Find's Treasure Forums
Welcome to Find's Treasure Forums, Guests!
You are viewing this forums as a guest which limits you to read only status.
Only registered members may post stories, questions, classifieds, reply to other posts, contact other members using built in messaging and use many other features found on these forums.
Why not register and join us today? It's free! (We don't share your email addresses with anyone.) We keep email addresses of our users to protect them and others from bad people posting things they shouldn't.
Click here to register!
Need Support Help?
Cannot log in?, click here to have new password emailed to you
Differential PI front end
- Thread starter moodz
- Start date
The differential front end shows some ingenuity, I admit it. However the horses aren't drinking because the arrangement confers so little benefit in relation to the additional complexity required, complexity which has the potential to impair performance and in any case it increases cost.
The differential arrangement provides some rejection of electrostatic interference: however searchcoils are usually electrostatically shielded anyhow and that would probably necessary even with the differential arrangement. Given that most searchcoils are electrostatically shielded, the primary kind of electrical interference a metal detector fights is magnetic field interference which the proposed differential arrangement does nothing about -- that would require a "gradiometer" or far field null receiver coil arrangement, which kills depth, and that's why you rarely see it.
I appreciate moodz' ingenuity and look forward to more posts of his ideas. Just gotta keep the perspective that most ideas seem pretty good when you first have them, and that over time when you've slogged through the problems there aren't very many ideas which are still left standing as good ones. I've been designing beepers for 28 years , and even now, my bad ideas greatly outnumber the good ones.
--Dave J.
The differential arrangement provides some rejection of electrostatic interference: however searchcoils are usually electrostatically shielded anyhow and that would probably necessary even with the differential arrangement. Given that most searchcoils are electrostatically shielded, the primary kind of electrical interference a metal detector fights is magnetic field interference which the proposed differential arrangement does nothing about -- that would require a "gradiometer" or far field null receiver coil arrangement, which kills depth, and that's why you rarely see it.
I appreciate moodz' ingenuity and look forward to more posts of his ideas. Just gotta keep the perspective that most ideas seem pretty good when you first have them, and that over time when you've slogged through the problems there aren't very many ideas which are still left standing as good ones. I've been designing beepers for 28 years , and even now, my bad ideas greatly outnumber the good ones.
--Dave J.
Tinkerer said:
My opinion in this case is expressed by what I actually do. Of the metal detectors I've ever designed, pulse induction or VLF, mono or induction balance, all have used a single-ended preamp. A few of my transmitter designs have been differential (i.e., push-pull).
--Dave J.
Moodz, from your description it sounds like your machine (prior to conversion to the differential arrangement) was picking up electrostatic 60 Hz fundamental. Your proposed differential arrangement would certainly help big bunches with that. But ordinary proper shielding in a single-ended arrangement would also help big bunches. Single-ended and balanced receiver arrangements are equally vulnerable to magnetic field interference.
Most metal detectors have pretty good rejection of electrostatic interference and of low frequency (below the fundamental transmit frequency) interference whether electrostatic or magnetically induced. However some PI designs are markedly inferior from the standpoint of low frequency interference rejection. This may have been factor in your experience. Also, in a home-brew PI it is tempting to not bother with electrostatic shielding of the searchcoil since you can generally get the thing to work without shielding (which is not to say it'll work as well as it could if it had shielding). I don't know how much of this applies to your particular situation.
My favorite PI was the "CodFisher", built as a joke for Jim Lewellen, then president of Fisher. Still have it, but the circuit died a few years ago and I never fixed it. Searchcoil wound on a wicker basket tied to a crooked tree limb, and the electronics in a wooden dried codfish box. Jim didn't know I was building it: I tricked him into buying it at the local fish market. No shielding of anything, ran almost forever on a rectangular 9 volt battery. It worked pretty good by mid 1980's standards and eventually morphed into the Impulse (no longer manufactured). Main thing I noticed with the lack of shielding was increased ground pickup in wet grass or on the wet part of a beach. No particular problems were observed with 60 Hz fundamental because this design had good rejection of low frequency interference. Of course like any PI it would alias higher frequency interference down into the baseband, but it had some tricky stuff in there to reduce aliasing compared to a conventional design. .....This design may have been less sensitive than most PI's to its own electrostatic transmit field because whereas most PI designs use flyback systems in the range of 50 to 500 volts, my PI's used no high voltages, everything ran between the power supply rails which were typically single-ended 6 volts.
