Advanced PI coils

[Tinkerer]

Originally Posted by Ferric Toes Correct. Too much reliance in the past on the simple mono coil, because of its simplicity, where the constraints are for the benefit of the TX.

Eric.





The good old mono coils have served their purpose for a long time. However, they are far from ideal.
Lets look at some alternatives to the simple mono coil configuration.
Separate TX-RX of same or different amount of turns and inductance.
IB configurations, concentric and not. Different amount of turns or different inductance.
Differential coils
etc.

Personally I prefer concentric, co-planar IB coils. This configuration is not very popular for PI detectors. Why?

Better alternatives?

Ideas?

Tinkerer

[WM6]

Personally I prefer concentric, co-planar IB coils. This configuration is not very popular for PI detectors. Why?

Better alternatives?

Better alternative is mono coil (except for our friend mikebg).

If we are speaking of depth, of course.

Complex coils (structured out of two or more coil) cannot go so deep as mono, cause "more coil in one" represent load to each other. Coil under load cannot go so deep as "free" mono coil.

If we are speaking of discrimination etc. things can go different. As I understand our main interest still remain depth. So mono in its variations.

[Davor]

Considering the differential coils' ability to cancel EMI, EF and ground, plus the opportunity to wind a Rx coil to a better S/N performance, and a possibility of a real discrimination, well, I'd say the monocoil deal is not such a clear-cut after all.
With the best preamps performing at ~1nV/sqrt(Hz) noise (~60ohm source resistance) monocoils are not performing optimally. You can fix that either by providing a much better preamp, or winding a coil to higher impedance, latter being a more trivial task.

[Ferric Toes]

The good old mono coils have served their purpose for a long time. However, they are far from ideal.
Lets look at some alternatives to the simple mono coil configuration.

Personally I prefer concentric, co-planar IB coils. This configuration is not very popular for PI detectors. Why?

Better alternatives?

Ideas?

Tinkerer

Good idea to have a separate thread for this subject as I reckon there will be a lot to debate. Strange as it may seem, the first PI's in the 1960's had coplanar concentric TX and RX coils of different inductances for a very good reason. The mono TX/RX coil originated when we were developing diver held underwater detectors. Underwater coax cable and sealed connectors were limited in their availablity so we opted for the simplest system. Coils had to be open centred and neutrally bouyant and not act like a huge paddle. Mono coils were easy to make and the idea stuck for subsequent land detectors, except for a few custom made units and special industrial applications.

I agree that the coaxial coplanar arrangement is good because each coil can be optimised for the circuit it is coupled to. RX could be just inside, on top of, underneath, or just outside the TX coil if we want both coils to be about the same diameter. RX can be thin wire relative to TX as there is virtually no current. We have to bear in mind the stepup in high voltage spike though, that could result. I'm currently doing some experiments with Moodz bifilar CT coil. Anyway, I'm glad the ball is rolling on an interesting subject.

Eric.

[WM6]

Where it goes Moodz bifilar CT coil?
I consider it still as mono coil.

[Ferric Toes]

My 1968 PI which used IC's/ RTL logic and 709, 712 Linear devices. DECCO stands for Decay of Eddy Currents in Conductive Objects. Coil array is a small one for lab testing. Field coil was 1m square for TX and 0.5m RX. My initials and date are in the bottom right hand corner. EJF 2 68.

Eric.

[WM6]

Interesting bit of MD history.

[Ferric Toes]

The reason the the 1960's PI's had separate TX and RX coils was that it reduces the ground signal considerably, particularly from the very surface. These PI's were developed originally for surveying archaeological sites, not only for metal objects but for ground anomalies too. A strong ground anomaly could indicate the site of an old pottery kiln where the heat from firing had enhanced the viscosity response. Those detectors were not auto-zeroing and a measurement was taken with the coil array sitting on the ground that had previously been gridded out. A mono coil would have just given the response from the very surface, but a dual coil ignored the first few inches where anomalies can be smeared by ploughing etc.

For a discussion of this (primarly for maths freaks), Google "The Transient Electromagnetic Response of a Magnetic or Superparamagnetic Ground" by T. Lee. WBGB Aziz will love it.

Now for modern coil development. The 8in bifilar CT coil that I wound, with external screening, has a resonant frequency of 1Mhz. That is with it not connected to anything except its own distributed capacitance and 1ft of plaited 3 wire lead. I was exciting it by holding near to a running PI coil and observing the induced ringing on the PI front end. Comparing this with a 10in standard mono, screened in the same way and wound with the same 0.25mm Teflon wire, the rfreq was 0.9 Mhz. This indicates that the bifilar construction does not add appreciable capacitance. Inductances were 320uH (total) and 335uH respectively. The bifilar winding ended up about 5 twists per inch as the Teflon insulation is more slippery and springy than pvc and tends to untwist even when stretched a bit. Both coils just bundle wound - no fancy basketweaving.

