Laptop Metal Detector utilising Digital Lock-in Amplifier

Is a powerful metal detector possible by using only laptop and sensor
(search-head) without any additional electronics?

See the generation and progress of this experimental project on:
http://www.thunting.com/geotech/forums/showthread.php?t=14102

The metal detector uses only the sound-card (output/input) for
interfacing the search-head.
;-)
Aziz

 

[John Larkin]

On Wed, 13 Aug 2008 17:29:33 -0700 (PDT), oeguet@gmx.de wrote:

Is a powerful metal detector possible by using only laptop and sensor
(search-head) without any additional electronics?

Powerful in what sense? As compared to a real metal detector, no.


John

 

[Michael Black]

On Thu, 14 Aug 2008, Bob Masta wrote:

On Wed, 13 Aug 2008 17:29:33 -0700 (PDT), oeguet@gmx.de wrote:

Is a powerful metal detector possible by using only laptop and sensor
(search-head) without any additional electronics?

See the generation and progress of this experimental project on:
http://www.thunting.com/geotech/forums/showthread.php?t=14102

The metal detector uses only the sound-card (output/input) for
interfacing the search-head.
;-)
Aziz

I haven't looked into metal detectors, but my Daqarta system has a
signal generator that can probably create any sort of driver signal
you want, plus real-time spectral (or waveform) analysis of the input
signal. It might be useful for development purposes, to try out
concepts before you devote a lot of time to writing your own code.

If you can explain the basic principles involved (or point me to a
Website), I'll be able to give you a better idea of whether Daqarta
can handle the task. (And if it can't do it now, it might be
something to add to the next version!)

That page was too much to follow, too little detail and too many

links.

Many metal detectors work by having an oscillator with a coil that
is open to the world. When metal comes close to the coil, that shifts
the oscillator's frequency and that shift is the giveaway that there
is metal.

That won't work by merely feeding the coil with a signal, the coil
has to be part of the signal generator.

I seem to recall there are things where the metal acts as coupling
between two coils, one excited by an oscillator and another a pickup,
and then the amplitude in the second coil varies depending on the
unknown metal. I suspect that is less workable than the first method,
though the excitation coil does not have to be a part of the oscillator.

From a memory of a Carl & Jerry story, I think the magnetometer uses
a standalone excitation, and not that much more than a milk bottle filled
with water and wound with wire. I can't remember what they used as a
pickup.

The problem is the question is based on some discussion somewhere, rather
than based on a foundation of what a metal detector is. If the original
poster had done some basic reading before asking here, he'd have a better
idea of what was required, and maybe an idea of whether it's possible.

Michael

 

Hi Bob,

my experimental measurement system has already a scope, spectrum
analyzer, digital lock-in amplifier, signal generator and much more.

So this experiment is a typical proof of concept which is saying: it
is working!

Basic operation of the laptop metal detector:
Harmonic sine wave is sent through the earphone output. The impedance
of the earphone output is round about 20 Ohms which delivers enough
current for the transmitter coil. The transmitter coil has a capacitor
connected defining a LC resonant tank. The transmitted frequency is
the resonant frequency of the LC tank.
The receive coil is in inductively balanced position (less coupled to
transmitter coil) and has also a capacitor which also defines a LC
resonant tank (same of transmitt frequency). The receive coil is
inducing a small signal. The signal amplitude and phase will change
upon a metal target nearby the search coil appears. The sound-card is
used in full-duplex mode (while transmitting a signal, the receive
signal is acquired).
The digital lock-in amplifier (a two channel I & Q lock-in amplifier)
detects the signal magnitude and phase of the receive coil. While the
laptop knows the reference frequency (internally generated), it
detects the changes by the receive coil.

This is the most simple and sentive metal detector ever designed. You
don't need any active electronics between the laptop and search coil.
Search coil has only some capacitors and the inductors (transmit &
receive coil). All the rest is done by the software using a high
definition sound-card operated at 96 kHz sampling rate and 24 bit
resolution.

The experiment is showing amazing sensitivity results. It can compete
with professional VLF detectors.
Aziz

On 14 Aug., 14:15, NoS...@daqarta.com (Bob Masta) wrote:

On Wed, 13 Aug 2008 17:29:33 -0700 (PDT), oeg...@gmx.de wrote:
Is a powerful metal detector possible by using only laptop and sensor
(search-head) without any additional electronics?

