GROUND CONTROL PI CIRCUIT

[mickstv]

An alternative method to amplifying the result of subtracting S3 from S2, is to widen the GB and EF sample pulse widths to increase the gain. However, the variation of gain with sample pulse width is non-linear, making this method even more difficult to implement.

So you can use S1+S3-S2. With the sample width of S1+S3 being the same sample width as S2.

i.e.

S1 = 10
S3 = 100
S2 = 90

[Waikiki_Sweep]

Fisher CZ is VLF with square wave TX.
PI usually has one polarity. Anyway we have to compose something new.
So why not to use that two mosfets - two capacitors driver.
Just replace 100uF capacitors with 1000uF and make some schematic to drive mosfets separatelly to make that short + and - impulses and proper delays.

[mickstv]

Fisher CZ is VLF with square wave TX.
PI usually has one polarity. Anyway we have to compose something new.
So why not to use that two mosfets - two capacitors driver.
Just replace 100uF capacitors with 1000uF and make some schematic to drive mosfets separatelly to make that short + and - impulses and proper delays.

It will still only be a VLF. The only options are H-bridge using four Mosfets, or use a centretapped coil with two Mosfets to generate the bipolar TX.

I also experimented with using P and Nch Mosfets and a virtual ground into a mono coil, but the differences in internal capacitance and resistance between the P and N Mosfets mean't that you couldn't get the first sample in less than about 12us, so a waste of time.

[Waikiki_Sweep]

May be four mosfets is a solution. Just twice more smoke and holes in PCB if it shored once

Probably make a separate RX coil will help to keep that schematic in reasonable size.

Also in most cases mosfet capacitance is not a problem if you put fast diode like HER208 in series with mosfet.

[mickstv]

Putting a diode in line with the Mosfet and coil, might help slightly with speed but it won't help with the difference in Coss between a Pch and Nch Mosfet.

Another problem with bipolar on a monocoil is you cant use blocking fets to gate the RX to the pre amp, so your stuck with using the typical resistor and diodes method.

Using a separate RX coil, will reduce depth.

[Qiaozhi]

So you can use S1+S3-S2. With the sample width of S1+S3 being the same sample width as S2.

i.e.

S1 = 10
S3 = 100
S2 = 90

It's more complicated than that.
S1 = main sample, S2 = GB sample, and S3 = EF sample
The value of S1 (for example) is the amplitude of the main sample, not the main sample pulse width.

[Waikiki_Sweep]

Using a separate RX coil, will reduce depth.

Yes. In air. In real ground separate RX may even make it deeper depending on ground conditions and noise.
Some very good commercial PI detectors using separate RX like Garrett Infinum LS or Sea hunter Mark II and others.
And last one most advanced and most expensive from Garrett (Garrett ATX $2100) using very tricky looking DD coil:

[mickstv]

The value of S1 (for example) is the amplitude of the main sample, not the main sample pulse width.

Scratching head re this, not following.


I have a detector built that seems to GB fairly well over mineralized ground. It uses a timing sequence similar to what I had posted previously.

EF works because S1 and S3 are combined into one integrator, and S2 GB is inverted and combined with the integrator as well.

[mickstv]

So you can use S1+S3-S2. With the sample width of S1+S3 being the same sample width as S2.

i.e.

S1 = 10
S3 = 100
S2 = 90

Correcting a mistake. Values only approx


S1 = 10 Target
S2 = 100 GB
S3 = 90 EF

[Qiaozhi]

Scratching head re this, not following.


I have a detector built that seems to GB fairly well over mineralized ground. It uses a timing sequence similar to what I had posted previously.

EF works because S1 and S3 are combined into one integrator, and S2 GB is inverted and combined with the integrator as well.

In your previous post you quoted your equation as: x = S1 + S3 - S2

Now I can see what you're actually doing in practice, I assume you must be amplifying the GB sample (S2) prior to being combined with the integrator, otherwise it wouldn't work at all. So that your equation is in fact:

x = S1 + S3 - A(S2) ..... Eq.1 (ignoring the gain of the integrator)

When the gain (A) is adjusted, ground balance can certainly be achieved to some degree, but it does not give the same result as:

x = A1(S1 - S3) - A2(S2 - S3) ..... Eq.2

The problem with your solution is that x does not remain at zero for all values of the Earth field.

