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

My theory on why optical heart rate monitor locks with cadence

Former Member
Former Member
  • optical hrm work by Photoplethysmography (ppg)
  • this is the volumetric measurement of an organ.
  • LED in sensor illuminates the skin; and the amount of light transmitted or reflected is measured; which reflects change in blood volume
  • change in volume --> change in pulse pressure --> heart beat
  • SO! when you run in cold conditions where there is a temperature gradient in between your watch and the ambient running environment
  • this causes liquid vapor from your body heat to condensate onto the lens of your optical watch sensor
  • as a result, the LED transmission and reflection is altered
  • arteries are located deeper in the body than veins, which can be close to the surface of the skin.
  • Photoplethysmogram optically generated from your sensor is going to be off; and instead of detecting blood volume changes in your artery; it is detecting volume changes in your veins.
  • veins are a low pressure system relative to arteries, an the hemodynamics are more easily influenced by arm motion
  • hence, optical sensor now detects vein hemodynamic changes from arm pumping that occurs with every step, causing CADENCE LOCK
  • that's why your HRM locks onto cadence-- it isn't a software bug in your garmin but a hardware limitation.


some questions my friends asked me:

1. How does the PPG, when it's working correctly, know to look at the arteries instead of the veins?

PPG actually uses an averaging algorithm between your veins and arteries. The veins are much less responsive to changes in blood volume; and as noted earlier are a lower pressure system so influenced more by up down arm movements so you get the cadence lock. However, when the reflection and transmission of the light to sensor is altered by condensation, the veins are picked up more than the arteries and are therefore represented as a higher percentage of the average-- this is why when you get cadence lock the HR doesnt coincide directly with the cadence; you still get some averaging in from detected arteries

2. Do you think the condensation is forming on the inside or outside of the watch? If it's inside, isn't the watch supposed to be waterproof? If it's outside, wouldn't it be removed from your skin running on it?

The condensation is forming on the outside of the watch. Your watch is much warmer (assume body temp since it is worn close) than the ambient environment; so it is radiating heat. The atmospheric moisture condenses at a rate greater than that at which it can evaporate, resulting in the formation of water droplets. You can see a similar effect on leaves in the early morning forming dew.

This is also why cadence lock is more commonly experienced on humid days/days with high dew point. If you chill your watch before your runs you can stave off HR cadence lock at least until your watch warms up to body temperature. Do this by leaving your watch overnight near a window sill, or sticking it in a fridge for 1/2 hour.
  • Former Member
    0 Former Member over 8 years ago
    Cool theory, thanks for sharing.
  • Interesting theory.
    Is it backed up by any practical evidence where instances of cadence lock are aligned with the weather and/or the likely "warmth" of your watch?
  • Well.... I would say that the sensor is in close enough proximity to your skin that the temperature gradient is very low and condensation is not the issue in this case.

    My view is that the cold weather causes constrictions in the blood vessels near the surface of your skin which makes the measurement much more difficult the colder your skin is. Constricted vessels have lower flow overall and therefore constrain the ability for the device to measure the differences at much lower overall flows. Just my 2 cents.

    EDIT: The theory regarding cadence lock may still be valid though.
  • the watch is great, but not the OHRM.

    I need to log a 30min workout for my medical aid provider to allocate points based on heart rate - multi-sport does not work as they only take the one highest/longest activity of the multiple. I had hit my goal already for the week so though I would do a short experiment... set watch to indoor run, do a run then a bit of rowing... the distance would be low as the cadence would be much lower during the row... but the heart rate should still be fine?

    I pushed reasonably hard and this work out should probably average 158bpm... but yeah... think the watch was accurate for about 20 seconds around the 17min mark... This is in the gym so no low temperatures to worry about. Heart rate looks awfully close to the cadence?

    https://connect.garmin.com/modern/activity/1423869612
  • Can you change that GC activtiy link to public please so we can see it
  • Former Member
    0 Former Member over 8 years ago
    I pushed reasonably hard and this work out should probably average 158bpm... but yeah... think the watch was accurate for about 20 seconds around the 17min mark... This is in the gym so no low temperatures to worry about. Heart rate looks awfully close to the cadence?

    https://connect.garmin.com/modern/activity/1423869612


    Can't see the activity shared on the link, it's still private.
  • Can't see the activity shared on the link, it's still private.


    Oops, sorry still quite new :)
    I've made it public. So it should work now?

    Here is a run I did after a good 15min warm up on a WATT bike, then straight into this run... HR here is close to spot on.
    https://connect.garmin.com/modern/activity/1403024532 (made it public as well :o )
  • I had the cadence lock even in summer, so it's not weather related in my case.

    But there can be another theory.. OHR sensor is just a special type of camera and cameras are sensitive to shaking when recording. Each step can affect the current capillary sample by small shaking and the algorhitm may be confused by these repetitive changes. Moreover, the capillary widening is weaker in cold, so the algorhitm is predisposed to lock on another periodical anomalies. That can be the reason why I never had the cadence lock while biking...
  • Former Member
    0 Former Member over 8 years ago
    But there can be another theory.. OHR sensor is just a special type of camera and cameras are sensitive to shaking when recording. Each step can affect the current capillary sample by small shaking and the algorhitm may be confused by these repetitive changes. Moreover, the capillary widening is weaker in cold, so the algorhitm is predisposed to lock on another periodical anomalies. That can be the reason why I never had the cadence lock while biking...


    Yeah i think it's something more along these lines.

    When i first got my 735XT i ran home from the shop with it in my running backpack (had a 235 already and wanted to play around and set up the 735XT at home before using it). In the store the sales guy demonstrated the HRM for me but forgot to turn it off. It went back in the box, then snug in my backpack, and then jiggled around on my back for the 20+km run home. I didn't know the watch/HRM was on at the time.

    To my amazement when i got home and pulled the watch out, looking at the HRM widget it showed a HR trace for the last ~2hrs perfectly aligned with my cadence.

    So it's certainly something to do with the camera and a repetitive behaviour that it picks up on in the absence of a stronger signal. May even use the accelerometer, perhaps as a correction, but also susceptible to over-correction in certain cases.