• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 11 subscribers
  • Views 2215 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.

TRF370417EVM: TRF370417EVM - low LO and image rejection at high frequencies

Nir Flysher
Expert 3287 points
Part Number: TRF370417EVM
Other Parts Discussed in Thread: DAC38J84, TRF370417

Dear Team,

My customer is using the TRF370417EVM  with the following modifications (see attached)

LO : 2 to 6 GHz

I/Q:  50 to 500 MHz

While the conversion loss seems right and matches the datasheet, the LO and image rejection they receive in the high frequencies is only 20-25dB.
As these numbers are getting worse as the frequency increase and as they're far from what is described on the datasheet I wanted to have your advise on what might be wrong in the setup.

Best Regards,

Nir

TRF370417EVM.pdf

  • 0
    TI__Intellectual 2245 points

    Hi Nir,

    I will look into this.

    Regards,
    Oscar

  • 0
    TI__Intellectual 2245 points

    Hello Nir,

    After looking at the schematics, I believe that the decrease in the LO and image rejection may originate from the I/Q inputs which are given.  Sideband supression depends on how well the phase and amplitude of the I/Q signals are in match with one another. Tiny mismatchs  can definitely impact performace. From what I believe to be a hybrid coupler may be introducing some phase and amplitude errors with the I/Q signals. Even the system set-up; cables, sources, PADs, ect. can also contribute some additional mismatches into the modulator as well. Also by having the I/Q inputs being driven singled-ended, internally the device may experience some imbalances by seeing a 50 ohm termination on one input and the I/Q souce on the other. It is recommended that the signals be driven differentially to check if performance improves.

    As for the LO feedthrough, this is usually caused by the imbalance of the DC offset at the I and/or Q input. Which results in LO leakage. One thing to check for this is to measure the voltages at the I/Q inputs to ensure that they are equal ot each.

    By using a DAC, such as the DAC38J84, one can mitigate the phase and amplitude imbalances by having the DAC digitally adjust the offset and phase of the I/Q signals.

    Please consider the TI design

    This solutions provides the DAC38J84 interfacing with the TRF3704.

    Regards,

    -Oscar

  • 0 in reply to Oscar Guevara
    TI__Expert 3287 points

    Dear Oscar,
    Customer is interested in working with 50ohm on the I/Q Inputs from 5-400MHz.
    LO is single ended 50ohm.

    Can you suggest an example of how to work with the TRF370417 properly ?
    How to match the inputs correctly ?
    How to add the bias voltage suggested on the datasheet ?

    What is the conversion gain we should expect ?


    Best regards,
    Nir.

  • 0 in reply to Nir Flysher
    TI__Intellectual 2245 points

    Hello Nir,

    The image shown above is an example of an interface network between the DAC and the TRF3704 modulator, where the DAC would need a 0.25V common mode voltage and the modulator needs 1.7V. The TI Design 00335 goes more into depth on how to properly set up the interface network, along with the calculations that are done. By placing in whatever common mode voltage required of the DAC that the customer is using, you can calculate the values of the components. 

    Looking at the datasheet of TRF370417, the conversion gain can vary from -8dB at the lower frequencies to -1.9dB at 400 MHz. 

    Regards,

    Oscar