New Concentric Co-planar Coil with Strong MF.

[satdaveuk]

...and VERY cheap!!! Even new blades are really cheap.....a dozen at a time for pennies......

As a kid in the '50s, I made jigsaw puzzles for my younger brother with one!!!

There are electric ones around, some on ebay....liken here:-

http://www.ebay.co.uk/itm/FRETSAW-Dr...item3cbaf4e0fd

regards

Andy

ditto we made jigsaw puzzles with a fretsaw, and my dad made them for us when we were knee high to grass hoppers, I thought most people owned one.
Talking of comparing your DD coils with other types, I can only comment on personal experience of owing and using many different brand detectors over the years that to be honest with you even in are uk soils it largely depends on the detector design on whats good or better on which particular machine.
For instance the Whites DFX which is not so popular over here and never was for what reasons im not sure but at a guess I think your avarage detectorist just found them to complicated to use, but if they had grasped the way it operates and had done some good field tests by people like myself with my own custom settings which took a good 18months and hundreds of hours to master then the tables would have turned because even by todays standards of machines on the market there still way past half way to the top end machines.
Coil wise they work at there best with the circular coil they came out the factory which was a 950cm, I have done exstensive tests with the Whites 14x10 DD coil which Whites themselfs told me should give me that extra depth which never happened.
All it did was make pinpointing less accurate and ruin the balance of a perfectly made machine.
So tech boys work that one out because that coil been so much larger should in theory have at least produced more depth .
On the other hand I have a Minelab Quattro which is multifrequency with a 10DD coil factory fitted ,it gives a tad more depth with silver finds, most detectors are good with copper and lead so not worth the mention, but gold its crap, and that no disc and all metals, even my Whites 1v pics up better again with the circular coil.
So my point is between the two types of coils you have been chatting about its horses for courses, I really have come to the conclussion that some work better than others on different machines and yes even maybe the same makes and models because two machine will never be identical due to lots of factors, and thats why different people use different settings, its not just the types of soils etc there detecting on, its also the machines, What do you think?
Ive never made a circular coil for a induction balance machine because they seem so complicated, has anyone here got some uncomplicated instructions to build one say for the TGSL most of us built here because it will be interesting to see the results, Ive made loads of DD coils with very good results, or at least I think so.
May a add ive been reading this thread with great interest like so many others on here, your such a clever bunch I thought I was good but now I feel like a apprentice again
So a big thank you to all of you and as ive said previously if you was girls id give you all a big wet kiss.
On a serious note wish I had known about this forum years ago its been a education and such a learning curve.
Regards
Dave

[simonbaker]

...and VERY cheap!!! Even new blades are really cheap.....a dozen at a time for pennies......

As a kid in the '50s, I made jigsaw puzzles for my younger brother with one!!!

There are electric ones around, some on ebay....liken here:-

http://www.ebay.co.uk/itm/FRETSAW-Dr...item3cbaf4e0fd

regards

Andy

Thanks, I'll keep an eye out for something like that. I have a problem with space however, so will have to do some major organizing at some point.

Regards,

-SB

[der_fisherman]

Nice - I knew of coping saws, but this is better.

...and VERY cheap!!! Even new blades are really cheap.....a dozen at a time for pennies......

As a kid in the '50s, I made jigsaw puzzles for my younger brother with one!!!

There are electric ones around, some on ebay....liken here:-

http://www.ebay.co.uk/itm/FRETSAW-Dr...item3cbaf4e0fd

regards

Andy

[simonbaker]

That sounds like a great idea for anyone without a router. Maybe cutting bits of wood or ply with a hand fretsaw.....

For anyone who does not know what a fretsaw is or how to use it, it does finer work than an electric jigsaw, look here:-

http://en.wikipedia.org/wiki/Fretsaw

regards

Andy

Nice - I knew of coping saws, but this is better.

[der_fisherman]

That sounds like a great idea for anyone without a router. Maybe cutting bits of wood or ply with a hand fretsaw.....

