Digital Pitch Changing instead of mechanics?

[J D Sauser]

In the late 90's I JOKINGLY suggested... well... LIED, that I had just come back from the biggest steel guitar show in Asia (yeah, Asia!) and they had introduced a fully digitally pitch changing PSG made of PLASTIC and which had pre-programmed setups like C6th, E9th and setups like BE's or Jimmy Day's.... as I was going, I added that they had effects built in, which the called the "Emmonsator" and "Dayizor" which made any average player SOUND like BE or Jimmy Day at a flick of a soft switch!

To my surprise the thread found mostly only opposition from the "If It Ain't Broke-Don't Fix It"-Crowd. Yet my PM-boxed got hit by several inquiries for WHERE to buy it. Ha! Sneaky Son'uffa'... you know what! Didn't want to ask about it on the thread.

So, recently somebody brought up the subject of Stepper Motor/Solenoid changing mechanisms which had been discussed several times over this Forum's life.
I highjacked the thread and suggested considering Digital Pitch Changing and thus ELIMINATE ALL mechanics.

Turns out, that during my sabbatical from steel guitar from about 2001 thru 2020 this subject HAD come up too.

I think that TODAY, this ought however to be re-envisioned.
Smart Phone Apps being used nowadays to control the settings on a variety of machines like amps etc via Bluetooth would allow for setup changes without having to incorporate the setup controls IN the guitar.
The system would evidently require individual pickup outs for each string.
But it would allow for "smart"-tuning-compensation for EACH individual change and ANY combination of various pedals and knee levers. ANY combination could become "in-tune" to any tuning approach (JI, ET or MeanTone).
Changes could reach different fine tuning depending of what other change they are combined with and even non-affected strings could be micro-tuned in tune with the rest.
Mechanics, Body-Drop-Detuning would be eliminated and the bridge could be shaped to the most ideal. Strings would last for as long they sound good (oxidation) as they would not suffer pull-release stress.
Pedals and levers would all have the same travel, resistance able to be set and stops rock solid.
The instrument would become very much lighter, likely cheaper once a certain production number is reached.
Only customization would be Sound Board material, Color, number of strings and necks.

In 2010 Ed Packard suggested:

Alternative PSG design shown here:

http://s75.photobucket.com/albums/i287/ ... ON%20DATA/

And here:
http://www.sierrasteels.com/pages/index2.html

Two pickups, both tapped, either or both can be switched to series or parl, phase aiding or opposing.

Strings are the same distance apart at both ends.

30” scale length tuned to C as in C9 = E9 is at 25” scale length.

No neck block. Glass drop in fret board = back light or?

Fret board will be LCD computer driven. Could use LCD picture frame now for info.

Integrated changer and keyless/gearless tuner, on players left. No string puller needed.

Tuning screws can be Allen wrench (shown…keeps tweakers away), or finger knobs (not shown).

String can be removed and put back on.

The rod (under the strings) on the string retainer end can be changed to a variety of materials to change the tone.

Two pedals are added to get the 13 series tuning structures.

Room under the drop in fret board for lots of circuitry.

It will be interesting to see what the maker of the worlds most elite PSGs will do with the concept (don’t ask who!).

NEXT GENERATION:

Circuit boards are being made for the pedal/lever activation sensors. Object is to eliminate the rods, bell cranks, cross shafts, and mechanical changer.

Individual string programmable pickups proven out…packaging being worked out.

Ed had a deep knowledge of the physics behind Sound & Music, and built a "Left Change PSG" and many other projects. He is sadly missed.


So, in order to not hijack other threads, I felt it was time to bring this back up to discussion.


Thanks!... J-D.

[b0b]

It's doable, but I think it would sound like crap.

[J D Sauser]

It's doable, but I think it would sound like crap.

I doubt that with today's sampling rates it would sound worse than most any of today's top singers.
Erm... wait a minute! Now, come to think of it, most any of today's top singer DO sound like crap! Ha!

