[Laser] optical communications, new developments

[Tim Toast]

I was forwarding this on behalf of Stuart Wisher G8CYW. He should 
be on the email list soon, i sent him the sign-up url.

``````````````````````````````````````````````````````


optical communications, new developments
Wednesday, October 13, 2010 8:05 AM
From: "stuart.wisher at talktalk.net" <stuart.wisher at talktalk.net>

Tim,


Thanks very much for getting all the optical stuff together. I do 
not know where in the world you are, but you are doing an excellent 
job. 


I have been experimenting with optical comms at times throughout my
life, and two years or so ago, started using a 1mW laser and PWM in 
a similar manner to G0MRF. I then searched the web before I was aware 
of your site and got stuff from the VK's, K3PGP and KA7OEI, to name 
the most useful sites.


I started with a 1mW laser and PWM as already stated, with a small 
group of local amateurs here in NE England. Then tired of the QRM 
from streetlights and cars, noticed that VK7MJ's modulated light
receiver had a frequency response up in the tens of kilohertz region, 
so we then built 25kHz FM rigs (4046 TX, NE566pll RX) which both 
avoided the 100Hz hum and made up for some of the scintillation 
effect by limiting it out. We then got fed up of aiming problems 
with lasers and went to LEDs. This went fine and contacts out to 
15km were made. I must point out that here in this damp land with no 
dry desert air and no mountains to perch on we are not in a position 
to challenge on distance and the core group have as our aim, 
immediate microphone to loudspeaker communications. 


I have now built six different heads, two AM complete rigs (one PWM 
one linear), several stand alone FM rigs, and then.....


One of our group is keen on microwave communications, and has a whole
array of transverters.This gave me the idea to make an "optical
 transverter" to enable an FT817 on 3.6MHz to operate an LED and 
receive a signal from a head unit which has now progressed to a
 modification of Clint's design. Locally five of these have been 
made, three more are under construction, and they are proving very 
good, giving us FM and SSB contacts currently up to 34km, but at 
very high signal strengths so they should go a lot further.


Just last week I saw your latest additional link to the Italian 
research report on reverse biassed LEDs as photo diodes. I tried 
to repeat this work and found one LED in the junk box that operated, 
not in avalanche mode, but as a normal photodiode when biassed to 
about -45V. I was limited due the unavailabilty of anything higher 
than 50V to hand.


I then converted an old K3PGP head to use this LED and incorporated
 switching to make it transmit on forward bias, receive on reverse 
bias. The only thing was that I found it would work with no bias at 
all.


This has now been tested over a short range of 6.5km that we use, 
and a new member of our group who wanted to test his newly completed
 transverter was unwittingly used as test pilot.
Early indications are that this works as well as our two separate
 receive and transmit systems, and is the first true optical 
transceiver using the same device as RX and TX known to me. 


What follows is a report on tests carried out using a two fresnel 
lens array (the "big rig") and the LED transceiver using a 100mm 
dia glass lens last night.


I hope you find this useful, it would not have been possible without
 your vigilance in spotting the research report in the first case.


If you are interested I have sound clips and photos, I do not have 
a website so this may be a problem?

Thanks,

Stuart Wisher G8CYW





A memorable Night in optical communication.


First optical contacts from G8CYW (home QTH) to G4MSF (Throckley 
Fell) 6.5km distance.


As most of you know, Keith G4MSF has been constructing opto gear to 
my design. We arranged to test it over the 6.5km path last night
 12.10.10. 


Since I have worked this path several times I thought I would 
combine this first contact for Keith with a few little tests of 
my own.


Firstly I would try to use the "big rig" (2 fresnel lenses, 28cm 
by 20cm, A4 "page magnifiers"), since on its last outing Gordon 
G8PNN heard the TX LED at 34km but I could not get the RX to work 
(I forgot to remove the protective photo diode cap as I told you
 recently).


Secondly, since I mentioned the Italian research report on using LEDs 
as sensors, I have been secretly working away at making a true LED
 transceiver, the one Nick said would be "cool" if it could be made 
the work forward biassed on transmit and reverse biassed on receive.


I started with a spare receive head (mk3 out of six I have made) 
which is a modification of K3PGPs design on the web. It has a 
smaller pcb and had a bit of room in the box for "extras".


