Most of you are probably skeptical about a review written and sold on the open market. Let me assure you, this was done neither for fame nor fortune. I am simply doing it as a service to all detectorists who I feel deserve honest and factual information. My background is in electrical engineering and when I first got involved in this hobby, I was appalled by the lack of data available to assist in the selection of a good detector. Quite to the contrary, this hobby is beset by more that its share of charlatans and snake oil peddlers. So some three years ago, I set out to learn as much as I could about detectors. I read literature, studied patents, looked at schematics, and read every book I could get my hands on. It was a lot of work (and fun). My wife says it is an obsession. It is in this spirit that I plunged into these reviews. I started by contacting all the major vendors of detectors, most of whom I already knew from my previous studies. Most of them were very supportive of the project. I asked them to contribute their favorite tests and knowledge. Two of them also volunteered to supply me with evaluation detectors. To ensure that these detectors were not “enhanced” in any way, I carefully compared their performance to detectors that were bought off the dealer’s shelf. There were not any noticeable differences. When the tests were finished, I returned the detectors to the manufacturers. I neither asked for nor received any monetary or other consideration from any manufacturer to avoid any conflict of interest. I am confident you will find these tests accurate and, I hope, worth reading. I welcome your comments.
Note: It is assumed that the reader will have a working knowledge of the terms common to detector use and operation. If you are less than comfortable with these terms I suggest that you get a copy of the best book I have ever read on the subject, “The Detectorist” by Bob Sickler. This is a great book not only in explaining the technical details, but Bob also has included many tips that I have not seen elsewhere.
The detectors were chosen for these tests by popular demand. They are the Fisher CZ6a, The Garrett GMH CXIII, and the White’s Spectrum XLT. Most of you requested these models because they represent the top of the line units from each of the manufacturers. With retail prices up to around $1000, they are also expensive and as such represent a large investment. Make a mistake here and you might be making a costly one. You will note the absence of Minelab and Tesoro from the tests. The Minelab was not tested because it is not as popular nor as available. The Tesoro Toltec was left out because it was not requested and because the manufacturer preferred that this detector be compared with others more in its price class. These detectors share a lot in common; yet as you will see, there are distinct differences that you are unlikely to discover by reading the data sheet or even by testing a detector at the dealer’s shop. The Spectrum XLT by White’s is by far the most versatile detector in these tests- every parameter of the detector being user adjustable. In the hands of a skilled user this allows superior results under a wider variety of conditions than the other detectors. This versatility does not come without a price though. The XLT requires a thorough knowledge and can easily be mis-adjusted to produce rotten performance. It does have a number of preset programs that work quite well and are easy to use, it is only when the user THINKS he knows what he is doing but doesn’t that he really gets into trouble. The Fisher CZ6a is very easy to use by comparison but interestingly the preset positions of the knobs produce mediocre results. With just a little knowledge the CZ 6 and its cheaper cousin the CZ5 are capable of great results. The Garrett CXIII is mid way between the CZ and the XLT in terms of both versatility and ease of use. It is a menu driven instrument like the XLT, but there are a lot fewer menus and choices. The preset settings work quite well, and the machine even talks to you about the settings and what you have found.
The tests are a mixture of those suggested by the manufacturers, those invented by some of the local detectorists, some that I read about in magazines, and a sprinkling of those I dreamed up myself. Most of the tests were conducted in my backyard garden which is moderately mineralized clay common here in Houston, Texas. The garden is in the suburbs and the electrical noise level is quite high in part of the garden. This garden contains rows of pennies and nickels buried at depths from 2-8 inches. The targets have been buried more than two years and so have had some chance to create a halo. In addition some of the tests were repeated in heavily mineralized red clay from East Texas, and in salt saturated sand with some black sand mixed in as found on the Gulf coast. These latter tests did not employ buried targets but used the same technique used by Fisher, which is a long slanted PVC tube buried in the ground with a sled on which the target is placed. This allowed me to vary the depth from 0-12 inches and also the orientation of the target in the ground. I conducted testing with the target parallel to the ground as well as at a 45 and 90 degree angles.
