Announcement

Collapse
No announcement yet.

Quad Bilateral Switch

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Quad Bilateral Switch

    Always debating about some projects in general, how about we to analyze some details?
    I made dozen various md projects so far, pretty often there is some bilateral switch included (Classic-4016, Delta Pulse-2x4066 ...etc..).
    So..would be good to know details and differences between various ic's.
    Looking at datasheet man can see detailed technical explanations. But personal live experience is something much more valuable and important.
    So... for a start i would put only 4016 and 4066 on debate here.
    Few times i used one instead other...and it works. I haven't noticed some significant difference.
    So i wander if i can generally always use one instead other?
    Difference is that 4066 has lower "ON" resistance, 80 Ohm at 15V. When 4016 has 280 Ohm at 15V. Meaning what? Meaning what in some particular design like one of our projects here?
    I would like somebody here to explain me what this difference exactly means in ...let's say particular design as Delta Pulse? If i put CD4016 instead noted CD4066 , what differences i should expect than, for example in DP behavior?


    P.S.
    Questions like this one are much more beneficial and educational, than just some posted project without any explanations.... Personally i would like to attract attention here with this one and start series of similar small questions and answers, and later to form a sort of small "cookbook,hints and tips" for everyday use and especially good start for beginners...



  • #2
    I think some of the important parameters of switches are switching speed, the on resistance, input capacitance.
    The higher these values the worse, assuming the ideal switch has zero on resistance, capacitance and ideal switching times.
    Of course the rate of input signal has to be taken into account.

    Comment


    • #3
      Very good!
      Next question; is the lowest "ON" resistance always desirable?

      Comment


      • #4
        Well, maybe we have think of the equivalent resistance circuit. If R1 is the on-resistance and R2 is the load, when the switch is ON a resistive divider is formed and thus the output voltage would depend on the R1 and R2 values. If R2 is large enough and R1 is low the input signal will not be altered much. So if R2 is much bigger than R1, the on-resistance may be neglected.
        Attached Files

        Comment


        • #5
          I may be jumping ahead of where you're going with this thread, but there are many analog switches available now from various manufacturers that have much better specifications (wider bandwidth, faster switching speed, lower on resistance) than the aforementioned.

          As one example, take a look at the specs on the Pericom PI5A123:

          http://www.pericom.com/pdf/datasheets/PI5A123.pdf

          Comment


          • #6
            Originally posted by lucifer View Post
            Well, maybe we have think of the equivalent resistance circuit. If R1 is the on-resistance and R2 is the load, when the switch is ON a resistive divider is formed and thus the output voltage would depend on the R1 and R2 values. If R2 is large enough and R1 is low the input signal will not be altered much. So if R2 is much bigger than R1, the on-resistance may be neglected.
            Maybe adding extra biasing can solve higher On resistance issue?
            And how should look comparision between use of bilateral switch and simple fet as switch? (i am referring this to exact case we have at older Tesoro detectors where fet is used as simple switch).
            Further; i noticed certain "delay" (signal alteration) at some detectors where some (maybe improperly chosen) bilateral switches are used. This further leads to more complicated issues in signal processing in late stages of detector (now you see why i started this thread! )

            Comment


            • #7
              Originally posted by hobbes_lives View Post
              I may be jumping ahead of where you're going with this thread, but there are many analog switches available now from various manufacturers that have much better specifications (wider bandwidth, faster switching speed, lower on resistance) than the aforementioned.

              As one example, take a look at the specs on the Pericom PI5A123:

              http://www.pericom.com/pdf/datasheets/PI5A123.pdf
              Excellent!
              Actually this thread is just about that - to mark differences, benefits and drawbacks between some choices in process of designing new design.
              Me personally...i would like to reconsider some better options and apply possible changes on some designs i already made.
              For example; at TGSL we have simple fet used as switch, is it best possible solution? Could it be better?
              What if we replace fet with some "fast" bilateral switch?
              etc...etc...

              Comment


              • #8
                I remember another difference between the 4016 and 4066. The leakage current is lower on the 4016. So on a sample and hold circuit with a long time period you would select a 4016 over a 4066. An example would be on the old no motion detectors. Where you would push the button and tune the threshold. The lower the leakage current, the less drift you will have in your threshold over time. I doubt it makes much difference in a demodulator, switching in the khz range. No one ever mentions the 74VHC4316, it has logic level switching, good for microprocessor circuits.

                Mark

                Comment


                • #9
                  Good info!
                  Thanks!

                  Comment


                  • #10
                    bilateral?

                    I ask a question: looking at various schematics of MD I wonder whether it's really necessary that the switch is "bilateral"?

                    Comment

                    Working...
                    X