Oscillator power!!!!

[Geo]

Hi.
From time to time i hear that if i will increase the power of the oscillator i will make the detector to goes deeper!!!!.
Today i supplied the output transistors of my Sovereign with more voltage so the wave output to be about 50% more. I did not see any difference to the depth of the detector (air test). I want to learn if this method help to increase the depth inside the earth.
Any info?????

Regards

[Qiaozhi]

Hi.
From time to time i hear that if i will increase the power of the oscillator i will make the detector to goes deeper!!!!.
Today i supplied the output transistors of my Sovereign with more voltage so the wave output to be about 50% more. I did not see any difference to the depth of the detector (air test). I want to learn if this method help to increase the depth inside the earth.
Any info?????

Regards

Here is the root of the problem:
In order to double the detecting distance you will need to increase the TX power by 64x. This is because the received signal is proportional to the 6th power. So even doubling the TX power will give you very little depth increase. Also there will be problems caused by saturating the soil matrix and possibly overloading the pre-amp.

[Geo]

I remember when i builded the Anker SS60, it has about the same High oscillator power but better depth.
It had problems caused by saturating the soil matrix but only at rocks.
So..., you recommend me to close the box!!!

Regards

[mikebg]

Hi.
From time to time i hear that if i will increase the power of the oscillator i will make the detector to goes deeper!!!!.
Today i supplied the output transistors of my Sovereign with more voltage so the wave output to be about 50% more. I did not see any difference to the depth of the detector (air test). I want to learn if this method help to increase the depth inside the earth.

Geo, the correct term is "TX power" because there are metal detectors where oscillator drives TX end stage.
When you increase TX power (ampereturns) 64 times, the AIR and GND signals also increase 64 times and you should decrease the gain of preamp to avoid its saturation. However if you suppress AIR and GND signals 64 times without increasing TX power, you can increase the gain of preamp 64 times or to a value limited by noise and EMI in input. This can by made by ABC (Automatic Balance Control) and AGC.
http://www.geotech1.com/forums/showt...eferrerid=2910
http://www.geotech1.com/forums/showt...eferrerid=2910

[maikl]

I remember when i builded the Anker SS60, it has about the same High oscillator power but better depth.
It had problems caused by saturating the soil matrix but only at rocks.
So..., you recommend me to close the box!!!

Regards

Hi Geo. "Anker SS60" has a good detection of the air test, but that proved to be the ground? Is it stable? At what depth in the soil to detect coin 25mm, larger items ...?

[Aziz]

The magic 64 fold myth

Hi friends,

it's really time to reveal the myth about this. The 64 fold applies only in the infinite distance to the coil (far distance region). In the near detection distance region (we usually have), the factor is less, which makes sense to increase the TX power.

If we have a coil with radius R and double the target distance from distance N to 2N, where N is the multiple times of R (N related to coil radius R), then the exact diminish factor DF is:

DF = ( (1+4*N²)/(1+N²) )³

Example:
If we have N=1, which means target distance doubling from R to 2R,
then the target response diminish factor is DF = 15.625. So the target response is diminished by this factor DF. For this distance region, we have to increase the TX power by a factor of 15.625 to be able to detect the same target at double distance (2R).

You will find for different distances the diminishing factors below in the table. Look, how fast the diminish factor convergates to the magic 64 fold.

Aziz

[Aziz]

Hi all,

want to follow a discussion on this topic?
here:
http://australianelectronicgoldprosp...hp?topic=587.0

(I suspect, you have to be a member of this forum to read the whole discussion. I recommend it.)

You can also see the evolution and the proof of the formula above. The formula is correct, as I have made multitude numerical magnetic field simulations with my coil software.

So, what's the important result?
1. If you have a big coil, then you gain from the tx power increase.
2. If you want to detect small targets with big coils, then you gain from the tx power increase.
Both cases tend to be a near detection distance problem and the 64 fold does not apply anymore.

This formula is only valid for signals at the RX coil. Any detector specific non-linearity effects (limitting, saturating, etc.) are not taken into account of course.


Aziz

[Qiaozhi]

If we have a coil with radius R and double the target distance from distance N to 2N, where N is the multiple times of R (N related to coil radius R), then the exact diminish factor DF is:

DF = ( (1+4*N²)/(1+N²) )³

Hi Aziz,

Can you show the derivation of this formula?

[Aziz]

Hi Q,

let's start with the following (BTW correct) formula:

"Thus the there-and-back signal is proportional to
1/(a²+z²)³,
a=coil radius, z=target distance from the coil"
(source: Metal Detector Basics and Theory, page 14, written by Bruce Candy)

f(z) = 1/(a²+z²)³

Who is able to make the proof with the formula?
(You need to follow the thread in the other forum.)
I have done it. It's so much trivial, that it will bore me now.