One PI prototype had near-perfect ground balance in Jim's yard, but no place else. The reason was that there was a defect in the searchcoil shielding, and the negative pickup from lack of shielding in his irrigated yard just happened to balance out the maghemite pickup from the iron minerals in his soil. Took me a long time to figure out why it worked so good in the boss's yard but not so good in anyone else's yard!
--Dave J.
Most metal detectors have pretty good rejection of electrostatic interference and of low frequency (below the fundamental transmit frequency) interference whether electrostatic or magnetically induced. However some PI designs are markedly inferior from the standpoint of low frequency interference rejection. This may have been factor in your experience. Also, in a home-brew PI it is tempting to not bother with electrostatic shielding of the searchcoil since you can generally get the thing to work without shielding (which is not to say it'll work as well as it could if it had shielding). I don't know how much of this applies to your particular situation.
My favorite PI was the "CodFisher", built as a joke for Jim Lewellen, then president of Fisher. Still have it, but the circuit died a few years ago and I never fixed it. Searchcoil wound on a wicker basket tied to a crooked tree limb, and the electronics in a wooden dried codfish box. Jim didn't know I was building it: I tricked him into buying it at the local fish market. No shielding of anything, ran almost forever on a rectangular 9 volt battery. It worked pretty good by mid 1980's standards and eventually morphed into the Impulse (no longer manufactured). Main thing I noticed with the lack of shielding was increased ground pickup in wet grass or on the wet part of a beach. No particular problems were observed with 60 Hz fundamental because this design had good rejection of low frequency interference. Of course like any PI it would alias higher frequency interference down into the baseband, but it had some tricky stuff in there to reduce aliasing compared to a conventional design. .....This design may have been less sensitive than most PI's to its own electrostatic transmit field because whereas most PI designs use flyback systems in the range of 50 to 500 volts, my PI's used no high voltages, everything ran between the power supply rails which were typically single-ended 6 volts.
One PI prototype had near-perfect ground balance in Jim's yard, but no place else. The reason was that there was a defect in the searchcoil shielding, and the negative pickup from lack of shielding in his irrigated yard just happened to balance out the maghemite pickup from the iron minerals in his soil. Took me a long time to figure out why it worked so good in the boss's yard but not so good in anyone else's yard!
--Dave J.
Hi Dave .... well you would have probably seen quite a few things over 28 years in the detector business ... gotta laugh at that codfish box .... I have been around the electronics side of things for about the same time but not with detectors .. well not since building a couple of VLF units about 25 years ago. A guy at work said he was interested in getting a professional detector but after seeing the prices he asked me what would be involved in building one and I ( stupidly ) said ... 'how hard can it be ?' So here I am .... It was not my intention to try and develop some new type of circuit .. my background is in professional RF , video / audio etc and balanced systems are the norm there .. so I will most probably learn the hard way and recommend that my friend just bite the bullet and buy one. 
Regards,
Paul ( moodz )
Regards,
Paul ( moodz )
moodz, the schematic doesn't show a differential amplifier, it shows a composite amplifier in which the top channel has a gain of -1 and the bottom channel has a gain of +2. The way to make it differential is to configure the top op amp as a noninverting amp with a gain of +2, not as a voltage follower. Then the composite amplifier arrangement comprises a differential amplifier with an absolute gain of 2 for both inputs.
Note that the top channel has a little bit more delay than does the bottom channel, since it's going through 2 op amps. So sharp spikes may not balance out well.
--Dave J.
Note that the top channel has a little bit more delay than does the bottom channel, since it's going through 2 op amps. So sharp spikes may not balance out well.
--Dave J.
Dave J. said:
Hi Dave .... you are right ... the amp bit is just a muck around I did not clean up before posting. The waveforms are all pre / post the fet switches. I should have knocked the amp off the diagram. I am actually using a THAT1512 differential amp ... but dont have a spice model for it.
Regards,
moodz.