduced it to 10 degrees to investigate what effect these first test has. I startet to draw a minimal wallthickness of 0,1 mm and increased it by 0,1 mm to 1,3 mm. I have put this file in to the Ultimaker, but it didn't print anything. It startet to print somewhere in the sky. I checked if the Ultimaker bed was leveled well. After checking and adjusting the bed, I startet the print again, and again it startet printing about 6 mm above the bed. So for this reason I changed the the minimal walltickness of the 2 smallest cirkels. The first cirkel I filled complete and the cirkel 0,2 mm I increased to 1,3 mm. I started the print again and now the Ultimaker worked fine.
The first print was printed with the standard settings of the Ultimaker. Shell tickness of 0,6 mm ans speed of 60 mm/s. This resulted in a to thin walltickness. The quality of te top side of the
print was worse. I also noticed that the smallest corner was not printed well. It is not fully filled. The length is also not correct.
CAD dimension= 13 mm
60 degree = 12.83 mm
50 degree = 12.75 mm
40 degree = 12.65 mm
30 degree = 12.50 mm
20 degree = 12.24 mm
10 degree = 11.01 mm
This is a suprising outcome to me. The sharpness of the corner has a big effect on the length of the model.
The cirkels are also not printed correct. The walltickness of 0.3 mm is not printable. It is to thin. All the letters have not been printed as well.
Walltickness of the cirkels:
0.3 mm = -
0.4 mm = 0.71 mm
0.5 mm = 0.68 mm
0.6 mm = 0.79 mm
0.7 mm = 0.88 mm
0.8 mm = 1.00 mm
0.9 mm = 1.12 mm
1.0 mm = 1.08 mm
1.1 mm = 1.17 mm
1.2 mm = 1.38 mm
1.3 mm = 1.55 mm
The outside dimension of the block should be 80 x 94.5 mm. The measured dimensions are 79.54 x 94.25 mm.
The outcome of this test is that the wall thickness of the samples totaly don't match with CAD model dimension. The outside dimensions also showing a big diviation to the cad model
The sharp corners 10 and 20 degree gives problmes with the filling of the model. Open gabs, like the Dimension and Up, are visible.
The next test was to test with a thicker shell walltickness of 1.2 mm instead off 0.6mm. Also the speed is decreased from 60 to 40 mm/s.The result is much better surface quality
The dimensions of the triangles and the cirkels are very close to the 1 test sample. Also the outside dimension are similar
CAD dimension= 13 mm
60 degree = 12.79 mm
50 degree = 12.80 mm
40 degree = 12.64 mm
30 degree = 12.52 mm
20 degree = 12.21 mm
10 degree = 11.02 mm
Walltickness of the cirkels:
0.3 mm = -
0.4 mm = 1.04 mm
0.5 mm = 0.81 mm
0.6 mm = 0.78 mm
0.7 mm = 0.86 mm
0.8 mm = 1.17 mm
0.9 mm = 1.03 mm
1.0 mm = 1.23 mm
1.1 mm = 1.21 mm
1.2 mm = 1.40 mm
1.3 mm = 1.38 mm
The outside dimension of the block should be 80 x 94.5 mm. The measured dimensions are 79.85 x 94.38 mm.
The overhang up to 60 degree is stil ok, below the 60 degree the surface becomes worse and worse. Ending in that the tip of the corner is very poorly printed. The same issue as with the other draft testing.
The test is for me not finished yet. I like to do some more test to check the different parameters, to test an other slicer software, and to do some more test with design of the part itself to improve the print quality. For now my time is up.
At this moment I do a project with the Hospital MST in Enschede the Netherlands. As FabLabEnschede we work together with the department Radiology to use a CT scan of a frontface of a patient to make a 3D printed mask, like the mask of the Phantom of the Opera. Tis is needed to get radiation in to the skin instead of deep into the body. The mask must provide resistance to the radiation so that the radiation comes in to the skin where the abscess is located. The mask must fit perfectly to the outside shape of the patient, because air between the mask and skin is bad for the result of the radiation. Below pictures of the CT scan. The first test have been performed on a test head. This is a head there is a real skull inside. So in the CT scan result there is a skull visible. This test head is used to check the CT scanner.
The scans have a dicom extension. With a open source software the files is saved as STL.
In Rhino the STL file was imported and a section was cut out of the face. This cut out was offset with 8 mm.
This is the mask for the patient. Below the mask the abscess is located. Work is still in progress. The mask will soon been printed if we can find the right 3D print material that comes close to the density of the human skin. The material must also be tested how much radiation it can take before is fails.
With the sense scanner I scanned my cycling shoes. My goal is to make my own designed cycling shoes out of carbon fiber. I try to do that in May during the week assigment composites.
The scan is not perfect, but I use the scan only to check the shape when I design in SolidWorks my new shoes
This is the bottomside of the shoe, to be able to copy the shoe profile.
This is the inside sole of the shoe. I use this to copy the sole profile.
In the sense scanner software I used the special settings. The object side 50 x 50 cm. So I had to scan within this area. If the focus piont of the scanner is not in this area the software will lose it track. This was the most difficult to hold the scanner in this scan area. Several times I had to stop the scanning proces and start again. In the software of the Sense I cutted the debri and surrounding away.
Files of week 5
Because the hospital MST Enschede is owning the CT Scans I cann't put the 3D scan files in my website