For anyone who does not know what a fretsaw is or how to use it, it does finer work than an electric jigsaw, look here:-

http://en.wikipedia.org/wiki/Fretsaw

regards

Andy

[simonbaker]

Hi,

Although I am far away from being an expert, I have made 6 or 7 coils using plywood and a router, the first time was at least 10 years ago and I still have that machine, though all the ones since have either been sold or given away or weren't mine in the first place.....and I haven't built a new detector in years.....but I will dig out the old Twin loop PI that I made and make some photos of the head......light, but strong, maybe not pretty, but prettier than a mud fence!!

After reading your reply, I do feel that such construction details need to be better documented and put on Geotech in a blog for construction details (assuming that nobody has done that up to now!!), we have a coil blog, perhaps that is the right place, but for electronics cases and the expanding leg (I am useless on this bit!!) ideas, maybe another blog should be started.....what do you all out there think?

When I make the coil (this year) for my IGSL, I will make pictures and post them here on the Coil Blog, its really simple and easy to do........the trick being to cut a slot only as deep and wide as needed as the epoxy quickly adds weight. I sand down and fill any tiny voids with car filler fiber glass, slightly flexible version.

It should not be used to actually hold the coil down itself though, then a good sand, followed by one coat of spray paint, sand lightly, then two coats.....it keeps moisture out of the wood. I have also sometimes coated with epoxy to seal.....it depends on how fluid the epoxy is, some are too thick.....

Where coils crossover, the depth of cut needs to be increased (doubled) to account for that so that the coils are not "sticking out" of the wood. I also try and make the coil winding itself a bit wider and therefore thinner at the point where they crossover in a Double D config (If I don't forget to do that when Super glueing!!), to stop major thickness buildup.....

Tools needed are an electric sander with various grades of sanding discs, a router with various different cutters, ones with a ball end are particularly good as you have a cut that is rounded at the bottom and this reduces the amounts of epoxy needed....eg. weight!
But even if you don't have a rounded cutter, a square one will do as well.......

Try and measure the thickness of the coils and use a cutter of that width, if the slot is then too tight, you can always run the cutter around again, slightly offset from the first time.....its usually helpful to cut a millimeter deeper than you think you need, then when sanding, you won't hit and damage the coil......

The biggest problem is that you need a piece of plywwod far greater than the coil, so that you can hold it down well, you also need some throwaway wood underneath, so that you can cut out areas to lighten the assembly up, and you need to cut deeper than the plywood. So there is a degree of waste that one has to accept as well, so plan carefully to keep that to a minimum......

I buy my plywood at a big company where they take peanuts for good big quality offcuts.....that keeps costs down.

The overlap areas need to be more open to allow the coils to be nulled easily, be prepared to cut some wood away to assist in that.....

Do try the coils in the head BEFORE YOU TAKE IT OFF THE ROUTER TABLE. (been there and got the T-Shirt!!), you may need to take off more wood and its easier when its still held fixed.....

As a rule of thumb, you need to find plywood at least twice as thick as the coil, but make sure that its not too thick (too heavy), though you can make it narrower where/when you cut out unwanted wood.......you can add another piece of ply on top of the nulling area, to add extra strength nif needed.....properly glued over the whole area......and clamped while setting. Use the router to make it look better and to blend it in.....

Now who here knows a great way to make a telescopic leg for the detector......????

Regards

Andy

I like it.

I'm also interested in an alternate construction -- instead of routing, cut pieces and build them up like a sandwich, gluing pieces, leaving channels that would be same as routing. The idea is to do it with minimum tools, just a saw (hand scroll saw or jig saw or band saw).

The bottom piece would be a very thin veneer plywood or even cardboard since the stacked pieces would eventually make a rigid structure.

Always searching for extreme DIY techniques for those who just want to get an MD made no holds barred.

-SB

[simonbaker]

You will have probably noticed that IB VLF motion detectors have an optimum sweep speed. Some detectors like to be swept very slowly, whereas others (such as the Tesoros) can be swept very quickly. So your analogy to driving a car with knackered suspension is quite apt. Only when your head hits the roof have you found a bump of the correct size.
Of course, driving more slowly would miss the bump. As would driving a lot faster.

