A new fully analog FM modulated wideband optical transport Open Standard.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Read post #36 and then forwards from this thread:
http://www.diyaudio.com/forums/tubes-valves/220271-glowing-plates-doom-4.html#post3230168

Important posts are:
http://www.diyaudio.com/forums/tubes-valves/220271-glowing-plates-doom-4.html#post3230185

http://www.diyaudio.com/forums/tubes-valves/220271-glowing-plates-doom-5.html#post3231390

http://www.diyaudio.com/forums/tubes-valves/220271-glowing-plates-doom-6.html#post3231674

http://www.diyaudio.com/forums/tubes-valves/220271-glowing-plates-doom-6.html#post3231684

I'm hoping that the diyaudio community can collaborate and construct some kind of working prototype, which will then spur further development later on for higher analog bandwidth over fiber optic cabling.

So people will

#1, Have a way of optically isolating their TT or DAC's from their amplfiers and transport those low level signals over from one part of the room to the other with little loss or degredation of the signal by placing that low level analog signal onto an optical fiber.
#2, Do away with the expensive cable industry once and for all, much like how Optical TOSLINK has done for home consumer grade audio.
#3 And finally retire the somewhat unpredictable RCA connectors, even though RCA is great and all, and even works for RF on the HF bands, I think we should develop a new optical transport interconnect, you never know it might turn out to be half good.

Base it upon a currently existing connector so adoption is widespread, Some suggestions are (for stereo) a pair of LC/SC or ST (Bayonet fiber connectors), or a pair of TOSLINK connectors. Each carrying 1 High quality mono channel or 2 lower quality mono channels for 2 or 4 channel analog.

HJWeedon has made it pretty clear already that this should be an FM modulated standard, with a 200kHz carrier deviation. And as we all know FM can easily carry Stereo, so catering to the 5.1 or 7.1 crowd shouldn't be too difficult.

The astronomy/physics community is already doing it: http://www.skatelescope.org/public/... Technology for Astronomy Instrumentation.pdf

So why not us?

Here is a few more interesting links to look over:
http://electronics.stackexchange.com/questions/25375/current-modulator-for-laser-driver
http://www.ilxlightwave.com/appnotes/AN 22 REV01 Modulating Laser Diodes.pdf
http://www.its.caltech.edu/~atomic/publist/diodecontroller.pdf
 
Last edited:
ok ok, I'll give it a shorter name, how about analog laser-based open sound interface or ALOSI

My contribution to this idea has ended, I'm far too unwise to be contributing to this thread and I hope some smart sparks will come along and make it a reality with a working prototype, then publish the schematics.

I certianly hope the same thing doesn't happen as has happened in the other thread over. And by this I'm referring to "A New Interconnect Design".

Its been a long time coming thats for sure. and by that I mean this project.
 
Last edited:
Sounds completely barmy to me. You want to include a highly linear analogue FM or pulse-width modulator and corresponding high quality demodulator for every interconnect? Every line stage becomes an FM tuner? Have you ever designed an FM tuner? What is wrong with a bit of copper cable?

You will never do away with the expensive cable industry, because such purchases are not based on need but perception. If your idea got off the ground then within a few months people would start offering expensive fibres for it. That is human nature.
 
Hello.

First we have to identify the customer. Is this going to replace cable? My answer would be NO. Copper cabling works pretty well. It is only when there are significant AC ground-loop problems that we have to do away with the copper connection.

In my own case, this was primarily due to the shortsighted Ground. Neutral and Hot wiring of incoming power. This system is inherently unbalanced, capacitive and inductive coupling introduces 50/60Hz hum in every wire. In my own case I installed an isolation transformer where the output is 2 X 65V AC. The center-tap of the isolation transformer is connected to GROUND. This power configuration gives about 20dB lower sensitivity to line frequency HUM.

But that is not what we have set out to do.

A suggestion has been made to use 98MHz carrier I.E. right in the middle of the FM band.
Why do that? is that so all our neighbors can listen in on their FM receiver? I do not like this choice for that reason.

My selection would be: 2MHz to avoid interference in the standard AM band, and not in the way of Amateur radio which is generally located at harmonics of 3.5MHz. The 2MHz wiring is low enough so that regular wiring technique can be used. Also circuit design can become relatively mundane.

My second choice would be 10.7MHz where a plethora of circuits exist off the shelf. This is the standard FM IF frequency.

My third choice would be 60MHz in the RF-"no-mans land" second harmonics is 120MHz etc not disturbing anyone.

10.7MHz would require some careful wiring, but not excessive. 60MHz would require solid state and printed circuit-board technology. An experienced constructor could handle any frequency, but the less experienced builder would have to limit himself to the 2MHz circuit techniques.

That is all for now.

Hans J Weedon.
 
