Verator front end

[Sergey_P]

and at the end the phase discrimination will always be more or less fuzzy

not discrimination, the ability to balance the the ground to zero.
GB: S=(G+T)*sin(dф) (G- ground; T- target; dф- (phase(G+T)-phase(G)) )
Verator: S=(G+T)-G*|cos(ddф)| (ddф- phase window (=>0) )

[Davor]

On the other hand, a series tank does not provide impedance transformation, but instead it features band pass characteristics, and a very small phase change over 50% inductance increase. It would, however require a better op amp on input to maintain the same noise figure, but it provides great improvement over the original layout in terms of phase and amplitude stability.
See for yourself...

[Davor]

And now the verdict. I assumed Verator to have 10mH Rx coil because I have no idea what the actual inductance is. I was very merciful with placing the resonance having the 50% inductance change in my mind. And I assumed 10kHz as operating frequency. As inductance rises, the Rx tang gets closer to resonance, so the amplitude rises, and phase also shifts. It is obvious that the voltage on the input of an op amp is 3-18 times larger (the impedance transformation effect), but the most important feature is a rapid change of amplitude and phase. It is also apparent that the parallel tank is in effect a 2nd order low pass filter...

[Davor]

No, it is the same. A coil placed in environment with rising μr will have inductance rising. if there is a double μr the inductance will rise twice, regardless of the initial absolute inductance. Same with resonant circuits, with twice μr the resonance will go down by square root of 2. It is the same effect for both Tx and Rx coils, as they share the same environment.

I played with a voltage controlled inductance, so called "varicoil" which may become handy for modelling behaviour of coils in varying environment. It is controlled by a voltage in a way that the sensitivity is 1H/V. In the example circuit below 1mV of control voltage stands for 1mH inductance. This is an idealised component with no possibility of magnetic coupling. It is a hierarchical model so it must be accompanied with files Varicoil.asc and Varicoil.asy

Let's see what will Sergey come up with. I'll proceed with modelling a Verator Rx front end to see what happens with phase when inductance changes up to 50%.

[Verator]

As a receiver you have nothing else, and do not apply except for the coil. It does not matter what it is round or square. Important as it is set up. In the Tesoro used as resonance at the TX power will not get sensitivity. Whites did not use resonance. Nothing new there you do not do and do not offer. Otherwise does not exist. No low-noise amplifiers do not solve the problem of the suppression of the soil. This is the first task. If there are no ideas, it does not matter what's the amplifier circuit and to the switched capacitor. This is the little things that do not solve. Only for beauty.When you and I will learn even be that there is, then perhaps we can offer something better, or even be worse.

You say that both the input circuit and a transmission change their inductance of the same and in the same direction. For example, but then these circuits should be identical and in the same conditions. It is possible to realize a sensor DD. But what about in a planar sensor when the transmitter 2 big the size of the land area and is attached to it for more than a receiver circuit? Here is all the other way. This point can not be used as an advantage.

[Davor]

Funny but that's almost exactly what I'm saying. Verator uses a parallel tank in Rx which worked perfectly well for old Tesoro, Whites, and others, and these designs are quite old and clunky considering what you can do with nowadays components.

[Verator]

I would like to draw attention to the fact that at high frequencies Q.faktor increases and the phase slope of the up and up. At a lower frequency and the quality factor of less than a linear phase slope But this does not mean that the resonance of the system always work well. It is only in Tesoro and similar devices, the development of 20-30 years ago. There is nothing good in those schemes do not. All primitive and cheap. Commerce.

[Davor]

Sergey is right about the phase noise. Unfortunately this is not the only source of phase noise and at the end the phase discrimination will always be more or less fuzzy.

The effect Sergey is describing is a second order phase change. It happens because the ground vector is not at exactly the same angle at different frequencies. It is however a second order contribution because the ground vector rotates very little over a very wide frequency span. Important thing to note is that the first order phase change related to phase difference between Tx and Rx frequency resonance difference is at least an order of magnitude larger than the ground vector rotation over the same frequency change. With fixed Tx frequency it is directly proportional to the Rx tank Q.

Sergey, I propose you to analyse a phase difference contribution over resistive ground and, say, 50% inductance change and corresponding frequency change of a free running Tx. I'll do the same for fixed frequency Tx and Rx with parallel and series tank input. We may both learn something from this.

[Verator]

but changes the phase of the signal from the ground, which causes the phase noise...

That's right! Distorted the response object itself. But all these devices in phase with notch irrelevant because the accuracy of such systems is low, even using low-noise amplifiers. These solutions are used in cheap commercial devices

I do question 10 years and realized something ..

[Verator]

Verator front end

http://www.youtube.com/watch?v=plueAJwsUF4 (Verator3) We take only the outline TX receiver circuit is disabled, input shorted. In this video you can see how the TX resonant circuit behaves on the purpose or effect of the soil. Work on the timing synchronization detectors TX. Static principle allows us to estimate this impact. If there is an oscillation circuit and changes the inductance of the phase current (field) will change in parallel inductance effects.Noise input stage is not fundamentally important, since signal ground is much higher and does not have a decisive influence, especially knowing that port X is no suppression of the ground effect. It simply is not!MD is the first phase meter but not a seeker of metal. For the phase measurement have a good reference. No oscillator circuit does not provide. Everything else - the lyrics. What there are comparators and how it is soldered, it does not matter. When not making simple requirements then it's just a commercial solution.On the principle of change of inductance built all the devices Q-meter cares what this circuit TX or RX.These contours are different, working under different conditions, have different impedance. How they can work together to change the action of the soil? Sensors such as DD and coplanar totally different. And is it permissible to change by itself? These changes are then amplified inductor channel amplifiers in 1000! This is the negative of this approach. This is not the best in the world ...If you recall the principle of IB. there is no inductance, about her question. Work only coils coefficient. transformation method for isolating the target signal, the secondary induction. Everything. There are no other resonances and inductors. Resonances is the technical implementation with advantages and disadvantages.

