• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 56 subscribers
  • Views 2186 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.

CC2630: Moving our design from 7x7mm RGZ package to 5x5mm RHB package - radio RX not working?

Ben Renneberg
Prodigy 130 points
Part Number: CC2630
Other Parts Discussed in Thread: TIMAC

We have been using the 7x7mm RGZ package successfully in our designs for almost 2 years. We designed a new PCB with the 5x5mm RHB package and now have an interesting radio issue.

CC2650RGZ (devkit) when communicating with CC2630RHB (our new PCB) OR CC2630RGZ (our original PCB), with everything sitting on a desk.

Test 1:

CC2650RGZ measures about -50dBm RSSI

CC2630RHB measures about -100dBm RSSI

Test 2:

CC2650RGZ measures about -50dBm RSSI

CC2630RGZ measures about -50dBm RSSI

When using a sniffer, we see similar RSSI from both transmitted signals, so it seems that the issue exists in the receive pathway on the CC2630RHB PCB only. The only components (on both CC2630 boards) between the RF_MCU_SE signal in the schematic below and the CC2650RGZ devkit are an RF switch, two antennas, and air ... all of which should work with reciprocity, which makes me suspect we have something wrong in the MCU RF settings. Are there differences in the RF settings between the CC2630 RGZ and RHB packages?

See schematic snip below:

  • 0
    TI__Mastermind 27145 points

    Hi,

    I observed that you didn't put DC Blocking Capacitor (12pF) at the output of the filter (in series with L10T). This could be the reason for not working properly.

    Please check.

    Thanks,

    PM

  • 0 in reply to PM
    Prodigy 130 points

    Hi PM, I had posted an incomplete schematic image. RF_MCU_SE connects to the RF switch below, which has DC blocking capacitors.

    The "No Load" components are optional tuning components which we have not loaded yet.

  • 0 in reply to Ben Renneberg
    TI__Mastermind 27145 points
    Hi,

    Did you check the Switch operation (RFC TO RF3 path)? You can check the loss from RFC to RF3?

    If it is possible you can disconnect C62T from the Switch and the feed the RF Signal into C26T and check the Receiver performance.

    Thanks,
    PM
  • 0 in reply to PM
    Prodigy 130 points
    Hi PM,

    I have the RF switch on a test coupon, and see ~1dB loss through the switch (S21 and S12) for all ports. I only have 5 test PCBs, and adding a pigtail to C62T will sacrifice a board, so I'm hesitant to do that until I'm desperate (2 more days from now, probably).

    Does TI have an eval board with the RHB package that I can purchase for comparison?

    Thanks,
    Ben
  • 0 in reply to Ben Renneberg
    Prodigy 130 points
    Another note: The test coupon with the RF switch was a copy-paste of the layout of the full test PCB, so I have high confidence that it should behave the same way as on the main board. Even when testing invalid control signals, the RF switch has the same loss bidirectionally and the issue we are observing seems to only be on the receive side.
  • 0 in reply to PM
    Prodigy 130 points

    Hi PM,

    I modified a board to completely bypass the RF switch (soldered a coax pigtail to an antenna from the COM trace), so the schematic is the same as the first image in my original post with a 12pF series cap to an antenna.

    The RSSI on the RHB package is 46dB less than the RSSI on the RGZ package for this COM link. This proves that the RF switch is not the issue.

    Please let me know if there are any register settings on the receive pathway that I should be looking into to resolve this issue, or if you have other ideas.

    Thanks,

    Ben

  • 0 in reply to Ben Renneberg
    TI__Mastermind 27145 points

    Hi,

    Are you using SmartRF Studio?

    If not, did you generate the Register Settings from the SmartRF Studio?

    You have an option to select the Type of the Device (7ID or 5XD). In your case, you need to select 5XD. Please try with THE 5XD Register settings.

    Are you getting the same performance on all 5 boards?

    Thanks,

    PM

  • 0 in reply to PM
    Intellectual 550 points
    Hi,

    I work as a firmware engineer with Ben, and we finally resolving this issue. First off, internal bias was chosen for the hardware design. This was different than the default setting in the TIMAC library for the 5x5 package, which is set up for external bias. I had to update the bias setting in the Macs.h file to match the hardware. Also, after I made the change, it turned out that the following definition of the internal bias bitfield doesn't work with my IAR compiler version 7.40.3:
    #define RF_FE_INT_BIAS (0<<3)
    This definition, when I printed it out, resulted in the value of 8. So I had to change it this way to make it 0:
    #define RF_FE_INT_BIAS (0ul<<3)
    After that, everything worked as expected with comm RSSI values being reported the same as on our previous products with 7x7 package chip.