Announcement

Collapse
No announcement yet.

Circuit Diagram SPI-MX 3b

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

  • Circuit Diagram SPI-MX 3b

    Here is the complete circuit diagram of the SPI-MX 3b

    I will be posting the source code next week.

    Dual phase pulses:
    All the mosfets are used to switch the coil in both directions. If there is a better way let everyone know !

    This helps to cancel noise;
    External signals will shorten (or lower) half the target pulses, and lengthen (or increase) the other half.. thus the average will not be affected.
    i.e Noise or emf will induce a voltage on the coil, the affect on pulses in one phase is the opposite on pulses of the other phase.


    Conductivity.
    The conductivity is calculated by comparing the long response with the short response.
    i.e Working out the shape of the discharge curve.
    This is accurate enough to tell the difference between a bottle cap and a $1 aus coin at any detectable depth %100 of the time
    It can't however tell the difference between a vertical tent peg and a $2 aus coin for example.

    The sound-scape or audio can be explored refined or changed in code.
    Currently I use two tones.. one constant, and one according to conductivity.
    This is because it is easier for the ear to compare two tones than to recognize a single tone with nothing to compare it with.
    Having two tones enables fairly accurate target discrimination,..but I am sure there are other possibilities.

    The two guiding rules of the SPIMX design are:

    PERFORMANCE
    SIMPLICITY

    These are different requirements that sometimes conflict and sometimes compliment
    Currently
    There are no knobs to adjust, other than volume.
    There is no display to look at, all information is available in the sound..through the ears, which are quite sophisticated and under-utilized instruments.

    Tec
    Attached Files

  • #2
    I finally got a chance to look this over. I'm curious about your TX driver, appears you are using cascoded IRF740's for the P-side switch. Why not just use a PMOS device?

    Comment


    • #3
      P-mos, maybe even better ...

      Originally posted by Carl-NC View Post
      I finally got a chance to look this over. I'm curious about your TX driver, appears you are using cascoded IRF740's for the P-side switch. Why not just use a PMOS device?
      Hi Carl,

      yes good point.. it would be even simpler to have P type mosfets on the high side.
      As you know, a pmos of a given voltage max (400V) usually have much higher on resistance (7 ohms instead of 0.550 Ohms for example).

      I used the N channel mosfets because of that.. those two in series should have a combined resistance of 1 ohms.

      But you make me think

      The whole purpose of having a clockwise (cw) pulse and counter clockwise (ccw) pulse is to cancel out noise.
      The pulses have to be identical, and the topside resistance has to be the same (left resistance should be equal to right resistance).
      The lower the better.
      (Compared to 3 ohm coil resistance, 1 ohm mosfet is large)
      Because no two mosfets are the same, there is a variation between the resistance of the left side and the resistance of the right side, which will result in slightly unbalanced pulses meaning less noise cancellation.

      Perhaps an ultra low ON resistance P-mosfet (with 30V flyback voltage), in series with a DIODE,.. will not only make a simpler circuit, but result in much more identical ccw and cw pulses.. resulting in closer to perfect noise cancellation.

      I will try this and see if it works, and perhaps improves (the already good) performance (a process of a few weeks)
      Then I will update the circuit.
      Please let me know if you have a better suggestion..


      Tec

      Comment


      • #4
        Originally posted by Tec View Post
        Perhaps an ultra low ON resistance P-mosfet (with 30V flyback voltage), in series with a DIODE,..
        That's the approach I prefer.

        Comment


        • #5
          Hi Tec

          Congratulations on the project!
          I look little on schematic and i think you have one error.
          As you can see on picture on connector 5.1 leftphase don't exist, but it is used in next part of schematic.
          Cheers
          Detecto
          Attached Files

          Comment


          • #6
            Originally posted by detecto View Post
            Hi Tec

            Congratulations on the project!
            I look little on schematic and i think you have one error.
            As you can see on picture on connector 5.1 leftphase don't exist, but it is used in next part of schematic.
            Cheers
            Detecto
            Hi,

            Thanks Yes, the connector is not needed it is just used as test points on the PCB.
            The left and right signal do go to the correct place on the the circuit.
            The next revision I will remove the connector to simplify the diagram.

            Tec

            Comment


            • #7
              My next question will be, why are you using connector? But you already give answer.
              Cheers
              Detecto

              Comment


              • #8
                Originally posted by Carl-NC View Post
                That's the approach I prefer.
                A diode adds a voltage drop, but the coil need to be tied / referenced to ground as it discharges. Because of the way the signal is measured on the comparators. (referenced to ground), can't have a voltage drop caused by the diode. I just remembered why can't use diodes.
                no voltage drop allowed. hence the convoluted mosfet configuration.

                lower on-resistance mosfets would be an improvement.

                Comment

                Working...
                X