Metal Detector question.

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

Reply to
Larry Snyder
Loading thread data ...

"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.

Reply to
Robert Baer

beat frequency metal detector two oscillators one fixed and one variable connected to the coil the last one I made used two tunnel doide oscillators. and some digital logic as a mixer. the main problem I had with that one was that when the frequencies became very close the oscillators would injection lock to each other.

Bob

Reply to
sycochkn

Hi Bob. Thanks for your response. There is a way doing it with a crystal that allows for frequency multiplication. The coil and oscillator must be very well isolated from the rest of the circuit. I put the oscillator down on the search coil. By mixing with an exclusive or gate and then filtering and amplifying the result, you can get multiple harmonic frequencies. This in turn can be mixed again with another similar circuit to get frequency multiplication. 1000:1 is very reasonable. Oscillator stability is the limiting factor. Larry

Reply to
Larry Snyder

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

Reply to
Larry Snyder

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.

Reply to
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?

--
Joe Leikhim K4SAT
"The RFI-EMI-GUY"©
 Click to see the full signature
Reply to
RFI-EMI-GUY

Hi Joe. That's about what I'm doing. Here's schematics(not updated lately)

formatting link
formatting link
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

Reply to
Larry Snyder

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.

Reply to
Robert Baer

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?

Reply to
Robert Baer

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

--
Joe Leikhim K4SAT
"The RFI-EMI-GUY"©
 Click to see the full signature
Reply to
RFI-EMI-GUY

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

Reply to
Larry Snyder

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

formatting link

Reply to
Vladimir Vassilevsky

The original poster is designing at well below microwave frequencies.

--
Joe Leikhim K4SAT
"The RFI-EMI-GUY"©
 Click to see the full signature
Reply to
RFI-EMI-GUY

GPR is best when you already know there is a buried object, but aren't sure of it's exact location. The human brain is an excellent tool for looking at the GPR readout and seeing patterns in the garbage.

Reply to
Don Bowey

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.

Reply to
JosephKK

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

Reply to
Robert Baer

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.

Reply to
Robert Baer

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.