F7 temperature laboratory compare just for fun

Thank you for posting your experiments, I enjoy this a lot. It would be interesting to chart individual differences between watches as well.

Some years ago I kept a Fenix 1, 2, 3 and 5 and a couple of Tempe sensors in a standard household refrigerator for half a day and there was 2 degrees Celcius variance between the watches internal temp sensors. They were located close together, but recently I saw a video of someone on temperature differences inside a refrigerator and that still might have been a factor. Fenix 1 and 2 were only accurate to 1 degree as far as I recall, so not too useful in that comparison. I used no reference, just compared them amongst themselves. You really need proper lab stuff for things like this!

Thank you for these tests!!!! Very interesting!
If you have the opportunity to test with the watch on your wrist (when the situation allows it) to be able to test if the body temperature disturbs the watch sensor.
Thanks again!

The temp sensor should be right next to the barometer sensor in the barometer cavity, since it is used to compensate barometer temperature drift.

For the Fenix 3, it looks like in this image, the temperature probe is the spiral thingy (at least that is what I was told by a vendor when I needed a replacement)

In the Fenix 5 it looked different but I did not disassemble that part of it, and I haven't opened up any 6 or 7's (yet).

Definitely the body temperature affect the sensor. No need to test it. Watch on the hand shows 31°C in ambient 21°C temperature. That's the reason why Garmin says you should take it off for 20 to 30 minutes to get most accurate temperature.

Thanks. I thought that. So this could maybe make some "error" in some scenarios. Now I'm waiting if Garmin repair the temp drifting (ticket created) and then I'll test it again, but maybe only in ambient temperature so that my colleague doesn't complain that I'm blocking his temp chamber :-).

Just for update. I got second F7Pro and compared the temperatures. It´s still drifting, but the F7Pro shows about +0.5°C higher value. So it seems perfectly accurate!

It would be good to know at what point body temp no longer influences the reading of the watch. For instance, when I'm out training here in the tropics, temperatures easily start from 32­°C/89.6°F in the shade and slowly work its way up to 40°C/104­°F sometimes peaking at 42°C/107.6­°C in direct sunlight. I use the onboard temp and sometimes the Tempe's for reference during heat acclimatization training.   

Maybe it will, but I don't want to sit outside in the sun with calibrated thermometer I think, it could be around 30°C which is +- measured temp on wrist