[Sergey_P]

that a free running frequency is a bad thing. Reality is quite opposite. When you approach the ground with your coil, both Tx and Rx coils will be affected by same rate of inductance change. In a fixed frequency system the Rx tank will detune against the fixed frequency drive, and you'll have a phase change. If frequency is free running, a phase difference between Tx and Rx will remain the same, before and after.

but changes the phase of the signal from the ground, which causes the phase noise...

[Davor]

Phase slope steepness is related to the tank Q, so it is of less consequence if tank Q is low. In case of Tx tank it is almost irrelevant because it is the voltage across the coil and the resulting current through the coil that matters, not the current through the tank capacitor, hence Tx Q is of little consequence if you drive your coil with voltage. In case it is tuned badly - it will just use more current. (PI machines do that all the time )

Rx front end can be realised with parallel tank, which is in effect a low pass filter, as the voltage source is in series with the inductance, and with a capacitor it forms an L topology low pass filter. It improves input noise by means of up-transforming the input impedance and cheap op amps can deliver good results. It must be operated off-resonance because at the very resonance the phase slope is very steep.

Using a better op amp makes series tank a better solution because of very low Q (<0.5) but it does not perform up-transformation so there is no noise improvement in this topology - the op amp must be low noise. The way to fix it is using transistors to reduce input noise. I think this is a far superior solution because there is no AM->PM problem related to phase slope.

Now, you often emphasize a wrong assumption that a free running frequency is a bad thing. Reality is quite opposite. When you approach the ground with your coil, both Tx and Rx coils will be affected by same rate of inductance change. In a fixed frequency system the Rx tank will detune against the fixed frequency drive, and you'll have a phase change. If frequency is free running, a phase difference between Tx and Rx will remain the same, before and after.

There is a big advantage in Verator that it drives a Tx by symmetrical signal, and thus avoids all the 2nd harmonic problems. It is unfortunate that Verator's Rx is not designed to accommodate for fixed frequency operation, e.g. by operating at minimum Q.

I also think you did not understand my tank locked loop approach. PLL chip, and flip-flops if needed, are providing a 50% duty cycle drive at or very near the tank resonance. In effect it uses the best of both worlds, a free running frequency, and punctual 50% duty cycle drive. As a plus I can also extract phases digitally by simply dividing a VCO frequency with a ring counter and use phase shifts in discrete steps. So it is really the best of both worlds.

[Verator]

Verator1 live http://youtu.be/reGVKNx6f1g
Pay attention to the sound of the lid of a beer
The rest is a dynamic instrument of little interest. The trouble all dynamic schemes is that the scheme of differentiation makes his interpretation of the response, which does not actually have. The scheme matches the logical output channels are not perfect. Because of the different values VDI. there are windows errors. The channels are overloaded
Frequency separation TX-RX. is not for the frequency and the difference of course is irrelevant. This is done in phase linearity. At the resonance phase curve! Ground inductance moves the receiver circuit and violates the parameters of the goal! Will not help the PLL. Why would he need a PLL in this case? If we look to the characteristics of the circuit as in the picture is for phase linearity to disrupt the contour of 2 kHz. either upwards or downwards in frequency. Better down.

[Davor]

IGSL lacks a few things to wish for, but also has two discrimination channels that I like very much. That's why I am playing with balanced Tx and other things.

The digital shift is tightly bound to the TX. Lack of flexibility of choice relations XY.

Verator has fixed values of shift for X and Y, so there is nothing wrong about digital phase shift. Besides, you can achieve phase shifts other than 90° by means of Johnson counter, so you can include nickel, foil and stainless steel directly. It would still use less power than the comparators.

You see, having a Tx tightly bound to Rx shifters makes accurate frequency of no consequence within very wide limits. I'm planning using LF frequencies for my next build. LF will broaden the response of small targets, foil, nickel and such, but comes with strong interferences, mostly from Russia hence I need working frequency as flexible as possible. To improve balance and have the most out of the more silent LF band, I'll use both Tx and Rx in balanced mode.

[Verator]

The digital shift is tightly bound to the TX. Lack of flexibility of choice relations XY. At low frequencies, 2-6kHz have to shift the phase of the signals for low-conducting objects ... Nickel, shallow gold alloys. The ratio is not 0 \ 90Using the shifted resonance receiver RX feature allows you to get. how to filter LPF. and align the phase. You can even do without resonance but worse sensitivity and a lot of garbage. For example the receiver coil in the sensor Blue 950 max. has 400 turns and is set to 4,6 kHz. TX circuit has Q = 5 (30 turns of aluminum) 6,6 kHz.Now remember this is a vector unit working in statics. Conversion function of the target signals of all nodes device must be linear phase, Resonance is a bad phase. As a result - bad signal vectors, poor balance of soil. In simple devices such IGSL no phase linearity (linearity and it is not necessary). So always say this is the ground he so strongly influenced by the depth decreases. But the influence of soil were concluded in the ground and in the instrument of his abilities.