Synchronous Perforating
By David Wasson
[ Editor's Note:
[ I met David Wasson because I liked the musical arrangements he played
[ on his band organ. David was cutting his rolls by hand at that time,
[ with an Xacto knife on a light table. David has since built a perforator
[ with a quite interesting design. It has freed him to concentrate on
[ the arrangements without having to worry about the mechanics and minutia
[ of accurately hand cutting little holes in paper. It also allows him
[ to test certain musical ideas on the synthesizer first, before comitting
[ to making a roll. David is considering improving the system, primarily
[ to get uniform chain perforations.
[
[ David shares Mike Ames's account. Mail for him can be sent to:
[
[ ames@foxtail.com
[
[ Jody
This letter is intended for Jody, Robbie, Wayne Stahnke and Richard Tonneson, but anyone else willing to comment is welcomed.
I have decided that I need to have my roll perforator run synchronously from the computer instead of asynchronously. The current mode of operation is to slow the MIDI file to a speed that is the same as the perforating speed of the perforator, and to shorten all of the events in the file by a certain amount depending on the time base and tempo, so events are not stretched by the perforating process. This new "perforator file" is then sent to the perforator in real time, although much slower, and a music roll is cut.
The purpose of the perforator is to cut music rolls from MIDI files. It has never been my intention to make copies of old music rolls. It presently is set up for 9 to the inch with .0625" holes. The rolls are 11 1/4 inches wide with a maximum of 100 positions left to right. The MIDI files are typically my own arrangements that I do for my 98 key band organ. The perforator has become my MIDI interface for the organ. It also eliminates the labor of cutting rolls by hand, which is how my first one hundred tunes were made.
The perforator is not a production machine and cuts only one copy at a time. Compared to hand cut rolls, this number of copies per run is perfect for me. For tempo compensation I am using a take-up spool which is the same diameter as the take-up spool in my organ. This also allows me to use, within certain limits, any thickness of paper.
The number of steps per inch with this take-up spool is 60. This gives me a resolution of .015" per step. With a step this small I have never before been concerned with being in synchronization with the computer. Since the MIDI file is the "master roll" the maximum error in hole position would be only .015". Subsequent copies would never be made as a copy of the roll, but always from the original source which is the MIDI file. Presently I have no provision for automatic chain perforations, as my hand cut rolls never had this feature. But I would like to be able to do this. In order to really have control of the perforator and to do things like chain perforations, it will be necessary to run the perforator in sync with the computer. With the present set-up, every 5th step makes a nice chain. This leaves about .013" of paper between holes in a chain pattern. Towards the end of the roll, when the roll is moving a little more than the .015" per step at the beginning, this will still probably make a good chain pattern. To make things easier, it would be probably good to eliminate tempo maps and use a some time base that would make calculations more simple. Any tempo change would be accomplished by duration of the events instead of changing the MIDI tempo. I need a way to do this without rerecording from one computer to another. Rerecording tends to skew the data, and not make an accurate master file from which to perforate. I realize the first thing to do is to probably put some sort of switch or sensor on the crankshaft of the perforator, but what do I do with the wires?
Would conversion to a Wayne Stahnke type file help this matter? Does his format handle tempo maps or would a conversion program to remove the tempo map do this? I do not have infinite amounts of money to spend on this hobby project but I do have a feeling some of this rather custom software may already exist.
David Wasson |
(Message sent Fri 22 Mar 1996, 08:25:39 GMT, from time zone GMT-0800.) |
|
|
Mechanical
Music Digest™ Archives
