You are reading a single comment by @Gordon and its replies.
Click here to read the full conversation.
Espruino is a JavaScript interpreter for low-power Microcontrollers. This site is both a support community for Espruino and a place to share what you are working on.
@Gordon
You'd expect them to differ by a few degrees - the nRF52 datasheet says at around the 25 degrees C mark it should vary by a maximum of 2.5 degrees C in either direction. However what you're seeing there does look particularly extreme - and it's odd that one of them changes temperature over time while the other doesn't move at all... That looks a bit fishy to me.
It seems most IC manufacturers factory trim the actual sensitivity of the sensor, but not the temperature offset which they leave up to the user... So if you calibrated just the offset yourself, they'd end up being reasonably accurate, but out of the box they're not amazing.
I've actually just added a
Puck.magTemp()function (which will be in the cutting edge 'travis' builds, or firmware 1v93 when it's released). That allows you to check the temperature according to the magnetometer - but if I'm honest that looks even more inaccurate than the microcontroller!