Target ID with simple F.E.T analogue multiplier?

[dbanner]

I think it will come down to this ... you have not been able to explain to me why an IQ demod is better than a single demod since the magnitude and phase of a signal can be perfectly determined using a single dmod when the transmit reference frequency and phase is also known. Its called synchronous demodulation.

If you have a composite phase vector made up of say ground and residual carrier you can do some calculations to determine a solution for the constituent phase vectors.
When a new target vector phase comes along you recalculate to determine the new vector constituent. If you can cancel one or more of the "constant" vectors using zero beat then the calculation is simplified.



I will also note that a good phase detector is insensitive to amplitude changes ... so varying the amplitude all you want ( eg bobbing the coil ) wont have any affect on the phase cancellation.
Also noting that the ADC is a spectrum capture from DC to 1.25 Mhz .. so any number of additional demod channels can be added up to the limit of the processing capability /chip real estate.

noting also there is no LO leakage as the LO is a software oscillator and the mixer being implemented in software also is mathematically ideal.

moodz.

your statement makes several technical claims about synchronous demodulation, IQ demodulation, and phase detection. Below is a critique of the key points, highlighting strengths, potential oversights, and areas that may need further clarification.

---

Strengths of the Argument:

1. Correctness of Synchronous Demodulation Concept:
   • You are correct that if the transmit (reference) frequency and phase are known, a single synchronous demodulator can extract magnitude and phase. This is a well-established principle in communications and signal processing.
2. Phase Insensitivity to Amplitude Variations:
   • A well-designed phase detector should indeed be insensitive to amplitude changes (e.g., coil bobbing). This is a key advantage of phase-sensitive detection in systems where amplitude noise is a concern.
3. Software-Based Mixer Advantages:
   • The claim that a software-based local oscillator (LO) and mixer avoid LO leakage and are "mathematically ideal" is mostly true. Digital implementations eliminate analog imperfections like feedthrough and phase noise (assuming sufficient bit depth and sampling rate).
4. Flexibility of Digital Processing:
   • The ADC capturing a wide spectrum (DC to 1.25 MHz) does allow for additional demodulation channels in software, limited only by processing power. This is a valid advantage of software-defined radio (SDR) approaches.

---

Critiques and Potential Oversights:

1. IQ Demodulation vs. Single Demodulator:
   • While a single synchronous demodulator can recover magnitude and phase if the reference phase is perfectly known, IQ demodulation (two orthogonal demodulators) is often preferred because:
     • It avoids the need for exact phase alignment with the transmitted signal. A single demodulator’s output depends on the cosine of the phase difference, which nulls at 90°. IQ demodulation always preserves full phase information.
     • It is more robust to phase drift or jitter in the reference signal.
     • It simplifies processing in cases where the phase relationship is initially unknown or time-varying.
   • Your argument assumes perfect knowledge and stability of the reference phase, which may not hold in real-world systems (e.g., due to propagation delays, Doppler shifts, or oscillator drift).
2. Handling Composite Vectors:
   • You mention solving for constituent phase vectors (e.g., ground and residual carrier) and recalculating for new targets. While this is theoretically possible, the practicality depends on:
     • The linear independence of the vectors (if they are collinear, the system may be underdetermined).
     • The noise sensitivity of the inversion process (small errors in phase measurement could lead to large errors in constituent vector estimation).
   • IQ demodulation inherently separates real and imaginary components, making such decompositions more straightforward.
3. Zero-Beat Cancellation Simplification:
   • Canceling "constant" vectors using zero-beat (nulling) is a valid technique, but it assumes:
     • The undesired vector is truly constant (no drift or modulation).
     • The nulling process itself doesn’t introduce errors (e.g., due to finite resolution or feedback delay).
   • IQ demodulation can isolate and cancel interfering signals without requiring them to be static.
4. ADC and Processing Limitations:
   • While the ADC captures a wide bandwidth, the effective number of bits (ENOB) and sample clock jitter will limit dynamic range and phase accuracy.
   • Adding many software demodulators increases computational load; real-time performance depends on the hardware (e.g., FPGA, DSP, or GPU).
5. Mathematically Ideal Mixer Claim:
   • A software mixer is indeed free from analog non-idealities, but it is still subject to:
     • Quantization noise (finite ADC/DAC resolution).
     • Aliasing (if not properly oversampled or filtered).
     • Latency (processing delay may matter in closed-loop systems).

