High Speed optical transmissions thru air ?

CounterRotatingProps

{ Pulled out of another thread to avoid confusion --- was http://forums.parallax.com/showthread.php?p=811645·}

I would like to use optics for data transmission - but through air, not a fiber.

Let's say the distance would be < 10 or 20 feet using an IR LED. (Or IR Laser, if they're not to expensive - I'm intentionally avoiding radio.)

1. This is pretty generic, but, how fast can the transmission get before noise and / or device switching speeds becomes a factor ?

2. What parameter(s) in a data sheet should I look at to find a device's optimal speed ?

3. Is 400 Khz to 1Mhz realistic ?

(There's got to be a ton of threads already on this subject, but searchs aren't yielding much ... )

thanks
- H

---

To Which Phil kindly replied:
[noparse][[/noparse]...]Short answer: at those speeds, a photodiode would be a more appropriate receiver.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
No matter how much you push the envelope, it'll still be stationery.

Post Edited (CounterRotatingProps) : 6/2/2009 11:56:23 PM GMT

Humanoido

Some considerations: Seven points. One, correct, there is a ton of scattered information on the subject but it may be dated and slower to find or embedded in Physic papers.

Two, I have run data though the Optoelectronic Interface with OPTICS language for some years, due to its reliability, and ease of use, for remote robotics control for telescopes and observatory systems, though I did not formally publish the design of the OI or language. These designs depend on and are limited by the speed of the receiver sensing element.

Three, for fiber optics, experiment with the system by just removing the fiber cable and use higher powered light sources which remain modulated. There's plenty of data on FO systems.

Four, focus the beam optically for distance transmissions at increased distances. A convex objective lens of flint and crown glass to reduce optical achromatic aberration, whose focal ratio has a center of focus placed at the point source of the light wave transmitter will emit parallel aligned light at infinity. To keep degradation down, use straw coated anti-reflective high precision lenses.

Five, infrared behaves a bit different and the lenses are not the same compared to the lenses for optical wavelengths. The Stamp can handle this effectively but the IR is often more broad band due to this reason, like a TV controller.

Six, at the stamp level, you can get typical high speed baud ranges. Select the Stamp which will do the job. Another unit I built handled 7,000 to 12,000 Hz at optical wavelengths, so these speeds were typical. I was not trying to push it to higher parameters.

Seven, There are inexpensive laser systems at the front end, but you would need to home-brew the remainder of the project finding a way to modulate the signal and decode it to keep the cost down. Hope this helps.

humanoido

Post Edited (humanoido) : 6/3/2009 2:05:03 AM GMT

CounterRotatingProps

Hi Humanoido,

nice to hear from you again. Thanks... these are good points. Interesting that you mention the astronomy aspect... One of the things I'm going with this·is remote telemetry of solar observations [noparse]:)[/noparse]
Actually, I'm planning on using an SX.

Taking the fiber out of a fiber system is a clever idea.

The laser route may be the cheapest actually as I've got dozens of laser pointers I got for $1 at a hardware store sale. Modulating it is straight forward, but I've not tested them for speed yet.
Phil mentions using a photodiode generally. Would this be appropriate for a red laser. I'd like to use IR (just so the beam isn't visable)... what do you think?

thanks
- H

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
No matter how much you push the envelope, it'll still be stationery.

Mike Green

Years ago, Datapoint developed an IR optical link that could handle network (ARCnet) traffic at around 1MBaud over a line-of-sight distance of about 1 mile. It used conventional LEDs and photodiodes with Fresnel lenses around 18" in diameter. I think there was some automatic adjustment of transmission power based on ambient light levels. I don't remember any other details, but it was slick.

They specifically avoided using lasers so they didn't have to deal with regulations on their use.

Mike Green

Here's a link to a project on this subject. The actual project website appears to be down.

en.wikipedia.org/wiki/RONJA

CounterRotatingProps

Wow Mike --- that's some cool stuff --- looks like Science Fair material too!

The way they fire the IR LED's is worth study.

and, from the free-space optical link: "An 8-beam free space optics laser link, rated for 1 Gbit/s at a distance of approximately 2km"

(If I put one of those on my roof, my neighbors will think I've installed a spy cam :-P )

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
No matter how much you push the envelope, it'll still be stationery.

jazzed

Did you know that lazer optic data communication began as an off-shoot from lazer listening conducted by the CIA ?
After the Russians found out what was going on, they demanded a second pane of glass be installed at their embassy.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
--Steve


Propalyzer: Propeller PC Logic Analyzer
http://forums.parallax.com/showthread.php?p=788230

Humanoido

It's amazing what you can do with a simple laser pointer. At one time I bought 25 of these for my projects. It was possible to shoot the laser pointer across the street and all the way down the block to turn off each city light for doing astronomy work at night. Beyond a city block, the pointing sensitivity increased and a tripod mount was added to the pointer but the circle of confusion diverged with atmospheric scattering. They also work well for telescopic finder scopes by beam pointing into the blackness of space. They are unsurpassed for amateur interferometry experiments too. With solar, the lights are not a concern. But for remote communications, it appears this range may fit your requirements and the price is right. There were also numerous articles about building simple LED "laser" systems in past issues of Popular Electronics. Some of these articles may be reproduced on the web. For "powerful enough" lasers, range will be limited to line of site, which is about 24 miles where the curvature of the Earth drops about 6-feet.

humanoido

CounterRotatingProps

jazzed said...
... listening conducted by the CIA ?

...·the Russians found out ... demanded a second pane of glass be installed at their embassy.

Ah, that·would be·interfering with the interferometry? ;-P

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
No matter how much you push the envelope, it'll still be stationery.

Tracy Allen

You can purchase an OPV382 1.5 mW VSCEL IR laser from Digikey for about $8. VSCELs are the champs for high speed optical transmission, up into gHz. That particular VCSEL has a beam divergence of about 6 degrees, and they are easy to collimate. But you can't overdrive VCSELs with as much abandon as an LED.

Some IR LEDs rated 100mA continuous are also specified for light duty cycle with short pulses up to 1 to 2 amps. That is how the RONJA project appears to be driving the HSDL4220 LED. Another one in this catagory is the Osram SFH4550, which concentrates its beam into +/- 3 degrees without any additional optics. And it is only $0.60.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com