• State Suggested Answer
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 2 subscribers
  • Views 2159 views
  • Users 0 members are here
Options
Related

GPS signal loss mid run

darnich
over 4 years ago

I did a trail run this morning and although the first 5K were recorded correctly the watch lost signal for a great part afterwards, mainly between 5K to 9K. It then recovered for the most part but you can still see that there are some long straight lines in the activity around 10.5K and 11.5K, while the part from 12K to 13K is completely off track. I'm on the latest beta fw13.81 and used smart recording with gps+glonass. I always delete the activities after the watch syncs and usually connect the watch to sync with garmin express almost daily. What may have caused this? I own the watch for two years and this was the first time.

Here is the activity link: connect.garmin.com/.../6609388270

Thanks

  • 0 over 4 years ago

    I'd like to add that the weather was clear and the vegetation is not that dense that would obstruct the signal for so long

  • 0 over 4 years ago in reply to darnich

    Hard to tell. It happens to me up and then too. I'd tell the likely culprit is an outdated ephemeris file. They are generally valid for 4 hours, so if the file was not sucessfully updated directly at the beginning of the Activity, it may have expired during it.

    The EPO is being downloaded over Garmin Connect when you activate the GPS button to start the activity. The problem is often, that it happens just as you are stepping out of the door, having the phone still connected to your home WiFi, but losing the connection shortly thereafter as you move away. Before the phone stops trying to reconnect the WiFi and switches to the mobile data, the attempt to download the fresh ephemeris file fails, leaving you with the expired data.

    When I do not forget, I use to connect the watch over the cable to a PC with Garmin Express, just before leaving, to make sure it gets the fresh EPO, and then I turn off the WiFi on my phone, before starting the activity. That guarantees having a fresh EPO.

  • 0 over 4 years ago in reply to trux

    Thanks for replying, I had read your advice for the wifi in one of your previous answers and ever since I try to remember and switch to data before starting an activity. I will try to connect to Garmin Express beforehand as well, as you suggest. I also switched to every second recording to see if that helps.

    What buffles me however is why it happens mid run, perhaps it is a corrupted part of the ephimeris that concerns a different set of satelites

  • 0 over 4 years ago in reply to darnich
    darnich said:
    What buffles me however is why it happens mid run, perhaps it is a corrupted part of the ephimeris that concerns a different set of satelites

    As I wrote, the ephemeris validity time is 14,400 s (4 h), more precisely, WikiPedia tells:

    "An ephemeris is valid for only four hours; an almanac is valid with little dilution of precision for up to two weeks.[7] "

    So if it was last updated 3 hrs before the start of the activity, then it indeed might have expired one hour after the start.

    The ephemeris data is also transmitted in the satellite messages, but typically it takes up to several tens of minutes before all the necessary data is collected. I am not really an expert in GPS, but as far as I understood, each satellite sends only its own ephemeris data. And since the visible satellites change with the time, and with your position (especially if in an accidented terrain), the device has to pull new ephemeris data from individual newly appearing satellites again and again, potentionally causing the drop of signal, or position errors.

  • 0 over 4 years ago in reply to trux

    Yes, that's a possible explanation

  • 0 over 4 years ago in reply to trux

    Just plugged the instinct in my PC, and tried to read the EPO file's timestamp before it got overwritten (should have killed Garmin express, so that it would not update) and managed to see that the filesize was 0. So maybe it was a corrupted EPO after all.