[time-nuts] Re: Seeking info on Ericsson NCD 901 56/1 devices

[John Miller]

My units came in yesterday and I have had good luck with them, so far. Very fast lock.
Interfacing with them is easy enough. The pinout on the osmocom site is correct indeed,
and RS422<->UART devices are actually not required - the voltage levels are 3v3, so
standard USB-UART devices connected to the "+" sides of Rx and Tx work just fine.

If you connect a standard T-568B ethernet cable to the jack, the pinout is:

(2) Solid Orange - PPS
(4) Solid Blue - Tx (Data from GPS)
(6) Solid Green - Rx (Data to GPS)
(7) Brown White - 12v
(8) Solid Brown - GND

Each RJ45 connector outputs identical signals and has identical pinouts.

The default UART data is sent 9600 baud, 8n1. The data sent is a curious Ericsson 
protocol, and I have not been able to find complete documentation on it as of yet. 
I believe that the document that may contain the specs for this protocol is titled 
"IWD L2 L3 GPS Digital Interface" but I have not been able to find reference to it other
than in a thesis written in 2012. 

Every second, two sentences are sent. $PERC,GPsts which is alleged to be the GPS
status, specifically if it's in acquisition mode, surveying, position-hold, or surveyed in. I have
had one unit running for over 12 hours continuously and it has not left "position-hold"
mode. The second sentence sent is $PERC,GPppr, a "periodic PPS report" which 
includes GPS TOW, GPS Week, number of satellites used to compute the fix, and
maybe more - the documentation is incomplete.

Every 60 seconds it sends a handful of standard NMEA sentences: GNGSA, GPGSA,
GLGSA, GPLSV, GPGSV, and GPGGA. A few other Ericsson PERC sentences are
sent as well. an "averaged position" sentence, and a sentence with date & time.

I don't yet know if there are any commands we can send to the unit to enable it to 
begin a survey or alter the strings transmitted. I imagine that there are but we need 
to get our hands on the protocol documentation first.

At this point, my suspicion is that to get the best value for money out of these things,
one should tap the UART & PPS signals directly off of the RES SMT 360 module itself and
bypass the SMT32 entirely.  


Regards,
John
KC1QLN



> On Aug 15, 2022, at 8:53:16 PM, Peter Bell via time-nuts <time-nuts at lists.febo.com> wrote:
> 
>  There is also another version of these Ericsson GPS receivers floating
> around - this one is marked "GRU 04 01" - physical construction is very
> similar, but this one appears to be manufactured by Trimble.
> 
> 
> 
> Operationally, it's almost identical to the GPS 0301 described earlier, and
> produces mostly the same output, except that it has an extra
> $PERC,GPpps sentence that the earlier model didn't.  From the look of the
> board, it's using the same (or very similar) GPS module, but
> with modified internal firmware so that it directly outputs the Ericsson
> messages rather than having to have the STM32 to translate the output.
> 
> https://i.ibb.co/VmSLFSP/IMG-3625.jpg
> https://i.ibb.co/gg86jQw/IMG-3626.jpg
> https://i.ibb.co/7R1TBR0/IMG-3627.jpg
> 
> From a hackability point of view, the GPS 0301 looks like a better bet.
> 
> Regards,
> 
> Pete Bell
> 
> 
> On Wed, Aug 10, 2022 at 9:39 AM John Miller via time-nuts <
> time-nuts at lists.febo.com> wrote:
> 
>> Hey all,
>> Does anyone know anything about these "Ericsson NCD 901 56/1 GPS 03 01"
>> devices? They are all over ebay and are fairly cheap.
>> 
>> https://www.ebay.com/itm/185337146688
>> 
>> https://www.ebay.com/itm/294113323964
>> 
>> https://www.ebay.com/itm/194342806640
>> 
>> They have one SMA connector and two RJ45, probably RS485? I haven't been
>> able to unearth any documentation on them at all. A few pictures of boxes
>> suggests they were made in 2016-2017 in Poland.
>> 
>> Any insights?
>> 
>> Thanks!
>> John
>> KC1QLN
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at lists.febo.com
>> To unsubscribe send an email to time-nuts-leave at lists.febo.com
>> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe send an email to time-nuts-leave at lists.febo.com