• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 28 subscribers
  • Views 771 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.

DLPC350: DLPC350 lock-up with usb-communication while repeated pattern sequence running with frame-time > (exposure-time+230us)

Anders Guldahl
Intellectual 385 points
Part Number: DLPC350

We need to add a small delay between exposures because our 2D camera is not ready to capture a new frame before 400us after the last exposure ended. We've tried increasing the frametime to 8733, leaving exposure at 8333. This works fine if we make sure not to send any usb-commands, but if we start writing usb-commands to toggle GPIO-pins while running with this timing, the dlpc350 locks up. sometimes we need to power-cycle, other times it is enough to hit reset i lightcrafter gui. 

The usb-communication works fine with exposure-time == frame-time, or frame-time exactly equal to exposure-time+230us (as per manual).

Higher delays like frame time = 16ms, exposure=8.333ms also gives same issue when sending usb-commands. 

Is this a known bug, is there a workaround?

  • 0
    TI__Mastermind 28005 points
    Anders,

    Welcome to the E2E forums, and thanks for your interest in our DLP technology.

    Are you using Windows 10? I am aware of some driver issues users have encountered when operating on Windows 10. Please try again with a different PC/OS if possible.

    I hope this helps.

    Best Regards,
    Philippe Dollo
  • 0 in reply to Philippe Dollo
    Intellectual 385 points
    We are aware of the windows 10 driver issue, that is another problem which we are also struggling with. This issue does not seem related, we can connect and send commands to the dlpc350 without issues. It is only dependent on exposure-time related to frame-time, we do not re-connect usb between changing frame-time up to 400us longer than exposure. When we then send usb-commands while patterns are displayed, it will lock up after a while. With 0 or 230us difference, it does not lock up. and if we make sure not to send any usb-commands, it also doesn't lock up.
  • 0 in reply to Anders Guldahl
    Intellectual 385 points

    4456.issue_with_dlpc350_lockup.iniSome more information, I can reproduce this issue with only the lightcrafter GUI 3.1.0 on windows 10:

    1. Configure some pattern sequence with repeat, set 8333us exp-time, 8733 frame-time, see attached ini-file with solution I've used.
    2. send, validate, play.
    3. go to peripheral control, use a mouse autoclick tool to send as many requests as possible, I sent I2C-read requests with autoclick-tool set to 1ms click interval.

    After 5-10 minutes, the pattern sequence projection stops. Without spamming the dlpc with requests, it can run for days.

  • 0 in reply to Anders Guldahl
    TI__Mastermind 30875 points
    Hi Anders,
    Definitely the pattern display sequence going to have influence by sending commands in the middle of the pattern display sequence. We recommend limiting the period of sending the command.
    One approach would be to stop the pattern sequence then query the information; to know whether the controller is alive, you can look at the trigger o/p coming continuously.
    Regards,
    Sanjeev
  • 0 in reply to Sanjeev
    Intellectual 385 points

    Does this mean that we cannot reliably send any usb-commands during a pattern-sequence? We are depending upon using continuous pattern display, but we also need to change the state of various GPIO-pins while the pattern is displaying. Is this limitation documented somewhere? 

  • 0 in reply to Anders Guldahl
    TI__Mastermind 30875 points
    Hi Anders,
    Unfortunately the behavior is not documented. Can I ask you why you need to change the GPIO state real time that too when the pattern sequence in progress.
    Regards,
    Sanjeev
  • 0 in reply to Sanjeev
    Intellectual 385 points

    Hi again Sanjeev,

    In the current design we are using the GPIO-pins to control fans, and also control a stepper-motor for the 2D camera iris. We are running patterns for an indefinite amount of time in our application and need to turn fans on and off, as well as move the iris while running.

    We have found another workaround for our core issue, so we can use the 230us delay that seems to be more stable. So this issue can be closed for now, but one more question:

    Would it be correct to assume that you will not release a new firmware for the dlpc350 that addresses this issue? 

  • 0 in reply to Anders Guldahl
    TI__Mastermind 30875 points
    Hi Andres,
    Thank you for updates. Yes there is no immediate plans in near future to make the changes in the f/w to fix the issue.
    Regards,
    Sanjeev