AD633 Control voltage polarity

Hi Hopefully this is an appropriate place for the following question. Apologies if not.

     I hope someone can help me find a smarter way to resolve a small problem. I've been using the AD633 as a voltage controlled low pass filter. The problem is that the control voltage works the wrong way round for my purpose, the crossover frequency goes up as the voltage increases. I want the frequency to go down as the voltage increases. There is a way round by inverting and offsetting the control voltage but there's the potential for the control voltage to go below zero. Again there's a way round this by using a single supply rail to rail op amp with a false ground but it's all getting rather complicated.

Any clever way to make the AD633 work the other way round, or use something else?

Any suggestions much appreciated.


  • Not a circuit as such and can't manage a pdf but I've attached a png of the LTspice simulation. It's essentially a simple automatic loudness  control. Reduces the top end as the volume gets louder.

    Thanks for taking an interest


    P.S. Just realized the circuit is slightly confusing as there's a voltage source for the control voltage. It would normally be connected to vCtl (U3 output). The voltage source is there to run an AC analysis. You can't do it using the op amp output.

  • OK, thanks Harry. So long as I know I haven't missed something obvious I'll just have to stick to the hard way. BTW seems the LM13700 is still available.

    Thanks for the PDF tip.



  • 0
    •  Analog Employees 
    on Sep 5, 2012 12:29 AM

    One idea: try using a current mirror to pull current through a resistor string that biases the AD633's control input and use your control voltage to drive the current mirror. Depending on the accuracy you want, you could use a generic pair of 2N3904s or an SSM matched pair from ADI.

    As the voltage input to the current mirror increases, it will pull down on the control voltage to the AD633, which is what I think you want.

    Just a thought.



  • Thanks Bob. That sounds like a plan. Accuracy isn't important as the gain can be trimmed easily enough.



  • 0
    •  Analog Employees 
    on Sep 5, 2012 3:25 AM

    Hi Gordon,

    Sorry for the late response, I was on vacation when your EZ question arrived.

    I'm a bit confused, in your original email you said you;ve been using the AD633 for a voltage sontrrolled LPF. Thus if you've been using it, where lies the problem? Did you mean you have been evaluating this particular applicaiton?

    You didn't mention if you tried reversing the input polarity of the X or Y inputs, note these inputs are fully differential so if you just need to reverse the polaity of one or the other it's pretty simple. You did mention using a single supply. This isn't really an option with the AD633 or any other multiplier due to symmetrical supply needs. An exception would be if you created a common supply at half the value of the single supply but I've not tried this. It would have to be pretty elaborate, when people bring this question up I usually say it would be easieer to create a negative supply using an inverter chip.

    If you send a copy of the circuit you're using,along with requirements I could check the basic requirements.