Announcement

Collapse
No announcement yet.

Modular Metal Detector

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

  • Modular Metal Detector

    Several years ago I developed a modular metal detector platform whereby individual subcircuits are broken out into plug-in modules. Here is the motherboard:

    Click image for larger version

Name:	Rev1Bare.jpg
Views:	958
Size:	559.6 KB
ID:	408808
    The modules are Preamp, Demod, post-demod Filter, ADC, and Micro. The micro was an STM32F767 Nucleo (Cortex M7), the ADC was the ADS131E08 (24-bit 8-ch simultaneous sampling).
    I had built maybe 30 modules for this but all of this was on my moving truck that was stolen 2 years ago. All I have left is this blank motherboard.

    I had already identified some deficiencies in the design so I redesigned everything last year. Here is the new motherboard:

    Click image for larger version

Name:	Rev2Bare.jpg
Views:	748
Size:	668.4 KB
ID:	408809
    I eliminated the ADC and Nucleo making this an all-analog motherboard. I may eventually design a digital motherboard to mate with this.

    There is a Power module (AVCC) that supplies +/- V to all the modules.
    There is a TX module that can support a basic VLF, PI, or multifrequency transmitter. Bigger, more complex TX circuits will have to use a separate board.
    There are 2 Preamp modules that can have dedicated RX coils or can share a single RX coil (jumper selected).
    There are 4 Demod modules, each one is a quadrature demod. This means you can run a 4-frequency MF design, or do an 8-sample PI design. Each demod can accept inputs from either Preamp (jumper selected).
    There are 4 Filter modules, each follows one of the demods.
    Every module has a clock header.
    A final header contains the 4 (X,R) signal pairs for exporting to an ADC.

    The real purpose of this (for me) is to build parallel channels and compare, ferinstance, different demod topologies.
    Here is the new motherboard with some unfinished modules:

    Click image for larger version

Name:	Rev2Built.jpg
Views:	763
Size:	668.8 KB
ID:	408810
    Most of the components are SMT except where I want to easily swap opamps or caps. It's obviously not meant for high-performance or for outdoor portability but it's a lot better than an ACE breadboard.

  • #2
    My take on a modular platform.
    Post 10.
    https://www.geotech1.com/forums/foru...875-new-member
    Steve

    Comment


    • #3
      Hi Carl,
      I would be very keen to purhase and evaluate your modular concept. I have an SMD reflow oven so all I need are the PCB's and BOM.
      Cheers

      Comment


      • #4
        Originally posted by Stevedbk View Post
        My take on a modular platform.
        Post 10.
        https://www.geotech1.com/forums/foru...875-new-member
        Steve
        I tried that approach a long time ago but never could get the backplane figured out in a way I liked. Looks like you got it solved.

        Originally posted by Dean Sarelius View Post
        Hi Carl,
        I would be very keen to purhase and evaluate your modular concept. I have an SMD reflow oven so all I need are the PCB's and BOM.
        Cheers
        I think what I'll do is move this to the Projects section and start posting schematics and Gerbers.

        Comment


        • #5
          Originally posted by Carl View Post
          I tried that approach a long time ago but never could get the backplane figured out in a way I liked. Looks like you got it solved.
          I think what I'll do is move this to the Projects section and start posting schematics and Gerbers.
          Good!
          I might convert some parts into through-hole for my own experiments.
          I don't prefer such tiny "invisible" components.
          1206 is the all I can do.
          Not only bad eyesight but also shaking hands too.
          Schematics and Gerbers will do.



          Comment


          • #6
            At half sampling rate than what you used; there is ADS1256 ready modules, affordable here:
            https://www.aliexpress.com/item/4001...02200146%21rec

            Comment


            • #7
              Originally posted by ivconic View Post

              Good!
              I might convert some parts into through-hole for my own experiments.
              I don't prefer such tiny "invisible" components.
              1206 is the all I can do.
              Not only bad eyesight but also shaking hands too.
              Schematics and Gerbers will do.