Naw... At one point, over 20 years ago, I thought about a PSG with "neutral" stings. So even the basic tuning would be "Pitch Shift Generated" and from there on any change. Hence you could have had C6th or E6th of Bb6the of F#m7th (Speedy ) with the corresponding setups pre-programed at a flick of a switch... today at a "swipe" on your SmartPhone.
I think THAT would likely sound like crap. Still, a long touch screen could become a PSG... where you're bar only slides on the touch screen and everything else is just virtual. Still not my goal... but possible today.

But, I think that pitch shifting digitally up to 3 semi-tones in either direction each string would not loose the quality of each string's timbre with today's sampling rates.
Recorded musicians are today constantly pitch-adjusted digitally. I played acoustic Swing Rhythm along with an acoustic bass and solo guitarist and I complained about some tuning "wobble" (these old Selmer Maccaferri style guitars can be a challenge to intonation tune along the neck!) ... The recording engineer told me not to worry "I'll fix it, y'all will sound just perfect, just keep on playing like that!". And the "dry" playback sounded just not well to me, but a week later I got the final mix and Django Reinhardt would have envied us so bad and replaced his brother with me on the spot. Ha!

Electronic Pianos have come to a point where their only issue is the speakers to compete with acoustic instruments.
Chorus and other effects, which I admit have a history to sound like crap, have in high quality versions come a long way.
Remember the first CD's and cymbals, pianos and violins early one... and the Vinyl crowd professing it would never reach "HiFi"... ? That was the 80's.
You know I am a "Tone Nut" just most any steel guitarist... but tone would not worry me today.

In my opinion hurdles are two in my opinion:

1- Initial developing cost (and I would not even know where to go ask).
2- sourcing single pole pickups (not to use piezo or "light beam" sensors, which would seem the lesser problem.

Thanks!.. J-D.

[Donny Hinson]

People, it seems, are always looking for an easy way out. The big question here (in my mind, anyway) is that even if the instrument were perfected so that any imaginable chord could be played, and everything was corrected electronically so that it was in perfect tune, and the instrument were far more affordable, would that make us better players?

As I see it, the biggest limitation would still be there - that "thing" between the seat and the steel.

[Georg Sørtun]

In my opinion hurdles are two in my opinion:

1- Initial developing cost (and I would not even know where to go ask).
2- sourcing single pole pickups (not to use piezo or "light beam" sensors, which would seem the lesser problem.

Hall effect sensors would work as single-string PUs. Ed Packard's idea.

[Karlis Abolins]

J D, Have you looked at the NU pickups from CYCFI? https://www.cycfi.com/projects/nu-series/ There are a lot of options, including 6, 7, 8 string pickups as well as modular sets of 4 individual pieces. Their engineering staff is oriented to the software side as well.

Karlis

[J D Sauser]

Hall effect sensors would work as single-string PUs. Ed Packard's idea.

J D, Have you looked at the NU pickups from CYCFI? https://www.cycfi.com/projects/nu-series/ There are a lot of options, including 6, 7, 8 string pickups as well as modular sets of 4 individual pieces. Their engineering staff is oriented to the software side as well.

Karlis

Thanks!
I must admit, at electronics, I am a total idiot. So, it might seem that I am the wrong person to envision such a PSG. But I think, one is still allowed to have "ideas".

I long wished for single output pickups. Maurice Anderson once spelled it out differently: He wished that every string would have the same timbre... and only different pitches. I am not sure it would be desirable to that extent, but I always felt and still do feel that with the huge difference of mass (cross section) from the highest to our lowest string and the fact that we used the same scale length for all of them (unlike a piano or harp, where heavy strings also are longer and thus sound more "balanced" among each others), it would be "nice" to have the ability to PRE-EQ each string before the signals is being bundled into one and EQ'ed as a whole.

For an electronic PSG, single output pickups is an evident must.

I will contact CYFI, Karlis.

Thanks!... J-D.

