Announcement

Collapse
No announcement yet.

TINKERERS_TEM-TX

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

  • #31
    Originally posted by Halo View Post
    Chan 2 is definitely set to DC. Probes properly calibrated. Hmm? I wish I had another scope on hand to compare. Maybe someone could post what their TP1 looks like?

    A couple other question Tinkerer? Did you have ideas for the 4 FETS and their driver (U3710)? I was going to try 4 IRF740 FETS and a TC4421(inverting). I'll LTSpice this circuit first. For the 24V IN would be battery or boost? I'll post some pics as I progress.

    Thanks.
    Hi Halo,

    Don't worry too much about the overshoot. It should not be a problem with the TC4421. The IRF740 will work, but it will give you only half as much maximum power than an 800V Mosefet, because the maximum Flyback voltage is the limiting factor of this circuit.

    The TINKERERS_TEM-TX is a TX drive platform that allows for a very wide variety for experimenting. If you have a specific goal in mind, I can help you defining the best output for that purpose.

    Tinkerer

    Comment


    • #32
      24V IN.

      You can use anything from 12V to 30V.

      It is good to have a regulated supply. The voltage of the batteries, specially the Li-Ion ones, can vary a lot between fresh charged full and low and this will change the TX power.

      Tinkerer

      Comment


      • #33
        Hi Tinkerer,

        I was wondering whats your reasoning behind having a resistor and pot in between the coil and tank caps? If is power control then you wouldn't you be much better off sticking them in between the coil and power supply? That way you more efficiently regulate the power going into the system. Where they are they are sapping power every time the current sloshes back and forth from the coil to the caps. Even better would be not to have them at all and just run the coil of the right voltage to get the power level you want.

        Midas

        Edit: Sorry scratch all that its only on the bucking control isn't it... Think twice.. post once....I need coffee

        Comment


        • #34
          Originally posted by Midas View Post
          Hi Tinkerer,

          I was wondering whats your reasoning behind having a resistor and pot in between the coil and tank caps? If is power control then you wouldn't you be much better off sticking them in between the coil and power supply? That way you more efficiently regulate the power going into the system. Where they are they are sapping power every time the current sloshes back and forth from the coil to the caps. Even better would be not to have them at all and just run the coil of the right voltage to get the power level you want.

          Midas

          Edit: Sorry scratch all that its only on the bucking control isn't it... Think twice.. post once....I need coffee
          The Bucking control:

          The Bucking control is not needed for the functioning of the detector. It is only there for compensating for slight induction balance shifting during the casting of the coil.

          When we assemble the coils, TX coil, Bucking coil and RX coils, we carefully balance the induction to obtain a minimum of residual voltage on the RX.

          The actual amount of residual voltage varies according to the size of coils, the type of wire used and the precision in the coil diameters, amount of turns and inter-wire capacitance.
          With the larger coils, like 1 meter and more, I am happy to get a residual voltage of 50mV peak to peak. However, I know a coil builder who builds his coils to such a high precision, that he gets 1 to 2 uV residual.

          This is the induction balance that we desire. Once we have balanced the coils, we cast the assembly. If we get a minor shift of the coil balance at that time, there would be no way to correct that. This is where the Bucking coil control comes in. It makes it possible to adjust the induction balance by about 1%.

          The induction balance adjustment may also help compensating the effect of extreme magnetic ground, although the TINKERERS_TEM, has such a wide dynamic range that this probably will never be needed.

          Tinkerer

          Comment


          • #35
            Quaqsi-resonant PI

            The good old ideas..........

            quasi-resonant discriminator coil for PI

            http://www.findmall.com/read.php?34,130077

            Tinkerer

            Comment


            • #36
              Originally posted by Tinkerer View Post
              The Bucking control:

              The Bucking control is not needed for the functioning of the detector. It is only there for compensating for slight induction balance shifting during the casting of the coil.

              When we assemble the coils, TX coil, Bucking coil and RX coils, we carefully balance the induction to obtain a minimum of residual voltage on the RX.

