AMX TX

[KingJL]

The superD coil assembly sounds very interesting. What kind of RX frontend did you use?

Tony, Currently experimenting with direct sampling using low noise differential input buffer with 26db gain and Audio DAC... a work in progress. First issue I had was a suitable coil cable... now that is solved

Did you use any ground compensation?

Only ground compensation is in cpu code... call it DSP if you want, but I call it preprocessing before DSP.

Is there any way to simulate the SuperD coil?

Probably, but that is beyond my capability... I find it difficult to simulate any IB coil setup with the simulator(s) that I have available.

Adding to the AFE mentioned above: I am following Carl's approach to the AFE and subsequent processing closely. I like the "current mode" configuration and I have always been a proponent of the AD797 (A few years ago, I actually modified my HH2 implementation to use a two stage preamp with a AD797 @ 12x as the first stage... I piggy backed 2 boards to implement the 2 stage preamp and supply the extra supply current that AD797 needed).

[Carl-NC]

But I wonder; why use of driver?
Can't logic level fet be used without driver?
IRLZ44NN comes to mind, I have those...

The IRLZ44N has a breakdown of 55V. You can use this (and maybe drive it directly from a micro, I don't know) but you will need to add capacitance and slow down the transitions to avoid avalanche. I'm using gate drivers to give me plenty of flexibility to use whatever FETs I want, even SiC. If you can't get gate drivers then make your own, this has been done countless times in other PI circuits.

[Carl-NC]

And to minimize any effects of a "tilt" keep the low side resistance to an absolute minimum... that is where the losses occur... if you need resistance for current sense, put in the high side where it does not involve the coil discharge current!

I don't understand this, can you explain?

[KingJL]

I don't understand this, can you explain?

"that is where the losses occur"... I probably should have said "that is where the regenerative losses occur". They are both losses (high side and low side). But the moodz circuit relies on the regenerative effects of the collapsing magnetic field of the coil to maintain the constant current. Just as an increase in coil R increases the loss of regenerative energy so does the resistance in the ground side of coil circuit (at regeneration time the high side path is isolated from the coil). It does little good to source out sub 100 mohm RDSon FETs (in my case; sub 10 mohm in moodz's case) and then put another 100 mohm in it's path. In the high side it is still a loss overall, but not on the regenerative component. "Tilt" can be caused by regenerative loss (usually IR losses) and change of effective inductance (there may be others factors also). The objective I think, is not to add to the losses and strive that they remain insignificant.

[ivconic]

The IRLZ44N has a breakdown of 55V. You can use this (and maybe drive it directly from a micro, I don't know) but you will need to add capacitance and slow down the transitions to avoid avalanche. I'm using gate drivers to give me plenty of flexibility to use whatever FETs I want, even SiC. If you can't get gate drivers then make your own, this has been done countless times in other PI circuits.

...comapring to IRF740 400v... 600N60 600v ... yes, of course.
stupid question...
Most of my questions like this one ; stem from not being able to get components easily and quickly. Sometimes even some trivial ones to get…
Oh yes, I can make simple gate drivers with transistors.
I have bunch of those xx00Nx00 salvaged from broken devices. Full "jar" of those.
But usually the labels are so compicated (various manufacturers put various prefixes) and I have to use glasses + magnifying glass to decypher the writtings and than to "google" it out for datasheet.
It is somewhat tedious "labor" , so that's why I leave it as last resource.

[ivconic]

"that is where the losses occur"... I probably should have said "that is where the regenerative losses occur". They are both losses (high side and low side). But the moodz circuit relies on the regenerative effects of the collapsing magnetic field of the coil to maintain the constant current. Just as an increase in coil R increases the loss of regenerative energy so does the resistance in the ground side of coil circuit (at regeneration time the high side path is isolated from the coil). It does little good to source out sub 100 mohm RDSon FETs (in my case; sub 10 mohm in moodz's case) and then put another 100 mohm in it's path. In the high side it is still a loss overall, but not on the regenerative component. "Tilt" can be caused by regenerative loss (usually IR losses) and change of effective inductance (there may be others factors also). The objective I think, is not to add to the losses and strive that they remain insignificant.

Please explain that to Carl a bit more clear... and later I will ask him ; he to explain it to me!

[ivconic]

You people ... (this did not sound good! oops!) keep mentioning current a lot,
I still don't understand what you want; constant current? (logical)
... or unstable, changing, shaking current!?


I saw in a couple of places that it figures about 500mA...
Here is a "poor man's solution" for both a voltage of 2V and a constant current of 500mA.
And since Carl likes to control everything through code (bosses are usually like that); I added that possibility too.
I'm sure Carl won't use this but I will, I'll give it a try and see what happens...

[moodz]

You people ... (this did not sound good! oops!) keep mentioning current a lot,
I still don't understand what you want; constant current? (logical)
... or unstable, changing, shaking current!?


I saw in a couple of places that it figures about 500mA...
Here is a "poor man's solution" for both a voltage of 2V and a constant current of 500mA.
And since Carl likes to control everything through code (bosses are usually like that); I added that possibility too.
I'm sure Carl won't use this but I will, I'll give it a try and see what happens...

good work ... there is a lot of spurious noise about the tilt and constant current .. fact is you make a detector with any sort of "moving" current.

[moodz]

PS .. if anyone can point out why tilt does matter AND point to a detector design ( not theories ) that uses tilt control I might reconsider that opinion ...
r
Did you try to change the energy absorption of the coil by passing a large target over it?
This is not a theory,.
A tiny variation of tilt generated by this type of change generates in turn a large signal offset change after all the receiver stages.
The variations of XMIT current tilt are not really disturbing but the large variations of signal offset at the receiver end are really disturbing the DSP.
Everything was OK while testing on the bench.
It is a surprise we got a few months ago on our real square wave project as soon as we went over a real ground.
It is not even when there is a large target going under the coil, it is also happening while the coil is swinging with its slight variations of height over the ground.
This behaviour is specific to the square wave system and does not appear on traditional triangular wave systems.