[b0b]

I always felt and still do feel that with the huge difference of mass (cross section) from the highest to our lowest string and the fact that we used the same scale length for all of them (unlike a piano or harp, where heavy strings also are longer and thus sound more "balanced" among each others), it would be "nice" to have the ability to PRE-EQ each string before the signals is being bundled into one and EQ'ed as a whole.

2 rows of adjustable pole pieces, like the Telonics X-10 and X-12 pickups, come pretty close to accomplishing that.

[Andy DePaule]

I had this idea 44 years ago, but I'm a guy with electrical skills limited to changing light bulbs and batteries.

However, 44 years ago I went into a music shop in Oakland, California (Way over on the left coast) and tinkered around on a keyboard they had. It had what I'd describe as a 2" X 2" pad, kind of like the pad on the lap top computers today.
You hit a note on the keyboard and slid your finger up on the pad and it raised the note up to an octave higher and any spot or note in between. If you slid down it would lower the note down to the next octave and and spot or note in between.
Naturally I was thinking the same things as everyone else on this thread.

I mentioned this to Bob because he is way smarter than me about electronics and many other things too.
He thought it would not sound like a pedal steel so I dropped the idea and went back to changing light bulbs. I like bright rooms.

Anyway, my guess is that with all the amazing things now in the computer world and CNC and what not, I don't think it will be long before someone develops a light weight pedal steel with just these king of electronics. Hope someone does because my back hurts!

[Mike Preuss]

How about a MIDI pedal steel? Kind of like how a wind controller can turn a sax or flute into a MIDI synthesizer. I think that would be pretty cool.

[Robert B Murphy]

I think digital signal processing is fine for sonar and communications technology but I cordially detest voice auto-tune, especially when it's used as a gimmick on a recording. Applied to a steel, I shudder to think. It's the random variations that give a steel's sound the depth that I love. Big E's touch is unmistakeable. The way he moved the bar around and voiced phrases. That sassy attitude that came through. I just can't believe that sampling, digitizing, processing, and converting back to analog could ever do him justice. Something would be lost.

When the subject of pickups comes up, my experience with hall effect sensors, in effect a BJT where the base utilizes the magnetic field from a permanent magnet instead of an input current, makes me wonder where the magnetic field is coming from. I think those devices the cops and spies use that point a laser beam at a window and the acoustic vibrations inside the room vibrate the window and modulate the beam has more promise to sample a string without affecting the string. That brings up the problem of sampling the string at one little slice along it's length. When a string is plucked it vibrates in a precessing ellipse just like a free pendulum. But it's not just the fundamental doing this, it's every overtone in decreasing amplitude. Regular old bar magnet pickups do a pretty good job of capturing a swath of that complex vibration along the string. Sure each design gives us tonal coloring and other intangibles but it allows us to go on chasing the holy grail forever too.

Stepper motors in my experience come in 200 steps per revolution and 20,000 steps per revolution. They also come in two quadrature coils that require a bipolar drive through an H bridge, and center-tapped coils where only one half is used at a time. You can drive the coils with a sine/cosine signal and ease your way into the next detent but when you shut off the current you still have that 200, 20,000 resolution. That might be fine enough for any tuning but there's also how fast you can drive the motor too without skipping 4 detents at a time.

I listen to Bessie Smith and Ma Rainey sing and think about the power behind those voices that could fill an auditorium without a PA system. I love pursuing technological innovations but it is the dealing with the shortcomings of the real world that can make someone truly great.

By the way JD, your original post was truly sneaky. I think you are your own evil twin!

[J D Sauser]

I think digital signal processing is fine for sonar and communications technology but I cordially detest voice auto-tune, especially when it's used as a gimmick on a recording. Applied to a steel, I shudder....

I fully understand the feeling, but still we're in 2022 and DECADES into enjoying digitalized music from CD's, mp3 etc.
Just a $99.99 stomp-box pitch shifter (evidently shifting ALL) has a sound that is difficult to discern from the original timbre.
After all, most music we listen to, CD's, mp3 are ALL digitalized sounds. Singers, violin sections, cymbals (which were horror on early CD's), opera voices... we're far beyond "HiFi". Any of these formats can be pitch shifted without affecting timber... because... it's just NUMBERS, not just the notes, but over-tones, noises, "Touch And Tone" too.

