Audio mag sensor switching circuit brainstorming circuit assistance

[4tune8]

Hi there, first post....so, hi there, hope this is the right section for this thread...

I am looking for people to contribute ideas or develop this project in the circuitry department...or at least give clues as to approaches one might take for this kind of project/design...

The idea is for sensor on the low 3 strings of a guitar. I have developed such sensors which are low impedance that can sense individual strings. The intention is to send them to an output and drop the signal by an octave using a pitch shifter or similar effect. A preamp to boost the signal will also be required.

The circuit help is required is about 'logic' for want of a better word. The end result is that I want to only have the lowest note of a chord sent to it's own output.

So if we label the sensors E6, A5, D4 for the low three strings of the guitar, low to high. If E6 is sounding I want this to turn off A5 & D4. If E6 is not sounding, A5 to out and D4 to be turned off. If lower strings are not sounding, I imagine there is no harm in leaving them 'active'.

So, any ideas to achieve this kind of thing. Needs to be low noise, simple and compact and low power requirements. My intitial thoughts are some kind of electronic switch, perhaps a transistor that will sense the incoming signal from say, E6 and turn off the signal to a preamp from A5 and D4...that kind of thing.

I have a bit of electronic experience and can solder basic circuits and such, but more of a concept guy and not good at the details nor know quite the technology that might be appropriate and keep it all 'happy' for such a task. Power is likely to be from a 9v battery. Quiet and low noise characteristics will be essential as for any audio circuit. Small and simple and low power drain will also be necessary to be practical.

An example of the kind of 'sound' one might get or to clarify the 'intention'. Guitarist 'steven king' uses a magnetic pickup on the low two strings simulaneously and the sound created can be found here... http://www.guitarbystevenking.net/King_CD_sound_clips.html

I can do this kind of thing with a pair of active 'sensors' but there are some compromises in this kind of approach. The two low strings are always on and so low as to sound musically 'muddy'. Many octave dividers are not ahppy and will glitch with more than one signal at a time, the octave drop is always on these strings so one needs to ensure that the tunes arrangments do not hit both as these low notes tend to sound 'muddy' and perhaps glitch things as well.It relies on finding teh bass notes only on the two low strings which can create large stretches for the player to reach and so another string would be an advantage for the player.

In case it is not obvious, the guitar will have it's own output/pickup system for all strings like a conventional instrument an that these low notes will double the normal octave sound of the instrument. The purpose then of this idea is to send out a separate output of the lowest string being played alone to be processed separate from the rest of teh guitars output.

So, any help, ideas or direction to aid achieving this kind of result would be appreciated....thanks in advance...

 

[Raven Luni]

Interesting. Hadnt heard of this guy before. So basically what you're looking for is a fancy bass boost. As a guitarist myself I was actually looking for a similar sound but was thinking more along the lines of a custom neck with a few bass strings (alot of my stuff uses different tunings and 2 or more simultaneous melodies). I havent really experimented with audio circuits much since amplification is such a huge field with good sound being a fine art (but I still have a cool idea for a quasi-digital ringmod I want to try out one of these days).

Anyway, from a top level view, there will always be some crossover and / or accidental hitting of strings so its probably best to go by which of the 3 is sounding the loudest. Then again what if you want 2 simultaneously? In order to measure loudness, you will have to rectify and smooth the signal. Pass it to a comparitor set for your trigger level but give it a decent hysteresis gap since it will never be 100% smooth. From here the logic for determining the lowest should be simple enough:
( D4 AND NOT (A5 OR E6)) XOR (A5 AND NOT (E6)) XOR (E6) )

Use the logic output to switch the audio signal from each pickup - which brings up another issue: the sudden turning on and off will sound ugly so you can smooth this either with capacitors in the switching circuit for a simple fade or have some kind of crossover area (which makes things far more complicated).

As for the octave division, the only way I know of doing this effectively is in software. Its not like a bat detector where you can throw away either the harmonic relationship (hetrodyning) or the amplitude information (frequency division), the audio quality has to be preserved.

Anyway - hope that helped at least a bit. You say you developed your own pickups. I'd be interested in hearing more about this. I'm always looking for stuff like that (currently experimenting with surface transducers). Who knows maybe I'll buy some from you

 

[4tune8]

Hello there, thanks for the quick reply...

Well, I wouldn't call it a 'bass boost' which could be misleading, this is pitch, not tone of course. S King calls his 'bass expansion'. He uses an analogue octave divider stompbox effect to get the octave down. You can buy a digital pitch shifter from behringer for instance for $60, so there is no desire at this stage to design an octave divider to build into the device...a separate output would allow one to use any effect one might choose on this signal.

Here is a pic of my prototype sensors that will amplify the E and A strings only. While Mr King uses big conventional pups with high impedence, mine are very small and surface mounted and wont pick up signals from the D string nearby. A slightly different design would be required to ensure that each string is separate and not two as in this case but believe I've developed this end enough to be successful at this part which is no small...well is small...achievement...for a prototype!

