[Laser] serial port PDM and PWM

[Art]

OK, good on all Chuck!

At 01:06 PM 12/7/2006, you wrote:
>On 12/7/06, Art <KY1K at verizon.net> wrote:
>>I'm not sure how useful a data link will be for weak signals. And, if
>>the signals are strong, you probably want to just do USB to USB anyway::>
>No, this is not weak signal, it is all LOS, the idea is to look at
>alternate links
>and optical links are a interesting alternative. If I can get 3-5
>miles out of them
>I am good. My next step is to test a 3 mile link between my home and 
>a building
>in down town Tampa.

My guess is that you could do much much faster than 9.6 K over a 3 mile link.


>>But, I did set up tnc's over laser years ago. There is a problem that
>>might not seem obvious to those who haven't tried it.
>>
>>If you use packet tnc's, you will need software for the computers. To
>>the best of my knowledge, there is no kiss based software that will
>>allow you to send a constant stream of beacons or connect attempts.
>>If you use cmd mode, the most often you can make a connect attempt is
>>once per second and once every 10 seconds for beacons. That's becasue
>>these are made for shared channels, and the vendors don't give you
>>absolute control of the duty cycle on trasnmit.
>Here in Florida we are in the process of changing our packet switches to
>Linux based machines, the network is something called FPAC and one
>of the test features is the ability to send a continuous string of data or run
>full dux links. I can make the thing send a beacon once/sec or even faster,
>or I can cause it to send a continous string of packets, the tools have all
>manner of softknobs to twist to I can get a constant stream if I so need.
>For the TNC based test, replace the eprom in the TNC (needs to be a
>TNC 2 type machine) with a 6pack eprom and you can make it do all
>of that fast stuff at whatever timing you need. By using 6pack or KISS
>all of the timing parameters are passed to the PC and so you are no
>longer tied to the TNC internal parameters.

OK, I did not know about FPAC, yes it should be an improvement.

Here in Maine we had a very thoughtless pack node dictator, he drove 
everyone off who tried to use the network, so we all basically gave 
up. SO, I am not aware of the newer advances in packet.

I also don't know of any KISS mode software that allows near 
continuous beacons or long repeating packets to be sent....except for 
the FPAC you mentioned.


>>So, just beware that alignment and signal acquisition will be limited
>>by the hardware and you'll spend long periods standing around waiting
>>for the units to transmit.
>See above, since these links will be full dux anyhow (there should be
>no reason not to run them FDX) I can insert commands in the data stream
>to the far end and control it too.

Yes, agreed.

I also didn't know that I could upgrade to FPAC with a ROM chip-I 
have an old MFJ-1274 that i have run Tiny II firmware in for years. 
With the Tiny II firmware, I lose the 300 baud HF capability, but 
gain alot of immunity from bugs and it have a very nice personal mailbox in it.



>>On the other hand, USB might be more practical. The usb transceivers
>>are simple plug in units with the transmitter LED under one 'bubble'
>>and the receiver under the same type of 'bubble' on opposite ends of
>>a DIP chip package.
>Which USB device are you talking about here? I have thought about
>getting my hands on a couple of the cheap RS-232 comm port devices,
>the ones that in many cases only have TXD/RXD and ground active
>and using two of them as the data source to run the laser TX/RX head.
>
>>The receiver can be used as is.....just focus your receiver lens on
>>the bubble. For the transmitter, I think you can mount a high speed
>>photodiode over the LED bubble to intercept the transmit data, and
>>then use that output to drive your laser transmitter.
>I am interested in the device you are describing, is it like a IR transceiver?


OK, I am sorry-I jumped ship without telling you, and I can see where 
it's confusing.

The hardware transceivers I was talking about are IRDA trasnceivers, not USB.

The USB comments I made are still valid, but the small self contained 
transceivers are IRDA, not USB.





>>I'm not sure how far you can go on USB. But, I know that 10 feet of
>>cable is about as far as you can go when using hardwired links over
>>USB. So, you probably need a honkin' big signal to make USB work from
>>computer to computer.
>
>The first step is just to use the serial port, it could be a USB serial port
>or the real thing (getting rarer and rarer these days). The trick is 
>to get the
>bits into the laser and out of the receive end. Then see just how far it will
>talk at reasonable data link speeds. Remember that most packet stuff
>(for those of you who are still doing it) is fast if it is above
>9k6... So if I can
>get it to go at 96kb then I will be happy. Of course we are running parallel
>radio links, but getting these things going is just part of amateur radio, and
>a part that a lot of hams do not think of, but it is all RF, just in 
>some cases
>high enough to see B=]. Besides if I can get a few miles out of a laser as
>opposed to a RF link well, it is one more path I have available in the event
>that we need to have data links.
>
>Also some will point out the issues with optical links such as rain, fog etc,
>well if the sites are still standing the air after a hurricane (the 
>major issue
>in Florida) is usually clear and rather dry, so as long as the two sites can
>see each other they should be able to pass data. I also figure that the size
>of the piece exposed to the weather is smaller for the optical 
>device as opposed
>to the RF device, antennas in order to get the gain at lower frequencies tend
>to be big wind catchers.
>
>


OK, one issue you might not have considered much was beam steering. 
Believe it or not, those lasers DON'T travel in a straight line. If 
you set up a test range with a 1 milliradian (for example) laser at 1 
kM and set 4 other identical receivers 1 meter away on each axis 
(relative to the center), you can watch the atmosphere steer the 
transmitter laser, even over a moderately short path. It's for this 
reason (I believe) that the very expensive and high throughput 2 way 
data links for laser ALWAYS use a 2 watt laser. If they could 
sacrifice some reliability, they could achieve very good speed with 
10 or 20 milliwatts.

I wonder though....Are you going to use baseband or will you modulate 
this on a carrier to achieve daylight immunity? If you are 
considering doing baseband, you should take a look at Burr-Brown's Ap 
note on using an integrator to cancel constant state daytime light 
levels. Without using a second detector or a second photodiode, you 
can effectively achieve daylight operation without dumbing down the 
receiver sensitivity (which is the common method of achieving 
daylight immunity).

Regards,

Art