CC3351: What are the effects of using the internal slow clock of CC3351?

Part Number: CC3351


Hi  TI‘experts

   The slow clock of CC3351, We plan to use the internal clock provided by CC3351. Will this have any impact on the WiFi during the sleep mode? Also, if we do this, will there be any effects at a temperature of 85℃?

  • When choosing the slow clock to be driven internally, which is less accurate that , the device is configured to wake up earlier and for a longer period of time, so that the device can readily receive the beacon from a connected AP at the correct time, resulting in higher power consumption. This will be seen when measuring the average current.

    I will need to check internally what are the effects at 85C, but I assume the internal clock frequency shifts very slightly, thus causing the device to wakeup earlier and again stay awake longer. Again, an external clock would be much more accurate.

    You can use our boosterpack to measure the current with the Slow clock used externally, and then remove it and measure the current again. 

  • Hi Sabeeh Khan

        I see the Operating Ambient Temperature specified in the CC3351 datasheet is -40℃ to 105℃. Under such environmental temperatures, can the internal

    clock accuracy still meet the working requirements of the CC3351 chip ?

  • Hi,

    We do test the CC33xx at the full range of temperatures from -40C to 105C. Let me include our hardware team to help get the response on this issue to see how we test internal vs external slow clk. 

  • HI Sabeeh Khan

       Let me include our hardware team to help get the response on this issue to see how we test internal vs external slow clk. 

    ---- When will these have results? and you test full range of temperatures from -40 to 105℃,Does this mean that under such environmental temperatures,  the internal clock accuracy can meet the working requirements of the CC3351 chip.

  • Hi Yixiong,

    We are still working with our internal R&D team to understand what data we have on this. I hope to have an update by early next week. 

  • Quickly mentioning that customer requires sysfs API to determine internal/external SLW clk.

  • Hi:

       Any update?

  • Hi Yixiong,

    I'm still working on getting the correct performance data from our R&D team - I have accuracy data and current consumption data that is quite old so I'm verifying if this data is still accurate with the FW improvements we have had over the last year.

    I do want to note that no direct RF performance is impacted by the slow clock accuracy - this is due to the Slow clock not being used in the RF chain. For direct RF impact, the fast clock would be the main contributor so its critical that the accuracy of this clock is kept as tight as possible. To summarize: typical RF characteristics such as TX power, EVM, Spectral mask, frequency accuracy, RX sensitivity, adjacent channel rejection, and PER performance will not differ between using the internal slow clock or the external slow clock. 

  • Hi Dylan Hubbard

       any update?Have you received the data from your R&Dteam?  I need know how you test internal vs external slow clk and the test date this week.

  • Hi Yixiong,

    Some of the information needed to be provided is sensitive in nature. I have sent the information to your field support team to get it to you offline.