Good point -- how many people know the optimum speed of their detectors?

Also gives some insight into ways in which a detector can be fooled -- such as small bumps spaced apart at just the right intervals might build up a signal, like a pendulum. Next time you sweep, it's gone -- then its back -- drives you nuts...

This does depend on the damping factor of the filter of course, some more sloshy than others I imagine.

-SB

[Monolith]

What about a small TX coil inside a large RX coil, would that improve matters more?

Regards

Andy

Hi Andy,

I never tried a second small TX in the center. The RX works best in a position of weak magnetic field density. There is a French patent where they call it a "magnetic vacuum", because the residual voltage from the TX coil coupled to the RX coil is a limiting factor.

What causes the residual voltage?

Can the residual voltage be calculated?

Monolith

[der_fisherman]

I like your way of making coils, I have been interested in the idea of using router to cut grooves in plywood (or other wood) for the coils with just enough room to adjust for nulling. Then cut away excess wood. Currently I make some coils just gluing coils to a flat platter of thin veneer plywood. Uglier than a mud fence but easy to make -- except attaching the pole requires some construction. But I have a long way to go at making a great MD.

I'd like to see some pictures of your coils.-SB

Hi,

Although I am far away from being an expert, I have made 6 or 7 coils using plywood and a router, the first time was at least 10 years ago and I still have that machine, though all the ones since have either been sold or given away or weren't mine in the first place.....and I haven't built a new detector in years.....but I will dig out the old Twin loop PI that I made and make some photos of the head......light, but strong, maybe not pretty, but prettier than a mud fence!!

After reading your reply, I do feel that such construction details need to be better documented and put on Geotech in a blog for construction details (assuming that nobody has done that up to now!!), we have a coil blog, perhaps that is the right place, but for electronics cases and the expanding leg (I am useless on this bit!!) ideas, maybe another blog should be started.....what do you all out there think?

When I make the coil (this year) for my IGSL, I will make pictures and post them here on the Coil Blog, its really simple and easy to do........the trick being to cut a slot only as deep and wide as needed as the epoxy quickly adds weight. I sand down and fill any tiny voids with car filler fiber glass, slightly flexible version.

It should not be used to actually hold the coil down itself though, then a good sand, followed by one coat of spray paint, sand lightly, then two coats.....it keeps moisture out of the wood. I have also sometimes coated with epoxy to seal.....it depends on how fluid the epoxy is, some are too thick.....

Where coils crossover, the depth of cut needs to be increased (doubled) to account for that so that the coils are not "sticking out" of the wood. I also try and make the coil winding itself a bit wider and therefore thinner at the point where they crossover in a Double D config (If I don't forget to do that when Super glueing!!), to stop major thickness buildup.....

Tools needed are an electric sander with various grades of sanding discs, a router with various different cutters, ones with a ball end are particularly good as you have a cut that is rounded at the bottom and this reduces the amounts of epoxy needed....eg. weight!
But even if you don't have a rounded cutter, a square one will do as well.......

Try and measure the thickness of the coils and use a cutter of that width, if the slot is then too tight, you can always run the cutter around again, slightly offset from the first time.....its usually helpful to cut a millimeter deeper than you think you need, then when sanding, you won't hit and damage the coil......

The biggest problem is that you need a piece of plywwod far greater than the coil, so that you can hold it down well, you also need some throwaway wood underneath, so that you can cut out areas to lighten the assembly up, and you need to cut deeper than the plywood. So there is a degree of waste that one has to accept as well, so plan carefully to keep that to a minimum......

I buy my plywood at a big company where they take peanuts for good big quality offcuts.....that keeps costs down.

The overlap areas need to be more open to allow the coils to be nulled easily, be prepared to cut some wood away to assist in that.....

Do try the coils in the head BEFORE YOU TAKE IT OFF THE ROUTER TABLE. (been there and got the T-Shirt!!), you may need to take off more wood and its easier when its still held fixed.....

As a rule of thumb, you need to find plywood at least twice as thick as the coil, but make sure that its not too thick (too heavy), though you can make it narrower where/when you cut out unwanted wood.......you can add another piece of ply on top of the nulling area, to add extra strength nif needed.....properly glued over the whole area......and clamped while setting. Use the router to make it look better and to blend it in.....

