High heart rate values
#11
(07-20-2021, 04:09 AM)danielt Wrote: Tightening the band is unlikely help. Loosening it might... the sensors can work even when rested on the wrist using only the weight of the watch to hold it on your arm.

I didn't have it that tight, but good point. Resting the watch on my arm under its own weight only doesn't help either. But I did find that if I hold the watch and move it up and down off my wrist about once per second I can get it to read 60bpm.
  Reply
#12
(07-20-2021, 08:40 PM)wilsonjw Wrote:
(07-20-2021, 04:09 AM)danielt Wrote: Tightening the band is unlikely help. Loosening it might... the sensors can work even when rested on the wrist using only the weight of the watch to hold it on your arm.

I didn't have it that tight, but good point. Resting the watch on my arm under its own weight only doesn't help either. But I did find that if I hold the watch and move it up and down off my wrist about once per second I can get it to read 60bpm.

If you are moving the watch periodically then the system might just start measuring the periodic movement rather than the heart rate. The current system is pretty primative. It works by filtering out high and low frequencies and discarding "stupidly big" samples (which are usually caused by movement of the watch chassis relative to the arm) and after that it looks for signal auto-correlation to find the lowest frequency periodic signal it can find.

Getting the sensor to work has always been a bit hit and miss, even after I walked though all the documented sensor modes trying to find the most useable signal. In the end that's the reason wasp-os has a graph. It was not added because it looks neat... it is there to help the user see if they are actually getting a periodic signal from the sensor or not (the current algorithm is not so good at saying "none" if the signal from the sensor is junk... and instead often settles on actual HR*1.5 or HR*2).
PineTime: wasp-os and MicroPython, Pinebook Pro:  Debian Bullseye
  Reply
#13
Unfortunately I have no news. I tried to record and download some heart rate data with waspOS according @danielt 's documentation here, but I failed. I managed to flash the new waspOS and set the debug variable to True with wasptool, but after using the graph for a while, I was not able to download the data. I don't remember the error anymore, but I could try again and report back. I checked with a print statement that the debug variable was indeed true, so that wasn't the problem.
  Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)