[ML-Hams] APRS / Digital Comm

[Tim N8DEU]

> On Tue, 29 Mar 2016, Jake Polatty wrote:
> > I have a GE Delta that I set aside for something like this.  It would 
> > take some modification, but it would work great as a APRS 
> > gateway/repeater.
> 
> > > Would anyone be interested in an "APRS / Digital Communication via
> > > Ham" infrastructure class / presentation?
> > 
> > Do any of us have equipment capable of getting a two way communication going? :)
> 
> I strongly suggest you involve Tim Cunningham, N8DEU. Tim has more or 
> less lead the implementation of APRS in the area for the better part of 
> a decade. 
> 

The Hunstville Hamfest has carried a introduction to APRS and whats new each year for the past 15 years or more.


> At this stage of the game, more power isn't needed. APRS is a 
> single-frequency network, and there are multiple nodes with overlapping 
> coverage areas.
> 

Think of it like a cellular telephone network.


> The I-Gate requires an internet connection, but a digipeater doesn't. 
> 3.4 GHz is the better route to go for IP connectivity since it's not 
> shared with anyone except USAF. YAAC & AGWPE are possible solutions for 
> that, but one has to watch the packet traffic on 144.39 MHz to make sure 
> that packets aren't being repeated excessively. It is entirely possible 
> to configure AGWPE to act as a digipeater while connected to an 
> OpenTracker acting as both a KISS TNC and a digipeater, resulting in two 
> sets of repeats, one from the OT and one from the AGWPE software. That 
> is not a desirable condition on a slow-speed simplex channel. 
> 
> I'd suggest finding one of the historical channels (145.01, 145.65, 
> 144.97, 145.05) and doing some packet from TNC A to TNC B while watching 
> from TNC C to get a feel for how things work. There is likely a local 
> BBS to connect to, or at least Winlink running somewhere in the area. 
> 
> http;//www.harc.net/programs/Winlink2000_RMS%20Express-Intro-.pdf

Local Huntsville activity for Winlink is on 145.09 Mhz.

> 
> There are also caveats from radio to radio and TNC to TNC with respect 
> to TXDelay values (330 - 500 ms) as well as amplitude values of the 1200 
> Hz and 2200 Hz tones. Some TNCs won't reliably decode if one tone is 
> louder than the other, but this is a different problem than the TX 
> audio.
> 

Many of the new radios can perform down to 125ms TXDelay times. This encompasses tarnsmitter onj, freqencie stable, and output up to power. You also have to take into account somebody using a squelch circuit ahead of their audio, so setting a TXDelay below 200ms may not get received by some stations using a squelch circuit ahead of the audio output. It is best to connect audio ahead of the squelch to avoid this added delay on the receive end.


> I'd also suggest reaching out to someone with the test equipment to set 
> deviation and check the levels of both tones with a given radio to make 
> sure that they are more or less level. The pre-emphasis of PM and FM 

Pre-emphasis and de-emphasis is critical and generally the reason packets are not decoded. You do not want to bypass the preemphasis or de-empahsis circuitry for 1200 baud AFSK packet. It is important to make sure the TNC (if using a TNC) does not add pre-emhpasis or de-emphasis to the transmitted or received signal. If it does, you will need to modify the TNC or use a different audio path in the radio to bypass it. This is generally the reason a packet is no decoded. It was either transmitted with no pre-emphasis added or double pre-emphasis added. The ssame is true on the decoding end with de-emphasis.

Pre-emphasis adds a 6db octave upward slope to frequencioes above 2122 hz and de-emphasis adds a 6db octave slope to frequencies below 2122 hz.


> tend to start kicking in around 2KHz and excessive pre-emphasis results 
> in no decoded packets on those XR2212-based TNCs. (Ergo, if you have a 
> true-FM radio, you shouldn't have issues provided you're using the RX 
> audio output and not the discriminator output.)
> 

3.5 khz to 4.5 khz is OK. Most satellite users run about 3.5 Khz to reduce Doppler shift tuning effects when working fast moving spacecraft.


> David Kelly, N4HHE, may also be able to help. He has extensive 
> experience in this area as well and can tell you more about op-amps and 
> radio impedance matching than you care to know.
> 

Ground loops are big deal if you are using a TNC or external hardware. I connect no ground wires between my TNC and radio equipment and let the ground make it's path to commonly connected power supply. I have not had any lightning damage for over 20 years since this change at home or at any digipeater site. Otherwise, you might want to use isolation transformers or infra-red switching circuitry for audio and PTT signals. Ground loops are easy to resolve. If you have audio hum or lightning damaging equipment you will know where to start resolving issues as the weakest is always what gets damaged.


Tim - N8DEU


> --
> Kris Kirby, KE4AHR
> Disinformation Architect, Systems Mangler, & Network Mismanager
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.makerslocal.org/pipermail/hams/attachments/20160329/0f4e0979/attachment.html>