I play a Yamaha WX11 MIDI wind controller, I use it with a Yamaha VL1-m synth. The VL1-m wants as many continuous controller channels as it can get, to control various parameters of the physical model. The WX11 can only transmit two channels - breath pressure and lip pressure. More channels are needed.
Art Whitfield has experimented with adding more controllers using Force Sensing Resistors (FSRs) from Interlink Electronics. FSRs are force to resistance transducers. The harder you push on them, the lower their resistance drops. Art has put FSRs into a neckstrap and into the toe of his shoe, among other places.
This page describes an experiment of my own, where I've embedded an FSR into a neckstrap hook. This is an experiment and a work in progress. I think I'm onto something, but I don't know what it is yet. (-:
The FSR neckstrap hook snaps onto the back of the WX11. A neckstrap attaches to the hook. When I push the WX11 away from my body (primarily with my right thumb), the FSR varies some parameter in the VL1-m, affecting the synth's sound.
Here are some pictures of the hook in JPEG format.
The hook's mechanical construction is fairly simple. It's two pieces of metal, an FSR, and some electrician's tape.
The first piece of metal, the eyelet, is made from a small hasp. I trimmed the base down using a hacksaw, and cut through the loop at one end.
The second piece is the clip. I cut it from a sheet of 0.025" aluminum and bent it into shape. It clips onto the back edge of the WX11 just below the built-in hook, and it holds the eyelet in place.
The FSR is sandwiched between the clip and the base of the eyelet, and the whole thing is held together with electrician's tape. When the neckstrap pulls on the eyelet, the FSR is squeezed.
I am using a 1" round FSR. But a 0.25" round FSR would fit better. I started out using a 0.25" FSR, but I bent its leads too many times and it broke. I didn't have another 0.25" FSR, so I'm using the larger one.
Here are some JPEG pictures of the pieces and my pink fingers.
Currently, the FSR, in series with a 50K resistor, is plugged into the BC input on the VL1-m. The VL1-m maps the FSR's resistance into MIDI CC 2 (breath controller) messages. The CC 2 value varies from 0 (no force) to about 123 (high force).
The WX11 is routed through a Silicon Graphics Indy which is running a simple mapping program that converts the WX11's CC 2 messages to CC 9, passing other MIDI through unaffected. (CC 9 is not assigned.)
I have remapped the controllers on several VL1-m patches to use CC 9 to for breath control, and CC 2 for whatever I want to vary using the neckstrap.
Each of these patches has been modified to use CC 9 (the remapped WX11 breath pressure) to control breath related inputs and use CC 2 to control something else.
BB+ Tenor3 is based on Bill Busch's Sax:Tenor3 patch. I control Scream and Growl with the neckstrap. It is hard to make
a convincing transition from not growling to growling, but I think it could
sound good with more practice.
ThickTenor is based on the Tenor Sax demo voice in the VL1-m ROM. I control embouchure with the neckstrap. This
patch does not work very well.
AverysHarp is voice G12 in VL1LIB.ALL, distributed by Yamaha. It's a harmonica style
voice. I control the Dynamic Filter with the neckstrap to mimic moving the
hands away from the harmonica for the "wah" effect. It works very well.
Fuzzball is a distorted electric guitar patch created by Randy Hudson. I use the
neckstrap to control Scream, Dynamic Filter, and Harmonic Enhancer. It works
fairly well, though it is not very controllable. But who expects subtlety
on a fuzzed-out guitar?
IgneouSax is my own combination of Igneous (voice E7 on VL1_VER2.ALL on the VL factory disk) and PlastiSax (voice E12 on VL1_VER2.ALL). I use the neckstrap to control Igneous's Dynamic Filter, Damping, and Harmonic Enhancer (yes, I turned on the
H.E.). The neckstrap does not affect the PlastiSax's sound at all. This is a fairly complex voice, and I get a huge range
of sounds from various combinations of attack, neckstrap, and lip pressure.
Sometimes it sounds vaguely like a sax, sometimes vaguely like an electric
guitar, and sometimes vaguely like a Hammond organ. But it's too weird for
real use.
The patches I started with either are included with a Version 2 VL1-m or are available from Ken Barry's patch archive.
It works. Sort of. I don't think I could use the hook to control a high precision parameter like pitch bend - it's not that accurate. But the hook is good for translating coarse gestures into coarse parameter changes. It works well for a harmonica's wah or for an electric guitar's distortion.
The WX11 moves around more than I expected it to. I'm currently using a cheap, thin neckstrap (about 3/4 wide), and it sinks into my neck when I push. If I try to play a fast wah, the end of the WX11 shakes a couple of inches. I plan to try a wider neckstrap.
The one FSR neckstrap patch that I consider an unqualified success is AverysHarp. I have played it long enough that I've forgotten that I'm testing the neckstrap, and have just been enjoying playing the harp.
I think I can make the clip a little smaller and lower profile. I want to get another 0.25" FSR to replace the one I broke.
I need to package the wiring better. I noticed that one of the conductors in the WX11's MIDI cable is not used, so maybe I can run a wire inside the WX and tap into the unused conductor.
I also need to get a separate resistance-to-MIDI converter so that the hook will use a different controller than Breath. Art recommends an Anatek Pocket Mapper. Yamaha has recently announced the MFC10 MIDI Foot Controller, which has a WX style input as well as three voltage to MIDI converters.
I have some patch ideas too. I'd like to try to simulate moving a brass instrument's cup mute in and out. I'd like to try some more embouchure experiments with a sax patch, or maybe combine embouchure with growl.
Bob Miller, kbob@jogger-egg.com, January 14, 1997