[etherlab-users] etherlab dc sync check

[Jun Yuan]

Hi Martin,

Is that really neccessary for you to call
ecrt_master_application_time() prior to your main loop? What would
happen if you don't? I'm hungry for the log file with the warning "No
app_time received up to now, but master already active."

> ecrt_master_set_send_interval(pMaster,cycleDuration/1000);

So your master's interval for the FSM is 250us, 4kHz, too.

Let me guess, the typical error of the calculation in the your master
app with 4kHz is 250us, and you typically have a debug log like
"DC 64 bit system time offset calculation: system_time=xxxxx
(corrected with 0), app_time=xxxx, diff= approx. 250000"

The time offset calculation in the current etherlab is theoretically
totally correct, yet failed in the real world of discrete time. The
larger your time interval is, the larger error it has, and hence a
longer time for synchrony.