Jump to content

Tune Out Nails - You Will Miss Gold!

Recommended Posts

VLF metal detector discrimination works well on isolated targets in an air test. The problem is in the field what is reported by the detector is the sum of everything the coil "sees". This means the ground mineralization, the gold nugget (or any other item you are trying to find), other metal under the coil at the same time, and even electrical interference. Sweep speed matters also as does the angle of the item in the ground and the direction from which the coil approaches it. Add it all up, and it is a miracle discrimination works at all, and the reality is it is wrong very often.

Almost any ground with iron mineralization will cause non-ferrous items to read as ferrous. Usually it is something that happens right on the edge of detection depth. However, the more iron mineralization, the less depth it takes for the item to flip over to ferrous. It does not matter how large the item is either. Small non-ferrous items are more prone to reading ferrous but even very large items will flip in very bad ground. Bury a two ounce nugget deep enough in bad ground, and it will read ferrous. The ground mineralization pulls the VDI numbers down, and the deeper the item is buried, the lower the VDI number gets until it passes into the ferrous range.

This happens with coins when coin detecting. A person using discrimination is looking for items that read in a certain number range. The problem is that mineralization pulls those numbers lower and then the items reads instead as a trash item, and is left behind.

The simplified explanation is the detector is seeing a little bit of non-ferrous signal and a lot of ferrous ground signal. The White's GMT is a rare machine that tries to show you this graphically. It will say a target has a 40% chance of being non-ferrous. Most machines have to call it one way or the other and in this example just go ahead and call it ferrous. Which is it? Ferrous? Or 60% chance of being ferrous? Would you dig something if you knew it had a 40% chance of being a nugget?

A picture says it all. See the one below. This is such a well known thing that White's has for a long time shown it on their simplified VDI (Visual Discrimination Indicator) scale. On most White's 1 through 95 indicates non-ferrous, and the negative numbers -1 through -95 indicate ferrous numbers. Notice how ferrous readings as low as -20 could indicate gold. Yet nearly everyone using any discrimination at all will tune out this range to eliminate finding small ferrous trash.

This happens on all VLF metal detectors that employ discrimination.


Good old Ganes Creek, Alaska is a VLF test bed on a massive scale. Tons of ferrous trash is buried intermingled with gold nuggets in tailing piles. The ground is not all that mineralized and VLF detectors work well there. Because the hunting was pay-to-mine competition style a VLF made more sense than digging hundreds of ferrous targets with a PI while your buddy was cherry picking nuggets around you with a VLF.

The reputation of the White's MXT as a nugget finder was largely built at Ganes Creek, but many other VLF detectors did well. I saw a couple things over and over at Ganes Creek.

First, we ran detectors in either of two modes. Dual tones with low tone ferrous, high tone non-ferrous was the most popular. Or there were some who ran in all metal mode then analyzed the target VDI once located. I did both.

In either case you could get two results very often. In my case running a F75 in all metal I would get a target. I would then sweep it and watch the meter VDI numbers. What I learned is if there were five sweeps that said ferrous and just one that read non-ferrous, then dig it.

Most people are looking for reasons not to dig. Getting five readings ferrous and one non-ferrous, most people would take that to mean a ferrous target. What I found more often than not is that if I could just get the target to read non-ferrous even once, it was worth digging. These were usually items reading borderline anyway. The only stuff relatively safe to walk away from is items that give strong ferrous readings repeatedly.

With dual tones the same thing applies. If you get five low tones but then get it to bounce high tone even once, better dig it.

The real proof was the simple kick test. You go along, get a ferrous indication, and kick a couple inches off the surface and bingo! the ferrous item turns non-ferrous. Pretty amazing stuff. Keep in mind folks, this is relatively mild ground!

What not to do. If you take any detector and tune it to completely ignore ferrous targets, you are in big trouble. The Gold Bug 2 for instance. If you flip to Iron Disc mode iron targets are ignored as if they are not there. Yes, some will click or pop but most simply get ignored. Any detector with a simple knob, like a Tesoro Lobo in disc mode, you can turn up until a small nail will be ignored. Hunting directly set up to reject ferrous is very problematic.

Now you can find gold doing this. I have and many people have. The problem is as you go along you only get one chance at the target. If you hunt in all metal, you will always get the target, and then you can analyze at your leisure, or just dig it up. In dual tone you will always be alerted to the target, so you can check it again.

But if you set to reject, and the detector makes a bad call on the first sweep, you pass over the target and never know it was there at all.

Part of the problem is in targets that you only get partly over on the first pass. For the discrimination to have its best shot, you need to be centered on the target as much as possible to get the strongest signal. An on edge pass will usually be wrong, but if you are alerted you can make multiple sweeps to get centered on target for the best reading.

