Hello All,
We are unable to get direct support from Invensense, perhaps we can get some help over here. We have designed (actually moved from ST) 20649 in our application that works for sports in high-g applications. We choose this part due to its high dynamic range in acceleration and rotation.
We have designed and shipping product but we encounter an issue in perhaps 20% of the chips out there. Under high G (>800, <1500g), one of the channels in accelerometer stucks. Literally. It always show 30g. Restarting the IC (we have had provisions to cut power and reapply), or reset (via registers) doesn't help. The only way the situation reverse itself is, if the product is applied another shock, than that channel gets unstuck and life is back to normal.
To me, this seems like a reliability or poor design issue and I wanted to hear from the fellow designers. Did you have such issue and how did you over come it? I am really hoping my startup sequence etc. is wrong. Since we are not supplied any reference code, I am at a loss if our code is solid or not.
Please let me know.