Whenever possible, I made measurements using accurate distances and signal levels measured with sophisticated instruments. Some tests were subjective, but when this was necessary, I tried to use arbitrary but constant points of reference. For example, maximum depth with correct ID was that depth that produced accurate ID on 50% of the sweeps. I will try to explain each of these assumptions as I come to them.
What you will not find in these tests is a listing of all the targets I found while testing. Since most of my time was spent in a test garden and I didn’t want to dig up my test targets, the finds weren’t all that interesting.
Ground balance is that function that allows detectors to mostly ignore the soil matrix that they are searching in and consider only the targets. The way this is implemented will strongly affect the maximum depth that a detector will deliver. This becomes more important as the mineralization of the soil increases or the sensitivity of the detector is turned up. All of the reviewed detectors have ground balance that is user adjustable. The CZ6a has a manual adjust that works in both motion and non- motion modes. It provides a simplified technique to do this that Fisher calls push-button ground balance. This only requires that the user raise the searchcoil 6-12” off the ground, push and hold a button, then lower the coil to search height and adjust the ground control until the tone just goes away. This works OK most of the time, but didn’t work well for me under mineralized conditions. There I resorted to the well known bobbing technique which was almost as fast and proved more reliable. The CZ also features a salt switch which extends the range of the ground control to balance wet salt saturated sand. The Garrett and the Spectrum XLT both have automatic ground balance which have sufficient range to accomplish the same goal. This is not “preset” by the factory like most that claim “automatic ground balance” but rather a true microprocessor controlled adjustment. However, in the motion mode only, the CX III reverts to a factory preset for ground balance. This will reduce its depth in soils other than those it is preset for. In the all metal non-motion mode one only has to push a button twice on the XLT or once on the CXIII to properly balance the detector. The CXIII must be bobbed up and down when this balance procedure is carried out. The XLT typically requires 8 seconds, whereas the CX III is much slower- taking 15 seconds or longer. It has been argued that a skilled operator can ground balance a detector more accurately than a computer. This may be true, but even if it is as soon as the ground changes the manual GB detector will be mis-balanced. Both the CXIII and the XLT “track” the ground and so are very close to optimized at all times. This is an advantage. The XLT will also allow positive and negative offsets in ground balance for special purposes such as nugget and beach hunting. Ground balance worked well on all machines except at the beach, where the salt saturated sand dramatically cut the depth on the CXIII and to a lesser extent, the XLT. Neither machine was able to completely balance these conditions. The CZ6a loved the beach and went even deeper there than on dry land. This is largely due to its SALT mode switch which moves the range of ground balance so that it can compensate for the positive response of salt.
Threshold is the background audio tone that is heard on the CXIII and the XLT in all modes of operation. The CZ6a is a silent search machine with no tone in the discriminate mode and has a threshold only in all metal mode. It has been a point of contention that hunting without a threshold is convenient and less distracting but is less sensitive due to the fact that without a tone the operator does not know how far down below audibility the threshold really is. Therefore some small signal is required to raise the level until it can even be heard. I do not believe that this is a factor any longer as very little depth appears to be lost in the CZ. It seems to be stable enough so that the internal setting is just fine for normal use. Threshold has an additional function in the XLT. It also tells the operator if a target has been rejected by temporarily going silent. This is considered to be a benefit because it alerts the operator to the number of targets he might be missing. This might be an indication to sweep slower, choose a smaller coil, or lower the discrimination level. This helped me identify a masking problem with the CZ. (See details in the performance section). It needs to be pointed out that the CXIII and the XLT can both be set for silent search. The CZ in the all metal mode has an automatic threshold that self adjusts. This is referred to as auto tune. Both the CX III and the XLT share this characteristic. This mode is good for areas where you are trying to achieve maximum depth and/ or the ground has a high mineral content. It has the disadvantage that it will autotune out the target if you attempt to hold the coil motionless, as when pinpointing. Autotune can be defeated on both the CXIII and the XLT to give a true non-motion mode. Additionally the XLT allows a wide range of autotune speeds for more flexibility when searching.