Now who here knows a great way to make a telescopic leg for the detector......????

Regards

Andy

[Qiaozhi]

I have to correct myself -- sweep rate-of-change is insignificant since it is the oscillator rate-of-change that is responsible for the target currents. The sweep rate however does affect the filter electronics in the detector - and the filter definitely can act like a bowl of jello -- or at least a car with lousy shock absorbers cruising over bumpy ground looking for a bump of a certain size.

Regards!

-SB

You will have probably noticed that IB VLF motion detectors have an optimum sweep speed. Some detectors like to be swept very slowly, whereas others (such as the Tesoros) can be swept very quickly. So your analogy to driving a car with knackered suspension is quite apt. Only when your head hits the roof have you found a bump of the correct size.
Of course, driving more slowly would miss the bump. As would driving a lot faster.

[simonbaker]

Hi Monolith --interesting analogy!

The larger coil will also have less rate-of-change when sweeping typically, which gives the smaller coil an advantage up to a certain depth.
-SB

I have to correct myself -- sweep rate-of-change is insignificant since it is the oscillator rate-of-change that is responsible for the target currents. The sweep rate however does affect the filter electronics in the detector - and the filter definitely can act like a bowl of jello -- or at least a car with lousy shock absorbers cruising over bumpy ground looking for a bump of a certain size.

Regards!

-SB

[simonbaker]

Hi SB,

intuitively it would seem that the small RX coil will cover only a small search area. In reality this is not so. The area covered is still the area of the TX coil.
Ho does this work?
Look at the TX coil field as a large sphere. Consider this sphere to be jello. Now consider the target to be a bug that fell into the jello. The bug shakes to try to get out. The shaking is shaking the whole sphere of jello, but the movement is stronger near the bug.
If you have the coil sensor balanced within the sphere, it receives the signal directional, but it will sense any movement of the jello.
A large RX coil will be more sensitive for the same amount of inductance, but a smaller coil will generate less residual voltage and therefore can have more inductance to compensate for sensitivity.

Doubts on these fine points of this relationship between TX, BU and RX coils were nagging on me for a long time, so finally I built a few really large coils, like 0.5m, 1.2m and 1.6m. Then I experimented with small and large RX coils, from 300uH to 1200uH.

My findings are along the lines described above. I am sure all this could be proven mathematically and would welcome if somebody would do it for us.

SB, you say the proof is in the pudding. Will jello do the trick?

Monolith

Hi Monolith --interesting analogy!

Although the proof is in the pudding, experiments can be deceptive because I believe there is an optimum coil size for a given target at a given depth. This is because although a larger coil will have its magnetic field extend deeper, the field is more spread out and weaker to begin with -- so compared to a smaller coil, it starts out weaker, but then is stronger at a greater depth. The larger coil will also have less rate-of-change when sweeping typically, which gives the smaller coil an advantage up to a certain depth.

So it depends on the depth of the target what is the best size coil. But since shallow targets are easier to pick up with any coil, it makes sense to optimize for the deeper depths.

For a particular depth, I believe that it probably works best when the TX and RX coils are the same size if possible. I think that is because the equations that calculate the maximum stimulation of the target by the TX coil are just the same as the equations that calculate the maximum stimulation of the RX coil by the target. With concentric design, it is not possible to have equal sized coils, so you live with different sized coils.

The concept of "disturbing the balance" is just different way of visualizing superposition of fields. Really, the target just sees the TX coil simulating it, and the RX coil just sees the electromagnetic fields from the currents in the target. The balancing is just to remove the TX component from the RX reception so it is easier to amplify the target signal. (It also prevents coupling of the low impedance TX coil circuit to the high impedance RX circuit, which would rob some gain).

That is my intuition, but maybe something else happens too! Wish there were time to really do the math and experiments, because this is pretty essential to designing an MD. I'm sure the commercial vendors like Tesoro, Minelab have tried their best to figure this out, so copying their designs is pretty safe. I'd just like to figure it from scratch.

