Target response tester

[green]

I measured 10 nickels on the conductivity meter ranging in dates from 1964 - 1997 and they all read 5.xx% IACS. The lowest is 5.22 and the highest 5.63%. Doing a quick test for TC the difference appears to account for about 1uS between the two. Further tests still to to be done on the quarter.

Eric.

What is the peak Tx current and time when you test for decay? Wondering what we are doing different that I get a tau around 10us and you get around 8us for the nickel.

[Ferric Toes]

Another test. When I tested the 19mm length wires the other day I noticed orientation didn't seem to make much difference. Tested 3 wire sizes and a US nickel. Can get a lot different slopes with the nickel. Wondering if you see much change with 19mm of copper wire with the VLF.

I measured 10 nickels on the conductivity meter ranging in dates from 1964 - 1997 and they all read 5.xx% IACS. The lowest is 5.22 and the highest 5.63%. Doing a quick test for TC the difference appears to account for about 1uS between the two. Further tests still to to be done on the quarter.

Eric.

[green]

This initial rapid fall-off you see on your slopes -- it's not skin-effect. When you excite your target, circulating currents can flow in any which direction they like. Those that go in tight loops, because they are constrained by a thin sheet, or a wires' edge-profile etc, decay quicker. What's left later on are the 'big' circulating loops, whose size is related to the larger dimensions of the target. It's nothing to do with skin effect.
Remember that pdf Eric linked to, about the test jig that measured long rods of metal decaying ? The maths of that showed the multiple decays, and then stated that the 'big' one is the one they want to measure and analyse. It involved Bessell Functions, which are called upon when things happen in two (or three) dimensions ... vibrating drumskins? heat transfer through a solid? Digital filtering of images? I've little exposure to their intricacies ...

Skin effect is something really relevant to continuous AC signals, like a VLF detector, or 60Hz/50Hz power line currents. The currents only flow in the outermost layer of the item. At microwave freqs, it's just the skin, microns thick that conducts, nothing at all goes on in the core of a wire, it could be a tube, with no difference. You don't see 'the effect of the skin' , it's always there, it doesn't 'show itself' under any particular given circumstance.

[ I hope all that is correct, I don't want to have to ask Carl to edit/delete it..]

Re: your turn-on / off measurement differences. What's different about the circuit? An I correct in saying the damping is different between the two states, one is damped by a resistor, typically 500 Ohms, the other will be a near-zero Ohm turned-on Mosfet ?

Another test. When I tested the 19mm length wires the other day I noticed orientation didn't seem to make much difference. Tested 3 wire sizes and a US nickel. Can get a lot different slopes with the nickel. Wondering if you see much change with 19mm of copper wire with the VLF.

[Skippy]

"I've thought the copper clad acted as a star point"
The clue is in the name - Point. If a high current flows through a conductor, even a sheet of copper, it creates a voltage drop, bouncing the ground level to other parts of the circuit. Worse still when considering step changes in current, such as may occur with digital logic, or PI detectors. Then the inductance of the conductor also comes into play, as you're aware; put a dI/dt through an inductor, and a V is generated.
"Have you checked if target position or orientation affects your VLF measurement?"
I'm very aware that it can do, so I try and be consistent. Not too far away from the centre sweet-spot for static measurements. Avoiding changing meter ranges, as the calibration is slightly different. EMI and temperature variations are always a pain.
One of the problems with using a commercial machine is that you don't really know the 'specs' .. what does/doesn't drift with temperature etc.

You're correct about having nothing/no-one to compare to. I get the feeling that I'm the only ever to have made some of the measurement I've done. Clearly I'm not, I'm sure George Payne had to do a lot of this kind of stuff back in the 1980's, but the work of professional/commercial designers is very closely guarded.

