Round vs Square shaped coils?

[brucester]

Hello,

I want to build a large coil for detecting medium sized objects relatively deep. Like a big saucepace at 1-2m.

Would a 1m round coil be better than a 1m square?

What effect does shape have on the field?

Or does it make no difference?

Also, is there a calculator that computes number of turns for square coils? Or can i use the circle calc with an adjusted diameter for the extra perimeter of the square?

[Davor]

You'll find that wire length is strongly related to the coil inductance, and shape much less. Same goes with the field and everything else. Round coils look coo though.

[brucester]

Thanks Davor.

So essentially, I have to square a circle to find its inductance using the circle calculator

One last question.

If i have a really large coil and dont provide enough amps, will there be a blind spot in the center? I mean, when the field encircles the wire but there is not enough power to radiate into the center?

Is there a method to calcuate how many amps are require for a magnetic field to reach the center of a coil? Sorry if that is worded poorly

Image a 10x10m square coil of 200uH being pulsed by a 1 volt battery. Would there be a big blind spot in the middle? But it would still detect metal close to the perimeter.

[Carl-NC]

Round is always better than square, but not by much, and large square coils are easy to build with PVC pipe. So go for square. Davor is right, a round coil "squared up" doesn't lose much inductance.

The "extent" of the magnetic field is the same regardless of current, only the strength varies. Strength, coils size, and target size all work together in determining sensitivity; a 1m coil typically won't detect a single small coin at all, but it may detect a larger coin at the perimeter. As you weaken the field larger targets will also start to drop out.

[brucester]

Cool, thanks Carl.

So, regardless of current, the distance from the coil to the tip of the inner magnetic field is the same for all currents?

And its ability to generate eddy currents in targets is dependent on the strength of the field.

So my question now is, is there a way to calculate how many amps are required to detect your target at the tip of the inner magnetic field? For example, if you wanted to detect a coin at 2m with a 2m coil, would supplying 10000 volts and 1000 amps be enough assuming you had a coil made of a material that wouldnt burn out?

When people say, you need 4 times more power to detect an extra inch, how much power is needed to detect to the max distance of a coil? And how do you calculate the max distance of a coil? How far can a 1m coil detect at infinite amps? :P

Is there any point experimenting with high power sources? Or does it max out at around 12-24 volts for practical purposes? I noticed in one thread you were interested in pulsing a coil with 1000 volts by a huge capicitor so it would seem more power equals greater sensitivity at greater depths up to the maximum distance.

Any help calculating that maximum distance for power would be great

[Davor]

If you look around you, most probably you'll see that the technology which stays around is the one that uses less energy. Say GSM - it is very battery friendly.

What you propose is a direction in which PI technology is already going, based on a mentality that the athlete with bigger muscles will run farther, and the one with enormous muscles even farther, while in reality they will only eat much more, and complain more often.

In case you are a person of built up and very muscular physique, just stick with the bigger bang approach - your muscles are going to swell even more in a process.

[Carl-NC]

I like Davor's reply... however, depth/sensitivity is largely an empirical process. Put lots of knobs on your design and see what happens. Simply adding more power gives rapidly diminishing returns, you'll likely find a point where the extra 1/4" of depth isn't worth another 10 pounds of batteries & heatsink.

P.S. -- my desire to use 1000V is simply to improve the turn-on slew rate, the TX current will still come from a 5V supply.

[Tinkerer]

I like Davor's reply... however, depth/sensitivity is largely an empirical process. Put lots of knobs on your design and see what happens. Simply adding more power gives rapidly diminishing returns, you'll likely find a point where the extra 1/4" of depth isn't worth another 10 pounds of batteries & heatsink.

P.S. -- my desire to use 1000V is simply to improve the turn-on slew rate, the TX current will still come from a 5V supply.

With large coils, the depth is no problem, large coils go deep. The sensitivity is. To get an idea about sensitivity, use the calculator at the link below.
Multiply by turns
Compare the field strength with that of a smaller coil at the depth/distance required.

http://hyperphysics.phy-astr.gsu.edu...ic/curloo.html

[Geo]

With large coils, the depth is no problem, large coils go deep. The sensitivity is. To get an idea about sensitivity, use the calculator at the link below.
Multiply by turns
Compare the field strength with that of a smaller coil at the depth/distance required.

http://hyperphysics.phy-astr.gsu.edu...ic/curloo.html

Hi Tinkener.
Using the calculator i see that if i want to have the same field at double distance i need to make the field 4... 8 times stronger.
For example a field created with 1A at a loop with radius 10cm, at 0.1m out along the centerline of the loop, the field is 0.022G. If we want to have the same field with the same coil at double distance (0.2m) we need to flow 4A current throu the coil.
At another thread Carl has wrote that if we want to make the same field at double distance we need 64x times the first field.
Who is the right or where i made the mistake???