Sometimes mistakenly called power, this function has nothing to do with how much signal is sent into the ground. Rather, it controls how sensitive the detector is to returned signals. Almost every detector on the market that costs more than $150 has a sensitivity control and they all function much the same. The sensitivity should always be set as high as possible without causing false signals. It is common for many users to set this control too high causing many false detects. The CZ has only one sensitivity control that adjusts both the Non-motion (all Metal) and the motion (discriminate) sensitivity simultaneously. The XLT and CXIII allow the user to adjust these independently, and this is an advantage in that a different setting can be selected to optimize pinpointing of deep targets or for other reasons. The XLT additionally has a control called signal balance. This control allows the user to adjust how much the return signal is amplified before it is sent to the signal processing circuits. In good ground, away from stray electrical signals, this allows the XLT user to achieve a little more depth than would be possible with a fixed gain. Like many of the controls on the XLT, it is possible to increase the amplification to ridiculous levels that render the machine unusable.
This is probably the most important and overused control on a detector. Interestingly, no matter how fancy or simple a detector is, they all discriminate in the same way, based on the conductivity of the target. The way this function is implemented will cause the feature to perform with different degrees of competence. All of these detectors provide powerful discrimination compared to their less expensive cousins. The CZ has a simple control that looks like a normal discriminator. However, it doesn’t work the same. The targets are arranged more or less according to their desirability and thus this simple control allows the user to reject pull tabs and still get nickels. This in the past required the user to use a second control called a notch to reject part of the conductivity spectrum while retaining targets both above and below the notch. This function is implemented in such a way as to prevent the ID from disagreeing with the selected discrimination. In fact, if a target is discriminated out, it will not ID on the display. This is a slight disadvantage, but is an outgrowth of the implementation on the CZ. This is offset by the fact that discrimination has almost no impact on depth. It does have the further disadvantage of versatility. For example it is not possible to reject coins and dig only rings. It is probably adequate for most applications and it is easy to use. The XLT has 191 discrete zones in the conductive spectrum that can be set to accept or reject. This system, while versatile, is not easy to use because there is no quick way to tell what you have set without scrolling through the whole number range, one at a time. This is more resolution than is necessary as most targets in soil are probably +/- 5 zones wide anyway. The Garrett CXIII has perhaps the best discrimination system going. It divides the targets into 24 zones and allows individual accept or reject of each zone. This is enough to give good selectivity and still be manageable. Further, it employs a graphic system of target selection and ID that is visually intuitive as well as easy to use. Finally it allows the user to switch rapidly between 2 complete detector setups that include all control and discrimination settings. This is the single strongest feature of this detector and an important one. On all detectors it is important to use the least amount of discrimination possible for several reasons. The first is that you can unknowingly reject good targets (like rings) when you reject the common pulltab, the second is that it takes a finite amount of time for the discriminator to recover after rejecting a bad target before it will accept a good one. Since junk is frequently found around desirable targets, it is possible to miss a good target near a rejected one. This effect is called masking and is checked in the performance section.
