Announcement

Collapse
No announcement yet.

My Homemade VLF Detector

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

  • My Homemade VLF Detector

    Hi guys, I'm attempting to build my own VLF detector. Well, almost homemade. I did have to buy the search coil. I tried for awhile to wind my own coil, but decided to buy the Excelerator EQ2 coil. I've been working on my circuit, but I can't seem to get a good detection signal on the Rx side of the coil. In order to get a good signal of about 100+ mV peak to peak, I have to put a big piece of metal inside the ring. Normally when there is no metal around the coil I see about 20 mVp-p.

    Can you guys help me understand the secret? I've varied the frequency. I also varied the parallel capacitor across the Rx coil. Nothing seem to help to be able to detect metal unless I'm right in the coil ring.

    I'll see if I can attach my schematic here. I hope to be able to share my work with everyone as I build this circuit. But I'm really stuck right now.

  • #2
    Here's a schematic showing my Transmit circuit and my Receive circuit.


    Comment


    • #3
      Heres some pics of my waveforms.

      First, this shows the Tx and Rx with the coil laying on the carpet.



      Next, this shows the Tx and Rx when I put a mini aluminum flashlight inside the Double D coil. The flashlight causes the Rx waveform to be almost 400 mV p-p.

      Comment


      • #4
        On the RX amp, 0.22uF is too low, try 1uF... and 0.01uF is too high, try 470pF or less.

        Comment


        • #5
          Ok, thanks Carl. That Rx circuit came from the ACE250 schematic that is floating around here. Although they used a 0.018 uF. But I changed it to 0.01 uF. One thing that really surprised me is that the preamp gain on that ACE250 schematic is 1000! wowser! I kept that 1000 gain setting (the 47K/47 resistors) but doesn't seem to help me any.

          Will let you know what happens...thanks again!

          Comment


          • #6
            Originally posted by JamesPicard View Post
            That Rx circuit came from the ACE250 schematic that is floating around here. Although they used a 0.018 uF. But I changed it to 0.01 uF. One thing that really surprised me is that the preamp gain on that ACE250 schematic is 1000! wowser! I kept that 1000 gain setting (the 47K/47 resistors) but doesn't seem to help me any.
            I wouldn't trust that schematic. The component values look completely wrong. As you correctly pointed out, the gain is way too high (it's over 60bB), and the cap values give a break frequency of 15.4kHz (too high) and a cutoff frequency of 186Hz (too low). In fact, these should be the other way round. No wonder you cannot get any response from this circuit. There's also another "clue" here - the feedback resistor is shown as 47.5k, which is not a standard value - which implies that the person who back-engineered the Ace 250 schematic measured everything in-circuit. Try using the TGSL preamp for better results ->
            http://www.geotech1.com/forums/showthread.php?t=15710

            What inductance values do you have for your coil?

            Comment


            • #7
              Yes, your right. That 47.5K isn't standard, so I used a 47K. This is the Excelerator EQ2 Pro coil. I found the specs somewhere online and saved it, I think the Rx side is 6 mH. I'll look through my notes.

              Comment


              • #8
                The 47K might not be good either, as Qiaozhi stated in the post above.
                "which implies that the person who back-engineered the Ace 250 schematic measured everything in-circuit"

                This means the values vary dramatically when measured on the board like that. Quite often parts need to be isolated from the board to be measured properly.

                The "47K" might not be any where near 47K. 47K could be the combined sum of two or more resistors etc.

                Comment


                • #9
                  While I agree there are circuit errors, I feel that you have unrealistic expectations of the signal level you will obtain from a metal target. On one of my commercial detectors, the static output of the coil is about 2mV peak-to-peak, when driven with 12V pk-pk on the transmit coil. The minimum change in output needed to give a 'detectable' beep is about 0.5 microvolts pk-pk. To obtain a change that could be seen on a 'scope screen, say 0.1mV, would require the object to be within a few cm of the coil, as you observed.

                  Comment


                  • #10
                    Well, I think that ACE250 schematic floating around here has quite a few errors. But I did change my receive circuit over to the design that Qiaozhi showed me. So this is what I have now for the Rx side...




