VL53L4CX Sensor Measurement Issue – Incorrect Readings Beyond 2 Meters

[Saidattu]

Hii,

       I am writing to seek support regarding an issue I am facing with the VL53L4CX Time-of-Flight sensor. In my current setup, I’m consistently encountering incorrect distance measurements beyond 2 meters, where the sensor tends to report values that are approximately 75-85 cm more than the actual distance.

Details of the Issue:

• Setup: Indoor environment, with stable power supply, targeting a white sheet-covered wall.
• Timing Budget: Set to the maximum value of 200ms.
• Observations: While the sensor occasionally reports the correct distance, more often than not, the values are incorrect, with a significant error margin of 75-85 cm. The status returned is also zero, indicating no error according to the API.

Steps Already Taken:

1. Environmental Factors: The sensor is used indoors with no direct sunlight or ambient IR interference.
2. Cross-talk Compensation: Using cross-talk compensation function provided in driver.
3. Field of View Adjustments: Limited any possible interference from surrounding objects, but the problem persists.

I would greatly appreciate any guidance or suggestions you might have for addressing this issue, or if there are any firmware updates or additional calibration procedures I should follow.

Looking forward to your expert advice.

[liaifat85]

Even though you're using the sensor indoors with no obvious IR interference, ambient light calibration might still be necessary. Try to recalibrate in the current environment,

[John E KVAM]

Getting ranges that are too far is not a common problem. I don't think I've ever heard of it before. 

You did not mention what you had for a coverglass. One only needs to do the crosstalk calibration if you have the coverglass. 

With a coverglass and NO crosstalk cal, you will get a LOWER range than you should- for distances up to 80cm.

But getting longer distances is kind of hard. Could it be that you did a crosstalk cal, and then removed the coverglass? 

On the L4CX we use a histogram algo, and the crosstalk only affects the distances up to 80cm. After that the crosstalk can be separated from the actual target. 

But that would only affect distances up to 80cm - maybe a bit more. 

So that leaves Field of View. The FoV in rough terms is 1/3 of the distance. At 3M a full FoV target would need to be 1M in diameter. But if you had a sheet-covered wall, I'm sure it's quite large enough. 

Do you get the same result using the Evaluation Kit? The p-Nucleo-53L4A2 is only $56 and using it you can log all the data the sensor outputs. Using that you might be able to see what the full data is showing you. 

Is it possible that even though you have the VL53L4CX, you used the code for the VL53L4CD? The code would run but distances above a couple of meters would be all screwed up. 

I don't have any real answers. Perhaps if you posted some data you collected. But I'd need all the information the sensor returned - not just the range and status. (It's why I suggested going with the evaluation kit.)

- john