[green]

Oh! That's ... a work in progress. It raises a number of questions ... is that why you have noise concerns ? Is that why you get odd, inconsistent data?.
More specifically: long leads flapping around when there's a high-current pulse generator nearby isn't good. And the general spread-out nature of it raises issues about the quality of the ground(s) and the power supplies.
And all the advice about power supply connection, decoupling connection, having the star point at the 0.5R ( I assume it's two 1R0's on top of each other?), needs implementing.

And re: my idea of there being too much coupling between target and Rx/Tx coils, I've just taken a fresh look at some charts you posted a few days ago:

http://www.geotech1.com/forums/attachment.php?attachmentid=45184&d=1548613325

..and noticed that targets on top of the coil gave slightly lower TC's than those 16mm away, which supports my thinking. Though it is only a slight change.
Seeing as the 40mm test coils are very similar in size to half of your 1.5" 8-coil, I wonder if they show this effect more strongly? Just an idea, I guess too close and you'll just get overloaded RX amp problems?
And re: the latest 100/200/400us slope, I found it hard to work out which was which at first. The shortest transmit time results in the biggest response from the target? This is related to the business of 'optimum' TX time being 4 times longer than the target TC ? But this is a 200 us target, so why doesn't 400 us give a bigger response.? Non-expert here, don't forget.

Some thoughts, maybe correct or not. When I design a circuit it usually needs modification or additions. Building on copper clad boards makes it easier for me. Not a good way to build a circuit if making very many. I've thought the copper clad acted as a star point and where thought it was necessary connected components at same place on buss(750R, .5R and the caps for current fdbk and command as an example). Building this way might not be best for noise. The noise thread was about determining if noise was what was expected or worse. Example: I use a differential amplifier with 1k to common at each input, gain about 450 and BW about(DC to 1MHz). Input open, p-p output noise=15mV minimum. Input shorted, p-p output noise=2.7mV with 1nV/rt Hz input noise. Something to compare my amplifier with to see if improvements could be made. You are probably right about leads flapping in the air, not sure if worse than Rx next to Tx with 500V spikes on Tx. There are no spikes after the amplifier has settled. Any thoughts or corrections appreciated.

Inconsistent data: not listing ambient temperature, target position and orientation, coil size, Tx time, constant rate or constant current might have some effect on decay rate along with circuit calibration. You have done some measurements with a VLF. Have you checked if target position or orientation effects your VLF measurement? Eric is the only one that I can remember charting decay with a PI. Keep hoping for some more input to compare with.

[green]

Sorry, I thought that you were looking at the RX signal during the TX time OFF.

I am still looking for the discussion mentioned above. I am sure I have it on some backup disk.

The current wave shape as well as the di/dt, during TX ON, can have a considerable influence on certain targets.

Nothing to be sorry about. Good to see you are reading the thread. Any comments welcome.

[Tinkerer]

My reply was about Rx signal vs constant rate Tx when Tx is on. Before the kick.

Discussion after the kick could be interesting also.

Sorry, I thought that you were looking at the RX signal during the TX time OFF.

I am still looking for the discussion mentioned above. I am sure I have it on some backup disk.

The current wave shape as well as the di/dt, during TX ON, can have a considerable influence on certain targets.

[green]

Some 17 years or so ago, Eric Foster and David Johnson had a discussion about PI TX on http://www.findmall.com/list.php?34 Basically the question was: What kicks the target. If I remember right, the two opinions were different, but both very interesting. I know I have saved some of the discussion, but am unable to find it right now. It could be interesting to compare the output relative to the 2 different opinions.

My reply was about Rx signal vs constant rate Tx when Tx is on. Before the kick.

Discussion after the kick could be interesting also.

[Tinkerer]

Amplitude is proportional to rate, Tx on. One half the Tx time, twice the rate(1A peak). Should be twice the signal. Close to 3divisions near start of decay(3divisions=twice the signal). Same with your VLF, double Tx amplitude doubles Rx signal?

Noise is lower with Rx input shorted. Coil pickup causing a lot of the noise, some coming from the scope.