                    However, I notice that it seems like the gain is still too high. I say this because even when the coil is in the air with no metal around it I measure almost 700mVp-p across the 1K resistor. When I put my little aluminum flash light in the coil ring I measure about 2Vp-p. I don't think I should ever measure that high of voltage on the Rx side.

                    So I disconnected the Rx coil from the op amp and left the 0.018uF in parallel. When I measure across the rx coil (not connected to the circuit), it measures basically like Skippy says....about 1-2mVp-p. Then when I put my little flashlight into the coil ring, I measured a sine wave of about 40mVp-p.

                    So, it seems like there way too much gain in the op amp circuit, don't you guys think?

                    Comment


                    • #11
                      Did some more measurements to be sure....with the coil in the air and not connected to the op amp, its about 10mVp-p. When I put my flashlight in the coil ring, I measure 80mVp-p.

                      I notice that if I can swing the flashlight about a inch above the coil, and I can see the amplitude of the signal increase a little bit. So it does see something even when it's not in the ring. But swinging a quarter above the coil about an inch away and i can't see any change in the signal. I have to put the coin right on the coil where the double D's overlap to get the signal amplitude to increase.

                      More studying to be done.

                      Comment


                      • #12
                        Originally posted by JamesPicard View Post
                        Did some more measurements to be sure....with the coil in the air and not connected to the op amp, its about 10mVp-p. When I put my flashlight in the coil ring, I measure 80mVp-p.

                        I notice that if I can swing the flashlight about a inch above the coil, and I can see the amplitude of the signal increase a little bit. So it does see something even when it's not in the ring. But swinging a quarter above the coil about an inch away and i can't see any change in the signal. I have to put the coin right on the coil where the double D's overlap to get the signal amplitude to increase.

                        More studying to be done.
                        With an RX coil inductance of around 6mH and a tuning cap of 18nF, the receiver is tuned to 15.3kHz. So I assume your TX is also somewhere near that frequency.
                        The gain of the TGLS preamp should be ok (it's 33.4dB).

                        A voltage of 10mV at the RX coil, with no metal targets, is a good value.

                        Comment


                        • #13
                          Could you give us more info about the coil you bought, such as what machine it is for, or what frequency it is intended to be used at? You should use the correct frequency in your circuit, because during manufacture, the coil is nulled (induction-balanced) at that frequency, and the null will be worse at other freqs. The receive coil is normally tuned with a load capacitor so that it is resonant at a frequency 30% higher or lower than the transmit frequency (this is a guide, others may have more detailed suggestions)

                          Comment


                          • #14
                            The coil I bought is a 10" x 14" Double "D" Excelerator Search Coil for the Garrett ACE 250 and 150. Kellyco Detectors calls it the EXcelerator II EQ2.

                            I guess I let that ACE250 schematic sway my thinking too much. It was saying the Tx frequency should be 7.5 KHz. So that's what my Tx is currently. But I just did some searching in the internet and websites are saying the Tx frequency of the Garrett ACE250 is 6.5 KHz.

                            Comment


                            • #15
                              Originally posted by JamesPicard View Post
                              The coil I bought is a 10" x 14" Double "D" Excelerator Search Coil for the Garrett ACE 250 and 150. Kellyco Detectors calls it the EXcelerator II EQ2.

                              I guess I let that ACE250 schematic sway my thinking too much. It was saying the Tx frequency should be 7.5 KHz. So that's what my Tx is currently. But I just did some searching in the internet and websites are saying the Tx frequency of the Garrett ACE250 is 6.5 KHz.
                              If your TX frequency is 7.5kHz, then your tuning capacitor needs to be 75nF for a 6mH coil, if you want to match the TX frequency. However, it is much easier to design your detector if the frequencies are slightly offset, resulting in an initial phase-shift between the TX and RX waveforms. I would try using 68nF, which would tune the RX coil to 7.9kHz. Or, alternatively, use 82nF to get 7.2kHz.

                              Comment

                              Working...
                              X