Now, if you still want to ignore this fact, go ahead, you will have the same surprise as us!!!

... I reckon your DSP algorithm is broken ... come to Australia and try the ground signals here

[moodz]

For more explanations, See post of Tony here:
https://www.geotech1.com/forums/foru...352#post411352

What I think is missing is a clear understanding of what causes tilt.
Tilt is caused by losses that the coil sees ( ie loads across the coil ... either coupled or physical parallel loads ) NOT series loads ... as series resistive load will reduce peak current but not cause tilt.
Measuring at the RX coil only complicates matters if you are trying to compensate tilt and I am not sure why you are referring to measuring peak flyback voltage ... those ccts cant do it. ( not Post #1 in AMX RX you referred )
CCPI from startup shown below 0.5 ohms series and 2 ohm series vs 1K parallel load on the coil for the two cases. Absolutely clear that a parallel load on the coil causes loss of flyback and tilting whereas a change in series resistance in line with the coil causes a change in peak current but no tilting.
moodz

[Carl-NC]

"that is where the losses occur"... I probably should have said "that is where the regenerative losses occur". They are both losses (high side and low side). But the moodz circuit relies on the regenerative effects of the collapsing magnetic field of the coil to maintain the constant current. Just as an increase in coil R increases the loss of regenerative energy so does the resistance in the ground side of coil circuit (at regeneration time the high side path is isolated from the coil). It does little good to source out sub 100 mohm RDSon FETs (in my case; sub 10 mohm in moodz's case) and then put another 100 mohm in it's path. In the high side it is still a loss overall, but not on the regenerative component. "Tilt" can be caused by regenerative loss (usually IR losses) and change of effective inductance (there may be others factors also). The objective I think, is not to add to the losses and strive that they remain insignificant.

Sorry, I still don't get it. Whether the sense resistor is high-side or low-side should make no difference to the transition energy losses. Either position just looks like battery resistance. The only difference might be that the low-side R very slightly alters the gm of Q5/Q6 but for 0.1Ω it's probably not noticeable.

[Carl-NC]

What I think is missing is a clear understanding of what causes tilt.
Tilt is caused by losses that the coil sees ( ie loads across the coil ... either coupled or physical parallel loads ) NOT series loads ... as series resistive load will reduce peak current but not cause tilt.

I'm not entirely sure if I understand what you're saying, but series resistance within the coil does affect tilt. During the transition, the currents that convert the coil's magnetic field energy to electric field energy flow through the coil's series resistance, and that causes a real power loss. It happens again within the same transition as the electric field energy converts back to magnetic field energy. So the magnetic field energy coming out of the transition always has to be less than it was going in. Yes, more series resistance will reduce peak current, but for the same peak current it will also increase tilt.

Transitional energy loss also has to increase with eddy targets or viscous ground, same as with any other TX circuit.

Addendum: A parallel resistor across the coil becomes a series coil resistance during the transition, when the isolation FETs are off. So if you put in a 1k resistor you will see a massive amount of tilt. A series R also affects the tilt, but you have to look more closely.

[detectormods]

What I think is missing is a clear understanding of what causes tilt.
Tilt is caused by losses that the coil sees ( ie loads across the coil ... either coupled or physical parallel loads ) NOT series loads ... as series resistive load will reduce peak current but not cause tilt.
Measuring at the RX coil only complicates matters if you are trying to compensate tilt and I am not sure why you are referring to measuring peak flyback voltage ... those ccts cant do it. ( not Post #1 in AMX RX you referred )
CCPI from startup shown below 0.5 ohms series and 2 ohm series vs 1K parallel load on the coil for the two cases. Absolutely clear that a parallel load on the coil causes loss of flyback and tilting whereas a change in series resistance in line with the coil causes a change in peak current but no tilting.
moodz

Is that a Noah's Ark for detectorists?

[Tinkerer]

What I think is missing is a clear understanding of what causes tilt.
Tilt is caused by losses that the coil sees ( ie loads across the coil ... either coupled or physical parallel loads ) NOT series loads ... as series resistive load will reduce peak current but not cause tilt.
Measuring at the RX coil only complicates matters if you are trying to compensate tilt and I am not sure why you are referring to measuring peak flyback voltage ... those ccts cant do it. ( not Post #1 in AMX RX you referred )
CCPI from startup shown below 0.5 ohms series and 2 ohm series vs 1K parallel load on the coil for the two cases. Absolutely clear that a parallel load on the coil causes loss of flyback and tilting whereas a change in series resistance in line with the coil causes a change in peak current but no tilting.
moodz

The circuit at Post #1 in AMX RX does indeed measure the changes in peak Flyback voltage with great accuracy. It can detect the changes caused by very minute targets.
Why precisely measure the changes in Flyback voltage?
Because the Flyback voltage is proportional to the Peak TX current. The Flyback voltage can therefore be seen as a proxy for the peak current.

[moodz]

The circuit at Post #1 in AMX RX does indeed measure the changes in peak Flyback voltage with great accuracy. It can detect the changes caused by very minute targets.
Why precisely measure the changes in Flyback voltage?
Because the Flyback voltage is proportional to the Peak TX current. The Flyback voltage can therefore be seen as a proxy for the peak current.

No that is completely wrong .... if I have a flyback of say 500.0 volts in the TX coil .. that circuit will not give a reading of 500.0 volts .... The RX coil responds to changes in magnetic field ( from anywhere ) ... how can it measure the peak flyback voltage in the TX coil ????