Using Air Tests To Gauge Detector Performance

[Steve Herschbach]

I see this over and over on the forums. People requesting that this machine and that machine be air tested on a dime to see which gets better depth.

All it really does is tell you how well the detector can find things in the air, but I do not use a detector for that myself.

I do a lot of air tests, but I am only looking at depth from a relative perspective and not as how two different models compare for depth in the ground. In other words, if I have two Gold Bug detectors and set up identically if one gets far less depth than the other there is a problem with the detector. Detectors will rarely detect deeper in the ground (I did not say never) than they do in the air, so if a detector air tests on a dime at 10" I am not going to expect it to do better than that in the field as a general rule. If a detector cannot detect a tiny nugget in an air test, I do not expect it to get that ability in the ground. Air tests generally tell you about a detectors maximum possible performance, and in the ground things will usually go downhill from there.

I also like air tests to learn how a discrimination system may identify certain items since different detectors use different sounds and id number scales. Very often I just want to hear how the detector sounds. Is the signal modulated or boosted? Does it employ VCO audio? Short chirps or big booms? I have to listen to the thing all day long so I prefer something that is pleasant for me to listen to. Air tests can also teach me a lot about how a coil responds, especially to small targets which do radically strange things under DD coils at close range in particular.

Depth though, when we are metal detecting, is all about how detectors handle ground. In fact there is almost nothing more important than how the detector handles the ground. Without the ground, depth tests can mean nothing at best and at worst can be quite deceiving.

With VLF detectors low frequency models generally handle ground better than high frequency detectors, but high frequency models often air test very well. A Fisher Gold Bug 2 air tests really well, but the depth in ground drops off faster than most any other detector made due to the very high 71 kHz operating frequency. When I was up at Moore Creek I had a guy insist that the Fisher Gold Bug 2 would detect a large nugget deeper than a Fisher F75 because, by golly, he had air tested them. No amount of talk would convince him otherwise, so we buried a 1/4 oz nugget and the F75 easily hit it deeper than the Gold Bug 2. The guy was amazed by something I thought would be very obvious. I forget people do not know how metal detectors work and the effects of the ground at different frequencies.

It is not always true but a generality is low frequency detectors will tend to retain depth better in ground than high frequency detectors.

Pulse induction detectors are sort of like super low frequency detectors in that ground is relatively invisible to them. Not totally so by any means but compared to a VLF a PI has built in ground capability just because of the way they work. Often a PI detector will not air test very well compared to a VLF, but put them in bad ground, and the PI loses very little if any depth while the VLF takes a big hit. The worse the mineralization the worse the VLF does by comparison to the PI detector.

So another generality is that pulse induction detectors do not air test well compared to VLF detectors.

A lot of very good but less expensive coin detectors have no ground balance control. They air test just fine against far more expensive detectors. Put them on bad ground however, and the lack of ground balance control just kills them.

Something rarely ever discussed is ground balance systems and how they work. Ground balance methods vary and often are proprietary and closely guarded secrets. These days it is far more than just a knob. Detectors like a White's GMT or MXT were among early models employing software algorithms to ground balance the detector. Minelab multi-frequency detectors employ very sophisticated ground balancing methods that help account for how well those machines work in differing soils and even saltwater environments. They are designed with the goal of delivering accurate target id information as deep as possible as opposed to absolute depth and the accuracy they deliver is cutting edge. Ground balancing and accurate target id go hand in hand. It should be obvious that a detector that has factory preset ground balance is going to suffer in bad ground. But past that point, what do you really know about the ground balance method employed by a detector and how good it is? More importantly, how is air testing going to help teach you about it?

Yet another generality is that multi-frequency detectors do not air test well against single frequency detectors. You can go farther and just say air tests teach you nothing about ground handling capability.

The degree and efficiency with which a metal detector handles the ground conditions it encounters is the most important thing there is when it comes to depth in the ground, and air testing does nothing to reveal this most important attribute. It is very easy to have detector A go twice as deep in an air test as detector B and see the situation reverse in the field.

In theory if you air test two detectors, both the exact same model, and they air test the same then they should get the same depth in the ground, right? But what if one detector has a ground balance system that is not functioning properly? They air test identically but one still performs poorly compared to the other in the ground.

