E-Trac/SE/DFX Data: Spreadsheets, Charts and Emulator Files

[The Beep Goes On]

Wanting to understand the E-Trac more fully I gathered some data by comparing it to machines I was already familiar with, namely the SE and DFX.

I bench tested the same targets on all three machines, created spreadsheets and charts and then created E-Trac and SE emulator files from the data. Three sets of data were created: US Coins (29 targets), Gold Jewelry (43 targets) and Silver Jewelry (56 targets). For the US Coins I tested the coins flat and on edge.

If you are familiar with any of the machines you should be able to get up to speed on the others by reviewing the files. The charts really help with this. The three E-Trac emulator files are the same as the three SE emulator files (except for the FE-CO numbers, of course) which allow you to have both emulators running at the same time with the same targets. It is an interesting way to do a side-by-side comparison.

You can download these files from http://www.thebeepgoeson.com. Click on the link to the DOWNLOADS page. There you will find two pages: "Target ID Spreadsheets" and "E-Trac and SE Emulator Files" which contain the respective files.

There are three tables, ten charts and six emulator files. As an example, two of the charts from the Silver Jewelry data are shown below. The charts graphically depict the difference in the axis orientation between the E-Trac and the SE as well as where the targets fall.


HH!
TBGO

 

Amazon Forum Fav 👍

A Handbook of Civil War Bullets and Cartridges - Grab it through Amazon!

[The Beep Goes On]

I have no idea what any of that means..It just seems like a lot of work to find out if one detector goes one inch deeper than another.!!!!

None of the tests had anything to do with depth Jim.

TBGO

 

[go-rebels]

Nice charts!

I'd love to see a subset of the coin chart with "masked" items, like taking a reading of a silver dime WITH a small steel brad (nail), then with a bottle cap, with a pull tab 'tail' and with a current pull tab handle (with hole in it).

Also, I'd like to see where the common trash stuff above lies on you Etrac and Explorer charts both flat and on edge.

Keep up the good work!

 

[The Beep Goes On]

Nice charts!

I'd love to see a subset of the coin chart with "masked" items, like taking a reading of a silver dime WITH a small steel brad (nail), then with a bottle cap, with a pull tab 'tail' and with a current pull tab handle (with hole in it).

Also, I'd like to see where the common trash stuff above lies on you Etrac and Explorer charts both flat and on edge.

Keep up the good work!

Thanks, go-rebels. Probably the best you could do is a dataset on trash items. The emulators only take one set of values, but it would be nice if they could take two per target so you could make the cursor jump around like a masked target. I guess you could expand the tables themselves to show multiple readings per target. The problem is that there is an infinite set of relative positions of the target/target pair making the data somewhat questionable (and hard to present in an understandable way).

Jim_K - This is just target responses placed on a graph that is configured the same way as the E-Trac and SE screens. The spreadsheets contain the data from which the charts and emulator files are created. I collected data on three machines so that, theoretically, if you knew one machine well, you could see how the other machines responded and make the learning curve a little less steep.

HH!
TBGO

 

[Tom_in_CA]

I don't think a chart like that, to study where various coins and specific pieces of jewelry fall, would do you any good. For starters, on the coins for instance, the TID's would be immediately skewed (even if ever so slightly) the moment you add several years in the ground, esp. in funny soils (minerals), and then get further skewed when you add decades, and put them several inches deep. Ie.: the only time the cursor would land on the exact spot you computerized, is when you're holding it in your hand, in an air test.

As for gold jewelry, you can scan infinated pieces and find infinate places on the TID they will land (on both axis). But for every one of those cursor spots, you will also find out that there's a heck of a lot of foil wads, can slaw, bent tabs, etc.... that can land all over the spectrum of those TID's too.

About all you can do is to know the approx. TIDs (highs, lows, mediums), combined with your knowledge of a site, the "size" of the object (ie.: the audio length of the beep), the depth bar, etc.... to give you educated guesses of what to chase.