( I assume it's two 1R0's on top of each other?)1RO's? I used one board, copper both sides?

Some 17 years or so ago, Eric Foster and David Johnson had a discussion about PI TX on http://www.findmall.com/list.php?34 Basically the question was: What kicks the target. If I remember right, the two opinions were different, but both very interesting. I know I have saved some of the discussion, but am unable to find it right now. It could be interesting to compare the output relative to the 2 different opinions.

[Skippy]

Two 1R0's means two one Ohm resistors. It's common engineering practice to replace the easily-lost decimal point with the multiplier, example 2K2, 1M5, 4R7, R22 for: 2.2K, 1.5M, 4.7R, 0.22R.
These fellows, one hiding under the other:

[green]

Oh! That's ... a work in progress. It raises a number of questions ... is that why you have noise concerns ? Is that why you get odd, inconsistent data?.
More specifically: long leads flapping around when there's a high-current pulse generator nearby isn't good. And the general spread-out nature of it raises issues about the quality of the ground(s) and the power supplies.
And all the advice about power supply connection, decoupling connection, having the star point at the 0.5R ( I assume it's two 1R0's on top of each other?), needs implementing.

And re: my idea of there being too much coupling between target and Rx/Tx coils, I've just taken a fresh look at some charts you posted a few days ago:

http://www.geotech1.com/forums/attachment.php?attachmentid=45184&d=1548613325

..and noticed that targets on top of the coil gave slightly lower TC's than those 16mm away, which supports my thinking. Though it is only a slight change.
Seeing as the 40mm test coils are very similar in size to half of your 1.5" 8-coil, I wonder if they show this effect more strongly? Just an idea, I guess too close and you'll just get overloaded RX amp problems?
And re: the latest 100/200/400us slope, I found it hard to work out which was which at first. The shortest transmit time results in the biggest response from the target? This is related to the business of 'optimum' TX time being 4 times longer than the target TC ? But this is a 200 us target, so why doesn't 400 us give a bigger response.? Non-expert here, don't forget.

Amplitude is proportional to rate, Tx on. One half the Tx time, twice the rate(1A peak). Should be twice the signal. Close to 3divisions near start of decay(3divisions=twice the signal). Same with your VLF, double Tx amplitude doubles Rx signal?

Noise is lower with Rx input shorted. Coil pickup causing a lot of the noise, some coming from the scope.

( I assume it's two 1R0's on top of each other?)1RO's? I used one board, copper both sides?

[Skippy]

Oh! That's ... a work in progress. It raises a number of questions ... is that why you have noise concerns ? Is that why you get odd, inconsistent data?.
More specifically: long leads flapping around when there's a high-current pulse generator nearby isn't good. And the general spread-out nature of it raises issues about the quality of the ground(s) and the power supplies.
And all the advice about power supply connection, decoupling connection, having the star point at the 0.5R ( I assume it's two 1R0's on top of each other?), needs implementing.

And re: my idea of there being too much coupling between target and Rx/Tx coils, I've just taken a fresh look at some charts you posted a few days ago:

http://www.geotech1.com/forums/attachment.php?attachmentid=45184&d=1548613325

..and noticed that targets on top of the coil gave slightly lower TC's than those 16mm away, which supports my thinking. Though it is only a slight change.
Seeing as the 40mm test coils are very similar in size to half of your 1.5" 8-coil, I wonder if they show this effect more strongly? Just an idea, I guess too close and you'll just get overloaded RX amp problems?
And re: the latest 100/200/400us slope, I found it hard to work out which was which at first. The shortest transmit time results in the biggest response from the target? This is related to the business of 'optimum' TX time being 4 times longer than the target TC ? But this is a 200 us target, so why doesn't 400 us give a bigger response.? Non-expert here, don't forget.

