PI Power Output

[Carl-NC]

The myth says that the ON transient can not generate as much Eddy Currents as the OFF transient.

I think you will find this depends on the tau of the target relative to the turn-on tau of the coil. Yes, I have seen a good RX response in the turn-on region from a nice thick coin. I don't recall now, but I don't think fast-tau targets (like small nuggets) do so well during turn-on. This is why at turn-off we try to get the flyback out of the way ASAP and sample early, and also why VLF gold detectors run at high frequencies.

I don't have a PI-IB rigged up right now, so someone else will have to verify.

From what people say, an aluminium foil can be cut to a specific size, that corresponds to a desired TC, which means size of the target changes the TC.

Almost true... thickness.

I use a 1"x1" aluminum (sorry, aluminium) foil square as a standard test target. On V3 it has a VDI number of 4 and hits hardest at 22.5kHz. Now sandwich 2 of those squares together and I get a VDI of 8. I have these sandwich stacks (laminated in clear packing tape) with thicknesses of 1, 2, 4, 8, 16, 24, and 32x. All still 1"x1" mind you. When I get to a 32x thickness the VDI is 60. When I stack the 16x, 24x, and 32x together (total of 72x) I get a VDI of 78 (US dime) and it hits hardest on 2.5kHz.

You can do the inverse with silver... foil-thin silver will read like a low-conductive target and respond better to high frequencies.

As any VLF coin hunter knows, trash responds literally all over the scale. So does jewelry. Gold nuggets do, too. So, skin effect probably plays a greater role in target response than does the metal alloy itself. We tend to calibrate on standard targets (coins) that are consistent (and consistently thick), so we deceive ourselves on the importance of "conductivity". It is really admittance, where skin effect plays a prominent role.

- Carl

[Tinkerer]

Hi Tinkerer,
I don't understand why is the difference between targets with different TC important to you. Is it useful for discrimination?
Correct me if I'm wrong, but can't I have a gold target with the same TC as an aluminium target (or any other metal).

From what people say, an aluminium foil can be cut to a specific size, that corresponds to a desired TC, which means size of the target changes the TC.

Regards,
Nicolae

Hi Nicolae,

after you have spent a few thousand hours metal detecting and have dug up 500,000 pieces of trash, to find 1000 pieces of treasure, you will understand why people would love to have information about the target before digging.

Consider this, how long does it take on average for you to dig up a target?
If you could just save 10% of the wasted time would it be worth .... what?

VLF people prefer the VLF to the PI because of its ability to discriminate.

In the de-mining business they consider the cost of each dig to be about 100 dollar. This is why they spend enormous amounts of funding to develop ways to obtain
information about the targets.

Even if discrimination is not perfect, it still helps.
All the best

Tinkerer

[Tinkerer]

I think you will find this depends on the tau of the target relative to the turn-on tau of the coil. Yes, I have seen a good RX response in the turn-on region from a nice thick coin. I don't recall now, but I don't think fast-tau targets (like small nuggets) do so well during turn-on. This is why at turn-off we try to get the flyback out of the way ASAP and sample early, and also why VLF gold detectors run at high frequencies.

I don't have a PI-IB rigged up right now, so someone else will have to verify.

As any VLF coin hunter knows, trash responds literally all over the scale. So does jewelry. Gold nuggets do, too. So, skin effect probably plays a greater role in target response than does the metal alloy itself. We tend to calibrate on standard targets (coins) that are consistent (and consistently thick), so we deceive ourselves on the importance of "conductivity". It is really admittance, where skin effect plays a prominent role.

- Carl

Thanks for the feedback, Carl.

Actually it is the opposite. The eddy currents of a 5 TC or 10 TC target will
have mostly decayed by the time you sample them after switch OFF.
Sampling during TX lets us sample the target at the moment of highest response. This is where the advantage is.

In general VLF detectors are better in detecting small targets. This is a way to close the gap between VLF and PI.

The second advantage is in the possibility of discrimination.

I fully agree about the skin effect.

Tinkerer

[Tinkerer]

Gday Tinkerer, i was just reading your last post where you say about the Myth as you put it:

Why not just test this & see, it should be a pretty easy test, all the physics in the world can't explain a lot of things.
In a perfect mathematical enviroment where most of this stuff comes from is a far cry from real world testing.
I was having a look at a similar thing while playing around with things.

Hi B^C,

Well, I tested it hundreds of times. I know what the results are. However, every time I open my mouth about it, I am being told that it is against the laws of physics.
Now, I am old enough to have learned a lot of things at school that have been proved different since then. I am also old enough to be suspected of senility, so I would really like for others to try and to confirm if what I say is right or to convince me to check in to the "Friendly Home"

All the best

Tinkerer

[nick_f]

Hi Nicolae,
after you have spent a few thousand hours metal detecting and have dug up 500,000 pieces of trash, to find 1000 pieces of treasure, you will understand why people would love to have information about the target before digging.