Insane idea, FM even with very wide deviation introduces distortion (It has to, you have to band limit the bessel sidebands somewhere), and FM discriminators are not exactly distortion free, that is before you consider that you have just doubled the number of parts in most consumer audio kit to no benefit.
Hideously complex compared to just doing a properly thought out balanced interface, which will NOT hum if both ends are implemented even mostly competently (Even an unbalanced interface will usually be perfectly acceptable given line lengths typical of HIFI usage).

Hell, if you want optical, SPDIF or ADAT are both simpler to implement than this monstrosity, or do a simpler delta sigma modulator at the transmitter and a low pass filter at the receiver.....

Moving line level audio over HIFI sort of distances is very nearly electrically trivial, and doing the same over hundreds of meters of twisted pair is only a little tougher, it does not really take anything except competent line drivers and receivers.

Sorry, but I don't buy there being any problem with sending raw baseband for most audio applications (Certainly including all HIFI uses), the fact that some folks cannot design an audio input or output stage that works properly should be taken as a sure indication that the same folks cannot design high speed FM discriminators. Design your kit right and more or less any old bit of twisted pair (Ideally with a screen but you can often get away without) will do just fine for moving audio around.

Regards, Dan.
 
Gentlemen.

Dan (dmills) is actually quite correct, the simplest interconnect is in all honestly the wired connection. It is just that the circuit designer buried deep inside us wants to get out and do something complicated and tricky. My preference for interconnect under reduced circumstances would be an Ethernet type of cable. This cable has 4 well twisted pairs and a convenient locking connector. Each channel gets its own cable where all 4 pairs are connected in parallel.

We drive this cable with about 100 Ohm source impedance to limit capacitive hum pickup. the receiving end should be a high quality differential receiver circuit with about 40dB of common mode rejection (1% resistors, obviously). The signal level should be about 3V nominal full scale, not the More "modern" 1V full scale. I pick 3V (RMS) because that is 18V in the very highest peaks (6* RMS to account for the unfortunate sum of several instruments playing full blast) This can be done to high performance with +/- 20V rails on the driver and receiver side. The load impedance at the receiver end should be be in the 22 to 47k range, mainly to limit the thermal noise of the load resistor.

This interconnect is simple and has, believe it or not, already got gold plated connection points as part of the definition. Since we plan to put 4 pairs in parallel we have great redundancy for continuously reliable interconnections.

For the freak among us this cable also comes in a more expensive shielded variety.

Anyway, this is how I plan to connect up my remote signal processor to my living-room signal sources. My listening room is 11*11m and the speakers are in one end of the room and the source equipment in the other end. I can not put the speakers next to the source equipment because that is where the piano and the book-cases are.

I can not put the Audio equipment near the speakers because there is a kitchen and a bathroom there. Instead of rebuilding the house, I prefer the less expensive way of rewiring the Audio equipment.

Years ago I had my Audio system closer to the speakers but every time the electric water-heater turned on I had induced hum in the audio channels. That turned out to be the power-feed to the house magnetically inducing objectionable hum in the Audio system.

Wife acceptance is also important, heavy shielded cables across the floor, simply does not fly.

Well that is my reality, guys.

Hans J Weedon.
 
Haha.

I initially suggested AM modulation, but then I adopted HJWeedon's FM idea because it made sense if we're going to be using fiber, and he knows better. And even on Laserdisc I truly did love the sound that the FM tracks put out, especially when they were played back on Tube equipment.

And the main reason why I suggested that we use 98Mhz was because people out there in the field can protect their neighbours by simply using an FM radio that they might have handy inorder to better shield their circuits. If these circuits are leaking /at all/ then there is something seriously gone wrong.

As far as I understand it all we need to do is produce a carrier modulator which will make its way into a laser diode, whatever the frequency is, and then modulate that modulator with an FM modulator with an IF of some ungodly low number. so all of that could be on the order of 0.005 watt of RF, tops, and it would be heavily shielded.

But if we are going to go back to a very wideband AM modulated signal then I strongly suggest that you keep away from 1.6MHz-30Mhz and instead either take it down to 1MHz or straight up to 98MHz. The shortwave bands are extremely sensitive to even microscopic amounts of wattage, hence the whole invention of the QRP crowd. and they /will/ come down on us like a ton of bricks.

Think about it this way, if we put the carrier on a band that isn't a Broadcast Band, and someone who is building their kit out there in the field all of a sudden has a bias resistor out of specification, then all of a sudden you've got anywhere from a petawatt to 1 watt of RF spattering itself all over a critical government or corporate infrastructure.

And that my friends would be utter disaster to this project.

Remember this line, the RF bands are already chock full of every single service you could imagine, there is no point in finding a free part of the spectrum because you have to take into account every single country in the world, the only bands that we could conceivably use are the ISM bands. Either that or 455KHz. or the BCB's, because 1 Mhz and 98Mhz are both internationally known as broadcast bands /and/ people can self-diagnose their circuits without any understanding or knowledge of how to use a HAM radio or a Shortwave radio.
 
