Related
Options

F7 temperature laboratory compare just for fun

honza_81
over 1 year ago

So here are the results as I promised. First test was in temp chamber. The value on watch is drifting (in all cases) in range of 0.3°C, so I selected the value, which holds most of the time.

   

So It´s the same story as with BAROMETER. If we have possibility to set an offset, it could work maybe a bit better.

Next test was done at one temperature in water bath. Etalon value was 25.2°C vs. watch 24.7°C so again 0.5°C lower value.

   

I also put the watch in empty glass with temp probe and waited at least 30 minutes. One day I got 0.1°C lower value, next day I got exact value and today I got 0.4°C lower value. This measurement is not consistent, but the temp probe reacts almost immediately with changing temperature, so maybe that could be one of the reasons. The watch has more inertia.

     

The drifting value suprised me a bit. Garmin could add some averaging algorythm of last 10 measured values for example. The watch is changing a value in 0.3°C range very fast.

  • 0 over 1 year ago in reply to honza_81

    Thanks for posting your experiments!

    (Joking: Maybe an idea for Garmin to build in calibration routines for pressure and temperature to perform at the factory. And a printed calibration report packaged in the box. Awesome!)

    jan.shim's comment made me think, the opposite question basically; if and when does body temp stop affecting water temperature during open water swimming (without wetsuit) in cold water, but that depends for a large part on the physical properties/response to cold of the person wearing the watch, so not a good experimental design even if 100 people participate.

  • 0 over 1 year ago in reply to Data
    Data said:
    when does body temp stop affecting water temperature during open water swimming (without wetsuit) in cold water

    I’d say almost immediately. I’ve been swimming through winter here in NZ with water temp (according to the watch) going down to 10C. The temperature showing on the watch will quickly drop 10C as soon I get in the water and lose the rest within a minute or so. I have the watch directly on the wrist not over the wetsuit. Certainly, after an hour swimming I’d be confident the watch is showing the water temperature largely unaffected by my skin temperature. 

  • 0 over 1 year ago in reply to Dr Phil

    Good to know that it settles so quickly. I keep my time in 10-15 C water limited, and check on the watch what the water temperature is after 10 minutes to determine when I should get out. Always wondered if the reading after 10 minutes was still influenced by body temp in a significant way, but that seems not to be the case.

    I will test my Fenix 7 in a container with water and ice to see if it shows 0 C, or if it has an offset.

  • 0 over 1 year ago

    So, the standard accuracy for most thermocouples is +/- 1 degree C, and that is before you combine the uncertainty of an associated display.  This is for ones that are new and/or calibrated regularly.  Considering what is on our wrists is commercial vs. professional use, and not regularly calibrated, I would say 0.5 C is pretty impressive.  Would be nice to see the results of a better sample size for both this and the barometer (which was equally impressive for a comercial, wrist worn device) 

    Would also be nice if Garmin actually published specifications on those measurements, because then we would know what they got for both accuracy and repeatability/reliability.  (Signed - Someone who also works in a lab and has > 30 years of experience with this)

  • 0 over 1 year ago in reply to TWE
    TWE said:
    Would also be nice if Garmin actually published specifications on those measurements

    Agree. And it would be nice if it were know which component is used so we can retrieve the datasheet.