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.

DLPDLCR470TEVM: Can't get image on DMD, some readouts in DLPC4422GUI seem to point to problem

Part Number: DLPDLCR470TEVM
Other Parts Discussed in Thread: DLPC4422, DLP470TE

Hello TI team,

we want to set up  the DLPDLCR470TEVM together with the DLPC4422EVM. We did so far according to user guide:

- Connect DMD board to DLPC EVM board

- Successfully flashed the DLP470TE-8.1 firmware as downloaded from TI website (file Flash_DUAL_DLPC4422_DLP470TE_LED.img)

- USB communication between DLPC4422 GUI and EVM board works fine

- However:

  - No TI splash image is displayed

  - In "Display" page of DLPC4422GUI

         - "Projection Mode" cannot be changed from "INACTIVE" to anything else

         - "Power Mode" is always RESET (J29 on EVM board is NOT installed)

  - In "DMD" page of DLPC4422GUI

         - "DMD Park" is TRUE, can be changed to FALSE

         - "DMD MIrror Lock" is "Locked", can be changed to "Unlock", if we do se, the DMD shows a somewhat noisy quadrangle on one half and a "clean mirror" image on the other half

Thank you for any help and best wishes,

Matthias

  • Matthias,

    Every one is out of the office until next week.  In the meantime, I do have one thing to check to start….

    Are both Flex cables attached the same direction, with the end labeled “DMD End” nearest the DMD?  Are the screws on the cable connectors tight?

    Regards,

    Gary

  • Thanks Greg for your quick reply, we double-checked both assembly direction and screw tightening, also had a close look (microscope) on the contact, but everything looks OK.

    Best wishes, Matthias

  • i guess you are using the DLP470TE, which uses an interposer and socket plate wit’s 4 screws to connect the DMD to the board … so there are 4 additional screws to check.  

    Have you checked the the DMD socket screws as well as the cables?

  • Thanks Gary, yes, it's the 470TE, and we checked the screws only now, indeed they weren't tight. So far I only tried to tighten the screws (No.1 in attached pic), and saw that the hexnuts on one side (2) turn with the screws, so they are also loose.  Is it safe to remove the screws or even disassemble the whole Heatsink/interposer/DMD thing to check alignment and reassemble with all hexnuts and screws tightened correctly ? (don't want to break anything)

  • Hi Matthias.

    It should be fine to re-assemble the mounting hardware. The main items to keep in mind are uniformity of pressure on the DMD as well as on the PCB. 

    Getting a uniform assembly should then enable solid electrical connection, which was the starting point for Gary's questions.

    We can also try in parallel to get you more info on the mechanical attributes of this assembly.

    Thanks,

    Matt

  • Hi Matt, thank you for your reply. We now disassembled the 

  • 470TE module and carefully re-assembled it. The hex bolts are now uniformly tightned (not to much), as well as the heat sink. However we still see the same behaviour: No TI splash screen on startup, 

      - In "Display" page of DLPC4422GUI, "Projection Mode" cannot be changed from "INACTIVE" to anything else, and  "Power Mode" is always RESET (J29 on EVM board is NOT installed)

      - In "DMD" page of DLPC4422GUI, "DMD Park" is TRUE, It can be changed to FALSE, afterwards the DMD shows a pattern you can see in the attached picture. Everything else (Power Mode, Display Mode) remains the same and cannot be changed.

    We also tried to swap the Flex cables (to check whether one of them is defective), but the pattern stays the same, i.e. the quad does not move to the other side. - What could we try next?

    Thanks, Matthias

  • Matthias,

    Just to confirm, you tightened #2 in your previous drawing, which should tighten the DMD/Board interposer interface....   I always try to squeeze the stack-up  together with my hand/fingers and lightly finger tighten the 4 screws; then I tighten the opposite corners with a driver with tight finger tips only.  

    To confirm, J29 and J31 are not installed.  Have you checked all of the Jumpers are set as indicated in Table 2-1 in the User's Guide?

    Also, can you do a "Get" on everything on the DDP Control/Stats page and let us know what is reported there?

    Thanks,
    Gary

  • Hi Gary, yes, we assembled it as you describe, same method of tightening the hex screws. All jumpers are set according to Table 2-1 in the UG. One note: It looks like SW1 behaves the opposite way than described in the UG, USB access through the DLPC GUI works when SW1 is in OFF rather than in ON state, so we do everything with SW1=OFF. . Here is a screenshot of the DDP Control/Status page after doing a GET on all entries.

