Optical cables

[zpyder]

Using optical cables to transmit instrument signals......


IM SO FREAKIN EXCITED because I've been undertaking an indie recording project with my current band (it'll be our first release that's not a demo) and all is going quite well.  Nothing is quite as rewarding as putting a couple years into something, having very little time or money (only a small stockpile of well-selected gear, beers, and skills), and hearing the preliminary results for the first time and saying - "shit... that sounds good"   

Anyways...

I bought an E-MU 0404 PCI audio card (http://www.musician.com/product/EMu-0404-PCI-Digital-Audio-Recording-System?sku=240390) yesterday to facilitate with the transfer of optical S/PDIF signals from my Dig. 8-track (where the raw recordings live) to my computer (where I'll master).  No longer will I worry about stray EM fields in my house.  No longer will I worry about not plugging my PC and my 8-track into the same outlet.  Not to metion, aside from interference, the signal stays digital until it's played in our audiences' speakers - I'm using one less pair of DA/AD converters - just a little more lossless (phunny frase).

WAIT A SECOND!!! I just bought a 3-prong out let tester last week, became at least aware (if not a little concerned) with electrocution potetional when touching my strings and a mic, I get zapped every time we play at the drummer's place and I'm singing - and all because my guitar is grounded!  I read somewhere that the only foolproof way to prevent electrocution due to ground voltage differences between a guitar and a PA is to use a wireless system - duh, no ground!

Sooo... I'm curious what others think - what would the be the downsides (other than a little loss) to putting an AD converter in my guitar, sending an optical signal to my pedalboard, and then running it through a DAC into my pedal chain?
What would be the upsides?

I haven't read anything about it before, and a search for optical on this forum turned up very little...

cheers,
zpyder

[Mark Hammer]

Sometime in the late 1980's there was actually an optical guitar cable project in Electronic Musician.  I may just have the article in one of my binders.  Just note that fibre-optic cable may not be quite as flexible as some mic cables.  Still, in theory it ought to be a nice alternative to wireless (as far as going greater distances with no treble loss).

[JimRayden]

Hmm, but why convert to digital at all? I mean, you can let the guitar signal control the intensity of the red light. No digital sampling needed, and nearly no mojo lost.

----------
Jimbo

[zpyder]

I mean, you can let the guitar signal control the intensity of the red light. No digital sampling needed, and nearly no mojo lost.

Ahhh... interestring.  This is the kind of thing I like to see... So one could simply use a red LED at your guitar's output, sending light through the pipe based on your signal.  At the other end would be an LDR.... hmmmmmmmmmm. 

zpyder

[Ardric]

A phototransistor, not an LDR.  It needs to respond quickly.

Don't need a LED on the guitar; pass laser light down the same cable in the opposite direction.  This makes it a passive, analog optical pickup.

Modulate the light source at a frequency well above the highest audio frequency before it goes down the cable, then demodulate the recieved version.  Maybe that'll make it easier to subtract any ambient light source noise.

[zpyder]

A phototransistor, not an LDR.

sweet.

pass laser light down the same cable in the opposite direction.

how?  what?

zpyder

[JimRayden]

pass laser light down the same cable in the opposite direction.

how?  what?

zpyder

Just like you have two conductors on a regular cable, you can have two glass fibres in a light cable.

------------
Jimbo

[zpyder]

hmmm... I guess I'm still confused as to how you'd imprint your guitars signal onto the laser light passively... with regular pickups

zpyder

[Ardric]

hmmm... I guess I'm still confused as to how you'd imprint your guitars signal onto the laser light passively... with regular pickups

Maybe the passive magnetic pickup has enough energy output to power a nano-sized array of series wired voice-coil driven mirrors....

But really, I meant an optical pickup.  Shine a LED on the string, put a phototransistor opposite the LED, and we have an electrical signal from an optical pickup.  Now put a pair of optical fibers between the string and the LED/phototransistor, and the "pickup" is passively optical.  It's just two fibers pointing at the string.  The electronics are in the box/amp/whatever at the other end of the fibers.

But how to do the guitar knobs...  A graded translucent density wheel as volume control?  Maybe a pair of tinted sunglasses for tone control? 

[Paul Perry (Frostwave)]

All these things - including wireless - are non-trivial for home construction, if you want "studio quality" noise floor. Which, unfortunately, is the point.

