Dragon's Den - Analogue Optical Guitar Cable

[Phil Starr]

If they say it's analogue then it probably is analogue.

Anyway here is my guess as to its workings. You can do this with an ordinary torch bulb (with huge bandwidth restrictions and distortion due to hysteresis/time lapse) connect a bulb to your speaker output and it will light up in time with your playing. As one or two people have already pointed out it was commonly used in the past to couple/decouple and compress a signal. A lot of old 70's disco sound light machines ran off the speaker jack and used this technique to decouple the speakers and act as a bit of automatic volume control. Using a led would give you a much faster response time and you'd just drive it off a pre amp circuit at the guitar end, hence the battery. At the other end you need a light dependant resistor or transistor to turn the signal back to an electrical one, hence the second battery.

For me it's just a gimmick though. None of the 'problems' with guitar leads have ever worried me in practice and my very ordinary G30 radio connection works great too. Long term I'm not convinced of the stability of an optic fibre and connectors gigged every night and chucked into a lead box at the end of every gig after being wound up by the drummist.

[luckydog]

I'm curious how it might work. If analog, guess it must be FM on a carrier, that at least might work !

LD

[Guest bassman7755]

[quote name='tauzero' timestamp='1470235502' post='3104370']
They're digital rather than analogue - it would require either transmitting and receiving devices having very linear responses, or for them to at least match, or some form of compensating circuitry at one or both ends to cope with the non-linearity. Come to think of it, I've got some laser diodes and photodiodes, maybe I could do some experimenting.
[/quote]

Even commodity DACs would surely have much better linearity than this sound/light conversion process in which case a decent wireless system would be better (with the added bonus of no actual cable to restrict movement).

[4stringslow]

True, but even DACs are not necessary for a digital transmission system over optical fibre. PFM is another option.
https://en.m.wikipedia.org/wiki/Pulse-frequency_modulation

Edited August 4, 2016 by 4stringslow

[Sausages]

I was more concerned that they said that they couldn't afford to manufacture 100 of them; what the hell is the manufacturing cost? Not much use if you have to sell your bass to cover the cost of a cable

[TimR]

[quote name='4stringslow' timestamp='1470296103' post='3104779']
True, but even DACs are not necessary for a digital transmission system over optical fibre. PFM is another option.
https://en.m.wikipedia.org/wiki/Pulse-frequency_modulation
[/quote]

That would only transmit the frequency data though wouldn't it? Everything would just be at the same volume.

[tauzero]

The way it would work would be significantly different to opto-compressors and light-controlled wah pedals. Those are used to provide a variable resistance in a signal path or a filter network, so don't themselves need to carry the signal. This is more akin to a MIDI opto-isolator (which is carrying digital information at 31kHz or so), except that the signal is analogue and not digital.

You'd have a laser diode or LED at one end driven by the signal from the guitar. This points down a fibre cable to a photodiode. You then use the photodiode in a potential divider circuit (I should think) which will give an output voltage varying with the intensity of the light from the laser diode. This (allowing for all the distortions introduced by the laser diode and photodiode) is a replica of the signal output by the guitar.

[440Hz]

I was watching Dragon's Den on catch up and thought what's new?

I Googled and landed on this forum which, as a retired electronics technician working in education, I found interesting.
I agree that varying the brightness of the light entering a fibre optic cable is amplitude modulation of an extremely high-frequency carrier. (light is just an extremely high frequency radiation travelling at the same speed as radiowaves in free space)

I remember years ago (probably late 60's) putting a 15W pigmy lamp on the output of a valve audio amplifier via a step up transformer to match it.
The audio was effectively modulating the light.
However if you consider a single note say an A at 220 Hertz this will light up the lamp not 220 times a second but 440 times a second an octave higher(since the lamp lights up the same irrespective of the direction of the current).
A photocell (I actually used a cell from a photographic light meter) which sees this light will thus produce a voltage at 440Hz and if connected to an amplifier and speaker will sound very distorted as it is mainly second harmonic distortion.
A very simple trick was then applied; a 90 volt battery (from an old portable valve radio) was put in series with the lamp and with:[list]
[*]no signal the lamp glowed
[*]on the positive parts of the signal the lamp became brighter
[*]on the negative parts of the signal the lamp became dimmer.
[/list]
The complete 220 Hertz signal was causing 220 fluctuations of brightness each second.
The resulting audio via the photocell was then reasonably undistorted and with the pigmy lamp (thin single non-coiled filament) the frequency response was sufficient for intelligible speech to be carried across the link.
Coiled coil lamps had too much thermal storage for a good frequency response.

