Post Go back to editing

HMC602LP4E and HMC346AMS8GE for ALC

Hi,

Can you provide some application note regarding the controller mode of the HMC602 (or equivalent log detector/controller) ? We recently acquired evaluation boards for the HMC602LP4E and HMC346AMS8GE in order to implement an analog ALC loop in an experiment (1-8GHz), unfortunately I can't seem to get the controller mode working properly.

The HMC602 works flawlessly in detector mode, so does the HMC346 (wired in single-line control as shown in page 8-5 of its datasheet), however, when decoupling Vset and LOGOUT, the error amplifier output either swings between 0V and 4V (with significant hysteresis), or shows large oscillations, almost regarless of the voltage applied to Vset.

The RF signal exiting the HMC346 is fed to the HMC602 through a 16dB directional coupler. The loop itself is implemented in a similar way to Fig. 43 of the AD8318 datasheet, with the difference that, in our case, LOGOUT is multiplied by 2.5 and -5V is added to it before being fed to V1 of the HMC346: an increase in LOGOUT translates to a V1 closer to 0V and therefore less attenuation (and vice versa).

Since either device work independently, I don't think I have defective parts, so I'm obviously either missing something important or doing something terribly stupid and we'd appreciate some input on how to get it working.

Best regards

Parents
  • Thank you to both of you for your replies and suggestions. I am indeed applying between ~0V and 2V to VSET and the HMC346 is receiving between 0V and -5V. I also doubled check the loop polarity/slope and it is correct: an increase in the LOGOUT output (corresponding to a lowering RF power) is translated into a larger voltage applied on V1 (closer to 0V, so a reduced attenuation).

    However, upon further inspection/debugging of the circuit, I am (heavily) limited by the slew rate of the op amp I use to condition the loop signal, and the oscillations are most likely due to the HMC602 overshooting/undershooting because the HMC346 doesn't react fast enough. I unfortunately don't have a faster op amp on hand, but I'll order a few and hopefully everything should be fine :)

Reply
  • Thank you to both of you for your replies and suggestions. I am indeed applying between ~0V and 2V to VSET and the HMC346 is receiving between 0V and -5V. I also doubled check the loop polarity/slope and it is correct: an increase in the LOGOUT output (corresponding to a lowering RF power) is translated into a larger voltage applied on V1 (closer to 0V, so a reduced attenuation).

    However, upon further inspection/debugging of the circuit, I am (heavily) limited by the slew rate of the op amp I use to condition the loop signal, and the oscillations are most likely due to the HMC602 overshooting/undershooting because the HMC346 doesn't react fast enough. I unfortunately don't have a faster op amp on hand, but I'll order a few and hopefully everything should be fine :)

Children
No Data