Announcement

Collapse
No announcement yet.

How to judge shield effectiveness?

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

  • #31
    Cable connected to the coil will have small but finite effect on coil performance. Adding 1.5 or 1.8m of coax will add some capacitance, like 150-200pF for common cable types, to other circuit capacitances, but you well may not notice that. This added capacitance will slow down coil a bit, but not dramatically, will change SRF somewhat lower, and if you want to tune detector “on knife edge” small variation of dumping resistor value is needed. No need for some exotic cable types etc. for this coil.

    Best way to adjust detector is with cable, shield and all in place, then you can optimize Rdump value and minimal sampling delay.

    10” coin range is not good at all. In order to utilize full potential of this coil, minimal sampling delay must be set, then all advantage is from this shorter delay compared to standard coils, increased sensitivity to small objects\gold, larger “coin” range etc, no other magic involved.

    If you have scope, this adjustment is simple, if not, try iterative way.
    First, set Rdump to some convenient value, like 330R, set delay pot. to minimum, adjust delay trimmer (reduce resistance) to the point when detector stops responding, if needed, change component values to get that point, then increase delay just a bit above this to get response. Delay now will be around or below 10uS.

    Then try to tweak Rdump value. Reducing it will overdump the coil, delay will increase and at some value detector will again stop responding, so you need to increase delay a bit to get response. This is too low value. From this point, goal is to increase Rdump to the point when minimal delay adjustment can be achieved, but not above that. Too large Rdump value will underdump the coil, some ringing will appear on waveform, but you wont notice that in air tests, analog detectors are very tolerant. However, in ground tests, significant ringing will produce some instability and loss of range.

    Use some pot. or trimmer for this rdump adjustment, then, when you find optimal value, replace with resistor(s) to within 5% of measured value.

    If everything is done right, outdoors tested on battery power, you have to achieve like 11-12” for small gold ring, 13-14” for coin (18mm 1E) or better.

    Please note that this ranges are not guaranteed or absolute, may vary in different designs due to different PCB layouts, noise problems (7660 synced or not etc), component quality\type, and other factors, so i not claim that every construction or kit will work like this.

    Comment


    • #32
      Thank you for that explanation tepco, thats the easiest walkthrough on how to adjust the Rdump ive seen on this forum! Does it matter what pot or trimmer spec I use? I see a bunch with 100k, 500k, 1m ohm ranges. I dont know what those mean but I assume I need to get somewhere around 390ohm (the dumping resistor on the Surf Pi 1.2) so am I correct in thinking that I should probably use a 1k or 10k ohm trimmer?

      I also had one other very beginner question. I find at times when I have the coil laying on a telephone book that the coil will not respond to metal objects. I sometimes have to lift the coil or keep it at an angle. Is this related to shielding or it somehow grounding out? I noticed when I move the coil farther from the pcb board, this is less noticeable?

      Comment


      • #33
        What I’m using for adjustment is dual 2x220R linear potentiometer, series connected, and 100R resistor, providing 100-540R adjustment range. Some coils need higher resistance, so another set using 2x470R pot is good idea. You can use small trimmer too, but not recommended, it will not last for long. Surf and similar detectors are not too “energetic” but some other designs can produce over 1KV flyback with significant energy, naturally, small trimmer will be instantly destroyed.

        Problem with coil sensitivity is obvious, some metal, like steel reinforcement or something below telephone book. You must ensure that no metallic object is present at lest in coil radius, more if possible, when you make adjustments. Also, interference encountered indoors can significantly degrade detector performance, only outdoor testing is relevant.

        Comment


        • #34
          Thanks Tepco,

          Ill order some parts and give it a try, but even when turning the delay pot all the way to the right (maximum turn, till theres the clicking), I get gold ring detection about 10-11" very thin small white gold ring about 5-7". Seems like other surf PI users are hitting this depth as well.

          Comment


          • #35
            Problem with Surf kit is that you wont be able to adjust minimal delay time suitable for this coil type without component value change. You need to reduce R25 (15K in original) set to 4K7 for example, then fine adjust with trimmer, or decrease C9 (1nF in original) in order to decrease minimal delay. With too low values, entire timing circuit will stop working and you loose audio completely, but critical delay adjustment will happen before that point. Simply, component values are selected for slower coils. When you make precise adjustment, you will see improvement in small gold\jewelry detection range, more important than some spectacular increase in sensitivity to coins etc.

            (I never built one from kit, neither I have working one right now, so I can’t compare)

            Comment


            • #36
              Anyone still watching this thread? I'm going to try a flat coil along with the compact type. Was wondering if the old TV rooftop lead-in wire: Low loss 300-ohm 20 AWG foam-insulated twinlead wire would work? It would give greater separation between the coils, which I think was inferred from earlier message. See image.
              Attached Files

              Comment


              • #37
                All of the shielding stuff can be subjective.


                Things I would avoid are:-

                Highly conductive things like silver paint,
                Thick metal shields which can themselves support eddy currents.

                If a detector can detect a sample of your proposed shielding material - I wouldnt use it. e.g. Thick ally foil. or self adhesive copper tape.


                Nickel spray is popular - but it is a good conductor with metal in it - with good conductive layers you would create current flow around the shield, sourced from the Tx magnetic field - this high conducting shiled stops mag feild from the coil getting out. These sheilds require a gap, so no short circuit effect.


                The other way is total coverage with a layer that cant support high circulating Tx currents - so no short circuit, so the mag field gets out. These are the mid conductors like graphite and carbon.


                This effect is popular on the mats used on electonic test or soldering benches - and are static dissipative, - they leak charge accrued from the presence of an E or voltage field though carbonated vinyl rubber to ground.


