View my account

D430 depth module and Vision Processor D4 Board info

Comments

5 comments

  • MartyG

    Official Intel CAD files in the STEP format for the 400 Series depth modules, cameras and Vision Processor can be downloaded from the link below.

    https://www.intel.co.uk/content/www/uk/en/support/articles/000026841/emerging-technologies/intel-realsense-technology.html 

    Details of the 'interposer' for connecting the Vision Processor D4 and Depth Module together can be found on page 43 of the current edition of the 400 Series data sheet document.

    https://dev.intelrealsense.com/docs/intel-realsense-d400-series-product-family-datasheet?elq_cid=1206596&erpm_id=1567221 

    The mechanical drawing for the D430 module and Vision Processor D4 are on page 104 and 105 of this data sheet.

    The interposer can either be a flexible cable, or a pre-made rigid 'Interposer Board' with connectors to plug the Vision Processor D4 and Depth Module onto.  One of the suppliers for this board is FRAMOS.

    https://www.framos.com/en/d400-rigid-50-pin-interposer-20737 

    Otherwise, if the HSF interposer cables sold by FRAMOS are too short for your needs then it is possible to create your own custom cable using the details in the data sheet document or have a company that creates custom cabling as a service produce it for you.

    If you also require information on the second 'Tracking Module' 50 pin connector on the Vision Processor D4 Board, this can be found in the January 2018 edition of the 400 Series data sheet document.  The information was removed from the data sheet in the March 2018 edition onward.  The details of the second connector, receptacle and cabling are on pages 50-55.

    http://ttps://www.rutronik.com/fileadmin/Micropages/Intel/Intel_RealSense_D400_Series_Datasheet.pdf  

    0
    Comment actions Permalink
  • Roman Leuzinger

    Hi

    Thanks for the good overview. 
    I found many information in the datasheet chapter 7 regarding integration of modules (D430) into a custom case.
    Are there any more information, guidelines, how to's regarding integration of modules into custom case?

    Best regards

    0
    Comment actions Permalink
  • MartyG

    Hi Roman Leuzinger  The field of view (FOV) of the camera module's sensors should not be obscured by any non-transparent part of the custom housing that is on the front of the housing.

     

    If you are placing a Depth Module in the housing rather than a cased camera then you will likely need to choose a cover material to place in front of the camera sensors, similar to the glass on the front of the cased camera models.  Examples of materials that can be used for the cover material include acrylic (plexiglass) and Gorilla Glass.  

     

    The data sheet's guidelines about cover materials are on page 156 (Appendix B) of the current edition of the data sheet document, whilst the link below provides guidance about Gorilla Glass.

    https://github.com/IntelRealSense/librealsense/issues/2566

    0
    Comment actions Permalink
  • Roman Leuzinger

    Cool thanks.

    We will design the case that the FOV will not be affected.
    But 10mm in front of the camera we have additional parts. At the moment I am not sure that they will never block the FOV of the camera. It will be in an area, where we do not need the information of the camera. So the top of the depth image is not used and this is the section where parts might block the FOV.

    Does the depth calculation need the full FOV to generate the depth, even though we the do not need the full FOV of the depth image?

    Example:
    D430 has FOV H:87 V:58
    Our application requires H:87 V:45
    Obstable might block a bit in the area which is not used but still covered by the sensors.
    Would that work?
    Could reflection from the dot projector disturb the result?

    0
    Comment actions Permalink
  • MartyG

    The camera will generate depth data for whatever areas of the scene that the cameras can see in the FOV and analyze for depth.  If part of the FOV is obscured then this does not prevent the generation of the depth image. The obstructed area may simply be rendered on the image as plain black (no depth detail).

    The D430 model has a minimum depth sensing distance of around 0.1 meters / 10 cm.  So an obstructing part may not show up on the depth image if the obstruction is less than 10 cm from the camera sensors, but I can foresee that laser light might be disruptively reflected back to the sensors if the obstruction has a reflective surface.

    0
    Comment actions Permalink

Please sign in to leave a comment.