Hi everybody. I designed a custom down-conversion board comprising an HMC8411 for the amp stage and an ADL5830 for the I/Q demodulation. The board is realized on RO4350B substrate and correctly dimensioned to be matched at 5.8GHz. I tested the board without the HMC8411 and the demodulator is working just fine.
The problem comes when mounting the HMC8411. The interface circuit to the DC bias and the RF was designed following what is prescribed by the datasheet, also inspired by the schematic of the EVB. However, many spurious are generated, that seems to be without meaning.
When sectioning the circuit by "breaking" the RF branch before the ADL5380 (so having just the HMC on the RF branch), when I connect the INPUT to the Spectrum Analyzer I can see a pure clean tone at ~5.3GHz, with 10dBm power. It seems like a problem on the feedback loop, but I cannot figure out how to suppress this harmonic that is feedthrough towards the input.
How can I arrange some tests or a solution to suppress this spurious that I think is generating some issues on the demodulator, and so, to the whole circuit?