CC2652P7: RF issue during stability test temperature shock

Part Number: CC2652P7

Tool/software:

Hi, 

Customer is doing stability test, fast temp change between -40~85. 

Zigbee protocol. ZR role. 

if 30s network sync, the device's RF has issue to communicate. looks the PLL is unlocked based on the snapshot of Spectrum (right one is the issue happens). the M4F is still working well. after reset, all is recovered. 

if 5s data transfer, almost works normally. 

guess there is a temperature calibration/compensation mechanism? how to optimize it?

BR. Albin

  • Hi Albin,

    There is temperature compensation for the PA, see 25.3.3.2.16 from the TRM (https://www.ti.com/lit/ug/swcu192/swcu192.pdf). But this is only a compensation coefficient to modify one of the other TX power parameters depending on the delta from 25C. 

    Please provide some additional detail on the testing:

    • How rapid is the temperature change
    • In what direction is the temperature change
    • Is the entire system seeing the temperature change (i.e. HFXT + CC2652P7)?
    • how is the temperature test performed (i.e. cryofluid, temperature chamber swap, cold spray, etc.)
    • Is external load capacitance being used with the crystal?
    • What is the customer requirement here?
    • What is the TX power being used?
    • Is this seen on multiple devices or just a single device?

    Thanks,

    Jake

  • Hi Jake, 

    Thanks for reply, i will ask customer to input the questions directly. 

    but one confusion from me, looks the PA temp coef only impact the tx power right? but based the picture customer shared, looks the frequency is not expected. 

    BR, Albin

  • Hi Albin,

    Like I said above, I agree with you that the temp coefficient only affects the TX power setting. Beyond this there's no additional temperature compensation I can direct you to. So it seems that this is something we need to understand what the customer is trying to do and if there's debug required to identify the source of their issue.

    Jake