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.

OMAP3503 65930 not booting

Other Parts Discussed in Thread: OMAP3503, TPS65950, TPS65930

We are seeing a problem booting our OMAP3503 with 65930 also. We have clocks and reset signals that look ok. We also verified sysboot straps.

We have set MMC1 as the first boot device and verified the same boot setting configuration works on our EVM kit. We are not seeing any VMMC1 voltage (0) but all other voltages look OK.

We have our SD/MMC card detect brought into a GPIO on the OMAP instead of the more common GPIO0 on the 65930. I understand this is OK as I believe the firmware boot loader does not depend on the CD input and the "disable VMMC1 voltage on no CD" feature of GPIO0 is not enabled by default. Are these statements true?

Can you think of anything else we can try?

DV

 

  • Hi,

    I dont think card detect is your problem. Can you boot OMAP in other modes and check if the I2C writes are proper. If in MMC mode the I2C write doesnt work properly then the I2C command from OMAP to TPS device will not configure VMMC1.

    You can try to check/probe if the I2C command is ok on the bus.

    Are the connections on the problematic board and EVM kit identical? I dont know what you refer when you say EVM kit.

    Let me know some details, those may be helpful.

     

    Regards,

    Gandhar.

     

  • By the EVM I am refering to the TI/Minstral OMAP35XX EVM kit. We have the SYS_BOOT switches on the EVM set to match our board. EVM boot proper but our board does not. The EVM uses OMAP 3530 and 65950 and our board uses OMAP3503 and 65930.  We have an external oscillator so we have sys_boot6 pulled up to 1.8v.

    We have both sys_nrespwron and sys_nreswarm directly tied from OMAP to 65930 with pullups on each line. CLK_REQ is also tied from OMAP to 65930 with pullup. 26MHZ clock and and 32khz appears to be OK.

    We get inconsistant operation with JTAG attached to OMAP. Sometimes it appears to be reset cycling approx. every 5 seconds but we see nothing on the nreswarm line (its always high). The OMAP appears to never get initialized properly and Code Composer Studio fails trying to execute the initialzation gel file. When JTAG attached to EVM I can see gel file complete and I can step through the on chip firmware boot code. I can not duplicate this on our board.

      

  • Hi,

    The two devices TPS65950 an dTPS65930 are HW and SW compatible. You shouldnt be seeing any difference in behavior due to this.

    Which external oscillator are you referring to above? I dont know whatthe sys_boot settings should be on the OMAP. Why would the external oscillator change the sys_boot settings. If the sys_boot is not the same as on Mistral EVM then I am not sure if OMAP is going to follow the same boot up procedure. I dont know if sys_boot[6] pin is in any way related to MMC booting.

    Did you check your I2C bus to see if you can see the correct I2C command from OMAP to TPS65930? Which I2C port are you using on OMAP and TPS65930?

     

    Regards,

    Gandhar.

     

  • Hi,

    Do you still have this problem. Were you able to check I2C writes and I2C port for OMAP?

     

    Regards,

    Gandhar.

     

  • Yes, we found the problem a trace on the Vcc_PLL of the OMAP was not connected therefor internal clock was not being generated.

     

    DV

     

  • hi

    i'm facing the same problem. Could you tell me how to reslove it?

  • Hi,

    If you can share some observations then that can help.

    If its the same problem then try the same fix.

     

    Regards,

    Gandhar.