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.

DLP9500 Max Pattern Rate

Guru 10570 points
Other Parts Discussed in Thread: DLP9500

Hello,
I identifies that Max Pattern Rate is binary 23KHz in DLP9500.
But, Micromirror Switching Time is specified as 140 usec in datasheet(dlps025b: Table4), which means just 7KHz pattern rate.

Question)
- To perform 23KHz pattern rate, are there any restriction to control DLP9500?

Best regards, RY

  • Hello,

    There is note(10) about Micromirror Switching Time like following:

    I need to achieve 23KHz Max Pattern Rate.
    - Can I have support for the detail of system implementation and representation described above?

    Best regards, RY

  • Hello,
    I assume high speed pattern rate depends on Application FPGA.
    But, there are some special tips and technic to use FPGA.
    Is my understanding right?

    How can I get its information?
    Do we need to get NDA with TI? or get special support from 3rd Party?

    Best regards, RY

  • Hello RY,

    Our sincere apology that your request got overlooked.

    The 140 µs is a copy and paste error when the datasheet was updated for public release.  This is already in our list of items to correct on the next revision of the datasheet, which should be sometime this year.

    The correct number is  56.2 µs for Global operation.  Global operation is when data is loaded to the array and then a Mirror Clocking Pulse [MCP] is sent to the entire array at the same time.  This is 43.2 µs loading time at 400 MHz DDR (LVDS) plus 2 µs MCP time plus 11 µs µ-mirror settling time (13 µs total for MCP + Settling].  This only gives approximately 17.8 k Binary Patterns per Second [BPPS].

    To achieve the 23 kBPPS you must use "Phased" operation.  This works in a manner similar to a rolling shutter on a digital camera.  You can be loading data in one section of the DMD while waiting for the MCP + Settle Time in another section.  This allows the 13 µs MCP + Settle Time to be "hidden" during the loading of another section.  In this mode the refresh time is just the load time of 43.2 µs which gives ~23 kBPPS. 

    The caveat is the same as a rolling shutter on a digital camera - The image is not displayed all at the same time.  This is not a problem for a human eye or a camera with a sufficiently long integration time.  See this Wikipedia article on Rolling Shutter.  <-- click

    I hope this helps.

    Best regards,

    Fizix

  • Fizix-san

    Thank you so much for your reply.
    It was helpful for us. Thanks!!

    Best regards, RY