For example: If I'm working a junky park, and get a surface tab or foil signal, odds are: It's going to be a tab or foil - doh! But if I'm working an 1800s foundation site out in the wilderness, where no human has set foot in 100 yrs, a tab signal will probably be a pistol ball, button, etc... If I'm working the wet salt beach where erosion has washed away all the light stuff, leaving only the heavier items, then a pulltab signal can be a gold ring, a mid-sized key, a small fishing sinker, etc... In each case, the cursor can read exactly the same. But based on where I'm hunting, I get excited or not

 

[sniffer]

I believe, the charts represent the display on the SE, and the marks on the chart, are different targets and where it was found on the display of the SE

 

[The Beep Goes On]

I don't think a chart like that, to study where various coins and specific pieces of jewelry fall, would do you any good. For starters, on the coins for instance, the TID's would be immediately skewed (even if ever so slightly) the moment you add several years in the ground, esp. in funny soils (minerals), and then get further skewed when you add decades, and put them several inches deep. Ie.: the only time the cursor would land on the exact spot you computerized, is when you're holding it in your hand, in an air test.

As for gold jewelry, you can scan infinated pieces and find infinate places on the TID they will land (on both axis). But for every one of those cursor spots, you will also find out that there's a heck of a lot of foil wads, can slaw, bent tabs, etc.... that can land all over the spectrum of those TID's too.

About all you can do is to know the approx. TIDs (highs, lows, mediums), combined with your knowledge of a site, the "size" of the object (ie.: the audio length of the beep), the depth bar, etc.... to give you educated guesses of what to chase.

For example: If I'm working a junky park, and get a surface tab or foil signal, odds are: It's going to be a tab or foil - doh! But if I'm working an 1800s foundation site out in the wilderness, where no human has set foot in 100 yrs, a tab signal will probably be a pistol ball, button, etc... If I'm working the wet salt beach where erosion has washed away all the light stuff, leaving only the heavier items, then a pulltab signal can be a gold ring, a mid-sized key, a small fishing sinker, etc... In each case, the cursor can read exactly the same. But based on where I'm hunting, I get excited or not

I definitely agree. I think you miss the point. This excercise was done to map one machines response to the other machines' response. If you know the SE well, for instance, and know that machines FE-CO numbers well, the data will tell you what the corresponding E-Trac and DFX IDs are. It is kind of like the Rosetta Stone...if you know one language you can decipher the other two. It was not my intention to imply that the targets presented always respond as shown. It was also done to show the difference between the E-Trac and SE screen configurations and where an identical target will fall on each machine.

However, having used all three machines I find that, unless the target is deep, the numbers are close to what I experience in the field. Gold is a crap-shoot and always will be, but coins and silver usually hit around the numbers presented (at least in the areas I have hunted). The response might be buried in iron tones, etc., but if I get that one peep out of the machine that I recognize as a good target I will dig it. This is not a blanket statement...I know some soils can drastically alter a target response.

On my download page I say...

***
Bench testing metal detectors using air tests is good for some things, but not for others:

• Air tests can provide a good idea of a shallow target's response - deep targets, targets in close proximity to other metallic items, targets that have been in the ground for a long time and extreme soil conditions can produce very different responses
• Air tests are not a reliable indicator of a detector's depth capabilities
• Air tests can provide a good method for comparing different detectors regarding their response to the same target
• Air tests indicate that, for some targets, the orientation of the target can be as important as its composition
• Air tests teach you that, in the end, it is probably best to dig everything within whatever you consider to be reasonable limits

NOTE: It is entirely possible that you may not get the same results as presented here. The environment, manufacturing variability, the coil used, testing methodology and many other factors can cause different test results.
***

HH!
TBGO

 

[go-rebels]

Thanks, go-rebels. Probably the best you could do is a dataset on trash items. The emulators only take one set of values, but it would be nice if they could take two per target so you could make the cursor jump around like a masked target. I guess you could expand the tables themselves to show multiple readings per target. The problem is that there is an infinite set of relative positions of the target/target pair making the data somewhat questionable (and hard to present in an understandable way).


HH!
TBGO

I would be interesting to take a small nail and tape it to a dime, then pass the tightly bound pair under a coil. Switch the nail with an aluminum tab, etc... I wonder if the signal from the Explorer or Etrac would be consistent, or would it significantly vary based on the orientation and distance from the coil. Just curious...

 

[The Beep Goes On]

Some have suggested that I do a trash target file...I admit that seeing how the machines handle target masking/contamination would be interesting.

HH!
TBGO