Separate Oscillator

[pebe]

Can anyone point me to a circuit that uses a driven TX output stage - rather than a self-oscillating one?

[Carl-NC]

Any PI circuit uses a driven loop. You can use a similar method in VLF by adding a parallel cap to improve efficiency and get a sine wave.

- Carl

[pebe]

Thanks. Can you tell me the duty cycle of the driver? is it 50-50?

[porkluvr]

Thanks. Can you tell me the duty cycle of the driver? is it 50-50?

You are now asking about IB? Yes, 50/50 drive signal is usual for an externally driven coil. Look at the Tesoro Lobo, and Fisher 1265, 1266, CZ5, CZ6, and Gold Bug detectors to see some different schemes. Use the "Search" button to find those schematics.

If there is a possibility of confusion you should specify whether asking about PI or IB detectors because they are two different animals.

[pebe]

Thanks, but a search only came up with a circuit for the Lobo.

Yes, It was an IB detector I was referring to - I am new to this hobby. I assumed that PI detectors used a pulsed TX field. I didn't know that 'self-oscillating' could apply to them also.

[mikebg]

TX for Sine induction

Can anyone point me to a circuit that uses a driven TX output stage - rather than a self-oscillating one?

Peter,
Instead transistor type IRF510, can be used IRF540, IRF640, IRF740, IRF840.
Resistance of R11 is chosen so as to obtain a duty cycle about 0,2. It depends on TX coil resistance.
Resistance of R6 is selected so as to obtain a minimal settling time without overshot at test point tp2.
Adjust R2 for minimal duty cycle at test point tp1.

[porkluvr]

sine induction?

Can you tell us more about the coil inductance and what range of frequencies this circuit works best at?

Is this perhaps something that would make a high output sinewave for a two-box?

This is very unusual and some description would be useful.

[pebe]

Mike,
I want to use a microcontroller to generate the Tx pulse, and control the amplitude of the Tx waveform by varying the pulse width.

I want to try a radically different design for ground elimination, to the circuits I have seen so far.

[mikebg]

Porkluvr,
Why not "Sine induction"? Since there is "Pulse induction", must have antonym of that term. The design of the TXs and RXs are generally carried out in frequency domain. From this perspective, the metal detectors are of two types: Narrow Band MD and Wide Band MD. This is "TX for narrow band metal detector".
The maximum operating frequency of this circuit diagram depends on the operational amplifiers used. I experimented it at 6500Hz, but I think with TL072 and TL082 will work almost to 18KHz. The greatest need for speedy action is required by the comparator U2A, which acts as a pulse width modulator.
If you want your TX to work at a higher frequency, use for U2A an integrated circuit designed for comparator instead of one that is designed for operational amplifier.
If the term TWO BOX understand orthogonal coil configuration, no sense of this amplitude stabilizer because TX coil moves relatively far from the ground so that the parasitic modulation is weak. The circuit is suitable for sensors in which the TX coil is moving closer to the ground. Soil and salty water act as moving core. More on the issue of unwanted modulation of the TX, you can read my postings in section "Modifications".
If you need an extremely powerfull TX for orthogonal coil configuration, I published in forum an idea.
The block diagram, which should be implemented for faultless operation of the TX for Sine induction, is the following:

[WM6]

If you need an extremely powerfull TX for orthogonal coil configuration, I published in forum an idea.

Not only TX but RX too. Complete schematic and no PIC please!. Thanks.

[technos]

Mike,
I want to use a microcontroller to generate the Tx pulse, and control the amplitude of the Tx waveform by varying the pulse width.

And then by definition you no longer have a VLF but a PI as Carl suggested. While some kind of "hybrid" design might be possible, it should not be forgotten that measuring the phase angle of a reflected signal becomes very hard with the demodulators (i.e. phase detectors) if the wave train changes radically from one pulse to the next. I'm not saying that it's not doable, but it might be useful to keep in mind the scope of your project before you try to design a Mark I like device.
However a VLF/TR type is a lot more straight forward, i.e. you would have a VLF/PI. But I question the usefulness of that.

[pebe]

Technos, I think it is a VLF/IB. Let me describe how I intend it to work.

The heart of the circuitry will be a Wien Bridge oscillator giving an almost pure sine wave at 14kHz whose amplitude will be stabilised. An output from that will be squared (amplified and converted to a square wave) and that will be fed to a PIC controller. Its leading edge (P1) will mark the zero crossover of the sine wave. After a delay (Td) the PIC will send a pulse to generate the sinewave TX pulse. The Rx pulse will be amplified and squared (P2) and sent to the PIC where its timing is compared to P1. If the two are not coincident, Td will be altered until they are. The Rx and Wien waveforms are then phase locked.

Another output from the Wien is taken through an ‘L’ attenuator comprising a resistor and a FET controlled by the PIC to give a variable resistance. Initially, the PIC sets the FET to its lowest resistance to give maximum attenuation.

The Rx sinewave and the 180degree phase shifted Wien sinewave are then added in an amp (A1). Its output is fed to an A/D converter in the PIC. The PIC will then modify the FET’s attenuation to alter the Wien output until the amplitudes of both waveforms are the same and there is no output from A1. That will be the state during searching. This action will be carried out every cycle so the status is continually updated.

The PIC will keep a short-term record of the maximum (Max) of the A1 output level. When the coils pass over a target there will be an output from A1 and there will be a faster-than-normal rise of Max and this will tell the PIC to stop updating, temporarily, until the signal reduces again. This will be used to give an audible signal. The operator then presses a button, telling the PIC to stop updating and start measuring phase of RX signal. At the same time the amplitude of the signal will be displayed on a centre-zero meter. There can be a setup procedure for the PIC whereby it can be given limits to the phase angle it will accept as a valid target.

The PIC will run at 20MHz giving a minimum timer period of 200ns. The period of a 14kHz signal is 71.4microsecs. So a resolution of about 1degree of phase angle will be possible in a measured ½ cycle. If all operations cannot be accommodated within the timespan of a half cycle, then the duties of TX generation and FET control will be shared between two PICs – they are cheap enough!

That is by no means the end of the story, and the circuit will be refined as development progresses.

[mikebg]

Not only TX but RX too. Complete schematic and no PIC please!. Thanks.

WM6,
The RF amplifier for RX is posted today in section "General Electronics".