Eric.

[Aziz]

...
For a discussion of this (primarly for maths freaks), Google "The Transient Electromagnetic Response of a Magnetic or Superparamagnetic Ground" by T. Lee. WBGB Aziz will love it.
...

Hi Eric,

WBGB *LOL*

Unfortunately, I couldn't find the whole paper. Only abstracts... (they want some money.., krauts want some money too, damn it all!)

Got from the abstract (compressed impression):
1/t bound onto conditions
1/t² *LOL*
1/t^2.5 *LOL*
Anyway.

Would be interesting to fly over it. But I'll hold onto the real measured data instead of abstract theories from strange people. That's a good advice to look a the real measured data.

Ooops, it's getting off-topic. Sorry guys.

Cheers,
Aziz

[Davor]

There are some newer papers and dissertations related to human demining with thorough theoretical background. I even found some detailed analysis of ferrous targets. Hopefully it will enable me to tune up the spice target models for Fe.

[Ferric Toes]

Would be interesting to fly over it. But I'll hold onto the real measured data instead of abstract theories from strange people. That's a good advice to look a the real measured data.

That's not exactly fair to Lee who is well respected in the field of geophysics and who was very willing to answer some of my questions by email a few years ago. His "theories" match exactly what I have found with mono coils and concentric TX/RX on the very highly mineralised Australian ground, both thin surface layer and uniform ground. Also viscosity plots from my Magnetic Viscosity Meter add confirmation.

Eric.

[Davor]

So far I had no problems obtaining just about any kind of information/papers/sound files from all kinds of authors, and without any finances complicating the exchange. It is the corporations that force us to believe that intellectual property is difficult to come by.

Eric, will it be possible for you to nudge Mr. Lee to share his paper with us? I've already noted that it was quoted in several other papers that are full of goodies, so why not learning from the source

[Aziz]

That's not exactly fair to Lee who is well respected in the field of geophysics and who was very willing to answer some of my questions by email a few years ago. His "theories" match exactly what I have found with mono coils and concentric TX/RX on the very highly mineralised Australian ground, both thin surface layer and uniform ground. Also viscosity plots from my Magnetic Viscosity Meter add confirmation.

Eric.

There is a misunderstanding Eric. I'm not criticizing Lee.

The *LOL*'s being for people, who strictly are holding onto the 1/t response (d log(M)/dt ).
(I'm not criticizing the 1/t response as well.)

There has been written a lot. Which one is correct? Best to look at the measured data.

Aziz

[Aziz]

Hi all,

has anyone built the Tophat(c)(r)(tm) coil yet?
I find it to be the most powerful IB coil and one of the best IB coil inventions in this century. *LOL*
(When PJ says it won't work -> take the opposite -> it will work very very well )
<Advertising and bashing mode disabled now>

It is looking very complex (four coils: TX/RX+, BX/RX-, TX = transmit coil bottom, BX = bucking coil top, RX+ = receive coil bottom, RX- = receive coil top) but is very easy to build and balance. It allows a better (mechanical) fine induction balancing feature without the need for extra loop turns too.

Just meet the following criteria:
Height (co-axial) distance between TX/RX+ and BX/RX- coil bundle > 2* RX+/RX-/BX coil radius (better more but not really required to be more than TX coil diameter)
The RX+/RX-/BX coil radius factor: 0.5 - 0.7 times the radius of TX (no problem, it works even with smaller radius factors)
A higher radius factor requires more bucking coil turns and a higher inductive coil coupling coefficient between BX and RX- coil.
The BX/RX- inductive coupling can be used as a rough IB (or adjusting the desired co-axial distance), whereas the co-axial distance offers a fine IB feature. If you add one turn more to the optimum BX coil, you have enough room to make the rough IB'ing and co-axial distance setting. It is quite easy to set the BX/RX- coil bundle distance IMHO.

Cheers,
Aziz

[Ferric Toes]

Eric, will it be possible for you to nudge Mr. Lee to share his paper with us? I've already noted that it was quoted in several other papers that are full of goodies, so why not learning from the source

I could upload a .pdf of the full paper provided there are no copyright problems. It was published in Geophysics and Geophysical Prospecting in 1984 with slight differences.

Eric.