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.

DLP2010NIR: I am trying to find out the DMD Row buffer size.

Part Number: DLP2010NIR
Other Parts Discussed in Thread: DLPC150

I have read the DLPA0008 document and it shows the buffer size for a XGA DMD.  See this image:

I am trying to find this same information for the WVGA version, DLP2010NIR.  The DLP2010NIR is 854 pixels wide and our team was unsure if it would require 32 clock edges or less (32-bits*27= 864 pixels > 854). 

Our team is trying to preform a raster scan as fast as possible. We are trying to determine the maximum speed at which a single pixel (or small group of pixels) could be scanned across the DMD if the display is loaded row by row with the DLPC150 controller IC and utilizing the max speed 75 MHz 24-bit parallel interface.  Based on what we have read so far the minimum total time to toggle a single pixel appears to be:

(6uS (micro mirror switch time) + time the load the row on the SubLVDS 532 MHz bus + time to transfer data to DLPC150 on 24-bit parellel bus @ 75 MHz) x 2

x2 because you would need to enable the pixel and then clear it again before moving to the next pixel.  We have seen 2880 Hz binary pattern rate in multiple sources.  Although this is quite fast to display a pattern we need to raster each micro mirror much faster than this.

Thanks,

  • Hello Richard,


    Could you please clarify which document is the DLPA0008? I was unable to find it.

    Also, are you able to clarify a little bit more about your application? If  it's for something other than spectroscopy you may be better off using the DLP3470 controller to interface with the .2 WVGA DMD. We do not support single row updates on this platform to my knowledge .

    Thanks,
    Kyle

  • Hi Richard,

    Thanks for clarifying. That document is actually specifically for a different type of DMD (although many of the general principles are still applicable). Are you able to answer the follow-up questions:

    Also, are you able to clarify a little bit more about your application? If  it's for something other than spectroscopy you may be better off using the DLP3470 controller to interface with the .2 WVGA DMD. We do not support single row updates on this platform to my knowledge .

    Thanks,

    Kyle

  • Kyle,

    I know the document is for a different DMD.  I was wanting more detailed information on the DMD I want to use.  Like does it have a 32 bit row buffer?  The documentation on this product is very scarce.  It does not really show this part works.  

    I am trying to raster the dmd, so I want to control each mirror.  I am wanting to do this in 10FPS., but the documentation does not really  show fast this can do and if you can even send patterns at a bit, row, or whole image.

  • Hello Richard,

    It depends on which controller you are using. The DMD is not directly sent data from a front end. The utilized controller will affect how you can address the DMD. Please let me know about the follow-up questions so I can help.

    Thanks,

    Kyle

  • I am trying to find this same information for the WVGA version, DLP2010NIR.  The DLP2010NIR is 854 pixels wide and our team was unsure if it would require 32 clock edges or less (32-bits*27= 864 pixels > 854). 

    Our team is trying to preform a raster scan as fast as possible. We are trying to determine the maximum speed at which a single pixel (or small group of pixels) could be scanned across the DMD if the display is loaded row by row with the DLPC150 controller IC and utilizing the max speed 75 MHz 24-bit parallel interface.  Based on what we have read so far the minimum total time to toggle a single pixel appears to be:

    (6uS (micro mirror switch time) + time the load the row on the SubLVDS 532 MHz bus + time to transfer data to DLPC150 on 24-bit parellel bus @ 75 MHz) x 2

    x2 because you would need to enable the pixel and then clear it again before moving to the next pixel.  We have seen 2880 Hz binary pattern rate in multiple sources.  Although this is quite fast to display a pattern we need to raster each micro mirror much faster than this.

  • Hi Richard,

    We do not support individually raster scanning DMD rows on this platform. If you need something faster I would have to recommend a different platform.

    Thanks,

    Kyle

  • Kyle,

    Thanks, I understand that.  

    Can you verify the timing we have presented in our original question?

  • Hello Richard,

    The DLP2010NIR uses a different non-public loading technique. It is required to use our controller to address the device.

    Thanks,

    Kyle