I'm sending at most a single 9 byte command per second. I currently wait for the TX reg empty flag, which means I'm sending them all back-to-back. I'd have to instead wait for the TX complete flag and then add a delay after that. It's doable, but it would astonish me if 9 bytes in a row were causing it heartburn.
I'd love it if there were some way to restore the unit to factory settings / firmware. I'm on the cusp of sending an email to FEI to see if they do anything besides laugh at me. Sent from my iPhone > On Apr 29, 2016, at 7:05 AM, Chuck Harris <[email protected]> wrote: > > Likely what is happening is you are overrunning the RS232 > interrupt buffer, and that is causing a write into code > space. Odd that the firmware is changing, though. > > Does this thing have Flash RAM, or battery backed NVRAM (Dallas)? > > Oftentimes, NVRAM using devices use the NVRAM for both firmware, > and data RAM space. Things can get really ugly when that is done. > > Also, to test my guess, put a delay between each character > sent to the FE-5680A. Say, 2 or 3 ms. > > Needless to say before you do this, you need to develop a way > of restoring the firmware. > > -Chuck Harris > > Bert Kehren via time-nuts wrote: >> Sadly yes >> The reason we have not released our GPSDO for the FE5690/50 and FE405 is >> that we have experienced the same on all three devices. You are the 9th case >> that I know of. We know very little as to what exactly causes it but we are >> incorporating circuitry to prevent it. What we know is that it is not a >> particular code sequence on the RS232 port, but what happens the operating >> code is turning to mush. Some how the RS port is involved. Serious flaw on >> all FE devices since most likely the code was written by the same >> individual. >> Bert Kehren >> >> >> In a message dated 4/28/2016 9:02:28 P.M. Eastern Daylight Time, >> [email protected] writes: >> >> Just as I started testing the GPS discipline board prototype, my FE-5680A >> seems to have developed a very odd problem. >> >> If I give it power, it outputs a “kinda” 10 MHz sine wave while it sweeps >> around looking for a physics lock. This is as expected. >> >> About the time that I would expect the lock light to turn on, instead the >> output just stops. The lock light never comes on. >> >> This unit’s been working fine for months now. It’s conceivable that I’ve >> sent it some sort of serial command it couldn’t digest, but using the >> Windows calibrator software seems to work - albeit there not being any >> output >> from the oscillator at all. >> >> >> _______________________________________________ >> time-nuts mailing list -- [email protected] >> To unsubscribe, go to >> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. >> _______________________________________________ >> time-nuts mailing list -- [email protected] >> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. > _______________________________________________ > time-nuts mailing list -- [email protected] > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