Jump about 40 years[list]
[*]Replace the valve amplifier with a guitar pickup.
[*]Replace the pigmy lamp with an LED, The 90 volt battery with about 1.4 volts bias on the LED or better still drive it with a constant current around 1mA, vary the current source up and down with the audio signal
[*]Pass the light through a fibre optic cable
[*]Demodulate the light at the other end with a photo-diode or photo-transistor and you have a link with a frequency response extending well beyond audio.
[/list]
The quality of the link is dependent on the linearity of the devices. More signal less noise but greater distortion and vice-versa pick a good compromise and you have a good quality analogue AM link.

If you were to use FM modulation you could achieve a greater dynamic range and would not get the distortion caused by non-linearity in the diodes. FM does not introduce a delay as encoding a digital signal would and the system would still be sort of analogue as an FM signal is a continuous analogy of the original signal and not broken up into samples as it is with a digital a to d and d to a system.
For an AM system performance could be improved by adding negative feedback over a separate fibre optical cable using matched transducers but this would complicate things. Anyway why not use an FM wireless link and do away with all the cables on stage?

Edited August 4, 2016 by 440Hz

[Count Bassy]

[quote name='4stringslow' timestamp='1470155918' post='3103688']
Agreed, not entirely necessary, though even a high impedance 5m unbalanced cable is susceptible to noise interference and the longer the cable the higher the risk. It would do no harm to standardise on balanced cables. Not all mic leads are particularly long (eg PA amp on stage with the band) but they are mostly balanced XLRs these days so why not guitars?
[/quote]

Another difference is that a typical 'good' microphone will output up to about 50mv. Line level is typically 1v. Guitar outputs will vary widely according to many factors (particularly active/passive), but will generally be significantly higher than the microphone output. Through the same cable the induced external noise level will be the same, so is less of a problem the higher the signal level.

A balanced signal will be better than unbalanced in all cases, but becomes less necessary as the signal level increases, more necessary as the cable run increases.


In theory optical cable is brilliant because it is totally immune to external electrical noise, but seems an overkill when there are other cheaper & more practical solutions available.

Now, what about 'Starquad' cable?

Edited August 4, 2016 by Count Bassy

[EBS_freak]

I do hope somebody spawns a sister thread about how copper cables work...

[opticfibre]

Hi Guys!

Its looks like the most popular question here is How do we do it?
The answer is simpler than you might think, and may cause a lot more questions from you.

We use IM to send the analogue optical signal Intensity Modulation not AM or FM or PWM or...

https://en.wikipedia.org/wiki/Intensity_modulation
when the voltage signal goes up, our light goes up, when the voltage signal goes down our light goes down.

Our wavelength is 660nm but i don't think that's important, the reason we use 660nm is because the bespoke optical fiber we use is most sensitive in that range.

Hope this helps

[EBS_freak]

Thanks for joining us!

You wouldn't get this level of techy conversation on a guitarist forum would you?

[TimR]

Thanks. But isn't intensity dependent on amplitude and frequency?

You're fixing the frequency so...

Anyway, due to the 'wave' being light and not really a wave I suppose it's right that you should have another way of describing it.

Thanks.

Edited August 4, 2016 by TimR

[opticfibre]

My Pleasure to join, took a couple of days to get approved by the admin

Don't think i should compare basschat to guitarchat

[gapiro]

[quote name='opticfibre' timestamp='1470311322' post='3104983']
My Pleasure to join, took a couple of days to get approved by the admin

Don't think i should compare basschat to guitarchat
[/quote]
One of the questions I asked before was on the cable strength - one of the things that is typically associated with optical fibre is a low flexural strength - that is it is prone to snapping with low radius bends in the cable - what is the tightest bend you can do with your cables? Are you able to do something like this:

[opticfibre]

The frequency of the LED isn't changing just the brightness.

Some people will say AM amplitude as its the amplitude of the light.
people assume we are transmitting a frequency i.e 100khz and changing the intensity of the 100khz and modulating the signal on that.. but we don't transmit a frequency. The 660nm is just the frequency of RED light, but i see your point about 660nm being the carrier frequency and us changing the amplitude of it.

