ZP ACTIVE DAMPING and NOISE CANCELLATION

[moodz]

The ZP uses 3 samples an early X sample and then 2 samples A and B to resolve targets ( and cancel Earth Field ) ..... it can also sample a fourth variable ... the control loop voltage from the X samples.
The VRM ground is occupying the half volume space of the coil field ... the X sample is drives a negative feedback loop that tracks fairly much the field losses and keeps the coil optimally damped.
The A B samples pick up the remaining target .... for long responses the A and B pick up proportionally the same value.

[Mr.Jaick]

Yesterday I took the detector to the beach went up to the waters edge and found half a dozen bits of metal rubbish in 5 minutes ... so it works fine in salt water / sand.

that's some good news, however, I'm wondering how it works on wet salt because taking one sample at under 5us will produce a response for salt; are you overlapping the sample X and A or maybe A and B?
B should be the EF sample

Look at the PIC below ... the blue trace is no target .... the green trace is large ferrous VRM ground and the red trace is small gold coin near the coil ( 50 mm distance ).
Take particular note of the where the zero volt reference is the polarity of the responses and also how the voltage level does not return to zero volts.
Its should be clear that there is a clear signal distinction between target and ground :-)

Thank you for the photo that is much clear now.
now a few questions:
are the signals before the pivot point all X-signals? (the point where the signals converge)
after this pivot point do both signals become pure R responses or not?
the ground looks like it goes negative after the pivot.

[Mr.Jaick]

There are two magnetic responses: instantaneous and viscous. PI detectors normally ignore instantaneous magnetic responses (ideal ferrite) because it dies faster than the decay. Viscous magnetic responses are time-delayed decays so they are not ignored. But the decay curve is not exponential, it is t^-n where n is typically between 1 and 1.1. This is what the 2-point subtractive GB removes. I suspect that ZP damping doesn't change this at all.

I was pointing to the viscous one hence the name VRM.
I think it won't remove it from the equation but maybe hinder it to some degree.
can't wait to test it on the bench and see the waveforms.

Not necessarily. I use a 1k damping resistor in my probe design, otherwise the probe can ring a bit. That's because I'm using a 215MHz opamp.

yes as I thought; it rings but not as much as a voltage mode measurement.
thanks for the advice on the probe, I need to make a low-noise high bandwidth scope probe like yours, it's very useful.

[Mr.Jaick]

The A B samples pick up the remaining target .... for long responses the A and B pick up proportionally the same value.

I was just thinking about that, the B sample will hinder the detection of high conductors for any sample after the pivot point but improve it before the pivot point.

[moodz]

that's some good news, however, I'm wondering how it works on wet salt because taking one sample at under 5us will produce a response for salt; are you overlapping the sample X and A or maybe A and B?
B should be the EF sample

The X sample is taken early so the feedback loop "tracks out" the salt response. All the samples are fully adjustable in the code ...
A and B samples can be balanced around the pivot point. The pivot point is where the signal crosses 0 volts.
Feeding the A and B samples into a differencing integrator will produce A-B or B-A depending which samples go to which input

For example ..... if A samples before the pivot we get 1 volt and B samples after the pivot we get -1 volt .... because its A-B .... its 1 - ( -1 ) = 2 ... so the samples always add. ( for targets )

The EF always in the same sign ( before and after the pivot ) 1 -1 = 0 ... so it will cancel with the right sample length.


Thank you for the photo that is much clear now.
now a few questions:
are the signals before the pivot point all X-signals? (the point where the signals converge)
after this pivot point do both signals become pure R responses or not?
the ground looks like it goes negative after the pivot.
Are they all X signals ?? dunno .... have not done a enough research on this method ... have to remember we are looking at one RX cycle and the integrator will average out the A-B response.

If the ground changes very rapidly I suspect the feedback loops wont track. I have not played enough int field or with all the timings and component variables in the lab that could be done.

... unless there is a lump of reactive clay in benign ground .. but then the clay is a target and not ground.
I have dug up lots of gold nuggets made out of clay with my ML gear. :-)

[Mr.Jaick]

a lump of reactive clay in benign ground

one of the biggest problems right now, there is no tracking GEB that can track this, we've seen the ML best detectors can not track these quickly enough, or can not at all as the signal is massive.
I'm talking about pottery on castle tops and how they embedded many little bits and pieces of iron(sometimes big ones) in these clays to perhaps increase the strength or something.

Paul thank you for the time, work, and everything really.

alright, enough of me talking here let's get into the hardware mode!

[Dean Sarelius]

Yesterday I took the detector to the beach went up to the waters edge and found half a dozen bits of metal rubbish in 5 minutes ... so it works fine in salt water / sand.

