AUDIO TARGET CLASSIFICATION ... or ATC
Helpful Tips for Non-Metered and Metered Detector Operation
It should be acknowledged that each brand of detector has some operating characteristics that are unique to their design. Therefore, not all of the following suggestions will apply to all detectors, although most will assist anyone who searches with a motion discriminate detector. This is slightly altered from a forum post I made several years ago in response to some posts for those who had questions about using several Tesoro models, however, it really can benefit ANY operator of a good motion discriminate detector. Most of these tips and techniques stem from my field use of non-metered detectors through the years, relying on only the audio response, but they can benefit those who use metered Target ID detectors as well.
I started doing metal detecting seminars in 1981 in which I incorporated several terms, such as Audio Target Classification, or ATC. We all know that a detector's visual target identification is only somewhat accurate, and in-field use might only fare 15% to 35% correct for coin and jewelry hunting. Desired targets might be too deep, at odd angles, mixed with one or more additional targets, or some other masking problem to prohibit the TID circuitry from being right all the time. Let's not forget the number of 'copy-cat' targets that are out there as well. The good news for those who want to learn is that there is more audio information available than they might realize. Most of us are coin hunters, so we therefore search for coin-like target signals. All of us, regardless of the type of detecting we engage in, might be interested in knowing how to help classify some targets to help in the dig, no-dig decision making process. Here are some tips I hope will be of help:
SIZE & DEPTH ESTIMATION: In a metal and target-free area, place a small, medium and large coin on the ground. Using the Disc. mode, with your discrimination set at a point to just reject a surface nail, and any manual ground balance adjustment made, sweep over each target at about 2" and "listen" to the response. Then, raise the search coil while making consecutive passes directly over the target, noting what the audio sounds like as the distance is increased. Finally, note the maximum distance (depth) you can achieve. Do this with rings, too, and you'll note similar results, except a little less depth, perhaps.
Repeat the testing using various types of junk you might encounter, such as a crushed aluminum can, aluminum screw cap, etc.
Now, if you get a good, repeatable signal when searching, simply raise the search coil while making a few sweeps across the target to compare the response with your "best possible" in-air response. If you can get a coin at "X" inches, yet you can raise your coil 4" to 12" more than that over a located target, then it is a larger object and not a coin-sized target. You have now used a form of ATC to Audibly investigate a detected Target and then Classify it as either a potential coin-type target or too larger to be a coin-type/size target.
NARROW vs WIDE TARGET SIGNAL: This popular method is done in a conventional, threshold-based All Metal mode. Once a target is detected and pinpointed, the search coil is moved to the side so the coil's edge is past the target. Switch to the All Metal mode and slowly sweep towards the target noting where the leading edge of the coil is when you hear a response to the target. Then, do the same from the other side of the target, again noting where the leading edge of the coil is when a good signal is heard. The operator visualizes these two 'marked' spots where the leading edge of the coil was when a signal was heard, and if they are almost over-lapping or within an inch or so, it is a narrow target signal. A coin-like target signal. If you get a noticeable increase in audio starting when the coil's leading edge is out from the pinpointed spot by a couple of inches or more on either side, it is either a larger target, such as an aluminum can, or an iron-based (magnetic) object like a round washer, etc.
This simple technique to Audibly analyze a Target and Classify it as a potentially coin or ring-sized object or perhaps larger, unwanted junk, has been in use for quite a while, but it is still a valuable technique that is quick and easy to use with either a metered or non-metered detector. This same audio technique is used to "size and shape" a located target as well.
PROBLEM TARGETS and "QUICK-OUT": A good, old-fashioned TR-Discrimination type detector will easily reject a bottle cap, rusty or not. Other very low-conductive trash can be rejected very cleanly with a TR- Disc. model because that circuitry design is what I refer to as a "true, progressive" discriminate design. Nails, old iron bottle openers, hair pins, and perhaps the most annoying of all, the crimp-style bottle cap, can all be rejected quite easily.
The more modern motion discriminators, however, have some difficulty with a lot of rusty iron just, but the biggest headache for many coin-hunters is that blasted, annoying bottle cap! I am NOT referring to the taller, higher-conductive aluminum screw caps, but to the older pry-off and current twist-off bottle caps with the scalloped, crimp-on edge. Often you will get a rather good-sounding audio signal from a bottle cap in the discriminate mode. If the response is from what appears to be a fairly shallow target, say surface to 3" or so, then try the All Metal mode test for Narrow vs Wide signal. Generally, the rusty bottle caps will produce a wider signal than a coin-type target. If you still question the target, pinpoint the target more precisely and then use a technique I have referred to as "Quick-Out." This is where the operator uses a short but quick sweep across the well centered target and notes the audio response.
