Re: [gpsd-users] Ublox EVK-M8T

[Denny Page]

Okay,

I can confirm that gpsd doesn’t work with the EVK-M8T. I’ve encountered several 
issues...

The first issue is that the gpsd auto-baud detection doesn’t appear to work 
well with the unit. At 115200, gpsd will settle on 4800 if the unit is in 
binary mode. It’s unclear why. This means that gpsd without the -b option will 
work once, but fail the second time. For the time being, I’ve locked the unit 
at 9600.

The second issue is that a standard serial cable does not work. The EVK has the 
time pulse tied to both pin 1(DCD) and to pin 7 (RTS). I don’t know why they’ve 
also tied it to RTS, but it wreaks havoc on the kernel’s PPS detector, with 
ppstest reporting more than a thousand events per second. A custom cable fixes 
this. I am currently using a cable with just pin 1(DCD), pin 2 (TXD), pin 3 
(RXD), and pin 5 (GND).

The third problem is where things get interesting. While gpsmon run against the 
tty device happily reports a functioning PPS with both the NMEA or UBX 
protocols, gpsd itself does not. Gpsd is rejecting the PPS signal:

Sep 18 23:36:50 k9 gpsd[15453]: gpsd:PROG: KPPS:/dev/ttyS2 assert  
1505777810.100060219, sequence: 8, clear   1505777810.000026753, sequence: 8 - 
using: assert
Sep 18 23:36:50 k9 gpsd[15453]: gpsd:PROG: KPPS:/dev/ttyS2 Assert cycle:  
999962, duration:  100033 @  1505777810.100060219
Sep 18 23:36:50 k9 gpsd[15453]: gpsd:PROG: PPS:/dev/ttyS2 Assert cycle:  
999962, duration:  100033 @  1505777810.100060219
Sep 18 23:36:50 k9 gpsd[15453]: gpsd:PROG: PPS:/dev/ttyS2 Assert rejected 1Hz 
trailing edge
Sep 18 23:36:50 k9 gpsd[15453]: gpsd:PROG: KPPS:/dev/ttyS2 assert  
1505777810.100060219, sequence: 8, clear   1505777811.000005806, sequence: 9 - 
using: clear
Sep 18 23:36:50 k9 gpsd[15453]: gpsd:PROG: KPPS:/dev/ttyS2 Clear cycle:  
999979, duration:  899945 @  1505777811.000005806
Sep 18 23:36:50 k9 gpsd[15453]: gpsd:PROG: PPS:/dev/ttyS2 Clear cycle:  999979, 
duration:  899945 @  1505777811.000005806
Sep 18 23:36:50 k9 gpsd[15453]: gpsd:INFO: PPS:/dev/ttyS2 Clear hooks called 
clock:  1505777811.000005806 real:  1505777811.000000000: accepted
Sep 18 23:36:50 k9 gpsd[15453]: gpsd:PROG: PPS:/dev/ttyS2 Clear accepted @  
1505777811.000005806 offset -0.000005806
Sep 18 23:36:51 k9 gpsd[15453]: gpsd:PROG: KPPS:/dev/ttyS2 assert  
1505777811.100012657, sequence: 9, clear   1505777811.000005806, sequence: 9 - 
using: assert
Sep 18 23:36:51 k9 gpsd[15453]: gpsd:PROG: KPPS:/dev/ttyS2 Assert cycle:  
999952, duration:  100006 @  1505777811.100012657
Sep 18 23:36:51 k9 gpsd[15453]: gpsd:PROG: PPS:/dev/ttyS2 Assert cycle:  
999952, duration:  100006 @  1505777811.100012657
Sep 18 23:36:51 k9 gpsd[15453]: gpsd:PROG: PPS:/dev/ttyS2 Assert rejected 1Hz 
trailing edge
Sep 18 23:36:51 k9 gpsd[15453]: gpsd:PROG: KPPS:/dev/ttyS2 assert  
1505777811.100012657, sequence: 9, clear   1505777812.000028886, sequence: 10 - 
using: clear
Sep 18 23:36:51 k9 gpsd[15453]: gpsd:PROG: KPPS:/dev/ttyS2 Clear cycle: 
1000023, duration:  900016 @  1505777812.000028886
Sep 18 23:36:51 k9 gpsd[15453]: gpsd:PROG: PPS:/dev/ttyS2 Clear cycle: 1000023, 
duration:  900016 @  1505777812.000028886
Sep 18 23:36:51 k9 gpsd[15453]: gpsd:INFO: PPS:/dev/ttyS2 Clear hooks called 
clock:  1505777812.000028886 real:  1505777812.000000000: accepted
Sep 18 23:36:51 k9 gpsd[15453]: gpsd:PROG: PPS:/dev/ttyS2 Clear accepted @  
1505777812.000028886 offset -0.000028886
Sep 18 23:36:52 k9 gpsd[15453]: gpsd:PROG: KPPS:/dev/ttyS2 assert  
1505777812.100033753, sequence: 10, clear   1505777812.000028886, sequence: 10 
- using: assert
Sep 18 23:36:52 k9 gpsd[15453]: gpsd:PROG: KPPS:/dev/ttyS2 Assert cycle: 
1000021, duration:  100004 @  1505777812.100033753
Sep 18 23:36:52 k9 gpsd[15453]: gpsd:PROG: PPS:/dev/ttyS2 Assert cycle: 
1000021, duration:  100004 @  1505777812.100033753
Sep 18 23:36:52 k9 gpsd[15453]: gpsd:PROG: PPS:/dev/ttyS2 Assert rejected 1Hz 
trailing edge
Sep 18 23:36:52 k9 gpsd[15453]: gpsd:PROG: KPPS:/dev/ttyS2 assert  
1505777812.100033753, sequence: 10, clear   1505777813.000005331, sequence: 11 
- using: clear
Sep 18 23:36:52 k9 gpsd[15453]: gpsd:PROG: KPPS:/dev/ttyS2 Clear cycle:  
999976, duration:  899971 @  1505777813.000005331
Sep 18 23:36:52 k9 gpsd[15453]: gpsd:PROG: PPS:/dev/ttyS2 Clear cycle:  999976, 
duration:  899971 @  1505777813.000005331
Sep 18 23:36:52 k9 gpsd[15453]: gpsd:INFO: PPS:/dev/ttyS2 Clear hooks called 
clock:  1505777813.000005331 real:  1505777813.000000000: accepted
Sep 18 23:36:52 k9 gpsd[15453]: gpsd:PROG: PPS:/dev/ttyS2 Clear accepted @  
1505777813.000005331 offset -0.000005331
Sep 18 23:36:53 k9 gpsd[15453]: gpsd:PROG: KPPS:/dev/ttyS2 assert  
1505777813.100069296, sequence: 11, clear   1505777813.000005331, sequence: 11 
- using: assert
Sep 18 23:36:53 k9 gpsd[15453]: gpsd:PROG: KPPS:/dev/ttyS2 Assert cycle: 
1000035, duration:  100063 @  1505777813.100069296
Sep 18 23:36:53 k9 gpsd[15453]: gpsd:PROG: PPS:/dev/ttyS2 Assert cycle: 
1000035, duration:  100063 @  1505777813.100069296
Sep 18 23:36:53 k9 gpsd[15453]: gpsd:PROG: PPS:/dev/ttyS2 Assert rejected 1Hz 
trailing edge
Sep 18 23:36:53 k9 gpsd[15453]: gpsd:PROG: KPPS:/dev/ttyS2 assert  
1505777813.100069296, sequence: 11, clear   1505777814.000002372, sequence: 12 
- using: clear
Sep 18 23:36:53 k9 gpsd[15453]: gpsd:PROG: KPPS:/dev/ttyS2 Clear cycle:  
999997, duration:  899933 @  1505777814.000002372
Sep 18 23:36:53 k9 gpsd[15453]: gpsd:PROG: PPS:/dev/ttyS2 Clear cycle:  999997, 
duration:  899933 @  1505777814.000002372
Sep 18 23:36:53 k9 gpsd[15453]: gpsd:INFO: PPS:/dev/ttyS2 Clear hooks called 
clock:  1505777814.000002372 real:  1505777814.000000000: accepted
Sep 18 23:36:53 k9 gpsd[15453]: gpsd:PROG: PPS:/dev/ttyS2 Clear accepted @  
1505777814.000002372 offset -0.000002372

