• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 98 subscribers
  • Views 331 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Clock overdriving spec for Sitara?

Wallace Ly
Prodigy 240 points
Other Parts Discussed in Thread: AM3352

Hi,

I have a customer in the bay area with a question regarding overdriving the Sitara clock lines. Please see below. Any comments are greatly appreciated!

The short of it is that customer is concerned about overdriving the crystal which has a 100 microwatt MAX spec.  By changing the 330 ohm series resistor to 0, we risk overdriving if there is no current limiting function inside the Sitara.  So can you verify that there is a current limiting feature within the Sitara and that we have no issue with overdrive?  Also does the Sitara ESR spec apply to the raw crystal, or to the whole crystal circuit?

 
Thanks in advance,

Wally

  • 0
    TI__Guru 62855 points
    Please specify which Sitara device you are asking about. Also provide more details related to your 330 ohm series resistor reference. I assume you are asking about a crystal reference circuit with a 330 ohm series resistor. However, I'm not familiar with any crystal reference circuit that define a specific series resistor value.

    Regards,
    Paul
  • 0 in reply to peaves
    Prodigy 240 points
    Hi Paul,

    Thanks for feedback. Here is additional clarification from the customer.

    The customer is using the AM3352. The 330 ohm series resistor (inline with the crystal noted as Rd in Fig 6-9 of the AM335X datasheet) was added by them, as previous experience showed us that conventional CMOS inverter-oscillators will overdrive the crystal if the output current is not limited. The intended crystal has a 100 microwatt max spec. According to the xtal Pd calculations from table 6-3 in the datasheet, they would be driving at about 600 microwatts with Rd at 0 ohms. T.I.’s schematic review wanted us to remove the 330 ohm resistor, saying that the crystal ESR should be 50 ohms max.

    They need clarification on the ESR spec (in ref: Figure 6-9/table 6-3) as to being the xtal only or the whole xtal circuit. My assumption is that the ESR spec applies to the raw xtal.

    Best Regards,
    Wally
  • 0 in reply to Wallace Ly
    TI__Guru 62855 points
    The ESR specification in the data sheet is for the crystal.

    Our recommended crystal circuit has an optional series damping resistor inserted between the oscillator output and the resonant crystal circuit. We have seen examples of damping resistors being inserted in series with the crystal such that it is part of the resonant crystal circuit and this effectively increases the ESR of the crystal. Adding the resistor in series with the crystal increases the impedance of the resonant circuit. Adding a resistor on the oscillator output doesn't increase the impedance of the resonant circuit, but reduces the drive level of the feedback required to maintain oscillation.

    The data sheet provides a formula for crystal power dissipation, but it assumes there is no damping resistor. If they insert a damping resistor as shown in our recommended crystal circuit, the voltage across the crystal would need to be measured and used in place of VDDS_OSC to determine the power dissipation of the crystal. They should adjust the value of Rd to maintain a safe power dissipation for the crystal. Once they determine the value of Rd required to provide a safe power level to the crystal, they should increase the value of Rd by a factor of 3 - 5 times while testing the product across all environmental conditions to confirm there is enough gain margin to start oscillation.

    Regards,
    Paul