CONCENTRIC SEARCH COIL DESIGN

[DOOLEY]

hello all ,
have been reading allot on search coil design , concentric in particular , including one by carl moorland and one by dave emery ,

dave emery's artice on coplanar coils makes it all look quite easy , with the straight forward ratio of 4:1 regarding the bucking coil.

two pictures bellow i found on carls article are of great interest........





right then , the design with 2 tx coils , looks like it would waist energy / have a lower magnetic field due to poss cancelation of the bucking coil wound in the opposite direction , or am i wrong ??

the one that uses two rx coils interests me , as on the face of it , it looks like there would be no waisted power / field interaction and canceling.

so here is the big question , what is the ratio between coil turns ???

is it the same as the design with the bucking coil ??

would the design with the two rx coils offer bennifits over the bucking coil design ??

more sensitivity ???

i have drawn a blank regarding getting information online , cheers g@@gle !

so here i am , asking for help.........
PLEASE HELP !!! OWH..........PLEASE.

[Carl-NC]

Without giving it much thought, I would expect the RX+/RX- ratio to be 4:1 if the diameters are 2:1. Easy to try it and see.

[DOOLEY]

cheers carl , was a bit unsure due to the different arrangement , makes sense , thought i'd ask before getting started.

[Aziz]

Hot Tip

Hi Dooley,

forget the figure 12 configuration. It's not going deep enough.

Figure 11 configuration is not optimal.

One can get more depth sensitivity, if:
a) coil coupling between receive coil (RX) and bucking coil (TX-) is maximized (as much as possible = as close as possible to each other)
b) provided that a) is achieved, you can increase the radius of TX- and RX (same radius).
(Radius 0.7 .. 0.8 times the outer radius of TX+ coil gives a good range for experimenting, but is dependent on the condition a) )

It's too diffucult to build these coil. There are much simpler and more efficient IB coils available (not invented by others yet but I have found them ).

Cheers,
Aziz

[DOOLEY]

cheers aziz for feedback,
was thinking fig 12 looked the most efficient due to not having a "live" bucking coil , which i rekon lowers the strength of the field.
so fig 12 looked the best overall choice , was thinking of sticking to the 1/2 diameter dimentions to retain the 4:1 windings to make it easy.

so you rekon most efficient design would be outer rx- touching tx inner , touching rx+ final inner ???

might give up the idea , was thinking of using it for a PI idea and sample during on time using the rx + and rx - using fig 12 configuration, and poss sample tx duing decay time.

one proplem i thought of was , if we make a PI that samples duing on time , havent we just made a higher powered vlf ??? in which case we are back to the problems inherant to vlf's , noise from hotrocks and sea salt et....etc....

PI's "miss" most of that due to sampling after the main of the noise has ended , so thinking of sticking to what we know for now.

still looking for perfection / the best idea before i get started on a build , by doing this i can see me sat here toying with idea's anf thinking for the next 10 years , and still not started anything !!!

[DOOLEY]

going to add a question ,

is it "best" for a PI to be mono coil , or do i carry on experimenting with TX and RX coils ??

[Aziz]

going to add a question ,

is it "best" for a PI to be mono coil , or do i carry on experimenting with TX and RX coils ??

Yes, mono coil achieve more depth sensitivity than any IB coil (regardless of type).

Regarding figure 12:
Let me make more EM field simulations and I can tell you more later on.


Design guides for IB coils:
- (dominating = more inductivity) TX coil should be as big as possible (high flux area coverage) and should have as much turns as possible
- (dominating = more inductivity) RX coil should be as big as possible and should have as much turns as possible
- bucking coil should have generally less turn count and should have a higher inductive coupling to the cancelling coil

With these simple rules, one can optimize an IB coil. The typical concentric co-planar coil with the 0.5*diameter bucking/RX is definitely not the optimal coil design (RX coil has less flux area coverage).

To be continued ...

Cheers,
Aziz

[DOOLEY]

cheers aziz , can't wait for field visualisation on the differance between fig 11 and 12 , this will answer the question me and tinkerer had a long time ago on another post , as to weather opposite fields cancel out or distort.

if the typical concentric design is not the optimal design , are we being short changed by manifacturers regarding performance.

i would have thought that sepperate RX and TX would have had better sensitivity / performance than mono coil especialy if RX coil has more turns , transformer effect.

carl was saying a while ago about the "ballance" between number of turns and decay time , and it's all a compromise ,

if we have the "optimal" TX coil , and we add an RX coil with say 4 times the amount of turns to get more sensitivity , will the decay time be extended due to the RX having more turns ??? or does decay time governed by the TX coil ???

[Aziz]

Hi Dooley,

still trying to find a better coil of figure 12 configuration. Figure 11 type is performing always better.

I should have mentioned, that I am always taking the same coil inductivity and same coil size (10" outer diameter) for performance comparisons. Inductivity TX(total) = 300 µH and RX(total) = 300 µH and TX & RX = induction balanced. Fair fight between different configurations (like the box fight Haye vs. Klitschko today. BTW, a good fight. Delayed the analysis today ).

It's of course easy to add more RX winding turns in an IB coil but the RX coil in this case would also induce more EM interference noise. This step can always be done to increase the performance until noise decreases the signal-to-noise ratio or other parameters gets worse (i.e. capacitance, resistance, weight due to more copper). Or you can just increase the gain of the RX amplifier.

