Week3

Week 5: Printing and Scanning
Click here to go to home page

	This week's 3D printing and scanning exercise shows one how items are constructed and deconstructed digitally. I chose to construct a tool holder for the tweezers and the XActo knives we seem to use all of the time. Instead of having to hunt for them I wanted to buy a holder that mounted to pegboard we have at our lab stations. I didn't find what I was looking for. This then became an item I could print and custom make to my liking. This fits in with the idea that the Fab Lab is intended for a market of one. I made the original model in Inventor. I have experience with the program so this portion did not take long.
	The file went to the Stratosys 3D printer. I loaded the file easily and located the printing location on the base plate using the software linked to the printer. I ran through all of the layers to make sure there weren't any errors.
	The model took 5 hours to print which is longer than what I had anticipated. I would make the walls of the box thinner next time to save on material and print time.
	Lessons learned when printing included: When using the shell command the embossed text shells as well causing protrusions inside the box. Two holes in the back and the embossing in the front required extra support material. I knew there would be some but I was not anticipating this much extra material. The extra material required about 2 hours for removal. Overall the printing process was straight forward.
	Abu, Mike, Scott and I worked to get the Next Engine scanner up and running. It took 2 hours to load the software for the scanner. The scanner would twitch but not actually work to scan an object. We finally figured out that the software did not run on Windows 8. When loaded onto a machine running Windows 7 it took less than 5 minutes to load the software and get the scanner to work.
	My 3D XActo holder is about to become a digitized box. I chose to do 12 scans after watching Mike do 8 scans and having to spend a lot of time piecing his model back together. 12 scans may be a little excessive given the flat geometry of the box but hopefully increased scan time saves time in the long run.
	My scanned model was centered fairly well and scanned with few flaws. Items learned from this scan included: Scanned objects need to be centered Black backgrounds are used to prevent background clutter Holders keep objects from vibrating as they are rotated
	The problem with proprietary software - specific to a particular machine - is that retrieving files is difficult once the computer, hardware and software are gone. I tried exporting files from the scanning computer to my USB sticks. Only one or two images were saved. The original files are attached to the computer hooked to the scanner and are now inaccessible. I went on to trim the excess appendages off of my model after the scan was complete.
	I attempted to make my own 3D printing setup in addition to working with a purchased scanner. I used a lazy susan bearing and hung the bottom platform off of the bottom of the bearing. The center platform is held stationary on a pillar that is attached to a base resting on a table. It was important to find the center of a square before cutting the corners to make a circle. By making squares, finding centers, then trimming boards to circles, the center point through all of the pieces was maintained to some degree of accuracy. I looked at links online and at the commercial printer. and saw that there had marked concentric circles to indicate center. I have rings to help align objects. I included saw kerfs around the edge of the plywood circle in 15 degree increments to help determine camera angles. In addition to using scrap plywood from home I had $8.19 in 1/2" x 2' x 4' plywood; 1- 6" metal lazy susan bearing; about $8.00 in a cheap camera tripod; and $ about 4.00 in bolts and washers. The action is smooth enough. It is easy enough to align. I used a Nikon CoolPix L4 for taking pictures.
	I tried taking 9 shots and importing them into Autodesk 123D Catch. The upload went fine. The photos were compiled but there was an error at the 77% mark. I think there are several problem that I will continue to work on. They are: Inconsistent exposure among photos Unlike the purchased scanner that uses a black background the Autodesk software wants a detailed background to help stitch photos together. I need to remove the white background -- maybe.
	I wanted to make sure that the error message I was getting was not due to the scanned object being too similar to the background. I took pictures of a bass bird/paper clip to be sure. The same error appeared. I tried taking photos using a paper plate on a kitchen table and the error still appeared.
	In both cases 123D Catch indicated that the photos were uploaded. During the create capture I received: Process Error: An unknown error occurred during the upload process. Please check your connectin and try again later. I can try the process at a different time during the day. I am pretty sure my home made image capturing device will work. I think I will need to visit the tutorials and spend more time troubleshooting the issue. I would also like to try using the Kinnect sensor when time permits. February 26: I tried using 123D catch from school and came up with the same error. I then tried the "email" response instead of the "wait" selection. The email message indicates that "The application was unable to create your capture because not enough of your images could be stitched together". It refers me to the 123D Catch site and the video tutorials. I will review them as I believe my exposures aren't consistent enough and that there is not enough background information to allow the stitching to occur.
I took the opportunity to photograph an indoor carousel on a recent excursion. So far it is working to stitch the photos of the carousel together using 123D catch. It is taking a bit of time. I am afraid if I quit then I will lose my work. We'll see what happens.
	The photos I am stitching together have more and more green arrows indicating completion but the stitch looks much the same as when I started. The first picture is the stitch at the beginning of the process. The second shot shows the stitch with photos halfway stitched. I wish I would have taken more photos. There is too little overlap in some of the photos to get good reference points. I think I will also need to move more dynamically between photos when I am stitching in order to get the 3D effect.