Announcement

Collapse
No announcement yet.

WT metal detector model

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

  • WT metal detector model

    Hi, I am a year 3 student studying in the UK and I am doing my final year project now. At the beginning of my project, I want to build a small size multiple zone WT metal detector model using 6 metal detectors to detect 6 zones to show the location of the object and I build a metal detector using the circuit design of 'Matchless metal locator'. I found it needs to be adjusted frequently and the detection area is limited so I want to use PI type to achieve it. However, I am not sure if I can build the system using 3 or 6 PI metal detectors and could someone who has experience of it to give me some advice to achieve? In addition, I found that the current product can detect 33 zones but at the airport, it still requires the staff to detect the whole body using a handheld metal detector again so does the multiple zone WT metal detector be applied widely now? My supervisor also asked me to think about how to improve the performance of current WT and I think it can already achieve powerfully functions now. So could someone tell me about the limitation of current product and which aspect that is suitable for me to improve it? My idea is to increase the detection distance by using two several coils which is similar to the wireless charging technology, does someone have any better idea?
    I really appreciate your help and time.

  • #2
    ok you tell us about industrial type detectors that schematics are unaccesable. we are living in world now with no schematics on devices
    while in my youth we had have all schematics on all devices. at first of all you have to go in your university library and take
    Inside The Metal Detector copy to read.

    Comment


    • #3
      what I mean is that the design of the WT metal detector is difficult to achieve since it is complicated and what I want to do is two use several metal detectors to build kind of system like the real product

      Comment


      • #4
        real product WT does Carl M. you see I am only an amateur, i never did work in industrial field. problem of your generation is absolute inaccessibility of schematics you need. take PI way, learn Carl's HH article for start.

        Comment


        • #5

          Comment


          • #6
            .. .. ..
            Attached Files

            Comment


            • #7
              My supervisor also asked me
              ---
              Whites Electronics company - based in UK. ask directly https://www.facebook.com/WhitesUKLTD/
              copy their answer to here.

              Comment


              • #8
                Originally posted by kt315 View Post
                at first of all you have to go in your university library and take
                Inside The Metal Detector copy to read.
                Originally posted by kt315 View Post
                take PI way, learn Carl's HH article for start.
                Both the best way to start.

                The HH is a great introduction to PI detectors and easily modified.
                Here are the Details of my version of the HH with lots of discussion:
                http://www.geotech1.com/forums/showt...ake-on-the-HH2

                As for running multiple "Zones" this might be done with time domain multiplexing of multiple coils. Many inexpensive micro-controllers could do this with a little extra circuitry.

                Comment


                • #9
                  DJSUOL, if you are going to build this with multiple copies of a regular metal detector then they will all need to be synchronized, which is a chore. The way most WTs are designed is with a single TX coil on one side and multiple RX coils on the other side. All the RX coils are processed in parallel channels with their own analog/demod/ADC/DSP. You can do this with either PI or CW but PI is overwhelmingly easier IMO.

                  As you move across the gate from the TX side to the RX side the TX field gets weaker but the RX sensitivity increases. Overall, the cross-gate sensitivity isn't perfectly even but it's not too bad. A high-end WT uses a cross-fired system, where each panel has a TX coil and multiple RX coils, and they are fired in an alternating manner. This is where PI helps.

                  You are right, 33 zones is overkill but multiple zones has been written into a lot of government requirements. Many of those zones are interpolated. That is, the Garrett PD6500 has 3x11 zones, but each panel has 6 RX coils; the other 5 zones are interpolated between the 6 coils, and the entire 11 center zones are interpolated between the left & right side. Hardware-wise, it is a 12-zone WT.

                  Limitations depend on the application. At airports, they have to strike a balance between detecting guns & knives and ignoring watches and eyeglasses. Generally the sensitivity is set to an unaggressive level so that getting a stainless steel knife through is not too difficult. In prisons, they want to detect tooth fillings so small target sensitivity is critical. Ceia is the only detector I know of that meets NIJ requirements for prisons.

                  Other than that, it's mostly bells & whistles.

                  Comment


                  • #10
                    Ťhanks sir. So your suggestion is that I use a bigger coil as the transmitter coil and 3 small coils to receive the signal, am I right? But I think the signals received by the different coils are not the same so I need to design different circuits?

                    Comment


                    • #11
                      my suggestion is to go in an aeroport near your placement and to speak with technicians there. your sitting for PC alltime and gaming in World of Tanks and GTA 5
                      does nothing. you have to action. amigo YOU are new generation, NOT me.

                      Comment


                      • #12
                        Thanks for the overview of how these gates work Carl.

                        Yes, mush easier with one TX Coil and multiple RX Circuits. Many micro-controller chips can easily handle the timing.

                        Comment


                        • #13
                          Originally posted by DJSUOL View Post
                          Ťhanks sir. So your suggestion is that I use a bigger coil as the transmitter coil and 3 small coils to receive the signal, am I right? But I think the signals received by the different coils are not the same so I need to design different circuits?
                          Yes, each RX coil gets its own preamp and demod at a minimum. You can possibly mux them all into a single ADC or use a multi-channel ADC. DSP should treat them separately. It's almost like having 6 metal detectors running, but with only 1 (or 2) TX circuits.

                          Comment


                          • #14
                            so is it possible that I use the current design such as surf or HH to achieve this system and why should I connect the different analog circuit to each Rx? Is it due to the different received signals? Should I set a different threshold value for them? If I want to use this circuit to detect different zones, how can I know if the object is in the left or right?

                            Comment


                            • #15
                              You still need to READ and STUDY the circuits.
                              The heart of a PI detector is the Integrator stage. Carl has written a detailed description in this forum on how and why the stage works.
                              Your first step is to get a basic PI detector working so you can then experiment with different coils and their placement.
                              As to detection in different Zones, this would be where the RX coils are located relative to the metal object excited.

                              Comment

                              Working...
                              X