TARGET ID AND THE PIVOT

[Tinkerer]

If we can get information about the TC of a target, we have a much better idea of the identity of the target that lies buried in the ground.
we can then decide if it is worth the effort to dig it up.

The TC of the target is the time it takes for the exponential signal to decay about 63 %. So we can sample the peak target response and sample again some time later and then calculate the TC of the target.
This sounds very simple and will work perfectly with the FPGA, where samples at every uS are taken.

Of course, to be able to work out the algorithm, the behavior of the target signal needs to be perfectly understood.

So below I copy an old post of mine where I show THE PIVOT.

There is an interaction between the target TC and the coil TC and the TX TC. This interaction produces a PIVOT POINT.

This PIVOT POINT is shown in the pictures attached in my old post below, copied from another thread.

Let's look at what happens during the transmit time.
The TX pulse has about 12 V and a duration of 90uS.
The DC resistance of the Mosfet, the cable and TX coil is about 2.5 Ohm.
The coil has an inductance of 375uH, that means 375/2.5= a TC of 150uS

At 35uS, the coil current has reached about 1A, about 9.6W into the coil.

At 90uS, the coil current has reached about 2.15A and 14W into the coil.

The pictures show the response of different targets when exposed to the changing magnetic field.

I have changed the time scale to 10uS/div, so now the Tx or Transmit Time takes nearly the whole screen.

#1 No target
#2 A piece of half inch square alu foil is the target. It has a TC of about 5uS
#3 A piece of one inch square alo foil, about 10uS
#4 A Nickel ($0.5 coin) about 15uS
#5 A Penny ($0.1 coin) about 70uS
#6 A Quarter ($0.25 coin) this one has a TC of more than 90uS
#7 the last one is our steel lug again. It has a TC of more than 90uS and its response is negative, (saturating the preamp) while all the non ferrous targets gave a positive response.

The distance of the targets varies, to avoid saturation of the preamp, while still making it possible for the naked eye to see the responses of the different targets during TX.

Tinkerer
Attached Images

[ZED]

Ok so where on the waveform is the pivot point ?

Zed

[Tinkerer]

Ok so where on the waveform is the pivot point ?

Zed

Hi Zed,

The pivot is at a different time with each target. The time is dependent on the TC of the target.

What you see is a combination of signals generated by the coil, the ground and the target. This makes it difficult to spot the Pivot. After some time getting used to it, the eye can spot the Pivot easier.

Taking several samples along the time, and comparing the voltage level of the samples the Pivot is very easy to see.

Now, on these pictures:

First the FE target. There is no Pivot because the signal saturates.

On the other targets:

The target signal first increases, then decreases, then there is a spot that changes very little, then it increases again. In fact it is a sine wave superposed on the signal composed of the coil response and ground response.

I will try to post some pictures where it is easier to see with an untrained eye.
Tinkerer

[Mechanic]

Hi Tinkerer,

Could you also post some pics that is showing what the tx coil is doing at the same time.

Mick

[Tinkerer]

TWO TAU METHOD

Hi Tinkerer,

Could you also post some pics that is showing what the tx coil is doing at the same time.

Mick

Hi Mick,

attached are pictures of the controlled Flyback of the "TWO TAU METHOD" as well as pictures where the PIVOT is easy to see, made with the "TWO TAU METHOD"

I am sorry to say that the pictures in post #1 are not made during Flyback, but during TX. Small mistake, but wide ranging results. So the cat is out of the bag.

You see, the response during TX and the response during the Flyback, are the same. It is just that the response during the Flyback is many times higher.
How many times?

Take the energy of the TX, spread over say 100uS and then discharge the part that is stored in the coil and discharge it in 8uS and you have about 10 times more signal amplitude.

The interesting part is that the signal itself behaves the same way., just concentrated into a shorter time.

As mentioned elsewhere, there are many factors influencing the signal. Once we know them all, we can separate these responses and look at them individually.
I am not quite as far yet, but I know how to separate the ground response, the FE response and the non magnetic target response.

To separate the responses, it is important to understand the interaction of the TC of the coil, the TC of the target and the duration in TC of the TX pulse.

Now, with the "TWO TAU METHOD" we have the results of TWO TX pulses within one cycle. the responses are of the same kind but different amplitude as explained above.
Tinkerer

[Tinkerer]

TWO TAU METHOD PICTURES

The first set of pictures shows the voltage and current ramps of the TX and the Flyback.

The second set shows target response.

The pictures showing the NO TARGET signal, superposed with a target signal, show the PIVOT well.

The alu square targets are cut out of a thin houshold alu foil. the one inch square alu foil target has a TC of about 10uS when measured in the traditional PI way.
The 2inch by 1 inch alu foil target has a TC of about 20uS when measured the traditional PI way.

The gold ring is high carat gold, very thin and small.

There are pictures of all these targets shown in other threads.

The PIVOT is the point where the target signal crosses over the NO TARGET signal.

Since the whole signal window has a time of only about 10uS, the TC of the target has to be translated to a relationship between TX ramp and the traditional target TC. The ramp is about 10 times shorter, so the target TC is about 10 times shorter too.

The sampling window shows the signal from the switch off. The first spike is the switch off noise. Right after the noise the targets can be sampled.

These pictures show a very noisy signal because of some loose wire in the coil, as well as the loose connections on the breadboard, where the wires were just stuck into the solderless breadboard.

The fact that in spite of the primitive setup, impressive results can be shown, proves that the method has great potential of refinement.

These pictures are quite old. Much improvement has been done since.

Tinkerer