• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 54 subscribers
  • Views 328 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.

CC1190: New design with CC1190 has no gain with PA on and HGM on ?!?

citizen
Expert 2840 points
Part Number: CC1190
Other Parts Discussed in Thread: CC1120, BOOSTXL-CC1120-90

Tool/software:

Hi,

I'm a little bewildered to why I'm not seeing any gain on a new design using the CC1190 PA/LNA following a CC1120 transceiver.

Essentially, testing with a CW carrier at 868MHz, the expected output from the PA is over 26dBm, whereas I'm only measuring just under 0dBm ?!?

Thinking it could something on the CC1120 transceiver side, I decided to remove that stage from the circuit by removing a 10nH inductor from the LC matching on the SAW filter output, going into the CC1190 input port, and added a signal from a tracking generator on the SA at this input just to see where the gain is lacking on my radio.

The result is, with -20dBm TG signal input all I'm seeing on the CC1190 output is a signal slightly below -20dBm ?!? So essentially getting the TG signal going straight through the CC1190, no gain, zilch... if anything a slight loss of about 1dB from input to the SMA output ?

This is unexpected as the CC1190 design is based on a TI reference design schematic with identical components (albeit a variation on the routing and stack-up) that we have used before in a 99% same configuration.

I've checked the HGM pin and it's at +3.3V, the PA port is also high, and the PA power rails all check out fine at +3.6V. So, even if the assembly house botched up part (or two) of the output impedance match slightly (for whatever reason, but I doubt it as they are a reputable PCBA), I would expect to see some gain, say with at most 3 - 4 dB penalty on output power at most, but still having an output over +20dBm with +7dBm input power.

Just wondering what might have caused this (or could cause it) and should I be worried the CC1190 was deep-fried during assembly or is this unlikely ? no other issues on the PCB otherwise, all good and other PCB digital and analogue tests pass with flying colours, just the CC1190 gain being 0dB is a real bummer ! Tests so far on 3 boards, identical issue, no gain !

With the PA on, the current drawn is about 100mA at +4,2V supply input. I was expecting more current drawn also, if the output were closer to +27dBm maximum.

Any other possible checks I could do on the CC1190, other than replacing the device with a new one, as a last resort ?

Any thoughts, advice, suggestions would be appreciated...

Regards, MM

  • 0
    TI__Mastermind 18485 points

    Hi MM,

    Please can you perform the following initial tests (or confirm that you have already performed them):

    • Conducted testing using SmartRF Studio 7 to configure the device(s).
    • Conducted output power directly from the CC1120 using SmartRF Studio. 
    • If you have a TI CC112x-CC1190 LaunchPad or another design that uses a CC1190 that is confirmed to work as expected, swap one of the "bad" CC1190 devices and test on a working board (again using SmartRF Studio).

    Regards,

    Zack

  • 0 in reply to ZC
    Expert 2840 points

    Hi Zack,

    Thanks for the initial suggestions. Unfortunately I cannot do those tests as the CC1190 is on a custom PCB basd on the reference design as I mentioned.

    Also with regards to the configuration of the devices, the CC1190 is configured by a TMS320 device (not a Launch-Pad or other), so essentially what I can confirm are the following :

    1. The PA_En input pin on the CC1190 is logic high as required for the PA device to be ON. LNA_En is low to have the LNA block OFF.
    2. The CC1120 is out of the circuit for the current testing. The CC1190 is actually driven by an RF signal generator directly with signal levels -20dBm, -10dBm and 0dBm so as to observe the CC1190 output directly.

    At the moment I do not know if there is a bad device as I am seeing a carrier on the CC1190 output port at the expected frequency - just that there is no gain on the signal, but a slight loss of about 1dB or so. with the input level going to 0dBm, I am seeing the start of saturation as the output is now about 3dB lower than input.

    As I said I am not sure if there is a bad device, but would be surprised if there were 3 bad devices ?!? that would be unexpected but am willing to rule that out if needed.

  • 0 in reply to citizen
    TI__Mastermind 18485 points

    Hi,

    Please can you submit your design your design for a review using: Sub-1 GHz Design Review Submission: https://www.ti.com/tool/SIMPLELINK-SUB1GHZ-DESIGN-REVIEWS

    Regards,

    Zack

  • +1 in reply to ZC
    Expert 2840 points

    Hi there Zack,

    Thanks for getting back to me regarding this issue.

    The good news is that last week this issue has been resolved.

    Unfortunately my only last resort was to identify, by trial and error, possible erroneous placement of components on the CC1190 output.

    Fortunately, there were not that many components to look at, and finally it was established that one 7.5pF capacitor was obviously not a 7.5pF capacitor after all. This component was subsequently correctly replaced with a 7.5pF Murrata NPO capacitor and the gain has now been measured to be over 26dB on all the boards with measured output powers around 26dBm.

    The assembly house's quality department has been contacted to be alerted of this systematic problem with the 7.5pF capacitor. It would appear the value was much higher as the VNA measurements were showing S11 to be way out of the acceptable range of all previous manufactured boards, including the BOOSTXL-CC1120-90 eval kit we had in the office.

    So everything's cool for the moment.

    Regards, MM

  • 0 in reply to citizen
    TI__Mastermind 18485 points

    Hi MM,

    That's good to hear (that you've found the root cause) and thanks for sharing your results on the forum - it helps other customers with similar issues when looking for solutions.

    Regards,

    Zack