As a control, I repeated the test with a GlobalTop PA6H module.
Again, the reference is a Thunderbolt, so it’s not completely clear how much of the wander to attribute to the reference and how much to the DUT. The residual this time is on the order of 60ns - still surprisingly high. The DUT is an AdaFruit ultimate GPS module with its PPS pin directly connected to the TIA’s input. Removed from this dataset were two brief periods where the phase showed E+38 errors - probably missing PPS pulses. I’ve noticed that there are sidereal-daily periods where the PA6H would lose its tiny little mind briefly. On the plus side, the module has the good taste to indicate this by dropping from 3D fix to 2D for the duration. My GPSDOs were able to use that to drop down one “mode” level and reestablish a good lock more quickly. If you zoom in to the data, there are hanging bridges to be found: And more critically, there is no sawtooth correction data that could be had to avoid them. GlobalTop’s sales literature claims that timing firmware is available, but I don’t have any information about availability or the potential to upgrade existing modules. Still, the ADEV is nearly the same as the Venus838. Slightly better at lower tau, likely because it lacks the extreme excursions I found on that module. Just in case the graphs don’t come through this time, I’ll save them to http://www.kfu.com/~nsayer/PA6H-graphs.zip _______________________________________________ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
