• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 54 subscribers
  • Views 567 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.

CC1310: system occasionally goes into a state of perpetual restart

Zhiyong Li (FT)
Expert 2910 points
Part Number: CC1310

Hi,

The CC1310 based radios we are working on will occasionally go into a seeming perpetual state of reset right after: 1) we write some data into NVS, 2) trigger a reset via toggling reset pin. If this happens, we can recover the radio by downloading a fresh copy of firmware. After that, everything would work just fine. This doesn't happen often. In the past 3 years, this happened ~ 10 times in total. When we tried to reproduce it, none of the units being tested was bricked after 1000 cycles of NVS writing and reset. To be safe, we added a delay of a few seconds between writing into NVS and toggle reset pin, but I am not sure if that actually helps.

My question is, should this be a concern before we launch? Our system needs to run in the field without human intervention for years, it would be big financial drain if we have to deal with occasionally bricked units, even if it doesn't happen often.

Update: we gathered more information from the failed units. It seems system going into perpetual reset happened before external MCU started to force it rejoin network. The environment is quite noisy, surrounded with all sorts of heavy machinery such as flood pump, power supply lines, building HVAC etc. Presumably they were just shot dead by sudden electrical shot. 

Please advise,

Thanks,

ZL

  • 0
    Expert 2910 points

    On a side note, RF modules in general seem susceptible to ESD or other external interferences. When we were working with RF modules from other vendors, such as Xbee radios, we would brick some once in a while. Their manufacturer even had a service to help their customers to recover bricked modules, and there are numerous online tutorials teaching users how to recover bricked radios as well.

    Is there any good ways to prevent or minimize this from happening? 

  • 0 in reply to Zhiyong Li (FT)
    TI__Mastermind 20407 points

    Hi Zhiyong,

    Please check the errata and advisory regarding the CC1310 device. 

    There are some advisories and possible theories regarding your resets. 

    https://www.ti.com/lit/er/swrz062f/swrz062f.pdf

    Please check advisory 06, advisory 12. These might be something you are seeing.

    It is hard to comment on and provide suggestions on what exactly could be the issue without having a reliable way to reproduce it ourselves.

    Regards,

    Sid

  • 0 in reply to Sid
    Expert 2910 points

    Hi Sid,

    Again thanks for your reply.

    We don's use sensor controller so advisory 6 doesn't seem to apply. Also when the CC1310 units we have go into perpetual reset, neither reset or power cylcle can bring them back to life. Only downloading a fresh copy of firmware can resurrect them. So advisory 12 doesn't seem to apply either.

    Other posts on this forum mentioned that voltage drop when accessing flash could lead to flash corruption. I think this may have something to do with our case. We use primary li-scl2 batteries plus a supercap as power source. In theory, when sensor nodes attempt to join network, as a result of continuous packet sending which draws up to 380mA, voltage from supercap can drop below the minimum requirement 1.8V. If flash writing or erasing attempt is made at this point, I can see this could lead to corruption of flash. If there is no other alternative, I will go by this theory.

    My follow up question is, will attempt to read flash when voltage is below 1.8v also cause flash corruption? Because can manage to avoid flash writing into NVS and erasing, but it is very hard to avoid reading at all times.

    Best,

    ZL

  • 0 in reply to Zhiyong Li (FT)
    TI__Mastermind 20407 points

    Can you please point me to the threads you mention, I need to ask RnD regarding flash corruption at 1.8V. It would be great to get some context regarding this issue.

    Regards,
    Sid

  • 0 in reply to Sid
    Expert 2910 points

    Hi Sid,

    This is the most relevant post I can find now.

    https://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/787215/cc1310-cc1310-firmware-is-getting-corrupted-when-drop-the-voltage-under-1-8v

    I probably have read several other posts, but unfortunately I cannot locate them now.

    Thanks,

    ZL

  • +1 in reply to Zhiyong Li (FT)
    TI__Mastermind 20407 points

    Hi Zhiyong,

    The datasheet also suggests 1.8V and above supply voltage. As the thread suggests, if you actually use it below 1.8V, there is a chance of misbehavior. 

    Regards,

    Sid