Post Go back to editing


I have an evaluation board EV1HMC252AQS24 on hand. It is a new board. I tested it by connecting the RFC port to a waveform generator and RF1-6 port to an oscilloscope. The input signal was a square wave with a frequency of 1MHz and a Vpp of 4V.

  1. At the beginning, there is a serial capacitor at the input and output port. The output wave has many peaks (see Image_1). Obviously, it is the output of a high-pass filter.
  2. Then I removed these capacitors. The output wave is a square wave (see Image_2). It is correct except for a little insertion loss.
  3. But when I measured the output signal from those ports that was not selected, I also got some square wave (see Image_3). It is not correct. There should be no waves on those ports.

Is the chip broken? Please help me to solve this problem. Thank you.

  • 1. Can you measure the current of VDD, A, B and C? There is typical value for the current in the datasheet. Does the current look OK?

    2. The capacitors you removed are the DC blocking caps (C1-C7 in the datasheet). Since HMC252A is positively biased, the RF channel voltage is not 0V, therefore, there should always be the DC blocking caps at the RF ports, otherwise the part might be damaged;

    3. The original C1-C7 is 100pF and you tried to apply 1MHz signal. It is true that 1MHz signal will be blocked by 100pF cap. You can replace with a much larger cap on the evaluation board, or replace with a short wire and use external DC blocking caps;

    I would like to make sure the part is still alive and will not be further damaged.

  • Thank you for your reply.

    1. The current of VDD, A, B and C is 6.43mA, 55uA, 55uA and 1.6-1.8mA, respectively. Based on the datasheet, the current of C is not correct. It's too large and always changing.

    2. When I removed the DC blocking capacitors, the output waveform has DC voltage, as below. Now, the output waveforms of every RF channel are the same. Using a multimeter to check the connectivity of these channels, they are connected to each other.

    3. In fact, my input signal has a large frequency range, e.g. 1Hz - 100 MHz. The input signal can be a fully AC wave without any DC. Which kind of DC blocking caps can I use for C1-C7? Do you have any recommendations?

    In addition, if the HMC252A is damaged, how to replace it? Are there any cautions?

  • The original DC blocking caps on the EVB is 100pF and it should be good down to 100MHz. If you are going to reduce the frequency further, larger DC caps are recommended. You should look for 0402 package in order to replace 100pF caps on the EVB. Or like I mentioned previously, you can use wires to connect the gaps for the caps and use external SMA DC blocking caps.

    In terms of replacing damaged IC, you only need to reflow the solder and remove the IC. Then you should remove all the solder residue and apply new solder/flux before replacing a new IC. It is a pretty typical soldering process.