Last edited:
Hello.

First we have to identify the customer. Is this going to replace cable? My answer would be NO. Copper cabling works pretty well. It is only when there are significant AC ground-loop problems that we have to do away with the copper connection.

I think it could be applied in a surprisingly large number of areas like for example in intercom and CCTV systems, all of those systems require to some extent the coverage of audio, but cabling is usually using copper/coax/ethernet, and therefore prone to lightning strikes.

Another possible avenue is the Pirate radio & HAM crowd (I don't condone what the former do but that doesn't mean you can't sell them a kit), they want to protect their source equipment, but they have to run coax cable to a tall antenna, or in the case of HAM's they want to send something an audio signal but they don't want to introduce ground loops or RF into whatever they are sending it to, so with the invention of a fiber replacement for RCA they could quite easily just run a length of TOSLINK down and power the transmitter from Solar at the top, and /never/ be worried about lightning strikes ever again. I doubt this crowd would be a very large customer though, maybe in the middle east or europe (france), yes.

Those would be the low cost end of the spectrum.

It is true that either end needs to be powered, however Solar is coming out in a big way, so that and the advancements in batteries pretty much make this a null and void issue.

The high cost end of the spectrum would of course be the expensive cable crowd and hopefully with a bit of good thinking we can beat the performance of 90% of these cables when it comes to the audio spectrum. (We couldn't beat copper on HF performance).

Where I think you guys need to concentrate on is inbetween there, what would get the average joe kind of people off of the analog RCA interconnect and over to fiber. Which is contradictory to what HJWeedon has said, and I'm sorry if I cause any offence.

We will probably have to do this eventually anyway, copper is getting more expensive every year, and how do we know its not the same quality as we used to have anymore? How do we know that its even copper? Shill out $30-$50 for a pair of RCA cables? or $10-$20 for a ALOSI kit?

Don't know about you guys but Lightning is a huge issue down here in Aus, at least it is where I live, and I like to unplug my equipment from the wall but if anyone has looked behind their audio equipment lately and seen just how close their RCA cables come near-to that power cable for your tv set should no doubt be worried about equipment damage from lightning jumping from your tv set or set top boxes antenna to your very expensive transport or DAC...

Most people don't really care about cable positioning, this would quell those fears, it would also allow the use of unearthed equipment with earthed stuff, and completely obliterate the ground loop issue WITHOUT having to go to digital.
 
Last edited:
Ontop of that, the garage band audio/studio crowd would apsolutley LOVE this idea.

Converting 1/4"/XLR to optical and back to 1/4"/XLR while remaining fully analog would sell like hotcakes.

Tho if a copper cable is cut it can be mended when out in the field, if a fiber one is cut it can't be mended, its gotta be replaced.
 
Last edited:
This could also solve the age-old folly of mixing analog signals in a noisy digital environment, say for example, oh I dunno, THE HOME COMPUTER.....

Imagine having your I2S bus coming directly from a PCI-E tap, then into your DAC which is located /inside of your computer/ and powered from battery.

Then using ALOSI it is then taken outside of your computer and into your amp/preamp....

And while you sleep, it recharges.
 
Last edited:
Sorry still not seeing the advantages given the costs, simplest is usually best, and simplest is a bit of wire with appropriate interfaces at each end over the sorts of distances and channel counts hifi uses.

Large installations and outdoor events have gone to ethernet or fibre, but it usually carries a digital protocol (Cobranet, Dante, something like that), here the weight savings compared to a conventional analogue snake are compelling, as is the ease of producing things like broadcast splits.

BTW: The most linear FM discriminators are usually the PLL type which are essentially digital in some respects, for carting audio and control data over glass just digitize it already, it is no harder then this scheme and will almost certainly have better performance.

Cat 5 does actually make a perfectly usable cable for line level audio (4 pairs so you can send 4 channels trivially, and up to 7 channels if you are prepared to play it a bit clever (Look up 'phantom circuit).

The closed circuit guys are usually handling audio as a simple FM carrier, but they do not really care about quality which makes things easier.

It seems to me that all of this complexity is a band aid for kit with poor audio line receivers and drivers, far cheaper to put the fix in there then in some kind of powered 'wonder wire' which still has to connect via those sub standard interfaces.

There is probably much to be said for just moving audio as ethernet frames and converting only once you reach the final power amplifiers.

Regards, Dan.
 
Gentlemen.

I have a really selfish reason for joining this discussion. I have an audio system of my own that does require about 120dB of dynamic range connection link. I presently use power amplifiers about 50feet (15meters) remote from the speakers. I am planning to use my Behringer Model DCX2496 Loudspeaker management system that my wife gave me for Christmas last year to drive 6 power amplifiers to Tri-Amp my K2 Klipsch Horns.