See the generation and progress of this experimental project on:
http://www.thunting.com/geotech/forums/showthread.php?t=14102

The metal detector uses only the sound-card (output/input) for
interfacing the search-head.
;-)
Aziz

I haven't looked into metal detectors, but my Daqarta system has a
signal generator that can probably create any sort of driver signal
you want, plus real-time spectral (or waveform) analysis of the input
signal.  It might be useful for development purposes, to try out
concepts before you devote a lot of time to writing your own code.

If you can explain the basic principles involved (or point me to a
Website), I'll be able to give you a better idea of whether Daqarta
can handle the task.  (And if it can't do it now, it might be
something to add to the next version!)

Best regards,

Bob Masta

              DAQARTA  v4.00
   Data AcQuisition And Real-Time Analysis
             www.daqarta.com
Scope, Spectrum, Spectrogram, Sound Level Meter
           FREE Signal Generator
        Science with your sound card!

 

Hi Bob,

indeed, this is really an interesting and fascinating project. The
digital lock-in amplifier is a pure software implementation which is
applied on the input signal (A/D converted receive signal). The lock-
in amplifier is a very sensitive phase detector even the signal is
buried in high noise. Any small changes can be detected with it (ľV
measuring). Using a 24 bit sound-card at 96 kHz sample rate increases
the signal-to-noise ratio and dynamic range. It will also work on 16
bit and 44.1/48 kHz with reduced SNR and dynamic range.

It is quite time-critical application. You must not loose the
synchronisation of the transmitted to received signal. The continious
wave form is buffered to the sound-card to avoid signal gaps due to
operating system task switches. But this is easy to handle and only
the DMA is busy and relieves the CPU. Laptop has enough CPU power for
doing this and much more in real-time (FFT, digital filter, lock-in
amp, detection, signal generation, synch generation, graphical
output ..).

The operating frequency for the sensor is between 5 to 24 kHz (VLF
range). It depends only on the resonant frequency of the search head
(L,C resonant defining elements) and sampling rate (fmax=SR/4). The
higher the operating frequency, the better the sensitivity of the
sensor (Faraday's law). So it is mostly defined by the sensor
specification.

The sensors are typically D shaped coils with same inductivity L for
transmit and receive coil. This will allow a simple matching of the
capacitors (same for transmitter and receiver). The coils are in
overlapped co-planar position and forming a circle (two D's). The
receive coil should have a minimum of signal level (10-50 mV rms).
This position must be found by moving one of the coils.


What about digital lock-in amplifier for your application? This would
be a quite useful feature.

Regards,
Aziz

Aziz:

Thanks for the explanation.  This sounds like a very intriguing
project!

There are a couple of details I am still not sure about.  You mention
using a digital lock-in amp, and it sounds like you mean an external
hardware device when you say "two channel I and Q lock-in amplifier".
But then you mention running the sound card in full-duplex mode, which
sounds like you are doing the operations in software... which would
certainly be the way to do it.  Is that the case?

Does the sound card really need 24 bit, 96 kHz?  What frequency do you
send out to the transmitter coil?  

Finally, how difficult is it to tune the two coils?  As I see it, it
would be simple enough to automatically tune the sound card frequency
to match the transmit coil, using any capacitor that was "close
enough",  by watching the amplitude of the received signal.  But then
you'd have to manually tune the receive capacitor.  True?

Best regards,

Bob Masta

              DAQARTA  v4.00
   Data AcQuisition And Real-Time Analysis
             www.daqarta.com
Scope, Spectrum, Spectrogram, Sound Level Meter
           FREE Signal Generator
        Science with your sound card!- Zitierten Text ausblenden -

- Zitierten Text anzeigen -

 

[Don Bowey]

On 8/17/08 5:47 AM, in article 48a81818.1430283@news.sysmatrix.net, "Bob
Masta" <NoSpam@daqarta.com> wrote:

On Sat, 16 Aug 2008 11:06:52 -0700 (PDT), oeguet@gmx.de wrote:

Hi Bob,

indeed, this is really an interesting and fascinating project. The
digital lock-in amplifier is a pure software implementation which is
applied on the input signal (A/D converted receive signal). The lock-
in amplifier is a very sensitive phase detector even the signal is
buried in high noise. Any small changes can be detected with it (=B5V
measuring). Using a 24 bit sound-card at 96 kHz sample rate increases
the signal-to-noise ratio and dynamic range. It will also work on 16
bit and 44.1/48 kHz with reduced SNR and dynamic range.

