Ground Balance Theory

[Ferric Toes]

Hi Aziz,

I'm not sure that requesting 'scope measurements will give you the accuracy that you need. That is where some results for the exponent have fallen down in the past. Also, if the TX pulse has insufficient duration the decay will take a dive at late times. I could extend to milliseconds and guarantee it will not have deviated, but I am concentrating on the time range in which a practical nugget hunting detector will be working. My latest viscosity meter pulse timing is accurate to 25nS, thanks to my friend BandWidth who is working with me on this. Watch this space for exponent results from around the world. I already have a prototype PI that utilises the results of my viscosity measurements to give superior GB over a wide range of pulse timings without the need for further adjustment. No vapourware in this neck of the woods.

Eric.

[Prospector_Al]

I have never observed a difference between the two. At high TX field strengths there may be a difference on some Australian ironstones which can support remanent magnetisation. Remanent magnetisation affects the susceptibility, which in turn affects the viscosity. Bipolar would prevent this.

Eric.

Hi Eric,
You have a magnificent setup there!
deemon asked you a while ago about the difference between unipolar and bipolar excitation and you responded that you hadn't noticed any, except for Australian ironstone samples.
I think the bipolar vs. unipolar question has another aspect that deserves investigation and you have the equipment and expertise to provide the answer.
We know for a fact that many rocks and soils have coercivity. If it weren't so, the paleomagnetic aficionados wouldn't be in business and the Chinese would not have invented the compass. Remanent magnetism shows up only in solid rocks, but when the rock has eroded into sand, the manifestation of a field has disappeared because of the random orientation of the grains, but the coercivity is still there.
When you impose a bipolar field on a magnetic material, it will travel through the whole hysteresis curve, but with a unipolar pulse, all the action is in the first quadrant. This has energy implications: Travelling through the hysteresis curve takes energy which is sapped from the imposed field. One would think therefore that if you stay in the first quadrant, the excursion is much smaller and less energy is lost. That should translate into deeper penetration of the field into the soil.
The question is: Is this attenuation of the field insignifcant? According to Lenz's law, the effect must be there, but perhaps it's so small that it's not worth worring about. A measurement would provide the answer...

Regards,

Allan

P.S. I bought a dental oven on E-bay for heating the samples. It goes up to 2000 F.

[satdaveuk]

*LOL*
Patent examiners are dumb! Incredible dumb and aren't skilled in the art.
You can mind **** and dribble them easily by putting some (complex) formulas into the description.

The only good advice is to save/print the forum posting and keep it save in case of patent trolling procedure. Collect as many publications as you can do.
(Like I do it.)

Aziz

Good on you Aziz
And I for one would be a willing and participating witness with dates and facts for any of you here that had problems with these company's that steel your designs.
It makes me angry because although theft of patents has been going on even before the invention of the Davey lamp which came to light years later, pardon the pun, these days its done so blatantly and stamped as legal without the authority's doing any research, which may I add is so easy these days what with the Internet.
Its just one big gravy train where the fat cats are making easy money out of someone else's hard work.
As stated further up the posts, in the old days you had to demonstrate the workings of your finds and how you came to them conclusions, these days it doesn't really have to work in practice as long as it looks good on paper.

Regards

[Ferric Toes]

Hi Eric,
You have a magnificent setup there!
deemon asked you a while ago about the difference between unipolar and bipolar excitation and you responded that you hadn't noticed any, except for Australian ironstone samples.
I think the bipolar vs. unipolar question has another aspect that deserves investigation and you have the equipment and expertise to provide the answer.
We know for a fact that many rocks and soils have coercivity. If it weren't so, the paleomagnetic aficionados wouldn't be in business and the Chinese would not have invented the compass. Remanent magnetism shows up only in solid rocks, but when the rock has eroded into sand, the manifestation of a field has disappeared because of the random orientation of the grains, but the coercivity is still there.
When you impose a bipolar field on a magnetic material, it will travel through the whole hysteresis curve, but with a unipolar pulse, all the action is in the first quadrant. This has energy implications: Travelling through the hysteresis curve takes energy which is sapped from the imposed field. One would think therefore that if you stay in the first quadrant, the excursion is much smaller and less energy is lost. That should translate into deeper penetration of the field into the soil.
The question is: Is this attenuation of the field insignifcant? According to Lenz's law, the effect must be there, but perhaps it's so small that it's not worth worring about. A measurement would provide the answer...

Regards,

Allan

Hi Allan,

Somewhere I have a paper on this, but from memory the hysteresis loop of SPM particles is so skinny as not to be relevant; hence their coercivity is almost zero. Oz rock that I have has coercivity and if magnetised will pick up pins but this is due to additional larger grains that are above the SPM threshold. I notice that if I demagnetise Oz rock, the susceptibility and viscosity goes up, but the decay exponent is unchanged. There are some odd things that happen which I haven't yet found an explanation for, and that don't occur in normal soils, but more on that another time.

It's best to regard SPM particles as something separate from very long term viscosity (paleomagnetism), which results from large single domain and multidomain grains. Rock can have all of these which exhibit separate effects. In the first case TCs of 100s of uS, second case 100s millions of years.

Eric.

[Ferric Toes]

Allan, Have a look at this and scroll down to Superparamagnetism.

http://www.irm.umn.edu/hg2m/hg2m_d/hg2m_d.html

[Prospector_Al]

Allan, Have a look at this and scroll down to Superparamagnetism.

http://www.irm.umn.edu/hg2m/hg2m_d/hg2m_d.html

This is an excellent reference. The question about depth loss owing to the "work" performed on the domains is apparently not relevant to the viscous signal, but that's only part of the problem. The other magnetic species must also be considered. The theory seems so complex that the only way to settle the issue quickly is to perform experiments.