Firstly, I retested it with my beacon as a benchmark. Second, I
 converted the photo diode stage to KA7OEIs RC coupled reverse 
biassed design and retested, same result with my beacon.


Next some major surgery (butchery really) to the pcb to allow the
 addition of a type 47 relay to switch the LED round from RX to TX, 
a 4049 "dead bug" style to control the relay and a few extra sockets 
and components. A 4mm socket for ptt, BNC for transmit, and because 
I was running out of space, a 4.7nF bolt in feedthrough capacitor 
for the anticipated 50V reverse bias, see photo.


The water clear high brightness 5mm LED was installed in place of 
the photo diode making what was mk3 receive head into a full 
transceiver (in my dreams). The LED runs a maximum of 100mA straight 
off a 3025 coin cell when I got it from poundland. I installed a 22 
ohm resistor to make it compatible with the transverter output.


I opted to test the transmit function first, so I set it up and hit 
the ptt. The LED worked, ran about 30mA quiescent and peaked up to 
60mA or so on FM or SSB. I did try a lower value resistor but it 
seemed to overdrive the LED so 22 ohms it was. On test it seemed 
to produce good FM on my stand alone FM optical RX, and on SSB, 
guess what, it sounded just like SSB does on a FM receiver.


I was using my FT817 and transverter and I had the audio turned 
down since I was just transmitting. I then turned my attention to 
the difficult bit, receiving. I had not got my 50V bias gen working 
so I was just going to use the bench psu I had used for the tests 
last week.


I turned up the audio on the FT817 before connecting up the bias and 
to my surprise heard a tone coming from the loudspeaker, surely
 impossible but I had left the beacon on from the earlier tests 
hours before. It was some time before I confirmed it was not direct
 induction from the beacon and was indeed light hitting the 
component side of the board (the LED is on the track side). The 
reverse bias circuit seemed to be open circuit so I could not figure 
out how it was working, but sure enough it was. I put the lid on the 
box and tested it as before. It seemed to work nearly as well as a
 photodiode, I thought I was dreaming.


So there we are I thought, I have a true LED txr, low light level 
on transmit and a little less sensitivity on receive, jack of all
 trades, master of none, I thought.


Last night when we were both set up and ready, Keith saw my big 
rig light, we both assumed it was pointing correctly and proceded 
with our first QSO, strong signals both ways and confirmed my RX 
was aligned to my TX. 5 and 8/9 both ways, after easy alignment 
between us.


I then said I would like to try another rig and Keith guessed what I 
had been up to. I had it already pointed correctly and Keiths comment
 was that instead of being much dimmer, it was much brighter than the
 big rig, what a puzzle. Keith gave me some SSB and immediately it 
came thundering out of my receiver, using an LED and no bias in an
 incomplete circuit. The are two clips of us on FM in the
 contact,....but....I had already placed a 6dB attenuator in front 
of the 100mm lens (cardboard disc with a 50mm dia hole in), then 
with a 25mm hole for 12dB attenuation. So there we were having a 
QSO using effectively a 1 inch diameter lens, using a dim LED on
 transmit and the same LED on receive, one lens, one tube, and 
one head.


I definitely thought I was dreaming by now. I then realised that I 
must have had the big rig misaligned so we went back to it and 
indeed a few degrees of movement and Keith reported a very bright 
indeed signal with totally end stopped both receivers. He was 
obviously unaware of just how bright these things look when 
pointing in the right direction. There are short FM and SSB clips 
of these contacts.


So there we have it, The big rig lives up to expectations, and the 
LED txr just amazes me. Well done to Keith, I have seen the quality 
of his handiwork and his finderscopes excelled in being virtually 
spot on from the word go, making the whole process very easy to 
line up.


I still need to know how on earth the LED is working in this 
circuit, the cathode goes nowhere, except that I put a capacitor 
from cathode to 0V to limit the transient rise of reverse bias 
going from transmit to receive. A 12k resistor then goes to the
 unconnected feedthrough capacitor. The anode goes to a 2.2meg 
resistor to 0V, the junction of these two components is also 
connected to a 1nF capacitor to couple the signal to the fet gate. 
I can only think that the LED is self biassed by building up charge 
in the cathode capacitor.


Answers on a postcard please.....


Further contact news.....


Keith and Jeff are going to explore the lizard on Thurs, Jeff has
 spotted that my home location is not as on Nicks maps and we may 
indeed have a LOS path.