Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Astroberry + GPS. No fix.

  • Please log in to reply
10 replies to this topic

#1 TelescopeGreg

TelescopeGreg

    Gemini

  • -----
  • topic starter
  • Posts: 3,219
  • Joined: 16 Jul 2018
  • Loc: Auburn, California, USA

Posted 07 June 2021 - 09:59 PM

Has anyone had any luck getting one of those small GPS modules to get a fix on the GPS satellites on an Astroberry-based Raspberry Pi?  I recently purchased a GNSS GPS/BDS/GLONASS module ( It appears to be recognized by the system, but never gets a fix.

 

I have another Raspbian Buster-based Raspberry Pi 4B that I use it with, for a non-astro application, and it gets a fix there in a few minutes.  Same physical module, positioned in a similar position outside.

 

If not this one, are there any that do work with Astroberry?  My desire is to use it for setting the Pi's clock, as well as providing a location fix for the mount if there is a way to convey it.  (Yes, there are RTCs available for the Pi, but I already had the module...)



#2 gmiller123456

gmiller123456

    Explorer 1

  • -----
  • Posts: 96
  • Joined: 25 Dec 2020

Posted 07 June 2021 - 10:28 PM

The module works pretty much independent of what it's plugged in to, so the fix issue is probably something with the environment (like being near a large metal tripod/pier, telescope tube, and motors).

 

For the time, unless you are adding another wire and software to capture the 1pps pulse, you are way better off using NTP (network time protocol) to set the clock.  The jitter from the serial transition and CPU usage will cause more inaccuracies.  With NTP, the lag might be larger, but it can be measured and accounted for.



#3 TelescopeGreg

TelescopeGreg

    Gemini

  • -----
  • topic starter
  • Posts: 3,219
  • Joined: 16 Jul 2018
  • Loc: Auburn, California, USA

Posted 07 June 2021 - 11:19 PM

The module works pretty much independent of what it's plugged in to, so the fix issue is probably something with the environment (like being near a large metal tripod/pier, telescope tube, and motors).

 

For the time, unless you are adding another wire and software to capture the 1pps pulse, you are way better off using NTP (network time protocol) to set the clock.  The jitter from the serial transition and CPU usage will cause more inaccuracies.  With NTP, the lag might be larger, but it can be measured and accounted for.

NTP, sure.  But when I'm imaging out of range of the home WiFi, it's annoying to not have the clock set right. 

 

The module is physically down on the tripod's eyepiece tray.  If the stuff above it is enough to prevent reception, that's a pretty deaf GPS module...  I might try a USB extension and put it up on top of the scope, just to verify.  The other Pi I normally use it with has it out in the open, positioned sideways.  On the telescope mount, it's vertical in a USB hub.
 



#4 rdw4176

rdw4176

    Sputnik

  • *****
  • Posts: 26
  • Joined: 07 Oct 2012

Posted 08 June 2021 - 10:27 AM

On My setup with Astroberry + external GPS, I had to disable the virtualgps first for my external gps to work, might help in your case too:

1. Update gpsd deamon, edit /etc/default/gpsd script and made:

START_DEAMON="false"
USBAUTO="false"

and also put # in front of the rest...

 

2. Disable vgps

sudo systemctl stop virtualgps
sudo systemctl disable virtualgps

 

3. Reboot



#5 TelescopeGreg

TelescopeGreg

    Gemini

  • -----
  • topic starter
  • Posts: 3,219
  • Joined: 16 Jul 2018
  • Loc: Auburn, California, USA

Posted 08 June 2021 - 03:20 PM

Thanks for the thought, rdw.  I'll keep that in mind, but I think the issue is actually what gmiller said.  Poor view of the sky.  That said, Yikes! this is one deaf GPS module.  I am NOT impressed.

 

I got a short USB extension cable, and positioned the GPS module on top of the scope, and rolled the scope out to a place where it has a better view of the sky (it was partly under a fluffy tree).  After a while, it did get a fix on the sky.  Then lost it.  Then found it again.  And so it went.  Sort of settled down for a while, then lost it again.  Sheesh...

 

So, I think there's nothing about Astroberry that prevents it from working.

 

But, we're not done yet...  Drifting the thread slightly, next issue...

 

During the time the GPS has a fix, running cgps in a terminal window shows the correct lat / lon for my location, just as it did on the other system.  However, indi-gpsd displays the correct latitude, but incorrect longitude.  It's 360 degrees off (shows 238 instead of the expected -121).  Configuring the mount's indi driver (indi-celestron-gps) to use the GPSD location, I see that the odd location is fed across to it.  So, apparently the display is not exactly cosmetic. 

