Cylinder Coils?

[Smirnov-Arta]

...

Sorry friend). I went through all this at school), not counting the universities. Why do I need your reference?). You have given one of the formulas. There are several of them. Any physical quantity can be expressed using different parameters, is this news to you? I pointed you to another one. They are equivalent if translated to the area of a specific coil. If you are not lying, then take textbooks and watch / read, measure the parameters and calculate. And then take the oscilloscope and look. And then take a spectograph and take a look. And everything will fall into place... Therefore, there is no need to "reinvent the wheel", everything has been open for a long time. You just need to take it and do it, based on really measured physical processes occurring both in the coil and in the ground, etc., and not hypothetical assumptions about some kind of "Magic Coil".

[Detectorist#1]

No "magic coil". For every different purpose have optimal coil as shape, measures, inductivity, capacitance and soo further.

[Carl-NC]

The equation is

which states that total flux is proportional to inductance and current. This suggests that a higher inductance always wins. However, for a fixed power supply voltage, current is usually inversely proportional to inductance. In a traditional PI the turn-on slope is VTX/L where VTX is the TX power supply so peak current is VTX/L*t_on (assuming R_L = 0). In a sinusoidal TX the current is proportional to VTX/(2πf*L). This suggests that total flux is ideally independent of inductance regardless of the TX driver.

Even if it were, flux distribution can vary with winding geometry. For example, a scramble-wound coil vs a flat spiral. And then you have the RX side, where ε = N*A*dB/dt, so here you want more turns on the coil. In a split-coil design you have independent control of the TX and RX coils, but in a mono coil PI you don't. If you were to increase the turns on the coil then, yes, RX coupling would increase but it would also slow down the coil decay and push out the earliest sample delay. This reduces depth. All of this suggests that, for a given coil type (say, round scramble-wound), there is an optimal inductance that produces the deepest depth on a given target.

[Detectorist#1]

Thank you Carl for this " All of this suggests that, for a given coil type (say, round scramble-wound), there is an optimal inductance that produces the deepest depth on a given target.

[Detectorist#1]

In the words " on a given target " is cashed the dog. For small targets the fastest coil is important, for medium size targets medium speed coil with optimal inductance is preferred.

[Tinkerer]

The equation is

which states that total flux is proportional to inductance and current. This suggests that a higher inductance always wins. However, for a fixed power supply voltage, current is usually inversely proportional to inductance. In a traditional PI the turn-on slope is VTX/L where VTX is the TX power supply so peak current is VTX/L*t_on (assuming R_L = 0). In a sinusoidal TX the current is proportional to VTX/(2πf*L). This suggests that total flux is ideally independent of inductance regardless of the TX driver.

Even if it were, flux distribution can vary with winding geometry. For example, a scramble-wound coil vs a flat spiral. And then you have the RX side, where ε = N*A*dB/dt, so here you want more turns on the coil. In a split-coil design you have independent control of the TX and RX coils, but in a mono coil PI you don't. If you were to increase the turns on the coil then, yes, RX coupling would increase but it would also slow down the coil decay and push out the earliest sample delay. This reduces depth. All of this suggests that, for a given coil type (say, round scramble-wound), there is an optimal inductance that produces the deepest depth on a given target.

Since we have all the experts assembled, let's look at the possibilities of a coil for a CCPI. (round scramble-wound), diameter 300mm, 70 turns, 2 Amps current,
di/dt = 140/10us. At a distance of 600mm this gives us 0.0836 Gauss.
This is possible for a CCPI, but would be an unusable coil for a traditional PI.

[Carl-NC]

Since we have all the experts assembled, let's look at the possibilities of a coil for a CCPI. (round scramble-wound), diameter 300mm, 70 turns, 2 Amps current,
di/dt = 140/10us. At a distance of 600mm this gives us 0.0836 Gauss.
This is possible for a CCPI, but would be an unusable coil for a traditional PI.

This is where directly comparing CCPI with monopolar PI is tricky. It is dB/dt that determines the induced target kick and, in your example, that would be 8.37uT/10us = 0.837uT/us. Let's compare that with a monopolar design, where the coil is limited to, say, 20 turns but can turn off in 2us. The magnetic field at 600mm (for the same 2 amps) is now 2.39uT so dB/dt = 1.195uT/us, which is actually higher than the CCPI case. However, the CCPI can likely run a much higher N on the RX coil without compromising the timing. The bottom line is, it's hard to see all the gotchas involved so you almost have to build everything out and compare with real targets to get an accurate answer.

