Signal analysis VLF-Detektor

[Qiaozhi]

Actually, the received signal is parallel connected to a 1k resistor.So, that's the parallel signal from the resistor and the coil?!
And yes, I expected to see a chance in the amplitude?!?! But there is nothing as long as the metal is not in the "crossing section" of the coils...

The most sensitive part of the search head is where the coils cross over. This is where the target is detected. The voltage change will be very small. You should zoom into the RX waveform on the scope at the point where the signal has almost settled to 0V. Try looking at the 20us time point after the MOSFET is switched off.

Oh, before I forget: The shielding of the coils. I put both shielding wires together (as you can see on the picture) but if I connect the shield
with any GND, the signal is clear away....

Did you leave a break in the shielding? If you did not it will act as a shorted turn and kill the received signal. Also the shields on the two coils must be insulated from each other where they cross over.

[SIMONN]

Okay, but I can deinfitely put both shields together and connect it with one GND?

[SIMONN]

Okay... I got some new data again.

I already optimized the coil and enhanced the circuitry... But the reuslt isn't really good....

Here is some data:

Tx:
Frequency: 847.4 Hz
Period: 1.18ms
Mean: 2.56V
Pk-Pk: 5 V
Coil resistance: 22.3 Ohm
Inductance: 6.3 mH


Rx:
Frequency: 63 kHz
Period: 13 us
Mean: 20 mV
Pk-Pk: 1.2 V
Coil resistance: 24.4 Ohm
Inductance: 6.5 mH

Please see the attached pic. I don't really know why the received signal is oscillating... I also put a 100k resistor parallel to the rx-coil...

Best, Simon

Sorry, the smaller size of the pic doen't work...

[SIMONN]

and again the new signal

[Qiaozhi]

Okay, but I can deinfitely put both shields together and connect it with one GND?

Yes.

[Qiaozhi]

If you are trying to build a PI detector, which I assume is the case, your inductance values are far too high. Please read the Hammerhead article in the Projects section. Try using around 26 turns of wire to get an inductance about 500uH. The actual value is not really critical.

The ringing is happening because you do not have a damping resistor across the coil. Again, look at the HH article for details. 690 ohms is a good value to use as a starting point, and don't forget the limiting diodes.

I've reduced the size of your image to a more sensible size. Please use a graphics program such as GIMP to reduce the images to 800x600 before posting.

[mikebg]

Qiaozhi, I want to write formulas for the forum as you can in the example above.
I believe that other participants in the forum would like to write formulas in this way. Please explain me how you do this?
I can write formulas in this way, but they are in *. doc file, which can only be attached.

[holyuser]

I want to write formulas for the forum...

look here: http://www.artofproblemsolving.com/W...LaTeX:Commands

examples: http://elevatorlady.ca/doc/refcard/expressions.html

refefence card: http://elevatorlady.ca/data/tex-refcard-letter.pdf

[Qiaozhi]

Qiaozhi, I want to write formulas for the forum as you can in the example above.
I believe that other participants in the forum would like to write formulas in this way. Please explain me how you do this?
I can write formulas in this way, but they are in *. doc file, which can only be attached.

This forum supports Latex. I see that holyuser has already posted some links that can help you with the syntax.

[mikebg]

This forum supports Latex. I see that holyuser has already posted some links that can help you with the syntax.

TNX, Holyuser and Qiaozhi.
I will try to learn how to post formulas for MD designers.
Holyuser, I submit an idea - avatar for you. I think you have the original :-).

[holyuser]

Holyuser, I submit an idea - avatar for you. I think you have the original :-).

[mikebg]

To be honest, I don't understand how to use the signals. Therefore I need assistance, please. Does somebody have experience with data evaluation?Another question:
- Can I already use the signal (if phase difference is available) to discriminate between different metals?
Additionally, I'd like to digitize the whole process via DAQ.

SIMONN, the block diagram for narrow band metal detector in posting #23 is too primitive. It can not achieve deep detecting because not uses real GND signal for synchronous demodulation. The GND signal is artificial. It is produced from voltage across TX coil.
The attached block diagram is better but not the best.
Best principle for narrow band metal detecting is described in patent US 3,614,600.
The principle of wide band metal detecting with DSP is described in patent US 4,506,255.

[mikebg]

TEST FOR LATEX

Resistor for critical damping

where
is equivalent capacitance of circuit
is coil resistance.

SPECIFIC CASE: Self-damping coil
Formula for calculating resistance of self-damping coil: