gandalfg8--- via time-nuts
2018-09-20 13:55:10 UTC
I'm not sure, but to borrow from Orwell, I do suspect that some Samsung UCCM modules may be more equal than others, so here's a warning to the curious, sorry, M R James that time:-)........
Anyways, after Greg posted information about these earlier this month I bought one of the STP2945LF versions from Ebay auction 332792680396 and the one that arrived a couple of days ago is the actual item currently pictured in that auction, condition is very good and a flatstrip interface connector was an unexpected bonus so certainly no problem there.
Whether or not I've just missed something and ended up reinventing the wheel I don't know, but after first believing wrongly that this unit must be faulty here is what I've found so far......
The supply voltage seems to be quite critical, inasmuch as the alarm light is on if the supply is below approx 5.5 volts, it certainly doesn't like 5 volts, and it comes on again if the supply is increased to just over 6 volts. At the upper limit there's no immediate obvious other effect but at the bottom limit the unit seems to drop immediately into holdover, this from observing plots on a Pendulum counter.
Even when the supply is ok the alarm light is amber rather than green but not sure what that's trying to tell me. Trying different antennas and even a resistive load at the antenna socket makes no difference.
At the moment Lady Heather doesn't seem to be hapy with this one, despite forcing baud rate and receiver type she just didn't want to know at first and insisted there was no data to be seen from the COM port, and initially the unit itself didn't seem to be doing anything very useful either
However, running it into Hyperterminal did show it booting up at 57600, 8-N-1 as expected, and eventually I discovered that this unit has two options for Reference type, GPS or EXT, EXT presumably "external" and I assumed referring to what was supposed to be feeding whatever this was used in, and it defaults to EXT whenever the unit is started or after a hard or soft reset. Fortunately it's possible to run a "Ref : Type GPS" command and the activity LED quickly confirmed that it's now behaving as expected.
If I run a diagnostic:loop command though it's so far always returned a figure of +1.062E-7 for the OCXO but also states that EXT is "unavailable" which might even suggest that the EXT input is an expected input to the GPSDO itself.
Purely speculation of course, and the 1PPS is definitely an output at the moment...
Next oddment, trying yet again to let Lady H see it I left the GPSDO running and terminated the connection in hyperterminal and started Lady H again with the receiver preset as UCCM, still nothing until I brought up the command line help screen and then Lady H started taking data in behind it only to stop again when I closed the help screen.
This was very erratic though, sometimes she'd stay after the help screen was closed but sometimes lock and sometimes randomly updating, either way she certainly wasn't having a good day:-)
What Lady H did manage to do was to switch the output to the continuous data stream mode and the only way I've found so far of stopping this is a hardware or software reset, either of which then sets the Reference typ to EXT again:-(
So that's it for now, still playing but I did think it was worth alerting other potential buyers as to what they might expect.
Not all doom and gloom though, and believe it or not I'm even inclined to recommend it, Lady H has shown over a dozen sats with good signal levels at times, that's pretty remarkable for here, and it does lock quickly and conditioning looks good. It would be interesting to see if holdover improves though, and if it could be persuaded to default to GPS at start up that might perhaps be a bit more useful:-)
time-nuts mailing list -- firstname.lastname@example.org
To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.