Has anyone else seen this issue on Pebble Time 2 with official firmware 4.36.2?
When I start a Run workout, the heart rate is initially accurate and matches another HR monitor. After roughly 20 minutes, the Pebble HR suddenly drops to around 77–82 bpm and then stays in that range for a very long time, even though my actual HR is around 140–150 bpm.
It does not look completely frozen because the value still changes slightly, for example 77 → 80 → 82, but it is clearly wrong.
Things I have already tried:
tighter and looser strap — no difference
stopping running — HR still stays around 80
stopping and restarting the workout — sometimes it still stays around 80 and only recovers much later
comparing with another HR monitor — the other sensor continues to show the expected 140–150 bpm
The interesting part is that I did not have this problem during summer. It started appearing when outdoor temperatures dropped to around 10°C.
Has anyone seen similar behavior, especially in colder weather? Could this be related to the Goodix HR algorithm losing the PPG signal and getting stuck on a low-but-plausible value instead of reporting invalid data?
Just to say that anecdotally I have seen the same recently (two days ago cycling) but I’ve not investigated the issue at all, just seen that the heart rate was not what I expected after climbing a significant hill!
I was doing 4x1200m at a 4:30min pace (you can see the 4 blocks on the blue graph), which should push me to 170-190bpm (yes, I have a high heart rate). Instead it completely glitched out. It’s massively underreporting pretty much the whole time and has long “plateaus” where it reads the identical heart rate for a minute or so.
PS: Yes, I made sure it’s firm against my wrist the whole time. It might have been a little loose for a few seconds here or there but especially during the walk-phases there’s no way the reading should be THAT off.
PPS: While I did the recording using Iron instead of the pebble “workout” app, I did start the latter half way through, which reported equally nonsensical heart rates.
I’ll go back to v4.36.2 (which is the one I used before that worked well) and hope there’s gonna be a fix for this. The way it’s recording right now is entirely worthless data.
Edit: I just noticed I misread the version number. For me v4.38.x is the step that seems to mess things up.
Little update on my case. It seems for me it’s rather the type of training that just breaks the HR sensor. Today I tried to reproduce the issue by doing some short intense training at home, switching between firmware versions in-between sets and every single version broke. Like completely.
I went from “low” heart rate (80 on the first set, 100 on subsequent ones) to high (~170-180) within a few seconds, stayed there for a minute and then rested until I was back to ~100.
The recording showed a constant-ish pulse for a minute (until after I was done with the set) then climbed up the the then already declining pulse (about 130) and slowly went down as I rested.
Here’s the graphs for the first two sets (first is v4.33.1, second v4.36.2):
Weirdly enough during my interval training (previous post) I was at a high heart rate for way longer than a minute, so I don’t think the issue is time related. more something like “if delta between old and new heart rate is greater than 20, ignore”.
Huh, this might explain why I’ve had it sporadically do the same thing (HIIT workout with resistance bands). It’s only happened once, but I suspect the delta in heart rate might have caused this (my workout today showed the correct values).
