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.

DLPC900: LED Synchronization

Part Number: DLPC900

Hi Team,

Our customer is working with DLPC900 and would like to seek assistance if on this issue. Please see his query below:

"I have some trouble in synchronization of my LED hat with DMD. I used an RGB matrix hat which is controlled with a raspberry pi. I send a trigger out from the raspberry to the DMD via trigger input_1 to have PWM with desired duty cycle.

My aim is after sending the input trigger the DMD starts to display a set of images, four 1-bit plane images, and when finished return to the beginning and wait for the next trigger.

Tor testing the synchronization, I tried to turn "ON" the led just for one image while when the DMD display the remaining 3-image be turned "OFF"; this is repeating till I stop the process manually.

Unfortunately, the image is not synchronized and I think I need a little more declaration of the different parameters with their exact timing.

images1-2:

We know in the pattern set, the minimum delay for the trigger input is 105 us but may I know which one of the blue, orange, or (orange minus blue) arrows are 105us? whether the triggering duration from my raspberry should be also 105us duration? and one more important one the DMD is sensitive to the rising edge or falling one, image 1 or 2?

For trigger out 1, while it is raised the DMD display patterns sequentially; how long the green arrow takes? the tie is free to choose or it is dependent on the trigger in?

For trigger out 2, indicate the start of each pattern. whether it is started automatically after falling edge of the input trigger? in here that I do not use any trigger-out whether I should consider it. how much time the red arrow would take?

In the timing box, we should define the exposure and dark time. whether for the first frame that I have trigger_in, the 105 us should be reduced from exposure time if I want to have the same sequence of LED on?

I tried many scenarios but unfortunately in some loops, I see a delay in displaying the 4th image, and in the next repetition the DMD stop to display images till the next 2 input trigger; I think because the DMD frames are not synchronized the DMD stop displaying the frameset that can start it while the next one arrives.

Images3-4:




I tried to put the two repetitions of my sequent after each other for better observation. now I can clearly see there is some gap between the DMD image sequence. what should I do meanwhile these times?

I really do appreciate it if anybody could help me with this until I do not know the right configuration I could not manage my errors."


Thanks in advance!


Kind Regards,

Jejomar

  • Hello Jejomar,

    Thank you for the query.

    Our first recommendation is to use an oscilloscope to characterize the delay that is present. With this information, make the desired adjustments to the delay in the GUI or in your code. 

    You can also consider using the three LED Enable signals provided with the DLPC900 controller to turn the LEDs on/off in sync with the LED driver. See the DLPC900 Single EVM Guide or the DLPC900 Dual EVM Guide for more details.

    Finally, which DMD are you using?

    Best,

    Maxine

  • Hi Maxine,

    My apologies for the delay. I was busy fixing and checking the output of my LEDs, and I have been putting a lot of effort into it to check the exact frequency o TRIG_IN_1 with an oscilloscope as advised.

    But whether we could check the same thing for DMD beyond the software? I am using a DLPLCR6500EVM model with a DLPC900 microcontroller. However, I am unsure whether it is necessary to consider TRIG_OUT 1-2 when I am only using TRIG_IN?

    at least for TRIG_IN_1 Could you please guide me on how to ensure that the DMD starts to display 105us after the rising edge or falling edge? Also, if the DMD is sensitive to the falling edge, is it important to consider the width of the trigger, as shown in the figure up_right?

    Thank you for your assistance. I look forward to your response.




    Kind Regards,

    Jejomar

  • Hello Jejomar,

    We will look into this issue in the lab. Please give us a few days to make a thorough investigation.

    Best,

    Maxine

  • Hello Jejomar,

    We got our system up and running and will look at this tomorrow.  Sorry for the delay on this.

    Fizix

  • Hi Fizix,

    No worries. Maxine is now helping us with this issue and she's already in direct contact with the customer.

    My apologies for the late update. We can now close this thread. Thank you!


    Kind Regards,

    Jejomar