Announcement

Collapse
No announcement yet.

Let's talk about avalanche

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

  • Let's talk about avalanche

    Let's talk about avalanche
    This is my opinion, the way I see it, I would like to hear other opinions.
    When we surpass the current limit of the cicuit, the Mosfet avalanches.
    Mosfets have this inbuilt safety feature. The voltage limit, for example 1000V, is the working voltage of the Mosfet. A bit higher, somewhere between 10 to 25% the avalanche diode, like a Zener diode, shorts the mosfet to ground.
    This protects the circuit and the Mosfet from overvoltage damage.
    So, what does this mean for a PI TX circuit?
    Insert picture of TX circuit:
    The TX "charge" current comes from the battery, through the coil, to the ground. This current builds a magnetic field around the coil. This changing magnetic field induces eddy currents in the target. For better visualization, we consider these eddy currents to spin in the clockwise direction. (Eddy=Vortex)
    Now we block the current with the Mosfet.
    When the current stops flowing, the magnetic field collapses.
    The current that is induced in the coil by the fast change of the collapsing magnetic field now has to find a different path, since the path to ground is blocked.
    The path back through the coil has high impedance. The resistance of the path through the damping resisistor is relatively high. Therefore the Flyback voltage builds up.
    When the Flyback voltage reaches about 1000V plus whatever the safety margin ( 10 to 25%) is, the avalanche diode discharges the surplus.
    So now the Flyback has 3 different paths to discharge.
    a) the coil with it's impedance
    b) the damping resistor (including the RX path)
    c) the avalanche diode
    What exactly is the Flyback? I will leave my understanding of that for another day.
    The only path that is useful for the PI TX, is through the coil. The Flyback current builds a new, fast changing magnetic field. The fast changing field induces new eddy currents in the target. These eddy currents now spin in the counter-clockweise direction. Initially these eddy currents have to cancel the eddy currents that are still running in the clockwise direction.
    The decay of these eddy currrents is what gives us the target response.
    The current going through the damping network bypasses the coil. It does not add to the magnetic field of the coil, but it helps making a higher rate of change of the magnetic field.
    The current through the avalanche diode is a total waste, but it protects the circuit like a fuse.
    To make it clearer, let's put some numbers (not calculated or meant to be exactly correct) to it:
    Coil charge current: 3A
    Avalanche current: 1A
    Damping current: 1A
    Useful coil discharge current: 1A This is the only useful current that induces a target response
    Please correct me where wrong.
    Let's talk about avalanche.pdf

  • #2
    Coil current at turn off is what matters. Doesn't matter what it discharges thru, Rd, Rin or the Mosfet.

    Comment


    • #3
      https://www.geotech1.com/forums/show...rget-Responses

      The above link to a thread in this forum is a good read,see what you think. Some P.I. designs even recycle the current from the TX period for the following TX period.

      Comment


      • #4
        There was more I was going to add to that thread but now can't remember what it was.

        Comment


        • #5
          Originally posted by Carl-NC View Post
          There was more I was going to add to that thread but now can't remember what it was.
          I don't blame you Carl, but at least you left a mark on me.

          As Carl had clearly stated, the best scenario is one which allows the initial Eddy current to degenerate to nill before the collapsing field initiates a secondary, detectable Eddy current response. Otherwise they will subtract.

          Target saturation is achieved by means of the peak current, you'd like the current to " hang" for a brief moment to allow the initial Eddy current to subside. The coil itself now becomes reflective in nature, it's inherent Z acts like a buffer to absorb the energy.

          The mosfet is relegated to merely being a switch. Avalanching being a mere consequence. Now the opamp comes into the picture, to measure the reflective energy by way of decay time.

          What saves everything from going into meltdown is the PRF and the pulse width.

          Now ramping becomes obvious and intuitive. This is where Carl had left it.

          There are several patents which address this issue, but the language is muddled in technical jargon.

          What I like about Carl is that he can address these issues in the simplistic language possible. So that one may grasp the concept, in other words, have a conceptual understanding of the fundamentals. The math can come later on.

          Comment


          • #6
            AVALANCHE.zip
            Same calculation about avalanche.

            Comment


            • #7
              Agreed ! Carl's capacity to articulate that which he is describing brings clarity to what can be a confusing subject.

              Comment

              Working...
              X