The only parameters, which will be changed in my EM simulations are:
- geometric (diameter, position, winding direction)
- number of turns of the individual coils

Unfortunately, the performance improvements end up in a few cm or inches only. But the pure mono coils can be challenged by the IB coils, if you are using the pulse-on response as well. This will compensate the depth loss. Note, mono coil PI's use only the half of the spectral response energy of the targets (pulse-off time sampling).

I'll do some more EM simulations on figure 12 configuration.
Stay tuned...

Cheers,
Aziz

[Aziz]

Bad News for Figure 12 Design

Hi all,

I have tried many many figure 12 coil configurations (a lot of virtual copper has been wasted ). They didn't achieve the same level of the depth performance however. The figure 11 coil is always better than figure 12 configuration.

On the other hand, it is very difficult to build both coil configurations. A simple DD (double-D) coil design has considerable more depth performance compared to figure 12 design. And it is very very easy to build.

Figure 11 design can however be optimized to be more efficient for depth performance. Just take my hints into account and you get a good working concentric-coplanar IB coil.

Cheers,

Aziz

[Aziz]

Hi all,

as you can see (below), one can make more efficient and deep seeking concentric coplanar coils. But it's very important, that the coil coupling between bucking and RX coil should be maximized (coils as close as possible). In this case, the number of turns of the bucking coil decreases and the number of turns for the TX coil increases (for a given inductivity). Additionally, you can make the RX/bucking coil diameter slightly bigger, which effectively has more flux coverage area. In the example below, I could increase the inner diameter up to 0.8 times of the outer diameter. If you make it bigger, the depth performance decreases then. So you should experiment in the range of 0.7 .. 0.8 times the outer diameter.

Anyway, this coil is not for beginners. It's quite difficult to build (to me too). I personally wouldn't invest the effort as there are much simpler coils there.

Cheers,

Aziz

[DOOLEY]

cheers aziz for putting in the effort into an old design.

i'll think i'll poss move away from the idea then , thank you for you input.

[Carl-NC]

Figure 11 configuration is not optimal.

One can get more depth sensitivity, if:
a) coil coupling between receive coil (RX) and bucking coil (TX-) is maximized (as much as possible = as close as possible to each other)

True, you want extremely tight magnetic coupling, but you also want zero capacitive coupling, which means having a little spacing. Most manufactured coils wind the RX right on top of the TX-.

b) provided that a) is achieved, you can increase the radius of TX- and RX (same radius).
(Radius 0.7 .. 0.8 times the outer radius of TX+ coil gives a good range for experimenting, but is dependent on the condition a) )

Maybe, maybe not... a larger RX means you will also pick up a stronger ground signal, which might result in no net SNR improvement.

As examples, the old Fisher concentrics (from the CZ's) had an RX that was 0.6*TX+, yet Tesoro concentrics have an RX that is 0.3*TX+. The Tesoros go every bit as deep, maybe deeper. I realize this isn't a fair comparison as they are completely different detectors, but the fact that Tesoro uses such small RX coils suggests they found an advantage in doing so.

It's too diffucult to build these coil. There are much simpler and more efficient IB coils available (not invented by others yet but I have found them ).

This kind of concentric is remarkably easy to build, at least in a production environment. However, give us more hints on the more efficient method!

- Carl

[Aziz]

True, you want extremely tight magnetic coupling, but you also want zero capacitive coupling, which means having a little spacing. Most manufactured coils wind the RX right on top of the TX-.

That's true. It's more critical, if the TX coil is carrying a high voltage. A good shielding should minimize the capacitive coupling. A minimal gap is always present and necessary however.

As examples, the old Fisher concentrics (from the CZ's) had an RX that was 0.6*TX+, yet Tesoro concentrics have an RX that is 0.3*TX+. The Tesoros go every bit as deep, maybe deeper. I realize this isn't a fair comparison as they are completely different detectors, but the fact that Tesoro uses such small RX coils suggests they found an advantage in doing so.

Small RX coil is not efficient (fact). RX coil needs some flux coverage area to go deep enough.

This kind of concentric is remarkably easy to build, at least in a production environment. However, give us more hints on the more efficient method!

No, I don't think it is easy to build. I am speaking for the amateurs here.

The coaxial coils are more efficient than the concentric coplanar, 4B and DD coils. Particulary the ones, I have mentioned in the other thread (having lesser thickness now). And these coils are far more easier to build. Particularly, if somebody needs bigger IB coil.


This might be interesting to everybody:
The mono coil is equivalent sensitive (depth performance), if you put the same mono coil with another same as RX coil in the OO IB configuration. This however occupies significantly more surface area of the coil. But easy stuff.

If I try to optimize the DD coil (yes, it's also possible), there isn't much room for optimization (performance gain is not much significant as the coils TX and RX are already big enough). It's still possible to make the overlap area bigger and hence the TX and RX coil slightly bigger for the same coil coverage area. This can be achieved with a bucking coil in the overlap region of the DD coil. (same applies to the OO coil)

Funny stuff. But we are talking about very small depth gain. All the effort isn't really worth. I still prefer simple and easy to build coils.

Aziz

[ApBerg]

coils...

In ' clean ' ground the open centre coil is good to use with sensitivity that is almost just as good as a 2D coil, but with better disk and pinpoint options...
In fresh water sears (not salt beach) I prefer the open centre for good disc and pinpoint, look at the open centre coil in the pdf at post no:2 The cable is in the rod so less resistant in the water...http://www.geotech1.com/forums/showthread.php?p=77266#post77266
The open centre coils I found easier to make, this because the nulling is not so hard to make as a 2 D coil, a 2D coil is more mechanic work...
Best regards.
Ap
Off course this is a personal opinion !! I have respect for others ! And are very..very curios in what is new....