Standardized Tests for Metal Detectors

[WM6]

Perlite is something different. This is Liapor (made out of clay):

http://www.liapor.com/en/basic.php?n=0101

Here is example how test perform:

http://www.geotech1.com/forums/showt...-EURO-Sabre-II

[ivconic]

Eric asked first for schematic here on forum .... and at the end developed it for himself (and probably patented too):

http://www.geotech1.com/forums/showt...tibility-Meter

Commercial portable version:

http://www.zhinstruments.com/?gclid=...8z0RugjT7w_wcB

Deus in best configuration appear to be cheaper than SM-30 meter.

Usable if you wish to test extreme soil (mineralization) conditions like in some extra heavy terrains in Australia which calls for specials detectors design (even better than Minelab GPX series). EU terrains are not so critical for most nowadays (GB) detectors. Wish to have one susceptibility meter, but it is not "must have" for proper testing. But for developers like Eric, ambitious to develop detector capable to cope with extreme Australian gold fields it could be "must have".

Fur us main testing premises are in first line two:

1. Tests should be repeatable, no matter if we are testing right now or after years test conditions should be the same.

This mean that we need to forget on soil test, cause soil is living thing and in combination with wetter conditions next week will be already different than today - not to say years after. Soil tests are OK, if we perform comparison tests of different detector at the same day (like present Bulgarian tests). Instaed of soil we need standardized testing soil surrogate.

2. Tests should be comparable which mean that same testing criteria should be used for all tested detectors.

This include same testing sample with same orientation in testing substrate. Same depth measurement means. Same signal criteria (e.g. clear repeatable sound in both coil sweep directions). If we intend to make comparisons, we need to chose comparable coils. New (or full) batteries in all tested detectors. Relatively same wetter conditions (overheated or frozen detectors can lead to malfunction). Proper and optimal settings according factory instructions manual. At least 3 repeated test sessions for all detectors. In time documented results. Good performed depth test should be accompanied with real terrain searching impressions.

Ok now i see.
Insisting on such delicate and expensive method of measuring soil is pointless in our case here.
We better try to keep it "human" as much as possible, so huge number of people to be able to join in this.

"...we need standardized testing soil surrogate..."

Well... i tend to agree on this... but also to disagree too.
What's the point if we take only one or two cases of soil surrogate? What's the goal?
We must know that various detectors do perform different on various soils, on some are better performing and on some worse.
So we may easily end up in a false conclusions about some detector performances and real value.
We can't accurately anticipate all the possible cases if using surrogate, that's why is better to perform all our tests outdoor on real soils, but at the same time we surely must provide accurate soil readings too.
Using surrogates will dramatically limit the number of people involved in this "action", because needed material for such surrogates is not available the same in all countries and easy obtainable for all people involved.
It's too much complicated to approach it that way, trust me.
Let the soil be any natural soil from anywhere, let us focus more on soil measuring methods.
Deus is expensive and not easy obtainable for many people, Minelab too, E.F.'s apparatus probably too.
Can't we figure out simple and DIY repeatable device here on Geotech, which will serve all of us here for such purpose?

[ivconic]

Perlite is something different. This is Liapor (made out of clay):

http://www.liapor.com/en/basic.php?n=0101

Here is example how test perform:

http://www.geotech1.com/forums/showthread.php?21590-TEST-Tesoro-EURO-Sabre-II

Wow! Would be splendid if you "measured" that surrogate with Deus than!
Now i would have much clearer "picture" by knowing the GB "VDI" number related to that specific surrogate.

[ivconic]

I am having those two.
IBM one is gold plated.

[ivconic]

[ivconic]

Oooops he did it again!?

[ivconic]

I just performed skim and simple air test with those two Pentium processors.
Results were more than surprising for me!?!?!?!?
Ok... now we have TWO target references!

[Nupi]

I just performed skim and simple air test with those two Pentium processors.
Results were more than surprising for me!?!?!?!?
Ok... now we have TWO target references!

This is a fake one.

[ivconic]

This is a fake one.

I assume it's the genuine one too but not gold plated as IBM.
Anyway, considering that it is easy to obtain it worldwide; it also can be used as target reference.
As far as we are aware of it's magnetic and conductive features.
That's why is good to have metal detector with accurate VDI.
Luckily most of modern detectors do have VDI feature.
Those two processors used as target references can also show us differences in VDI's at different md models.
Ain't those indeed easy obtainable and cute targets!

[Nexus]

There are very few ways of testing detectors head to head that can actually give an accurate idea. Any kind of tests in air are a waste of time, because they do not correlate with the realities in ground.

This is how we do these tests with Nexus vs other brands.

Test 1 - practical field test.

Certain ground area is surrounded or marked by any means and the deepest detector in the comparison set is used to search that ground for all signals present.
If the sate is relatively clean then the deepest of all detectors in the set is to be used first.
If the site is a trash loaded one then the most efficient for such conditions detector is to be used first.
Every signal detected is then marked as Iron or non-ferrous let say by small flags - red for non-ferrous, blue for Iron (just a suggestion, can be anything else). The detected signals are numbered from 1 to what ever number turns up.
Every next detector goes over that patch of land over the already detected signals and over the "clean" soil to confirm that there are no more or there are more signals.
The data from every detector is recorded accordingly to the relevant number of each signal previously detected or each new signal recovered by that detector.
In the end all signals are recovered and recorded as what actually they are.
The actual data is then compared to the data recorded from each detector and only then becomes obvious who wins.