Regards

[tony_av]

B =μ0Na^2I/2(a^2 + z^2)^3/2

where B is in tesla if
• μ0 = 4π × 10−7 is the vacuum permeability,
• N is the number of turns of the field coil,
• I is the current in the wire, in amperes,
• a is the radius of the coil in meters, and
• z is the axial distance in meters from the center of the coil.

We have here B because our coil generates magnetic field only.
That field decreases with distance as z^2^3/2 = z^3 and generates in target
emf, witch creates eddy current. That current also creates magnetic field.
That second field travels back to our receive coil and decreases also
with distance in same z^3. If distance is doubled, z becomes 2. So, we have
2^3=8 in one way and 8 in back way = 8*8=64.

Tony

[Qiaozhi]

Hi Tinkener.
Using the calculator i see that if i want to have the same field at double distance i need to make the field 4... 8 times stronger.
For example a field created with 1A at a loop with radius 10cm, at 0.1m out along the centerline of the loop, the field is 0.022G. If we want to have the same field with the same coil at double distance (0.2m) we need to flow 4A current throu the coil.
At another thread Carl has wrote that if we want to make the same field at double distance we need 64x times the first field.
Who is the right or where i made the mistake???

Regards

In fact you will need 64x, as the signal produced by the eddy currents in the target (Lenz's Law) need to return to the RX coil. (i.e. 8x8 = 64)

[Tinkerer]

Hi Tinkener.
Using the calculator i see that if i want to have the same field at double distance i need to make the field 4... 8 times stronger.
For example a field created with 1A at a loop with radius 10cm, at 0.1m out along the centerline of the loop, the field is 0.022G. If we want to have the same field with the same coil at double distance (0.2m) we need to flow 4A current throu the coil.
At another thread Carl has wrote that if we want to make the same field at double distance we need 64x times the first field.
Who is the right or where i made the mistake???

Regards

You can also use 2A and 0.2m diameter coil for 0.022G
Or you can use 1A and 2xturns and 0.2m diameter coil for the same 0.022G

For the RX path, I think that not only the distance, but also the coil coupling is important.

One question I am trying to find an answer, is how to calculate the coupling of the target signal to the IB coil.

[brucester]

the TX current will still come from a 5V supply.

5V? That is surprising. I did a simple test on the pirate detector. Running at 5V, it halves the distance on a coin that 9V gives.

I can see why increasing tx voltage beyond what is practically available by modern batteries is kind of pointless, but using half or a third of whats available seems to be self defeating. Unless you are building a detector that is sensitive to sub gram gold on the surface, but how is 5V strong enough for any kind of depth?

The "extent" of the magnetic field is the same regardless of current, only the strength varies.

Then why do people keep talking about doubling the distance of the field with 64 times more power?

Is there ever a point where a coil is maxed out? I mean, reached its max distance? If you keep increasing power by a factor of 64, does its distance keep doubling? Or does it max out distance and just keep adding strength or flux or whatever you want to call it :P

[PiTec]

For the RX path, I think that not only the distance, but also the coil coupling is important.

Exactly … and this depends on the distance and the ratio of coil and target diameter (plus target angle and more, but to keep it simple let’s assume the target is a small loop parallel to the TX coil). The above expression calculates the TX field strength at a point on the coil axis, but the target has a certain radius. And then for the RX path, this target radius is a different (usually much smaller) ‘R’ in the expression.

One question I am trying to find an answer, is how to calculate the coupling of the target signal to the IB coil.

We could have a look at contactless power transfer. Usually with spiral coils for both TX and RX, but still useful to get an idea about coil coupling. Example (also dealing with coil misalignment):
Wireless Power Transfer in Loosely Coupled Links: Coil Misalignment Model
http://thesciencedude.com/projects/F...nt%20model.pdf

Let’s heat up the targets ;-)

Thomas

[Geo]

In fact you will need 64x, as the signal produced by the eddy currents in the target (Lenz's Law) need to return to the RX coil. (i.e. 8x8 = 64)

Tony and Qiaozhi, thanks.
I found my mistake. I forgot to calculate the return signal....

Regards