The VL53L1x stopped after 13 days of continuous running. It didn't detect the target. After the HW power cycle, it works fine.

[SJ.11]

Hi Sir,

I am testing the Pololu VL53L1x with TI Controller.

The test detects and counts the people ( normally 300 to 500 per day) in the test unit which is a 24x7 functioning unit.

I never turned off the power of the sensor. it works for continuously 13 days. after that, it fails.

Once power cycled the sensor it works well.

Is that an FW in sensor problem? or something else I have to do.

Kindly provide suggestion

[gbm]

I would rather assume it to be the famous I2C hang problem. You may find some discussion on this and the method of recovering from I2C comm hangup on this forum.

[SJ.11]

@gbm​ : Thanks for the swift reply sir. Can you share any specific thread link related to I2C hang problem in VL53L1x?

[John E KVAM]

The sensor should not stop - unless you tell it to.

So it's either noise getting into the sensor and causeing a reboot or the I2C bus hanging is vastly more likely. Google "I2C bus stuck low". The problem is pretty famous.

Both these issues are due to noise - and keep in mind your long I2C traces act like an antenna.

Shielding, stronger pullups, decoupling caps, shorter traces, wider traces with more isolation - are all things to consider.

It's also possible the sensor had a power glitch. This would cause a partial reboot.

Noise on the XShut pin would do the same thing.

I'd pull that up kind of hard.

• john