Announcement

Collapse
No announcement yet.

PI metal detector with energy recuperation

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

  • #16
    deemon

    Thats fantastic work, congratulations.

    Comment


    • #17
      exellent job deemon and tidy work too this is just my cup of tea!

      Comment


      • #18
        Wow!
        Looking at photos i am already discouraged!
        Congratulations!
        It was hard labour!
        Any video?

        Comment


        • #19
          Thank you deemon. Great job and nice job. Here, the forum has a lot of experts in the electronics, which will hopefully contribute this circuit to simplify the insertion of integrated circuits, because now there is no any... Best regards.

          Comment


          • #20
            Yep, that's really art work.

            (Deemon must be one of the good old school guys. The newer "facebook generation engineer's" even don't know, how the op-amp is internally working. Good work. Hats off!).


            Déjà vu
            BTW, the sound card detector controller platform is already supporting one of the proposed (transmitter) designs (left transmitter configuration in the above circuit). The TR1 & Rb isn't required however. The bipolar transistor and the reverse biased diode is the low-side n-ch mosfet with inherent body diode. (The high-side p-ch mosfet isn't installed.) The coil need no parallel capacitor installed (but can be) and another capacitor (series C) is connected parallel to the low-side n-ch mosfet.

            Cheers,
            Aziz

            Comment


            • #21
              Very interesting approach. I have no idea if I'll ever build something with this many transistors. There are a few too many gain stages to my taste. If forced to do gain with transistors I'd use mu-amp instead, or something. A 40dB design is in attachment.

              Maybe I'll do it in LTspice in some parts.
              Attached Files

              Comment


              • #22
                this is how things had to be done at one time as someone else said maybe condensing a lot of those trannys into ics will clean up the component count
                even though i like transistor designs i'm not apposed to ics
                today they have their place like any component if we help deemon to get this a proper c-diagram and board layout etc i'll have a go at building this

                Comment


                • #23
                  Originally posted by Davor View Post
                  Very interesting approach. I have no idea if I'll ever build something with this many transistors. There are a few too many gain stages to my taste. If forced to do gain with transistors I'd use mu-amp instead, or something. A 40dB design is in attachment.

                  Maybe I'll do it in LTspice in some parts.
                  the gain stages and such is a feature of deemons regenerative design

                  Comment


                  • #24
                    That may be so, but i prefer solutions with less components anyway. Less is more.

                    Comment


                    • #25
                      Great thanks , guys , for your kindly response ! I see that I didn't waste the time to draw it ... By the way , about this amount of transistors - it's not a whim but just a purposeful decision . You see , it's a good idea to use IC when we need to solve a common problem , and a great amount of IC's are such a typical blocks for many typical solutions . Amplifier , demodulator , counter , etc . But sometimes we meet some special requirements in our design , and cannot find a special IC for this .... and the best thing to do here is to use old and good transistors
                      And if I try to use IC here - I don't think I will obtain some better parameters . Of course I can get all my gain using only one good OP-amp , but this OP-amp must have enough speed to process those pulses with suitable precision , and this amp may consume the current more than entire signal processing circuitry ... I think that it's not good anyhow . But now this "analog signal processor" does consume only 1,5 mA current ( due to pulse operation of all stages , of course ) , and almost all power goes to the power chain , to pump the energy to the coil . Of course , if someone find the IC that can do the job better with less energy consumption - let's use it , why not ?

                      Comment


                      • #26
                        Originally posted by deemon View Post

                        But now this "analog signal processor" does consume only 1,5 mA current ( due to pulse operation of all stages , of course ) , and almost all power goes to the power chain , to pump the energy to the coil .

                        You are real Deemon Mountain King Cobra of power saving MD design. Congratulations again.

                        Comment


                        • #27
                          Originally posted by Aziz View Post
                          Yep, that's really art work.

                          (Deemon must be one of the good old school guys. The newer "facebook generation engineer's" even don't know, how the op-amp is internally working. Good work. Hats off!).


                          Déjà vu
                          BTW, the sound card detector controller platform is already supporting one of the proposed (transmitter) designs (left transmitter configuration in the above circuit). The TR1 & Rb isn't required however. The bipolar transistor and the reverse biased diode is the low-side n-ch mosfet with inherent body diode. (The high-side p-ch mosfet isn't installed.) The coil need no parallel capacitor installed (but can be) and another capacitor (series C) is connected parallel to the low-side n-ch mosfet.

                          Cheers,
                          Aziz
                          Of course , this capacitor connection doesn't give a significant difference - one can connect it parallel to the coil , or parallel to the mosfet ( or bipolar ) ... but what I can't understand is how they obtain the receiver signal from the coil without a current trans ? Or maybe they use another receiver coil ?

                          Comment


                          • #28
                            Originally posted by deemon View Post
                            Of course , this capacitor connection doesn't give a significant difference - one can connect it parallel to the coil , or parallel to the mosfet ( or bipolar ) ... but what I can't understand is how they obtain the receiver signal from the coil without a current trans ? Or maybe they use another receiver coil ?
                            Hi deemon,

                            yep, there is another receive coil. Usually an induction balanced coil configuration.
                            Aziz

                            Comment


                            • #29
                              Very interesting work deemon but I must say, mostly for educational purposes. People can learn a lot from your electronics experience, including myself, but I am (was not!!!) the "simple" approach guy and have learned one thing with metal detecting: Simple, clever solutions that do not degrade SNR and add into the drift equation.

                              Comment


                              • #30
                                Originally posted by straton View Post
                                Very interesting work deemon but I must say, mostly for educational purposes. People can learn a lot from your electronics experience, including myself, but I am (was not!!!) the "simple" approach guy and have learned one thing with metal detecting: Simple, clever solutions that do not degrade SNR and add into the drift equation.
                                Maybe you are right , maybe not , who knows .... but as for me , I just found a strange and as it seems to me beautiful path in a big world of metal detecting What I did - I followed this path , explored it as I could ... but whether people go this way - I don't know

                                Comment

                                Working...
                                X