Sensitivity too high?

[Chris(SoCenWI)]

We discussed this one before and I suggested a couple of experiments. The first was to set the sensitivity at 1, then switch from manual to semi-auto. Semi-auto gets MUCH more distance than manual 1. If you increase the sensitivity in manual you have to get up into the mid teens before you get the same distance. Therefore in Semi-auto mode the explorer will hugely iINCREASE the sensitivity over the set point.

Also, in a room with high EMI(TV, computer, etc) turn on the machine and bump up the sensitivity in semi-auto until it gets unstable, probably can get into the mid twenties. Now switch to manual mode and it should be going crazy. Bring the sens down in manual until it is as stable as it was in Semi-auto; you will probably have to decrease the sensitivity by 12 to 16 clicks.....Therefore in Semi-auto the explorer will hugely DECREASE sensitivity from the setpoint.

The bottom line, and which your statement from Minelab echos is that in Semi-auto the explorer strives for stability. I think that in clean ground far from any sources of EMI that there is very little difference between the two, but close to sources of EMI, or (and this is very important) in trashy sites the explorer sees all the signals from sweeping the coil across the ground as noise and drastically backs off sensitivity. I know I can detect an area that has many signals and manual and switch to semi-auto and they all disappear after a couple of swings.

I think Minelab says Semi is better because they are selling detectors. They still say that patterns are the best way to detect and I don't know any experienced users that consistantly make good finds that use them, most use as little discrimination as possible and sort the signals out in their heads. There are some users that use Semi-Auto sens and DO make good finds, Golddigger cheif amoung them. I suspect they would find even more in manual, or perhaps they detect in different conditions, or perhaps I and others are wrong.

Cody, the only thing I'm not sure of is if in Semi-auto mode something is being tweaked other than the sensitivity setting, that running in manual somehow cripples the automatic ground balancing feature of the machine. I don't think it does.... Almost everywhere I hunt is full of trash, and some people (me included at one time) thought some of the noise you constantly hear is mineralization. But I've detected up north in Minnesota (Close to RickND territory) and there are places where the land has never been built upon or farmed and it was amazing how quiet the machine ran there, even at high sensitivities in manual. Once I got close to a cabin or area of human activity all the chirping and beeping started up again. Same soil mineralization, but people and all the junk they produce is what makes the signals.

I too am geeky enough to want to understand how the machine works. Any experiments you can devise to demonstrate a point I will try duplicate.

Chris

 

[Chris(SoCenWI)]

Hey Dsurcy and Tim,

Minelab's and Andy's books spend quite a bit of time on learing and editing. I thought it was cool as beans when I first got my Explorer and like almost everyone else learned that it was pretty close to useless.

It works fine in areas of no interference from power lines, no trash in the ground, and no ground mineralization. And you won't loose depth in those conditions.

Unfortunately you will never hunt in those conditions. Most of the coins that hit exactly where they should have long since been found. The explorer excels at "sniffy out the iffy" ones that got left behind by other detectors but to do this you have to let it tell you more information, which means opening up the screen and listening to all the noise.

Yes, you can find coins using patterns, but you are probably not even hearing the majority of the ones you swing your coil over.

The whole concept of telling the detector "Here Fido, sniff this and find me another one" is compelling but just doesn't work in the real world. Which means finding the good stuff still requires hard work and lots of experience, which sucks. But all of the tones of the machine combined with the 2 dimensional screen really does give you much more info to work with then any other detector.

Chris

 

[Cody]

I prefer manual and at this point in time lean to thinking it is stability that is being sold with semi-auto. There is no doubt it is nice and stable in semi-auto but think stability can have a lot of different meanings. I don't think stability necessarily translates to more depth and that is the implication. I will test this to be sure but I am fairly certain there are a lot more true false signals in manual than semi-auto if the sensitivity is set up real high in the 28 or so area.

I don't think we can expect an answer to this question as I could not pry one out of Minelab other than go by the Manual. It is a question that is interesting even if it will not be answered. I have been using manual for many months now and it would be difficult for me to be comfortable with semi-auto due to being uncertain of the depth and sensitivity issues.

I thought you might have run across something which was the reason for my post.