                You can total cover in this, with no gaps required, as they cant support high circulating currents.



                As for cheap Litz, one source is Induction Hobs - often stripped from kitchen refits and given away. Ebay is good source, old stock or defective units..
                Click image for larger version

Name:	220px-Induktionskochfeld_Spule.jpg
Views:	2
Size:	15.8 KB
ID:	335570

                Steve

                Comment


                • #38
                  My mind is getting kind of boggled on this whole shield thing. Can someone explain in simple terms what the ideal material behavior should have? It seems it shouldn't be metal but carbon or graphite is fine. How about the carbon rubber used for pcb storage bags? I hate the thought of making an awesome coil but having it trashed by doing the shield wrong. So I want a shield tester. How about someone come up with a design for a shield material tester we can all agree on? Then come up with a test method and we all try different shield materials?

                  Regarding the induction hob, what strands/awg is the wire used and how much is in one? A typical coil may need 90 feet. Also does the wire come off that without a problem? Sometimes mfr's glue the wires together and they won't come off without damage.

                  Comment


                  • #39
                    Originally posted by bklein View Post
                    My mind is getting kind of boggled on this whole shield thing. Can someone explain in simple terms what the ideal material behavior should have? It seems it shouldn't be metal but carbon or graphite is fine. How about the carbon rubber used for pcb storage bags? I hate the thought of making an awesome coil but having it trashed by doing the shield wrong. So I want a shield tester. How about someone come up with a design for a shield material tester we can all agree on? Then come up with a test method and we all try different shield materials?

                    Regarding the induction hob, what strands/awg is the wire used and how much is in one? A typical coil may need 90 feet. Also does the wire come off that without a problem? Sometimes mfr's glue the wires together and they won't come off without damage.
                    Always a good question, and I think you'll get a lot of different opinions about it. I personally don't feel I've seen the kind of field data that convinces what is "best".

                    For example, Davor has put forth good arguments for omitting the shield altogether, when using differential coil inputs, and feels his models work well without one. Many commercial models seem to use a "resistive" shield made with carbon. Don Bowers likes conductive mylar for shields. I did some experiments that seem to show different shield materials (mylar vs aluminum) affect the "null phase" of a coil, but no conclusions about depth or usefulness in the field. Aluminum foil is a time-honored material that may not be the best but can give good results.

                    It may be true that it depends on where you will be using the MD. Desert vs wet grass may change which shield works best. If I had to choose, I'd try either conductive mylar (space blanket, but be careful about conductive "gaps") or graphite / glue mixture.

                    Anyway, you should get a lot of advice and opinions here. But if you have the time, experiment and tell us which shields you find work best.

                    Regards,

                    -SB

                    Comment


                    • #40
                      Unknown wire info sought

                      Whilst waiting for my Surf PI kit, thought I'd play around with the coil. This is not the wire I'll be using for the final coil, just wondering what different wire will do/look like. This I think is FM, or maybe some manner of computer wiring. I picked it up at a thrift store. The OD is 12mm x 2mm. There are four strands of fine wire (two to an end) each of which are approx. 4mm x .5mm (actual wire size - flat). The image shows one end split down centre. There are 27 winds, with an OD diameter of 22.7 cm. The insulation is very pliable, almost rubber-like. The wire is very slippery, so must have some manner of coating. It looks like pure copper with unknown clear coating. The stuff is like fuzzy hair and have no idea how many strands per line (in excess of 50?) The only markings on the package were: Part No. 32283 2/16 Gauge - 50FT. Any ideas? My multimeter gives a resistance reading of 0.1
                      Attached Files

                      Comment


                      • #41
                        I think I saw this wire at a Local HI FI shop. They recommend as speaker wire.
                        Henrik

                        Comment


                        • #42
                          Thanks. If it works will be pretty heavy! However, would overcome buoyancy problem and work well in water. That is, if these old arms could stand the resistance for an hour or so!!

                          Comment


                          • #43
                            Originally posted by bklein View Post
                            My mind is getting kind of boggled on this whole shield thing. Can someone explain in simple terms what the ideal material behavior should have? It seems it shouldn't be metal but carbon or graphite is fine. How about the carbon rubber used for pcb storage bags? I hate the thought of making an awesome coil but having it trashed by doing the shield wrong. So I want a shield tester. How about someone come up with a design for a shield material tester we can all agree on? Then come up with a test method and we all try different shield materials?

                            Here is a video to go with the simple circuit posted earlier in the thread. This is just testing the effectiveness of carbon graphite on paper.

                            http://www.youtube.com/watch?v=CiCRF_Bch1k.
                            Attached Files

                            Comment


                            • #44
                              And thoughts as to the effectiveness of: Conductive Cloth Fabric Tape For LCD Cable EMI Masking Material: Polyester fiber cloth Size: 50mm*20M, Coating high conductive copper and nickel in high strength ployester self acylic adhesive. With excellent flexibitity, conduction, abrasion and high temperature resistant

                              Comment


                              • #45
                                Originally posted by dfbowers View Post
                                Here is a video to go with the simple circuit posted earlier in the thread. This is just testing the effectiveness of carbon graphite on paper.

                                http://www.youtube.com/watch?v=CiCRF_Bch1k.
                                I've watched this several times and just can't visually/mentally grasp it. The led is not aimed at the camera so I can't tell the difference between when it may be on or off or in a shadow or not....
                                Unless the series resistance of the fet is really crappy I would think this circuit needs a current limiting resistor for the led.
                                If this is proving that rf energy is building on the fet input enough to bias it on, would ferrite beads on the cable junction (discussed in the patent thread) have the effect of eliminating the rf?

                                Comment

                                Working...
                                X