Issues with VL53L3CX in Simple Ranging app

[SidPrice]

I am trying to use the VL53L3CX with an STM32G070 for water level sensing. The app is based upon the Simple Ranging example.

What I am seeing is that the distance read by the sensor alternates between two values, quite far apart. The sensor is mounted in a stable position for this testing, a little less than an inch from a flat white target.

Pointers on how to debug this would be much appreciated.

Sid

[John_Kvam]

this is an interesting one. if the sensor is pointed at a white target one inch away, I'd claim there was no chance of finding a target at a farther distance. 

But physics can be tricky. If the target were translucent to 940nm, then maybe a second target is possible. Plastic is sometime transparent to 940, but generally not white. 

We need more experimentation. Can you switch to white paper? What if the target is 6" away.

Something very unexpected is going on. 

My guess is there are several paths the photons are taking, but without seeing the experiment, it's hard to guess what is going on. 

Does a black target have the same effect?

Sorry for not having a solution, but we need more data. 

- john

[SidPrice]

Thank you for the reply John.

For other reasons my client has decided to switch to a different TOF device, VL53L1X, and I have ordered a couple of VL531X-SATEL breakout boards. If I experience similar results I will for sure check back in.

FYI: the target I was using was plain white paper, and I saw similar results over longer distances.

Again, thank you,

Sid

[John_Kvam · Accepted solution]

The VL53L3CX is the same ranging engine as the VL53L1CB. They both use histograms. 

The VL53L1X is interesting in that it does NOT use histograms. So, you can only get one target. 

(The L1X also has a lens - for getting more photons so it can range farther.)

The non-histogram chips have a lot to offer. Simplicity for one. And as they work completely autonomously you can set the sensor to only alert you when something comes below a threshold.

But if you really want to dig into what is going on, buy the Nucleo-F401RE board. (It's only 10 dollars.) You can wire up your breakout boards (any VL53 sensor) and run the Graphical User code you download from ST's website. It will take a graph of the results, and you can really see what is going on. And it has a logging feature allowing you to really dig into the data.

Good luck, 

- john

[SidPrice]

Good points, well taken. I have a couple of Nucleo-F401RE boards around, so that's an attractive path to discover baseline performance.

Sid