Its similar to when we say optical, people assume its digital, Peter Jones assumed we were encoding and decoding.

[opticfibre]

[quote name='gapiro' timestamp='1470311609' post='3104985']
One of the questions I asked before was on the cable strength - one of the things that is typically associated with optical fibre is a low flexural strength - that is it is prone to snapping with low radius bends in the cable - what is the tightest bend you can do with your cables? Are you able to do something like this:

The minimum bend radius of our bespoke optical fiber is 2mm. Its quite flexible.
[/quote]

[opticfibre]

The minimum bend radius of our bespoke optical fiber is 2mm. Its quite flexible.

[gapiro]

[quote name='opticfibre' timestamp='1470312081' post='3104989']
The minimum bend radius of our bespoke optical fiber is 2mm. Its quite flexible.
[/quote]
Assume thats for the fibre inside, so realistically the protective coating etc on the "cable" will have a higher bend radius due to the tensile strength etc.,

[TimR]

[quote name='opticfibre' timestamp='1470311322' post='3104983']
My Pleasure to join, took a couple of days to get approved by the admin

Don't think i should compare basschat to guitarchat
[/quote]

You'll be ok, just try not to get dragged into bizarre arguments. There are a few people who can get a bit 'overzelous'.

Would this be better aimed at the PA market? A 64 core snake would be lighter than the usual copper balanced one, and more relevant for your application.

I think your hardest job is selling it on the strength of clearer signal. The majority of us play standing next to drummers and opposite guitarists.

.

Edited August 4, 2016 by TimR

[stingrayPete1977]

Hi opticfibre, I think getting along to a bass bash would be a great idea, maybe once we hear it in action we will become more optimistic? My current lead is around 10 years old with no issues, I do look after them so that helps, for live use I'd be much more likely to go wireless at some stage.

[4stringslow]

[quote name='opticfibre' timestamp='1470311916' post='3104986']
The frequency of the LED isn't changing just the brightness.

Some people will say AM amplitude as its the amplitude of the light.
people assume we are transmitting a frequency i.e 100khz and changing the intensity of the 100khz and modulating the signal on that.. but we don't transmit a frequency. The 660nm is just the frequency of RED light, but i see your point about 660nm being the carrier frequency and us changing the amplitude of it.

Its similar to when we say optical, people assume its digital, Peter Jones assumed we were encoding and decoding.
[/quote]

Thanks for joining in - it's great to get some definitive answers!

I'll admit to not having heard of IM (and I worked for an optoelectronics company for nearly 10 years!) but I suspect that's because it really is just another term for AM. You say you're modulating the power (intensity) of the light, I'd say you're modulating the amplitude (brightness) of the light. We're saying the same thing but just using different words. It does explain why your website states it's not AM (or the others) but I think it's a source of confusion.

I can understand the digital assumption because almost all modern optical communication is digital, but digital doesn't necessarily mean ADCs and DACs because the signalling could be achieved by PFM (see my earlier link) which is a form of FM with various advantages over AM (or IM ). But perhaps the technical advantages are not necessary for this application and might even be a marketing disadvantage if you believe the market values a completely analogue system. That's a trickier judgement call!

May I ask if the Dragon's offer made it through the contracts stage and you really did receive the funding for more prototypes?

[4stringslow]

[quote name='TimR' timestamp='1470313506' post='3105006']
Would this be better aimed at the PA market? A 64 core snake would be lighter than the usual copper balanced one, and more relevant for your application.
[/quote]

Depending on the value to the market, it would be technically possible to send 64 separate signals along a [u]single[/u] optical fibre. I know this to be the case as the company I used to work for was doing this with 128 separate video signals about 20 years ago, using Wavelength Division Multiplexing (WDM) - simplistically, multiple different colours of light

[opticfibre]

I started with guitar, because that's how i thought of the idea, but yes other areas are being considered, such as DMX512 and also a seismic measurement company has been in touch with us!?

[EBS_freak]

[quote name='TimR' timestamp='1470313506' post='3105006']

Would this be better aimed at the PA market? A 64 core snake would be lighter than the usual copper balanced one, and more relevant for your application.
[/quote]
I originally thought that too - although I suspect that ship has set sail already. Any company wanting that sort of channel count is going to be in the realm of digital - so a couple of cat5s does the job - at a significantly cheaper price also.