VL53L1X orientation and ROI

[SBann.1]

Dear,

I want to implement a people counting algoritme using the VL53L1X ToF sensor. I want to build my own PCB for it. I found out that when the sensor is rotated by 90 degrees (gold filters in a row), the edges of a door are measured better then when I mount it straight (gold filters side by side). I'm using a ROI of 8x16.

Is there any explanation for that, or am I doing something wrong?

Thanks in advance.

Kind regards,

Stijn

[John E KVAM]

I've seen an issue like this, but only with a coverglass. If the coverglass is not exactly parallel with the sensor, you can get some issues wiht symmetry.

But the light goes out in a symmetric cone, and unless you have a tilted coverglass you should get the same answers.

(I'm assuming when you rotated the sensor you switched from an 8x16 to a 16x8.)

Try the same experiment without the coverglass. Or use the crosstalk free glass that was included with the Eval kit. Does that work better?

• john