Cheers,

-SB

[simonbaker]

Great and interesting answer, many thanks.

By the way, no coil, once adjusted, can be allowed any movement, which is why many people first saturate the coil with super glue or similar before any form of shielding is applied.

Some big mono coils are filled with builders foam to achieve that.

I personally first, use superglue, then I wrap the coil tightly in insulating tape, pulled as tight as I can, then I take a piece of uninsulated multi-strand wire and wrap this spirally round the coil, leaving a gap of a centimeter or so, then, observing the gap, I wrap the coil tightly with strips of kitchen foil, holding them in pace with a futher layer of insulating tape also as tight as possible.

Finally, I progressively immerse in epoxy resin, adjusting the null each time if it varies, till the whole is immersed in hard epoxy.....but as little as possible as epoxy is heavy....

I usually rout out a piece of plywood to allow the coil to be set in, then coat the whole coil in epoxy as waterproofing, then lightly spray paint to look good, usually black....

If the plywood shape is properly made, it makes a light but strong construction for a search head, I am amazed that I never see any others made that way.....for one offs its great!!

You should look at the detection distances achieved with 25cm DIY coils with both the TGSL and the IGSL on YouTube........really fantastic.

http://www.youtube.com/watch?v=Nx4Oxac0Ufg

http://www.youtube.com/watch?v=MS9Tp2xw7RQ

regards

Andy....

I like your way of making coils, I have been interested in the idea of using router to cut grooves in plywood (or other wood) for the coils with just enough room to adjust for nulling. Then cut away excess wood. Currently I make some coils just gluing coils to a flat platter of thin veneer plywood. Uglier than a mud fence but easy to make -- except attaching the pole requires some construction. But I have a long way to go at making a great MD.

I'd like to see some pictures of your coils.

I agree, no amount of movement should be tolerated, but even bumping a rock would send shock waves across a coil.

-SB

[der_fisherman]

Hi SB,

intuitively it would seem that the small RX coil will cover only a small search area. In reality this is not so. The area covered is still the area of the TX coil.
Ho does this work?
Look at the TX coil field as a large sphere. Consider this sphere to be jello. Now consider the target to be a bug that fell into the jello. The bug shakes to try to get out. The shaking is shaking the whole sphere of jello, but the movement is stronger near the bug.
If you have the coil sensor balanced within the sphere, it receives the signal directional, but it will sense any movement of the jello.
A large RX coil will be more sensitive for the same amount of inductance, but a smaller coil will generate less residual voltage and therefore can have more inductance to compensate for sensitivity.

Doubts on these fine points of this relationship between TX, BU and RX coils were nagging on me for a long time, so finally I built a few really large coils, like 0.5m, 1.2m and 1.6m. Then I experimented with small and large RX coils, from 300uH to 1200uH.

My findings are along the lines described above. I am sure all this could be proven mathematically and would welcome if somebody would do it for us.

SB, you say the proof is in the pudding. Will jello do the trick?

Monolith

What about a small TX coil inside a large RX coil, would that improve matters more?

Regards

Andy

[der_fisherman]

I enjoy discussing theory in hopes of learning or improving understanding, so never hesitate to question or disagree with my speculations! I always believe the proof is in the pudding, so if theory doesn't agree with practice, I'm interested in finding out why and getting better theoretical models that are useful. I also admit I'm too lame to do the math that would answer many of my own questions.

I see a double-D coil as two eliptical coils side-by-side. Eliptical coils each have a sort of squished magnetic field that will be stronger at shallow depth than a circular coil of similar diameter, but then probably drops off faster because of the smaller width.

It gets complicated to imagine the net sensitivity of two such coils side by side. If the target is directly under the TX coil it gets maximum stimulation, but the RX coil is off to the side and is less sensitive. Vice versa if the target is directly under the RX coil. If the target is in the middle, neither coil is optimally positioned. So that is the double-D's Achille's heel similar to the concentric's need for the bucking coil.

On the negative side for concentric coils, one coil, typically the RX coil, tends to be very small, like 1/2 diameter of TX coil. That is a serious down-side and probably is at least as bad as the offset nature of the IB coils. Combine that with the bucking coil, and the concentric coil does have a serious Achilles heel.

