• State
  • Locked Locked
  • Replies 7 replies
  • Subscribers 53 subscribers
  • Views 1006 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.

CC430f6137 + CC1190 Communication problems

Carlos Silva46
Prodigy 60 points
Other Parts Discussed in Thread: CC1190, CC430F6137, CC1101

Hello,

Using the schematic and layout from the CC1101-CC1190EM 869MHz evaluation module (http://www.ti.com/tool/cc1101-cc1190em869rd) my company designed a board with a CC430f6137 and a CC1190 range extender. We replaced some of the smaller components (0402) by bigger ones (0804) with the same characteristics, since hand soldering 0402 components would be a very time consuming task.

We adapted the “Fixed_LT_FIFO” code provided in (http://www.ti.com/lit/zip/slac525) to continuously send and receive data and the SmartRF Studio application was used to obtain the register values.

Using the SmartRF Studio version 1.16.0 we selected the CC1101 device and the CC1190 range extender and obtained the register values used in our tests.

Below it is possible to see the default RF parameters used:

Base Frequency: 869.524750 MHz,

Channel number: 0,

Channel spacing: 207.641602 kHz,

Carrier frequency: 869.524750 MHz,

Xtal frequency: 27 MHz,

Data rate: 38.4178 kBaud,

RX filter BW: 105.468750 kHz,

Modulation format: GFSK,

Deviation: 19.775391 kHz,

TX power 20dBm.

Communication wise the results obtained during the first tests showed that the communication range between our boards was far from desirable and when we compared the results obtained with our boards with the ones obtained with two smartrf05 + CC1110EM, it was possible to see that we had some serious problems in our setup.

Since we don’t have the means to test the hardware (spectral analyser) we performed various tests with different firmware configurations (changed the baud rate, changed the PATABLE_VAL, etc.). The tests showed that for smaller baud rates the messages were corrupted 100% of the time, and the only baud that seemed to be working fine, i.e., with no corrupted messages, was the 38.4kBaud.

Since we’re running out of ideas, any tip would be appreciated.

If necessary, we can provide the layout and the schematics via private message.

Thanks,

Carlos Silva

  • 0
    TI__Guru**** 317180 points
    Using 0804 is not a good idea since the parasitics will be completely different and the layout will be spaced out a fair bit compared with the reference designs.

    Also try to find out if you have a frequency offset on your boards. Send CW from one board and find the frequency where the RSSI on the receiver is at max.
  • 0 in reply to TER
    Prodigy 60 points

    Thanks TER for your quick reply.

    Today we manage to test our boards like you asked to, and the maximum RSSI value was achieved close to the base frequency. We tested all the frequencies between 869.4MHz and 869.65MHz with a step frequency of 1kHz.

    We would like to know, if you have any more suggestions or tests, that we could perform in order to identify the problem cause.

    Thanks,

    Carlos Silva

     

  • 0 in reply to Carlos Silva46
    TI__Guru**** 317180 points
    Do you have a signal generator enabling you to measure sensitivity?
  • 0 in reply to TER
    Prodigy 60 points

    Hello TER,

    Unfortunately no.

    However we did manage to measure our boards sensitivity using the SmartRF Studio and a CC1110 module with a low TX power selected and changing the distance between boards. Our results showed that the radio behaves as expected and we have a link budget of -109dBm.

    If we perform the same test with two of our boards, this value is achieved when the boards are 2 meters spaced apart.

    In your opinion, could this be a Balun problem, or this type of behavior can only be explained with other type of mistakes (configurations, etc.)?

    Thanks,

    Carlos Silva

  • 0 in reply to TER
    Prodigy 60 points
    Hello TER,
    Since our project is reaching a critical point and since we're running out of options, I would like to know if you have any more suggestions or ideas regarding this issue.

    Thanks,

    Carlos Silva
  • 0 in reply to Carlos Silva46
    TI__Guru**** 317180 points

    - When you use 0805 components I expect poor performance

    - If you don't have RF instruments it's difficult to debug issues like this. You need a signal generator to measure sensitivity and more importantly a spectrum analyzer to look at output power and harmonics. If the reference design has been followed in detail this is not that important but when you use CC1190 and your layout uses 0805 it's required.

    - Link budget equals -109 dB does not give any meaning. The link budget is defined as: Link budget = output power - sensitivity + antenna gain RX + antenna gain TX  eq a positive number.

  • 0 in reply to TER
    Prodigy 60 points
    Hello TER,
    I'm sorry for the imprecise information.
    On a final note, could you provide your insight on this two components we're using, and if our poor performance could be caused by them:
    In our design we're using this PCB antenna : www.mouser.com/.../ENG_DS_1513273_A-224581.pdf from TE Connectivity, and this RF switch www.skyworksinc.com/.../200186C.pdf from Skyworks.

    Thank you for your patience.

    Carlos Silva