Metal Detector question.

[Larry Snyder]

I'm getting into metal detectors again and was wondering if I missed
something. Here's what I have so far:
Dual coil transmit/receive wave around thing.
Single coil variable reluctance wave around thing.
Blast the dirt with microwaves and look for heat with infrared imaging.
Blast the dirt with a magnetic pulse and listen for sound and generate 3d
image.
I have about another month before the weather permits scrounging. Does
anyone have a scheme for metal detecting that I missed? Thanks in advance
Larry

 

[Robert Baer]

I'm getting into metal detectors again and was wondering if I missed
something. Here's what I have so far:
Dual coil transmit/receive wave around thing.
Single coil variable reluctance wave around thing.
Blast the dirt with microwaves and look for heat with infrared imaging.
Blast the dirt with a magnetic pulse and listen for sound and generate 3d
image.
I have about another month before the weather permits scrounging. Does
anyone have a scheme for metal detecting that I missed? Thanks in advance
Larry

"Blast" the dirt with EM wave(s) and/or pulses and do *NOT* look for
non-existant sound; look for phase changes or pulse reflections.

 

[neon]

blast with microwaves BLAST WITH MAGNETIC. You have no idea how it works do you ? it is a simple oscillator whereby the metal proximimity changes the freq. and you can detect that.

 

[Robert Baer]

Thanks for input. Essentially a ground search radar. I've avoided high
frequency stuff because my scope only goes to 500mhz. Also the ground is
not very transparent to high frequencies. There may be a way to do it any
way. As a thought, send a ramped fm signal. The return would be frequency
shifted from the transmit frequency by the distance. Mixing with the
transmit
would yield a signal that was a fixed frequency. Adjustable notch filters
could seperate out multiple targets. Seperate antennas could be used to
triangulate. What do you think? By the way, the sound thing already works.
Larry

FM radar is nice in that one can choose the distance with a simple
resonant-type filter.
Seems that almost nobody knows how they work...they are all in the
time domain paradigm.

 

[RFI-EMI-GUY]

Thanks for input. Essentially a ground search radar. I've avoided high
frequency stuff because my scope only goes to 500mhz. Also the ground is
not very transparent to high frequencies. There may be a way to do it any
way. As a thought, send a ramped fm signal. The return would be frequency
shifted from the transmit frequency by the distance. Mixing with the
transmit
would yield a signal that was a fixed frequency. Adjustable notch filters
could seperate out multiple targets. Seperate antennas could be used to
triangulate. What do you think? By the way, the sound thing already works.
Larry

Are you getting tin cans jumping out of the ground?

What about a circuit that discharges a capacitor into a hi Q coil with
an o'scope monitoring the coil for the primary pulse and any subsequent
ringing or reflection? If in vicinity of conducting object could you not
determine the relative size and distance to it?

--
Joe Leikhim K4SAT
"The RFI-EMI-GUY"©

"Treason doth never prosper: what's the reason?
For if it prosper, none dare call it treason."

"Follow The Money" ;-P

 

[Larry Snyder]

Are you getting tin cans jumping out of the ground?

What about a circuit that discharges a capacitor into a hi Q coil with an
o'scope monitoring the coil for the primary pulse and any subsequent
ringing or reflection? If in vicinity of conducting object could you not
determine the relative size and distance to it?

--
Joe Leikhim K4SAT
"The RFI-EMI-GUY"©

"Treason doth never prosper: what's the reason?
For if it prosper, none dare call it treason."

"Follow The Money" ;-P

Hi Joe. That's about what I'm doing. Here's schematics(not updated lately)
http://www.pacificsites.com/~snyder/Remotes.bmp
http://www.pacificsites.com/~snyder/a d converter schematic2.bmp
The idea is to put a 100kw pulse into the ground. Magnetic materials will
be pulled toward the coil and electrically conductive materials will be
repulsed(eddy currents). The effect is like tapping it with a hammer. The
sounds are digitized in a 4 channel a/d converter and put in a memory. The
waveform looks like a step with ringing after it for each thing. This is
dumped into the laptop and processed. I'm using qbasic(qb7) for software.
The process is stripping the step out and dumping the ring. Put the steps
into a seperate array and match waveforms. Once matched, they are located
in two arrays organized as right and left for a 3d display and subtracted
from the original array. The left and right arrays set red and blue
positions on the display. The glasses I have aren't an exact match for the
laptop screen colors. Waiting for red and green glasses to try.
Anyway, I would be interested in any thoughts you may have about the ground
search radar scheme.
Take care of yourself
Larry

 

[Robert Baer]

Are you getting tin cans jumping out of the ground?

