Sine wave multi frequency

[CrizzyD]

Just something to add that hasn't been mentioned yet. There's been some research conducted using MLC (Multi Level Chirp) signals that claim to increase sensitivity up to 17% as compared to regular balance coil methods. https://ieeexplore.ieee.org/document/9767474

[Olly]

Just something to add that hasn't been mentioned yet. There's been some research conducted using MLC (Multi Level Chirp) signals that claim to increase sensitivity up to 17% as compared to regular balance coil methods. https://ieeexplore.ieee.org/document/9767474

Sadly I've let my membership to the IEEE lapse so can't log in to read the full document.
The swept-sine (chirp) technique is failrly widely used in audio when analysing loudspeaker room responses.
It might be interesting to generate some chirp waveforms for the TX and then analyse the received signal of various objects at different depths.
We could also play around with mult-levels in the chirp waveform, perhaps ramping amplitude up from the lower frequency start of the chirp.

[moodz]

The problem with chirp is like trying to measure the velocity of a continuously accelerating object ...the phase is continuously changing due to the frequency changing so it is somewhat harder to process the target phase change information. Typically the coil is also moving so this just complicates things. Contrary to popular opinion targets are not highly frequency selective unless you are searching for tuned circuits.

[Olly]

The problem with chirp is like trying to measure the velocity of a continuously accelerating object ...the phase is continuously changing due to the frequency changing so it is somewhat harder to process the target phase change information. Typically the coil is also moving so this just complicates things. Contrary to popular opinion targets are not highly frequency selective unless you are searching for tuned circuits.

Yeah, I kind of agree with you Moodz - That's an extra level of complication that I don't need at the moment.
I think I'll keep the chirp experiments till later.

[moodz]

I am not knocking the quoted article but the claimed benefit of chirp seems to be on the basis of simulation results. Personally I like simulation but not sure it should be the foundation of an engineering paper without empirical results to back it up.

[Aziz]

Hi all,

it doesn't make sense to use a wide band chirp. But a very narrow bandwidth chirp in a resonant TX-coil configuration, which I am going to test it. The impedance measurement of the TX-coil delivers enough information for GB algorithm. And there will be the RX-coil for target signal extraction.
Why resonant TX-tank? The reactance must get 0 or very low, so there is enough current through the TX-coil.
Continious IB systems lack of producing very very large reactive response compared to PI systems. So we need new techniques to sample reactive response only (at least with minimum target response). The impedance change measurement of the TX-coil will solve this issue.

Aziz

[Olly]

In order to save some pins from the FPGA, I decided to try a PDM output instead of a 16-bit R2R.
So I've written a 2nd order PDM modulator in Verilog and am clocking the PDM output at 98.304 MHz.
(Equivalent to a decimation ratio of 128x at 768kHz)
The reconstruction filter is a simple RC low pass and this works very well even with output sine waves of > 100kHz.

[moodz]

Good job ... I used one for the target audio VCO. An NCO followed by PDM.