Self made 64 tracks scanner for card board organ books.
The idea of making an organ book scanner of my own has been in my mind for a longer time. I wanted to make an equipment, which was easy to build, i should work on a nowadays interface (so an USB port) and not demand too much knowledge of electronics. Although I had done this hobby for thirty years ago, I did not want to make the electronic plats of my own and solder sensitive components.
You can find several subjects about this on internet but often the old midi port or the parallel port has been used. After a while searching I found at www.midiboutique.com a mkc64u module. This module had an USB connection and can read up to 64 tracks. Price: about 180 euro including shipping
I have purchased such module and received it after 2,5 weeks. Connected on a USB port, the first LED started to burn an when I simulated a contact , a second LED gave an indication. It worked. Now just one problem left: the software. An old version of Cakewalk gave a lot of trouble on my Windows 7 PC. I discovered that Piet Paardekam (Midiboek / Noteur) als has written decoding software and after a small correction, this software was suitable for upper standing module. Installing, configuring and testing and YES, the software wrote a midi file. I opened it with Noteur, which showed exactly my results. Bad contacts made bad notes with many interruptions. So that was a problem, which should be solved. For transporting the book, I used rubber rolls from a copier and a windscreen wiper motor, which is strong enough for this job.
Then I have searched for an other solution for scanning and found it by using infra red cells and LEDS. Because they are available with a diameter of 3 mm, they are usable in bearings of 3 mm with a center distance of 3,5 mm (Limonaire). The only question there was how the electronics will react and how sensitive were the cells. As photo cells I have used the L-932P3C cells and as IR LED the L-934F3BT at Conrad Netherlands. Those two work very good together. Even on a distance of 10 cm the photo cell react on the LED. So suitable for my project. For mounting the LEDS and photo cells I used 7 mm thick high-pressure laminate plate. For the photo cells I used for the first 3 mm a drill of 3 mm and the rest a 2,5 mm drill. For the LEDS i drilled in the first instance 3mm. This gave a much better result but on some places the light beam should have been made much critical.
A new plate has been made with the first part of the bearing 3 mm deep with a 3 mm dril and the rest with a 1,5 mm drill. Due to this, the light beam was more accurate and the scanner was working perfect. I could even double the scan speed without loss of information.
On upperstanding image you se an unedited scan in Noteur. The lower grey tracks are not used tracks. Also you can see that the notes does not have interruptions. With a MPL (see manual Noteur) routine this scan can be translated into a usable midi file.
Concerning the connection of the photo transistor: the collector should be connected on bus 9-10 of the diode matrix per 8 connections (see documentations of the mkc64u module) and the other connection on the corresponding connection of the concerning bus.
Because the scanner works on an 12V 7A battery, the LEDs will be connected serial with a resistance of 330 ohms. The bearings on the used experimental print stands 2,5 mm from each other. To mount the cells and the LEDs correctly for a center distance of 3,5 mm, they should be grouped in turn in groups of 2 and 3 pieces and between them jumping over 1 bearing.
 |