Evidently, pitch shifting the tone that comes off a o.058 string (b0b who sells string on here suggests a o.o70 as a better fit! )
(bottom of a C6th) to the pitch of an E9th's 3rd string, will result to a questionable timbre.
But lets keep in mind that our steepest change is only 3-semi-tones (8th pedal C6th on the bottom string, again). All others are only one or 2 semi-tones. The change in expected timbre-change of the same string would seem negligible.

When we mechanically pitch-shift a string, we change timbre-characteristics too. Can you tell? I doubt it. The string's gauge may not anymore be the optimal for that tension and pitch ever so slightly.
A digital pitch-shifting system would keep the same timbre over all pitches of each string.

If BE or Jerry Byrd would come down from Heaven to play my or your guitar, they would sound like BE or JB... I am convinced the same would happen if they'd play a digitally pitch shifting guitar, as THEIR timbre would be shifted as part of the sound too. Keep in mind that the original sound would still come from a real string, and be picked up by a magnetic pickup. Only thereafter, the sound would be digitalized, just like most recordings we listen to TODAY of BE, JB et all on CD's, mp3's etc. The Touch And Tone never got lost on the records either. Only the fewest had the pleasure to hear most of them "LIVE".

I am NOT saying it's a must, but I suspect that Ed Packard's prediction will eventually go into history as prophetic, and that today the technology to do it right has already been around for a couple of years.

just some thoughts!... J-D.

[Robert B Murphy]

JD, how dare you refute my prejudices with reason and fact? The nerve!

I wasn't saying that stepper motor resolution and response time is a show-stopper, I'm just saying it's a design consideration that has to be weighed.

Digital resolution is another variable that will become less relevant as memory and processing speeds increase.

The funky quirkiness that is inherent in past recordings with limited technology has it's charms too. I'm old school, hisses, pops, string noise, and pitch imperfections don't bother me.

If you are talking about non-mechanical pitch shifting, what's the point of having a string at all if this is entirely in the digital world?

[b0b]

Evidently, pitch shifting the tone that comes off a o.058 string (bottom of a C6th) to the pitch of an E9th's 3rd string, will result to a questionable timbre.

But lets keep in mind that our steepest change is only 3-semi-tones (8th pedal C6th on the bottom string, again). All others are only one or 2 semi-tones. The change in expected timbre-change of the same string would seem negligible.

Actually, the 10th string is typically .070".

Are we saying that the strings would be tuned manually to the player's preferred temperament to provide the basic timbre, and then the pedals would be tuned electronically? In such a system, the pedal tension and throw would could be set mechanically for comfort, unhindered by the limitations of changers, rods, and bell cranks.

Hmmm. sounds familiar.
[center][/center]
I was thinking of a motorized changer when I made that, but it could also be applied to electronic pitch changing. https://b0b.com/wp/wp-content/uploads/2 ... roller.pdf

[Michael Brebes]

Unfortunately, I think most players of this gadget would sound like this, at best:

https://www.youtube.com/watch?v=AD0kGAzv5Ds

[J D Sauser]

Evidently, pitch shifting the tone that comes off a o.058 string (bottom of a C6th) to the pitch of an E9th's 3rd string, will result to a questionable timbre.

But lets keep in mind that our steepest change is only 3-semi-tones (8th pedal C6th on the bottom string, again). All others are only one or 2 semi-tones. The change in expected timbre-change of the same string would seem negligible.

Actually, the 10th string is typically .070".

Are we saying that the strings would be tuned manually to the player's preferred temperament to provide the basic timbre, and then the pedals would be tuned electronically? In such a system, the pedal tension and throw would could be set mechanically for comfort, unhindered by the limitations of changers, rods, and bell cranks.

