Announcement

Collapse
No announcement yet.

Ground Balance

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

  • Ground Balance

    Hi all, I've been building a version of the hammerhead PI with some changes including a second amp in the frontend and a Atmega8 to run the timing of the TX and both sample channels. This section is working correctly.

    What I want to do is add another sample channel for ground balance because the ground where I want to use it is highly mineralized the problem is I can't seem to find much information on ways to implement this type of mod.


    My first idea was to set the Atmega8 up to transmit 3 TX pulses at about 100us each with 100us receive with between each and sample two times for each receive time one for target and one for earth field cancel. Then have a receive period after the third TX pulse of say 300us and sample once during this time for ground balance, then amplify and invert and feed back into the main channel output.

    Would this idea work and does the ground mineralization take a long time to decay ?


    Thanks
    Mick_GD

  • #2
    Just the opposite... mineralization decays rather quickly. Here is how the TDI does it (this is all public knowledge, so NBD):

    Turn TX on for 100us
    Shut off, enter flyback
    Wait 10us, take the Target sample
    After the end of the Target sample, wait a short delay, take the GB sample
    After the end of the GB sample, wait a long delay, take the two EF samples
    Use a diff integrator to subtract Target-EF1
    Use another diff integrator to subtract GB-EF2
    Subtract A1*(Target-EF1) - A2*(GB-EF2)
    Adjust A2 for ground balance

    In TDI, all the samples have identical widths. I could tell you what to use, but you would learn more trying a range of settings, say 5us up to 25us. The delay between Target and GB samples as well; try 5us to 25us. Technically, this same delay should be used to space the two EF samples. The spacing between GB and EF1 samples should be 50us or more.

    There are other ways to do GB, even simpler, but this one has particular advantages that TDI uses to produce distinct audio responses.

    - Carl

    Comment


    • #3
      Hi Carl, thanks for the reply, i'll see if I can implement something similar and see what happens. When I have it finished I'll post some results and info about how i've set it up.



      Thanks
      Mick-GD

      Comment


      • #4
        Carl, I forgot to ask, the EF sample is that short for Earth Field ?



        Thanks

        Comment


        • #5
          Originally posted by Mick-GD View Post
          Carl, I forgot to ask, the EF sample is that short for Earth Field ?
          Yup

          Comment


          • #6
            HI,

            This idea is more or less like the picture?

            Regards
            Taktyk
            Attached Files

            Comment


            • #7
              Yup!

              Comment


              • #8
                Hi Taktyk, thanks for the timing picture.

                With the EF sample points I'm going to try sampling them further out and just before the beginning of the next TX cycle, doing this would lessen the chance of bigger rubbish targets being detected in the EF sample points.


                Hi Carl, just a question about the diff integration in the HH, it uses one IC TL072 for both channels. Before I start experimenting with this section what are your thoughts on sampling into four separate integrators then into other IC's to diff the signals, good or bad idea.





                Cheers
                Mick-GD

                Comment


                • #9
                  Generally (even in VLF) you want the integrators to be tightly matched, which usually means hand-matching at least the caps. The more integrators you have, the harder it is to get them all to match. By running 2 samples into a single integrator you reduce the problem.

                  However, a differential integrator (as is used in HH) uses 2 caps which need to match (although I never really said so in the documentation), so it is no better than using 2 separate integrators. A better (often used) method is to first invert the preamp output, then gate the normal and inverted signals into a single-ended integrator.

                  In a TDI-style GB design, if you don't match the delays through both analog paths then you will never be able to hit a quiet GB point.

                  - Carl

                  Comment


                  • #10
                    Hi Carl, thanks for the added info. Some time back I was looking at the SD2000 schematics posted in another location and I see they do pretty much the same in that they invert the output of the preamp and feed feed the inv and non inverting inputs into the 4066 and use a variable resistor on the output of each channel for balance then the output connects to the integrator. So I might go down that path.

                    Time for some experimenting.

                    Comment


                    • #11
                      Mick

                      something I've been doing in my experimenting is checking the noise characteristic's of op-amp's verses gain, i downloaded "Tina Ti" from Texas instruments, easy to use, kinda fun!

                      Philip

                      Comment


                      • #12
                        Here's a picture of my first post frontend design, it had been breadboarded and does work but if anyone has any ideas for improvement any suggestions would be grateful.

                        All the components used are what I had in stock.

                        The second inverted output going to the 4066 has extra gain to compensate for a lower level of signal plus it allow the 10k pot to be adjusted more towards the S1 sample point, and this gives more response to wanted targets.

                        The next part is to add the GB and EF2 sample circuits, I'll post up when done.




                        Uploaded with ImageShack.us

                        Comment


                        • #13
                          +9 volts?

                          philip

                          Comment


                          • #14
                            Originally posted by hdphilip View Post
                            +9 volts?

                            philip

                            Yep 9 volts. Although it will run fine on 12 or 14. The reason it was on 9 is because the 7660 I'm using to make the +5 volt line is only a 10 volt max version. Making a better +/- voltage circuit is one of my next jobs.

                            Comment


                            • #15
                              The way I see it - it is a weighted sampling. It must be interesting to see what set of weighting curves would produce the same result with arbitrary sampling. With proper weighting function it should work with large tolerance for timing errors.
                              (Just thinking out loud)

                              Comment

                              Working...
                              X