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

[Aziz]

More Sample Pics..

Hi all,

this is the result of converting green rust into maghemite.


#2: Green Rust (not aged)
#3: Maghemite converted from sample #2
Sample #3 is most magnetic susceptible sample of my maghemites.



More details.



#1: First maghemite sample (out of aged green rust, notice: a different color)
#4: Heated green rust above 300 °C -> hematite conversion from unstable maghemite occurs.
This sample contains still maghemite and if I would heat it further, it would get more red and dark (more hematite). It has happened due to uneven heat application in a pot and cooker. The bottom of the pot had a much higher temperature (>300 °C).
I have to reconvert this low susceptible sample using charcoal, some moisture and more heat back into maghemite.



Overview of the samples.
#5: Some real Australian hotrocks.
These hotrocks contain a lot of magnetite however.



More details.

End.

Cheers,
Aziz

[WM6]

But this sample contains a much higher purity of maghemite and does not contain magnetite and other minerals.

Great work Aziz, congratulations.

Why do you think it is important that your sample do not contain magnetite?

[Aziz]

Great work Aziz, congratulations.

Why do you think it is important that your sample do not contain magnetite?

Hi WM6,

this is owed to the process-related "rusting/oxidation" path following the Fe(2+) path (wet and alkaline rusting). I don't get (relevant) magnetite in this process. But I can mix all the samples up with different magnetic susceptible samples including magnetite and other minerals.

I also followed the Fe(3+) rusting path. This is the (almost) dry rusting process. But I don't get much maghemite with this process however. A more wet rusting process however delivers Fe(2+) and Fe(3+) mixtures of oxides/oxyhydroxides, which can deliver quite a lot maghemite.

It took a long time to get this "green rust" as this process takes a long time (it's an under water corrosion).

A much cheaper and faster process is the hematite (alpha-Fe2O3) into maghemite (gamma-Fe2O3) conversion. Hematite can be bought very cheap (on ebay). All you need is charcoal powder, a barbeque grill, some moisture in the mix (water) and a lot of heat to get the exotherm reaction starting (around 400 - 600 °C).
You even get a lot hotter magnetic susceptible matter (maghemite) with this process....

BTW, this process is happening in the nature in the few 10'th cm deep in the ground, where the hematite with organic matter (carbon) is converted by a (bush-)fire into maghemite. The oxygene depleted reaction, which generates CO (very aggressive) is supporting this process in the ground. The top ground level is full of organic matter of course (soil).

Cheers,
Aziz

[Sean_Goddard]

AzizLike THIS one??

http://www.cirrus.com/en/products/cs4272.html

AND I have an eval board, unused which I bought for a multi frequency evaluation.

If you "Ping-Pong" the inputs (sample left, sample right) with input connected in "mono" mode, you get a sample rate of 384KHz with 100dB s/n ratio. Add a variable gain amplifier of 3,6,9,12dB gain and your dynamic range extends to earth noise floor (around -123dB).

[Sean_Goddard]

I know, only I'm not sure whether it is a function of software or a soundcard itself.

I used this software to learn about my soundcard performance regarding aliasing etc. I introduced a sweep from 20Hz to 20kHz and a 3D (kind of waterfall) spectrum display with linear frequency. Well, after what I've seen I am in pursuit of a better soundcard. Xonar or something.

(artifacts are seen as a mesh of lines going everywhere)

The swing left and right are a result of "sample skew". This (as far as I am aware) occurs when your sample rate is NOT a multiple of your incoming SAMPLED signal so you get "sibilants" etc. Also, you might want to BW limit the input using a low pass filter ( if the soundcard isn't already doing this). Remember that filters play havoc with current / voltage phases

[Aziz]

AzizLike THIS one??

http://www.cirrus.com/en/products/cs4272.html

AND I have an eval board, unused which I bought for a multi frequency evaluation.

If you "Ping-Pong" the inputs (sample left, sample right) with input connected in "mono" mode, you get a sample rate of 384KHz with 100dB s/n ratio. Add a variable gain amplifier of 3,6,9,12dB gain and your dynamic range extends to earth noise floor (around -123dB).

Hi Sean,

a higher sampling rate (more bandwidth) and SNR would improve the signal response of course. But the poor guy has a simple sound card, a cheap Tablet PC/Netbook and a simple detector controller. All parts must be easily and cheap available to the poor guy.
But the principle remains the same (either low bandwidth or high bandwidth).

But the poor guy can use an advanced demodulator/modulator concept (advanced detector controller), which makes a high bandwidth processing with cheap sound cards possible (@48 kHz). And this knocks out every "overpriced" detector in the world.


Cheers,
Aziz

[Davor]

The swing left and right are a result of "sample skew". This (as far as I am aware) occurs when your sample rate is NOT a multiple of your incoming SAMPLED signal so you get "sibilants" etc. Also, you might want to BW limit the input using a low pass filter ( if the soundcard isn't already doing this). Remember that filters play havoc with current / voltage phases

When supplied directly from the soundcard output to the soundcard input it is the soundcard by itself and it's quality that shows in such tests. My soundcard sucks. I'm not much into the reasons why it is that bad, as long as it is. It is a mainboard soundcard, and the only thing I can do is - buy a better one.
I'll go for a 192kHz sample rate simply to have a broader bandwidth for virtual oscilloscope application.