Any detector running in disc mode will have a search field that is more limited in extant than that you experience with the detector in true all metal mode. In disc mode you need to be well centered. In all metal, the coil reaches wider and deeper to gather signals. The reason I usually run in all metal is it gives me the best chance of capturing the target, then I can decide what to do with it. Running directly in disc gives you more chance of missing the target entirely.

Keep in mind these issues vary wildly with the amount and type of iron mineralization in the ground. I saw some ground in Alaska recently that you would not think was very mineralized. No intense red colors, not much in the way of hot rocks. And yet there was something about the iron in the ground that made fairly large nuggets and even .22 shell casings read as ferrous when sitting directly on the ground in plain sight! Very, very scary stuff.

Now having said all that, there are times I will crank up the disc and eliminate the signals. Sometimes they are overwhelming in number and it is the only way to deal with it. Maybe I am just tired and not in the mood to analyze every target. Maybe time is very limited and I need to do a quick cherry pick run of the ground. There are no absolutes in metal detecting. The main thing is to have the knowledge required to make the best choices you can, to get the best odds for the situation. Hopefully this little article will help.

For more technical detail see Metal Detector Basics and Theory by Bruce Candy / Minelab. An excerpt:

In goldfields, discrimination is required only against ferrous targets, without any time constant discrimination, as gold nugget time constants include all values from very long to short.  Unfortunately, X discrimination in goldfields has several major problems:

Most productive goldfields are extremely mineralised, and thus the soil X signal is extremely large. As was stated earlier, it is only possible to assess the target X signal if this is comparable to, or greater than, the soil signal after filtering. In such extremely mineralised soil, this will only occur when the target signal is also very large which means the target must be close to the metal detector coil. Hence, discrimination in highly mineralised goldfields is only effective for targets buried at shallow depths.

The discriminator action must be very conservative so that gold nuggets are not falsely discriminated as ferrous targets. Thus, the metal target signal must not only be comparable or merely greater than the soil X signal after filtering, but significantly greater so that there is no doubt whether the metal target is ferrous or not. This further reduces the depths at which targets may be discriminated.”

Related Threads/Posts:

What Is Metal Detector Discrimination?

Target ID More About Size, Than Type Of Metal

Metal Detector Discrimination Really Sucks

Metal Detectors With Reliable Target ID Numbers

The Ferrous / Non-Ferrous Overlap

European Detectors Versus U.S. Style Detectors On Target Masking

Metal Detectors With Reliable Target ID Numbers

  • Like 4
  • Thanks 1
Link to comment
Share on other sites

Thanks Bob,

One thing bears repeating. Here is the same chart as presented by Garrett. Their VDI scale is similar to White's. I added the vertical yellow lines for emphasis.

Ferrous/non-ferrous overlap region

All the scales and charts will say the same thing. They tell you small gold and iron readings overlap. This is not actually true except in air testing scenarios. Almost all non-ferrous targets can read ferrous in mineralized ground regardless of size. It is not about the item being small, it is all about the signal being small or weak.

So bury that two ounce nugget a few inches deep, you get a big signal, like a big gold nugget. But bury it really deep, and now you have a signal just like that small gold nugget. And just like that small gold nugget, it can read ferrous. The ground signal is overwhelming the gold signal.

I want to make sure people get this because the manufacturers with their charts and scales all tend to imply this is only an issue with small gold items. It is an issue for ALL items, and not just gold. A silver half dollar will read ferrous if you bury it deep enough in bad ground.

The obvious solution is to use VLF all metal and dig everything, or better yet a pulse induction and dig everything. When nugget hunting this is recommended practice. But for most coin detecting in areas occupied by humans it is not practical for many reasons, and even for nugget detecting in some locations the amount of trash will try the patience and backs of the most die hard individuals. There is a place for VLF discrimination, but again, only as long as one is aware of the limitations. My best advice is to go more by site selection than anything. Use a VLF to deal with trash overall, but if you find good items in any area, shift to dig-it-all mentality. Any location that has good items worth digging is generally worth digging all items!


  • Like 2
  • Thanks 1
Link to comment
Share on other sites

OK, this is an attempt to visually quantify all this. It may of may not succeed in helping everyone but I had to give it a try. This chart I concocted shows increasing VDI numbers on the left, and how they can get pulled lower to the right by a combination of depth and ground mineralization.


The effect is not strictly linear. It depends on the detector and type of ground balance and discrimination system employed. The Minelab BBS and FBS units employ a method that tends to keep the VDI number as accurate as possible as deep as possible. They therefore enjoy a reputation for very accurate target identification. Unfortunately, they have lackluster depth in highly mineralized ground. That is not their fault - they are just eking out accurate target id as deep as it can be had, and in mineralized ground that is not very deep. In other detectors the effect is rather linear - the deeper you bury the item, the lower the VDI number.