Hmmm. sounds familiar.
[center][/center]
I was thinking of a motorized changer when I made that, but it could also be applied to electronic pitch changing. https://b0b.com/wp/wp-content/uploads/2 ... roller.pdf

o.068?!?... yes, b0b... I had to go back to read that I really had written o.o58 instead.

I think yes, b0b... your suggestion could be one of many ways.

I am convinced that the Digital Pitch Shifting technology has been around for a while for this. I am just no electronics guy. The technology would have to be adapted to the special application.

I think the most natural evolution would be to do it like a non-pedal steel guitar with individual pickups, each per string and a digital pitch shifting device controlling changes to pre-programmed final pitches (proportional control). For as little it takes to change from E9th to B6th or even force a C6th, even that could be programmed. Even different temperaments could be programmed. One could even imagine temperament to be adjusted by digital pitch shifting (shall we call it DPS?) depending on pedal-lever combinations.
On E9th, an JI E-to-F raise could be programmed shift to a different pitch when used alone vs. with the A-pedal or both A&B-pedals. Some levers could even be set to change their string assignation depending on pedals being depressed. Just to mention a few ideas!
I mean, the mechanical limitations would be out the window.

I also see more and more devices which don't even come with an interface anymore, because the builders assume that everybody has one of these stupid "smart"-phones... they, just let you have an app and use your phone's technical capabilities serve as the control-board to communicate with a device internal module via USB or even wireless via Bluetooth!
I just got some finger-print-"smart"-locks for an apartment. NO buttons, no interface. Download an app and connect via bluetooth and add finger prints, assign names and times, all on the app.

How easy is that? I see apps being put on-line for free by kids. Can't be that un-reachable to have a setup app created?

On the TONE end, which is the most important aspect besides playability of our instrument, we would benefit from a clean bridge, no more mechanical "clutter" and considerations as to stops and such (Push-Pull or All-Pull) etc., and no more varying tension on the body... actually, no more body-drop detuning, because the stops could be at the pedal rack.

HOW is it done? As I said... I DUNNO! If I only did, I would have given it a try a long time ago. I don't understand electronics much beyond too often hating it.
So, IF I'd build a guitar again, it would most likely be mechanical and with a left change-and-tune setup. THAT would give me the tone I want from a PSG. I know, because like Ed Packard, I have done it.

I merely brought up the subject because the stepper-motor thread brought up the subject and an old post of Ed Packard's I had missed during my Sabbatical from this all was brought to my attention where I found that Ed, whom I admired very much for all his research and prototyping had declared that he saw THIS... or SOMETHING like "This" as the naturally to expect evolution of our instrument. I happen to think he MIGHT be right.

I shall go and change that o.058 string now... before I hit the fretboard with my bar!... J-D.

[J D Sauser]

JD, how dare you refute my prejudices with reason and fact? The nerve!

Because I miss the old Forum Flame Thrower Parties, the Lynch Mobs.
Because I am a Dare Devil.... because I am... whooowhoo... "Controversial".

(we're joking, Folks!)

I wasn't saying that stepper motor resolution and response time is a show-stopper, I'm just saying it's a design consideration that has to be weighed.

Digital resolution is another variable that will become less relevant as memory and processing speeds increase.

The funky quirkiness that is inherent in past recordings with limited technology has it's charms too. I'm old school, hisses, pops, string noise, and pitch imperfections don't bother me.

If you are talking about non-mechanical pitch shifting, what's the point of having a string at all if this is entirely in the digital world?

Just like digital photography's resolution, digital sampling or bit-rates or what ever the correct term is, has long shot beyond what we can possibly discern with our 20'th Century ears. Evidently, quality is still a mater of choice.

When pedals became a practical add on to steel guitars, Jerry Byrd is documented to have declared that "now, anybody could play steel guitar".
And the instrument got taken to the next level and then the next by newer generations. Still, I'd agree with whom ever wants to join in claiming that JB has yet to be out played in musicality on any kind of steel guitar.
But, so why have pedals and knee levers, or was it entirely a "mechanical" world?