Give a more "brisk" sweep across the target dead-center. (Note: It should be a SHORT and brisk sweep to avoid covering multiple targets.) Quite often, a bottle cap or other problem target with magnetic properties will be "kicked-out" with this quick-sweep technique. It works best on targets that are within the typically shallow range, perhaps from surface to 3" or so, depending on the mineralization.
As stated, it will often, but not always, audibly reject these problem targets, but it only takes a couple of quick passes to try and "classify" the target this way. Although the audio response might still be there, those with metered units might note that the ID is now reporting "iron" or at least "jumpy or unstable."
EDGE PASS REJECTION: Like ATC and "Quick-Out", I coined another term to refer to techniques I've used since the early '80s with motion discriminators. Edge Pass Rejection, or EPR, can be used with "Quick-Out" techniques, if necessary, for really challenging situations.
Lay a couple of coins and a couple of bottle caps on the ground, spaced far enough apart so as not to interfere with a coil's sweep. Use a newer crimp-style (crown-type) bottle cap, and older one, and one that is rusty. Then, sweep the coil over the coins, dead-center, at about a 2"-3" height. Note the good audio signal. Continue to sweep over the target coin as you draw the search coil back towards you and note any audio change. You'll see that most of the time a coin will continue to signal well until its position is just in from the edge of the coil (perhaps an inch or so). Now, advance the coil and note that the same is true in from the back edge of the search coil. A coin-type target in a typical coin and trash target range will respond well when it is just in from the outer edge of the coil by about an inch or so.
Repeat this test with the various bottle caps on the ground. Dead center, with a slow, comfortable sweep, you will probably get a good audio response. However, you will note that as you back the coil off while sweeping and get near the "edge" where the coin still sounded off well, the problem trash item will usually be rejected! Advance the coil while crossing the junk target and you will note that the same result occurs from the back edge of the coil. Where the coin was still responding the trash is often rejected!
On some really difficult targets, like a few pesky bottle caps, combining "Quick-Out" and "EPR" will "classify" them as undesirable ferrous trash targets. Please remember .... the sweeps across the targets in question should be brisk and short so as not to cover additional near-by targets. In some sites, (such as drive-in theaters or picnic grounds or fairgrounds) these techniques are very helpful, especially when using a smaller coil, such as a 4" to 7" size, to handle the trash problem.
Once again, those who use any form of metered ID will benefit from this technique as it will tend to make the "jumpy" or "bouncy" target readout register and lock-on as an iron object. Perhaps a bit jumpy when you try "Quick-Out" directly over the target, but on the more problem bottle caps you use that together with "EPR" and these two techniques will almost always provide the operator with an accurate or proper classification of the junk target. Coin targets, on the other hand, will not produce the lock-on iron reading.
DISCRIMINATE CONTROL ADVANCING: This is a somewhat popular technique, but one which I seldom use and caution against. With this method an operator will set the discrimination low to allow the detector to respond to most targets and then, upon receiving a signal, they make MANY continuous passes across the target as they slowly increase the discriminate control to note where the target "falls out" and then decide whether it is maybe a nickel or maybe a tab or maybe a ????
I do not like this method for several reasons, one of which is because it's quite time consuming. Earlier explained techniques are done with only a flip of a toggle switch or push of a button to go into All Metal (to determine a Narrow vs Wide target response), or else with only a couple of quick sweeps over the target in the motion Discriminate mode.
The Disc. control-advancing method often requires two hands, a lot more time, and with some makes and models there can be some other 'changes' that take effect other than simply increasing the detector's Discriminate level. I have used some models that have been designed (I would bet by error more than by design) such that increasing or decreasing the Discrimination level can actually increase or decrease the Ground Balance setting. Honestly! You might manually GB to be 'right on' in All Metal for pinpointing with the Disc. level set at minimum or just to reject iron nails, yet increasing the discrimination increases the GB setting as well. I have used one that reduced the GB setting. No, not all makes and models are designed with such quirks, but there have been/are some on the market that do.
Another reason I do not care for this method to attempt to audibly classify or identify a located target is that some detectors will not produce a 'proper' audio response when the control setting is being shifted and the coil is in motion at the same time. I've used several makes and models in evaluating this technique and I found that you can hit a point of rejection during the adjusting/sweeping combination that is actually below or before the true rejection point. For example, you might be sweeping and increasing the discriminate control and a target drops out before what would normally be a 5