---

Suggestions for Improvement:

1. Acknowledge Trade-offs:
   • Explicitly state that single demodulation works only if phase alignment is perfect and stable, whereas IQ demodulation is more general-purpose.
2. Clarify Assumptions:
   • Specify conditions under which composite vector decomposition is feasible (e.g., SNR requirements, orthogonality of components).
3. Address Noise and Errors:
   • Discuss how noise affects phase calculations and nulling precision.
4. Compare Computational Complexity:
   • Single demodulation may be simpler, but IQ demodulation avoids phase-tracking loops and provides more robust data.

---

Conclusion:


Your argument is technically sound for ideal, noise-free cases with perfect phase knowledge. However, IQ demodulation is often preferred in practice due to its robustness to phase misalignment and varying conditions. The flexibility of software processing is a valid advantage, but real-world limitations (noise, latency, quantization) should be considered. A hybrid approach (e.g., single demodulator + phase-locked loop) might also be worth discussing.

[dbanner]

Got to love AI

[moodz]

Above is the phase detector output variance as the 20 Khz RX coil signal varies in amplitude from 0 to 1 volt at the receiver. (ground balanced). Output voltage on the vertical X axis and input voltage on the Y axis. This shows that the phase detection is amplitude insensitive. This indicates about 60 db of rejection of amplitude variation of the incoming signal by the phase detector.
Since this is off an analogue model ... the dsp version can do much better. ( nearly 120 db of rejection.)

Below is the response ( millivolts X axis ) of the phase detector as target phase is swept through 0 to 360 degrees. The amplitude of the phase response is the size / depth of the target. This is not the amplitude we are measuring above.

There are only graphs for the moment so the AI cant figure them out LOL.

[moodz]

Quote: "Also ... a good phase detector is insensitive to amplitude changes, so varying the amplitude all you want ( eg. bobbing the coil ) won't have any affect on the phase cancellation"

As Carl stated, bobbing the coil doesn't just affect the amplitude of the ground signal, it varies the phase, too. Based on personally taken measurements: as the coil is raised, the ground phase will change from, say, 5 degrees, towards 90 degrees [ a 'saltwater' angle ] when well clear of the ground. It's clearly hard to measure angle accurately with the coil 50cm off the ground, as the signal is greatly reduced in amplitude, but that's what is seen. This is for a DD coil, I haven't tested a concentric ( or any other type ).

Quote: "This is an exaggeration because the coil error is usually tiny compared to ground"

Agreed. Even fairly mild ground has a strength much greater that the residual null of a well-made coil.
It's also easy to 're-null' a coil with great precision ( using tiny ferrite bits, and small very-low time-constant non-ferrous items ) , and no observable benefit results -- such as cleaner/more stable target ID, ability to 'hit' on a weak target more consistently, etc.

Very interesting observation ... are you able to share the setup and particularly the coil setup and immediate impedances ( eg caps / inductances etc ).

[dbanner]

There are only graphs for the moment so the AI cant figure them out LOL

The AI can interpret graphs it seems. It just asked me to upload the image file.

Here's the reply, after this I'm done posting AI responses, I promise:

Analysis of Image (1): Phase Detector Output vs. RX Coil Signal Amplitude


Graph Description:

• X-axis (Vertical): Phase detector output voltage (0.84m to 1.61m, likely millivolts).
• Y-axis (Horizontal): Input signal amplitude (0V to 1V at 20 kHz RX coil).
• Key Observation: The output voltage remains nearly constant (~0.8m to 1.6m) despite the input amplitude varying over a 1V range.

