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.

TRF370417: Problem on reducing the carrier feedthrough and sideband suppression

Part Number: TRF370417

Tool/software:

Hi,

My customers are working on a RF signal generator design and plan to use TRF370417 and are facing some problems about the carrier feedthrough and sideband suppression, details are as below:

As in the schematic, they bypass/cancel the U44 and U47 and connect the BBIN/P and BBQN/P pin directly to the signal waveform generator to adjust the carrier feedthrough and sideband suppression. In figure 1 showed the best results so far the have under the local oscillator frequency is 2GHz. The right peak is the output and the middle peak is the carrier feedthrough, left is the low sideband signal. The best results is that -60dbm carrier feedthrough and the -50dbc sideband suppression. Customers target is -65dbm for carrier feedthrough and -60dbc sideband suppression. Currently they are adjusting mainly by the adjusting the I/Q inputs from signal generator. Do we have any suggestions on the schematic and any recommend methods to reduce the carrier feedthrough and sideband suppression? 

Besides, customers local oscillator frequency will be continuously change from 250MHz to 6GHz during their real use case. From the datasheet I noticed that the carrier feedthrough and sideband suppression varied a lot with changing Flo, does this mean that they have to make the calibration/adjustment experiment for each local oscillator frequency in order to have the lowest carrier feedthrough and sideband suppression? Any other better methods? 

BR,

Manu

  • Manu

    Do we have any suggestions on the schematic and any recommend methods to reduce the carrier feedthrough and sideband suppression? 

    You may utilize the DAC348x or the DAC38J8x family of devices with QMC and LOL adjustment feature. The features have fine level adjustment and can further reduce the QMC and LOL performance.

    does this mean that they have to make the calibration/adjustment experiment for each local oscillator frequency in order to have the lowest carrier feedthrough and sideband suppression? Any other better methods? 

    Yes, this is correct. Each frequency range will needs its own LOL and QMC settings. 

    The limitation for wideband application for TRF370417 is the LOL and QMC setup. As a way to improve this, TI now has RF sampling converters which natively do not have LOL or QMC limitations:

    https://www.ti.com/rf-microwave/transceivers-transmitters-receivers/rf-sampling-transceivers/overview.html

    -Kang