VLF MD with digital signal processing : Bee-Buzz 1

[Atul Asthana]

Great conversation, I am learning

which low cost 24 bit codec would be suitable in this design?
and which direct sampling design should I study?

[moodz]

The RX signal in the coil already contains some noise, which gets amplified in the first op-amp, along with the addition of the op-amp's own noise. The critical factor is the required SNR at the signal extraction stage to recover a reliable signal. To determine the allowable op-amp noise, you must work backward, accounting for all noise sources in the chain, ensuring the first op-amp's noise is low enough to maintain the target SNR.

Thumbrule : get the opamp with the lowest possible noise, at your frequency of interest, that you can afford.

The opamps you mentioned, are likely to have been used in this Garret detector.

Your proposed detector is a single frequency @ 6 Khz ... Op Amp data sheets quote broadband noise and its usually worst in the first DC to 1 Khz ... far away from 6 Khz. No need to spend money on expensive opamps based on CE ( Catalogue Engineering )

The only way that low frequency noise in the ( cheaper ) opamps is going to get into your 6Khz signal is if there is non linearity or overload ( ie convolution ) occuring through bad design.
Do you reckon they did moonshots back in the 60s with these uber opamps ... no they didnt.
moodz

[Repwoc]

thanks for bringing up this point.

zero crossing + peak detection is very susceptible to noise,

OK, thanks for the explanation.

Looks like I need to do more experiments.

[Detectorist#1]

I have modest experience in building of military radars and they have a very rapid change in the level of the received signal with distance as with metal detectors. There, logarithmic processing of the received signal is used. Trying to use linear processing of the received signal with multi-resolution analog-to-digital conversion leads to serious problems in my opinion.

[Altra]

Yes, I didn't think it would work with direct sampling.

If you google search "stacking a/d converters " there is some information pro and cons.

[Detectorist#1]

Altra also not knows the final solution of the proposed project. This task is more complex than AMX project where no requirements for the final price. Proposed aim from Altra is for good and low price solution at the same time. Not easy task!

[Carl-NC]

I didn't think either one used paralleled ADC inputs. That's most useful with a true multi-channel simultaneous-sampling ADC, which the micro ADC is not.

[Detectorist#1]

Maybe Carl knows the true in this situation.

[Altra]

While you are on the AT and Ace series. They use parallel adc inputs to increase the dynamic range. Not sure if this idea could be applied to direct sampling?

[Detectorist#1]

Thank you Carl. Also the opinion of Atul for this calculation will be interesting.

[Carl-NC]

Yes, Garrett has the NXP factory custom-label their micros.

[ivconic]

The AT Pro is not a direct sampling design, it uses analog demods. Micro is likely a Kinetis Cortex M0 @ 48MHz. ADC is 16b. The TX is driven by a relaxation oscillator using an LM386.

At ACE series too.


I always wanted to try it on some of my diys, yet never did it.





I also have .txt on my disk containing the inscription, can't remeber what's this and why am i keeping it.
Could it be the inscription on the mcu?

GARRETT
8155300 E OM48Z
CTYHBK15320
GARRETT 8155300 E 0M48Z

[Carl-NC]

How to calculate the real value of the voltage noise of OpAmp allowed to use for first Rx stage in VLF MD?

Suppose you want to direct-sample with an ADC with 16 ENOBs and a 3V range. The SNR is 98dB; the max signal is 3vpp or 1.06v rms so the noise floor is at 13.7uv rms. Let's say you have a preamp gain of 100; now the input-referred noise is 137nv rms. Let's also say that the eventual signal BW is 50Hz; this makes the input-referred noise density 19.3 nV/rtHz. But you want the preamp noise to be lower than this, by at least 3dB, so that becomes 13.7 nV/rtHz. This will be dominated by the opamp voltage noise and the input resistor noise.

[Carl-NC]

The AT Pro is not a direct sampling design, it uses analog demods. Micro is likely a Kinetis Cortex M0 @ 48MHz. ADC is 16b. The TX is driven by a relaxation oscillator using an LM386.

[Atul Asthana]

Hi Atul,

It is interesting how low noise OpAmp needs VLF MD for first Rx stage. You mention using of TL072 ( 18nV sqr Hz at 1KHz) and OPA2134 (8nV sqr Hz at 1KHz) is possible. How to calculate the real value of the voltage noise of OpAmp allowed to use for first Rx stage in VLF MD?

The RX signal in the coil already contains some noise, which gets amplified in the first op-amp, along with the addition of the op-amp's own noise. The critical factor is the required SNR at the signal extraction stage to recover a reliable signal. To determine the allowable op-amp noise, you must work backward, accounting for all noise sources in the chain, ensuring the first op-amp's noise is low enough to maintain the target SNR.

Thumbrule : get the opamp with the lowest possible noise, at your frequency of interest, that you can afford.

The opamps you mentioned, are likely to have been used in this Garret detector.