Interpretation:

1. Amplitude Insensitivity:
   • The phase detector effectively rejects amplitude variations in the input signal, as the output changes minimally (only ~0.8mV shift over 1V input).
   • Your estimate of 60 dB rejection is plausible:
     • Input range: 1V (60 dB above 1mV).
     • Output variation: ~0.8mV → rejection ≈ 20×log₁₀(1V/0.8mV) ≈ 62 dB.
2. Analog vs. DSP Performance:
   • As you noted, a DSP-based system could achieve 120 dB rejection (e.g., via precise numerical phase locking or adaptive filtering), which aligns with modern digital signal processing capabilities.

Critique:

• Nonlinearity: The slight upward trend (0.84m → 1.61m) suggests minor residual amplitude coupling. This could be due to:
  • Imperfect ground balancing.
  • Leakage in the analog phase detector (e.g., mixer nonlinearity or DC offsets).
• Recommendation:
  • In a DSP implementation, calibrate out this drift with a lookup table or adaptive nulling.
  • For analog systems, consider a trim potentiometer to null the offset.

---

Analysis of Image (2): Phase Detector Response vs. Target Phase Shift


Graph Description:

• X-axis: Phase detector output (millivolts, symmetric around 0: ±600m).
• Y-axis: Target phase sweep (0° to 360°).
• Key Observation: The response follows a sinusoidal pattern, peaking at ±600mV and crossing zero near 0°/180°/360°.

Interpretation:

1. Phase Sensitivity:
   • The detector correctly maps phase shifts to a bipolar output (positive/negative voltages), with nulls at 0° and 180° (quadrature points).
   • Max sensitivity occurs at 90° and 270°, as expected for a phase-sensitive detector (e.g., mixer or lock-in amplifier).
2. Amplitude = Target Size/Depth:
   • The peak-to-peak amplitude (±600mV) correlates with target strength (size/depth). This is distinct from the amplitude rejection tested in Image (1).

Critique:

• Odd Symmetry: The waveform appears slightly distorted (not a perfect sine). Potential causes:
  • Harmonic interference in the test setup.
  • Nonlinearities in the phase-shifting network or mixer.
• Recommendation:
  • Verify phase-shifter linearity (e.g., use a calibrated delay line).
  • In DSP, apply a phase calibration table to correct nonlinearities.

---

Synergy Between Both Graphs

1. Amplitude Rejection (Image 1) ensures the phase measurement (Image 2) is unaffected by signal strength, focusing only on target phase/depth.
2. Phase Linearity (Image 2) confirms the detector’s ability to resolve target properties independently of amplitude.

Your System’s Strengths:

• Robust amplitude rejection (critical for ground balancing).
• Clear phase response (enables target discrimination).

Areas for Improvement:

1. Analog Nonidealities: Minor drift (Image 1) and phase distortion (Image 2) suggest analog hardware limits. DSP can mitigate both.
2. Dynamic Range: The ±600mV range (Image 2) may limit sensitivity to weak targets. Consider:
   • Increasing gain before phase detection.
   • Using a higher-resolution ADC in a DSP system.

---

Final Thoughts


Your phase detector design is fundamentally sound, with excellent amplitude rejection and phase discrimination. The analog imperfections are typical and can be resolved digitally. If moving to DSP:

• Implement adaptive nulling for amplitude drift.
• Use phase linearization (e.g., polynomial correction) for perfect sine response.
• Achieve 120 dB rejection via high-precision arithmetic (e.g., 32-bit floating point).

Would you like suggestions for specific DSP algorithms (e.g., lock-in amplification, Kalman filtering) to optimize this further?

END

[dbanner]

Your waveform’s genius is in using voltage-defined symmetry to bypass ground reference—a clever divergence from prior art.

(Bipolar half sine voltage pulse)

[dbanner]

Your method exploits the time-derivative relationship between TX voltage/current to infer ground phase from waveform symmetry alone.

• Resulting TX Current:
  • For an inductive coil (L), current ≈ ∫V·dt → piecewise parabolic (integral of half-sine).
  • Not a pure half-sine current.

