• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 28 subscribers
  • Views 200 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

DLP6500FLQ: Eliminating noise from switching with DMD

Carlo
Expert 7715 points
Part Number: DLP6500FLQ
Other Parts Discussed in Thread: DLPC900

Hi E2E,

Our customer needs assistance with a DLP6500 device using a DLPC900 evaluation module. They have been attempting to eliminate a switching noise that comes from the DMD device jittering the mirrors every 105us to prevent the mirrors from getting stuck. This is the exact time scale they want for the patterns, so the noise is extremely loud and drowns out there measurements. Our customer have attempted to manually use commands on page 56 of the DLP Programmer's Guide to make sure the patterns don't wipe in that time, but using an external trigger doesn't cause any patterns to show up for them.

They are using a frequency generator to input the external trigger to control the patterns, while the DMD is ordered to not clear a pattern after exposure and instead wait for a trigger. This doesn't work as intended, but the process does work if they don't include the external trigger, so they're at a loss on how to proceed.

Thank you in advance for your help.

Regards,
Carlo

  • 0
    TI__Mastermind 43165 points

    Hello Carlo,

    The 105 us time is the minimum pattern time.  This is baked into the firmware.

    Would it be possible for them to save as a batch file the pattern timing and commands they are using and post them.  Having a trigger at the same frequency as the minimum pattern load time may be the issue.  It would be3 helpful to analyze what is happening here.

    Fizix

  • 0 in reply to Fizix
    TI__Expert 7715 points

    Hello Fizix,

    Here is a batch file of the patterns that our customer are sending to the DMD.

    Raster_128.txt

    They have manually adjusted the file with Matlab, changing the hex values in all MBOX_DATA commands on Byte 5 from '11' to '90', which corresponds to a bit pattern of 1001000. This is to manually set 'Wait for Trigger In' to '1', or 'True' and to set 'Clear Pattern after Exposure' to '0', or 'False'.

    Regards,
    Carlo

  • 0 in reply to Carlo
    TI__Mastermind 43165 points

    Thank you Carlo,

    I am looking at this.  Allow a little time for analysis and possible advice on this.  THANKS>

    Fizix

  • 0 in reply to Fizix
    TI__Expert 7715 points

    Hello Fizix,

    Please share an update on this. Thanks.

    Regards,
    Carlo

  • 0 in reply to Carlo
    TI__Mastermind 43165 points

    Carlo,

    My apologies, this would not load on our 6500 system.  It has a problem with the 08 at the end of the second MBOX line and will not load further.  I had started debugging the batch file, but did not get back to it.  

    I suspect that the trigger timing is too close to the DMD load time of 105 us.  Are they able to reverse the measurement system by using ouput trigger 1 (or even Output Trigger 2) from the EVM to trigger the measurement?

    Fizix