Target ID is visual form of discrimination. The target conductivity is presented on a meter on the CZ6 and a Liquid Crystal Display (LCD) on the CXIII and the XLT. The meter is generally easier to see and has icons to describe the various probable targets. The LCDs allow additional information to be displayed. For example, the XLT and CXIII can display several targets simultaneously. This is an advantage because it is possible for multiple targets to be under the searchcoil at the same time. They also are less likely to be fooled by iron, which often shows as a smear of targets across the display. Target ID is a good feature to have because it also allows you to use lower than normal discrimination to avoid the problems associated with masking as described above. In addition, the CXIII talks to you and audibly describes the target while the XLT draws pictures (icons) of the probable identity. These last two features slow the operation of the detector substantially. Fortunately, they can be turned off in both machines. Another useful feature is audio tone ID where the pitch of the audio tone indicates the target type. In the CZ6 this is done with three tones - low for iron, middle for most trash (and most rings), and high for coins. The XLT has a large number of different tones that it uses with pitch proportional to conductivity. The CXIII has several forms of two-tone audio which divide the conductivity range into 2 areas. In Bi-Level, nickels and up will sound a high tone. In Bell-tone, pennies and up ring a bell-like tone that Garrett calls the sound of money. It is possible to turn on both bi-level and bell tone simultaneously and get three tones. The CXIII and CZ6a implementation is superior. They have enough resolution to separate the targets without a confusing number of tones. The XLT and CXIII allow tone ID to be turned off. As attractive as it is, target ID is not perfect; neither is discrimination. As long as the targets are shallow and the ground is wet, ID works perhaps 80% or more of the time. For targets below about 6”, particularly in dry soil, ID is frequently in error usually producing iron results. All of these detectors experience this problem to some degree.
This is an all-metal non -motion mode that is the most accurate way to precisely locate a target. All models have VCO pinpointing, where the pitch of the tone rises as the coil nears the target. This is a superior technique, but bothers some people who say it sounds like a fire engine. The XLT and CXIII allow VCO to be defeated and to be replaced by a rise of audio level only. VCO is generally a better technique because the ear is more sensitive to small changes in pitch than it is to small differences in level. The CXIII is the best overall pinpointer with very quick audio and a narrow response. It is followed by the CZ6, which is generally as good except if a nail is the target. Nails often pinpoint off to the side of the coil. The XLT does not pinpoint quite as well. With practice and by detuning, it does a perfectly acceptable job.
Closely tied to pinpointing is depth reading. All of the models provide similar methods of estimating the depth of coin sized targets with the meter or LCDisplay. I found the reading to be accurate +/- 15% or so under most conditions. The meter or display used varies substantially in its responsiveness. The CZ and CXIII are quite responsive; the XLT is much slower and I find it annoying.
The CXIII provides a method to reject strong (shallow) signals. The rationale of this is that shallow signals are likely to be trash. I think there will be limited circumstances in which this would be true, so I did not thoroughly test this feature. Fortunately it can be turned off.
Under some circumstances of heavy electrical interference from radio and TV stations and other sources such as nearby detectors these models can produce false signals. The CXIII and the XLT allow the user to slightly shift their operating frequency and thereby avoid or at least minimize the interference. In testing, I had no opportunity to evaluate thus feature.
The XLT and the CXIII can shift the audio frequency that the user hears as a response to a target. This may help is some cases of severe hearing impairment, but is a relatively minor feature.
This is a mechanism to increase the volume of faint, deep signals while maintaining the loudness of shallow targets. All of these detectors have this feature, though it seems to work the best on the CZ6a. There is very little effect on the XLT; it seems to signal at the same volume regardless of the depth and whether the feature is turned on or off. The CXIII boost also has little effect, though more than the XLT. I prefer to leave boost off so that deep signals are weaker. The CZ6a has a good volume contrast between shallow and deep targets in the boost off mode.
Because many headphones have no volume controls, these machines have all included a separate control. On the CZ6 this is combined with the boost control and is a little less versatile.
When a target is brought too close to the searchcoil, the electronics are overloaded with signal. While in this state, they cannot detect other targets. The CZ6a rings a telephone sound when this occurs. This is a good feature but it sometimes rings on good targets which might cause you to skip them. The XLT indicates OVERLOAD on its display, but continues to function if possible. This is a better solution, but requires that the operator look at the display. If the ground is very mineralized, which can cause a constant overload condition, the XLT is the only model that will allow the transmit power to be reduced to reduce or eliminate the overload condition. This allows continued searching but at reduced depth. The CXIII has a similar overload warning.
