Let's made a PC-base metal detector with usb interface !!!

[Aziz]

This is my laptop MD. Ok it's not operated with a laptop but a PC.

Aziz

[simonbaker]

TN1000.pdf .. TN1007.pdf from

http://www.signalrecovery.com/ApplicationsNotes.htm

Implementing Digital Lock-In Amplifiers Using the dsPIC DSC

http://www.microchip.com/Stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=1824&appnote=en532447

and use google:
http://www.google.de/search?hl=de&q=...amplifier&meta=

A lot of stuff to read. But it is very simple.

Now the competition race can be started!
Go!

Aziz

Excellent, thanks. I'm inspired!

[Aziz]

Aziz thanks for the info on small computer and your progress.

What is a good reference to read for someone who wants to program a lock-in algorithm like you did?

regards,

SB

TN1000.pdf .. TN1007.pdf from

http://www.signalrecovery.com/ApplicationsNotes.htm

Implementing Digital Lock-In Amplifiers Using the dsPIC DSC

http://www.microchip.com/Stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=1824&appnote=en532447

and use google:
http://www.google.de/search?hl=de&q=...amplifier&meta=

A lot of stuff to read. But it is very simple.

Now the competition race can be started!
Go!

Aziz

[simonbaker]

Aziz thanks for the info on small computer and your progress.

What is a good reference to read for someone who wants to program a lock-in algorithm like you did?

regards,

SB

[Aziz]

Here is the link for some photos of the Asus Eee PC 901:

http://www.blogeee.net/2008/05/14/exclu-eeepc-901/

Really very nice portable PC. At least for wardriving (WLAN), reading, writing and much more...

Aziz

[Aziz]

Hi friends,

I was today in a shop and found the new Asus Eee PC 901. It has the Intel Atom N270 processor and runs on Windows-XP. It is fully x86 compatible.
The small PC is leight weight: 1.1 kg and very compact.
It has the Realtek HD sound-card built-in! Earphone output and microphone input plugs are available! It can be operated up to 4.5 h (full operation) or up to 7 h (idle mode).

It cost's here only 379 EUR. I will try to get one and will check the usability for the laptop detector.
Now this is getting very interesting for a compact and very powerful VLF metal detector.

Aziz

[Aziz]

!!! Be careful !!!

PI detectors nearby the laptop detector will disturb the operation dramatically. Reason: wide band spectrum sent through the switching coil. This can not be filtered out when the noise is covered on operating frequency. All other noise sources (except on operating frequency) does not disturb the operation. Laptop detector is totally immune to such noises. Lock-in amplifiers have built-in filter with very narrow band filter characteristics.

The input-lines have to be protected against high voltage induction through the RX coil with two anti-parallel diodes (1N4148 ). Be carefull when you experiment on line-input or microphone input! Particularly nearby PI detectors!!!

Microphone input lines must be decoupled with capacitors (they provide operating voltage to microphones).


Aziz

[Aziz]

New detection depth record (air-test):

1 Euro coin at 54 cm.

Experiment:
With an external amplifier driven in single channel (TX). RX not amplified. I had too much drifting and when this is tamed, there are some more detection depth centimeters.
There is easy room up to 60 cm or more distance for 1 Euro coin.

I will implement drift compensation functions in the coming days and will do a new measurement soon.

Aziz

PS: The battery voltage for the external power amplifier is not stabilized. This will cause drifting.

[Aziz]

Here is a picture of the frequency spectrum and time-domain signal on my HD sound-card. It is not compareable with standard 16-Bit sound-cards. No jittering and timing problems on the signal. This is the reason for my breaking the limits.

Aziz

[Aziz]

Hi Eugene,

Aziz , your results are amazing !!! When you get some time can you measure some parameters of your TX coil , like total current drawn across the TX coil and even the amplitude of your voltage across the coil ?? WOW , 1 euro coin at 50 c.m. in air is remarkable to me. Are you still using the "Troy-5" TX oscillator configuration with your sound card ???? Thank-You for any answers , when you have time of course !!!. I will be away from any computer , for 3 days anyway.............Take Care..........Eugene

No, I am not using either Troy-5 TX oscillator nor any other TX oscillator. I am using the differential TX feeder shown in the previous posts.

The search coil consists only of:
- TX coil + stereo TX feeder (2 x 10 Ohms current limiting resistor + 3 capacitors)
- RX coil + resonant capacitor

That's all what I am using at the moment. No external electronics. External oscillators are not stable enough for lock-in amplifiers.