              The actual amount of residual voltage varies according to the size of coils, the type of wire used and the precision in the coil diameters, amount of turns and inter-wire capacitance.
              With the larger coils, like 1 meter and more, I am happy to get a residual voltage of 50mV peak to peak. However, I know a coil builder who builds his coils to such a high precision, that he gets 1 to 2 uV residual.

              This is the induction balance that we desire. Once we have balanced the coils, we cast the assembly. If we get a minor shift of the coil balance at that time, there would be no way to correct that. This is where the Bucking coil control comes in. It makes it possible to adjust the induction balance by about 1%.

              The induction balance adjustment may also help compensating the effect of extreme magnetic ground, although the TINKERERS_TEM, has such a wide dynamic range that this probably will never be needed.

              Tinkerer
              When you say you achieved a residual voltage of 50mv is that some sort of test, exciting the coil with a standard sine wave? If so what frequency do you use?
              Because in you post here: http://www.geotech1.com/forums/showp...52&postcount=2 when you actually start pulsing it TEM style you were only able to knock it down to 6v for the switching noise, ~3.5v for the signal. Obviously with your method its not a problem anyway but do you think you could achieve a better balance while pulsing TEM style if you wanted to?

              Cheers
              Midas

              Comment


              • #37
                Originally posted by Midas View Post
                When you say you achieved a residual voltage of 50mv is that some sort of test, exciting the coil with a standard sine wave? If so what frequency do you use?
                Because in you post here: http://www.geotech1.com/forums/showp...52&postcount=2 when you actually start pulsing it TEM style you were only able to knock it down to 6v for the switching noise, ~3.5v for the signal. Obviously with your method its not a problem anyway but do you think you could achieve a better balance while pulsing TEM style if you wanted to?

                Cheers
                Midas
                Thanks for the feedback. You have sharp eyes and keen perception.

                Now, where do I start with the explanations.

                At the post you mention, above, I show a signal as it appears while balancing the coil. This coil assembly is open and not cast yet, the windings are loosely tied down, but can still be moved.

                At this point we use a full power TEM-TX to produce the maximum residual voltage. This helps us with the balancing.

                With slight movements of the windings, (remember, there are 3 different windings) we reduce the residual voltage until we have obtained the minimum possible. If this minimum is not good enough, we better dismantle the coil and build a different one, but it is good to make some tests first, to recognize the cause of the problem with the balance. A wrong turn count for example, would make the coil impossible to balance.

                I have a few coils and a few TX boards and a few RX boards. when I look for answers, I connect one of each and fire away. It may be a 2 meter coil or a 45cm coil, or a low power TX or any combination, depending on what the question is that I need to answer.

                The residual peak to peak voltage is depending on many factors, one important factor is the Flyback voltage. If the Flyback is about 800V, the residual will be higher than if the Flyback is 300V. In this instance one could call it a percentage of the Flyback voltage.

                However, there are many other factors involved that make up the residual voltage.
                The size of coil has a very strong influence.
                Coil capacitance seems to be a factor. This is a bit surprising, since the coil and resonant capacitor form a tank.
                The type of wire used to make the coils seems to have some influence too.
                The cables connecting the coils to the board have a significant influence.
                The Mosfet switching noise is important too.
                All the factors combined produce a residual voltage at the input of the pre-amp.

                If we use a +/-5V supply for the pre-amp, we can amplify a 50mV signal about 70-80 times without saturating the pre-amp.
                Reducing the bandwidth, effectively filtering out the high frequency noise, we can get a bit more amplification.

                Using a +/-12V supply we can still amplify more.

                In general we want to use the highest gain possible at the pre-amp, without saturation.

                If we clip and or saturate, we distort the signal and throw a good part of the information it contains away.

                I hope to have explained a few things, but there are many more things involved.

                Tinkerer

                Comment


                • #38
                  Originally posted by Tinkerer View Post
                  Thanks for the feedback. You have sharp eyes and keen perception.

                  Now, where do I start with the explanations.