Since my Heathkit AP1800 preamp (modified for low noise performance, to be sure) is located so far away, I need to ship the Stereo Output to the Behringer without introducing noise and hum. The system is normally used at the 100mW level for normal listening, I can drive the Klipsch-Horns to the 50W level without distortion. That is a 27dB differential. Since the horns are 35% efficient I need a very low noise link at 27dB below full scale. 120dB - 27dB = 93dB, which is about the theoretical SNR of a 16 bit system.

Yo see, The reason is that I leave amplifiers and systems on all the time and when things are "turned off", I expect there to be no noise or hum to be heard when you put your ear to the speaker. Bumps and pops when other systems come on is OK.

Having worked professionally on 120dB dynamic range systems for over 35 years, I do not want to do worse in my home Audio system. I know it can be done, because I have done it in Medical CT scanners for years. ( I have some US patents on the technology)

I just want to tap the experience of you super critical guys to do a "design review" on my approach. My regular sea of engineers blank out when I talk to them about this sort of stuff. The "dynamic range" of run of the mill engineers stops at 17 bits, and blank out at 24 bits.

Thanks for your highly appreciated comments.

Hans J Weedon
 
Last edited:
The Creative Science Centre - by Dr Jonathan P. Hare
THE LUXEON
http://www.modulatedlight.org/Modulated_Light_DX/OpticalComms4Amateur79.html


CAUTION!, written in 1978: Simpler transmission systems, requiring less than, say, 50 KHz bandwidth, may use any readily modulated light source. Short range systems may employ light emitting diode sources which, except for the green or white phosphor-activated types, have a linear modulation characteristic, and are readily internally modulated at low voltages. Long range systems could use modulated high intensity gas discharge lamps, carbon arcs, gas lasers or solid state lasers. A coherent (ie. laser) light source is not mandatory, and may prove to be economically unjustified where bandwidth and limiting daylight is not a critical consideration.

A resurgence of interest in optical communication came with the rapid advances in lasers and optoelectronics after 1960. In 1962, television signals were transmitted 18 miles using a modulated infra-red beam generated by a GaAs diode, prior to the general availability of the laser. The all-time distance record for terrestrial optical communication with speech modulation was set on 3-4 May 1963, when a 632.8 nanometer helium-neon laser beam was transmitted 118 miles by W6POP and W6QYY, from a point in the San Gabriel Mountains near Pasadena to Panamint ridge near Death Valley, California. An amplitude modulated 10-metre amateur radio transmitter was used for energising the laser.

http://www.modulatedlight.org/eos/Operation_Red_Line.html

Initially, we had no idea how to do this, but after some experimentation we determined that applying RF to strips of foil properly placed on the outside of the tube did the job.
 
Last edited:
I can't quite see the connection between experimental long-distance free-space optical communication (distortion maybe 10%, frequency response 300-3kHz, S/N 10dB?) and high quality audio interconnections. Modulating light is easy. Doing it to high quality is not.

Gee I dunno, they managed to put a TV signal over one... over 101km...

cTfpW.jpg


Pretty good picture too.

http://www.darc.de/index.php?id=24358
Translation: http://translate.google.com/transla...otf=1&u=http://www.darc.de/index.php?id=24358

Translated project page:
http://translate.google.com/transla...layout=2&eotf=1&u=http://www.mydarc.de/dj1wf/

http://home.arcor.de/W_Fritz/Laser/BluRay/BluRayeng.html
http://modulatedlight.org/optical_comms/LED_AM_video_link.html
 
Last edited:
Video needs bandwidth. Audio needs low distortion. Easy to swap one for the other using digital techniques; harder for analogue. A decent wideband FM discriminator can achieve about 0.5% distortion and 80dB S/N. It gets hard to do much better than this. Most people would want their interconnects to do considerably better than this.
 
Last edited:
I initially desired an AM modulation, feel free to brainstorm with that descision.

I can't see why we can't do a simple + to - inversion then convert V to I for driving a laser diode via current modulation, the voltage of the curve of a rising sinewave signal would be represented as a wide range of current modulation and the noise reduction would be the inversion of the signal, where quieter audio would be represented by a brighter laser diode and louder audio would be represented by a darker laser diode.

Surely there can't be that much noise on a TOSLINK /sealed/ plastic fiber?
There are other true-glass fiber optic cables out there too.

There would of course be range limits for representing a 1v p-p signal and testing to be done for a wide array of different laser diodes and resulting frequency response tests that need to be done, and resulting solutions to that problem.

A set of guidelines like for example the maximum length of a cable for a given desired/guaranteed noise level and signal level on the other side.

And it would have to be 1 channel per fiber optic cable.

Anyway I'm off to bed, have a good one.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.