Ok, some progress. I went back to my modifications to predict to see what exactly I had done, and one of the mods was to set the serial port stop bits to 2. I recall now that the Basic Stamp that drives my controller logic isn't fast enough to parse the serial stream with only one.
Setting -C stop_bits=2 on the command line brought the rotor to life!
Now the next problem... I notice that Gpredict freezes while tracking is running. No GUI updates, though the tracking (both rig and rotor) keeps going. The rest of the system is fine, nowever, and top doesn't show anything stuck in a spin loop. Disengaging the rotor control (which takes a long time sometimes) brings things back. Feels like a timeout is at fault.
I took a capture of the traffic on the serial port, and it appears that the program is polling the controller for its position (command C2). My controller doesn't support that command, so that's probably what's hanging things up. Poking around I find -C timeout=1 Ha! Now we're looking good...
Final runstring: rotctld -r /dev/ttyS3 -s 9600 -m 603 -C stop_bits=2 -C timeout=1
Thanks again,
Greg KO6TH
Jim Jerzycke wrote:
I would suspect it's using the Yaesu GS-232 protocol.
Jim KQ6EA
On 06/22/2014 06:08 PM, Greg D wrote:
Hi folks,
So all this talk about Gpredict got me to look at updating my own setup, which pre-dated the modern Gpredict capabilites. But being so archaic, I need some help connecting the dots...
I have Gpredict / Hamlib talking to my Yaesu FT-736R radio for Doppler tracking just fine, but the rotor is a problem. My controller is home-brewed, and emulates the subset of the "Yaesu serial protocol" as output from the original DOS version of FODTrack. The syntax is the letter "W" followed by the Azimuth, a space, the Elevation, and terminated with a carriage return (hex 0d). Both the Az and El are whole decimal integers. When I moved to Linux and Predict (the ncurses version) in 2002, I modified the software to copy that same syntax, and all was fine.
Tab forward over a decade to my question: What model "Yaesu controller" was FODTrack thinking it was talking to? I need to find an equivalent model for Hamlib's rotctl daemon, and none of the choices seem to match. They all seem to want to talk to a single rotor (Az or El, not both). Is there an equivalent model, Yaesu or otherwise, that accepts "Waaa eee\n" for Az + El positioning?
Thanks,
Greg KO6TH
Sent via AMSAT-BB@amsat.org. Opinions expressed are those of the author. Not an AMSAT-NA member? Join now to support the amateur satellite program! Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb
Sent via AMSAT-BB@amsat.org. Opinions expressed are those of the author. Not an AMSAT-NA member? Join now to support the amateur satellite program! Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb