• State TI Thinks Resolved
  • Locked Locked
  • Replies 12 replies
  • Answers 1 answer
  • Subscribers 29 subscribers
  • Views 434 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.

DLPC3478: The projector will occasionally not cast light in use, and can not be recovered, only power off to restore the normal work.

wang tong
Prodigy 10 points
Part Number: DLPC3478

As mentioned in the title, I used Jetson Nano to control the optical projector projection through I2C interface. There was a certain chance that the projector would not emit light and could not be recovered. It could only work normally after the projector was powered off and powered on again.

The problem has occurred on multiple devices at random times. In order to easily reproduce the problem, I configured the projector through the i2ctransfer command. According to the configuration Mode in use, the Test Pattern Generator Mode was executed first, followed by the normal operation in Internal Pattern Streaming Mode.

Then I added the mode switch to simulate the fault state, and found that the optical machine could not cast pictures normally, and we had no way to recover, unless the power was turned on again.

In order to ensure the normal use of the product, first of all, we want to determine whether there is a software reset method of DLPC3478 that can restore the normal operation of the optical machine under the condition of continuous power.

And then whether there is configuration recommendations, or other suggestions, to avoid this problem.Thank you very much!

Here is the python script running in nano, which is the test method mentioned above:

import os
import time

# enable_solid_field

print("enable solid field")
os.system('i2ctransfer -y 1 w2@0x1b 0x1a 0x01')
time.sleep(1)
os.system('i2ctransfer -y 1 w2@0x1b 0x05 0x01')
time.sleep(1)
os.system('i2ctransfer -y 1 w7@0x1b 0x0b 0x00 0x70 0x0f 0x00 0x0f 0x00')
time.sleep(1)
os.system('i2ctransfer -y 1 w2@0x1b 0x52 0x07')
time.sleep(1)
os.system('i2ctransfer -y 1 w2@0x1b 0x1a 0x00')
time.sleep(1)
# disable_solid_field
print("disable solid field")
os.system('i2ctransfer -y 1 w2@0x1b 0x1a 0x01')
time.sleep(1)
os.system('i2ctransfer -y 1 w7@0x1b 0x0b 0x00 0x00 0x0f 0x00 0x0f 0x00')
time.sleep(1)
os.system('i2ctransfer -y 1 w2@0x1b 0x1a 0x00')
time.sleep(1)

# init

print("init")
os.system('i2ctransfer -y 1 w2@0x1b 0x05 0x04')
time.sleep(1)

while True:

    # pattern run
    print("pattern run")
    os.system('i2ctransfer -y 1 w2@0x1b 0x05 0x00')
    time.sleep(1)
    os.system('i2ctransfer -y 1 w2@0x1b 0x05 0x04')
    time.sleep(1)
    os.system('i2ctransfer -y 1 w3@0x1b 0x9e 0x00 0x00')
    time.sleep(2)
It can also be reproduced by using DLP EVM GUI 3.1.0.5->Advanced->Batch Command to operate the optical machine directly.
  • 0
    TI__Expert 5290 points

    Hello User.

    What firmware version are you using in this device? Have you tried using the 3.1.0.7 GUI? Newer FW and GUIs have made improvements on this behavior.

    The team will investigate further into your questions. Thank you for your patience.

    Regards,

    John

  • 0 in reply to John McKnight
    Prodigy 10 points

    Hello John,

    Thank you for your reply. I updated the GUI version and firmware version and found that the test results were the same as before, the problem still existed. The versions used for the test are as follows：

  • 0 in reply to wang tong
    TI__Genius 12575 points

    Hello User,

    Is this on a DLP3478EVM-LC or a custom board? If it is able to be observed, do you see any change in state of the HOST_IRQ signal?

    Regards,

    Austin

  • 0 in reply to wang tong
    TI__Expert 5290 points

    Hello User,

    Do you already have internal pattern data saved on the EVM? This can affect how the EVM works if it is commanded to display patterns that aren't saved on the EVM already.

    Regards,

    John

  • 0 in reply to Austin Snyder
    Prodigy 10 points

    On custom board. It is no problem of the HOST_IQR signal because the projector can cast light normally. Even under abnormal conditions, iic communication is also normal.

  • 0 in reply to John McKnight
    Prodigy 10 points

    Yes, internal patterns data saved on the board.

    I'm going to do some tests about exposure time through the GUI. These tests might take a long time as well, since the actual production ran with only one problem for a month.

  • 0 in reply to wang tong
    TI__Expert 5290 points

    Hello User,

    Can you save your internal pattern data to a file and share it?

    Can you also include a video of the EVM displaying when this batch file is running? You are freezing and unfreezing the image in succession and having it switch modes.

    Regards,

    John

  • 0 in reply to John McKnight
    Prodigy 10 points

    Here is the pattern data saved with GUI, and the video of Batch file execution.

    It should be noted that this test method only artificially simulates the state that the projector does not cast light, and does not have such mode switching operation when it is used (the switching operation between Internal pattern and External Video).

    internal_pattern_data.zip

  • 0 in reply to wang tong
    TI__Genius 12575 points

    User,

    Thank you for the binary file and video. Please allow us until this time next week to attempt to recreate and debug this issue.

    Regards,

    Austin

  • 0 in reply to Austin Snyder
    TI__Genius 9971 points

    Hello Wang, 

    Thanks for the detailed issue description. Can you please confirm that you see the issue only with the internal pattern display and not with TPG display?

    Regards,

    Mayank

  • 0 in reply to Mayank Khandelwal
    Prodigy 10 points

    Yes, that is only found in the internal pattern mode. 

  • 0 in reply to wang tong
    TI__Genius 9971 points

    Wang, 

    Thanks for the confirmation. The team is working in fixing and validating the fix for this issue. The next firmware maintenance release will have a fix for it. We will circulate an e2e FAQ once the new firmware release is available. 

    Regards,

    Mayank