[time-nuts] Re: Datum PRS-50 monitoring software

Magnus Danielson magnus at rubidium.se
Fri Dec 30 23:24:13 UTC 2022


Hi,

The protocol is described in the 4040A/4065A manuals [1]. Does not look 
too hard to implement, as it's an fixed format ASCII.

As you look into the manual [1], it has an interesting feature in that 
you can hook up multiple clocks on the same RS-232 chain. It's 
illustrated in the manual. That way one can monitor multiple clocks 
using one serial port. Data is separated, as each "line" of data from a 
clock has the serial number in it.

My python-script [2] is for the EFOS-A/B to those clocks, but I've not 
come around to adapt it to the FTS 4065 style cesium protocol, but it's 
certainly on my list of things to do.

My python script pulls data out of the serial port and dumps it into in 
influxdb database. On the other end of that Influx database I use 
Grafana. I strongly recommend using that basic approach. A learning 
experience is to convert data into SI base units as it goes into the 
InfluxDB, since Grafana becomes messier to fix that after the fact. My 
script still has that problem, but as I restart it I aim to fix that.

Cheers,
Magnus

[1] http://www.rabel.org/archives/Symmetricom/4065c-manual.pdf

[2] https://github.com/tfritzon/masermon

On 2022-12-30 19:43, Stewart Cobb via time-nuts wrote:
> In theory the Monitor3 program can automatically figure out the correct
> serial port settings to talk to the clock. I've used Monitor3 to talk to a
> sick Datum 4040A cesium, and in my experience the auto serial settings
> feature rarely worked. The 4040A defaults to very strange serial settings.
> IIRC it's 2400 baud 7-E-1. I believe the PRS-50 is based on the same cesium
> module as the 4040A and 4065A. You might try entering those serial settings
> manually into Monitor3 and see if it connects. If that doesn't work, I
> think I have a copy of Monitor2 somewhere, which is supposedly the proper
> control program for the 4040A. I never found a copy of the original
> monitor.exe. But I don't think any of them do anything fancier than just
> implementing the comms protocol in the manual. Someone could probably do
> that in Python or similar pretty quickly, and then we wouldn't need to
> worry about obsolete DOS programs.
>
> Cheers!
> --Stu
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe send an email to time-nuts-leave at lists.febo.com




More information about the Time-nuts_lists.febo.com mailing list