question regarding operation of HMC1020LP4E RMS detector

hello,

i have an evaluation board of your HMC1020LP4E RMS detector product line. When all bias, GND and control inputs are properly connected using a 5V DC supply and a variable DC supply for the Vset input, i can only obtain a constant +4VDC output for either CW or pulsed RF inputs (no matter what RF input level i've input to the detector).

The +4VDC is present for both CW and pulsed RF inputs and remains at +4V for any  RF input drive level between -60 and 0 dBm. My understanding of this product according to the datasheet is that it should produce an output voltage (RMS) proportional to the RF input level.

My first attempt was to simply replicate the RFin versus RMSout curve in the datasheet (using the same SCI1 through SCI4 settings in the foot notes of the datasheet).

Can you please explain what i'm doing wrong? i've double/triple checked my connections and bias voltages. I've tried various max/min RF pulse widths and SCI1-4 settings with no luck. I always just get the +4VDC at the RMSout connector into a high-Z scope input.

The Vset input seems to have an effect as well. As the Vset input voltage is varied upward (from 0.25V to 1.5V) the output trip threshold seems to also increase as the RF input drive level is increased. But, again, this only produces a constant +4VDC output (either pulsed or constant) at the RMSout connector for pulsed/CW RF input signals respectively.

Thank you in advance.

My contact info is:

Dana Spinuzzi

MIT Lincoln Laboratory

781-981-0558

  • 0
    •  Analog Employees 
    on Dec 6, 2019 5:24 PM

    It sounds like you are operating the detector in an open loop mode.  Driving the Vset voltage with a DC voltage supply will make the output op-amp be a comparator.  Hence the RMSout railing to +4V and the trip-point changing as you change Vset and RF input power.

    If you want an RMSout vs. RFin power curve, you need to feedback the RMSOUT signal to the Vset pin.  The application circuit on page 15 does this via a resistor divider network of two 4.7k resistors.  The evaluation board should come setup this way, so all you should really have to do it disconnect the DC source driving the Vset pin to fix your problem.