DRAWS GPS not seeing satellites #gpsd #gps


 

Up until a couple of days ago, my draws hat GPS was working perfectly. I had it setup to be the time source for my HamClock. I noticed that my HamClock was no longer updating time. So I went to the Draws pi and ran gpsmon. I get no sat fix, even after more than 24 hours. I notice that GPSMON TOFF = >1 day.

When I run gps_util.sh from the n7nix/gps dir I get:

Verify gpsd is returning sentences: Sentence count: 3

Test nmea sentence: GPS data not valid

gps data: $GNGLL,0000.0000,N,00000.0000,E,120126.000,V,N*59

Invalid gps data read from gpsd

Test gpsd sentence: GPS gpsd sentences OK

GPS data not valid

gps data: $GNGLL,0000.0000,N,00000.0000,E,120134.000,V,N*5A

Read Invalid gps data read from gpsd, using canned values

Exiting, gps error

When I run gps_test.sh I receive:

gpsd: 3.23.2~dev (revision 3.23.2~dev-g34066876c)

Warning: Stopping gpsd.service, but it can still be activated by:

            gpsd.socket

Source file found, building

gcc -O2 -g -gstabs -Wall -I/usr/local/include -g -O2 -D_LARGEFILE64_SOURCE -pthread -I/usr/include/gmime-2.6 -I/usr/include/glib-2.0 -I/usr/include/glib-2.0 -DLINUX   -c -o gp_testport.o gp_testport.c

gcc gp_testport.o -o gp_testport

gps_test.sh: version: 1.1, gp_testport: version: 1.1

Running in sat count mode

Fri Nov 26 13:39:14 2021,  wait for first satellite view

It will then sit there waiting for a satellite to come into view, which does not seem to happen.

I have verified this same behavior with image 19 and the new image 20.2.

Buginfo.sh provides the following:

=== Versions ===

== Kernel:

Linux DRAWS 5.10.63-v7l+ #1459 SMP Wed Oct 6 16:41:57 BST 2021 armv7l GNU/Linux

 

== Firmware:

Oct 29 2021 10:47:33

Copyright (c) 2012 Broadcom

version b8a114e5a9877e91ca8f26d1a5ce904b2ad3cf13 (clean) (release) (start)

 

== Pi hardware:

 Pi 4 Model B, Rev 1.1, 4GB mem, Mfg by Sony UK with WiFi

temp=47.2'C

 

== DRAWS hardware:

Product id: 0x0004, ver: 0x0206, Assembly rev: 2, fab rev: 6

 

== image version

2021 11 14 13:08:17 PST: SD image version: nwdr20.2

 

=== Checks ===

== Codec driver check:

 

== DRAWS driver check:

udrc card number line: card 2: udrc [udrc], device 0: bcm2835-i2s-tlv320aic32x4-hifi tlv320aic32x4-hifi-0 [bcm2835-i2s-tlv320aic32x4-hifi tlv320aic32x4-hifi-0]

udrc is sound card #2

 

== Boot 'fail' check:

 

== GPS check:

Serial devices OK

gpsd OK

 

== /boot/config file

framebuffer_width=1920

framebuffer_height=1080

hdmi_force_hotplug=1

hdmi_group=1

hdmi_mode=16

config_hdmi_boost=4

[pi4]

max_framebuffers=2

[all]

dtoverlay=

dtoverlay=draws,alsaname=udrc

force_turbo=1

dtparam=audio=on

pi@DRAWS:~/n7nix/bin $


I have also attached a screen shoot of CGPS and CPSMON.

Any help would be appreciated.

 

Rich, KR4PI

 


Basil Gunn
 

1. Make sure gpsd is running:

systemctl status gpsd
# You should see Active: active (running)

● gpsd.service - GPS (Global Positioning System) Daemon
Loaded: loaded (/lib/systemd/system/gpsd.service; enabled; vendor preset: enabled)
Active: active (running) since Sat 2021-11-20 12:00:31 PST; 6 days ago

2. Check your gps antenna. It must face up toward the sky.
Sounds like you are in your house so make sure it is by a window.
Also check the SMA connector, make sure it is tight.

3. Check the battery voltage of the CR 1220 battery.
Should read 3.3V

Thanks for providing the console output.
It all looks OK.


Rich KR4PI <rich.schnieders@gmail.com> writes:

Up until a couple of days ago, my draws hat GPS was working
perfectly. I had it setup to be the time source for my HamClock. I
noticed that my HamClock was no longer updating time. So I went to the
Draws pi and ran gpsmon. I get no sat fix, even after more than 24
hours. I notice that GPSMON TOFF = >1 day.


