• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 439 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.

TM4C1290NCPDT: Passive crystal oscillator startup problem

Allen Lu
Intellectual 2460 points
Part Number: TM4C1290NCPDT

Hi Team:

 My customer trying to use TM4C1290NCPDT with passive crystal oscillator in their system.

As pre check with customer, after the reset signal is pulled high, it takes 15ms for the TM4C1290NCPDT to start up normally after the external passive crystal oscillator is turned on. Please help me to confirm the following questions, thank you

1. For the 12ms low level of the crystal oscillator after the reset signal is pulled high, can it be judged that the crystal oscillator is not driven to start oscillation at this time?
2. After the reset signal is pulled high and the chip starts to run from the main function, how long does it take? Is it about 12ms?
3. customer tested other TI`s  DSP chip on their side, and the passive crystal oscillator starts running as soon as it is powered on. There is no need to wait too much. Please briefly explain the difference between DSP and MCU for passive crystal oscillator configuration?
4. if we think that 1290N recommends the use of active crystal oscillators, please also provide reasons why passive crystal oscillators are not suitable.

 tks for your confirmation.

  • 0
    TI__Guru**** 191086 points

    Hi,

      If you see between 12 to15ms on the startup time then it is within the limit of the spec. Please find details in the datasheet under the MOSC Characteristics. 

  • 0 in reply to Charles Tsai
    TI__Intellectual 2460 points

    Hi Charles：

     Tks for the comments.  As I've seen it, we've found that it takes more than 30ms to start up.

    The attached waveform yellow is the reset signal, and blue is the crystal oscillator signal. There are two types of test conditions at present.
    1. Under normal circumstances, the start-up time reaches 15ms, which is very close to the 18ms required in the specification.
    2. When there is a fault, it will take 30ms to start the vibration.

    Normal case within 15ms:

    so i think the 15ms is in the specification, this is clear, but it is also critical, and has been found to exceed this time (the second picture in the picture ),
    So I want to know what is the reason for such a long startup time, what is the internal mechanism to control it, and is there any way to compress this startup time to a more reliable range, thank you

  • 0 in reply to Allen Lu
    TI__Guru**** 191086 points
    Allen Lu said:

    2. When there is a fault, it will take 30ms to start the vibration.

    Hi Allen,

      Can you elaborate what type of fault the customer is referring to? 

      Is the 30ms observed on the same board or a different board?

      I'm not an expert on oscillator. What is the load capacitance (C1 and C2) you place on the crystal. In the datasheet Electrical Characteristics section, there is description on how to calculate C1 and C2. The load capacitance must be chosen to meet what is specified by the crystal manufacture. The datasheet also has a table listing different crystals that have been simulated and confirmed to operate within the specifications for MOSC Input Characteristics. If the customer's chosen crystal is of the one that is listed in the table or ones that have nearly identical crystal parameters then they are expected to work as well.