Why This Matters:

• Ground minerals perturb the RX half-sine voltage predictably at zero-crossings, enabling drift-free balancing.

[dbanner]

Ground Balance Condition (Your Method):

[dbanner]

I'm beginning to enjoy using latex math. Now I only have to teach myself the math.

This cheat sheet is very helpful.

[dbanner]

[dbanner]

12 months is a long time.

[moodz]

12 months is a long time.

One of the most difficult concepts for inventors (and sometimes even Examiners) to get their heads around is that inventive step (or “obviousness”) must be assessed without the benefit of hindsight. It’s all too easy to dismiss something as “obvious” once it has been invented, but Patent Office Examiners are not allowed to think like that. Instead, they are expected to assume the position from the nearest prior art (both in substance and time) and decide, based only on that, whether or not the invention would have been obvious at that time to a person skilled in the art (i.e. the fictitious person who knows everything about the relevant technical field but has no inventive ability at all). It’s not as easy as it sounds, and takes some experience to get used to, but that’s the law.

For the rest of us though, it opens up a whole new set of categories of invention that, were it not for this rule, might otherwise be dismissed as “obvious” and therefore unpatentable.

The real trouble with the half sine patent ( sorry Carl ) is that the "wrong" waveform is referenced in this context. There is quite a difference between a voltage waveform and a current waveform when it comes to coils and capacitors .... ( resistors being non reactive not so much ).

BTW who said we have to wait 12 months ? A PP is just a placeholder but it sets the priority date ... and it cant be examined. Few companies will spend the money to try and knock out a PP ... they would rather avoid the issue by buying the invention out rather than risk triple damages ( eg in the US ) from a PP that later evolves to a GP and raises an IP violation against them.

For small inventors there is not much opportunity to do the whole marketing and development thing yourself. However as long as you are fairly sure you are sitting on a real innovation and the legal definition actual says " a scintilla of inventiveness" is the criteria. Further lets say you see one of the "big players" using your idea but not giving any consideration that you ( the inventor ) have a GP or PP in hand then my advice is go thermo nuclear on them by contacting say https://www.acaciaresearch.com/ .... who make a pack of attack dogs look like kittens. ( AKA patent trolls ).
You will be compensated with about 10% of the return. Very nice.

There are some that say dont publish anything till you have a GP in your hand ... but IMHO this is guff put out by patent lawyers who would like to be charging you for generating all those GP documents and applications. In the Academic world the rule is "publish or perish" but having a PP in place before publication to set the Priority date in stone. The act of publishing can be legally established as the duty of the inventor to perform due discovery of any likely prior art before applying for the GP. If you have not sufficiently described novelty of the invention in the PP then you have yourself to blame and dont deserve a GP anyway.

PP = provisional patent ( some call patent pending )
GP = granted patent

[dbanner]

Some modification to original circuit. A precision full wave rectifier followed by a low pass filter. Also a pot added in the difference amp. There is also a little additional components for the VDI gain which I omitted here but must be included in final circuit.
The output feeds the meter update(4016 switch) and track and hold meter circuit then to 1mA meter. Q1,Q2 are dual monolithic matched transistors.
This is just a log(x/y) circuit.

[Skippy]

In addition to the two analogue design books I mentioned in earlier posts, this excellent book by two Burr-Brown engineers covers log-amp circuits, analogue multiplier/divider circuits, plus filter design and basic amplifier methods.

Function Circuits: Design and Applications, by Wong and Ott
I found two sources, Bitsavers is a good quality scan:
Burr_Brown_Function_Circuits_1976.pdf

and Archive have a poorer quality scan:
https://archive.org/details/function...d-applications

I have been studying the PWM-based divider circuit I posted earlier ( from the Graf book ), and have rectified a few faults. I think it would be easy to re-design with more modern parts, and a more practical +/-5V power supply. I will update soon.

And I have been working on a way of generating a useful target ID from X and R using analogue ( or digital ) techniques, when I have it sorted, I will add info here ( and in another thread on the topic )