Please let me explain with an example:

TS1 (target signal at sample point 1 [main sample]) = 10
TS2 (target signal at sample point 2 [gb sample]) = 2
EF (Earth field signal) = 2
S3 (signal at sample point 3) = 2 (the Earth field signal)
S1 (main sample) = TS1 + EF = 10 + 2 = 12
S2 (gb sample) = TS2 + EF = 2 + 2 = 4
A1 (gain of 1st integrator/subtractor) = 1
A2 (gain of gb sample) = adjusted by user to achieve ground balance
x (result of equation) = 0 at ground balance point

Re-arranging equation 2:

A2 = (A1(S1 - S3) - x) / (S2 - S3)
A2 = (1(12 - 2) - 0) / (4 - 2) = 10 / 2 = 5
i.e. gb gain (A2) = 5

Checking:

x = 1(12 - 2) - 5(4 - 2) = 10 - 10 = 0 ..... correct (ground balance achieved)

Now, let's re-arrange equation 1:

A = (S1 + S3 - x) / S2
A = (12 + 2 - 0) / 4 = 14 / 4 = 3.5
i.e. gb gain = 3.5

Checking:

x = 12 + 2 - 3.5(4) = 14 - 14 = 0 ..... correct (ground balance achieved)

If the ground matrix is fairly homogenous, and the technique proposed is correct, then the only way the detector will provide a signal is if a target with a time constant different to the ground passes under the coil. However, what happens if the Earth field signal changes? This can easily happen if you swing the coil a little faster or slower, or the magnetic content of the soil changes.

TS1 (target signal at sample point 1 [main sample]) = 10
TS2 (target signal at sample point 2 [gb sample]) = 2
S3 (signal at sample point 3) = 4 (increased from original value of 2)
S1 (main sample) = TS1 + EF = 10 + 2 = 14
S2 (gb sample) = TS2 + EF = 2 + 2 = 6

First, let's try equation 2 with the increased value for S3.

x = 1(14 - 4) - 5(6 - 4) = 10 - 10 = 0 ... correct (no change for different EF)

But what about equation 1?

x = 14 + 4 - 3.5(6) = 18 - 21 = -3 ... a non-zero result!

[mickstv]

Well to answer that....

I'm not a mathematical genius so I'll leave that all to the boffins. I just build circuits and experiment.

Remembering that the values given are only approx to what I've been testing with, all samples enter the integrator at the same gain. I have two GB controls for testing the first one controls the delay between the end of S1 and the start of S2, the second GB control sets the width of samples S2 and S3 . The TX pulse for the circuit is 140us.

[Qiaozhi]

Well to answer that....

I'm not a mathematical genius so I'll leave that all to the boffins. I just build circuits and experiment.

Remembering that the values given are only approx to what I've been testing with, all samples enter the integrator at the same gain. I have two GB controls for testing the first one controls the delay between the end of S1 and the start of S2, the second GB control sets the width of samples S2 and S3 . The TX pulse for the circuit is 140us.

OK - now I've got it.

Your equation is not as I assumed. In fact, it is:

x = A1(S1) + A3(S3) - A2(S2)

where A2 = A1 + A3

In your case, you're using different sample pulse widths to achieve / adjust the gain, but the result is the same.

As a test, with TS1 = 10 and TS2 = 2, the gb gain (A2) needs to be 5 in order to cancel the ground signal, so we have:

TS1 = 10
TS2 = 2
EF = 2
S1 = 12
S2 = 4
S3 = 2
A1 = 1
A2 = 5
A3 = ?

Let's calculate the required value of A3 (EF gain) to set x to zero (ground balanced).

A3 = (x - A1(S1) + A2(S2)) / S3
A3 = (0 - 1(12) + 5(4)) / 2
A3 = (-12 + 20) / 2 = 4

If you try various values for EF, the value for x is always zero, and the detector remains ground balanced.

So the big question is - why does this technique work in the same way as equation 2?

Here's the answer:

Starting with equation 1 : x = A1(S1) + A3(S3) - A2(S2)

Since A2 = A1 + A3, then: A3 = A2 - A1, and substituting for A3 gives:

x = A1(S1) + (A2 - A1)(S3) - A2(S2)
x = A1(S1) + A2(S3) - A1(S3) - A2(S2)

and now I bet you can see what's coming ...

x = A1(S1 - S3) - A2(S2 - S3) ........................ Viola! Eq.2

So, x = A1(S1) + A3(S3) - A2(S2) = A1(S1 - S3) - A2(S2 - S3)
i.e. Eq.1 = Eq.2

How did you come up with this alternative scheme without working the numbers?

[mickstv]

How did you come up with this alternative scheme without working the numbers?

I acquired a few buckets of mineralized ground from a nearby Goldfield, I know this isn't perfect but it was the easiest way for me to be able to test different timing sequences. The circuit isn't perfect, but being micro controlled all the timing is easly changed.

[Qiaozhi]

I acquired a few buckets of mineralized ground from a nearby Goldfield, I know this isn't perfect but it was the easiest way for me to be able to test different timing sequences. The circuit isn't perfect, but being micro controlled all the timing is easly changed.

I wondered if you were using a micro, as there are two variables that need adjusting (A2 and A3) in equation 1, as opposed to one variable (A2) in equation 2.

[Waikiki_Sweep]

I have tried to adjust my MD but in my high-rise building EMI interference is hell too strong. I was surprised how "8" shape coil canceling EMI interference to nothing and even canceling Earth magnetic field influence.