The architecture that the instruments employ strongly affect how fast they can respond. To reject the ground (motion detector) detectors use what are called ground filters. These are classified as 2 filter or 4 filter. Two filter detectors respond faster than do four filter models. They work best on low to moderately mineralized soils. Four filter models generally are employed when the soil mineralization is heavy. These generally do a better job of ground rejection, but at the expense of response speed. Further, 2 filter models can be swept slower (so called slow motion) and still retain sensitivity. Detectors that use microprocessors, while generally more versatile, are also typically slower to respond. This is true because the computer has to make a decision whether to signal or not and how. The XLT is a four filter microprocessor based detector. It is quite slow to sound. This design works well when you are hunting deep targets in mineralized soil such as might be typical of relic hunting in the South East. The CZ6a, on the other hand, has two filters, no processor and is therefore very quick to respond. This is helpful when the area you are hunting is trashy. The CXIII is a two filter, micro based design and defies the rule that the microprocessor slows response speed. It appears to have the quickest audio in this group by a small margin. It is slightly noisier under normal search conditions, but this is not a problem.
All of these detectors have battery test facilities. In the CXIII, the batteries are tested continuously and the status displayed. The CZ6 and the XLT test their batteries only at the operator’s request. The Garrett system is better because you simply have to look down to see if your batteries are about to go dead. It even will talk to warn you of low batteries. The accuracy of all systems was found to be acceptable.
Each detector was tested for power consumption with all controls set for maximum sensitivity and volume and with a pair of 100 ohm headphones attached. The estimated times of operation are: CZ6a-18 hours per set of (2) 9 volt alkaline batteries, CXIII - 20 hour per set of (4) C size alkaline cells, and XLT- 10 hours per charge on (8) nicads in a pack (included standard). The XLT is also provided with a battery holder for alkaline cells. Using this the runtime would be about 20 hours. Of course Nicad operating costs per hour are perhaps 1/10 of alkaline cells, but there is less hassle with alkalines.
These machines all support a variety of Searchcoils that increase the usefulness of the detectors. See the Sickler book for a thorough discussion of the benefits of each type and where it might be used. The XLT has a 3.5” coaxial and a 6”, 8”, and a 15” concentric as options. These all supplement the 9.5”, open center concentric that comes standard and is a good general purpose coil. In addition Jimmy Sierra markets several special purpose coils that extend the versatility of the XLT. The CXIII comes with a 8” concentric and you can add a 4.5” coaxial as well as a 12.5” concentric. It is also possible to fit a 3x7” or 5x10” elliptical double d coil. These work well when the mineralization is high, but don’t discriminate as efficiently. Lastly, there is the depth multiplier which effectively converts the CXIII into a 2-box unit capable of great depths on larger targets. This complement of coils adds greatly to the versatility of the CXIII. The Fisher CZ6a comes standard with a nice 8.5” concentric spider which also woks well in the water. Optional are a 10.5” concentric spider and a 5” coaxial.
The XLT tips the scales at 4 lbs. Though White’s brags about the major weight reductions they have delivered, this is still the heaviest detector in the group. Next comes the CXIII which is only a few ounces lighter at 3 lbs. 14 oz. The lightest detector is the CZ6 at 3 lbs./ 11 oz. But weight does not tell all the story; balance is a major factor also. The center of gravity on the CZ6a is about 20 inches ahead of the arm cup making the detector feel hard to swing. This torque can be computed and is 59 oz. times 20 inches or 1180 inch oz . This same number for the XLT is approximately 8 inches time 64 oz. or 512 in oz. Said another way it takes twice as much energy to swing as a CZ6. The CXIII uses a different method to swing which involves twisting the wrist. Because the balance point is near the handle it doesn’t require much torque to swing it. However the wrist is not very strong and I found it to be the most tiring to use. There is an optional armrest available and it provides a swing torque slightly higher than the XLT. The handle is then at the wrong angle because it forces you to bend your wrist downward which I found awkward. No hipmount kit is available. The XLT and the CZ can be hipmounted. This reduces both the weight and the swinging torque by about 1/2 but does make you wish you had three hands (one to swing the coil, one to manipulate the controls and one to hold the digger). In this configuration both machines require about the same amount of energy to swing. The CZ6a is much easier to hipmount and it requires only a belt, whereas the XLT requires a $50 adapter.
