Galvanic isolation for heart wah

[Cursor]

Hi all.

So I have this heart-controlled wah going on. Think biomed preamp, built around a differential amp made up of cheapo opamps, providing control voltage for a Tim E. VCF. I'm a little wary of posting a schematic as I don't want anyone to kill themselves but realise I might have to in order to get assistance.

My query is, is there any way I can safely isolate this thing? I'm going for battery operation only. I'm not too sharp on what would happen if for example a massive spike appeared on the upstream audio. I want to be pretty sure that couldn't make it the electrodes. The prospect of a dodgy ground that floats higher than true ground is also a worry.

Best,
I.

[Cursor]

Hang on... Let me rephrase.

Is there a way I can optoisolate the input and output of a pedal?

[Johan]

there are such things as opto transistors..if you could get a LED to give a good representation of the signal( amplitude and frequency), you could probably get the optotransistor to remodulate that light into an electrical signal...but I wouldn't expect too much of such setup....

I would start something like this
for the LED part I would try a regular opamp buffer with a 100-470 ohm resistor to a LED connected to the ground/ battery negative( so it always shines. you want it to represent the whole signal, not just the positive or negative swing.)
the opto transistor..from +v through 10kohm to collector. LED to base and 1k to ground from emitter.. get your re-modulated output from the collector...
...and probably end up with a lot of tweaking...

...but like I said...I wouldn't expect too much from such circuit...
 
j

[Mark Hammer]

There are many ways to detect the presence of a pulse.  What method were you intent on using?

[Cursor]

Thanks Johan; I will try a simple opamp driving a 4N25.

Mark, I'm using a pair of skin electrodes, one on each arm, biased with a "driven leg". A preamp (not my own  design) uses an AC-coupled opamp arrangement and gain stage to amplify the detected pulse. Then I use a summing amplifier to give me the control voltage.

Electrodes make sense for me rather than an LED or other approach as the eventual goal is a brainwave-controlled filter. In a seperate application of this preamp I've set up a microcontroller with onboard ADC to pass the data in realtime over an optically isolated serial link. For a brainwave-controlled filter pedal, I'll need a couple of orders magnitude higher preamp gain using a different design and a microcontroller with onboard FFT code, but I may as well address the isolation and practical issues now with a simpler heart-controlled filter and see if those problems can be solved.

Plus, electrodes on stage is very much my aesthetic

[Mark Hammer]

While one obviously CAN provide pulse information that way, my sense is that the electrical environment in which you would be doing it is ridiculously contrary to taking effective recordings.  Consider that the electrical activity is in the range of microvolts and that you've got transformers, fluorescent fixtures, cellphones, blackberries, and all kinds of other stray signals.

My vote is to strike out on a different path and go optical, where all you have to contend with is noise from the circuit, rather than competing signals

As well, consider that you would likely need some sort of averaging circuit, because the beat-to-beat variation when someone is engaged in an activity (as opposed to lying on a table or bed, motionless, with relaxed breathing) would create "pattern havoc".  You want the filter to have some semblance of order to its action.

[Cursor]

Mark,

Thanks, I'll look at optical detection; but the actual capture is a solved problem for me. Noise-wise it works pretty well at home but I haven't tried it in a band practice environment yet. The summing amp doubles as a low-pass filter.

Captured trace below (soundcard capture I'm afraid).



The pattern clash aspect is interesting playing against a metronome and will indeed be chaotic in a band setting. That is a good thing The kind of biofeedback loop feeling you get playing on your own is amazing. I can't wait to try connecting the heart wah to our drummer too.

In any case, while I was typing this reply, I solved my problem. It's so obvious I feel stupid for even asking on here. I can, of course, optoisolate the CV voltage into the filter!

[R.G.]

Just make sure that there is no conductive path from the electrodes you're wearing to the equipment you play. Isolating the control voltage is great as long as the stuff you wear on your body has no other path to the AC line, safety ground or not.

PWM is a good way to modulate so you can get across a safety isolation barrier. Another is voltage to frequency (that is, fm) and then PLL to the received signal across the isolation barrier. A 4046 will follow an FM signal easily enough.

[Cursor]

R.G. - Many thanks, I was hoping you'd chime in Point regarding paths to equipment from electrodes noted.

I'll have a look at PWM, but to be honest I was thinking SUPER simple - a DIY Vactrol (capitalisation?). I'm off to make one. Such an easy method to interface with effects circuits too; I can play with heart tremolo etc too... of course that pedal would have to be called "heart tremor"