As for your other concern ... a picture says a thousand words.

Look at the PIC below ... the blue trace is no target .... the green trace is large ferrous VRM ground and the red trace is small gold coin near the coil ( 50 mm distance ).
Take particular note of the where the zero volt reference is the polarity of the responses and also how the voltage level does not return to zero volts.

The coil is damped in 1.7 microseconds.

Its should be clear that there is a clear signal distinction between target and ground :-)

moodz

Hi Gents,
While I am still digesting this thread I wanted to let you all know that several years ago when I built some PI prototypes I noticed the exact pulse shape which Moodz is showing.
The very fact that what it appears is a negative phase shift was intriguing to me and made me think about the potential for discimination.
As you can see from the edited picture below there are two portions of the return signal that need to be considered Phase and Voltage.
The reason I suggested using a high speed MCU like the ESP32-S3 is beacuse of its 240MHz clock. Such a high speed clock provides 4.16 nano seconds of timer resolution.
So the idea is to use a high speed timer to sample the phase and a seperate ADC perhaps to sample the Amplitude.
I wonder if anyone else has had these thoughts also..?

[lucifer]

Hi Gents,
While I am still digesting this thread I wanted to let you all know that several years ago when I built some PI prototypes I noticed the exact pulse shape which Moodz is showing.
The very fact that what it appears is a negative phase shift was intriguing to me and made me think about the potential for discimination.
As you can see from the edited picture below there are two portions of the return signal that need to be considered Phase and Voltage.
The reason I suggested using a high speed MCU like the ESP32-S3 is beacuse of its 240MHz clock. Such a high speed clock provides 4.16 nano seconds of timer resolution.
So the idea is to use a high speed timer to sample the phase and a seperate ADC perhaps to sample the Amplitude.
I wonder if anyone else has had these thoughts also..?

You can achieve even picoseconds resolution if using a dedicated IC. For example I've used TDC7200, it's a time to digital converter with SPI, resolution of around 55ps, built-in averaging, etc. Very easy to interface to any MCU. Even with a standard PI works well and allows for some rough discrimination. However, I could not go lower than 1ns because of circuit noise. So a careful design is a must..

[Dean Sarelius]

Hi Lucifer,
Thanks for letting me know about the TOF chip TDC7200 looks very capable. I think I may abandon the ESP32 now as I wasnt aware of this chip and stick to using an AVR.
Cheers

[moodz]

Hi Gents,
While I am still digesting this thread I wanted to let you all know that several years ago when I built some PI prototypes I noticed the exact pulse shape which Moodz is showing.
The very fact that what it appears is a negative phase shift was intriguing to me and made me think about the potential for discimination.
As you can see from the edited picture below there are two portions of the return signal that need to be considered Phase and Voltage.
The reason I suggested using a high speed MCU like the ESP32-S3 is beacuse of its 240MHz clock. Such a high speed clock provides 4.16 nano seconds of timer resolution.
So the idea is to use a high speed timer to sample the phase and a seperate ADC perhaps to sample the Amplitude.
I wonder if anyone else has had these thoughts also..?

Hi Dean ... excellent idea however you realise that I was illustrating the "exclusive" view of responses. In reality you have the passive ground combined with the VRM ground and targets / hotrocks etc all combined together giving a combined response signal ...what is the base reference signal ??

The target you dont want and cant resolve is a very faint signal from say a meter into the ground and when you dig it up it is a beer can ( spent half a day doing just that ).

But there might be a solution ... story for another day ;-)

moodz

[Dean Sarelius]

Hi Moodz,
When I was evaluating the flyback I noticed that the phase was typically negative for gold but poisitve for iron so would assume that the beer can would provide you with a positive phase response not negative.
This is also evident in the response in green compared to the no response in blue.

[moodz]

Hi Moodz,
When I was evaluating the flyback I noticed that the phase was typically negative for gold but poisitve for iron so would assume that the beer can would provide you with a positive phase response not negative.
This is also evident in the response in green compared to the no response in blue.

What I meant with the weak target ( beercan / gold etc ) was that the shift would be so tiny for deep targets that I would have to dig half way towards it before getting a strong enough signal to resolve a shift.

moodz

[Dean Sarelius]

What I meant with the weak target ( beercan / gold etc ) was that the shift would be so tiny for deep targets that I would have to dig half way towards it before getting a strong enough signal to resolve a shift.

moodz

I also noticed studying the flyback the majority of the phase shift is at a much higher threshold voltage around 10 to 30V well above the regular 0.6V diode limited response of typical front ends.
In order to measure the phase I have been simulating a high speed high differential voltage comparator to be triggered at the higher threshold voltage to provide the phase shift data as well it provides for an auto threshold mechanism.