[JimRayden]

hmmm... I guess I'm still confused as to how you'd imprint your guitars signal onto the laser light passively... with regular pickups

Maybe the passive magnetic pickup has enough energy output to power a nano-sized array of series wired voice-coil driven mirrors....

But really, I meant an optical pickup.  Shine a LED on the string, put a phototransistor opposite the LED, and we have an electrical signal from an optical pickup.  Now put a pair of optical fibers between the string and the LED/phototransistor, and the "pickup" is passively optical.  It's just two fibers pointing at the string.  The electronics are in the box/amp/whatever at the other end of the fibers.

But how to do the guitar knobs...  A graded translucent density wheel as volume control?  Maybe a pair of tinted sunglasses for tone control? 

Congratulations, the light show doesn't follow you guitar anymore, your guitar follows the light show! The humankind is about to hear optical guitar feedback.

------------
Jimbo

[zpyder]

haha wow... thinking outside the box...

Getting a little OT but I think this line of thinking could be productive - though probably not for what I was intending.

Okay, I really enjoy the idea of "passively" "picking up" a string's vribrations optically -
1) We create an "input box" which serves two purposes - producing our "carrier" laser (freq. well above audio) and sending it toward to guitar, and also recieving the resultant "signal" laser and sending it through a phototransistor and out to our signal chain.
2) We use a two-fiber optical cable to go between our guitar and our "input box".  The first carries the carrier laser to the guitar, the second sends the signal back to the box.
3) We route the carrier laser, once it gets to the guitar, into 6 seperate strands (not sure how to do this part...)
4) We route 6 return signal laser strands back into one (not sure how to do this...)
5) We build a mounting system which suspends one carrier laser output lens below each string and one return signal lens above each string.  This must be as unobstrusive as possible.
6) Make a freakin' switch so you could also use your guitar's ordinary signal!

All these things - including wireless - are non-trivial for home construction, if you want "studio quality" noise floor. Which, unfortunately, is the point.

Ohhhhhh, come on Paul, you really don't think DIYing nano-sized arrays of series wired voice-coil driven mirrors wouldn't be easy!!??!!??  But seriously, your point is very valid in that to do this with any level of fidelity would be pretty rediculous.  When am I going to read about optical resistors and capacitors!!??

I believe that if someone were to construct our above passive optical pickup and reciever, some very interesting results could come out.  I'd bet a very crude signal could be transferred (albeit quite tediously) - and hey - sometime's that's sweet...

As for my idea of using optical cable to transfer my guitar's signal to my pedalboard and therefore eliminate ground loop shocks and a little hum (hum's not so much an issue for shorter cables), I think I'll stick to the idea of a DAC and an ADC - seems to be a reletively inexpensive and simple way to go...

BRAIND CANDY!

...humankind is about to hear optical guitar feedback.

cheers,
zpyder

[Ardric]

3) We route the carrier laser, once it gets to the guitar, into 6 seperate strands (not sure how to do this part...)

A block of frosted plastic that goes behind all six strings.

4) We route 6 return signal laser strands back into one (not sure how to do this...)

Mix them with optics (mirrors/prisms/lenses), or return six fibers and mix them afterwards.

But seriously, your point is very valid in that to do this with any level of fidelity would be pretty rediculous.  When am I going to read about optical resistors and capacitors!!??

There are various people doing optical pickups already.  A commercial effort:

http://www.lightwave-systems.com/

Check out this DIY-ish violin:

http://hct.ece.ubc.ca/nime/2005/proc/nime2005_034.pdf

I was recently googling this stuff with an eye towards making a pickup for my bandmate's cello.  Woops, pretend I didn't say that.

I don't see why a DIY optical pickup needs to be crude or noisy.  The whole reason for taking this approach is to reduce noise, and by all accounts it does that very well.  The main complaint against it seems to be that it's too hifi; that it needs EQ to sound good, but it'll never sound the same as a traditional magnetic pickup.  I guess it's like a piezo that way.

As for my idea of using optical cable to transfer my guitar's signal to my pedalboard and therefore eliminate ground loop shocks and a little hum (hum's not so much an issue for shorter cables), I think I'll stick to the idea of a DAC and an ADC - seems to be a reletively inexpensive and simple way to go...

How are you going to power it?

