LPD Pick Attack Modifier (PAM)

[thermionix]

One of the guys at the office plays whale songs with a distortion, a chorus and a volume pedal.

That alone is grounds for dismissal.

Unless he doesn't actually do it at the office.

Well, maybe still...

[Transmogrifox]

I'm thinking some kind of capacitive proximity sensor that you wear like a wrist-watch.  Measures voltage drop between wrist and ground.  When you get your fingers (with any pick or no pick) within a certain proximity it senses a change in amplitude of high-frequency oscillation on your skin, and uses that to process whether it is sensing a pick attack. 

Then the data can be encoded in the oscillator and picked up by a receiver onboard the guitar.  That makes the system reduce to only the inconvenience of wearing a wrist band or watch on your pick hand.

[R.G.]

Of course, your other hand is probably touching the strings, so you may not be getting all that much info on proximity to (grounded) strings. That's what keeps me from suggestion the Millennium Bypass, which is a sensitive detector of the difference between open circuits and some resistance to ground.

[Mark Hammer]

One of the guys at the office plays whale songs with a distortion, a chorus and a volume pedal.

Back when there used to be the National Lampoon Radio Hour, Harry Shearer did a sketch where he portrayed a rather effete snobbish A&R guy from a record label, who proclaimed that the next big thing in popular music would be "whaling songs".

[Groovenut]

I would worry about any remote transmitting device causing interferance because of the proximity to the pickups. But I do tend to worry.... 

[Transmogrifox]

Of course, your other hand is probably touching the strings, so you may not be getting all that much info on proximity to (grounded) strings. That's what keeps me from suggestion the Millennium Bypass, which is a sensitive detector of the difference between open circuits and some resistance to ground.

I was imagining something similar in concept, but the assumption is there is some fixed minimum impedance through your arms so the whole thing would depend on a change in impedance, and not an absolute impedance to ground.  Of course, the whole circuit path depends on electric fields in 3 dimensions since the arm band has no fixed reference to compare against.  It's just connecting an oscillator across maybe 3/4" strip of your skin and hoping that the parallel path around your shoes, your other arm, the air and into your guitar make a low enough impedance path that you can detect impedance changes large enough that you don't have to use ultra-precision circuitry to process it.  Would it work?  Don't know.  Just hypothetical at this point.

This approach allows for the possibility that other parts of your body are in contact with a path to ground.  Using an AC (RF) detector eliminates the constraint that the skin must make contact with something grounded.  If it just gets close and perturbs the field then this can hypothetically be processed and generate a trigger signal.

The possibility of the left hand triggering it by making and breaking contact with the strings could be considered a feature since it would work for left-handed tap-on playing and still respond to right-hand picking.  For that matter you might trigger it by touching your elbow to your rack, or a metal face on the front of the amp, tapping your foot on a stompbox, etc. It's just looking for that rate-of-change signature to trigger and it only needs some skin to rapidly approach something grounded.

Along with the idea of something held in the palm of your hand it may be an accelerometer or pressure sensor would be better suited to limiting the trigger specifically to a picking or plucking motion.  These require instrumenting the fingers or something directly in the hand.

Perhaps a piezo element that could be stuck to a pick could work.  Nothing else attached -- just a little 1/8" diameter sticker that you put on your pick.  The receiver "listens" for the signature pattern when the piezo element is excited by stress/vibration on the pick and emits some kind of EMI at a level high enough to detect from 6" away.

I know Leviton has wireless light switch system that provide battery-less operation based on this principle.  (It's freakin' expensive, but not as expensive as installation of wiring in some remodel scenarios)

Backing down on the complexity scale, a little "tap pad" near the pick guard would be acceptable to a lot of guitarists.  It would allow them to emulate the volume knob swell, but would require less finesse since the only constraint is to make contact with a little metal patch on note attack instead of hooking a pinkie around the volume knob.  I have a lot of respect for guitarists who do the volume knob trick well.

After considering all those ideas, what Groovenut did with a conductive pick and a wire doesn't seem all that bad.  Especially with an on-board unit.  The pick tucks into a little pouch or slot on the pick guard when not in use.  Second slot available for normal pick so guitarist can easily access either.  It's a bit of a novelty, but definitely one of the most straight-forward strategies for this type of thing.

[Groovenut]

...
After considering all those ideas, what Groovenut did with a conductive pick and a wire doesn't seem all that bad.  Especially with an on-board unit.  The pick tucks into a little pouch or slot on the pick guard when not in use.  Second slot available for normal pick so guitarist can easily access either.  It's a bit of a novelty, but definitely one of the most straight-forward strategies for this type of thing.

I've been trying to run use scenarios on this and the one you mentioned plays well. Having a removable conductor and bypass switch would make it more unobtrusive when the circuit isn't used. I like the idea of an alligator clip at the end of the conductor cable so the player can use their favorite metal pick and not be limited. This has it's own set of issues and may not end up working in the field. 

[DDD]

The close enough circuit was invented in early 1970-s (maybe in 1972 or so) and published in the "Radio" magazine in the USSR.
I didn't manage to find the original article, but one can find the schematic in the link below.
Maybe there are some interesting tricks, maybe not.

http://www.texnic.ru/shems/ungrad/07-1.htm

[Groovenut]

Thanks for the link and the schem pic. That design is much more complicated than the original Tel-Ray design, but it looks like it has more functions to it. I believe the original Tel-Ray was early 70's as well. My design is even more compact than the original Tel-Ray and much more than this one but it only does the pick attack modification. There are many ways to skin this cat. I arrived at my method after trying three different approaches. Each having a different set of benefits and liabilities. I was going for simple, small and completely natural sounding release envelope.

Sorry for the delay in posting the schem, but I've been swamped with taxes and other.... soon...

[Groovenut]

Here's a sneak peak at the proto board renders. They should be here Monday.

[Jdansti]

Very nice. There are some electrical conducting plastics out there that could be used for a pick if the mechanical properties were similar to nylon or nitrocellulose.

[Groovenut]

Very nice. There are some electrical conducting plastics out there that could be used for a pick if the mechanical properties were similar to nylon or nitrocellulose.

Or possibly a standard pick coated in said conductive plastics.

[Groovenut]

Here's the populated pcb. I hope to get a chance to test it tomorrow and verify the layout.

[Hatredman]

Lawrence, even with the "wire problem", I think you have a winner. Very cool solution indeed, and simplifies the detection/tracking.

But I'm a bit confused by your videos. You refer to the effect settings as "short release" and "long release" and so on. Am I getting something wrong, or did you mean short and long attack times?

[Groovenut]

Lawrence, even with the "wire problem", I think you have a winner. Very cool solution indeed, and simplifies the detection/tracking.

But I'm a bit confused by your videos. You refer to the effect settings as "short release" and "long release" and so on. Am I getting something wrong, or did you mean short and long attack times?

Thanks Henry!

The release time is how long it takes the signal clamp to release. When the pick touches the string the signal is muted and I consider this the attack. The attack time is about 1ms set by an RC filter. The release time is set by a pot variable from ~5ms to 1 second.

[Bitinwood]

hey groovenut, can you accept to share the schematic of this little effect, I would really appreciate to breadboard it