 

Fortunately (I think), the mount still believes it's standing in front of me, i.e. the "updated" location hasn't made it to the hand controller.  So, two questions:

 

1.  Before I log a bug against indi-gpsd, has anyone else seen this?  Is it, in fact, cosmetic (i.e. a valid way to display latitude as a positive number instead of +/- for E/W), or will it mess up my GoTo slews?

 

2.  What is the GPS time / location used for in indi_celestron_gps?  When a slew is commanded, it's the mount hand controller's perception of time and location that is used, and that doesn't seem to be updating, so why does the Celestron indi driver even care about the GPS?


Edited by TelescopeGreg, 08 June 2021 - 03:23 PM.


#6 bips3453

bips3453

    Mariner 2

  • *****
  • Posts: 298
  • Joined: 05 Mar 2020
  • Loc: MN

Posted 08 June 2021 - 03:29 PM

...  It's 360 degrees off (shows 238 instead of the expected -121).  ...

360 deg??? Hmm, isn't that the same location on a circle... !?!



#7 gmiller123456

gmiller123456

    Explorer 1

  • -----
  • Posts: 96
  • Joined: 25 Dec 2020

Posted 08 June 2021 - 06:18 PM

1.  Before I log a bug against indi-gpsd, has anyone else seen this?  Is it, in fact, cosmetic (i.e. a valid way to display latitude as a positive number instead of +/- for E/W), or will it mess up my GoTo slews?

 

It is fairly common to express longitude as 360 degrees as opposed to +/-180.  The best way to see if that will fool your mount is to actually try it.  My bet is, it either rejects it and nothing happens, or it works.  Internally pretty much all software checks to see if the longitude is negative, if it is, subtract it from 360.  So entering a number above 180 will work exactly the same. 



#8 TelescopeGreg

TelescopeGreg

    Gemini

  • -----
  • topic starter
  • Posts: 3,219
  • Joined: 16 Jul 2018
  • Loc: Auburn, California, USA

Posted 08 June 2021 - 08:05 PM

360 deg??? Hmm, isn't that the same location on a circle... !?!

To you and me, of course.  The question is whether such a representation of western longitude is properly understood by the other parts of the Indi subsystem.  I've never seen longitude displayed that way anywhere else.



#9 TelescopeGreg

TelescopeGreg

    Gemini

  • -----
  • topic starter
  • Posts: 3,219
  • Joined: 16 Jul 2018
  • Loc: Auburn, California, USA

Posted 08 June 2021 - 08:36 PM

It is fairly common to express longitude as 360 degrees as opposed to +/-180.  The best way to see if that will fool your mount is to actually try it.  My bet is, it either rejects it and nothing happens, or it works.  Internally pretty much all software checks to see if the longitude is negative, if it is, subtract it from 360.  So entering a number above 180 will work exactly the same. 

Posts crossed...

 

I tried entering "500" to check the input validation, and it took it, so I was wondering what it would do down-stream.  Later I tried it again (after some more fussing with things), and Indi ignored it, saying that it "wasn't necessary" because the mount was "already aligned".  That suggests that there's no impact one way or the other for the GPS location, since I always start the mount with "Quick align", which uses the Hand Controller's version of the location, and I don't think that has been updated by Indi.

 

Still not sure what exactly is going on, but it's becoming clear that this GPS module isn't going to do what I want it to do.  Too deaf.  I put the module back on the eyepiece tray and left it out (under the edge of the fluffy tree) for several hours while I did yard work, and as far as I can tell it never got a lock.  I'll give it one more chance (for time-setting purposes) at the next imaging opportunity, and see if the actual imaging site across the street works any better.  That said, the module did get a lock on the other system when it was sitting inside the house, but near a window.  Got a lock with little difficulty, so apparently the structure of the mount's tripod is a bigger deal than I'm thinking it should.  My USB extension cable isn't long enough to permanently remount the GPS module up on top, and it's frankly not worth buying another one.



#10 bips3453

bips3453

    Mariner 2

  • *****
  • Posts: 298
  • Joined: 05 Mar 2020
  • Loc: MN

Posted 08 June 2021 - 08:45 PM

To you and me, of course.  The question is whether such a representation of western longitude is properly understood by the other parts of the Indi subsystem.  I've never seen longitude displayed that way anywhere else.

Looks like that is a standard INDI behavior: https://www.indilib....1888.html#61888

Should pass correct location to the mount. 


  • TelescopeGreg likes this

#11 TelescopeGreg

TelescopeGreg

    Gemini

  • -----
  • topic starter
  • Posts: 3,219
  • Joined: 16 Jul 2018
  • Loc: Auburn, California, USA

Posted 09 June 2021 - 12:55 AM

Looks like that is a standard INDI behavior: https://www.indilib....1888.html#61888

Should pass correct location to the mount. 

Ah, thanks for finding that.  So all I need to do is convince the fool thing to actually get a fix.




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics