Organettes, Perforators, Etc
By John Wolff
The recent discussions about organettes and roll punching have at last prompted me into writing rather than just reading. When I joined this list about 12 months ago, I enquired about MIDI or template systems for preparing rolls for a 14-note organette to entertain and educate some primary schoolkids. I got some encouragement from several members of the list (Robbie Rhodes, Mark Fontana, Horst Mohr); Ed Schmidt sent me his extensive catalog of recuts, from which I ordered a small selection (and have been very pleased with them).
Anyway, all this inspired me to "have a go" at generating templates for my own rolls. On the principle of working from the known to the unknown, I took one of my decrepit "original" rolls and manually transcribed the holes into staff notation. (In doing this I noticed that the timing and durations of many of the notes were "all over the place". I couldn't decide at first if this was sloppy coding or skilful arranging - now I suspect it was the latter). Next I entered this into a music notation program (I use Noteworthy Composer (shareware) and a Roland SCD-10), experimented with it a bit until the PC sounded pretty much like the original, and exported it as a MIDI file.
Next I tracked down the MIDI file specification (one version of which was posted here recently), and set about trying to decode the contents. Now I have admit that when I learned programming it was in Fortran on punched cards. But I have since taught myself a fair amount of Turbo Pascal, so it wasn't long before I had a working imitation of a MIDI dump program. It did much the same thing as MF2T (or MIDI2TXT that was mentioned here a while back), but writing one of these yourself is probably the only way to get to grips with the innards of the MIDI file structure.
Next came some "enhancements" like merging separate MIDI tracks into a single play list, adding a translation table to map MIDI notes to instrument notes, and checking that all the notes in the MIDI file actually exist on the instrument (the organette does not have a full chromatic scale). The final big step was to adapt some bitmap graphics stuff I did years ago (actually it was to do with Mandelbrot sets on a Kaypro CP/M machine) to print a roll template on an Epson dot matrix printer. The organette rolls are 7-3/4 inch wide, which just fits nicely. Another hour or so with the Stanley knife on the kitchen table and the reproduction "Ring the Bells of Heaven" finally saw the light of day. I reckon it sounded pretty much like the original, and I was rather pleased with it.
Since then I have made a dozen or so new rolls by this method - kid's nursery rhymes, folk songs, and a couple of things from my grandmother's hymn book. "Arranging" in 14 notes in A major is a whole new art that I'm having to learn real quick. (I play early music on the tenor recorder, where I've only ever had to worry about one note at a time). I could sure use a few arranging tips from a real musician!
Anyway, all this tedious cutting of holes had started me thinking about computerising this operation as well, and I had started roughing out some ideas for a perforator. The plan at the moment looks remarkably similar to those that have been described here recently - a single interchangeable punch and die set driven transversely, with a friction feed roll drive. I don't claim it as original - it's really just a mechanised version of the old Leabarjan manual perforator. I figure a surplus wide-carriage computer printer could be recycled to provide much of the drive mechanism - I have a stepper motor and 24" lead screw from an old accounting machine which looks like just the thing. I don't know if I'll ever get to building it, but I'm certainly interested in the discussion. In the meantime, the templates and the trimming knife are working just fine, and giving the kids a lot of fun.
John Wolff.
|
|
|
Mechanical
Music Digest™ Archives