It is a common misperception that the ground balance setting on a metal detector tells you how bad the ground is. It is only vaguely related to that. The ground balance setting is determined by the type of ground mineralization you are dealing with, but it does not directly report the amount of ground mineralization. In other words, you may be setting the ground balance to reject magnetite, the most common iron mineral in a lot of locations. What you do not know is whether you are tuning out a lot of magnetite or a little magnetite. Several Dave Johnson detectors like the White's GMT and the Fisher F75 and Gold Bug Pro plus Teknetics T2 have Fe3O4 amount readings to complement the ground balance reading. It is this combination of the TYPE of ground mineral as indicated by the ground balance number and the AMOUNT of ground mineral as indicated by the Fe3O4 meter.

From the Gold Bug Pro manual:

Understanding ground conditions assists the user in setting up the machine, knowing when to readjust ground balance, and in understanding the responses of the machine while searching.

This detector displays two kinds of ground data:

1. The type of mineralization (which affects where the ground phase should be set). This is GND PHASE

2. The amount of mineralization (the greater the amount of mineralization, the greater the loss of detection depth & ID accuracy; this loss is more pronounced in Discrimination Mode). This is Fe3O4.

The goal of ground balancing is to equate the GND BAL number to the PHASE number. PHASE is the measurement of the ground. GND BAL is the detector’s internal setting which calibrates the detector to the ground’s phase. Notice that the GND BAL number is three digits, with a decimal point. PHASE has only two digits. GND BAL is a higher resolution number, so may differ a bit from PHASE in a perfectly balanced scenario. After pumping and releasing GG, the exact measurement of the ground will be transferred to the GND BAL setting. The two-digit PHASE number displayed on the screen indicates the type of ground mineralization.

Some typical ground mineralization types are:

0 – 10 Wet salt and alkali

5 – 25 Metallic iron. Very few soils in this range. You are probably over metal.

26–39 Very few soils in this range -- occasionally some saltwater beaches

40–75 Red, yellow and brown iron-bearing clay minerals

75–95 Magnetite and other black iron minerals


The Fe3O4 7-segment bargraph indicates the amount of ground mineralization, independent of type, expressed as an equivalent volume concentration of magnetite (Fe3O4). It updates every second. It is sensitive to motion and will give the most accurate readings if you pump the searchcoil up and down several times over the ground. The presence of metal or “hot rocks” will cause the readings to be inaccurate. If you stop moving the searchcoil, the bargraph will go blank.


7 Bars -------- High over 1 over 2500

2 to 6 Bars -- Medium .026 - 1.0 61 - 2,500

1 Bar --------- Very Low 0.006 - .025 15 - 60

none -- less than .006 less than 15

Magnetic susceptibility is expressed in micro-cgs units. In a salt water environment in the absence of iron minerals, the bargraph indicates relative electrical conductivity. In soils with greater than 10,000 micro-cgs units magnetic susceptibility, the signal from the soil may saturate, or overload, the circuitry. This will not harm the detector but the machine will not be usable in that condition. The solution is to hold the searchcoil several inches above the soil surface so it is not “seeing as much dirt.” By listening and watching you will know how high you need to hold the searchcoil in order to avoid overload. The highest magnetic susceptibilities are usually found in soils developed over igneous rocks, in alluvial black sand streaks on beaches, and in red clay soils of humid climates. The lowest magnetic susceptibilities are usually found in white beach sands of tropical and subtropical regions, and soils developed over limestone.

The Fisher F75 and Teknetics T2 have a better defined Fe3O4 meter as explained in the T2 manual:

BAR GRAPHS Fe3O4 (magnetite)

This bar graph displays the magnetic mineralization factor, or magnetic susceptibility, of the soil. Magnetic susceptibility is expressed in terms of the percent volume of the iron mineral magnetite, which most black sand is made of. The depth to which objects can be accurately identified is strongly influenced by the magnetic susceptibility of the soil. High Fe3O4 values have a greater effect on detection depth in the Discrimination mode than in the All Metal mode. For the most accurate Fe3O4 reading, pump the searchcoil as though you were ground canceling.

