Grade detection during a ride?

9360578
1 month ago

I’ve been using an edge 1050 for a while now and while I’m overall quite happy with it, the grade detection seems way off.

For example, earlier I was doing a ride with some climbs and when cliimbpro kicks in, the grade field seems extremely slow to change and rarely gets above 8-9% even on climbs I know to be far more than that.

Sure enough when I got home and look at the activity in Garmin connect, the max grade is shown in the climbs as 19.5%, theres a number of 16-17% climbs, but I never saw more than 11-12% on the screen during it.  So somewhere the correct data is being collected, but its not being shown in real-time.

Now I know this is a first world problem since with my dodgy knees, 19.5% is at my absolute limits, and its basically everything I have to keep moving, so why should I care if its not showing on a screen when all I want to do is stop and vomit.

However, this is supposedly a flagship product and I expect more.  Is there something I’m missing?

Any help  appreciated?

  • 0 1 month ago

    Seriously speaking, in the garmin world, you can be absolutely happy when the computer shows positive gradient in a climb.

    Not so long ago it showed very consistently negative gradients even when you've climbed 10%+ for an hour.

    So if you climb and it shows positive values, that is the best you will ever get here.

  • 0 1 month ago

    Bumping this in the hope that theres a better answer than thank my lucky stars my flagship bike computer is hopelessly incorrect?

  • 0 1 month ago in reply to 9360578

     as far as I am concerned, it is technically impossible to create a reliable inclinometer. 

    gyroscope based sensors are sensitive to sudden changes in acceleration (that happens when you pedal), so every bike computer uses a barometer and GPS combination to detect the grade.

    the problem is that neither is perfect, so your computer must guess the grade from a *** of noisy data.

    if you smooth out the data, we will complain the significant lag in gradient (it happened), if less the smoothing the calculated gradient could be nonsensical sometimes (it happened, also).

    to insult the injury, garmin fired all engineers and hired purple hair vibe coders instead and outsourced the QA to the antarctica penguins, so that's that

    the best you can do is to upgrade the software frequently (because garmin randomly changes the parameters of the grade calculation back and forth so in some firmware it is way better than in other), and when you think it is not that bad, then stop upgrading anymore.

    I followed this strategy with the 1040, and now I have somewhat acceptable grades.

    that being said, the current accuracy of the grade calculation is about on a level that I can guess the grade myself just by my eye, so it is kind of a useless feature now ;)

  • 0 1 month ago

    Garmin have been screwing up the gradient response and accuracy since the 1030 series. Its one of the reasons I switched to Wahoo as my main Head Unit

  • 0 1 month ago in reply to pxl

    While I agree with your observation it can improve dramatically when incorporating data from a speed sensor. Some IQ-Datafields do this. I think Garmin should do this too, when a speedsensor is available. At least a step forward from GPS-Data.

    It would be possible to incorporate a real inclinometer in Edge units. Mounting angle of the Garmin can be automatically adjusted based on GPS/SpeedSensor/Barometer values over some time. Would be easy and accurate