New Version of Bipolar PI MD with GB and DD Coil

[Detectorist#1]

No! These two peek voltages are in totally different times - one is during of the high level of the "Phase" signal multiplied with TX pulse . Other peek voltage is during of the high level of the "Phase #" signal multiplied with TX pulse.

[vasilink]

Hi everyone,

Has anyone successfully replicated the circuit from the "New version of bipolar PI MD with GB and DD coil" thread? I’m trying to evaluate its reliability before building.

Also, does anyone have oscilloscope screenshots or waveform data showing the signals and the timing of the measurement strobes? That kind of information would be extremely helpful

[Detectorist#1]

Hi vasilink,
This scheme was proposed before 3 days. Do you think someone is able to replicate and to test the circuit for this time?

[Detectorist#1]

Hi vasilink,
Attached is information for timings of the control and strobe signals of this project. In the last version, only the duration of the GB strobe is increased from 40us to 44us for the moving of the GB point of the GB POT more near to the mid-point for ferrite target.

[vasilink]

Thank you very much for your quick and informative replies, and for sharing the timing diagram — that’s extremely helpful!

I’m new to the forum and still learning, so I really appreciate the support and insights from experienced members like you. Looking forward to exploring this project further and contributing as I go.

[vasilink]

I’d like to share a bit of theory and ask for your feedback to better understand the core principle behind this circuit. Please correct me if I’m wrong.

From what I gather, the discrimination mechanism is based on the difference in responses from magnetic targets under symmetrical bipolar excitation (+ and – phases). Ferromagnetic objects (like iron) tend to produce asymmetric responses due to hysteresis effects, while non-ferrous metals (like copper or gold) generate symmetric responses that cancel out when comparing both phases.

As for ground balancing, it seems to rely on subtracting the second strobe (40 µs) from the first one (10–12 µs), effectively isolating the ground signal and minimizing its influence on target detection.

I’d appreciate any clarification or corrections from more experienced members. Thanks in advance!

[Detectorist#1]

Hi vasilink,
In this project, "Fe" discrimination is based on the fact that 'Fe" targets generates secondary magnetic field under influence of the primary magnetic field of the Earth. This secondary magnetic field generates electric current (signal) in the two moving coils of the sensor according the low for electromagnetic induction. The second (equal) coil and the differential primary amplifier are necessary for eliminate of the influence of the voltages inducted from primary magnetic field of the Earth and for eliminate of influence of relatively near sources of 50(60)Hz electromagnetic fields. As for ground balancing - your explanation is right.

[pito]

and timing on TC4428 outputs A and B ?
looks like they are the same

[Detectorist#1]

Hi pito,
The "A" output of TC4428 is with negative (-8V) output during TX pulse and controls the gate of P-ch MOSFET.
The "B" output of TC4428 is with positive (+8V) output during TX pulse and controls the gate of N-ch MOSFET.
Pulse on "A" output is generated during the high level of the signal "Phase".
Pulse on "B" output is generated during the high level of the signal "Phase #".
You forgot to have in mind that the signals "Phase" and "Phase #" are multiplied (logical AND function) with the TX pulse by the two 1N4148 diodes and the two resistors in the inputs "A" and "B" of TC4428.

[pito]

google complementary pair mosfet
This bipolar it is actually complementary pair amplifier, for square waves dead time is required , for sin wave not.

[Detectorist#1]

Hi pito,
The case with bipolar PI with N-ch and P-ch MOSFET switches is very different from the case for motor control. The "dead time" is appr. half of the TX period in bipolar PI design!

[nejdet]

Hi Detectorist#1, we are eagerly awaiting a video with the final version of this project. Greetings!

[pito]

[Detectorist#1]

Hi pito,
The right work of the bipolar PI MD is exact as the yellow (coil's voltage) , red (gate of N-ch) and green (gate of P-ch) traces on your picture. Yes, this detector is acting as with two separate pulses in two opposite directions. This is clear explained in ITMD3 book.

[PahWiz]

Hi Detectorist#1
The input is a 2-cell 18650 Li-ion battery. Each cell’s full-charge voltage is about 4.2 V, empty around 3.5 V, and the nominal voltage is 3.7 V. In the power supply, U2 (MC34063AP) is used as a voltage booster (step-up). The input voltage is labeled 7.4 V, but in reality, it ranges from about 8.4 V when fully charged down to about 7.0 V when empty, and the output is set to 8.0 V. When the batteries are full (Vin = 8.4 V), this cannot operate as a boost converter anymore. A quick and easy fix is to increase the boost output to +9.0 V or +10.0 V (±9 V or ±10 V).





There is another error in the Mainboard schematic (bottom right corner): +8 V is connected to −8 V through L1 (330 µH). It looks like an LC filter, but it’s connected to the wrong power rail.