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

[simonbaker]

Another nice circuit, thanks!

It will be interesting to see if anyone can achieve better info, especially discrimination, from two frequencies.

-SB

Hi all,

look at this dual frequency transmitter example. It is optimized for two frequencies only and achieves better efficiency. It can produce really high voltages (>100 V), if your voltage source can deliver enough power (small power amp, low source impedance).

The two frequencies and their spectrum power content can be dimensioned by the four parts: L1, Cp, Cs and LTX (search coil). Note: L1 is a center-tapped coil wound on a ferrite ring core. All filter parts must be high-Q parts (low losses).

This type of transmitter is optimally suited for a DSP starter kit with built-in codec. You need a power amp, a receiver front-end, this filter and a sophisticated DSP software.

You can also drive it with digital clock feed (containing the two fundamental frequencies) instead of sine waves. Should give more power to the coil.

Cheers,
Aziz

[Aziz]

Hi Simon,

thanks. Well, time will tell, whether it will work or not. I am thinking of wiring this prototype next year. With a discrete mosfet driver, the coil current goes up to +/-1.2 A and coil voltage is achieving 500 Vpp (simulation result).
I can even push more current, if I reduce the impedance of the mixer.

Indeed, this is a nice and interesting dual frequency transmitter. I wish, I would have had it much earlier.

Merry christmas and happy new year to all friends.
Cheers,
Aziz

[WM6]

Merry christmas and happy new year to all friends.

Same wishes to you dear Aziz for 2011.

A lot of success in your life and as little as possible krauts.

[Aziz]

Hi all,

I have seen, that the frequency spectral power of the two frequencies was not right balanced. I think, the frequency spectral power of f2 was approx. 10 dB lower.

Try the following changes:
Freq1 = 12.3905k
Freq2 = 32.909k
Cp1 = 100n
Cp2 = 100n
L1 = 600µ
LTX = 650µ
RTXs = 0.9

This pushes more power (energy) to the coil and kick-*** the target. The LTC1693-1 achieves a coil current of approx. +/-1.8 A. With a discrete mosfet driver, I get coil currents of up to +/-4 A (approx. 14 Watt power consumption, approx. 6 W at the TX coil, 2x3 W at the mixer coil, less than 1 W at L1). The ferrite ring cores of L1 and Mixer coil (LM) + the capacitors should also have some losses (getting warm or hot).
Keep cooling.
Aziz

[Aziz]

Same wishes to you dear Aziz for 2011.

A lot of success in your life and as little as possible krauts.

Thanks my friend,
Aziz

[Aziz]

More power to the TX coil!

Hi,

we usually have a high impedance TX coil (longer wire, more resistance). If we remove the TX coil and put there an inductive impedance transformer, we can kick the target much stronger. At the end, the voltage across the TX coil must go high. As much as possible. This is done with a simple transformer coil. It's characteristics in conjunction with the TX coil however must be optimized. Also the current in the ferrite ring core coils should not saturate the core material.

There are much more configurations possible now. I did not find the minimal and optimal solution for it yet.

Aziz

[WM6]

.

At the end, the voltage across the TX coil must go high. As much as possible.

Across the TX coil or across primary windings of proposed impedance transformer?

I do not like transformer in MD but this can be promising.

[Aziz]

Now,

it's arriving an interesting stage now. The dual frequency VLF can be made with just simple electronics. You don't need any microcontroller or computer/laptop for it. So we can detect at much higher frequencies too.

We just need:
- two stable clock generators (frequency f1, f2)
- transmit driver and network
- simple front-end receiver
- two independent full-wave demodulators (measuring the RMS of the receive signal, hence full-wave rectifiers)
- ground balancing stage (summing amplifier)
- audio output stage

We don't need any phase information of the received signals. It's enough to process the RMS values (cheap and easy).

Should we break the limits with high power VLF's?

Aziz

[WM6]

Now,

it's arriving an interesting stage now. The dual frequency VLF can be made with just simple electronics. You don't need any microcontroller or computer/laptop for it. So we can detect at much higher frequencies too.

We just need:
- two stable clock generators (frequency f1, f2)
- transmit driver and network
- simple front-end receiver
- two independent full-wave demodulators (measuring the RMS of the receive signal, hence full-wave rectifiers)
- ground balancing stage (summing amplifier)
- audio output stage

We don't need any phase information of the received signals. It's enough to process the RMS values (cheap and easy).

Should we break the limits with high power VLF's?

Aziz

Very interesting Aziz

What of data we can get out of measuring / comparing RMS of those two frequencies.
Can we discriminate in such way? Need some more explanation.

[Aziz]

Across the TX coil or across primary windings of proposed impedance transformer?

I do not like transformer in MD but this can be promising.

Hi WM6,

we have to pass some energy through the transformer. The primary coil current must be high and it's voltage low.
The secondary coil current will be low but it's voltage will be high. So the TX coil resistance can be high, which makes light weight coils possible.

Transformers are good to match the impedances.
Aziz

[Aziz]

Very interesting Aziz

What of data we can get out of measuring / comparing RMS of those two frequencies.
Can we discriminate in such way? Need some more explanation.

Yes, we can add one simple processing stage to isolate the reactive resonse. The relation of reactive to resistive response gives good discrimination information.

Aziz

[WM6]

The secondary coil current will be low but it's voltage will be high. So the TX coil resistance can be high, which makes light weight coils possible.

Can we use car ignition coil or gas ignition coil out from gas stove?

[Aziz]

I should meantioned, that the synchronous full-wave demodulator is a lock-in amplifier to detect the RMS value of a specific frequency. It's reference frequency is coming from the oscillators. We need two of them for the dual frequency VLF.

The implemantion of it is really trivial but very efficient.

Aziz

[Aziz]

Can we use car ignition coil or gas ignition coil out from gas stove?

No, they must be made for specific characteristics (inductivity, resistance). But it isn't difficult to wind some wire to a ferrite core.

Aziz

[Aziz]

Warning

Hi all,

I do not want to miss to give you a warning. With the high power VLF, you are getting in the region of harmful to one's health. With high coil voltages (>100V) and the low source impedance, please don't touch the hot connections carrying a high voltage.
These voltages are as dangerous as the power lines.

Fortunatelly, the coil power can easily be adjusted. The coupling capacitors to the mixer coil must be changed: low cap -> low output power, high cap -> high output power. This power regulation is very efficient (low power-> low battery drain). It is also possible to make a weighted power spectrum adjustments. Just increase or reduce the coupling capacitor of the specific frequency input pin.

Of course, a low/high power switch could easily be made too.
Take care,
Aziz