Bug: Winter Sport HR Monitoring/Load Calculation Broken

There appears to be an issue with how HR is monitored, and thus, how workout Load is calculated in some or all of the "Winter Sport" activity types. This appears to be a known issue in the Backcountry Skiing community, as there are examples of forum posts about this issue, see below. 

https://forums.garmin.com/outdoor-recreation/outdoor-recreation/f/fenix-6-series/327524/load-values-for-ski-touring

https://www.reddit.com/r/Garmin/comments/tijxb5/is_training_status_load_and_load_focus_accurate/

To test this on my fenix 6 pro, I set up the following experiment--I did a 1 mile/750 ft gain ski tour three times; once on Backcountry Ski, once on a copy of Nordic Ski, and one on  a copy of the MTB activity. I paced my efforts and got about the same time on all three. The load values for the Ski activities was very low considering the nature of the effort, around 5 or 6. It had my AVG HR at 99... There were times when was putting in a higher effort going uphill, and I could physically feel my HR very high (using my hand on my neck), and watch was reporting ~100. On the MTB workout clone, all was well--AVG HR 131, with proper fluctuation. Load of 48.

Why is there such a large discrepancy in the heart rates for what is virtually the same activity. Why does the HR monitor not seem to work in the Backcountry Ski workout type? 

I originally posted this bug in the ConnectIQ forum, but I was directed here. 

https://forums.garmin.com/developer/connect-iq/i/bug-reports/bug-winter-sport-hr-monitoring-load-calculation-broken

  • Please help with this bug / issue. Really bummed to not be able to accurately track my Backcountry skiing for training purposes. 

  • I can't help you more than this.

    You are using poles aren't you? That usually means that fenix 6 doesn't give you a proper HR reading. In fact it doesn't for almost all sports.

    Do yourself a favor and  I use a polar verity sense. This will  give you almost accurate readings without the discomfort of a chest strap.

  • Hi Jose-- I used poles in all of the test cases above, the HR readings while using the "MTB" preset were still more dynamic and accurate. I will do a test with and without poles today or tomorrow, to try to further rule out your hypothesis, but I think the data I have collected so far paired with the communities experiences points toward the protocol in which the device is monitoring and recording HR. 

  • We all know, that wrist HR isn't the most reliable. But it works ok for most people and sports. But not for BC skiing, XC skiing aso. Problem with chest strap isn't comfort. It is about interference between strap and avy beacon. Polar verity sense can also interfere with beacon, so no real solution...

  • But not for BC skiing, XC skiing aso.

    It is not really surprising, since the blood flow in extremities is restricted in cold weather (due to so-called peripheral vasoconstriction), hence making the task really tough for the optical sensor.

    It is about interference between strap and avy beacon

    There should not be any interference as for the broadcasting goes, since they each operate on a completely different bandwidth. Avy beacons operate on 457 kHz, while the strap broadcasts either over Bluetooth or over ANT+ and both of them use the frequency of 2.4 GHz. It is more likely the proximity of the electronics and the metallic wires in the strap (and also the watch self, phone, etc) that can affect the beacon. Generally it is advised that the avalanche transceiver is 50 cm from any electronics or metallic objects. I did no tests, but would guess that the impact of the strap is lower than for example the influence of the phone, or of the watch self. Certainly, testing your configuration in advance would be a wise decision.

  • since the blood flow in extremities is restricted in cold weather (due to so-called peripheral vasoconstriction)

    This doesn't explain why I got markedly better HR data and increased load values doing the exact same thing, 30 min later. It was the same temperature out for all of the activities in the test. The effort was virtually the same, but when using the MTB preset to record the uphill skiing, the HR data was accurate. 

  • This doesn't explain why I got markedly better HR data and increased load values doing the exact same thing, 30 min later

    Already warmed up little bit perhaps? Or the sensor already learned the weak pattern better? Or different filter parameters used in individual activity profiles? Hard to tell.

  • I am planning on rerunning the test in reverse this weekend, so "MTB" ski tour first, followed by a "Backcounty Ski" ski tour. the temps should be around the same as last time, below freezing. If I am feeling up for it I will just keep alternating laps on the two Activity types until I tire out, so we can control for my fatigue. 

    Given what I have already tested, I genuinely think there could be a bug in the protocol/routine that gathers the HR data on the watch's software. Like, it is not grabbing it often enough. All a dev would have to do to triage this is do a side by side review of the code for the MTB protocol and the Ski sport activities protocol. 

  • Thank you a lot for testing. If you want, you can also try an app from connect iq - Skitour App Professional from RH-Sports.

  • If you decode an activity profile using the fit SDK, there is a field called "hr_calc_type". This could indicate that Garmin uses different methods for calculating HR from the sensor raw data. 

    If you send me your MTB and backcountry ski app (from GARMIN/ Sports folder), I could check if the value is different (and maybe apply the MTB value to the backcountry skiing profile). Or of course, you can try yourself Slight smile