I haven’t dug into this issue yet. Any guidance welcomed.

Denny



> On Sep 04, 2017, at 09:29, Denny Page <address@hidden> wrote:
> 
> Hey Eric,
> 
> I have an EVK-8MT on the way, which I am planning to use as a timing 
> reference for some NTP/PTP projects. Amongst other things, I am wanting to 
> use it with gpsd. I’ve read about some issues with gpsd and the M8T in the 
> past, so I was wanting to know if anyone was ever successful in using that 
> specific chip. I’m happy to help get it working with gpsd, and can get you 
> remote access to a dev system with the M8T attached for testing if that would 
> help.
> 
> Denny
> 
> 
>> On Aug 31, 2017, at 14:40, Eric S. Raymond <address@hidden> wrote:
>> 
>> Gary E. Miller <address@hidden>:
>>> Yo Denny!
>>> 
>>> On Wed, 30 Aug 2017 22:30:45 -0700
>>> Denny Page <address@hidden> wrote:
>>> 
>>>> Looking back through the mailing lists, it seem that there have been
>>>> some issues with the Ublox M8T chips. Did these ever get sorted? Has
>>>> anyone successfully used the EVK-M8T work with gpsd?
>>> 
>>> I'm unaware of any outstanding issues with those chips.  ublox is a high
>>> priority for us, give it a try.
>> 
>> Gary, I think I might know what he's talking about.  The T is the
>> version that is optimized for time service.  There was some talk on
>> this mailing list (I think) about enhancing gpsd to know about
>> the special features, but uBlox charges a hefty premium for the part
>> and I don't have one, so that never got done.
>> 
>> Denny, we're big fans of the uBlox chips here - I designed a device
>> around them once. We don't know of any issues when they're used as
>> location devices.  Can you be more specific?
>> -- 
>>              <a href="http://www.catb.org/~esr/";>Eric S. Raymond</a>
>> 
>> Please consider contributing to my Patreon page at 
>> https://www.patreon.com/esr
>> so I can keep the invisible wheels of the Internet turning. Give generously -
>> the civilization you save might be your own.
>> 
> 
>