• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 428 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.

Original question:

DLP lightcrafter 6500 input triggers in structured light mode

DLPC900: input triggers + repeating a binary pattern before advancing

MaPear
Prodigy 10 points
Part Number: DLPC900

Hi there, TI techs!

BACKGROUND

I am using a DLP6500 with DLPC900 controller in a pattern on the fly mode using the USB protocol.

My application requires that I repeat the pattern M, N times before advancing to M+1.

Previously, I had used the following workflow:

(1) merged/encoded/uploaded the bitmaps over USB, defined the patterns, and configuring the LUT with one of each image

(2) next, I reordered the LUT with N entries for pattern M and then start the sequence (which plays all N images without stopping)

(3) when the sequence ends, I then reorder the LUT with pattern M+1 (and repeat).

This works, but since the LUT is limited to 511 entries, I cannot do all of the M*N patterns I need for my application.

As a result, I am exploring an alternate approach using the external input trigger functionality and ran into some issues getting it to work. 

INPUT TRIGGERS APPROACH

Instead of programming the LUT with N entries for each pattern M, I want to use input triggers to accomplish a similar result.

In particular (new steps in bold)

(1) merged/encoded/uploaded the bitmaps over USB, defined the patterns with input triggers, and configuring the LUT with one of each image

(2) use input trigger 2 to start and stop pattern M, N times 

(3) use input trigger 1 to advance to pattern M+1 (and repeat).

So far I have been able to advance the patterns with trigger 1. However, I cannot repeat the pattern with trigger 2? 

QUESTIONS

  • Is it possible to accomplish this functionality with input triggers?
  • Can you provide a reference to the documentation with an example like this? 
  • Can you explain why trigger 2 will not repeat pattern M? (I am able to advance to the next pattern with trigger 1)

Thanks for all the quality support you provide this community.

thanks! MP

  • 0
    TI__Guru 54320 points

    MaPear,

    Trigger two is more like a gate than a trigger.  Trigger 1 will advance patterns as long as Trigger 2 is high.

    Here is a possiblity, which you will have to try:

    1. First Load the two patterns into memory as you suggested with the Update LUT button
    2. Use the Re-order LUT and put pattern M with trigger in with repeat.  
    4. HIt Run.
    5. Hold T2 high and use T1 to display the pattern N times.
    6. Hit Stop
    7. Use  the re-order LUT to load Pattern M+1 with triggers and repeat.
    8. Hit Run.
    9. Hold T2 high and use T1 to display the pattern N times.
    10. Hit Stop

    You can likely set up batch files to do the stop and re-order commands.

    NOTE:  Read the EVM Guide section on the LUT - reorder for more details on how to use the Re-order 

    Fizix

  • 0 in reply to Fizix
    Prodigy 10 points

    Thanks. I'd prefer not to reorder the LUT with the PC, and instead just program once and let the triggers handle the repeats and advancing. 

    After some trial and error, I can get the work flow I suggested above to work IFF I use trigger 2 to stop the pattern while it is being displayed. 

    In this instance, it appears I can restart the same pattern.

    However, if I let the pattern display for its entire exposure time, then I can only advance. 

  • +1 in reply to MaPear
    TI__Guru 54320 points

    MaPear,

    Although this is not how we conceived of the triggers being used, this sounds like it will work for you.  Looks like "thinking outside the box" enabled another way to operate.

    I assume you are using binary patterns.  This could be a little more challenging with grayscale patterns.

    Should we mark this as resolved?

    FIzix

  • 0 in reply to Fizix
    Prodigy 10 points

    You may mark as resolved for now. 

  • 0 in reply to MaPear
    TI__Guru 54320 points

    Thank you MaPear.