If it's just isolation you're looking for, why not use a matching transformer?  That's gotta be simpler than putting a computer in your guitar.

[zpyder]

I don't see why a DIY optical pickup needs to be crude or noisy.  The whole reason for taking this approach is to reduce noise, and by all accounts it does that very well.  The main complaint against it seems to be that it's too hifi; that it needs EQ to sound good, but it'll never sound the same as a traditional magnetic pickup.  I guess it's like a piezo that way.

I absolutely believe that anything done commercially can be done just as well DIY, DIY does not impy crudeness or noiseiness.  However, building an optical pickup that is anything but noisy and crude is well beyond my current time and skill restraints.

Actually, my original impetus for this idea was to isolate the guitarist from the electrical circuits his/her instruments connects to, not to reduce noise.  Probably more importantly was just to put some brain candy on the table for people to try.  Using light for picking up string vibrations has an obvious ability to create a very high-fidelity signal, which could also be explored.

The two examples you provided were very interesting...cool... I imagined that there were at least functioning attempts at this concept out there, just hadn't seen em.  Thanks for providing those!

If it's just isolation you're looking for, why not use a matching transformer?  That's gotta be simpler than putting a computer in your guitar.

Again, this thread was started with the intention of bringing up an idea that doesn't seem to have been explored very much or get much attention, not to specifically address an issue.  Using a matching transformer to isolate a player from the board is obviously much simpler and cheaper, however, I started this thread to start a discussion about optical transfer of signals between components in a chain - as a general concept - something that at least my first couple searches on this forum returned not a single thread for...!

With all of the ground loop issues that people seem to have, it seems that optical may be a viable alternative, at least in the future.

To keep the topic alive:

A lot of people use pedalboards, and some, like myself, power all/most of the pedals via a daisy chain connected to only a single PSA.  Since all pedals share their grounds via the shield of our 1/4" jacks, this creates a ground loop problem.  I'm looking at building an isolated supply (i.e. R.G.'s Spyder) soon, but just to raise the idea: With a good ADC and DAC, we could use S/PDIF (either optical OR coaxial) to go between the pedals on a pedalboard.  Aside from this effectively eliminating any possibility of picking up a radio signal across the patch cables and eliminating ground loops, what could this do to the sound of our pedals?  I'm thinking input/ouput impedances...

cheers!
zpyder

[R.G.]

Unless you can do the work all optically, no electronics on the guitar at all, you haven't solved much. The guitar still gets connected to an electrical widget.

On the other hand: Optical fiber is quite small. Run a dozen light carrying fibers up to the guitar, fed by an LED at the equipment end. Run a dozen fibers back to the equipment and couple them to phototransistors there in the same differential occlusion setup as the lightwave/violin system, but do all the electronics there at the equipment. The light source doesn't need to be a laser, an ordinary IR LED will do nicely. You probably need some special setup to make up for the sensitivity of fibers to being bent to make the cord more flexible.

RF is very effective for isolation.

ADC is also power hungry.

[Mark Hammer]

As I recall, the project in EM was pretty much all analog, not digital.  I'll try and find it later this week.

[Ardric]

RG's correct of course, it takes two phototransistors per string ("differential occlusion"), not one.  My mistake.

I like the idea of buying a commercial wireless guitar setup too.  Sure makes it quick and easy.

An ADC is not something I'd want to do more than once in the signal chain if I could help it.  Going through a long chain of ADC/DAC without extra efforts like clock sync is just asking for new and exciting sources of noise.  Putting an ADC/DAC pair between a couple of simple analog stomps seems like severe overkill.  Heck, even the power supply for such a setup would be more complex than the average stompbox.

I'm looking forward to seeing the article that Mark started the topic on.  My guess is FM modulation of a square wave carrier.

[JimRayden]

I like the idea of buying a commercial wireless guitar setup too.  Sure makes it quick and easy.

I thought the point of implementing analogue optical signal carrying was to remain as much of the signal as possible and not worry about conventional cable shortcomings, such as resistance and capacitance.

Or have you pondered to other fields while I wasn't paying attention?

-----------
Jimbo

[puretube]

optical pickups:



from here:

[JimRayden]

Puretube, you DO have a photo for every occasion.


And a gigantic KUDOS to your 5000 posts. I've got a long way to go to catch up.

----------
Jimbo