ADA4522-4 seems not work well

Hi, I designed a 8th pole sallen key filter with the filter wizard.

here is what I get from filter wizard.

I realize this schematic in OrCad

and finally I designed the layout

But, if I feed a sinusoidal signal to the input of this filter I get this situation.

I check all the metal connection on the board. power supply is present and gnd is correctly connected.

If I also try to simulate this circuit in LTSpice

I get this situation

where the green track is the input and the blue one is the output.

In the datasheet I read that this opamp could work at +5V single power supply but it is not working as expected.

Do you have any suggestion?

Thank you for your support

    •  Analog Employees 
    on Dec 2, 2019 5:49 PM over 1 year ago

    Hi the ADA4522-4 has an input range of 0 to 3.5V on a 5 volt supply, which explains what you are seeing for the LTspice simulation.  Even if you keep your input range between 0 and 3.5V, peaking of the filter stages may make the output voltage between stages bigger than this, lowering your total possible range further.  You can use the "voltage range" view of the Filter Wizard to get more insight into what will work.

    I'm not sure why the hardware design is not working.  Could you do the following?:

    1) Save your design from Filter Wizard, and then attach the .JSON file that's generated?

    2) Send in a smaller input signal (say 1Vpp) within the input range of the ADA4522-4.  Take measurements at the output of each filter stage.  It's likely you'll find a stage where the input signal looks reasonable, but the output is misbehaving.  Once you find the stage that seems to not be working, probe around on the nodes of that circuits to see if anything looks odd.

    Matt

  • Hi Matt, thank you for your answer.

    in the meanwhile you answered me, I read the detail of the input range.

    Now I changed the power supply of each opamp to 0-15Vdc to avoid the limitation of the input range.

    Even if my input signal is 0V, when I switch on the circuit, the ouptu of the filter is not predictable: sometimes it remains 0V, sometimes it goes to 6V. If I scope the signal of each filter stage, I can see that each input/output of each filter stage is 0V but the last filter stage goes wrong and provide me a 6V output.

    Keep in mind that this behaviour is random and I can verify it only switching on and off the circuit.