How to match VL53L1 orientation with RoI X, Y parameters?

[charles23]

My application hopes to detect objects on a flat surface (e.g. table). I hope to place the VL53L1X in a vertical plane, close to the horizontal table top. Then I hope to be able to ignore reflections from the table itself by using the RoI feature. I hope to look for signals only from the "top" half of the 16x16 array - which I hope will be from objects present on top of the table, ignoring any reflections from the table itself.

Q1. Is this a reasonable approach, and might it work?

Q2. What is the orientation of the VL53L1X with respect to the RoI settings? Specifically, if I am running the VL53L1X GUI program and set RoI Left=0, RoI Right=15, RoI Bottom = 8, RoI Top=15, then how should I orient the VL53L1X sensor to observe only objects located above a horizontal table?

(Documentation, including AN5191, does not seem to address the question of "what is up").

[John E KVAM]

As many times as I write 'which side is up' I still cannot explain it so that everyone understands.

But you have the GUI, so you can do it this way...

set your ROI the way you want it.

Then using a bit of paper, move the paper into the field of view, watching the the PC output. Carefully log where the target starts being seen.

Then select the other half and run the same experiment.

You will very quickly figure it out.

The basic problem seems to be that some people make a diagram as if the user were standing behind the sensor, and of course the other folks draw the same thing as view from behind the target. It's further confuseded because the optical engineers know the lens 'flips the data in a mirror image'.

So a practical method is to do it for yourself.

The GUI user manual (available from the 'about' tab says:

Due to the lens presence, when pin1 of the device is placed on the bottom right (see datasheet),

and the device is seen from top, the right part of the receiver appears in the left side of the above

picture, and the bottom part of the receiver will appear on the top of theis (sic) picture.

But I've never found that explaination particularly helpful.

[PWan.1]

Search for UM2356, Figure 7 has the answer to the XY direction question.

[John E KVAM]

It is a reasonable approach, but it might be easier.

If you mount the sensor at a 13 degree tilt, the bottom row will go out about parallel with the table.

The top row will extend out at 27 degrees, but it might be easier to ignore.

If you do it your way, and the table gets a little dirty, you might get unwanted reflections back.

And those reflections might cause an issue.

As for the orientaiton of the part, the best way is to buy the development kit and try it.

This developement is kit is only $56 dollars and you will save vastly more time than this being able to visualize the data and test your positioning. The part is the P-Nucleo-56L1A1 for the VL53L1X and the P-Nucleo-56L1A2 for the VL53L1CB.