Project Logs

Collapse

• Picking up work

  treibair • 05/28/2017 at 18:24 • 0 comments

  So, first thing to do is to put some pressure on this thing. I want to have the first digitized reel by September 1st.

  The recent arrival of my own 3D printer makes me want to pick up this partially shelved project again.

  I was always worried about the alignment of the parts left and right of the scanner. So I shall borrow from the Prusa i3's construction with threaded rods and printed parts. I will build a frame for the scanner and both reels to stand on. Everything connected and properly aligned by M10 and M8 threaded rods.

  So the bigger parts of the project from now on will be:

  • Build the frame itself: design and build connectors between the rods. The connectors shall double as feet for the entire frame and as support for the scanner.
  • Design and build attachments for the driving motor, the sensor, the two reels and two transport wheels as well as an enclosure for the electronics
  • Design and build the transport sensor itself (not the badly improvised one. This one is supposed to work properly)
  • Source a stepper and a driver board (on, easy, but is has to be on the list anyway)
  • Design and print a guide for the movie strip covering the entire width of the scanner

• Moving on the Super 8

  treibair • 04/05/2016 at 15:20 • 0 comments

  After the punched tape was sucessfully read, I moved on to actually scanning Super 8 Images. While I was doing my research about the exact geometry of Super 8 film I stumbled upon the CineToVid project which happens to do just what I am up to: scanning movies using flatbed scanners. I do not want to use anyone else's software, though but the CineToVid Wiki is a nice source for inspiration about the transportation device for moving the film segments forward.

  For the first steps I simply advance the segments roughly by hand, as for I just want to scan a few frames for a short (think two seconds) test sequence, since for the moment it's more about the software to extract the frames. The picture in the gallery shows a few scanned frames with less than perfect settings.

• Reading the tape

  treibair • 04/05/2016 at 15:12 • 0 comments

  Scanning the punched tape was rather straight forward: I just moved the tape over the scanner segment by segment, always advancing the tape by hand. The scanner Access is done using the sane command line Interface. The further processing of the captured Images is done using GNU Octave, an Interpreter language similar to Matlab.

  The analysis first searched for transportation holes to identify the x-Position of a byte on the strip of tape, then the individual bits are read and interpreted. In the end, the octave script writes its output to a text file. It turned out that - rather unsurprisingly - the punched tape contains G-code for some CNC-equipment.

  N1 G90
  N2 M71 
  N3 T0000 
  N4 R02 2000
  N5 R33 600 2000
  N5 R33 600 
  N6 R32 297 
  N7 G0 G90 G53 X160. Z190. 
  N8 M42 
  N9 T0113 
  N10 L999 
  N11 G59 X92. Z309. 
  N12 M75
  N13 L99
  N14 M33
  N15 S1000 M03
  N16 G0 X0 Z3.

• Punched tape

  treibair • 04/05/2016 at 15:08 • 0 comments

  The Inspiration for the Project is largely drawn from Constantin Gillies' great series of "Extraleben" books. In the first of which the heroes find some Information on punched tape. To get the Information from the tape, they photograph it in segments and then parse the pictures.

  My parents have quite a frew Super 8 movies from my childhood days and I always thought about having them digitized one day. I thought about the process that was described in the book and wondered, if it could be fesible to scan Super 8 movied on a flatbed scanner...but since this apparently would require quite a lot of precision I thought I might be better off whith replicating the punched tape feat from the book first so I went off to ebay to get myself a few meters of vintage punched tape.

View all 4 project logs