[Aziz]

When supplied directly from the soundcard output to the soundcard input it is the soundcard by itself and it's quality that shows in such tests. My soundcard sucks. I'm not much into the reasons why it is that bad, as long as it is. It is a mainboard soundcard, and the only thing I can do is - buy a better one.
I'll go for a 192kHz sample rate simply to have a broader bandwidth for virtual oscilloscope application.

BTW guys,

most "bad designed" sound cards limit the real bandwidth of the sound card by the internal low-pass filter even it is supporting a higher sampling rate. The low-pass filter corner frequency is around 20 kHz - 24 kHz (supporting only audio frequency range).

But the "Creative Sound Blaster X-FI Surround 5.1 Pro" (see link: http://www.amazon.de/dp/B0042RUEQW/r...SIN=B0042RUEQW )
is supporting 48 kHz signal input bandwidth at 96 kHz sampling rate. A direct signal demodulation is pretty fine with it.

Hey, but it isn't a problem using a slow 48 kHz SR sound card. We just need 2 - 4 (or more) signal samples from the fast PI demodulator and modulate it into the 48 kHz region. That is enough for processing.


Cheers,
Aziz

[Davor]

What you say is completely reasonable, but what I see as a problem that I'll need to solve are the "ghosts" in the spectrum. OK, you can treat them if they are related solely to your signal, but they are not. Upon close inspection of the "3D" graph, you'll see also artifacts related to PSU, and whatnot. In case you are into a demodulator only, you'd be able to fix all of the problems in a few easy steps. However, I expect my soundcard to provide me with superb sound, to work as an oscilloscope, and perhaps to demodulate a thing or two.
Thanks for the recommendation on the soundblaster. I had a "Live" card once, but it was a moderate disappointment.

[WM6]

Aziz, are there no tablet with build in sound-card direct usable (without need to additional external sound card)?

External sound card solution can be a little questionable composition regarding real terrain use.

[Aziz]

Aziz, are there no tablet with build in sound-card direct usable (without need to additional external sound card)?

External sound card solution can be a little questionable composition regarding real terrain use.

Beep, Beeeeep, Beeeeeeeeeeep sound?
Two sound cards necessary: One for signal acquisition and one for the signalling.
You don't want to look at the display every time.

Aziz

[WM6]

Beep, Beeeeep, Beeeeeeeeeeep sound?
Two sound cards necessary: One for signal acquisition and one for the signalling.
You don't want to look at the display every time.

Aziz

I was lost somewhere in this extreme long thread. So we need two SC.
Maybe something like this in combination with WIN-based tablet will suit well?
http://echoaudio.com/products/echo-2

[Aziz]

I was lost somewhere in this extreme long thread. So we need two SC.
Maybe something like this in combination with WIN-based tablet will suit well?
http://echoaudio.com/products/echo-2

Hi WM6,

you buy a Ferrari and you are not allowed to drive faster than 80 km/h!!!

Analog Input & Output:
Frequency response: 10Hz – 20kHz, +/- 0.1dB

Supports full duplex 2 channels in, 4 channels out operation at 24 bit, up to 192kHz sample rate

Up to does not specify exactly. Could be output only.

It doesn't make sense, if you can't detect frequencies above the 20 kHz region. The 192 kHz sample rate is an oxymoron to this nice feature. If the input/output filter limits your frequency response, you can use a low sample rate SC and save a lot of money.

The low noise specs are pretty good however:
Input: Dynamic range: 115dB A-weighted

The 20 kHz bandwidth (frequency response) limits the application however:
- Limitted direct sampling & decoding (well, it doesn't make sense for a PI application to use the low bandwidth SC; except for VLF applications of course)
- We can modulate a few sampled DC voltages from the PI demodulator into the 20 kHz audio frequency region. Then it would make sense to use this overpriced SC.

Bad designed SC's use different time bases for AD and DA section (two oscillators/X-TAL's for instance). Lock-in amplifier applications can't be implemented in this case. It is important, that the AD and DA section is synchronized to each other.

It is risky to trust into the specs. One have to try it out. The SC I have mentioned above (The Creative Sound Blaster X-FI ..) works up to 48 kHz bandwidth (not limitted to 20 kHz)!!!

And you have to switch off all effects!!! We don't need it. We need the plain unmodified stereo input/output channels.

Cheers,
Aziz


PS: This echo isn't convenient for this application!!!! We need a true SC!!!!!

[Aziz]

What you say is completely reasonable, but what I see as a problem that I'll need to solve are the "ghosts" in the spectrum. OK, you can treat them if they are related solely to your signal, but they are not. Upon close inspection of the "3D" graph, you'll see also artifacts related to PSU, and whatnot. In case you are into a demodulator only, you'd be able to fix all of the problems in a few easy steps. However, I expect my soundcard to provide me with superb sound, to work as an oscilloscope, and perhaps to demodulate a thing or two.
Thanks for the recommendation on the soundblaster. I had a "Live" card once, but it was a moderate disappointment.

Switch off all effects!!!!

[Aziz]

Does anybody know a good (and possibly a free) BLOG site (independent site)?
I'm thinking of to move over to start interesting new projects.
Aziz