It is quite time-critical application. You must not loose the
synchronisation of the transmitted to received signal. The continious
wave form is buffered to the sound-card to avoid signal gaps due to
operating system task switches. But this is easy to handle and only
the DMA is busy and relieves the CPU. Laptop has enough CPU power for
doing this and much more in real-time (FFT, digital filter, lock-in
amp, detection, signal generation, synch generation, graphical
output ..).

The operating frequency for the sensor is between 5 to 24 kHz (VLF
range). It depends only on the resonant frequency of the search head
(L,C resonant defining elements) and sampling rate (fmax=3DSR/4). The
higher the operating frequency, the better the sensitivity of the
sensor (Faraday's law). So it is mostly defined by the sensor
specification.

The sensors are typically D shaped coils with same inductivity L for
transmit and receive coil. This will allow a simple matching of the
capacitors (same for transmitter and receiver). The coils are in
overlapped co-planar position and forming a circle (two D's). The
receive coil should have a minimum of signal level (10-50 mV rms).
This position must be found by moving one of the coils.


What about digital lock-in amplifier for your application? This would
be a quite useful feature.

Regards,
Aziz


Aziz:

Thanks for the explanation. Yes, I am quite familiar with real-time
issues. Daqarta needs perfect sync to do synchronous waveform
averaging for noise reduction, so I've been there and done that!

A digital lock-in would be a definite possibility for Daqarta. I'll
put that on my "Wish List" for future enhancements. I probably won't
offer the "lock-in" (PLL) part that hardware lock-ins have, since I've
always thought that was pretty silly unless you really do need to sync
to an external signal.

For those who are following this thread and aren't familiar with
lock-in amplifiers, they are essentially a single-frequency Fourier
Transform. You separately multiply the incoming signal by the sine
and cosine of the reference signal, and low-pass filter the results.
From the old high-school formula for the product of sinusoids, you get
only terms at sum and difference frequencies. It's the difference
term we want here. The low-pass removes the sum and produces an
output only if the input is exactly the same frequency as the
reference (difference = 0), or very near.

(See <www.daqarta.com/eex01.htm> for an FFT explanation that
goes into more detail about this.)

The only real difference between a lock-in and the output of an FFT
is that the FFT has a very crude low-pass filter (one for each
spectral line) and the lock-in usually has a better filter (longer
time constant in lock-in terms). That statement assumes that the
FFT has a spectral line just where you wanted the lock-in reference
frequency. This is no big deal if you are generating the output
frequency yourself... just make sure it lands squarely on a spectral
line.

Daqarta can already do this. And it can get the noise reduction by
synchronous waveform averaging before the FFT. The only thing
is that it doesn't display the data in lock-in format, with separate
sine and cosine or magnitude and phase readouts. That would be a good
addition!

Best regards,


Bob Masta

DAQARTA v4.00
Data AcQuisition And Real-Time Analysis
www.daqarta.com
Scope, Spectrum, Spectrogram, Sound Level Meter
FREE Signal Generator
Science with your sound card!

Great discussion, excellent commercial. Thanks guys.

 

[YD]

Late at night, by candle light, oeguet@gmx.de penned this immortal
opus:

Hi Bob,

my experimental measurement system has already a scope, spectrum
analyzer, digital lock-in amplifier, signal generator and much more.

So this experiment is a typical proof of concept which is saying: it
is working!

Basic operation of the laptop metal detector:
Harmonic sine wave is sent through the earphone output. The impedance
of the earphone output is round about 20 Ohms which delivers enough
current for the transmitter coil. The transmitter coil has a capacitor
connected defining a LC resonant tank. The transmitted frequency is
the resonant frequency of the LC tank.
The receive coil is in inductively balanced position (less coupled to
transmitter coil) and has also a capacitor which also defines a LC
resonant tank (same of transmitt frequency). The receive coil is
inducing a small signal. The signal amplitude and phase will change
upon a metal target nearby the search coil appears. The sound-card is
used in full-duplex mode (while transmitting a signal, the receive
signal is acquired).
The digital lock-in amplifier (a two channel I & Q lock-in amplifier)
detects the signal magnitude and phase of the receive coil. While the
laptop knows the reference frequency (internally generated), it
detects the changes by the receive coil.

