Hexaphonic Pickup Project

[cycfi]

What about starting a new thread "Ideas of processing hexaphonic signals"?

Why, what's wrong with this thread? I'd love to hear your ideas.

[Jazz P Bass]

He did.
http://music-electronics-forum.com/t33999/

[Yury Oransky]

Hi, I am sorry, I thought, this thread is dedicated especially to Hex PICKUP making. Frankly I messed up with system of this forum.

OK, some thoughts, while I started a new thread.

I made my first hex in 1985. That was Soviet Union and that was complete lack of information, especially about electric guitar gears!!! but I could not interested in HEX if not that lack of info... Here, we had no idea about rock guitarists using amps and cabs! We thought that is all the DI setup with overdrives made of tubes and transistors. So I made a lot of it -tens. and still have tape recordings with the sound of it (will upload some later) I am not sure if it is so easy to guess there I used transistor devices only btw.
As about that hex pickup I made with a hope to get to better results... I solded hexaoverdrive and made some recordings, showed it to my cousin who was (and still is) a big fan of best rock guitar sounds and... we had a great disappointment. the worst thing was that hexa sound had something very similar to accordion! And that was something very difficult to get rid of it. My next step was to filter each separate channel with low passes adjusted some way neatly of hight order. and adding the pitch multiply chips after filtering. That result was something unexpectable - did not get us any closer to the great rock sounds but opened another dimension of strange sounds like very close to Adrian Belew guitar...That was not the best I dreamed about because my subjectively highest pattern of electric guitars sound was Alex Lifeson's from 1982 to 1991 (especially Power windows, Hold your Fire , Presto) These sounds for me are like a precious, so skilfully produced diamonds that are soooo difficult to approach to. That sound palettes seemed like consisted of several different layers mixed with a high ingenuity. You just can imagine that situation was overhere when a total disinformation about how music was making those times. So, all I could guess that was a thoughts about hexaphonic processing of Lifeson guitars! After many years, when we got the first info about how those sounds were made (even no matter how! but not with hex!!!) I could not forget about hex evermore. It's worth to say, in the era of my unawareness, best guitar sounds like Lifeson, Travor Rabin, later Jimmy Page, Doug Boyl, Ian Crichton... - all inspired me to have a future deal with hex pickup forever. After playing some with hexa overdrive I started to made VCF units for 6 channels on late 80-th (there were our chips for doing this) but I finished just 3 units which took 3 pieces of A5 size and eat lot of power! but tracked good enough with the help of additional filters controlled by a V back from a hand-programmed RAM chips. Now all can be neatly made in Reaktor, but I tend not to use generators for guitar play.

And btw, I do not understand the reason (maybe missed something here) Why not to use standard Roland hex for your neads? Good design and great characteristics. Why to make your own?

[cycfi]

And btw, I do not understand the reason (maybe missed something here) Why not to use standard Roland hex for your neads? Good design and great characteristics. Why to make your own?

I'll reply to your other posts, but for now let me reply to this one.

The main reason is to make a better pickup not constrained to the limitations of the Roland's physical design. I prefer a modular design that is self contained (no external hardware -- the preamps should be contained in the pickup itself). A modular design also opens up other possibilities such as being able to have 7 and 8 string or even 4 string, 5 string and 6 string bass versions, perhaps even an electric harp. So the project has now evolved into 2 parts: The Neo1 and Neo2 pickups are small single and double pickups; the Six-pack contains 3 Neo2s and has dimensions that will fit the ubiquitous Fender strat single coil. It is now very possible to have other packaging options thanks to the modular design.

[Yury Oransky]

Ah, now I see, it's going to be a big production.
Well, then I may imagine myself a a buyer of hex pickup.
As for me, I'd prefer yours over Roland or Yamaha if it will cost not more then 200 usd! But if it used NOT battery for power supply. That will be ok if that active pickup come with its little external box (connected to guitar with multi-wire cable of course) powered from the 220v (I hate to have batteries and think about recharging or changing it).
Preamps right in the pickup would be the best solution, especially if they work as differential amplifiers for double coils.
You know, Roland and Yamaha work as 6 passive humbackers and OP work there just as buffers-amplifiers.
And I faced a problem with my 13pin cable of 7 meters length when I use a very big gaining of the signal after the long cable ( for compression, distortion, etc) and then I have a problem with a hum noise inducted in the long cable wires. The only solution in this case is to invert 3 signals of 6 (if I use the simple mix of hex signals it works ok, but if I process each srting signal differently it helps about 50% only. So better solution could be the cable with 6 pairs of wires going from the differential schemes, and differential receivers placed in the box after a long cable

[Mike Sulzer]

I have done my hex fuzz electronics in the guitar, but now I have come to prefer purely passive instruments and will try that next. The little individual string coils are low in impedance, and should be able to run farther than a normal high impedance guitar pickup without excessive noise and hum. And no change in frequency response due to the cable. I resonate such coils to achieve a sound in the range desired. Typical capacitor values are .022 micro f. Thus the impedance is very low at high frequencies. The Q is low, of course, and so the impedance never gets very high over the whole range.

