Taking a look at GROUND BALANCE on the bench

[Aziz]

Hi guys,

there have been arised issues in this thread. They have been solved now and you have the right to know.
See post #476 in the following thread:

http://www.geotech1.com/forums/showthread.php?20264-Ground-Balance-Theory&p=179405#post179405

Cheers,
Aziz

[Aziz]

Aziz,
Can you show the phase characteristic of above diagram?

http://www.geotech1.com/forums/showt...988#post167988

No, I didn't make it at that time. It's I think two years old now (during the WBGB development at that time).

It's irrelevant and the time variable t (phase) is eliminated.

Well, you can do it for the complex plane as well (I (=Re) and Q (=Im) spectrum response). Everybody is invited to do it. Very trivial.

Cheers,
Aziz

[mikebg]

As we used to say here: "Abwarten und Tee trinken" (Wait and drink tee).
I'll bite back. The end-game is near.

Aziz

Aziz,
Can you show the phase characteristic of above diagram?

http://www.geotech1.com/forums/showthread.php?20235-Taking-a-look-at-GROUND-BALANCE-on-the-bench&p=167988#post167988

[Aziz]

..
Be careful about letting any cat out of the bag, it may turn out to be a lion that bites the person who put it in the bag.

Eric.

As we used to say here: "Abwarten und Tee trinken" (Wait and drink tee).
I'll bite back. The end-game is near.

Aziz

[Ferric Toes]

BTW,

if my favourite & greedy metal detector company doesn't leave us hobbyists alone, I might let the cat (WBGB) out of the bag.

The implementation of it on the Netbook/Tablet/Laptop PC is very trivial and but very effective. I might even give the full sources for it.


Aziz

BUT, as we have discussed before, it may be trivial on the Netbook/Tablet/Laptop, but until it is tried extensively in the field, preferably in the Victorian Goldfields, you cannot give it that prestigeous title. Until proved, it is WWGB, VGB (vapour), or SMGB (Scotch Mist) - take your pick.

Be careful about letting any cat out of the bag, it may turn out to be a lion that bites the person who put it in the bag.

Eric.

[Aziz]

BTW,

if my favourite & greedy metal detector company doesn't leave us hobbyists alone, I might let the cat (WBGB) out of the bag.

The implementation of it on the Netbook/Tablet/Laptop PC is very trivial and but very effective. I might even give the full sources for it.


Aziz

[Aziz]

The plot shows frequency dependent susceptibility and hence viscosity. The Bartington MS2B measures susceptibility at two frequencies to arrive at a figure for viscosity. The MS2B is quite a complex instrument and the frequency switching has to be done manually, with the sample removed and the instrument re-zeroed between measurements. A simple TD PI circuit measures the viscosity in one. I will say this though, I have to use a sample calibrated for viscosity (Xlf - Xhf) on the MS2B in order to calibrate my TD viscosity measuring system.

A good soft ferrite will not exhibit frequency dependence at least up to 100kHz, otherwise it would be too lossy for use in SM transformer cores. Commercial ferrite materials are not a good repesentation of real ground as they do not have a wide range of grain sizes. A fascinating fact regarding grain size is that a magnetite grain of 0.05um has a tau of 100 sec. while one of 0.07um is 4.5 billion years. Temperature too can cause the switch i.e. volcanic lava cooling through the blocking temperature freezes in the prevailing earth's field direction and magnitude - hence magnetic dating.

Your favourite metal detector company sticks to TD processing for its high end PI gold detectors. Same for PI mine detectors from various companies. Wonder why?

No dead horses yet, or even lame ones.

Eric.

Hi Eric,

of course, the horse isn't dead. TD GB works too.

But FD analysis will help to understand the matter better. You can compare magnetic samples by making a "finger-print" of it. But by looking at the TD/FD response will not necessarily lead to find a good GB algorithm. There is more work involved to crack it.

