• State
  • Locked Locked
  • Replies 7 replies
  • Subscribers 54 subscribers
  • Views 1357 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

CC2640: CC2640: Delayed init when device wakes up from sleep or completely discharged power up

Rohit VV
Prodigy 80 points
Part Number: CC2640

Hello, I am using CC2640 with standby mode and deep sleep mode configured.
I see an unexpected delay in the device to go to standby mode from the task when the device wakes up from deep sleep or when the device is fully discharged before powering up, or when I pull the reset PIN for more than 2-3 seconds and release.
The power monitor shows an increased power consumption during the particular time when compared to a quick PIN reset.

  • 0
    TI__Mastermind 33460 points
    Hello,

    There is expected latency when trasitioning from shutdown(deep sleep) mode to active mode. This is because there are different power domains and clocks retained in standby vs shutdown mode. When waking from shutdown mode additional powerup logic must run.

    You can find a table that summarizes each mode along with its associated latency and power consumption in table 6-5 of the TRM
    www.ti.com/.../swcu117

    Does this match what you are observing?
  • 0 in reply to Sean2
    Prodigy 80 points

    Hi Sean,

    Thanks a lot for the reply,
    As I told, I face the issue in delay when device goes from active to standby mode when a completely discharged power up is done, but the delay does not exist when a quick reset is done.
    Can you please confirm if this behavior is expected



    I just wanted to know what is the reason for the spikes and delay in the second case.
    This is taken from a project based on Simple broadcaster example from TIRTOS

  • 0 in reply to Rohit VV
    TI__Mastermind 33460 points
    Hello,

    Are you using a TI reference design or LaunchPad to measure this?

    If you are using large decoupling capacitors, which could be powering the system after the power is removed.
  • 0 in reply to Sean2
    Prodigy 80 points

    Hi Sean,
    This measurement was made with CC2650EM-7ID Evaluation Module Bluetooth chip (on SmartRF06 Evaluation Board)
    I used a power monitor to measure the power, please have  a look at the behavior

  • 0 in reply to Rohit VV
    TI__Mastermind 33460 points

    Hello,

    Thanks for the information. During the TI-RTOS boot process the SetupTrimDevice() function will be called which initializes the various power/clock domains of the device based on the reset type.

    I believe there is a possibility that during the shorter reset period some registers/memory is still retained and thus the trim process is less intensive or takes longer.

    I have reached out to our hardware team for confirmation on this.

  • 0 in reply to Sean2
    TI__Guru** 106705 points
    Hi guys,

    A pin reset will always reset the HW in the same way, there is no difference in register retention. What could be happening here though is that the crystal oscillator start-up time is different. With the short reset, there could be residual oscillations in the crystal which makes it start up faster.

    Regards,
    Fredrik
  • 0 in reply to Fredrik K
    Prodigy 30 points

    Hi Frederik,

    thanks for your reply. Just to make it clear again - we have an unexpected delay when going from ACTIVE mode to STANDBY (dependent on the reset time before - but we have always started the ACTIVE mode first and we then go to STANDBY).

    For my understanding in ACTIVE mode the HF as well as the LF is active.

    In STANDBY mode only the LF is active.

    So

    1. when going from ACTIVE to STANDBY after "long reset" this means that the LF is not started up properly yet (although the HF is already running), and this will cause the longer delay.

    2. when going from ACTIVE to STANDBY after "short reset" the LF is still oscillating which will result in the shorter delay.

    Correct? (for LF as well as HF we use external quartzes)