 

[dsurcy]

Well a big part of my confusion was thinking that the center bar
was the ONLY response area.. not the "hot" strip.. therefore you
can see that when I received multiple target responses on one
sweep... I interpreted this as multiple targets.. now I see that
multiple responses could be.. and probably is.. multiple responses
to one target.. you can see how confusing that could be..

I'm definitely seeing, due to the responses received here..
that the pinpoint mode is a useful mode and I will be working
on it using the information received here...

More than 10 seconds.. hah!!!... then there ain't no maybe about
it.

Thanks for the information.. It is greatly appreciated..
Man.. I sure hope I'm not the only one out here having such
"newbie" problems with the Explorer... how embarassing that
would be...

 

[dsurcy]

Interesting.. I started with the Factory preset/discrimination..
found that the nulling was so frequent that I was getting hardly
any "detecting" time between nulls.. so switched to ferrous,
IM -16 so I could at least hear what was going on.. (don't know
what is going on yet.. but, at least I can hear it... and know
something is going on)

I had definitely not gotten to the "learn/edit" phase yet.. but had
actually planned to once I had made more progress...

Do you hunt in IM -16 or some other form of it?!

Thanks for the info/advice.

 

[Chris(SoCenWI)]

I hunt in the discrimination mode with only crown caps rejected. So the whole screen is blank except for a small bit in the lower right corner. I use ferrous- you have to if you don't have any discrimination on the left side of the screen.

Lots of people run this way and many others use a little bit of iron mask, say -12 or so. You will not loose many signals if you run this way; it pretty much comes down to if you want to listen to all the low tones from the iron or have your machine constantly nulling. I'm used to all the sounds and nulling now drives me nuts, others that use some discrimination get driven batty by all the sounds. Pretty much a matter of personal preference and what you are used to.

 

[Chris(SoCenWI)]

And it is refreshing to hear someone reasoning out their machine and listening to advice. There have been a few hot heads who were convinced that their machine was defective and got quite angry about the learning curve.

chris

 

[Cody]

You set up pretty much like I do with a little IM-15 but a few pixels clear in the Nails zone. I also use ferrous and am very happy with the performance. I make some other setting changes but they fall in the personal preferences category. I will use the learn function to kill some of the iron with no more than a pixel for each hit in the areas up to about IM-6, -3 area but keep the rest of the right half clear except for the lower right as you do. If it is set up right we can select all icons to be displayed so if it is a nail we know it and using ferrous sounds makes it pretty easy. I do like Audio1 so use it for now with the Audio Gain at 4 to 6 and manual sensitivity at about 25 to 28.

 

[dsurcy]

Happy to report that I had 100% finds during my tests
this afternoon.. by using the advice given here.. by all..
Thanks... I hope all of you experience ones that take
the time trying to helps us here know that we greatly
appreciate it..
Still not exactly comfortable with it all yet.. but... getting
there..
Thanks again...

 

[dsurcy]

What if you found a circular area, free of signals, say the diameter of...
oh... let's say a trashcan lid..
Place a circular pattern of old rusty nails combined with dimes on the
outer perimeter.. say a ratio of two nails to one dime..and spaced..
hmmm.. .say 6 to 8 inches apart along the circle..
then a circular pattern within this circle at the center.. with the diameter
of.. ohhhh.. say.. the diameter of the stock coil.. and make the same
pattern of nail/dime targets along the outer perimeter of this circle.
Then make this same circular target pattern.. midway between the
inner and outer circle.(could even make some target spoke target
patterns outwards from the inner to outer circle?!).

Using the stock coil at at some specified sensitivity setting/auto..
and probably 5-6 inch above the circles do some test sweeps..
then remove auto and compare?!
What is the difference?

Could then remove some of patterns one at a time and compare...
I guess the initial test would be one with no targets in the circle?!
Heck I dunno... I just know I can come up with some goofy ideas!!! :b

I had to try the difference this afternoon of auto and manual (sens 23 or
so?!)... the manual was giving many more target signals.. I don't know if
they were false or not.. I kinda figured there was some sort of increase
in sensitivity.. that caused more signals from further outside the edges
of the coil.. but what the heck do I know?!

I just know is you guys have really help me along with your advice...
and I thank you again.