This test can be done in variation for non-ferrous or other specific targets only on trash sites or less loaded with metal object sites for everything that can be found.

Test 2, under controlled conditions for depth.

Basic rules.
1. All detectors used in such test have to be equalised in their audio output level by a SPL meter.
2. All detectors used in such test have to be equalised in their noise (chatter) level by a SPL meter.
3. All detectors used in this test have to be set to reject Iron reliably.
4. All detectors used in this test have to be equalised in their Threshold level by a SPL meter.

SPL stands for a Sound Pressure Level.

These settings will remove the subjective factor such as loud audio = deep or powerful detector, which is a very common mistake in judging detectors.

After these setting are done all detectors can go on the test pit, corner test or a parallel to the ground surface test hole. Depths are subject to personal interest and choice.
Multiple hole will determine which is the deepest detector by some margin of error. The corner test remains the most accurate way to measure every centimetre of difference between detectors.


Of course people can suggest all sorts of weird ways to test detectors, but great many of them would prove only a subjective point of view or a narrow practical view.

[Nupi]

I assume it's the genuine one too but not gold plated as IBM.
Anyway, considering that it is easy to obtain it worldwide; it also can be used as target reference.
As far as we are aware of it's magnetic and conductive features.
That's why is good to have metal detector with accurate VDI.
Luckily most of modern detectors do have VDI feature.
Those two processors used as target references can also show us differences in VDI's at different md models.
Ain't those indeed easy obtainable and cute targets!

I do not know if the processor is fake, it was a bad joke, I think that is a real one.
I rather think that the fault lies with Deus? Maybe he has a bad day today, we sometimes have to.
If you want I can prescribing Efexor and Ritalin with lithium, he recovers for sure.

Where is the schematic????????

[ivconic]

There are very few ways of testing detectors head to head that can actually give an accurate idea. Any kind of tests in air are a waste of time, because they do not correlate with the realities in ground.

This is how we do these tests with Nexus vs other brands.

Test 1 - practical field test.

Certain ground area is surrounded or marked by any means and the deepest detector in the comparison set is used to search that ground for all signals present.
If the sate is relatively clean then the deepest of all detectors in the set is to be used first.
If the site is a trash loaded one then the most efficient for such conditions detector is to be used first.
Every signal detected is then marked as Iron or non-ferrous let say by small flags - red for non-ferrous, blue for Iron (just a suggestion, can be anything else). The detected signals are numbered from 1 to what ever number turns up.
Every next detector goes over that patch of land over the already detected signals and over the "clean" soil to confirm that there are no more or there are more signals.
The data from every detector is recorded accordingly to the relevant number of each signal previously detected or each new signal recovered by that detector.
In the end all signals are recovered and recorded as what actually they are.
The actual data is then compared to the data recorded from each detector and only then becomes obvious who wins.

This test can be done in variation for non-ferrous or other specific targets only on trash sites or less loaded with metal object sites for everything that can be found.

Test 2, under controlled conditions for depth.

Basic rules.
1. All detectors used in such test have to be equalised in their audio output level by a SPL meter.
2. All detectors used in such test have to be equalised in their noise (chatter) level by a SPL meter.
3. All detectors used in this test have to be set to reject Iron reliably.
4. All detectors used in this test have to be equalised in their Threshold level by a SPL meter.

SPL stands for a Sound Pressure Level.

These settings will remove the subjective factor such as loud audio = deep or powerful detector, which is a very common mistake in judging detectors.

After these setting are done all detectors can go on the test pit, corner test or a parallel to the ground surface test hole. Depths are subject to personal interest and choice.
Multiple hole will determine which is the deepest detector by some margin of error. The corner test remains the most accurate way to measure every centimetre of difference between detectors.


Of course people can suggest all sorts of weird ways to test detectors, but great many of them would prove only a subjective point of view or a narrow practical view.

Using processor as target reference is indeed weird much! I agree.
But i suggested it because it is easy obtainable worldwide, any continent, any country, any area.
Such processors are made under strict standards and all of those (from same series) are 99.99% identical.
That's what will prevent jitters in our testing results which may occur if our testing targets are not exactly identical.
...

Your suggestions are alright but not fully acceptable and will not produce objective meritorious results.
Simple example; most of Minelabs are primary designed and made to suit Australian soil conditions.
If you test some of those models on quite different soil conditions; the final results will be far from meritorious.
Both; Test 1 and Test 2 proposals falls down from being objective in many cases.
For example; Minelab SD series will easily "loose" the battle in such testings against any simple diy project from these forums, though we know that it is not like that.
You may than say that SD lost against some funny useless toy (frequently seen on these forums) and that's what we want to avoid here.
Other words; it is very easy to mix pears and apples with such approach and adjust final testing results as suits to your interests.
We must be smarter than that.

[ivconic]

I do not know if the processor is fake, it was a bad joke, I think that is a real one.
I rather think that the fault lies with Deus? Maybe he has a bad day today, we sometimes have to.
If you want I can prescribing Efexor and Ritalin with lithium, he recovers for sure.

Where is the schematic????????

Most probably my Deus is suffering from long term inactivity!
It's been a while since i walked it outdoor...

[ivconic]

Forget schematic !!!!

[Nupi]

@ivconic

I hope it does not come through serious illness that you could not go out.
And the schematic I forgive you.