My fuzz electronics is controlled by a dc level, and so I will run that back into the guitar, along with volume and tone control lines, and so it would seem as if the electronics is inside to the guitarist (except for the non standard cable). The main limitation of complicated electronics in guitars is that very low noise circuitry still requires more current than I like to supply with a battery.

You know, Roland and Yamaha work as 6 passive humbackers and OP work there just as buffers-amplifiers.
And I faced a problem with my 13pin cable of 7 meters length when I use a very big gaining of the signal after the long cable ( for compression, distortion, etc) and then I have a problem with a hum noise inducted in the long cable wires. The only solution in this case is to invert 3 signals of 6 (if I use the simple mix of hex signals it works ok, but if I process each srting signal differently it helps about 50% only. So better solution could be the cable with 6 pairs of wires going from the differential schemes, and differential receivers placed in the box after a long cable

[Yury Oransky]

I have done my hex fuzz electronics in the guitar, but now I have come to prefer purely passive instruments and will try that next. The little individual string coils are low in impedance, and should be able to run farther than a normal high impedance guitar pickup without excessive noise and hum. And no change in frequency response due to the cable. I resonate such coils to achieve a sound in the range desired. Typical capacitor values are .022 micro f. Thus the impedance is very low at high frequencies. The Q is low, of course, and so the impedance never gets very high over the whole range.

My fuzz electronics is controlled by a dc level, and so I will run that back into the guitar, along with volume and tone control lines, and so it would seem as if the electronics is inside to the guitarist (except for the non standard cable). The main limitation of complicated electronics in guitars is that very low noise circuitry still requires more current than I like to supply with a battery.

This setup seems to be very interesting and maybe... winning. I love this idea. If I could start to make my own hex again - I would use low-resistant coils too.

So if I understand right, your next guitar will be without hexfuzz inside?
Have you tried a test with that chain of low impedance: your custom coil > cable > preamp > the result for noise/hum level??

PS. my first hexafuzz I made (in1985) was all placed inside guitar too. But later I did not like this because of lack of freedom in farther soundprocessing

[Mike Sulzer]

I am using #42 wire, so it is not really low, but 1000 turns a round a single core does not add up to too much. I have not tried these without a preamp in some time and did not check for electrostatic pickup back then. However, the current project that I am finishing up is a strat type guitar (no, I did not make the neck and body, bought them, but I am doing the electronics) with three "medium" impedance stacked humbuckers. The main and canceling coils have just 1000 turns (or 1500 for the bridge pickup) so this is very similar levels to the small hex coils. This is a very quiet guitar. The electrostatic pickup is small because of the impedance, and the magnetic pickup is very small because a stacked pickup can achieve very good cancellation. (Each core is two short pieces of ferrite rod with a neo disk magnet in between.)

The signal level is low, of course, but the amp noise (12AX7, 5K snuffer resistor) really only becomes noticeable at heavy metal gain. For fun, I will build a 1 nanoV/rtHz preamp with a low noise FET and an op amp, which needs to have noise about 5 nanoV/rtHz to not degrade the circuit significantly. There is a very simple good published circuit that requires a small change for guitar use. This will restore the expected gain, and eliminate amplifier noise.

So I am thinking that if I do the hex coil fuzz electronics external to the guitar, then there should be no significant hum, at least with some effort. I will modify the electronics, linked to somewhere in this discussion, to use the low noise FETs replacing the input resistors in the op amps, as in the low noise circuit just referred to.

This setup seems to be very interesting and maybe... winning. I love this idea. If I could start to make my own hex again - I would use low-resistant coils too.

So if I understand right, your next guitar will be without hexfuzz inside?
Have you tried a test with that chain of low impedance: your custom coil > cable > preamp > the result for noise/hum level??

PS. my first hexafuzz I made (in1985) was all placed inside guitar too. But later I did not like this because of lack of freedom in farther soundprocessing

[cycfi]

Waveshaping

A lot of people here and in the other forums expressed interest on hex-fuzz. That's as I expected. Hex-fuzz is indeed pretty neat. For me however, my goal was, and still is: polyphonic sustain using feedback drivers for each string. What I want to achieve is have full control over dynamics and sustain without having to resort to overdrive and distortion. I want the natural harmonics of the strings to sustain indefinitely and be able to control the attack, decay, sustain level for each string. I love violin-like slow attacks, for example. I also would like to be able to inject extra harmonics and sound textures through the feedback drivers --acoustic synthesis.