    Thanks, Matthias

  • Matthias,

    It seems like the sequencer is not running.  Since the system is reset, I wonder if "power good" ("PWRGOOD") may be low, which would cause things to stay in reset.  Could it be that your power supply voltage is low or perhaps your power supply is not able to provide enough current (constant or possibly peak)?

    Can you check the "POWERGOOD" signal as well as the "POSENSE" signal; both of these are related to the power supplies and both could cause the ASIC to be stuck in reset.  If these are low, you should probably look at your 12V supply t see what it it is doing during startup and while stable.

    Also, can you connect a serial port to RS_232_M (115200, 8, N, 1) and log all of the serial port data from start-up (using a terminal program like Putty or TeraTerm, etc.); that should give us some clues as well.

    Regards,
    Gary

  • Hi Gary, we checked both POWERGOOD and POSENSE: POSENSE is always HIGH when 12 power is connected, POWERGOOD goes HIGH when SW1 is flipped to OFF.

    SW1=OFF also starts the startup sequence and output on RS232_M, which is as below. The last line ("TempSensor Sample Environment:Temp Sensor Write Incomplete, returned f9") is repeated continuously.

    In this output, I saw an error line 60 ("illumination: DMD is invalid for this system config error in DMD_Power()")

    Thanks, Matthias

    -----------------------------------------------------------------------------------------

    In Bootloader: v6.0.0

    Debug opened on URT0
    EEPROM content is valid
    EEPROM: Initializing Slave EEPROM: 3320 bytes
    sysmon: System Startup State From AppCfg is 0
    sysmon: Notify uC ASIC is running.
    sysmon: 1.8V power enabled via PMD

    Serial Flash NVM = 0xEFCF

    sysmon: Low-power mode change cc = 1
    sysmon: Memory test cc = 1
    sysmon: System mailbox ID = f7ff542c
    API version: 08.01.00
    App version: 08.01.00
    ASIC ID: 50
    ASIC Configuration: 4422
    Configuration layout versions:
         Seq Map: 44.21.0019
         SW Map: 44.21.0000
         EEPROM: 08.01.0000
    *****Through DDP_Init cw & seq*****
    projectorCtl: Opening projector control on USB
    digcontrolf: Cfg complete
    digcontrolf: powerStandby complete
    pictcont: Setting the new system mode_num 1
    pictcont: Disable BrilliantColor
    pictcont: pictcont_CompleteSystemMode completed
    pictcont: Set lamp waveform index if URT enabled for sequence mode number 0, cw mode 11
    pictcont: Setting the new system mode_num 1 completed
    datapath: Set the default system mode complete
    datapathf: datapath_powerStandby - Transition to powerStandby complete

    illumination: Starting SSI Initialization...
      Configuring SPI Driver(s)...
      Configuring Sensor. Type = 1.
      Initializing CCI configuration...
    digcontrol: StopCurrentOperation complete

