Related
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Track run calibration?

nonparametric
over 3 years ago

It seems like Track Run is not working for me... I followed the Garmin instructions to calibrate: I ran 4 laps in lane 1 on a 400 m oval track, and then saved the activity. Here is the GPS track from the calibration run:

This is supposed to create a calibration file for the track, but on the watch the RunningTrack folder is still empty. Then I ran an interval workout (2x5x400m), with 1:30sec walking recovery periods in between each interval, and got this as the recorded GPS track:

This seems less than stellar, considering that I was in lane 1 for the entire workout, and the review on DC Rainmaker makes me think that it should snap my position to the actual track geometry. Any suggestions?

  • 0 over 3 years ago

    Are you sure you started the "Track Run" activity and not a regular "Run"?

  • 0 over 3 years ago in reply to tmk2

    I was in Track Run, but thanks for the suggestion.

    This morning I did another run on the same track... 16 laps this time thinking that longer might give it more data to calibrate.

    What's odd is that it did seem to lock on to the track after 2 laps, but there's still no file in the RunningTrack folder on my watch after saving the activity and syncing with Garmin Express.

    Any suggestions for what to try?

  • 0 over 3 years ago in reply to nonparametric

    This is strange, what your map shows doesn't match track running, it will look exactly the same as the "track", more flat. Yours look like regular running.

  • 0 over 3 years ago in reply to nonparametric

    Four more calibration runs, still no luck.

    It looks like maybe the track I’m running on is double-radius. Is that a problem for Track Run? I am getting conflicting info. From Garmin support site:

    > “Track Run expects a track with two parallel straightaways and two semi-circular turns of a single radius value”

    https://support.garmin.com/en-US/?faq=eH8V1VwjN338uo8C5PgDJ7

    But then DC Rainmaker’s review, quoting Garmin:

    > “Garmin says that it takes approximately two laps for them to work out the details of your track. They also say their algorithm can correctly detect any track as long as it has two parallel straightaways, and either a single or double-radius curve at two ends.”

    https://www.dcrainmaker.com/2020/09/garmin-forerunner-745-depth-review.html

    Garmin, can you confirm which is correct?

    Has anyone had luck using Track Run mode on a double-radius track?

  • 0 over 3 years ago in reply to nonparametric

    Hello .  I have just replied to the email that you sent to our US Product Support team.

  • 0 over 3 years ago in reply to nonparametric

    My guess is DC Rainmaker was working off a pre-release version, and Garmin realized a track with  double-radius corners didn’t work well. Which is why the official Garmin support page for Track Mode says you have to be in a single-radius track. 

  • 0 over 3 years ago in reply to jaweiss

    At the suggestion of Garmin Support, I repeated some tests after upgrading to the new firmware (5.20). I can report some improvement... after completing a warmup followed by an interval workout, there is now a file stored in the RunningTrack folder on my watch. I have not been able to get to this point before. However, it is pretty clear from the recorded track that the watch is trying to fit a single-radius track model to a double-radius track:

    Note that it is "cutting corners" and then overshooting at the midpoint of each radius. (I stayed in Lane 1 for the duration of the workout.)

    This leads me to believe that Track Run does indeed have a problem with double radius tracks, which is unfortunate as I suspect these comprise a huge percentage of tracks worldwide (double radius tracks are designed to be able to accommodate a soccer/football pitch in the middle). However, I have not yet gotten any definite confirmation of this from Garmin, and I will update when I do.

    Lastly, despite fitting a round peg to an oval hole, the overall distance recorded is fairly accurate. I started and stopped on the same line on the track and made a conscious effort to run the measured line (30cm in from the lane 1 line), for a predicted distance of 18 x 400m = 7200m, while the reported distance is 7220m, an overestimate of 0.28% which is quite good in my opinion.

    For reference, the steps I took were as follows:

    1. Upgrade firmware to 5.20
    2. Restore default activity settings for Track Run
    3. Go for a Track Run
    4. Allow GPS to “soak”—stand on start line of the track for 1 min after GPS signal turns green
    5. Start a Track Run activity in lane 1
    6. Run 4 laps as a warmup
    7. Run an interval workout without stopping the activity
    8. Save the activity
    9. Sync with Garmin Connect via bluetooth

    The interval workout was 2x5x400m, walking 1:30 during the recovery periods.

  • 0 over 3 years ago in reply to Garmin-Kevin

    Thanks Kevin. I was hoping the Garmin documentation was simply out of date, or was being overly conservative in terms of its requirements. Thank you for confirming that is not the case, and I hope Track Run is able to (better) support double-radius tracks in a future update.

  • 0 over 2 years ago in reply to nonparametric

    Hi folks. Sorry to wake this up after so long, but seeing similar problems. Second run on a track today, which I thought I'd calibrated properly, but it's all over the shop. Not sure if I've got a double radius situation, or just that the track mode is useless... May try resetting and calibrating again next time I go, but any advice would be really helpful.