Using AD8561 for 100kHz or higher frequency Sine to Square Wave Conversion (zero crossing detector)

Hi, I have a question about using AD8561 for 100kHz or higher frequency Sine to Square Wave Conversion.

The comparator I'm using is AD8561 and the circuit diagram I set up on breadboard is

Inline image 1

And the response I get is as follows, with sine wave as input and blue as pin7 output.

Is there a way to get rid of these ripple/oscillation and get 7ns response bandwidth? 

Inline image 2

Inline image 3

Inline image 4

Is there a way to get rid of these ripple/oscillation and get 7ns response bandwidth? 

  • soljiang,

      The images did not come through.  When I try to look at them, it takes me to gmail.

    Can you attach PDF files or jpeg?  You can generate a PDF file with www.cutepdf.com

    It will look like a printer that you print to, but it will give you a PDF file.

    As a side note, with very fast parts, you need a clean pc board layout, adequate bypass caps, etc.

    and a little bit of positive hysteresis as described in the apps section of the data sheet.

    Harry

  • Thank you Harry for your reply.  I apologize that the pictures didn't show up. I attached the pictures again here.

    I did bypass the supplies. 

    Thank you!

  • soljiang,

    Take a look at figs 5 and 6 in MT-083.  You need hysteresis.  Try 10 ohm in, 1k feedback.  If that

    doesn't work, change to 20/1k

    http://www.analog.com/static/imported-files/tutorials/MT-083.pdf

    Harry

  • Hi, Harry, I got weird looking signals.

    I tried out using 10 or 20ohm resistors. 

    The diagram of the circuit I test is

    The response is:

    One can see that even the input signal is distorted at the zero crossing. 

    Also I don't understand why the logic "0" looks so weird below 0V.

    Thank you.

    Best,

    Charles

  • Charles,

      It's a little hard to read the graphs, but if the purple trace is about 0.7V below the input, I would suspect your -5V is really zero, or not very stiff.  Measure directly at the part for +5, -5, and ground.  (which package?)

    It looks like you are running on +5V and ground.  Also look at pin 8; it should be an inverted pin 7.

    What is your source?  Source impedence??

    Also try a sine wave with +/-0.5V.  Looks like you are running +2.5V to -2.5V which should work, but

    drop the amplitude and see what happens.

    Are you using one of those white breadboarding strips? or a pc board?

    Harry