• State Verified Answer
  • Replies 13 replies
  • Subscribers 67 subscribers
  • Views 6245 views
  • Users 0 members are here
Attachments (0)
Nordic Case Info
  • Case ID: 101174
Options
This post is older than 2 years and might not be relevant anymore
More Info: Consider searching for newer posts

0xDEADBEEF, 722, src \ ll_adv.c

Peter

Hi all,

I use nRF51822, sdk_v5_0_0_34603 and s110_nrf51822_6.0.0-5.beta. In a different periods of time appears error "0xDEADBEEF, 722, src \ ll_adv.c". Error appears only with battery powered device (tested with different batteries). What may be the reason for this error.

Best regards, Peter

  • 0 in reply to Asbjørn Sæbø

    Peter, since you do not use flash write, this issue does not touch you.

    Also, what you do, changing adv params and starting advertising after disconnect or advertising timeout is perfectly OK. (After connection, there is no advertising to stop, the same for after advertising timeout.)

    Asbjørn

  • 0 in reply to Asbjørn Sæbø

    Peter, looking at your code, you are actually doing flash operations, although you may not be aware of it. You use the bond manager (from the SDK), and the bond manager writes to flash. So the issue and workarounds above do apply to your case. (This may not be the cause of your issue, though.)

    Asbjørn

  • 0

    Thanks for suggestions. I'll try workarounds and report results.

    Best Regards, Peter

  • 0

    Hi all,

    Here are results of some tests:

    • add 50ms before the execution of command sd_ble_gap_adv_start() - the error exist.
    • reducing supply voltage of mother board with nRFgo_Studio and test with PCA10004 - reduced to 2V and there are no errors. But when PCA10004 is powered with battery the error exist.
    • using of 2 batteries in parallel for powering my pcb - the error exist.
    • upgrade to sdk_5_1_0 and s110_6.0.0_softdevice - the error exist.
    • sdk_5_1_0, s110_6.0.0_softdevice and example project ble_app_hrs (add led in assert_nrf_callback()) - the error exist.
    • downgrade to sdk_v4_4_2_33551 and s110_5.2.1_softdevice - the problem disappeared. The last test clear my doubt of defect in used batteries.

    I noticed that when there is no error, DEC2 is ON for a short period of time until rx/tx are active and OFF for a long period of time until sleep state. But when the error appeared DEC2 is constantly ON.

    I hope it will be possible for Your team to reproduce the case and find out the reason for the error. Please use my code and battery for power supply. May be in my pcb exist increased contact or internal battery resistance. Capacitor on pin 1 (VDD) is 1uF according "nRF51822_Development_Kit_1_0" instead 4.7uF as proposed in "nRF51822_Development_Kit_2_0". Chip is N51822 QFAACA.

    Thank You very much for assistance. Best Regards, Peter

    ble_app_hrs_sdk_5_1_0_p_k.rar

  • 0

    Hi Peter,

    The issue you see is due to a slight incompatibility between your HW revision and the s110 6.0.0 SoftDevice. As you can see from the S110 release notes and the nRF51822 compatibility matrix in the nRF51822 migration document (nWP-018), this SoftDevice version is not production level tested on older revisions of the nRF51822.

    A workaround that enables you to do development wit S110 6.0.0 on older hardware, is to implement the fix for PAN ID 11 (see nRF51822-PAN) while advertising. In short, this is setting the power mode to NRF_POWER_MODE_CONSTLAT, see the nRF51 Series Reference Manual (nRF51 RM). The drawback is a slightly higher power consumption while the power mode is set to NRF_POWER_MODE_CONSTLAT.

    On newer revision hardware (see the compability matrix), no workaround is necessary.

    All the documents mentioned can be found in the "Downloads" section of the nRF51822 product page on nordicsemi.com: www.nordicsemi.com/.../nRF51822

    Edit: Typos

Related