The absolute most important advances in the metal detecting world have been in ground handling and most of the real breakthroughs in recent years have been in pulse induction detectors employing very advanced methods to deliver depth unmatched by other detectors. These advances are only apparent in the worst ground conditions and so leaving the ground out of the testing is nonsensical when you think about it.

To sum up, we use metal detectors to find items buried in the ground, and when prospecting in particular highly mineralized soil can have severe impacts on detection depth. Only in ground testing has anything like any validity, and even then freshly buried target responses are suspect due to the ground being disturbed. Even buried test targets however are far preferable to air tests. A lot can be learned from air tests, but how a detector will perform in the ground is just about last on the list when it comes to relevance.

[Steve Herschbach]

OK, so how do you test a detector against another detector?

I take two detectors into the field. I use detector A to locate a target. I check the target with detect B and note any differences in target response. I recover the target. I then find a target with detector B and check it with detector A, once again noting signal differences. I repeat this over and over, looking especially for weak targets that one detector can get and the other will not.

Yes, I do air tests. Yes, I bury test targets. Things can be learned doing that. But when I get serious about wanting to compare two detectors I believe only extensive cross testing of found targets has any true validity. Anything else is just crude simulations of varying value and unfortunately they can often mislead people not versed in detectors.

[gambler]

for this newb, thread =meat. thanks

[gambler]

why are you not writing a column for icmj magazine? 

[Reno Chris]

Yah, why not?

[Mike Hillis]

Steve, I disagree with you about the air tests.   Not bench testing, but the air test, which to me are different.

For me, an 'air test' is a quick judge of potential, while 'bench testing' is about understanding the feature set.

A air test will quickly tell me if the detector/coil combo is suited for what I want to use if for.  For instance, if I was looking for a coinshooter, I would want to know the distance it gets on a dime.  I don't want to get the air test from just one source either.  I'd want to see results from multiple sources.  If the dime test is poor, or, less than I deem acceptable, then I just saved some money from a buy and try for a coinshooter.  

I can pretty much tell how a detector should air test a dime just by the operating frequency and coil size, but if the air test is different than I am expecting that also tells me stuff about ground balance systems and design intent.

Coinshooters care about dime tests.    In fact, a US coinshooting should air test a quarter farther than it does a nickel.   You might think, well duh! but not many modern units will detect a quarter farther than a nickel anymore. 

Thats a dime.

Do prospectors air test?   Sure they do.  You just told us about the bic pin test.  Again, its just potential.  We all know the ground minerals are going to decrease depth, but at least the air test gave us a potential to work down from, both from a size perspective, as well as a distance perspective.

So, yes, I think air tests are a valid quick and dirty way to judge between units when looking for certain performance characteristics.

HH
Mike

[Steve Herschbach]

So, yes, I think air tests are a valid quick and dirty way to judge between units when looking for certain performance characteristics.

HH

Mike

I very much agree Mike. I am pretty sure I said just that but I guess I did not communicate it properly. My goal is simply to warn people that detector A getting an extra inch over detector B in an air test does not mean that is how it is going to work in the ground. And yet a great many people imply just that with air tests.

[Mike Hillis]

Totally agree...

 

My goal is simply to warn people that detector A getting an extra inch over detector B in an air test does not mean that is how it is going to work in the ground.

[doc holiday]

Air tests are Very important when comparing 2 or more of the SAME detectors. Many years ago I was a silent partner(I signed all the bank notes LOL) in very busy firearms/detector/antiques business.We would have a couple dozen detectors on hand and I would air test them to pick MY detector. It was Common for 2 identical detectors to have 25/30% difference in air test depth---these were from one of Steve's favorite companies.

[tvanwho]

Could you simulate bad ground by using bricks in an air test? Some of them tend to be loaded with iron minerals altho I don't know how to tell which ones? Cinders can make for pretty tough detecting too. My aunt  had a yard full from the burn barrel and it was next to impossible to detect there. Its kinda hard to ground test outdoors in mineralized ground at times but you could stack bricks for quick indoor mineralized air depth tests perhaps? Maybe even collect the black sands from gold dredging/ drywashing and make mineralized bricks...just a thought...

 

-Tom