Cheers,
Aziz

[Ferric Toes]

Betonköpfe!!! (*LOL*)TD is simple and cheap to implement. But not elegant.Below is a FD response of a piece of ferrite (soft ferrite I think). It's a typical low frequency response, whereas the resistive targets cause higher high frequency component responses. The graph is made of real measurement data (nothing theoretical) and is a normalized response.[ATTACH]23654[/ATTACH]The FD method gives you much more benefits, which you can't do it in the time-domain:- Noise reduction (low frequency -> high frequency harmonics)- Induction balance mismatch detection and correction- Much much better GB- Less resistive response loss (GB is lossy)Still want to ride the dead horse?Aziz

The plot shows frequency dependent susceptibility and hence viscosity. The Bartington MS2B measures susceptibility at two frequencies to arrive at a figure for viscosity. The MS2B is quite a complex instrument and the frequency switching has to be done manually, with the sample removed and the instrument re-zeroed between measurements. A simple TD PI circuit measures the viscosity in one. I will say this though, I have to use a sample calibrated for viscosity (Xlf - Xhf) on the MS2B in order to calibrate my TD viscosity measuring system.

A good soft ferrite will not exhibit frequency dependence at least up to 100kHz, otherwise it would be too lossy for use in SM transformer cores. Commercial ferrite materials are not a good repesentation of real ground as they do not have a wide range of grain sizes. A fascinating fact regarding grain size is that a magnetite grain of 0.05um has a tau of 100 sec. while one of 0.07um is 4.5 billion years. Temperature too can cause the switch i.e. volcanic lava cooling through the blocking temperature freezes in the prevailing earth's field direction and magnitude - hence magnetic dating.

Your favourite metal detector company sticks to TD processing for its high end PI gold detectors. Same for PI mine detectors from various companies. Wonder why?

No dead horses yet, or even lame ones.

Eric.

[WM6]

Aziz, probably problem is that piece of ferrite has constant characteristic, but soil hasn't ever constant characteristic, sometimes respond as your ferrite sometimes as useful target, and all this in one coil sweep only.

[ODM]

There are plenty of published graphs for materials, if we wish to review such. Pudding, proof, again - it's out there in the (mostly frozen) soil which varies, as does the mobile measurement setup. What the "old farts" keep saying is that it's less successfully applicable as the panacea it's purported when moving into the field outside of canned soil samples and uniform testing. But they might be wrong, too. All it takes is some practical field examples which show the value of the approach.

I'll explain my standing further: Theory and practice is easy to see for yourself in some situations, such as basic mechanics. Other theories which are less intuitive are easy to accept, since they are readily applied, like electronics and chemistry. The key is, there are often several examples of a theory being used in practice if it proves to be useful. Development times are shorter than ever - why are these principles not applied in devices today? Are you the only people with a clear enough understanding of the issue? With as many companies studying electromagnetic geophysics, or only making mine/metal detectors, why have these simple and universal keys to success not been applied? ( ...this is a rhetorical question, and the answer is not litigation )

Riding a dead horse? Nope, not riding! I'm suggesting that the dead horse should be buried. Preferably underground - and at the same time re-acquaint ourselves with all the variables of working dirt.

[Aziz]

Betonköpfe!!!
(*LOL*)

TD is simple and cheap to implement. But not elegant.

Below is a FD response of a piece of ferrite (soft ferrite I think). It's a typical low frequency response, whereas the resistive targets cause higher high frequency component responses. The graph is made of real measurement data (nothing theoretical) and is a normalized response.



The FD method gives you much more benefits, which you can't do it in the time-domain:
- Noise reduction (low frequency -> high frequency harmonics)
- Induction balance mismatch detection and correction
- Much much better GB
- Less resistive response loss (GB is lossy)

Still want to ride the dead horse?

Aziz

[ODM]

Analysis in both time and frequency domain are viable tools, I don't think it deserves ridicule - but it does deserve some proofing. I assume Mike doesn't have and cannot afford the kind of laboratory setups some Geotech members have, otherwise he would be providing us with experiment journals.

Without experiments, we'll just keep on flogging the same dead horse in different directions.

[Davor]

Mine is Magnastat, time domain

[WM6]

Go on man ... switch on the soldering iron, and show us the error of our ways!

Which soldering iron, Time domain or Frequency domain one?

[Qiaozhi]

Allan,
All for nonferrous conductivity is predictable in Frequency domain, because we can use normalized locus.
Look that a wideband metal detector can show if the target is as ring or not.

Mike - you're not listening to what other Geotech members are saying. Or, if you are listening, you're not taking any notice.

Please go back and read posts #86, #94, #97 and #100.
You are never going to convince anyone here of your frequency domain ideas, unless you can prove them in reality, by actually building something that doesn't only exist in a simulation.
Go on man ... switch on the soldering iron, and show us the error of our ways!