[green]

Puzzled me. I'm contemplating whether the target is coupling more than wanted with the TX and RX coil. These both have loads on them (resistors), so will absorb energy from nearby inductors, like our target. This would shorten the decay time of any current flowing in the target. I'm trying to work out, if the above is true, why there's more resistive load during Transmit than during Receive. The RX coil is unchanged, so I'm looking at the TX circuitry. If that's a perfect current sink, then it behaves like " a very large voltage source, through a very large resistor" , so it shouldn't be adding any damping to the TX coil. If it were flawed, and behaved like a current sink with a parallel resistance, then the opportunity to remove more energy from the target exists.
I'm a bit short of ideas. One obvious thing is the 12 V power rail. If that drops/rises, could it effectively increase the coil damping? I'm unsure about it from a theoretical viewpoint, I would just take the simple way out and try more PSU decoupling, of the right type, in the correct place.
The place being from the bottom end of the 0.5R current-feedback res, to the coil +ve end.
Another possibility is the op-amp current regulator is not working correctly due to bad wiring, eg. no proper star point for the -ve supply. The star point should be the bottom end of the 0.5R sense resistor in the fet source circuit. The main PSU decoupling goes there. The bottom end of the 750R/0.1u/1u current-setting components go there. The wire going to your bench PSU/battery(?) goes there.
Other things I thought of had the opposite effect unfortunately. Eg. if the fet didn't turn off fully, it would add damping during the off-time, shortening the target TC's during this period.
I suppose it would be rude to ask for a photo of the board.

Including photo of board.

Another test, Tx on with AWG10 40mm ring. Tx time 100us, adjusted space(target to coil)for near full scale reading. Recorded data 100 and 200us Tx, target not moved.

Then adjusted space(target to coil)for near full scale reading at 200us Tx. Recorded data 200 and 400us Tx, target not moved. Charted data. Inserted a line parallel to the shorter Tx, added a second line same slope and positioned it near the longer Tx. Appears the shorter the Tx the steeper the decay slope(shorter TC). Could skin effect be the cause?

Forgot to add, Y axis, log scale .4V/decade, 25mV in=0V out to chart.
12V battery
board, copper both sides, top three sections, left (12V), middle digital (5V), right (analog common), bottom(-battery)

[Skippy]

Puzzled me. I'm contemplating whether the target is coupling more than wanted with the TX and RX coil. These both have loads on them (resistors), so will absorb energy from nearby inductors, like our target. This would shorten the decay time of any current flowing in the target. I'm trying to work out, if the above is true, why there's more resistive load during Transmit than during Receive. The RX coil is unchanged, so I'm looking at the TX circuitry. If that's a perfect current sink, then it behaves like " a very large voltage source, through a very large resistor" , so it shouldn't be adding any damping to the TX coil. If it were flawed, and behaved like a current sink with a parallel resistance, then the opportunity to remove more energy from the target exists.
I'm a bit short of ideas. One obvious thing is the 12 V power rail. If that drops/rises, could it effectively increase the coil damping? I'm unsure about it from a theoretical viewpoint, I would just take the simple way out and try more PSU decoupling, of the right type, in the correct place.
The place being from the bottom end of the 0.5R current-feedback res, to the coil +ve end.
Another possibility is the op-amp current regulator is not working correctly due to bad wiring, eg. no proper star point for the -ve supply. The star point should be the bottom end of the 0.5R sense resistor in the fet source circuit. The main PSU decoupling goes there. The bottom end of the 750R/0.1u/1u current-setting components go there. The wire going to your bench PSU/battery(?) goes there.
Other things I thought of had the opposite effect unfortunately. Eg. if the fet didn't turn off fully, it would add damping during the off-time, shortening the target TC's during this period.
I suppose it would be rude to ask for a photo of the board.

[green]

Looking at schematic reply#168. Doesn't show MUR460 diode in series with Tx coil. Tried it both ways, ended with it included. Tx in a constant current loop or constant rate loop. Wasn't thinking about effect Tx(Rd) has on constant rate coil current. Need to determine if a problem.