Announcement

Collapse
No announcement yet.

Announcement

Collapse
No announcement yet.

Shielding of TX and RX coils

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

  • Shielding of TX and RX coils

    ​My colleague managed to find a non-working sensor from Manticore and disassemble it. The interesting points turned out to be: the coils TX and RH are wound with a litzendrate. The winding is done on a plastic frame. The graphite screen is applied only to the frame. Thus, there is a gap in the shielding layer at the point through which the litzendrate is wound onto the frame. I'm wondering how important it is to ensure that the entire coil is in a graphite screen? From the experience of manticore, it can be seen that the gap in the screen is acceptable and is due to the very design of the coil and the use of frames.
    Click image for larger version

Name:	RX.png
Views:	70
Size:	165.0 KB
ID:	447547
    Click image for larger version

Name:	ТХ.png
Views:	69
Size:	119.5 KB
ID:	447548

    Earlier, when I was making coils, I shielded the body from the inside and after pouring the final layer of epoxy, I applied a thin layer of graphite screen. Thus, the coils TX and RХ were located in a completely closed screen. And in the manitcore, it turns out that some part of the coil is not shielded.
    I want to try to do a similar experiment: make a winding on the frames and put them in the case. So far, I've modeled this option.
    Click image for larger version  Name:	3.png Views:	0 Size:	254.3 KB ID:	447544
    Click image for larger version  Name:	4.png Views:	0 Size:	197.0 KB ID:	447545
    Click image for larger version  Name:	5.png Views:	0 Size:	139.3 KB ID:	447546
    It's a pity, I found out later that the coils must intersect at an angle of 90 degrees. But for a test sample, I think this design is suitable. Jumpers are used in the coil frames for rigidity, after winding the coils and impregnating with epoxy resin and hardening, these jumpers will be removed.

    As a result: do I need an absolute coil shielding? there should be a gap between the coil and the graphite shield, does the winding frame just provide a gap from the coil to the aeran on 3 sides? You can try applying a medical adhesive tape to the open part of the coil, and then apply a shielding layer on it. Thus, the entire coil will be in the screen, but on the adhesive tape side (on the 4th side), the gap between the coil and the screen will be minimal: about 0.5-1 mm.
    Attached Files

  • #2
    In fact it isn't necessary for the coils to cross at 90 degrees. There are lots of successful designs that don't look like that, for example some types made by Detech (SEF) and Minelab. DD coils often do cross at about 90 degrees but that's just a result of having two overlapping coils that have more or less smooth curves. (It's not often used but two rectangular coils of the same size will also work, with a narrow rectangular overlap region. There is no 'crossing' point as such.)

    Comment


    • #3
      JoyJo, if possible can you post a photo of the Manticore pre-amp pcb? I would like to see how it compares to the Equinox.

      Comment


      • #4
        But there's a problem with the board: the broken coil was bought without a pcb. I would love to look at the board made of the manticore reel myself. But I think it's the same there as in Equinox. In the mantle and envinox blocks there are the same two-channel audio codecs pcm1861 and their binding is the same.

        Comment


        • #5
          Originally posted by Gwil View Post
          In fact it isn't necessary for the coils to cross at 90 degrees.
          No, but it will minimize the overlap capacitance which helps with broadband nulling.

          Comment


          • #6
            Originally posted by JoyJo View Post
            But there's a problem with the board: the broken coil was bought without a pcb. I would love to look at the board made of the manticore reel myself. But I think it's the same there as in Equinox. In the mantle and envinox blocks there are the same two-channel audio codecs pcm1861 and their binding is the same.
            Thanks, I think it's the same, but with some minor improvements. The coil windings also seem to be similar to the Vanquish/Equinox

            Comment


            • #7
              So what about not full shielding of coils?

              Comment


              • #8
                Originally posted by JoyJo View Post
                So what about not full shielding of coils?
                You'll have to test it and see. I always use full shielding on coils, never tried it this way. Back in the old days, some detectors only had shielding on the bottom half of the housing to eliminate ground capacitance. No shielding on the top half, but back then they were not as sensitive so EMI wasn't as much of a problem.

                Comment


                • #9
                  I also agree to the conclusion that full shielding of the coils is needed. Brand detectors don't do this because it's a more time-consuming process and it complicates the manufacturing process

                  Comment


                  • #10
                    I think shielding stabilize capacitance between wound and ground/operator/metal objects, between tx rx coils, etc.. And shielding canceling E field from EM field around sensor.
                    In classic sensor, coils has big inductance and this capacitance can greatly shift amplitude and phase characteristics of rx tx LC tank , and this can shift balance of sensor and (or) can make false beeping of detector. BUT manticore coils has very low inductance, so parasitic capacitance don't greatly shift amp/phs characteristics , and in a result full shielding not need.

                    Comment


                    • #11
                      To make sure that antenna is not responding to E file I put coil inside of copper tube, because I was not convinced that E field does not leak through graphite.
                      Poor shelding cause increase of ground balance.
                      Why you asking that question, whot is the problem with full shelding ?​

                      Click image for larger version

Name:	cu rx.jpg
Views:	0
Size:	23.6 KB
ID:	447611

                      Comment


                      • #12
                        Just add closing ring

                        Click image for larger version

Name:	image.png
Views:	0
Size:	812 Bytes
ID:	447613

                        Comment

                        Working...
                        X