This is the most simple and sentive metal detector ever designed. You
don't need any active electronics between the laptop and search coil.
Search coil has only some capacitors and the inductors (transmit &
receive coil). All the rest is done by the software using a high
definition sound-card operated at 96 kHz sampling rate and 24 bit
resolution.

The experiment is showing amazing sensitivity results. It can compete
with professional VLF detectors.
Aziz

Why lug along a lap-top if the needed circuitry fits in a box the size
of a cigarette pack? All you need is a LED display and possibly a
beeper.

- YD.
--
Remove HAT if replying by mail.

 

[Jasen Betts]

On 2008-08-17, Bob Masta <NoSpam@daqarta.com> wrote:

On Sat, 16 Aug 2008 11:06:52 -0700 (PDT), oeguet@gmx.de wrote:

Thanks for the explanation. Yes, I am quite familiar with real-time
issues. Daqarta needs perfect sync to do synchronous waveform
averaging for noise reduction, so I've been there and done that!

A digital lock-in would be a definite possibility for Daqarta. I'll
put that on my "Wish List" for future enhancements. I probably won't
offer the "lock-in" (PLL) part that hardware lock-ins have, since I've
always thought that was pretty silly unless you really do need to sync
to an external signal.

For those who are following this thread and aren't familiar with
lock-in amplifiers, they are essentially a single-frequency Fourier
Transform. You separately multiply the incoming signal by the sine
and cosine of the reference signal, and low-pass filter the results.
From the old high-school formula for the product of sinusoids, you get
only terms at sum and difference frequencies. It's the difference
term we want here. The low-pass removes the sum and produces an
output only if the input is exactly the same frequency as the
reference (difference = 0), or very near.

you make the lock-in amplifier sound very much like a synchronous
detector. in-fact if you're not implementing the PLL it seems more
like a synchronous detector than it is a lock-in amplifier.

Bye.
Jasen

 

[Jasen Betts]

On 2008-08-18, YD <ydtechHAT@techie.com> wrote:

Why lug along a lap-top if the needed circuitry fits in a box the size
of a cigarette pack?

well. the size of a mobile phone anyway

All you need is a LED display and possibly a
beeper.

you'd probably want to be able to tune it to discriminate between
scrap iron and lost coins.

Bye.
Jasen

 

On 18 Aug., 05:46, YD <ydtech...@techie.com> wrote:

Late at night, by candle light, oeg...@gmx.de penned this immortal
opus:

Why lug along a lap-top if the needed circuitry fits in a box the size
of a cigarette pack? All you need is a LED display and possibly a
beeper.

- YD.
--
Remove HAT if replying by mail.- Zitierten Text ausblenden -

- Zitierten Text anzeigen -

Hi YD,

the development platform is a PC/laptop. But this project can be
ported to a handheld Pocket PC.
The reason, why it can't be done to a size of cigarette box pack is
simple: the demand for high number crunching computing performance.
That's the reason, why this project becames very simple. All the
detector is done by pure software programming.

Aziz

 

[Michael Black]

On Mon, 18 Aug 2008, oeguet@gmx.de wrote

On 18 Aug., 05:46, YD <ydtech...@techie.com> wrote:
Late at night, by candle light, oeg...@gmx.de penned this immortal
opus:

Why lug along a lap-top if the needed circuitry fits in a box the size
of a cigarette pack? All you need is a LED display and possibly a
beeper.

- YD.
--
Remove HAT if replying by mail.- Zitierten Text ausblenden -

- Zitierten Text anzeigen -

Hi YD,

the development platform is a PC/laptop. But this project can be
ported to a handheld Pocket PC.
The reason, why it can't be done to a size of cigarette box pack is
simple: the demand for high number crunching computing performance.
That's the reason, why this project becames very simple. All the
detector is done by pure software programming.

But which comes first? Are you actually getting a significant

improvement by using a computer, or is it so you don't have to
wire up much circuitry?

In other words, unless the "high number crunching" is adding
soemthing to make this better than the average metal detector,
it's only purpose is to compensate for the lack of circuitry.

A software modem needs a lot of number crunching, but that
does not make it better than a hardware modem. Just cheaper,
though of course I can use my hardware modem with any computer
hardware, including something from 30 years ago.

Michael