Otherwise yes, it is a means to get that range and separation on an ordinary guitar without the need for extra bass strings (and wide strong neck of a dedicated instrument) or extreme down tuning and such (you can of course still tune how one wishes). It is also a doubling of the guitars sound, so the strings will still be heard as normal. Generally if you play close intervals in that register they will sound way too 'thick' to be useful. Bass lines are generally monophonic. These little sensors are in addition to the normal sound of the guitar so it is not as if you loose the string when a lower one is played, just the octave drop accompaniment because a lower string is sounding.

Anyway, I think you got the concept and perhaps the little sensor prototype will help you and others get the idea of this further and where I am at. To make the basic prototype work, a simple preamp is all that is required to boost the signal and this will be the case on any such miniture device.

...

Ok...well the 'logic' is not really hard, if the signal is coming from the lowest string E6 the other two are switched off creating a mono signal that will go through to a preamp and out to the processing and amplifier.

Now, you will have to hold my hand through the rest...lol

I was thinking of perhaps a compressor like circuit, something very simple and small. Something like a switching transistor on the sensors to be switched off, the gate triggered by the signal out of the lowest sensor. Perhaps a diode to filter out the appropriate side of the signal so it is only +ve or -ve and perhaps a cap that will keep that signal high between the 'dips' in the signal and smooth things out...if that makes sense. I am more of a concept guy and electronics I know a vague idea of how one might approach it, but not the details of designing the most effective way or other options. Like the idea of using a comparitor...you might need to expand on this for me...

You are probably right that some kind of 'threshold' will be needed, certainly careful playing technique to ensure clean triggering). I don't think it needs to be too complex, though I can be very naive in this, as this is not amplifying signals or in the signal chain, but a switching circut that hopefully wont (it cant to be effective) be heard in the audio.

Bear in mind, for this application things need to be small, power light, cheap and within my capabilities to construct...as well as being effective for the application.

But, perhaps you have struck on some of the potential pit falls. Clicking and popping of these switches obviously could not be allowed to enter into the sound so that would need to be addressed. Ther could not be too much of a delay though as the signal will potentially be behind the beat...so it needs to be fast.

Sure, every idea helps, I'd be working in a vacuum and I know there are people whose interest and expertise is well beyond mine in terms of electronics required to make this happen effectively. There are many applications, potentially, of these kinds of sensors but at this stage this is the application that I wish to have and use personally and most appeal and perhaps simplicity. Other avenues I entertained was to send signals in stereo where you could select string sets and process as one might in stereo. Say a chorus effect on the higher string set, or delay or distortion...etc. But, this "bass expansion" idea is something that really does offer a lot of potential, at least for the kinds of things that I want to achieve and potentially of some niche appeal commercially. But anything like that is a long way off if ever as I ahve found from other similar projects I have achieved in this kind of line.

If the sensors are more refined through this kind of application, others might have other ideas for their application. Presently hand building prototypes could be an expensive proposition, but you never know. If I was to develop the things to something that I would seriously think to 'sell' then I imagine that I would want them to be independently evaluated and perhaps tweeked by collaborators...

For now it is a cool concept, but not just a vague idea...the ability to octave drop is affordable and available (many will have such effects in their digital multi effect boxes), musically (though I thought of it independently), Steven King has done something similar and good enough to demo the effectiveness of the concept...this is just the next logical step in this idea. I have developed working tested sensors and a way of making them and very small and discrete from string to string which too a lot of time and different approaches to achieve. The switching is my elaboration of the concept so that one can play notes on the A string while the E string is being 'dropped' which is not possible with Steven King's approach and if you look at clips of him playing, you will see he has developed a 'style' that puts bass lines only on those two strings and requires a lot of stretching along the strings and fretboard to reach the required notes...

...I notice here that he does play bass notes on two strings but that live his bass sound processing is a little 'fuzzy' and indistinct...his approach is a fairly crude way of doing it but effective in his hands and stunning in recordings.

Ok, enough from me...any other ideas or want to flesh this out further?

 

[4tune8]

I admit that I might be completely naive and this is a concept rather than a 'circuit'...but this is the kind of approach I came up with....troubleshoot and develop it, or alternatives welcome...

So, sensor E6, if there is a signal, has it's audio wave chopped by a diode so that only the positive (or negative?) wave charges a cap, that keeps the signal roughly above zero when sounded...this triggers the gate of a transistor on A5 and D4. Similarly, if no signal from E6, signal from A5 will pass, but will trigger the gate of D4 and turn that off. Any signals that pass goes to a single preamp for output.

As I say, may be naive in this kind of approach, but seems to make sense and might inspire better ideas or save me from following an impractical path...any help is appreciated in this...

Thoughts? Errors? Alternatives? Help with details?

Thanks...

 

[4tune8]

Bump

Any thoughts, perhaps other sources to canvas for assistance with details on how to make soemthing like this work?

cheers