• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 54 subscribers
  • Views 1393 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.

RTOS/CC1310: RSSI and Oscillator Settling Time and Current Usage

Chad Christensen
Expert 1670 points
Part Number: CC1310
Other Parts Discussed in Thread: CC1200, CC1120

Tool/software: TI-RTOS

We are using the CC1310 Launchpad to measure average power consumption of a basic polling receiver using the sniff mode example code. We are measuring almost the exact same curve shown here (https://e2e.ti.com/support/wireless_connectivity/proprietary_sub_1_ghz_simpliciti/f/156/p/547760/2005088

While the actual RX active time looks pretty good, there is a lot of time before this that is taking about 4mA. I'm guessing much of this time is oscillator settling time, and this adds up to a large part of our power budget. The average power for checking the radio receiver, assuming a wake-up once per second, is about 10.5uA.

Is there any way to reduce this current consumption or time? We've experimented with the RX bandwidth and datarates, but these only slightly shorten the RX active time, and have no affect on the timing / current draw before the RX becomes active.

This is actually worse than our older CC430, which only took about 1mA during this time. The average power for checking the radio receiver, assuming a wake-up once per second, was only about 7.2uA, and this was for checking 2 channels! Here's a plot from our older CC430 performing the same radio check for two channels instead of just one:

  • 0
    TI__Guru* 79481 points

    Hi Chad

    When going from standby to RX in the WOR example, the Rf core must be powered on, the crystal must stabilize and the radio setup command and the CMD_FS must run. In addition all patched for the particular PHY you are using must be loaded.

    Unfortunately that takes typically around 1.6 ms and there is not much that can be done with respect to make this time faster/current lower.

    BR

    Siri

  • 0
    TI__Genius 14860 points

    Hi,

    slow wake-up time is indeed a weakness of the CC13x0/CC26x0 device family.

    Chad Christensen said:
    We've experimented with the RX bandwidth and datarates, but these only slightly shorten the RX active time, and have no affect on the timing / current draw before the RX becomes active.

    You are right. RX bandwidth has only minor influence. See this table about RSSI settling time for CMD_PROP_CS/CMS_PROP_RX_SNIFF.

    Chad Christensen said:
    Is there any way to reduce this current consumption or time?

    I don't think so. Before going into RX, the CC1310 enables the external HF crystal, powers the RF core up, loads firmware patches and switches to the external oscillator. This happens in parallel, but the chip wakeup takes 300 µs, the RF core power-up procedure takes about 1 ms where the CPU is in active state, that's why you see such a high current consumption. On the CC430, everything radio-related was solved in hardware.

    Chad Christensen said:
    This is actually worse than our older CC430, which only took about 1mA during this time. The average power for checking the radio receiver, assuming a wake-up once per second, was only about 7.2uA, and this was for checking 2 channels!

    I don't know the current consumption of the CC430, but for CC1310 sniffing on 2 channels once per second the average current consumption is about 12 µA at 3.6V.

  • 0 in reply to Richard W.
    Expert 1670 points

    Thank you for the detailed answer.

    However, it does seem a bit disappointing that the new CC1310 takes 66% more current as a polling receiver than the older CC430. The CC430 had an option to disable the frequency synthesizer calibration, which significantly improved the RF core power-up time. We would manually calibrate on power-up, store and load the calibration values with each RF sniff cycle, and then re-calibrate periodically or after temperature changes. 

    Is the CC1310 doing something similar, or capable of doing something similar? Would this reduce the RF core power-up time?

    This is a substantial part of our power budget - are there any other recommendations to improve this?

    I'm assuming the CC2610 takes the same time / power?  Are there any other radios offered by TI that would be better?

    Thank you,

    Chad C.

  • 0 in reply to Chad Christensen
    TI__Guru* 79481 points

    Hi Chad

    It is not possible to turn off calibration the way we do it on our older devices, so there is not much to do with the timing unfortunately.

    You can take a look at the Sniff mode feature for our transceivers (CC1120 and CC1200). There is an Excel sheet in the app note (http://www.ti.com/litswra428) that will calculate the current consumption for you for different settings/preamble lengths.

    BR

    Siri