[moodz]

.... coils are coils are far as I am concerned ...without a magnetic monopole it gets hard to move outside the physics constraints.
However there are still things that can be done in the electronics. I have evolved the CCPI circuit i orginally published and now get these results.
I can punch 6 amps bipolar pulsing through a 300 uH coil or for example a 3 millihenry coil with the same fall times and no excessive flybacks.

I am still working on it so dont ask me to tell you how its done yet .. the blue trace is a 300 uh coil ... the green trace is a 3 millihenry coil.
The current pulses now fall to zero before the next pulse. 50 microseconds pulse on then off.

[Mdtoday]

MdToday, yeah mate , i have done my historical research, "Trove' brings up most of the info on his field, and yes the fork area produced the majority of the chunky nuggets from 2oz to upto 22oz that the founder of the gully was named after, there is very little mention of the other gullies except for one mention of nuggets upto 2grams found in some of them, so we know that if they were digging 2 gram nuggets at the bottom of the gully at 3 to 5ft deep then what chance do we have?, BUT people like Nenad who got in first cleaned up the known/found patches so that leaves the rest of us little to no chance of scoring any, but i might take your advice and hit those areas you mentioned as i haven't detected up that way yet, what i would like to see is a map and all the patches that have been found marked (by all the finders), since the place has been smashed it shouldn't be considered blasphemous to ask this, it's not like some new tech will come along and suddenly allow you to hit the area again and start digging up ounce chunks ( what do you think of this suggestion, still a bit too much to ask?)

Nova, If you get into the areas I noted, you will find some yellow stuff. I use a small 7 inch elliptical coil which suits the area.
Less noise, sensitive to the small stuff, still gets to the shallow bedrock or wash in the places I seek out and easy to get in and under the hakea and other brush.
On your asking suggestion, I wouldn’t but that’s just me, however I do see your point mate.

[Nova]

MdToday, fair enough, just one more question about 'Forties" road, yeah i do have the yearly pass, but we can't use that now, is Forties road always closed by gate?, i think when i drive past it ,it's always open, but i'm not sure, and since we aren't allowed to drive inside now, how or by what road would you choose to get into that area, thanx mate?

[Smirnov-Arta]

....

Unfortunately, you still don't understand what I wrote to you about. The ideal coil for ANY MD is 1 turn. Can you implement this in relation to your MD? If YES, then that's great! If not, then you will always have a compromise between the ideal theory and the actual MD scheme.

[Tinkerer]

.... coils are coils are far as I am concerned ...without a magnetic monopole it gets hard to move outside the physics constraints.
However there are still things that can be done in the electronics. I have evolved the CCPI circuit i orginally published and now get these results.
I can punch 6 amps bipolar pulsing through a 300 uH coil or for example a 3 millihenry coil with the same fall times and no excessive flybacks.

I am still working on it so dont ask me to tell you how its done yet .. the blue trace is a 300 uh coil ... the green trace is a 3 millihenry coil.
The current pulses now fall to zero before the next pulse. 50 microseconds pulse on then off.

This is not CCPI, Continuous Current TX anymore. You have a long period with 0 current.
It is a bipolar pulsed PI. Also very interesting. It appears to be even better than the CCPI.

[Carl-NC]

For me, CCPI means "constant current PI," so it's a variant of CCPI. It's constant current in that the pulse is not a rising exponential, but rather a constant current over the pulse interval. This is the first kind of CCPI I built back at White's, and it can use a mono coil.

[hykj888]

I have tested this coil; I used flat cable, which is intended to be used with the DP3SP. This is because the output of a large coil made with regular cable does not match the waveform required by the DP3SP system. Using this coil allows for perfect operation. Compared to traditional, it has lower parasitic capacitance. I have also tested the bipolar single coil; it is not suitable for detecting large, deep targets with a large coil because eddy current duration is very short. It is best used in a gold detector, as it can quickly sample and has high sensitivity to small gold nuggets. Carl refers to as "constant current PI," is it the ZVT of the GPZ7000? I have tested this circuit using analog circuits, it can achieve the function, but it is not perfect. It requires the support of an FPGA
Translated by 360AI。

[Tinkerer]

For me, CCPI means "constant current PI," so it's a variant of CCPI. It's constant current in that the pulse is not a rising exponential, but rather a constant current over the pulse interval. This is the first kind of CCPI I built back at White's, and it can use a mono coil.

There is nothing better than a mono coil for depth.
I am using a DD coil at present. It gives FE discrimination close to the coil, but the discrimination does not go the full depth.
Does anybody have more information about the relative depth of the FE discrimination of a DD coil?