Aziz

[Qiaozhi]

Hi Q,

let's start with the following (BTW correct) formula:

"Thus the there-and-back signal is proportional to
1/(a²+z²)³,
a=coil radius, z=target distance from the coil"
(source: Metal Detector Basics and Theory, page 14, written by Bruce Candy)

f(z) = 1/(a²+z²)³

Who is able to make the proof with the formula?
(You need to follow the thread in the other forum.)
I have done it. It's so much trivial, that it will bore me now.

Aziz

According to Candy's document, Figure 6 shows the logarithm of receive signal strength for an 11” mono-loop with distance from the centre of the coil.
Note that the signal is close to being a 1/z6 law for distances greater than 40cm (that's less than 16" in old money).

I would not call 40cm as being in the far field.
So where is the 6th law myth?

[Aziz]

Hi Qiaozhi,

According to Candy's document, Figure 6 shows the logarithm of receive signal strength for an 11” mono-loop with distance from the centre of the coil.
Note that the signal is close to being a 1/z6 law for distances greater than 40cm (that's less than 16" in old money).

I would not call 40cm as being in the far field.
So where is the 6th law myth?

The figure 6 shows the relationship too. The 1/z^6 law for far distances would be a log distance of log10(64)=1.8 . As the right ends at 0, the half distance ends at approx. 1.8. Ok, a 64 fold law would have a signal strength offset (in log view) of 1.8 at double distances.

Now look at the pink line crossing 5. It's crossing the 4 at almost the double distance. But it's log offset is now 1 instead of 1.8. 1 offset would be a factor of 10.

The myth is, that the factor is not always 64 as supposed. It is a non-linear behaviour, which the 64 is only achieved in the infinite distance to the coil.

We can consider a near detection region below R or 2R (=diameter) distance. This is a matter of definition of course.

Aziz

[Aziz]

Just an example:

We have a 20" mono-loop coil, coil R=10".
We can detect a small target at 5" (0.5R distance).
We now want to detect it at double distance (=10" or R distance).

The required TX power factor is according to the formula:
DF = 4.096

This is definitely not 64 fold! Only 4 fold.
Here is the big difference!

Aziz

[Qiaozhi]

Just an example:

We have a 20" mono-loop coil, coil R=10".
We can detect a small target at 5" (0.5R distance).
We now want to detect it at double distance (=10" or R distance).

The required TX power factor is according to the formula:
DF = 4.096

This is definitely not 64 fold! Only 4 fould.
Here is the big difference!

Aziz

OK that may be true, but most detectors (even the cheap ones) can detect a coin which is at a distance of at least the diameter of the coil. So using a 10" coil (for example) you would expect a coin to be detected at 10" minimum. In this case most users who are thinking about increasing detection distance are thinking beyond the R/2 distance shown in your example. Then they are surprised when doubling the TX power only provides a tiny increase in depth.

I don't think anyone was thinking that the 6th power law was an exact (or fixed) figure, and I'm not surprised that it has a non-linear behavior. A more accurate calculation can be obtained by using the Biot-Savart law, from which Bruce Candy's calculations are derived.

[Aziz]

OK that may be true, but most detectors (even the cheap ones) can detect a coin which is at a distance of at least the diameter of the coil. So using a 10" coil (for example) you would expect a coin to be detected at 10" minimum. In this case most users who are thinking about increasing detection distance are thinking beyond the R/2 distance shown in your example. Then they are surprised when doubling the TX power only provides a tiny increase in depth.

I don't think anyone was thinking that the 6th power law was an exact (or fixed) figure, and I'm not surprised that it has a non-linear behavior. A more accurate calculation can be obtained by using the Biot-Savart law, from which Bruce Candy's calculations are derived.

Yes, I made the Biot-Savart Law proof.

But a lot of people are prospecting for tiny gold nuggets. They often detect it at the near distance region.

BTW, the distance R could be the near/far distance region limit.
According to the f(z) (see above), the a² and z² term should be focused. On z<a, only a² is dominating and on z>a, z² is dominating. Taking the distance R as the near/far distance limit is a good idea I think.

Aziz

BTW, the edit/correction time-out is too short. I could not edit my post.

[mikebg]

The formula is correct, as I have made multitude numerical magnetic field simulations with my coil software.

So, what's the important result?
1. If you have a big coil, then you gain from the tx power increase.
2. If you want to detect small targets with big coils, then you gain from the tx power increase.
Both cases tend to be a near detection distance problem and the 64 fold does not apply anymore.

This formula is only valid for signals at the RX coil.

Instead simulation with software, you can make a target size coil and real experiments with it. Then you will understand that at small targets, the receiving path in practice appears in very far distance region. Most important is diameter of eddy current loop in target. Both important results mentioned above are valid for all sizes monocoils. Should be expressed what means term "gain". Even at zero GND signal, increasung ampereturns lead to increased AIR signal and you should reduce RX amplification to avoid its saturation. If RX sensitivity is limited by thermal noise, then increasing TX power will increase SNR. However at metal detecting AIR and GND signals limit the gain of preamp.
The attached image shows experiment of signal path from target to RX coil. To experiment signal path from TX coil to target, you should connect TX coil to sine generator and target size coil - to indicator.