This is the single most important factor that is largely subjective. The CZ6a is a classic layout with knobs for every function. This limits the maximum number of functions available, but makes learning dead easy. I was able to understand all the controls in less than 10 min. This arrangement also makes it easy to see how your controls are set. The CXIII is a menu driven instrument with pushbuttons only for major functions set as mode, pin point, and setup of discrimination. Each menu item is accessed sequentially, but once you are through with the menus the status is available continuously by observing the LCD. This is a good compromise between versatility and simplicity. It took me about 15 min. to understand all the menu items. The CXIII also strikes the best balance between control and ease of use in the discriminate function. The XLT is the most adjustable detector in the group and also the hardest to get comfortable with. It also is a menu based instrument, but everything is in menus with the only exception being pinpointing and the last function that you have used. Worse, there is no status screen that shows all the settings so you have to go from menu to menu to see how you ( or a preset program) have set up the detector. There are about 50 menu items that can be set so this is not a simple learning experience either. It took me about 4 hours to learn all the functions provided in this versatile instrument. The only saving grace is that if you are willing to live with not knowing how everything is set, you always have the option of calling up a preset which works well in all but the toughest conditions. If only the XLT had a status button and screen!
Each of these machines is equipped with a way to set the detector to a known state- settings that are likely to work under a variety of typical conditions. On the CZ6, these are simply marks on the knobs that point to typical settings. These presets unfortunately yielded poor results, but can serve to get the new user going. The CXIII, being menu based, has more options: It has 6 sets of motion discriminate modes that store discrimination settings. Three of these can be altered as the user desires, and the changes will be retained even if the power is shut off. It also has a single button that allows you to toggle between any two of the stored settings. This is a versatile and easy to use system and the presets are quite good . The XLT has the most sophisticated setting storage system. It has 9 preset programs that store all the detector settings, not just the discrimination settings. Of these four are user definable and can be named anything that makes sense to you. This greatly simplifies detector setups since there are so many adjustments you can make on the XLT. Likewise, the presets are quite good and many users never go beyond them and program the detector. Programs are published from various sources that you can try in your hunting environment.
Only the CZ6a claims splashproofing. This means splashes of water will not damage the detector. It should also be OK to hunt in the rain if you use the waterproof headphones. Neither the XLT or the CXIII claim that they can be used in such an environment and I would agree. Detectors live a hard life. I would suggest using a cover on both of these machines if you are not a fair weather hunter.