              Do you have a hot air soldering tool? If yes, I can show you and anybody else interested, my crude and simple SMD reflow method. Not good, but good enough for initial POC assembly. The whole board soldered in 5 minutes. I like 1206. The 0604 only under the microscope for placement, at great sacrifice.
              Having separate modules, we have some liberty for the individual modules as long as the connections to the motherboard are respected.

              Comment


              • #8
                Originally posted by Tinkerer View Post

                Do you have a hot air soldering tool? If yes, I can show you and anybody else interested, my crude and simple SMD reflow method. Not good, but good enough for initial POC assembly. The whole board soldered in 5 minutes. I like 1206. The 0604 only under the microscope for placement, at great sacrifice.
                Having separate modules, we have some liberty for the individual modules as long as the connections to the motherboard are respected.
                I have one. Please do tell, do show.
                At the moment Mancha is exploring it:


                Comment


                • #9
                  I use that same blue silicone map to lay out the parts to be soldered.
                  When I order the PCB I also order the stencil for the PCB.
                  With the Stencil I apply the solder paste. For quick and dirty I use low temperature Sn42 Bi58 solder paste (Aliexpress). that makes it easy to remove the parts if I want to change. Place all the parts on their places on the solder paste.
                  Click image for larger version

Name:	SMD_soldering.jpg
Views:	636
Size:	281.9 KB
ID:	408850 Then I set the PCB on a thick aluminum plate and heat the plate with a clothes iron until the solder paste starts to run.
                  Then I run over it with the hot air, not on the components, just around them. As the PCB is already pre-heated, the solder fuses immediately.

                  Comment


                  • #10
                    Originally posted by Carl-NC View Post

                    I tried that approach a long time ago but never could get the backplane figured out in a way I liked. Looks like you got it solved.



                    I think what I'll do is move this to the Projects section and start posting schematics and Gerbers.
                    Hi Carl, do you have a date planned?

                    Comment


                    • #11
                      Sorry, this completely fell off my radar. I'll post more as soon as I can.

                      Comment


                      • #12
                        Originally posted by Carl-NC View Post
                        Sorry, this completely fell off my radar. I'll post more as soon as I can.
                        Why don't we use the cheap PCB 100x100mm / 3.9x3.9 Inches, from JLPCB https://jlcpcb.com/ as standard? In post #9 I show my bi-polar Square current Wave TX board. These boards were designed to be stacked vertically.
                        For development work on the bench it is better to use the boards placed side by side, so that one has access to every part with the oscilloscope.
                        The TX board from post #9 is working good, only small changes to the layout would be needed to optimize it with a back plane.

                        The preamp at post #62 as now been tested on a breadboard. It works really well. This is really the preamp of the AMX Project.

                        https://www.geotech1.com/forums/foru...387#post419387

                        So I propose we merge this Modular Metal Detector Project with the AMX Project.

                        Comment


                        • #13
                          Originally posted by Carl-NC View Post
                          Sorry, this completely fell off my radar. I'll post more as soon as I can.
                          Thanks, waiting.

                          Comment


                          • #14
                            Originally posted by Tinkerer View Post

                            Why don't we use the cheap PCB 100x100mm / 3.9x3.9 Inches, from JLPCB https://jlcpcb.com/ as standard? In post #9 I show my bi-polar Square current Wave TX board. These boards were designed to be stacked vertically.
                            For development work on the bench it is better to use the boards placed side by side, so that one has access to every part with the oscilloscope.
                            The TX board from post #9 is working good, only small changes to the layout would be needed to optimize it with a back plane.

                            The preamp at post #62 as now been tested on a breadboard. It works really well. This is really the preamp of the AMX Project.

                            https://www.geotech1.com/forums/foru...387#post419387

                            So I propose we merge this Modular Metal Detector Project with the AMX Project.
                            Is this schematic for only PI RX preamp

                            Comment


                            • #15
                              For my purposes, the MMD project is already complete. I have a motherboard and numerous modules already designed and PCBs fabricated. It really began as a VLF investigative tool but can be used for PI as well. It is less easily adaptable to complex designs like AMX or stuff moodz is working on, those require more dedicated hardware.

                              Comment

                              Working...