[aprssig] Expanded APRS telemetry to 10 or 15 channels?

[spam8mybrain]

Isn't that a violation of the original spec saying the values have to be in the range 0 to 255 (and be integers)? Or are you talking about the values after applying the quadratic coefficients? If so, you still only have one set of coefficients for the overloaded channel. 
    
Wouldn't it be easier to use a second SSID value for the port, so you could transmit a completely separate set of values, coefficients, labels, etc., under the callsign and second SSID?Alternatively, use the !DAO! encoded telemetry in the position report as independent of the original format telemetry message.Just an idea.Andrew, KA2DDO

-------- Original message --------
From: Robert Bruninga <bruninga at usna.edu> 
Date: 4/2/20  14:37  (GMT-05:00) 
To: TAPR APRS Mailing List <aprssig at lists.tapr.org> 
Subject: [aprssig] Expanded APRS telemetry to 10 or 15 channels? 

Expanding APRS telemetry from 5 to ten, 15 or 20 channels and still remaining within the protocol...On PSAT3, I needed 6 A/D inputs instead of the standard 5, so I used two inputs on channel 2.  If the count is above 500 then it means one thing, and if it is below 500 it means another.  On receipt, I can choose one telemetry equation or the other depending on the value above or below 500.This means you could double the 5 channels to 10 if you used 500 as the dividing line.  Or you could have 15 channels if you restricted one input from 0 to 333 and the next between 333 and 666 and the last between 666 and 999.  You can make the transition points be anything that matches the resolution of your sensors, because the thresholds would be part of the decoding equation.  Even if you used four ranges, your precision (from 0-250) is still better than a half percent!Wow, I wish I had thought of this on my earlier PSATs. Having ten or 15 channels of A/D telemetry while still using the standard APRS protocol would have been useful.On the sensor side, you simply switch in bias resistors on each of the two, three, or four sensors per input channel.  Think of it as bank switching the 5 channels...If anyone has use for this trick, let me know.  Steve Dimse is looking at building it into FINDU for trivial plotting... if/when he finds he time...bob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.tapr.org/pipermail/aprssig_lists.tapr.org/attachments/20200402/bd816d85/attachment.html>