• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 10 subscribers
  • Views 604 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.

TRF370417: IF Harmonics issue

Daniel Lukach
Prodigy 40 points
Part Number: TRF370417

Hello all,

I'm currently working on a design using the modulator as an SSB upconverter. My input for IF is 16 MHz, my LO is 607 MHz to generate 591 MHz.  By adjusting the IF level, pots R4, & R2 I'm able to calibrate out the LO from the spectrum. I'm running into an issue where I'm getting frequencies generated at values like Flo-2*IF.

After modeling the modulator in simulink, I was able to determine that a phase mismatch between the I & Q IF signals would cause poor upper sideband suppression.  A level mismatch on the other hand would cause poor LO suppression, as proved by the sim & calibration I did. 

What would cause these harmonics? I tried a very High Q 16 MHz filter & still saw these on my spectrum so it has to be something I'm doing wrong with the chip.

EDIT: Also would like to add, I am using the RF out. Did not populate my amplifier. I hand assembled this board and used a stencil & paste to get the chip on the PCB. Would a poor ground cause IF harmonics from the internal amp to come through?

Simulink Project

KiCAD Project

Spectrum Image

Schematic

Layout

  • 0
    TI__Guru* 78794 points

    Hello,

    Please refer to the following app note. You may see if a clean injection of the LO can help with the images

    https://www.ti.com/lit/an/slwa059/slwa059.pdf

    -Kang

  • 0 in reply to Kang Hsia
    Prodigy 40 points

    Hi Kang, I'm using a High End R&S RF Gen for my LO. I decided to add an LPF as suggested in the whitepaper and no difference was observed in the spectrum. Looking for some other ideas.

  • 0
    Prodigy 40 points

    So to give an update to anyone following or can help.

    I added a parallel LC shunt with a tunable Shunt on each output of the hybrid. This will allow me to get roughly +-5 deg of phase shift at 16 MHz.

    After tuning both inductors and the bias to the modulator I was able to get some of the images/unwanted spectrum down. This makes sense from my simulation that a perfect 90 degree IQ will eliminate the upper side band. What I found interesting is that this also decreased the unwanted Flo-2*Fif signal. Additionally, I found that adjusting one of the inductors will decrease the amplitude of one of those 2 signals, while increasing the other. So as Flo-2Fif went up, Flo+Fif went down & vice versa. What is also interesting is increasing LO power from like 6dBm to 8dBm helped decrease these two more.

    Now I have these other unwanted frequencies which I'm still unable to process out. Below -

    Marker 3 = 558 MHz = Flo- 3*Fif

    Marker 4 = 639 Mhz = Flo + 2*If

    It seems there is a relationship where Flo-(n+1)*Fif & Flo + n*Fif, n being the harmonic order are related.

    If any TI experts can help, I'm very interested in solving this problem.