(All summarized in 3 tables, CZ6a, Garrett GMH CXIII, White’s XLT)
Probably the single most asked question I get is: “How deep will it go?”. Traditionally this is not a question that can be easily answered because “ it depends” on soil, moisture, degree of target oxidization, and many other factors that are highly variable. In fact, this was probably the number one reason why I set off to write this report in the first place. The same variables would exist for me of course, but because I could test all the detectors at the same time under nearly identical conditions, I could do an accurate relative comparison between them. Further, since I tested in several different types of soil I could get a good relative sense as to how they would perform under a broad range of conditions. The most comprehensive testing occurred in a moderately mineralized soil in my test garden. It would have been nice to have tested as thoroughly in other soils but I could not control the conditions well enough to make the tests valid. In order to simulate mineralized soil, I filled a large 25” high 24” in diameter plastic container with heavy red clay from East Texas. This was the largest amount of soil I could get into my car without breaking my back. This did limit me to depth testing only because there was not enough horizontal distance to bury two targets for masking tests. I’m sure it would be possible to get a worse soil, but I could not get my hands on any in a large enough quantity to be useful. The third soil I tested in was salt saturated sand with some black sand mixed in. I knew from past experience that these conditions give a VLF detector such as these a good wringing out. I spent a least 5 hours with each detector in my garden learning their idosyncrocies and learning to optimize the controls. From this I learned that it is very important that you take your detector, regardless of the brand, and spend 5- 10 hours experimenting in your local soil before you expect to find anything. As expected, there were big differences in the performances in the detectors in this area. In the test garden (moderately mineralized clay), all detectors did fine with the CZ6 going the deepest in discriminate, the XLT going almost as deep but giving more reliable ID, and the CXIII performing the best in the all metal non- motion mode. I think the CXIII might have been able to detect a 9” nickel, but I have none buried that deep.
In salt the story was much different. The CZ6 blew the other machines away by detecting a penny at 9.5” with correct ID and essentially no falsing! This is remarkable performance in this harsh environment. I don’t know if it is just the salt mode or its dual frequency operation, but it sure works. The XLT worked as well as most VLF machines giving 6” on a penny with some falsing. The CXIII was a close third in this environment. I attribute this to preset ground balance, not at all optimized for the beach. The depths on all machines could be slightly increased by operating in all metal mode, but with a corresponding lack of accurate ID. By the way, for comparison purposes I also tried a White’s Surf PI and a Garrett XL500 which I have modified on the same tests. The machines, which are PULSE INDUCTION designs optimized for salt water beaches both did about 9” on a penny.
When the soil was highly negatively mineralized red clay, the tide turned. In this environment the XLT did the best at 6” with a very solid response. The CZ6a was second at 5” and the CXIII managed only 4.5” all with 50% correct ID. The same detectors did 8, 8.5 and 6.5” when ID was allowed to be IRON. The CXIII excelled in the all metal non- motion mode where it did a full 9” on a penny.
There were many more targets that were tested and results recorded. See the summary charts for other depths and conditions, paying careful attention to the notes at the bottom of each page.
As was mentioned in the features section, each of these machines has sophisticated discrimination capabilities. The discriminate as implemented on the CZ6 is quite good. By this I mean that when pulltab rejection is set, pulltabs are really rejected. The CZ6 also has a very strong bias against iron which it rejects very completely. The only problem with this is that when an iron target is in the search field at the same time as a good target, there is a tendency of the unit to reject both targets. The CXIII falls at the other end of the spectrum with very little iron bias. It will often click and pop over rejected iron targets and to some extent other rejected targets as well. It is still obvious the targets are rejects; the machine is just noisier. The positive side of this is that the CXIII is remarkably quick in its audio response. It recovers very quickly from rejected targets and does a better job of finding valuable targets near rejected ones. This is a bit of a problem on shallow targets which will double beep even when not on edge. The XLT takes the middle road. It is smooth in operation and still retains goods sensitivity. It does take a long time to recover from rejected targets. Study of the charts will reveal these differences depending upon which targets you wish to discriminate out.
One of the least understood characteristics, masking is caused when a detector passes over (or near) a rejected target and then over a good one. The detector will frequently not detect (mask) the target under these conditions even though it is well within detection range. I performed both ferrous (nail and bottlecap) as well as non-ferrous (pulltab) masking tests. The CXIII performed consistently better in this test than the CZ and the XLT. Masking can be reduced by slow sweep speeds but the XLT lost significant depth when moved very slowly, due to its 4 filter design. The CZ and CXIII both offer slow motion which helps this characteristic. The XLT recovery speed can be adjusted to be faster. If this is done it helps somewhat, but this can make targets double beep if it is set too fast. The correct setting will still yield the slowest recovery time in the group. It is this characteristic that helps make the XLT a better deep seeker. Other than the strong tendency to mask when near iron (especially bottlecaps), the CZ runs a close second to the CXIII.