I love fuzz/overdrive/distortion, but I want to explore and go beyond preamps, clippers and overdrives. What I am more interested with is what's more generically termed as Waveshaping where you pass a signal through a non-linear transfer function stored in a table or computed. This implies of course that my general direction is towards DSP. I want to be able to control the transfer curve dynamically, have it controlled by an envelope, for example. With a controlled sustain through feedback in place, the main purpose of Waveshaping is to alter the spectrum of the signal rather than providing sustain.

Of course all these is easy said than done. Now, let me go back to work :-)...

I made my first hex in 1985. That was Soviet Union and that was complete lack of information, especially about electric guitar gears!!! but I could not interested in HEX if not that lack of info... Here, we had no idea about rock guitarists using amps and cabs! We thought that is all the DI setup with overdrives made of tubes and transistors. So I made a lot of it -tens. and still have tape recordings with the sound of it (will upload some later) I am not sure if it is so easy to guess there I used transistor devices only btw.
As about that hex pickup I made with a hope to get to better results... I solded hexaoverdrive and made some recordings, showed it to my cousin who was (and still is) a big fan of best rock guitar sounds and... we had a great disappointment. the worst thing was that hexa sound had something very similar to accordion! And that was something very difficult to get rid of it. My next step was to filter each separate channel with low passes adjusted some way neatly of hight order. and adding the pitch multiply chips after filtering. That result was something unexpectable - did not get us any closer to the great rock sounds but opened another dimension of strange sounds like very close to Adrian Belew guitar...That was not the best I dreamed about because my subjectively highest pattern of electric guitars sound was Alex Lifeson's from 1982 to 1991 (especially Power windows, Hold your Fire , Presto) These sounds for me are like a precious, so skilfully produced diamonds that are soooo difficult to approach to. That sound palettes seemed like consisted of several different layers mixed with a high ingenuity. You just can imagine that situation was overhere when a total disinformation about how music was making those times. So, all I could guess that was a thoughts about hexaphonic processing of Lifeson guitars! After many years, when we got the first info about how those sounds were made (even no matter how! but not with hex!!!) I could not forget about hex evermore. It's worth to say, in the era of my unawareness, best guitar sounds like Lifeson, Travor Rabin, later Jimmy Page, Doug Boyl, Ian Crichton... - all inspired me to have a future deal with hex pickup forever. After playing some with hexa overdrive I started to made VCF units for 6 channels on late 80-th (there were our chips for doing this) but I finished just 3 units which took 3 pieces of A5 size and eat lot of power! but tracked good enough with the help of additional filters controlled by a V back from a hand-programmed RAM chips. Now all can be neatly made in Reaktor, but I tend not to use generators for guitar play.

[cycfi]

Ah, now I see, it's going to be a big production.
Well, then I may imagine myself a a buyer of hex pickup.
As for me, I'd prefer yours over Roland or Yamaha if it will cost not more then 200 usd! But if it used NOT battery for power supply. That will be ok if that active pickup come with its little external box (connected to guitar with multi-wire cable of course) powered from the 220v (I hate to have batteries and think about recharging or changing it).
Preamps right in the pickup would be the best solution, especially if they work as differential amplifiers for double coils.
You know, Roland and Yamaha work as 6 passive humbackers and OP work there just as buffers-amplifiers.
And I faced a problem with my 13pin cable of 7 meters length when I use a very big gaining of the signal after the long cable ( for compression, distortion, etc) and then I have a problem with a hum noise inducted in the long cable wires. The only solution in this case is to invert 3 signals of 6 (if I use the simple mix of hex signals it works ok, but if I process each srting signal differently it helps about 50% only. So better solution could be the cable with 6 pairs of wires going from the differential schemes, and differential receivers placed in the box after a long cable

The quietest guitar I have here has an EMG 85 and 81 PUs. Let me tell you that I've successfully surpassed the SNR of these active humbuckers even at heavy-metal gains. The hum emanating from the EMG is noticeable at heavy-metal gains. And it might surprise David Schwab that the EMG hiss is also noticeable (tolerable and a lot less than the hum actually, but still noticeable).

I'll post some sound clips comparing the noise floors of 1) A Strat 2) a Les Paul 3) Charvel with EMGs and the hex-PU. The hex-PU we have has the lowest noise floor.