                  At the post you mention, above, I show a signal as it appears while balancing the coil. This coil assembly is open and not cast yet, the windings are loosely tied down, but can still be moved.

                  At this point we use a full power TEM-TX to produce the maximum residual voltage. This helps us with the balancing.

                  With slight movements of the windings, (remember, there are 3 different windings) we reduce the residual voltage until we have obtained the minimum possible. If this minimum is not good enough, we better dismantle the coil and build a different one, but it is good to make some tests first, to recognize the cause of the problem with the balance. A wrong turn count for example, would make the coil impossible to balance.

                  I have a few coils and a few TX boards and a few RX boards. when I look for answers, I connect one of each and fire away. It may be a 2 meter coil or a 45cm coil, or a low power TX or any combination, depending on what the question is that I need to answer.

                  The residual peak to peak voltage is depending on many factors, one important factor is the Flyback voltage. If the Flyback is about 800V, the residual will be higher than if the Flyback is 300V. In this instance one could call it a percentage of the Flyback voltage.

                  However, there are many other factors involved that make up the residual voltage.
                  The size of coil has a very strong influence.
                  Coil capacitance seems to be a factor. This is a bit surprising, since the coil and resonant capacitor form a tank.
                  The type of wire used to make the coils seems to have some influence too.
                  The cables connecting the coils to the board have a significant influence.
                  The Mosfet switching noise is important too.
                  All the factors combined produce a residual voltage at the input of the pre-amp.

                  If we use a +/-5V supply for the pre-amp, we can amplify a 50mV signal about 70-80 times without saturating the pre-amp.
                  Reducing the bandwidth, effectively filtering out the high frequency noise, we can get a bit more amplification.

                  Using a +/-12V supply we can still amplify more.

                  In general we want to use the highest gain possible at the pre-amp, without saturation.

                  If we clip and or saturate, we distort the signal and throw a good part of the information it contains away.

                  I hope to have explained a few things, but there are many more things involved.

                  Tinkerer
                  Thanks for the kind words Tinkerer. So you have achieved a 50mv balance with the TEM transmitter then, just not in that particular post? As I understand the induction balance proccess its only really possible to achieve a 'perfect balance' at one particular frequency. Change the frequency and you change the degree of coupling between the RX/bucking and TX/RX. So given the vast difference between the switching noise frequency and the excitation frequency, I think 50mv is very good.

                  Midas

                  Comment


                  • #39
                    Originally posted by Midas View Post
                    Thanks for the kind words Tinkerer. So you have achieved a 50mv balance with the TEM transmitter then, just not in that particular post? As I understand the induction balance proccess its only really possible to achieve a 'perfect balance' at one particular frequency. Change the frequency and you change the degree of coupling between the RX/bucking and TX/RX. So given the vast difference between the switching noise frequency and the excitation frequency, I think 50mv is very good.

                    Midas
                    I think frequency is the wrong word to use for the TINKERERS_TEM. It is a pulsed TX.
                    It is better to speak of pulses per second PPS or pulse repetition rate PPR.

                    To confuse the matter a bit more, the TEM TX does have a resonant frequency too. I prefer a resonant frequency of about 50kHz, combined with 5kHz PPS. The decade proportion is just incidental, has no correlation.

                    The switching noise has a frequency of several MHz. Interestingly, it excites very small targets very well, but I doubt if it will penetrate deep into the ground.

                    For simplicity I just filter out all frequencies that are of no particular interest, including RF, since it is a common source of noise.

                    I have achieved much better than 50mV with some coils, but, in order to understand the causes and find solutions, I have tried extreme coil building methods, like using #14WG mono wire or #10AWG household cable to build coils. (Mostly just because I had it available) Observing the differences in the final coils helped me understand the reasons of the higher residual voltage.

                    The biggest TX coil was about 24 meters in diameter, all around the outside of the house.
                    This TX as not balanced, instead I balanced the RX coil. I was quite amazed how good it worked.

                    Tinkerer

                    Comment

                    Working...
                    X