This forum is run by the BATC (British Amateur Television Club), it is service made freely available to all interested parties, please do not abuse this privilege.
I can see that it has been quite a while since I last posted my thoughts on ATV over optical.I decided to opt for simplicity and keep the whole project analogue so that it does not involve using a computer. I was a little concerned that the frequency (11MHz) might be excessive for LED modulation but my concerns appear to have been unfounded. Since then I have built a couple of prototype units using the LA72910V mod/de-mod chips.I was surprised at the quality of the signal which I was able to achieve when the mod chip is directly fed into the demod chip, there appears to be very little degradation. The carrier is approx 11.5MHz (although this can vary from batch to batch) and the deviation is about 3MHz. I am using emphasis /de-emphasis. I have managed to transmit the signal across the room , using a 3watt LED driven from a class A amplifier (2N3866) running at 120mA, which is about the maximum limit with a heat sink attached. The receiver uses a BPX65 near-infrared photo diode, back biased at 12 volts and feeding an LM359 Norton Amplifier.The LED I am using is not the correct wavelength so there is room for experimentation but first results are encouraging.If anyone is interested in trying this, the chips are available from Farnell.
I used this design for Tx, the Rx was initially a pin photodiode and latter a photomultiplier tube with much better results.
Still plan on doing digital here to drop the carrier freq/bandwidth but not got round to it yet.... hopefully sometime soon!
Thanks for the last posting and the links contained therein. I decided to try the receiving circuit with a couple of small mods. To keep the drain voltages more or less at half Vcc I used bias resistors in the source, suitably decoupled.I had very little success using an inductor between TR1 gate 1 and ground and substituted a high value resistor instead. I am now able to get a couple of volts of FM signal (by bouncing the light off the ceiling) to feed to the demodulator. Unfortunately, the photodiode will not work correctly with the lights switched on so will need to find one which filters out visible light.So far as I can see, the biggest problem will be getting clean synch pulses to the monitor. I am wondering whether it might be possible to re-constitute the synch pulses in order to ensure a solid lock .I have not bothered to try using my laser diodes as they are all 650nM wavelength.
The results were as follows:
First the LM359 version receiver: 2 MHz...-18dBm; 3MHz..-18dBm; 4, 5, and 6 MHz...-20dBm ; 7MHz... -21dBm; 8MHz -22dBm and 10MHz...-23dBm. Second harmonics were all below -50dBm.
Now for the 4 stage BF981 receiver. I should add that only resistive loads were used; it would have benefited from the use of shunt peaking.
2MHz.. -21dBm, 3MHz... -23dBm, 4MHz... -27dBm, 6MHz ...-34dBm; 8MHz.. -40DBm, 10MHz... -44dBm.
Whether or not shunt peaking coils would have made the response usable is open to question.
The photo diode i used for these results was the BPW34. The others all gave an inferior performance. What these results don't tell us is the fall off due to the LED characteristics. I am sure that a laser would produce better results but might be more difficult to drive. Also, I am somewhat nervous of their us Equally, there are much better, and more expensive, photodiodes. By the way, it is worth mentioning that great care must be taken to ensure that what we measure is due only to the optical component, it is very easy to pick up RF interference in an enclosed room. For this reason I mounted the receiver boards in metal enclosures and checked that my signal disappeared when the light source was blocked.
I hope you will find these results interesting and would welcome your comments.