• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 54 subscribers
  • Views 840 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.

CC1310: RX problem

Milorad Podoaba
Intellectual 765 points
Part Number: CC1310

Hi,

We have manufactured until now several hundreds of sensors based on CC1310. Apparently we have high number of sensor board with receiving problems on the last batch. The sensor transmit but in doesn't receive.

In the past we had the first batch with no major RF issues.

The second batch was bad. We had multiple HW issues including RX issue. We thought than maybe some manufacturing problem caused these issues.

The third batch was again OK.

Now the current batch we have only RF receiving problems. It is a high failure rate like 50%. After a first analysis of the boards, we couldn't find any soldering problems.

In order to find HW problems during manufacturing we have a production test firmware. For the RF test we use 917MHz LRM and a "golden unit" which TI CC1310 EM board. It has a custom firmware inside. The test consists of sending multiple time a 30 bytes packet and wait to receive same packet form the golden unit. Before any transmit the target board is listening for clear channel. The RX timeout is set to 100 ms enough to catch the SYNC word.

On sensors with the RX problem,  we have seen that the golden unit received the packet and transmitted back but the target unit(sensor) didn't receive. Afterwards the target unit tried to transmit the next message without success because of the CCA It sensed the ongoing message from the golden unit. So we concluded that the RF chip did detect the SYNC word and we do not know what can cause it.

Has anyone had the problem before? Or maybe to point out what could be the cause?

Thank you and regards,

Milorad
 

  • 0
    TI__Guru* 79271 points

    Hi Milorad

    To me it sounds like a HW/RF problem, but to rule out SW there is a couple of things you could do:

    1. Verify that the golden unit is transmitting. This can be done by using a third CC1310 with SmartRF Studio and sniff the communication to make sure that it is the RX on your sensors that is the problem.
    2. You should try to download some simple code (rfPacketRX) on the units that fails that simply sets the unit in RX and wait to receive. You could use SmartRF Studio on the transmitter. A test like this would rule out SW issues related to timing/timeouts/the CCA algorithm etc.
    3. Have the debugger connected when running the code that fails would be useful to verify that sync calibration has been performed OK, that the radio actually enters RX state etc.

    On a failing unit, does it receive some of the packets or does it not receive at all?
    I am sorry that I could not give you any more help at the moment, but hopefully doing the tests suggested above will narrow down the
    problems so that it will be easier to find the root cause of the problems you are seeing.

    BR
    Siri

  • 0 in reply to Siri
    Intellectual 765 points
    Hi Siri,

    1. the golden unit is transmitting fine, some of the target boards are receiving from it.
    2. i have tried different modulation(2 gfsk), bigger rx timeout and no cca, still the same
    3. i do not have a jtag connection on board, we use the bootloader to load the firmware; i think that the chip is entering in rx mode, the cca is working

    Most of the failing units are not receiving at all, but some are receiving something.

    Well if it was a HW RF problem I would suspend the matching circuit, the balun filter(we use johanson 0850BM14E0016) or the CC1310.
    Which one would cause the problem?

    Regards,
    Milorad
  • 0 in reply to Milorad Podoaba
    TI__Mastermind 39860 points
    If you share your schematic and gerber files I can do a review. If you do not want to share this information on the forum you can befriend me and send the files directly to me.

    Do you have the possibility to perform conducted testing on the failing units? That is, measure output power with a spectrum analyzer and sensitivity with a signal generator. What are the numbers?

    If you have CC130EM's that are working you can unsolder a failing CC1310 from your sensor and solder on to the the EM. You can then check if it is the CC1310 that is faiiling (since you will then be testing on a good-known working CC1310EM).
  • 0 in reply to Sverre
    TI__Guru**** 317180 points
    Are you able to measure the accurate center frequency you send with and compare between the batches? In other words, make sure you don't have a frequency offset in the system.
  • 0 in reply to Sverre
    Intellectual 765 points
    @Sverre

    Thank you for the offer. I'm going to contact you via email to send any files.

    We don't have any spectrum analyzer. We used only network analyzer to do the matching.

    Replacing the RF chip is a good idea. We are going try that.

    Regards,
    Milorad
  • 0 in reply to TER
    Intellectual 765 points
    @TER

    In order get the center frequency we used carrier wave and check the rssi on different frequency. We used the golden unit as a receiver.
    Depending on the result we are doing adjustment. This procedure is done before any production test.

    I know that this problem is like a frequency shift but only on receiving. I do not understand how its possible.

    Regards,
    Milorad