Fe3O4 approx. Range micro-cgs Description

3 --- 7,500 --- uncommon but not rare, heavy mineralization

1 --- 2,500 --- heavy mineralization, not uncommon in goldfields

0.3 --- 750 --- heavy mineralization, but not uncommon in some regions

0.1 --- 250 --- medium mineralization, typical

0.03 --- 75 --- light mineralization, but common

0.01 --- 25 --- light mineralization, often low G.C. setting

blank <14 --- quartz & coral white beach sands

From Bruce Candy at https://www.minelab.com/__files/f/11043/METAL%20DETECTOR%20BASICS%20AND%20THEORY.pdf

"Thus, USA goldfields are typically different from Australian goldfields:

• The USA soils are mostly mildly mineralised but in some areas may contain either nearly pure magnetite black sands or rocks, which are problematic for metal detectors as they have very high X components (strongly attracted to magnets).

• Australian gold fields have highly mineralised soils, but very few black sands or rocks that contain nearly pure X magnetite. The magnetic materials are in the forms of magnetite-rich small pebbles and rock coatings, clays and general “sandy” soils. These all contain magnetic materials that produce high levels of X signals as well as R. The ratio of X and R is random, and the R component arises from extremely small magnetic particles called superparamagnetic materials..."

Predicting Soil Influence on the Performance of Metal Detectors: Magnetic Properties of Tropical Soils http://www.jmu.edu/cisr/journal/13.1/rd/igel/igel.shtml

Influence of Soil Properties on the Performance of Metal Detectors and GPR http://www.jmu.edu/cisr/journal/17.1/RD/takahashi.shtml

  • Like 2
Link to comment
Share on other sites

Final summary - the higher the magnetic susceptibility of the soil, the less accurate VDI numbers are and the lower they will tend to be pulled with depth. For coin hunters that means an item reading as a pull tab may be a silver coin. And for nugget hunters, that ferrous reading may very well be a gold nugget.

  • Like 1
Link to comment
Share on other sites

Question... shouldn't those that detect learn to listen to what their detectors are telling them vs. what they are showing them? My GMT has the LCD readouts, but it's very rare that I even look at the screen. I try to tune my ears to the ground rather than my eyes.


I would think it would be the same for the coin and relic hunters. Don't those detectors sound off differently on the various types of metals and ground mineralization?


Call me crazy, but I think all those fancy smancy digital readouts and color LCD screens are nothing but marketing hype. Am I right or wrong?  :blink:

Link to comment
Share on other sites

Like I said, go by tones if you want. The detector decides it is ferrous, gives you a ferrous tone. Same difference. In other words, the detector determines the target id number first, then displays it visually as numbers or audibly as tones. If the underlying decision is wrong, the number or the tone will be wrong.

I agree hunting by tones is better as you get fleeting audio indications and nuances that say a hundred times more than any number can. Digital read outs as worthless hype? Everyone is entitled to their opinion on that and what works for them. They are not worthless for me. Not everyone has great hearing and many people are tone deaf. I love the displays on my V3i and CTX 3030. Whether they make me hunt any better is debatable. My personal preference in VLF detectors is to have it available in addition to tones to use as I see fit as opposed to not having it at all. Note the key there - personal preference.

I like hunting in multi tone modes as it paints a tonal picture of targets as opposed to dual tones, which gives a binary high low way of hearing things. Visual displays are the same, in that simple systems just put a number on screen. The V3i or CTX paint pictures on screen that display signal intensity and cumulative effect of hundreds of VDI samples. In the case of the V3i you get a picture of the same item in three frequencies. All this can be used by experienced hunters to get very good at deciding what to dig or not dig in urban trashy parks. Frankly, it is also kind of fun to play with.

But is it better? Those very rare times I am just trying to dig silver coins, yes, it is. But much of my detecting boils down to just digging all non-ferrous, and so those systems are overkill. I still enjoy them though. The entire point of this article however is that even just digging all non-ferrous can get problematic.

Here is a great video that illustrates how even listening to tones is a problem in really bad ground. I read all the time about people digging coins back east at 8" or 9" or 10" plus. Yet everywhere I go it seems 5" - 6" is about it for accurate discrimination with the best machines. Here we have a video comparing four top notch machines - the Teknetics T2 (pretty much same as Fisher F75), Garrett AT Pro, Minelab CTX 3030, and XP DEUS. Now the goal here for me is not to show any machine is better than another. Just focus on what is going on with a large non-ferrous item buried about 5" deep in hot red soil. The machines struggle to identify it as non-ferrous just going by tone. Target id numbers are all over the map. Then toss just a single nail on the surface, and it gets even worse. I am not a big fan of comparison videos in most cases because they are so easily manipulated but this one clearly shows how bad ground and discrimination can really cause good items to be missed, and why ultimately using a VLF in all metal or a PI and digging it all is the only way to get it all.

Link to comment
Share on other sites

I dig all the nails I can stand. Especially in the mine tailings area. Those old wooden sluice boxes were nailed together and gold nuggets worked their way into the cracks. Nails mean gold to me.
  • Like 1
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Create New...