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.

DLPC3439: Rare image glitch after power up

Part Number: DLPC3439

We've been seeing this rare issue where after the DLPC3439 powers on, an image is sent to it, and it gets jumbled. This image:

When projected, looked like this:

Every image sent while it's in this state gets jumbled in a similar fashion. I believe the DLPC is in a bad at this point because I attempted to set the input source to be a test image instead of external video to see if the issue persisted. The test image was not jumbled like the previous images, but when I set the input source back to external video, the image being projected was also no longer jumbled.

It seems as if the DLPC was in a weird state, and changing the input source somehow reset it.

Is there anyway to detect this by querying the DLPC? Or is there anything we can do to avoid this issue?

  • Heather,

    Thanks for the images you have provided. I have a few follow up questions:

    What system software version are you running? Furthermore, what is the resolution of the image as it is sent to the DLPC?

    Regards,
    Philippe Dollo
  • If I understand the result of the Read System Software Version command, the version is 1.3.2. The image size is 1920x1080.

  • Heather,

    Perhaps you are having issues with the external video data being sent to the system. Have you tried setting the image above as a Splash screen to see if it displays correctly? If so, the issue is probably the external video interface for your system. You may need to verify that the connections for your external video interface are OK, and that the video timing specification you are using is good. Please share it if you would like me to take a look.

    Best Regards,
    Philippe Dollo
  • As I stated in my original post, the image from the external source was jumbled, and when I switched to a test pattern, the image was no longer jumbled, BUT when I switched back to an external source, the image was still not jumbled. All I did was switch sources.
  • Heather,

    Can you share your system schematic?

    Best Regards,
    Philippe Dolllo
  • Heather,

    Are you still encountering issues with your system? If so, please share your system schematic. If not, I'll tentatively mark this thread as resolved.

    Thanks a lot!

    Best Regards,
    Philippe Dollo
  • We can't share the schematic.
    This issue occurs very rarely when the projector powers on, and I was wondering if there was a way to detect it and correct it without the user knowing about it.
    Changing the input source made the issue go away, but I'm wondering if the DLPC powered up in a weird state, or if this is something that can happen with changing sources itself. If it's a power up issue, I'm not opposed to just telling the projector to switch to using a test pattern and back to an external source after everything has powered up. I'm concerned that the problem might actually be in the logic of setting the input source itself as the source is initially set after power up, so I can't rule it out. I just need some insight on this so I can figure out the next steps.
  • Heather,

    This chipset actually features a curtain functionality for this sort of purpose. If you want to hide the image temporarily you can enable the curtain while switching display modes, and re-enable it when it is safe to do so. This command is detailed in the software programmer's guide (page 32):

    http://www.ti.com/lit/ug/dlpu020c/dlpu020c.pdf

    Are you starting the system in external video mode by default while also sending parallel image data at the same time? Perhaps your particular frontend startup configuration is causing data to be sent the DLPC improperly. In this case having the system start up with a curtain is an easy fix.

    Best Regards,

    Philippe Dollo

  • Video data is being sent regardless of the projector being turned on. So if this has the potential to cause problems, then that's probably what's going on.

    So what we should do is start with one of the internal image sources instead? That's what I was originally thinking about doing because I had noticed switching sources seemed to fix the problem, but I was worried that setting the source might be what caused the issue because I didn't know the source of the problem.

  • Hi Heather,
    Please let us know if you still need help with this issue. I will tentatively close this thread.
    regards,
    Vivek