HMC525 IQ Mixer SSB Malfunction

Hello, I prepared a eval board of my own with an external 3db-90* Coupler + HMC525 IQ Mixer.

Amplitude imbalance between arms of the coupler is 0.4db while phase imbalance is +- 4 degrees.

I gave LO with  +15dbm power @4.3Ghz, and RF with -10dbm @2Mhz, same with values in datasheet except for IF=100 Mhz to IF = 2 MHz. I expected to see a single-sideband upconversion

with power difference of about 30db between  4.3Ghz + 2 Mhz and 4.3Ghz - 2 Mhz; however, they were about even power.

What may be the source of this error?

    •  Analog Employees 
    on Apr 18, 2019 9:08 PM


    A few more questions on this.

    1) What is the input power that is going into the 90 degree coupler?

    2) What is the insertion loss of the 90 degree coupler?

    3) What is the measured output power of the LO tone at the RF port?

    4) What is the measured output power 4.3GHz + 2MHz tone  at the RF port?

    5) What is the measured output power 4.3GHz - 2MHz tone at the RF port?


  • 1) My LO is +15 dbm. I expect it to be 15- (LO-IF Isolation). I dont have a test port between Mixer and Coupler.

    2) IL of hybrid coupler=0.6db

    3)-47 dbm @ RF port LO

    4)-50 dbm @ RF port 4.3GHz -2 Mhz

    5)-50.4 dbm @RF port 4.3GHz +2 Mhz

    I put three markers on those signals; Delta (LO-Upconverted signal) ~3db 

    Delta(UpperSideBand - LowerSideBand) = 0.2~0.4db 


    •  Analog Employees 
    on Apr 19, 2019 2:57 PM in reply to tersoz


    With respect to question 1, I'd like to know the IF input power.

    Just to double check your setup. Please let me know if this is correct.

    IF input (@ xx dBm , 2MHz) ->  3db-90* Coupler -> Quadrature inputs into IF1 and IF2 ports of HMC525A @ 2MHz and -3.6dB from the IF input power.

    LO Input (15dBm, 4.3GHz) -> LO port of HMC525A

    RF output (4.3GHz +/- 2MHz) from the RF port of the mixer.


  • -20 dbm @2 Mhz (Main tone) -70dbm @4MHz  2nd harmonic power due to signal generator imperfection.

    •  Analog Employees 
    on Apr 23, 2019 6:51 PM in reply to tersoz

    Based on the LO leakage at the RF port, I don't think the LO signal is getting through to the mixer- so no 'mixing' is happening.

    On ADI eval board even without any IF input when I provide 15dBm LO signal at 4.35 GHz the LO leakage at the RF output is -31dBm.