• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 2 answers
  • Subscribers 30 subscribers
  • Views 222 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

OPT3101: Safety Usage

Spark
Prodigy 80 points
Part Number: OPT3101

Hello TI team,

I'm trying to build a safety device that uses an array of ToF sensors to detect obstacles (people) and I wanted to use the OPT3101 IC.

The Requirements are:

  • It must work up to 4-5m of distance;
  • It must work both outdoors and indoors;
  • It must detect obstacles in real time and must be able to detect objects as small as a hand/arm;

To be more precise I would like to build a ToF array of 3x2 10° emitters that point to the ground and detect objects/people. Since the OPT3101 can only manage 3 emitters, I would like to put 2 emitters in series (3x2).

Is this type of application compatible with the OPT3101 IC? Would I have problems detecting those types of obstacles? What about parts of the body such as an arm?

Thank you in advance.

Best regards

  • 0
    TI__Mastermind 25895 points

    Hi,

    OPT3101 can detect both people and any other objects that are not transparent. It cannot distinguish between a person and any other type of object. I'm not sure how you intend to put the emitters in series. If emitters are in series then they will become one zone of detection.

    Best,

    Alex

  • 0 in reply to Alex Bhandari-Young
    Prodigy 80 points

    Hi,
    Yeah I'm aware of the fact that by putting them in series they become a singe detection zone.
    And of course I know that it doesn't distinguish the type of object detected.

    My only concern is about how big must the object be in order to be able to detect it. If for example I use 2xVCSELS with a FoV of 10° in series on TX0 (so together they cover a FoV of 20°) and I place an obstacle that only blocks 5° of the FoV of one of the sensors, what type of measurement would I have? Does it return the smallest distance it detects?

    Another scenario (more important): If I have just 1 VCSEL per TX channel, each with a FoV of 20° and the sensor is placed at 4 meters from a wall, facing the wall. What is the smallest object it can detect? Would it detect a 10cm hand? Is there any way to calculate this?

    Thank you in advance.

    Best regards

  • 0 in reply to Spark
    TI__Mastermind 25895 points

    Hi,

    The OPT distance will be a weighted average of the strength of the signal returned (as long is signal is high enough that crosstalk is insignificant). If there are no other objects reflecting light back to the sensor the sensor will give the correct distance with only 50% of the FoV filled. Again, you just need to check that the signal strength will be high enough. More details in our intro to system design doc online. We also have a system estimator tool for checking the error based on target reflectivity.

    Best,

    Alex

  • 0 in reply to Alex Bhandari-Young
    Prodigy 80 points

    Hi,

    Thank you for your reply. Just one more question: If I want to improve the distance noise and accuracy, can I use multiple photodiodes in parallel with smaller FoV?
    For instance using 3x PD with 20° of FoV each(tilted so that they cover 60° in total) instead of a single PD with 60° FoV.

    And if I can connect multiple photo-diodes in parallel (on the INP pin), is there any hardware change that I must pay attention to?

    Best regards

  • 0 in reply to Spark
    TI__Mastermind 25895 points

    Hi,

    Multiple PD in parallel will increase the device susceptibility to xtalk as the RX loop will be enlarged (see system design doc for more details). Additionally you will increase your PD capacitance and max PD cap is 6pF. 

    Best,

    Alex

  • 0 in reply to Alex Bhandari-Young
    Prodigy 80 points

    Hi,

    Let's say that my design limits the xtalk and it's not a problem. Will the capacitance be a big problem? Is there a solution that would allow me to use multiple PDs? Perhaps by using additional hardware?

    Best regards

  • 0 in reply to Spark
    TI__Mastermind 25895 points

    Yes capacitance will be an issue if you exceed the datasheet spec. I have not seen anything that will allow this for a 3ch design. It will be quite challenging for all these reasons.

    Best,

    Alex