 

[Bill_in_MA]

Periscopes are not currently being sold - only the used ones in the classified - Hope Jim Ott brings them back.
They really do eliminate the empty hole syndrome because if you can't touch it with the probe - it's not there - so no need to dig.
Worst case if it's very deep you take out a quick scoop of soil - try again if you still can't touch it - it's not there - so no need to dig - so eliminates digging ghosts.
Really the neatest probe.
C'mon Jim bring it back !

 

[dsurcy]

The nulling drives me nutso much more than the sounds..
I would much rather hear what is going on than not...
it also takes much too long to recover... and what is going
on in the meantime (I know, I know..targets could be missed)

I wish there was an option to discriminate without nulling..
such as.. "just ignore nulls.. leave it status quo"... without
a loss for recovery.. such as.. oh.. it's discrimanted.. so..
do nothing... waste no processing time.. drop it in the null
bucket quickly and continue on...

another one of my nutty ideas....

 

[dsurcy]

Yes.. I've been following your posts.. and have been considering the
"pixels clear in the nail zone" technique.. just battling with confidence
in "finding" the target for now.. I even made progress in that this afternoon...
100% finds on target.. I'm a first grader for now.. your information on
coil footprint and pinpointing and such helped me understand and
visualize what was happening...

I will have to try Audio 1.. I just don't want to stray too far for too long for
now.. but then... if I try it and like it.. that wouldn't be straying... would it?
The gain is at 5 (factory preset I believe).. so I should be fine there...

Whoa.. manual sensitivity 25 to 28?! I tried manual 23 this afternoon...
for just a little while.. and it was like there was a zillion targets all over?!
Any advice on how to handle/interpret that?!

Again.. I appreciate all the info and advice.. I have a feeling there are many
more silently following also... I sure hope so.. I would hate to think I'm the
only one with all these dumb questions...

 

[Cody]

Not so nutty in my opinion. I have often wondered why not have a circuit that does nothing to the audio if the target is rejected or simply goes slightly negative if they got to do something. I have stated for years that null discrimination is the worst way to discriminate. Never liked it from the first detector I saw it on which was a Whites way back in the early seventies. They tried to combine TR discrimination in the same box with all metal VLF so a target was found in VLF and then we had to switch to TR to discriminate after the VLF came out. What an absolute mess. The Bounty Hunter Red Baron was the first true VLF discriminator I saw but we had to swing the head so fast it you accidentally hit something it would have knocked the head off the shaft. Man talk about a workout after swinging those for a couple of hours.

You know these kind of discussion is what got us where we are. It is suggestions and complaints from users that drive the change as much as competition between manufactures.

 

[Rick(ND)]

On the test I have done in some trashy area or maybe it was mineralized as it was where a old sidewalk was removed. At 32 auto it ran very smooth,but couldn't pickup a dime on top of the ground. Manual was very unstable at 32, but when I dropped the sensitivity to 22 auto now could get it the dime 3 inches above it. Dropped it to 16 auto seem to get now a good 5 inches so it seems the higher the sensitivity is set in auto the more it will over compensate in bad ground and the more depth you will use. I found if I ran 20 manual I did the best and 16 manual or auto were the same, so I feel if you run auto 16 or lower is the best,but if you want the best depth than manual at 20 or above while stable works the best.
Does anyone else notice this??

 

[Cody]

What you see is correct and why there is some "mystery" about which is best. This is also why Minelab tells us to use semi-auto at the highest setting that has stability. Notice they tells that some experienced users will do better in high mineral soil. The sensitivity setting is not just sensitivity of an amplifier as we normally see. It adjust the Q of the coil and gain of demodulators which is a little complex but is important to the time domain processing. I adjust my own based on what the threshold is doing which is why I like to hear a threshold so use as little null discrimination as I can get by with. I think new user will stop having so many problem with false signals if they use semi-auto. That is a problem with using manual in with no compensation for the changing Q we will get false signals but the experienced user recognizes them as such. There are advantages to both semi-auto and manual so the user has to see which is best for different ground and the soil matrix. What users are not sure about is how quick and how far up or down semi-auto adjusts the sensitivity. I never had any problem with great depth with semi-auto or manual. I have located ultra deep targets in manual that I could not detect in semi-auto but did not attempt to find out why.