I think the double-D pinpointing is good for surface coins, but only working across the overlap. With the concentric you can scan both side to side and forward-backward to pinpoint. I always thought concentrics were easier to pinpoint with, but that may be a personal feeling.




The idea there is that the bucking coil is a smaller coil than the TX coil, but designed so that the magnetic field of the bucking coil is virtually the same magnitude as the TX coil at the plane of the coils, only opposite direction. This causes near cancellation of the magnetic field at the plane of the coils within the interior of the bucking coil.

But the field of the bucking coil, whose diameter is much smaller, will diminish much faster with depth compared to the TX coil. So as you go deeper, the "field cancellation" goes away and TX coil field will dominate and provided the desired target stimulation. But I agree, there is always some cancellation, perhaps 25% or so that remains.




My speculation there is that with double-D coils, a slight shift of the coils causes significantly more or less flux from the TX coil to pass through the RX coil -- because that is how balancing is done! But with concentric coils, the magnetic field from the TX coil is more uniform in the interior of the coil perimeter, so shifting the RX coil does not make as radical a change in the amount of flux passing through the RX coil.



I'm not sure what you mean by tighter focus. For the same current and coil inductance, a smaller coil may have a stronger field at zero depth, but the larger coil probably has stronger field at greater depth (circular coils); i.e. the smaller coil field diminishes faster with depth. I need to do the math to confirm the exact quantities there. But in general, the larger the coil, the more depth you can get for a given inductance. There are some trade-offs there so that practically you don't want to go above a certain diameter for practical detection of certain size targets.

When talking about "power" used, it is tricky because a high-Q resonant TX coil will use less power than a lower-Q coil to produce the same field. The bucking coil does limit the Q because it adds wire which decreases the inductance (wound opposite direction) at the same time it increases resistance. So to get the same inductance as an IB coil, you need to wind more wire - more resistance, less Q. The circuit you use to drive the TX coil also affects the power used, so a high-Q coil can be wasted if you don't drive it efficiently. But I do agree that the IB coil should be more efficient power-wise. Using thicker wire helps minimize power loss of course.

I believe on the whole that IB (double-D/double-O) coils are deeper coils than concentrics, but I just wanted to point out some interesting virtues of concentric coils that make them more competitive than expected.

The shielding question is deeper physics for me - I'm aware there are different-order terms in the equation for an oscillating electro-magnetic field, and shielding has frequency-dependent effects, so I'm not clear on the net result. Practice says that shielded coils work better, but I think there is a lot of misunderstanding out there as to why.

I'm actually a big fan of the "mono coil" because of its perfect alignment -- maybe moodz can get one to work well.

Regards,

-SB

Great and interesting answer, many thanks.

By the way, no coil, once adjusted, can be allowed any movement, which is why many people first saturate the coil with super glue or similar before any form of shielding is applied.

Some big mono coils are filled with builders foam to achieve that.

I personally first, use superglue, then I wrap the coil tightly in insulating tape, pulled as tight as I can, then I take a piece of uninsulated multi-strand wire and wrap this spirally round the coil, leaving a gap of a centimeter or so, then, observing the gap, I wrap the coil tightly with strips of kitchen foil, holding them in pace with a futher layer of insulating tape also as tight as possible.

Finally, I progressively immerse in epoxy resin, adjusting the null each time if it varies, till the whole is immersed in hard epoxy.....but as little as possible as epoxy is heavy....

I usually rout out a piece of plywood to allow the coil to be set in, then coat the whole coil in epoxy as waterproofing, then lightly spray paint to look good, usually black....

If the plywood shape is properly made, it makes a light but strong construction for a search head, I am amazed that I never see any others made that way.....for one offs its great!!

You should look at the detection distances achieved with 25cm DIY coils with both the TGSL and the IGSL on YouTube........really fantastic.

http://www.youtube.com/watch?v=Nx4Oxac0Ufg

http://www.youtube.com/watch?v=MS9Tp2xw7RQ

regards

Andy....