cancel
Showing results for 
Search instead for 
Did you mean: 

VL6180X trouble picking out the best target object

AGome.7
Associate II

Hello all,

I am confused on some test results that we performed while gauging the VL6180X's performance and am seeking some advice. We are testing various materials with the intention of using the best performing one for the target object in our application. Our application requires very precise measurements within a 30-65mm range.

So in our setup, we are using 2 ToF sensors with a custom made cover glass design which separates the TX and RX side of the sensor, idea being it will minimize crosstalk. Offset and Cross-talk compensation were also performed. We took 6 measurements between 29mm and 61mm and recorded the distance and mcps.

So in this case, we used a white, shiny, high reflectance material as the target object. Results were great, distance reported was within 3mm of the target distance, and the return rate was high. It looked very similar to an experiment ST performed; a distance vs return rate graph in the app notes of the VL6180X.

AGome7_0-1688075316771.png

 

We then performed the same experiment with a black, matte, low reflectance target object. Results were also great. Distance reported was within 3mm, but this time the return rate being very low. Again, it was very similar to ST's experiment.

AGome7_1-1688075642344.png

Ok, so now we have some data establishing a high reflective and low reflective bounds of what to expect. We then did the same experiment with a gray and matte target object that we desire to use, but results were surprisingly terrible. The reported distance was consistently off by about 6mm. Crosstalk could not compensate for this? The return rate is higher than the black material used, yet it performs a lot worse? 

AGome7_2-1688076236957.png

Thinking that the surface finish of the material had an impact to the results, we then polished the gray material so that it has a shiny surface rather matte finish. However, we actually saw that it performed WORSE than a matte finish.

So basically, we are sort of stumped as to why the material we would like to use for our target object is not performing as well as we believed it would. Any insight on how color, matte vs glossy finish, reflectance%, or any other variables at play affect the performance of the ToF sensors would be much appreciated. @John E KVAM 

 

 

 

 

1 ACCEPTED SOLUTION

Accepted Solutions
John E KVAM
ST Employee

It is not suprizing that a specular (mirror-like) surface performed worse. The VCSEL (laser) is not perfectly uniform, with the center coming on just slightly before the rest of the illumination. It's a physics issue. So you tend to get shorter distances in this case. If you are assuming the device can range on any target, this error is something you might have to live with. If you know which target you are aiming at, you can do the calibration with that target and you will get better results.

In the world specular targets are rare. We tuned the sensor assuming a Lambertian (matte finish) target. And those do work better. 

Compare your results with the accuracy statement in the datasheet. Most of the time, the sensor will do better than the stated accuracy statement, but there are targets that cause the sensor issues.


If this or any post solves your issue, please mark them as 'Accept as Solution' It really helps. And if you notice anything wrong do not hesitate to 'Report Inappropriate Content'. Someone will review it.

View solution in original post

3 REPLIES 3
Anne BIGOT
ST Employee

Hello,

It seems that for sensor2 (orange curve), there is offset issue. Did you retry the measurement after redoing an offset calibration ? 
For sensor1, it is more like cross-talk issue : less error at short distance, more error at longer distance. Can you try to redo a calibration too and see if the deviations are still there ?

Regards

 


Our community relies on fruitful exchanges and good quality content. You can thank and reward helpful and positive contributions by marking them as 'Accept as Solution'. When marking a solution, make sure it answers your original question or issue that you raised.

ST Employees that act as moderators have the right to accept the solution, judging by their expertise. This helps other community members identify useful discussions and refrain from raising the same question. If you notice any false behavior or abuse of the action, do not hesitate to 'Report Inappropriate Content'

Hi Anne,

Both of these sensors have already been calibrated using the recommended offset and cross-talk calibration methods and the the measurement read was taken from the "distance_true" register, which should include the offset and crosstalk compensation already applied.

John E KVAM
ST Employee

It is not suprizing that a specular (mirror-like) surface performed worse. The VCSEL (laser) is not perfectly uniform, with the center coming on just slightly before the rest of the illumination. It's a physics issue. So you tend to get shorter distances in this case. If you are assuming the device can range on any target, this error is something you might have to live with. If you know which target you are aiming at, you can do the calibration with that target and you will get better results.

In the world specular targets are rare. We tuned the sensor assuming a Lambertian (matte finish) target. And those do work better. 

Compare your results with the accuracy statement in the datasheet. Most of the time, the sensor will do better than the stated accuracy statement, but there are targets that cause the sensor issues.


If this or any post solves your issue, please mark them as 'Accept as Solution' It really helps. And if you notice anything wrong do not hesitate to 'Report Inappropriate Content'. Someone will review it.