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

some erratic repetitive occurrence in the speed registration

relevant devices:

- Garmin Edge 830 (bought and used since jan 2022)

- Favero Assioma DUO (bought and used since 2020)

First time posting so I hope you can see the picture clearly enough.  Just in case, I have it also on imgur: https://imgur.com/a/GLbjyTR

I noticed during my last ride home (post club ride drink) there was a repetitive pattern in the speed registration.  The disturbance itself is also the same each time.

Below is a zoom-in on one of them but they are all the same.

During the club ride, I noticed that my 3-seconds wattage data was also not as reliable as it should.  It seemed to follow a similar pattern whereby the value would be too high and then too low, too high and normal again but it was more continuously throughout the ride.  I don't know if this is related but it would be very coincidental.  But afterwards, looking at my power data I don't see any real issues or similar pattern although I must say that it does show a higher up and down pattern for the more stable stretches when comparing to previous rides.

When I look at the speed recording of the club ride, I notice that the recording is not very smooth.

Below you can see for example how it goes up and down slightly.  This is on a straight piece along a channel.  It should have been a very smooth line.

to compare, from another ride where there were no issues, you see how the curve is much more smooth on a similar straight piece

It almost looks as if at some moments, the speed recording is based on the wattage meter instead of gps.

It is also the first time that I rode the bike at below freezing temperature.  Maybe it is just an issue of needing power meter recalibration?  But would it influence the speed?

  • It is the result of the bad code in the firmware and the GPS LAT/LON values are duplicated ("duplicated GPS data" issue, I posted it many months, years ago, all Edge units are affected since many many years and only a speed sensor can "hide" it, the Garmin coders cant fix it "somewhy".) due to low priority for GPS chip. sometimes it is more frequent during a ride, sometimes not, but it is really annoying (especially if You dont use speed sensor on your bike and analyze your ride on Strava, on GC, it is "not problem" because that is unsuitable for 1sec chart analysis...) 

    But if You share your original FIT file (in zip here or in private message) , i can show You the problem from raw datas. 

    "to compare, from another ride where there were no issues, you see how the curve is much more smooth on a similar straight piece"
    It has this problem also just You didnt recognize... Slight smile (not so spectacular, ofc, but those are /the result of/ GPS values duplications...)

  • Thanks for pointing out in that other chart.  So I guess there is no easy solution for this?  It doesn't change with the smart recording vs 1 sec recording?

    I see it now on previous rides as well.. not a smooth ride Smirk

    It does look like I have 2 separate issues.  For my power meter, there are several who reported the readings going awry close to the 1.5~2 year mark.  Very unfortunate :(

  • if You have problem with power values too, that could be any other issue (pmeter-related or other connection-related between pmeter and edge). what kind of powermeter do You have and which protocol is used (ANT+ or BT) for connection? 

    yeah, there is NO SOLUTION... (just the speed sensor which is hiding this problem but not solving it)  With powermeter the recording rate is always 1sec so the "smart" is not working in your case (maybe You can set it manually for "smart" in settings but there will be 1sec recording if pmeter is connected and sensed). but the smart means less data, and less data gives you "less problems" , thats all, but that is not solution neither just harder to notice the issues :) 

    send me in private message a FIT file if You can and i check it.

  • I've been checking it out on my past rides and suspect it could also have something to do with disturbance of the GPS signal.  For example here the disturbance happened after I went through a short tunnel.  It is more pronounced in this case.  It is possible that in some cases there is something in the ground that reflects the signal?  Or when you ride through a forest or between houses that could all influence the signal.

    I have been comparing a couple of rides where I do exactly the same trajectory and at some points I get a fairly consistent reoccurrence.  Power lines, underground gas pipes, crossing water, houses, steep hills.. they could all influence the accuracy.  A few weeks ago, I did a club ride which partially went along a river with small lakes next to it and it shows a similar disturbance along those stretches.

    The power meter got fixed, just needed a recalibration

  • no, it happens on totally clear roads also where no powerlines, no trees, so nothing. Of course, the tunnels and other obstacles can cause gps signal drop but that is not duplication (when there is signal drop, there the GPS values are blank data records, where there is duplication, there are 2 totally same LAT/LON value in 2 rows (so for 2 secs), which is impossible if you (or your bike) is moving, the only thing can cause it the bad coding and low priority for GPS dataflow and the main cpu (or dont know what is the exact design) doesnt get new GPS values within time (so within less than 1000millisec, i dont know what is the exact time limit here, maybe 800ms or similar) and using the last known LAT/LON values again (which cause duplication, and summary issue in "distance" datafield)

    the problem is that the Strava is drawing the speed chart from the change of the "distance" data values  sec by sec (and not using the "speed" data values) and this cause this erratic strava speed chart ...