VLF people prefer the VLF to the PI because of its ability to discriminate.

Even if discrimination is not perfect, it still helps.
All the best
Tinkerer

Hi Tinkerer,
I am sorry for being a bit slow to catch-up, but I still don't understand.
By finding the TC of a target, it will provide me information about its size, not about its material, right? Or you can identify what metal is the target made of?
With VLF detectors, I've done a bit of testing (actually just with a two transistor circuit, consisting of an oscillator and a preamp) and I was able to identify the metals pretty well, no matter of the target size (I have to do more testing, now I know more than before). With VLF, diferent metals cause different phase changes.
I suspect you try to get information about the phase change (not about the TC of the target) with PI detectors, right? TC would be just a variable in the equation, with the other variable being the amplitude of the response?

Regards,
Nicolae

[Tinkerer]

Hi Tinkerer,
I am sorry for being a bit slow to catch-up, but I still don't understand.
By finding the TC of a target, it will provide me information about its size, not about its material, right? Or you can identify what metal is the target made of?
With VLF detectors, I've done a bit of testing (actually just with a two transistor circuit, consisting of an oscillator and a preamp) and I was able to identify the metals pretty well, no matter of the target size (I have to do more testing, now I know more than before). With VLF, diferent metals cause different phase changes.
I suspect you try to get information about the phase change (not about the TC of the target) with PI detectors, right? TC would be just a variable in the equation, with the other variable being the amplitude of the response?

Regards,
Nicolae

Hi Nicolae,

Yes! And this lag time in the impulse response corresponds to a phase shift in a sinusoidal response.

- Carl

The way I see it and the way I understand Carl's quote, it should be possible to get somewhat similar information as with the sinusoidal response of the VLF.
I am not there yet. For the time being I have observed different responses for different metals, but there are so many factors, size, shape, volume distance to the coil etc. it will take a lot of work to unravel the maze.

For FE it is relatively simple. The response is negative for FE and positive for non-magnetic targets.
But, if a target, lets say a rusty thin piece of steel sheet, presented in its flat position to the coil has the same amount of reactive response as it has resistive response, they cancel each other.
Now, if you present this same target on edge to the coil, you see mostly the reactive response that is characteristic to FE.

Either changing the angle of the target or changing the angle of the coil may help.

Tinkerer

[nick_f]

Hi Nicolae,

The way I see it and the way I understand Carl's quote, it should be possible to get somewhat similar information as with the sinusoidal response of the VLF.
I am not there yet. For the time being I have observed different responses for different metals, but there are so many factors, size, shape, volume distance to the coil etc. it will take a lot of work to unravel the maze.

For FE it is relatively simple. The response is negative for FE and positive for non-magnetic targets.
But, if a target, lets say a rusty thin piece of steel sheet, presented in its flat position to the coil has the same amount of reactive response as it has resistive response, they cancel each other.
Now, if you present this same target on edge to the coil, you see mostly the reactive response that is characteristic to FE.
Tinkerer

Hi Tinkerer,

Thanks for the explanation.
I didn't associate Carl's response with your experiments, although he specifically answered to your observations. From what Carl said, even for VLF it is very difficult to discriminate against junk, since junk can be found all over the VDI sca;e.
The fact that iron can be pretty easy identified is already a major advantage, since a very significant portion of the junk is iron.
About the resistive and reactive response, if you measure them separately (I think I read about that on some patents), they won't cancel each other.


Regards,
Nicolae

[ZED]

Gday Zed,

All targets are saturated with a magnetic field this is what creates the eddy currents, the extent of the saturation, well that's another story,
Old Boy!
These targets are being belted with a very strong magnetic pulse & they
only seem to produce so much even with a time variant so one would assume they can't be saturated any more?

Well its your universe dude,you can have any physics you like but to me saturation is the end of the road,no more,finished,maxed out !!!

Zed

[ZED]

Now B^C if you want more signal try increasing the avalanch voltage on your coil switch [ fet ],say if its 200v go for 400v.
What i see is not a target saturating but a coil field collapsing at a steady rate for a period of 300 ns i think
This is caused by the diode in your coil switch clipping the high voltage fly back and this controls the speed of the collapsing coil field and this speed will be constant for the duration,the target signal will reflect this with a flat horizontal zone at its peak signal .

Zed

[ZED]

Thanks Carl,

so I am not crazy after all. What I have observed hundreds of times, is real.

Now, let's tackle the other persistent myth. Hey, I like the "Myth busters"

The myth says that the ON transient can not generate as much Eddy Currents as the OFF transient. Laws of physics are blamed for that. After all somebody or something has to be blamed for our mistakes no?