[cycfi]

There is a very simple good published circuit that requires a small change for guitar use. This will restore the expected gain, and eliminate amplifier noise.

Any links, Mike? Appreciated!

[Rick Turner]

After several decades of loading everything imaginable into instruments in the way of active circuitry, I now think that all you should put in the guitar or bass itself is the most basic buffering (with gain if need be), balance or volume controls for the pickups, maybe passive tone (treble cut), and then EVERYTHING else in the way of signal processing should be done externally as a buffer against obsolescence. Maybe put some assignable pots and switches on-board that can control external processing...fuzz, EQ, oddities, whatever...but keep that stuff the hell out of the instrument where it will prove obsolete in twenty or ten or fewer years.

Example from me...David Crosby's Gibson/Alembic 12 string guitar for which I made medium impedance pickups, and for which Ron Wickersham made custom buffer/preamps in 1970. That, by the way, was arguably the first guitar with LED neck side dot markers...which still work after 43 years use. And the guitar sounds fantastic still. There have not been huge improvements in discrete solid state audio in the last four decades...at least not where this one is concerned. But if we'd built in any effects, they'd be history by now.

Keep the onboard electronics simple. You can control off-board electronics from the guitar or footpedals. Build now for 2053. That's the time span we built for in 1970. Onboard fuzz and shit won't cut it.

[defaced]

The quietest guitar I have here has an EMG 85 and 81 PUs. Let me tell you that I've successfully surpassed the SNR of these active humbuckers even at heavy-metal gains. The hum emanating from the EMG is noticeable at heavy-metal gains. And it might surprise David Schwab that the EMG hiss is also noticeable (tolerable and a lot less than the hum actually, but still noticeable).

I'll post some sound clips comparing the noise floors of 1) A Strat 2) a Les Paul 3) Charvel with EMGs and the hex-PU. The hex-PU we have has the lowest noise floor.

Outside of the casual nerdy interest I had with this topic originally, you now have my full attention from a musician's standpoint. Keep up the good work. I look forward to any further progress and hopefully building these once the design is finalized.

[Mike Sulzer]

Any links, Mike? Appreciated!

Sure, but I hope I have not mislead you into thinking this is for inside the guitar; my intent is to keep the guitar completely passive (25K or lower vol. pot.); this circuit is not low current.

The link: http://arxiv.org/pdf/0811.2866.pdf

I realize now that this is from the Cornell library. It appears to be public access, but if you have trouble, I can send you a copy.

The circuit is briefly discussed here as well: http://www.gellerlabs.com/JCan 20References.htm. Apparently there is a very similar circuit, which I have not yet seen, that uses ac rather than dc coupling of the FET to the op amp inverting input. There is a statement that the dc coupling is used to stabilize the drain current. It appears to me that only the drain source voltage is stabilized. By the way, this is accomplished with a battery (from which essentially no current is drawn). A reference is provided showing that batteries are very quiet. Also this analysis questions where the operating point is set. I think it is very clear that the battery voltage puts the drain source voltage on the flat part of the curves, and the FET goes to IDSS (VGS = 0). This is a reasonable way to operate for very small signals, usually why one needs a very low noise amplifier. However, for guitars, even medium impedance pickups, the signal is not that small, but the high gain is needed for non-linearity in the guitar amp. The FET preamp therefore probably should be operated with better linearity or you will lose the possibility of a clean signal. Looking at typical curves (Field Effect Transistors), one would choose a current of about 80% of IDSS. (I have not tried this yet, but soon, since I have received FETs from Linear Systems.) A source resistor bypassed by a tantulum capacitor could be used to achieve this. Tantalum Cs are noisy at sub audio, but the noise falls off very fast with increasing frequency, and there should be no problem at guitar frequencies.
(downloads.hindawi.com/journals/apec/1987/010769.pdf
TANTALUM CAPACITORS. D.T. SMITH. University of Oxford, Clarendon Laboratory, Parks Road, Oxford OX1 3PU. Noise has been measured in a number of ...)

[Mike Sulzer]

I now think that all you should put in the guitar or bass itself is the most basic buffering (with gain if need be), balance or volume controls for the pickups, maybe passive tone (treble cut), and then EVERYTHING else in the way of signal processing should be done externally as a buffer against obsolescenceit.

Yes, but I would take it one step farther: The buffering is unnecessary if you use medium impedance pickups. The vol. control effect on the tone (from the cable) is gone, and the first thing you go to is a low noise preamp. The medium impedance pickups also allow flexibility in setting the resonance, and so you can replace the treble cut pot with a rotary switch allowing the resonance to be set over the full range (both above and below the "normal" position).