TX and RX coils are made on a 20 cm diameter pot and bend to D form type shapes. TX coil has 2*30 windings (center tapped coil, 3 Ohms total) and RX coil has 140 windings (~60 Ohms). The current through the sound-card output is probably 2 x 20-30 mA (roughly). I can not measure the current through the output due to high operating frequency (sorry, I have no oscilloscope yet). My multimeter can only measure currents up to 1 kHz. This also counts for voltages. Maybe I should measure this with my PC scope one day.
The output voltage is: 2 x 1.4 V rms (differential stereo mode, volume set to maximum).
Output impedance: ~20 Ohms (earphone output)
Input voltage range: 110 mV rms (set to maximum sensitivity/amplification).
Input signal is fed into line-input (10 kOhm impedance).
RX signal is dependend on the balanced position and can be between 5 mV to 90 mV rms. So this detector will work also with very bad balanced coils. A minimum RX signal level is necessary (let's say 1 mV rms).

The additional VCO output (sound output on the second sound-card) made it possible to measure the detection depth more accurately. Even if there is some drift in the signal, the VCO output is clear to detect than the displayed numerical output.

I will make some experiments with external amplifiers next week. I am sure, there is more potential to get out.

Detection depth of 50 x 50 x 5 mm AL block: 70-80 cm.

Aziz

[amtech2005]

Let's make a PC-based metal detector with usb interface !!!

Aziz , your results are amazing !!! When you get some time can you measure some parameters of your TX coil , like total current drawn across the TX coil and even the amplitude of your voltage across the coil ?? WOW , 1 euro coin at 50 c.m. in air is remarkable to me. Are you still using the "Troy-5" TX oscillator configuration with your sound card ???? Thank-You for any answers , when you have time of course !!!. I will be away from any computer , for 3 days anyway.............Take Care..........Eugene

[Aziz]

Yes, not practical, but would be interesting to see if lock-in could cut through noise given enough time, and how long it would take (assuming noise is a stationary process).

Hi Simon,

one could do metal detecting also with 16-Bit sound-cards. But not with all of sound-cards on the market. Some of them has enormous clocking problems which makes them not convenient for lock-in amplifiers.

Normally, the performance is degrated by:
- sampling rate (the lower, the worse)
- noise (sound-card)
- input voltage range (suffering from low signals),
- input impedance
- output voltage range (delivering not high voltage)
- output impedance (delivering not enough current)
- total harmonic distortion
- jitter (timing inaccuracy, phase noise)
- clock generation inaccuracy
- software driver bugs (driver not correctly implemented)

So increasing the time constants of the lock-in amplifier is not allways the best solution to minimize the noise. You need often a reasonable fast response time to targets. Why taking a worse sound-card when there is a better one available?

Therefore I will use two different sound-cards for the future. One HD sound-card for analog signal processing and another one for detecting audio signalling (normally built-in in a laptop). Then I can drive the TX coil in differential mode for double power and can take stereo input signal for better signal conditioning.

There is still room for detection depth improvements:
- amplifiying the TX signal (reducing the sound-card noise, boosting TX power with low noise amplifiers)
- amplifiying the RX signal (reducing the sound-card noise with low noise amplifiers)

I will try again the boosting of the TX power first.

Aziz

[Aziz]

All metal mode (air-test):

0.33 ml coca-cola can (the old one - not the new slim line one) at:
>100 cm!

It's not bad. Or is it?
Aziz

[Aziz]

Incredible new detection distance record (air-test):

I built a very simple VCO on the second and cheap sound-card (mentioned above). The analog signal processing runs on the HD sound-card. I set the sensitivity level to max (one can hear the noise and instability almost).

1 Euro coin at 50 cm!

No, you wouldn't believe me! But there is still more to squeeze out.

Aziz

[simonbaker]

I have measured different things:

- jitter (total ugly)
- total harmonic distortion (total ugly)
- impedance (ok, 32 Ohms specified and measured much lower for earphone output, 1.1 Vrms output)

This codec-chip (CMI 8738/PCI-SX) is not recommend for such sensitive lock-in amplifier measurements. I have observed clock timing anomalies at 44.1 kHz sampling rate. In this mode, the lock-in amplifier can not be used due to low constant phase shift drift. This might be a chip-related clock generation problem or software driver. On 48 kHz SR, the lock-in amplifier is stable.

Line input is not much sensitive (1.1 Vrms) and the input signal needs to be amplified for more sensitivity (not acceptible).
Microphone input is mono (50..200 mVrms) but can be boosted 20 dB additional. I got more detection distance with mic-input.

This codec-chip is only for your ears - not for your coils. For a professional laptop MD one need really a true HD (high definition) 24 Bit/96 kHz or better sound-card.

Now I will be focusing on software coding (state of the art VCO). I have now a second sound card for audio detecting signalling. And no more plugging again and again TV, MD and so on..

Aziz

PS:
Hi Simonbaker,

I am measuring static signals on the RX coil at the moment (non-motion). When I increase too much the time constants, this will reduce the response time (reducing depth on faster moves). I have not changed the time constants.

Yes, not practical, but would be interesting to see if lock-in could cut through noise given enough time, and how long it would take (assuming noise is a stationary process).