I tested the ability of the detectors to reject noise by bringing a Radio Shack high powered eraser near the searchcoil. The detectors were tested with the controls set to the positions that yielded the best results in my garden. This test simulates a 60Hz signal much stronger than you would normally encounter. The instrument the least sensitive to this noise was the CZ6a, followed by the CXIII, and lastly the XLT.
This is largely a subjective impression, but I preferred the quick audio of the CXIII (but only in my garden). In both salt and red clay the audio popped and clicked a little more frequently than the other machines. The XLT was smooth and clean everywhere except at the beach and not that bad even there. It was subjectively slower to respond, but not unpleasantly so. The CZ6a was good everywhere. It was never noisy, and I found it to be the “quietest” detector in the group. In my mind it is the best compromise in this area.
The Garrett and White’s units are covered by a 2 year parts and labor warranty, the Fisher detector has a 5 year parts and 1 year labor warranty. Shipping is the responsibility of the user. To judge out of warranty repair and maintenance costs I phoned each factory. The CZ6 costs $35 to repair plus parts. It has 21 internal adjustments which are likely to require optimization just due to their sheer numbers. This costs $40 and I would suggest it be done once every 2 years. If the circuit boards become damaged such as would be the case if you dropped the unit in salt water the cost would be about $445 plus labor. Shipping charges are added onto these numbers. The XLT costs $35 to repair plus parts. Calibration is $35 and it is only 2 adjustments, so once every 5 years or so is probably fine. If the circuit board needs replacement it is about $120, downright cheap. Shipping charges are extra. The CXIII has a flat $40 repair charge, parts are extra. Calibration is $40 unless done as part of a repair. The main circuit board cost is $250. Return postage is included.
CZ6a comes standard with an 8” spider coil, a set of mid range headphones, and batteries. As options one can add the coils mentioned above, a carrying case, carrying bag, waterproof headset, control housing holster, and a thin clip on speaker( it has NO internal speaker).
XLT comes standard with a 9.5” open center coil, a nicad rechargeable battery with a fast/ overnight charger, and a holder for alkaline batteries. You can add a carrying bag or case, a longer lower rod (for tall people), a holster, hip mount kit, and several types of headphones.
CXIII comes with a 8.5” searchcoil , instructional video and batteries. Optional items are headphones, a carrying case, nicad batteries, an armrest, and the depth multiplier which is the kit that converts the CXIII to a deep seeker.
Each machine has its own strengths and weaknesses. The XLT is a good all around machine that really excels when the ground gets bad. The more mineralized the ground the better it does relative to its competition. It is also the most adjustable. If you really know what you are doing it may be able to give you some edge. If you don’t the chances are that any adjusting you do will only make things worse. Although not much was said about the spectrum display earlier it yields , with some interpretation the best ID of any of these detectors: deeper, more accurate and much less fooled by rusty bottlecaps and other iron trash. The CX III is best characterized by really good non- motion all metal performance. It is the only detector in this group that would easily detect both 8” nickels in my garden. It did a better job on these targets than any other detector I have ever tested. Further despite its sensitivity it is very stable. It is also the only tool in this group that can be converted into a two- box detector for hunting caches. Its performance in discriminate, while solid was no match for the other two detectors. It also excelled in the area of masking doing a better job of ignoring closely spaced trash near treasure. This was particularly true if the trash was ferrous. If you like the beach the CZ6a is the machine for you. It blew away the others at not only depth but also stability. It also has the best discriminate mode of the three machines. It will usually go deeper in discriminate than the other machines particularly if disc is set low. It is also the quietest product in the review. If you are both a dry land and beach hunter and can afford only one model it is the clear choice.
Well there you have it, I hope this review has lived up to your expectation. I again look forward to your comments. You can reach me by writing: Charlie Conger
The report was written in 1996