• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 54 subscribers
  • Views 502 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.

CC2510 Low Power RCOSC Frequency Accuracy

Zach12504
Prodigy 80 points
Other Parts Discussed in Thread: CC2510

Hello All.

The manual states that the accuracy of the calibrated low power RCOSC is +/-1%.  It also appears that temperature and voltage variations further affect the accuracy.  My questions:

1. Is there anything I can do to improve the accuracy, or are there any conditions under which the accuracy will be better?  

2. Does the accuracy vary at some predictable rate over time, or is it just a random inaccuracy every time it is calibrated?  Please provide a little more information on this.

I'm trying to avoid using the external crystal oscillator due to space/money constraints.

Thanks.

Zach.

  • 0
    TI__Guru 71526 points

    There are some techniques you can use to keep this oscillator (RCOSC) closer to CAL.  First, the 1-percent accuracy is dependent upon the internal oscillator circuit, so given all the variable such as TEMP and Vcc, at a given set of conditions, nothing you can do to improve this accuracy.   The 1-percent accuracy is specified at 25C and Vcc = 3v, so this is the best accuracy you will get, at that temperature and voltage.    

    To keep the frequency closer to the initial CAL, provided the normal HS crystal is enabled and running.  As shown in table 6.6, CAL is continuous during the ON time.  However, you have no direct control over the Low Power RCOSC, so, to keep the RCOSC CAL, the HS Crystal needs to be kept as accurate as possible.

    Although I have never tried this, and have no data to confirm this will indeed keep the RCOSC from drifting, here is the theory.
    The CC2510 has an onboard TEMP sensor, and with it, you can create a CAL table vs TEMP for the PLL.  Refer to these note, one on the TEMP Sensor, one on TEMP compensation:
    If you keep the PLL as accurate as possible, and it is used to CAL the RCOSC, then the RCOSC will also be kept as close to the initial frequency as possible.
    -Leonard