Sure, you can take it further and further. Replace strings with "active"-sensor bars, or what ever they could be called. I thought about that too when I saw the first "tablets" and wondered "how'bout no strings, and just sliding some bull-shit-material "bar" over a touch-screen?". Yeah, how'bout that?
Pre-recorded "sampled" sounds, like digital pianos!
Could even get that elusive "Emmons"-tone for an extra $9.99/month too (and it' be only $99.99 if you pay one year NOW). Think 'bout it... it's a great offer!

I don't think I would want THAT, but I can see how that could maybe be made to work.
And I'd gladly discuss it.
Is THAT the future? I don't know, but I wouldn't bet my money on something not eventually ending up being the future up to WW-III (after THAT, Einstein suggested the few that might survive would fight out WW-IV with sticks and rocks again anyways... steels may even be missing coat-hangers by then).


time to sleep! Thanks!... J-D.

[Robert B Murphy]

Since the only two variables to change a string's pitch are length and tension, both mechanical processes, are we talking about using electrical means to make them happen, or digitizing the signal and making the change in the digital realm?

[J D Sauser]

Since the only two variables to change a string's pitch are length and tension, both mechanical processes, are we talking about using electrical means to make them happen, or digitizing the signal and making the change in the digital realm?

Just the "changing".... replacing the changer.
In other words, imagine a perfectly "natural" non-pedal steel with more changing capabilities than, let's say an Excel Superb.

The tone originating from an original set of strings, bar, picks and pickup... just that the pickup would be wired individually for each string to capture the signal from each string separately (something I long wished for, for better EQ'ing). Only from there on, is the signal digitally (instead of mechanically) processed/pitch-altered WHEN ever there is a change being engaged.

"Only" replaces the changer and with that most of the undercarriage.
Pedals would only controlling "readers"... just like the gas pedal on most of today's fuel injected cars... there is no more physical linkage on most.
Very precise proportional "reader" systems exist since the 80's... I've seen them on RC controls for RC-model planes back then... 4 Decades ago.

That's the most "natural" first step. Evidently, one can fantasize further from there.

... J-D.

[Robert B Murphy]

So we are talking about precise linear actuators with a maximum force of around 40 lbs then? This calls for a little research to see what's available.

[Georg Sørtun]

So we are talking about precise linear actuators with a maximum force of around 40 lbs then?

No. Just control the output frequency(ies) of each string electronically, same as a good old vocoder does.

The string provides the tonal spectrum, overtones and all.
A frequency generator provides the controlling frequency. Now add 10 - 12 (or whatever the number of strings happens to be) such "vocoders", and start playing

[b0b]

Isolating the sound of each string could be a real problem, though. If any sound from an adjacent string is present in the signal, the digital pitch-bending pedal would affect that note as well, creating a very non-musical artifact. You might need wider string spacing. (That could actually be a good thing, as the string spacing is often seen as an impediment among guitarists considering the instrument.)

And if you solved the isolation problem, would the overall tone of the guitar suffer? It might be impossible to get "that sound" from single-string pickups.

[Robert B Murphy]

Yeah, and we're not just talking about electrical bleed over here, the strings are mechanically coupled through the chassis too.

I have trouble with the concept of digital pitch shifting because we are talking about sampling a complex wave and while retaining the same start, stop, and decay characteristics, increasing or decreasing the number of cycles within a given period. In other words, the processor is adding or cutting out little snippets of the signal at the zero crossing points.

I just don't see the difference between that and using a totally stored, sampled signal to start with. I'm with Michael on this one. I think it would wind up sounding like that goofy Al Cooper facsimile of a steel.

[Bob Hoffnar]

Jeff Snyder who works at Princeton University in design, fabrication with the computer music dept has built one that is surprisingly good. He has prototypes for a number of amazing instruments. He plays pedalsteel and is definitely on to something. There are a couple of other inventors scattered around in the academic world that are quite busy working on similar projects with the pedalsteel. .

[Georg Sørtun]

Good. List up all the potential problems. Then, maybe, someone will come up with working solutions for the entire thingie.