What about a circuit that discharges a capacitor into a hi Q coil with
an o'scope monitoring the coil for the primary pulse and any subsequent
ringing or reflection? If in vicinity of conducting object could you not
determine the relative size and distance to it?

The problem with RADAR is that one has a "dead time" equal to the
pulse width plus the TR/ATR/receiver recovery time.
At distamces greater than that, it is OK and useable.

 

[Robert Baer]

Hi Joe. That's about what I'm doing. Here's schematics(not updated lately)
http://www.pacificsites.com/~snyder/Remotes.bmp
http://www.pacificsites.com/~snyder/a d converter schematic2.bmp
The idea is to put a 100kw pulse into the ground. Magnetic materials will
be pulled toward the coil and electrically conductive materials will be
repulsed(eddy currents). The effect is like tapping it with a hammer. The
sounds are digitized in a 4 channel a/d converter and put in a memory. The
waveform looks like a step with ringing after it for each thing. This is
dumped into the laptop and processed. I'm using qbasic(qb7) for software.
The process is stripping the step out and dumping the ring. Put the steps
into a seperate array and match waveforms. Once matched, they are located
in two arrays organized as right and left for a 3d display and subtracted
from the original array. The left and right arrays set red and blue
positions on the display. The glasses I have aren't an exact match for the
laptop screen colors. Waiting for red and green glasses to try.
Anyway, I would be interested in any thoughts you may have about the ground
search radar scheme.
Take care of yourself
Larry

I have the most recent compiler M$ ever sold for BASIC; they called
it Professional Development System 7.0 (upgraded to 7.1 if i remember
correctly).
Have all of the docs, etc.
Interested?

 

[RFI-EMI-GUY]

The problem with RADAR is that one has a "dead time" equal to the
pulse width plus the TR/ATR/receiver recovery time.
At distamces greater than that, it is OK and useable.

TRue but its primarily a dynamic range issue that might be overcome in
practice.

--
Joe Leikhim K4SAT
"The RFI-EMI-GUY"©

"Treason doth never prosper: what's the reason?
For if it prosper, none dare call it treason."

"Follow The Money" ;-P

 

[Larry Snyder]

I have the most recent compiler M$ ever sold for BASIC; they called it
Professional Development System 7.0 (upgraded to 7.1 if i remember
correctly).
Have all of the docs, etc.
Interested?

Thanks for the offer but qbasic is comfortable for me. There is a speed
problem sometimes but machine code fixes that.
Take care of yourself
Larry

 

[Vladimir Vassilevsky]

RFI-EMI-GUY wrote:

TRue but its primarily a dynamic range issue that might be overcome in
practice.

The typical attenuation in the soil at radar frequencies is at the order
of 200dB/m. Thus it is not feasible to pull out the far signal from the
near reflections, no matter what the technology is. Ground penetrating
radar doesn't work unless in the ideal conditions (like dry sand) and
very shallow depth.

Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

 

[RFI-EMI-GUY]

RFI-EMI-GUY wrote:



The typical attenuation in the soil at radar frequencies is at the order
of 200dB/m. Thus it is not feasible to pull out the far signal from the
near reflections, no matter what the technology is. Ground penetrating
radar doesn't work unless in the ideal conditions (like dry sand) and
very shallow depth.

Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

The original poster is designing at well below microwave frequencies.

--
Joe Leikhim K4SAT
"The RFI-EMI-GUY"©

"Treason doth never prosper: what's the reason?
For if it prosper, none dare call it treason."

"Follow The Money" ;-P

 

[JosephKK]

FM radar is nice in that one can choose the distance with a simple
resonant-type filter.
Seems that almost nobody knows how they work...they are all in the
time domain paradigm.

I do not find them so difficult. you can even detect multiple targets
simultaneously. The first range vs gain compensation is a
differentiator just after the detector.

 

[Robert Baer]

TRue but its primarily a dynamic range issue that might be overcome in
practice.

Well, practice (that is to say, real radars) have not solved the
problems, just (partly) decreased the times involved.

 

[Robert Baer]

Thanks for the offer but qbasic is comfortable for me. There is a speed
problem sometimes but machine code fixes that.
Take care of yourself
Larry

Actually, if you are careful and nasty in your coding, you can
achieve almost computer speed, as well as "optimised" ASM code after
compiling.
Use integers = no calls to code routines; use only single loops where
ever possible, as the code for a second inner loop is not optimum (but
could be hand-corrected); use COBOL fall-thru logic; use CALLs to any
routines (get rid of spaghetti logic); use tables (internal or external)
for complex logic.