I keep insisting that this is a fallacy but my whining falls on deaf ears. (Didn't use the right language, dahh..)

So lets look at this explanation, does this make sense?

Hmm maybe I should make a sketch, they say a picture is worth a thousand words...

Looking at the sketch,

a) represents the TX coil current with about 1.5 TC, at switch ON.
b) and c), from post #43, the fast and slow Flyback.

Now, look at the "area under the curve" which one is the largest? Ooops, small mistake here, will have to come back on that later.

d) is the TC current in black and the target response in red. There is a lag between the target response and the stimulation. As the di/dt diminishes, the stimulation of the target and its response diminish. This slowdown is a bit exaggerated on the sketch.

e) is the TX charge curve with a "flat top", like a about 6 TC. The response of the 2 different TC targets (red and green) raises with a time lag, relative to their TC. The lower TC target (red) has reached its peak earlier and as the di/dt diminishes, the eddy currents start decaying.
The longer TC target (green) lags more, the time is too short for it to reach its maximum eddy currents.

OK, this is the first try at drawing the picture. It is not all that good, but maybe it gets the idea across and somebody with better skill and knowledge could improve on it.

Tinkerer

G`day Tinkerer

The delay in the target signal i imagine would be caused by the natural inductance of the target trying to resist eddie currents forming but with time eddies will grow.
Im with you in that the eddies appear to be stronger during on time as apposed to the left over eddies after the transmit.
The target signal you see at picture "E" will reflect the speed of the growing coil field.The growing coil field will be at its fastest initially and begin to slow as it approaches saturation,this will be reflected in the target signal where during the linear part of the coil field growth is where you will find the best eddie current growth,as the coil field growth slows the current eddies cannot be sustained because of the slower growth rate of the coil field and so the eddies will decay to a point reflecting the slowing growth rate of the coil field until when saturation is reached and then all you have left are very low level eddie currents that need a little more time to decay to zero.
Thats my interpretation anyway.

Zed

[Tinkerer]

Hi Tinkerer,

Thanks for the explanation.
I didn't associate Carl's response with your experiments, although he specifically answered to your observations. From what Carl said, even for VLF it is very difficult to discriminate against junk, since junk can be found all over the VDI sca;e.
The fact that iron can be pretty easy identified is already a major advantage, since a very significant portion of the junk is iron.
About the resistive and reactive response, if you measure them separately (I think I read about that on some patents), they won't cancel each other.


Regards,
Nicolae

Hi Nicolae,

You got that exactly right if you measure them separately the don't cancel each other.
It is all about separating the total lumped information into its individual branches and reading them separately.

Tinkerer

[Tinkerer]

G`day Tinkerer

The delay in the target signal i imagine would be caused by the natural inductance of the target trying to resist eddie currents forming but with time eddies will grow.
Im with you in that the eddies appear to be stronger during on time as apposed to the left over eddies after the transmit.
The target signal you see at picture "E" will reflect the speed of the growing coil field.The growing coil field will be at its fastest initially and begin to slow as it approaches saturation,this will be reflected in the target signal where during the linear part of the coil field growth is where you will find the best eddie current growth,as the coil field growth slows the current eddies cannot be sustained because of the slower growth rate of the coil field and so the eddies will decay to a point reflecting the slowing growth rate of the coil field until when saturation is reached and then all you have left are very low level eddie currents that need a little more time to decay to zero.
Thats my interpretation anyway.

Zed

Hi ZED,

I think we agree on the general idea. Lets get into a bit more details on it.
The time of eddy current growth should be readable from the shape curve. (as shown in green and red in my post #85)

Maybe there is information to be gleaned by comparing the "Time of Growth" with the time of decay.

What I have observed so far, is the moment in time where the response has a "PIVOT". This is a dead spot, where a measurement produces 0. It is the point where the "relative polarity" changes.
The PIVOT for FE is at a very different time than for non-magnetic targets. With the right kind of decoding, it seems to me that the same phase shifts as in VLF can be read in this "PIVOT SINE WAVE"

Tinkerer

[Tinkerer]

Phase shift measuring in PI

Zed,

I am going to start a new thread for this phase shift measuring idea. Would you please help me discussing this idea?

Tinkerer

[Zed,]

How do you diffirentiate the different inductances that targets have,wont this alone give you different pivot points ?
As for the delay in signal growth this could be used to get an estimate of target size,larger targets would have slower growth rates and so more phase where as smaller targets with less inductance would have a quicker signal response and less phase.

Where is this pivot point measured,at the coil or at the output of an opamp,your pivot scenario has me confused,im a dodo with vlf.

Zed

[Zed,]

Zed,

I am going to start a new thread for this phase shift measuring idea. Would you please help me discussing this idea?

Tinkerer

As best i can,im not very bright.