A comparison of the field strength in air vs. through a volume of soil would yield the answer. The principle applies to energy loss caused by eddy currents as well--the field strength on the other side of an aluminium plate is noticeably lower. A static field is, of course, unchanged.

Oh well, it was just a thought--back to the soldering iron...

Allan

[Altra]

That's a great source Eric.

I found the complete paper in PDF format. In case the link
becomes broken in the future, I'll attach it here.

[Ferric Toes]

That's a great source Eric.

I found the complete paper in PDF format. In case the link
becomes broken in the future, I'll attach it here.

Yes, that is a very good paper. Another one that is very readable is http://www.pmc.ucsc.edu/~njarboe/pma...netismBook.pdf Enjoy.

Eric.

[Tepco]

These plots were done in 1968 by C. Colani and shows the advanced thinking at that time. All were done with a logarithmic front end amplifier so that exponential decays from non-ferrous targets appear as a straight line decay. Top left is a brass test plate 10 x 10cm. This is the noise free trace. The lower trace is a derived negative voltage level that is proportional to the slope, and that gets noisier as the signal decays. There is some non linearity at the bottom end of the decay, hence the curve. Top right is the trace from a nail which gives a log lin response similar to viscosity. The bottom left photo is for three exponential TCs - 50, 100, and 150uS and also shows the three derived levels. Further details can be seen on the .pdf (pages 8 - 10) attached to Post 65 on The PI History and Theory thread.

Eric.

[ATTACH]24332[/ATTACH]

Very interesting, ahead of it's time definitively. Good old days of Tek 54x scopes and film cameras. Logarithmic display, then differentiated to get deviation from linear ramp, that is 1\t in original signal, fine. Just probably cannot be used for soil characterization this way due to S\N limitations. Notorious issue with log-amp, noise and interference problems became exponentially more visible. Considering bandwidth required, even limited and filtered to, say, 10kHz to 1MHz, this is still two decades, compared to 1Hz or less in integrated measurement, nothing can compensate for difference. Modern day equivalent of this measurement can be using DSO to average out noise from log-amp, transfer data to PC and process with LabView or something similar. Time ago, around 1975, Tektronix produced 7000 series “calculating scope”, with P7001 processor, capable to do this without any external computing. I'm just completing sample holder, something like 5cm diameter, 7cm long coil to put sample inside, will try to measure response directly, but what is expected is maybe 20-30dB above noise floor, not enough. Not even with some cheating, using thermite mixture for sample (and hoping PI cannot ignite it). Pure Fe2O3+Fe3O4 mixture, if we neglect Al, perfect stuff. Will see..

[redcaveman]

hi Eric can your new GB method be incorporated into the GS5b or will it require new platform. cheers red

[Ferric Toes]

hi Eric can your new GB method be incorporated into the GS5b or will it require new platform. cheers red

It is purely for my own interest and experiments into ground compensation.

Eric.

[Ferric Toes]

using thermite mixture for sample (and hoping PI cannot ignite it). Pure Fe2O3+Fe3O4 mixture, if we neglect Al, perfect stuff. Will see..

It will be interesting to see how you get on with the Thermite. My guess is that you will get very little viscosity response unless the particle size is <30 nanometres; which is the SPM threshold. I have some 100mesh 99% Hematite and Magnetite and the particle size is way too large, hence no viscosity although high susceptibility.

Eric.

[Tepco]

Ended up exactly that way. Particles are certainly not 30nm, more like 30um or larger, very uneven. This oxide part is usually waste product from some chemical factory process, can contain just about anything, not good for this purpose, also not good for reference, different mixtures may behave differently, actually some Thermite types contain no Fe at all. Trying to find some other suitable substitute in form of ferrite. Ordinary soft ferrite is useless, but other types, like low permeability RF materials, or very high permeability, usually highly conductive types, or noncrystalline cores may be.


Idea with log-amp and differentiated output is phenomenal. I never before considered using log-amps for any hobby project, chips were expensive and hard to find time ago, but funny things can be done this way. Tried this “discrimination” method as described, well, works perfectly for nails, tools , paperclips and similar stuff, but unfortunately not for bottle caps or any more complexly shaped object. Very strange effects can be observed at shorter pulse lengths. Most objects tend to anomalously fall below it's natural linear (1\t log) curve earlier, some have “early time” bump more visible, some (larger more complex) deviate from linear in both directions during decay... Too many variables here.

[Tepco]

Noise floor

This is best I managed to get, using brick, this %^&*~! thing is impossible to get above noise floor. TX width around 70uS, but coil intentionally underdumped, so it goes down in 4-5 fast cycles, increasing speed and visibility at early time. Two things I concluded so far, this response is also TX dependent, falls down with pulse width (rendering my double pulse idea useless), getting just noise floor at 3uS, also, process is probably not 1\t. Hard to distinguish from images like this, but ordinary soft ferrite is almost ideal 1\t, trace is just parallel (at some angle) to reference, this one is not (traces averaged x256). With some modifications, fixed TX time, more power, filtering, shielding etc, this can be improved for another 15dB or so, just not sure it is worth doing. Direct measurement may not be the best idea.

[Davor]

Interesting that you mention it, but log amplification was mentioned on a few occasions here. Heck, I made some simulations with a simple opamp/diode setup. AFAIK it could be used for this purposes, and yes, it loses some log features at low signals as diodes are not as perfect as chained log amps, and they are prone to temperature effects, but both effects are mostly irrelevant for normal range of dynamics in use. As this setup uses inverting configuration, it is additionally resilient to high input signal (feedback fights the flyback), and first samples can be taken very early, at 4us or so.