USING THE KANTRONICS TNC FOR SATELLITE CPU                  22 DEC 2006
-----------------------------------------------------------------------
                                                                 WB4APR

Now that we have 8 Kantronics TNC's in orbit serving as complete 
stand-alone satellite command and control systems (PCSAT-1,2, ANDE
RAFT and MARScom) here are some notes about how  their telemetry
systems work.

> could you explain the 8 status bits, please?  And the meaning of the 
> last 3 digits? 
> 
> T#529,083,002,036,035,071,10000000,000
>                           ^^^^^^^^ ^^^

This is the standard APRS telemetry frame which can report up to
5 analog inputs, and 8 digital discretes.

In the original 1995 APRS Mic-E MIM module they were a fully independent
8 bit I/O port.  Useful for discrete status bits.  In the Kantronics 
9612+ TNC, they are also a separate 8 bit I/O port (as used on PCSAT-1 
and PCSAT-2) good for both inputs and commanding.  But on the KPC-3+ 
TNC, they are mostly the same port as the analog telemetry pins.  

We find these Kantronics TNC's to be very powerful.  That is why we now 
have a total of 8 of them in space as the primary command/control system 
for several satellites, and use another few dozen in out LAB satellites 
for the students to learn elements of satellite comms.  See all about
our LABSATS on http://www.ew.usna.edu/~bruninga/labsats.html

Actually the KPC-3+ has 8 A/D inputs, but only 2 are brought to the 
connectors.  We bring out 3 more for a total of 5.  But 3 are used 
internally and so that is why we only have 5 channels of telemetry.  
The first two bits are used for RS-232 handshaking inputs, DTR and RTS.  
On RAFT we actually use the DTR line (the first "1" above") as the SIGN 
bit for the battery current sensor.  If it is 1, then the current into 
the battery is negative.  In this case, 002 shows no input current from 
the sun, so the 036 current is out of the battery.

On ANDE, we use both the DTR and RTS pins as a 2 bit counter to indicate 
which of the four sets of telemetry are being transmitted.  We call 
those the FRAME counter bits.  So here is the definition of those 8 bits 
on any KPC-3+ TNC:

   DR5I3210

   Where D and R are the DTS and RTS inputs
   I is used internally?
   5,3,2,1,0 are the 5 analong inputs which show in the telemetry
             frame as T#SSS,000,111,222,333,555

But we have always called them 1,2,3,4,5 in our documents for
simplicity.

So in the 8 discrete bits, the "1"'s and "0"'s besides the first two and 
the internal one, are simply a binary reflection of the same 5 channels  
of telemetry as the actual 5 analog inputs.  If they are above the digital 
threshold, they show as a "1" and if below the threshold, they show as 
a "0".  Since the "I" pin is used internally (something to do with the 
optional real-time clock chip, it is not invovlved in the TELEMETRY 
command.

However, you can see it's value and might be able to still use it as a 
on/off discrete indicator I guess.

Anyway, with the inputs, and outputs, and password protected logon built 
into every TNC, we find them a powerful tool for all kinds of applications, 
not just APRS...

OH!  And the last 3 digits on the end are a memory wash counter.  We are 
looking for SEU hits on the RAM chip.  All unused RAM is painted with a 
bit pattern, and periodically, the TNC checks for bit errors.  It 
acumulates a total  until the next reset.  This was custom code we asked 
Kantronices to add for this mission to give us some "science" data.  But 
there was never enough time to test it completely and we noticed on the 
ground that it would often jump to 999 on its  own even on boot up.  So 
we will probably ignore it...

Bob, WB4APR