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.

AFE7222: Swap P/N input/output of baseband signal

Part Number: AFE7222
Other Parts Discussed in Thread: TRF370333,

Hello!

I have question regarding connection of baseband signal to quadrature modulator/demodulator. In PCB mounting the IC have several layout, and I can't connect output of demodulator symmetrical to AFE, P to P and N to N. If i would cross lines in pcb I get some vias and not desirable turns on routing differencial quadrature pair.

In the AFE7222xEVM documentation I saw, that I P and N signal are swaping, but Q P and N signal have right connect for modulator TRF370333. In the layout respectively lines don't crossing and have direct connet to modulator.

Is this right scheme, and this swap don't have bad performance for quadrature signal?

I'm more interested in the case with the demodulator, but I've never found such connection examples.

  • Hello
    Someone will respond to your question on Monday.
    Best regards,
    Jim B
  • Yes, you can swap the P/N connections if needed to ease the layout connections. The swapping will cause the spectrum to be "flipped" compared to typical expectations. In the modulator, for example, the swapping yields the desired signal to be on the low side of the LO and the suppressed image on the high side. This is comparable for the demodulator. You can easily flip your incoming pattern to get to the high side if needed and similarly flip it the received signal within the FPGA or processor.

    --RJH
  • Hello!

    I think you don't understand my question. You say about swaping Hi and Lo input to modulator. In this case I know, that spectrum are flipped, and for this case i need swap digital signal at processor for get right spectrum. But my question about swapping P/N pins into differencial pair, that connect to the demodulator.

  • For my problem I find partically decision. If I swap lines for into differential pair (swap P/N pins), I get on input signal with 180 deg shift, and I need invert output digital code for getting source analog signal. But problem, if I get right I-signal, and 180 deg shift Q-signal, then will use process with internal features of AFE7222 (offset correction, balance control), I already can't simply invert digital code for Q signals. Or I don't right?
  • Swapping P/N on one of the input lines has the same effect as swapping the I/Q lines. I did not follow your last question. You can still use the digital QMC correction features related to DC offset control and I/Q amplitude and phase balance as normal with the input signals inverted appropriately.
  • Thanks for clarifying! Will use swapping for P/N pins.