DIYAPASON-L Digest #603 - Friday, July 19, 2002
 
DIY Combination actions
  by "Drew Taylor" <drewt@loritsu.com>
Re: DIY Combination actions
  by "Larry Chace" <RLC1@etnainstruments.com>
Free stuff and combination actions
  by <tmbovard@earthlink.net>
 

(back) Subject: DIY Combination actions From: "Drew Taylor" <drewt@loritsu.com> Date: Fri, 19 Jul 2002 01:36:02 -0400   With all this talk about DIY MIDI projects, it reawakened my sleeping = experimenting dragon (when he wakes up, he is always hungry for my time and money).   Does anyone have any experience building solid state combination actions = from components? I would imagine it to be a simple operation consisting of = storing ons and offs in an EEPROM or the like. Then retrieving the data by jumping an = output (piston). Can one be somewhat easily built?   Input is much appreciated!     -Drew Taylor    
(back) Subject: Re: DIY Combination actions From: "Larry Chace" <RLC1@etnainstruments.com> Date: Fri, 19 Jul 2002 09:02:53 -0400   Drew Taylor <drewt@loritsu.com> asked about DIY MIDI, combination actions, and other such fun things. I've done several of these, starting with pipe organ MIDI stuff in 1988 with my friend and Cornell colleague Tom Dimock.   Yes, you can do it.   Yes, it can be simple.   However... (didn't you *know* that was coming?!?)   Neither of those things wants to remain simple, it seems, especially if = you are trying to make something that will be attractive for other folks, perhaps as a commercial venture. The combination action I build for use with the Z-tronics relay (though it can be used "stand-alone") has a = rather full set of bells and whistles, I believe, and it runs about 11,000 lines of 8051-ish assembler code.   The Z-tronics MIDI facility, supporting sequencer and sound module and connecting with 16 Z-tronics datastreams (keyboards, groups of stops, etc.), runs about 7,000 lines of 8051-ish assembler.   If you are willing to cut back on the features, you can do this sort of stuff on a smaller scale, as is evidenced by the several offerings of little boards that convert a MIDI channel into some number of magnet-driving outputs or that encode some keyboard contacts into MIDI. A small combination action can be built around some EEPROMs and a number of ICs. Remember, though, that typical electronic components are often unhappy with the electrical environment found within a pipe organ. Watch out for all those magnets and their inductive kick-backs! Remember that there will be a (large?) number of stops, pistons, and other things to control and sense. Contacts will bounce. Magnets will stick. The power will fail during the 5 milliseconds that it takes to write into the = EEPROM. Much of your design will deal with error conditions that "never" happen.   In the course of these projects over the past 14 years, I've had to learn about digital design, board layout, board fabrication, real-time processing, and a host of other topics that go beyond the many years of assembler programming (IBM 360, etc.) that I thought would help make these projects "simple"!   It can be fun, though, and I don't want to discourage you from trying. It is especially gratifying to play back a MIDI-recorded performance or to operate the pistons and know that it is *your* stuff (hardware and = sofware) that is producing the results. We do need to remember, from time to time, to actually work on the sound-producing portions of our projects!   Larry Chace        
(back) Subject: Free stuff and combination actions From: <tmbovard@earthlink.net> Date: Fri, 19 Jul 2002 08:12:02 -0500   Hi, all --   Some talk about both these topics recently, has got me thinking of stuff sitting in storage.   Anyone have a use for a Reisner electro-mechanical remote combination machine? Four large plywood boxes full of switches and magnets and mechanical miscellany. Seems to be enough for a 3m/p + couplers console. Near as I can tell, the machines are in like-new condition, possibly even unused -- units were part of a long-gone organ that burned while still new (but being mounted remotely, the c/a survived unscathed).   If anyone has the slightest interest, let me know!   Cheers,   Tim Bovard <tmbovard@earthlink.net>