Announcement

Collapse
No announcement yet.

preamp types

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

  • #16
    Originally posted by detectormods View Post
    Most engineers do not take the actual ground noise into consideration, even at 17 c (290K) the noise figure from the ground is 3db so using a preamp with a lower noise figure just amplifies the ground noise. The important aspects of the input stage is dynamic range, slew rate, settling time and low distortion.

    http://cp.literature.agilent.com/lit...5952-3706E.pdf

    The Link provides some useful information on noise measurement.
    Although this is an interesting document, it is not relevant here..
    In the first sentence it states: "Noise figure is a key performance parameter in
    many RF systems.".
    A metal detector is not an RF system.

    Comment


    • #17
      Its Not RF but the mathematics do not change. One can view a VLF as RF but a P.I is different i know. Then again basic noise figure measurements are still valid.

      Comment


      • #18
        Then all this noise is averaged whit long time constant integrator (except maybe low freq. noise component, may not be totally random), EMI interference will unfortunately end the game long before noise can. Anyway, some form of bandlimiting may help, in extent it can be done. Not much in classic PI, only frequency band below detector pulse rate is not needed. Whit input amp blanked, using higher operating freq. and fully floating integrator, amplifier chain can be AC coupled and high pass filtered, this will help a lot whit at least low frequency (power line related) interference. In some other configurations, like Fisher Impulse style, heavy bandpass filtering can be used.

        Comment


        • #19
          Lets look at Bruce Candy's designs, every P.I detector from the SD2000 has state of the art lowest noise figure technology at the time. There must be a logical explanation for this design approach. To date I have not seen any Pulse induction detector beat a Minelab for sensitivity or depth. Even on the hotest mineralized soils of Australia, the Minelab wins every time. If noise figure is not an issue then what is providing all that extra performance? My latest mods that match the input circuit gain to the return signal from reactive minerals does increase the total performance but it is more an issue of dynamic range rather than noise figure.

          Comment


          • #20
            The integrator does a good job at attenuating HF noise.
            The one kind of noise that is hard to filter out, is the 1f noise. The noise figure we need to take a close look at, is the 1 to 10Hz noise.
            If it is at 1uV, how are we going to differentiate this noise with a target signal that is of the same frequency and same amplitude?

            Tinkerer

            Comment


            • #21
              Originally posted by detectormods View Post
              Lets look at Bruce Candy's designs, every P.I detector from the SD2000 has state of the art lowest noise figure technology at the time. There must be a logical explanation for this design approach. To date I have not seen any Pulse induction detector beat a Minelab for sensitivity or depth. Even on the hotest mineralized soils of Australia, the Minelab wins every time. If noise figure is not an issue then what is providing all that extra performance? My latest mods that match the input circuit gain to the return signal from reactive minerals does increase the total performance but it is more an issue of dynamic range rather than noise figure.
              Far from that noise figure is not important, just alone will not win the race. ML’s are bit specific in many ways, my comment is more about classic PI like HHd or surf etc. Some early detectors built whit LM709 amplifiers, not known to be especially quiet, when nothing else was available, worked quite well.

              Comment


              • #22
                I built many front end designs with lm5534 devices and they work fine, I can recall even using LM301 / 701 or something like that 10 years ago. Its so long ago i cannot even remember the number..lol Why is it that if Noise figure is non important in a P.I design then why does the lowest noise figure designs beat all other detector designs for overall performance? I really think that there is something else going on that is not understood.

                Is it the low NF or something else? The question remains unanswered.

                Comment


                • #23
                  Originally posted by detectormods View Post
                  Most engineers do not take the actual ground noise into consideration, even at 17 c (290K) the noise figure from the ground is 3db so using a preamp with a lower noise figure just amplifies the ground noise.
                  I don't understand this... ground doesn't have a "noise figure". And I think you are misunderstanding what "noise temperature" means, it's just an equivalent way of expressing noise figure. That is, a device with a NF of 1dB has an equivalent noise temperature of 75K, but that doesn't mean the device has a physical temperature of 75K (brrrr!). Conversely, a device (or soil) at room temp doesn't automatically have a NF of 3dB.

                  As I've shown in some simple examples, NF in metal detector designs don't mean squat. I can make the preamp NF look stellar by putting a horribly inefficient and noisy coil in front of it. And, by making real improvements to noise, the (otherwise unchanged) preamp ends up with a worse NF. Detector designers just don't care about NF.

                  That's not to say we don't care about noise. We do, but we're often sloppy about it, because the integrator helps correct our sinful ways. Tinkerer is right, 1/f noise can be a gotcha, but generally post-integrator/demod, not at the preamp. I don't disagree that noise needs to be considered more carefully, but the concept of noise figure is useless here.

                  Comment


                  • #24
                    I think output offset drift is more important than component noise. Buy nature high quality low noise components have a low temperature drift.

                    So when you see a pre-amp with a super matched transistor pair feeding a high quality
                    op-amp using low value metal film precision resistors. You think super low noise, Yes.

                    More important is offset drift. When the super matched transistors are used as a differential pair, they drift equally with temperature and maintain as close to zero offset as possible. So with modern detector designs using system gains in the tens of thousands. A small 50uV offset can become a volt at the speaker. Or rail the output of a later stage.

                    Anyone who used detectors before the mid 80's such as the Garrett groundhog. May remember having to ride the threshold knob and hitting the retune button often. The big break through was using ac coupled amplifiers, which maintained close to zero offset. It served two purposes, AC coupling acted as a high pass filter. Blocking low frequency ground signals and passing higher frequency targets. Second was to eliminate the effects of component drift, crappy designs and allow for higher gains.

                    Look at the GS4/5 and the SD2000 designs. All full of trimmers and ac coupled amps to eliminate offsets.

                    Comment

                    Working...
                    X