[evla-sw-discuss] TC11IB ASC Port

[Bill Sahr]

-------- Original Message --------
Subject: Re: [evla-sw-discuss] TC11IB ASC Port
Date: Thu, 06 Mar 2003 14:00:47 -0700
From: Bill Sahr <bsahr at cv3.cv.nrao.edu>
Reply-To: bsahr at cv3.cv.nrao.edu
Organization: National Radio Astronomy Observatory
To: George Peck <gpeck at revere.aoc.nrao.edu>
References: <200303062036.h26KacYt019191 at login.aoc.nrao.edu>

If module designers have not already created designs which
make use of the ASC port, or if a rework of such designs
would be a minimal sort of effort then I would definitely
endorse using RS-232 signal levels on the ASC port.

Of necessity, the ATI port of Nucleus was done to the
Triboard (the development board), and not to the MIB
prototype board.  To everyone's surprise, it is the ASC
port rather than the modem port (the 2nd serial interface
included on the TC11IB chip) which is brought out to a
DB9 connector on the Triboard.  The modem port is
implemented only as a header on the TriBoard.  ATI made
the reasonable decision to use the ASC port as the
serial port.  It is over the ASC port that the rtos
does its serial communications, and it is over the ASC
port that Nucleus Shell accepts commands and produces
its output (Excepting, of course, a telnet connection
to the Shell.  For that case, Shell I/O is automatically
redirected to the ethernet port.)  The ASC port is,
therefore, the "natural" port for debugging via a serial
connection.  The ASC port is also the port that must 
be used for downloads that are done via a serial port
rather than over the network.  There is the further
consideration that we have advertised a serial port as
available for interfacing devices that cannot be interfaced
via SPI or the GPIO lines.  The modem port rather than the
ASC port is more likely to satisfy the requirements
of such devices.

So, in summary, if it does not cause dislocations for the
Electronics Division, I would advocate converting the ASC 
port LVDS lines to single-ended RS232 levels and and I futher
suggest that this port be reserved as a debug and download 
port while continuing to make the modem port available for 
device interfacing.  The 2nd portion of my suggestion would
make the ASC port unavailable as a hardware device interface.

Bill



George Peck wrote:
> 
> Wayne and I are trying to decide whether or not to
> bring the ASC port out of the MIB using RS-232 levels.
> It seems that there may be more interest in doing this
> than we originally thought.  In the prototype MIB board,
> we converted these signals to LVDS.  It looks like a
> change to RS-232 levels could be made on the next iteration
> of the board, without too much difficulty.
> 
> I would appreciate any comments and opinions that any of
> you may have about this.
> 
> Thank you,
> 
> George
> _______________________________________________
> evla-sw-discuss mailing list
> evla-sw-discuss at listmgr.cv.nrao.edu
> http://listmgr.cv.nrao.edu/mailman/listinfo/evla-sw-discuss