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.

TMS320F28069 On-Chip Crystal Oscillator Internal Block Diagram

Hi Team,

   My customer using 3.3V external clock source directly connected to X1 pin with TMS320F28069PZP usage, and the XCLKIN used as input GPIO, and reported an that when the XCLKIN input 3KHz(also tried 1Khz-100Khz) Square waveform, the F28069 reported work at abnormal states.

  Checked with customer, and configured the software as below:  and if XCLKIN input 3Khz(also tried 1kHz-100Khz) square waveform,  the F28069 still can work. 

    EALLOW;

     SysCtrlRegs.CLKCTL.bit.XTALOSCOFF = 0;     // Turn on XTALOSC

     SysCtrlRegs.CLKCTL.bit.XCLKINOFF = 1;      // Turn off XCLKIN

     SysCtrlRegs.CLKCTL.bit.OSCCLKSRC2SEL = 0;  // Switch to external clock

     SysCtrlRegs.CLKCTL.bit.OSCCLKSRCSEL = 1;   // Switch from INTOSC1 to INTOSC2/ext clk

     SysCtrlRegs.CLKCTL.bit.WDCLKSRCSEL = 0;    // Clock Watchdog off of INTOSC1 always

     SysCtrlRegs.CLKCTL.bit.INTOSC2OFF = 1;     // Turn off INTOSC2

     SysCtrlRegs.CLKCTL.bit.INTOSC1OFF = 0;     // Leave INTOSC1 on

     EDIS;

Need you kindly support on below questions:

1.

 

     EDIS;

Reproduced the Step1 operation, no abnormal reported as step1;

 

Support Need:

As datasheet mentioned, the XCLKIN can connect to 3.3V external clock source, X1&X2 can used as clock source with on-chip crystal oscillator, and X1 voltage is limited in 1.8V level.   Currently customer hardware design is out of TI F28069 spec required(X1 as external 3.3V clock source), and plan to modify to correct way, but for the auto shipped out, customer want TI to provide below support to help them to evaluate their next step operation.

1. Share the On-chip crystal oscillator internal schematic or block diagram, as customer connect X1 to 3.3v, which required 1.8V,  customer need the schematic or diagram to evaluate the risk;

2. Give comments for the reason why on-chip crystal oscillator is off, but the X1 signal still get through and worked as clock source?

3. other solutions with current hardware?

 

Best Regards

Benjamin

 

  • Hi Benjamin,

    If the customer is using a 3.3V-level single-ended oscillator on X1, they are violating the datasheet in two ways:
    a. "The on-chip crystal oscillator X1 and X2 pins are 1.8-V level signals and must never have 3.3-V level signals applied to them."
    b. "The X1 pin is not intended to be used as a single-ended clock input, it should be used with X2 and a crystal."

    1. The oscillator schematic is sensitive design information and cannot be released to the public. The risk is in reliability. The datasheet is being violated and as a result could be causing damage to the device. It is necessary to inform the customer to not operate the device in this manner and that the board should be modified to meet datasheet requirements.

    3. I don't believe that there will be any other solutions with current hardware, but if can you share the clocking part of the schematic I can take a look.

    Best Regards,
    Adam Dunhoft
  • Hi Adam,

      Very appreciated for your comments. Has highlight the recommendation of datasheet required, and push customer for the next version hardware modification.

      The clock part number is KC7050A20.0000C3GESP;  and the detailed schematic has updated. Expect for your  further risk analysis, thanks.

    Best Regards

    Benjamin

  • Hi Benjamin,

    Just to confirm:

    The output of the above oscillator is now connected to XCLKIN? It should not be connected to X1.

    Also, did you send the schematic? I have not received it if you did.

    Best Regards,
    Adam Dunhoft
  • Hello Benjamin,

    Is this issue still ongoing or can I close this thread?

    Best Regards,
    Adam Dunhoft