    datapathf: Transition to SUSPENDED
    SSI_Calibration  EE_GetVAR command Successful...
      SSI initialization complete
    illumination: Completed illum_Init
    sysmon: Through _sysReset
    sysmon: System Startup State From AppCfg is 0
    sysEvent: SYSEVENT_START
    sysmon: MASTER ASIC
    Sysmon: Transition to normal operating mode
    Environment:Temp Sensor Write Incomplete, returned f9
    TempSensor Sample Environment:Temp Sensor Write Incomplete, returned f9
    Environment: Data not read from Temp Sensor, returned 249
    TempSensor has been initialized systemmode: There are 4 defined System Modes
    systemmode: TwoD System Mode Table Created
    systemmode: HighSpeed System Mode Table Created
    systemmode: XPR System Mode Table Created
    systemmode: ThreeD System Mode Table Created
    Illumination: Transition to operating mode
    TRP_DMD_FUSE_ID 5
    illumination: DMD is invalid for this system config error in DMD_Power()
    source: EEPROM defined
    source: SetUserSFGColor = 0
    source: DisplaySFG Color = 0
    source: DisplaySFG size = 960 x 1080
    pictcont: color profile defaults written to EEPROM
    illumination: Can't set up DMD
    sysmon: Fault in illumination transition to operating mode
    sysmon: Sysmon can't transition to operating mode
    sysmon: Transitionning back to Standby mode...
    illumination: Disabling SSI Illuminators.
    datapathf: autolock initialization passed
    digcontrolf: Init complete
    digcontrolf: powerNormal complete
    datapathf: dig powerNormal
    datapathf: datapath_ConfigFPGAatStartup complete - isUHD = 0
    pictcont: Setting the new system mode_num 1
    TempSensor Sample Environment:Temp Sensor Write Incomplete, returned f9
    illumination: API error in illumination transition to standby
    sysmon: Fault in illumination transition to standby
    digcontrolf: powerStandby complete
    pictcont: Setting the new system mode_num 1
    TempSensor Sample Environment:Temp Sensor Write Incomplete, returned f9
    TempSensor Sample Environment:Temp Sensor Write Incomplete, returned f9
    TempSensor Sample Environment:Temp Sensor Write Incomplete, returned f9
    TempSensor Sample Environment:Temp Sensor Write Incomplete, returned f9
    TempSensor Sample Environment:Temp Sensor Write Incomplete, returned f9
    TempSensor Sample Environment:Temp Sensor Write Incomplete, returned f9
    pictcont: Disable BrilliantColor
    pictcont: pictcont_SSI_CompleteSystemMode completed
    pictcont: pictcont_CompleteSystemMode completed
    pictcont: Set lamp waveform index if URT enabled for sequence mode number 0, cw mode 11
    pictcont: Setting the new system mode_num 1 completed
    datapath: DISP_SetDisplay failed (ret=0)
    datapath: Set the default system mode complete
    source: SPLASH_LoadImage reload addresses:  0xf9200000, 0xf7ff34b4
    source: SPLASH_LoadImage addresses:  0xf9200000, 0xffffffff
    source: Loading RGB -> RGB CSC table
    source: DisplaySplash size = 992 x 1080
    dispfmt: minPan, pan, maxPan:0, 0, 0
             data = { 992, 0, 992 }
    dispfmt: minScan, Scan, maxScan:0, 0, 0
             data = { 1080, 0, 1080 }
    ******* Try to set Color Profile 0 *******
    pictcont: Active Display is 2
    pictcont: Internal Source is displayed
    pictcont: Set gamma 0
    pictcont: Disable BrilliantColor
    pictcont: pictcont_SSI_CompleteSystemMode completed

    datapathf: Transition to SPLASH_AT_STARTUP
    pictcont: pictcont_CompleteSystemMode completed
    pictcont: Set lamp waveform index if URT enabled for sequence mode number 0, cw mode 11
    pictcont: Setting the new system mode_num 1 completed
    datapath: DISP_SetDisplay failed (ret=0)
    datapath: Set the default system mode complete
    datapathf: datapath_powerStandby - Transition to powerStandby complete
    systemmode: TwoDTable released
    systemmode: ThreeDTable released
    systemmode: XPRTable released
    systemmode: HiSpeedTable released
    sysmon: _sysStandby - Transition to powerStandby complete
    sysmon: Can't start projection
    digcontrol: StopCurrentOperation complete

    datapathf: Transition to SUSPENDED
    TempSensor Sample Environment:Temp Sensor Write Incomplete, returned f9
    TempSensor Sample Environment:Temp Sensor Write Incomplete, returned f9
    TempSensor Sample Environment:Temp Sensor Write Incomplete, returned f9
    TempSensor Sample Environment:Temp Sensor Write Incomplete, returned f9
    TempSensor Sample Environment:Temp Sensor Write Incomplete, returned f9

  • Matthias,

    The temperature read shouldn't be a problem; there may not be a sensor on the board.  

    The thing that is odd is this:

    TRP_DMD_FUSE_ID 5
    illumination: DMD is invalid for this system config error in DMD_Power()

    5 is the correct value for the p5/p47 UHD DMD; so something seems strange....  

    Can you go to the DLPC4422 GUI -> Projector Control -> DMD and click the little blue UP-arrow to read all items on the page (or click "Get" on all of the items on the page).  I am really interested in the Device Type; I think I expect it to be either 0x91 or 0x84.  

    Depending on the answer there, that may point us in the right direction.

    Thanks,
    Gar

  •  Hi Gary, see attached screenshot. Thanks Matthias

  • Matthias,

    That confirms my suspicion.  It seems the firmware deliverable on TI.com needs updating.

    In the meantime, we will need to discuss a near-term solution privately.  I will send you message request.

    Regards,
    Gary

  • Matthias,

    We confirmed offline that the DMDs were changed without updating the firmware deliverable that is posted on TI.com.  I provided you a package with new DLPC4422 firmware as well as new FPGA firmware.  I believe you were able to get an image on the DMD.  Hopefully this gets you going.

    Regards,
    Gary