 

On Fri, Nov 26, 2021 at 03:41 PM, Basil Gunn wrote:
systemctl status gpsd
pi@DRAWS:~/n7nix $ systemctl status gpsd
● gpsd.service - GPS (Global Positioning System) Daemon
   Loaded: loaded (/lib/systemd/system/gpsd.service; enabled; vendor preset: ena
   Active: active (running) since Fri 2021-11-26 14:50:12 EST; 55min ago
  Process: 643 ExecStart=/usr/local/sbin/gpsd $GPSD_OPTIONS $OPTIONS $DEVICES (c
 Main PID: 645 (gpsd)
    Tasks: 3 (limit: 4915)
   CGroup: /system.slice/gpsd.service
           └─645 /usr/local/sbin/gpsd -n /dev/ttySC0 /dev/pps0
 
Nov 26 14:50:12 DRAWS systemd[1]: Starting GPS (Global Positioning System) Daemo
Nov 26 14:50:12 DRAWS systemd[1]: Started GPS (Global Positioning System) Daemon

GPSD is running, Antenna has been in this position running perfectly for a couple of years.  
I checked the battery and it was very low voltage. I replaced the battery with a new one, but still, GPS does not get a fix. 

Thanks for the help!
Rich, KR4PI


Basil Gunn
 

Not sure what else to try, maybe power it down for a few minutes. Also
are you powering it through a 5.1V wall wart or 12V to the DRAWS hat?
Post the output of the 'sensors' command.

Rich KR4PI <rich.schnieders@gmail.com> writes:

On Fri, Nov 26, 2021 at 03:41 PM, Basil Gunn wrote:


systemctl status gpsd
pi@DRAWS:~/n7nix $ systemctl status gpsd
● gpsd.service - GPS (Global Positioning System) Daemon
Loaded: loaded (/lib/systemd/system/gpsd.service; enabled; vendor preset: ena
Active: active (running) since Fri 2021-11-26 14:50:12 EST; 55min ago
Process: 643 ExecStart=/usr/local/sbin/gpsd $GPSD_OPTIONS $OPTIONS $DEVICES (c
Main PID: 645 (gpsd)
Tasks: 3 (limit: 4915)
CGroup: /system.slice/gpsd.service
└─645 /usr/local/sbin/gpsd -n /dev/ttySC0 /dev/pps0

Nov 26 14:50:12 DRAWS systemd[1]: Starting GPS (Global Positioning System) Daemo
Nov 26 14:50:12 DRAWS systemd[1]: Started GPS (Global Positioning System) Daemon

GPSD is running, Antenna has been in this position running perfectly for a couple of years.
I checked the battery and it was very low voltage. I replaced the battery with a new one, but still, GPS does not get a fix.

Thanks for the help!
Rich, KR4PI


 

power is through 12 volts into the DRAWS hat. What concerns me is that the TOFF: >1 day. 

I will power it off for the night and see if that helps. I assume that is should also take the battery out for the night as well?

Thanks for the help!

Rich, KR4PI


 

here is the output of Sensors:

pi@Draws:~ $ sensors
iio_hwmon-isa-0000
Adapter: ISA adapter
+12V:         13.40 V  
 +5V:          5.07 V  
User ADC 1:    3.00 mV 
User ADC 2:    3.00 mV 
 
cpu_thermal-virtual-0
Adapter: Virtual device
temp1:        +47.2°C  
 
rpi_volt-isa-0000
Adapter: ISA adapter
in0:              N/A 

thanks for the help.

Rich, KR4PI


 

I removed power from the DRAWS hat and removed the battery from it as well. I left it off all night. when I put the battery back in and pluged in power to the DRaws hat, I still could not get a fix on satellites. 
Are there any commands I can run that do a reset of the GPS?

Thaks for the help!

Rich, KR4PI


 

After this procedure how long did you wait for a fix?  It can take a fairly long time to get a fix after a 'cold' start. (15 minutes or more sometimes)

On Sun, Nov 28, 2021 at 10:39 AM Rich KR4PI <rich.schnieders@...> wrote:
I removed power from the DRAWS hat and removed the battery from it as well. I left it off all night. when I put the battery back in and pluged in power to the DRaws hat, I still could not get a fix on satellites. 
Are there any commands I can run that do a reset of the GPS?

Thaks for the help!

Rich, KR4PI



--
John D. Hays
Kingston, WA
K7VE / WRJT-215

 


 

I am now past 3 hours after power on and still no fix. 

Rich, KR4PI


J P Watters
 

The two things, replace the battery with a known good one. 
Then i would check the time, date including the year.

..jpw J P Watters
KC9KKO
Morris, IL USA


 

Problem Solved.  

I had to power the draws pi down and take it outside. On reboot, it began to see the satellites and got a fix. I then moved it back inside to the same place it would not get a fix and on power up it got a fix immediately. TOFF is no longer >1 day. Until I moved it outside the year was off by 5 and TOFF was greater than 1 day. Once I moved it outside that was corrected and now it gets a fix inside no issue. 
I suspect that with the date off by 5 years and the TOFF >1 day it could not get a fix because it did not see the satellites where it expected to see them.

Thanks again for the help and suggestions.

Rich, KR4PI