DETECTING INVISIBLE NUGGETS

[Anonymous]

Recently, on the Nuggetshooter Forum, there has been some discussion about gold nuggets that are invisible to PI type detectors, but give a good response on VLF types. I got involved in this discussion and one of the forum participants, John Blennert, was kind enough to send both myself and Reg Sniff, a couple of nuggets each to test. These are not tiny specks of gold, but of the size that you would expect a reasonable detection range from.
Examination of the nuggets shows a very platey/spongey structure, which is not very good for supporting eddy currents. Due to the thinness of the contact points between the plates, and the very small size of the plates themselves, the internal resistance of the total nugget must be quite high. This would result in the eddy currents having a very fast decay, due to the resistance losses. In the first picture, John's two nuggets are on the left, and my solid 0.9gm nugget on the right.
To determine how fast the eddy current decay is, I dusted off my ultra fast PI circuit board. This was a prototype for a detector that would find broken off needle tips in textiles and garments. With a suitable coil, the sample delay can be set as short as 1uS after the TX switch off. The normal coil on this unit is wound on a ferrite pot core half of nearly 3in in diameter. Placing the nugget on the centre pole of this core generates a strong enough signal to observe the decay curve all the way from its maximum value, right down into the noise level. Both of John B

 

[Anonymous]

I would like to add a little more information about the

 

[Anonymous]

I followed the threads last night and this morning, all together a number of hours. I have to "Study" words to get an understanding of what is meant by the words and the message of it all. I am not a tect. person but I have to state that I learned much, got inspired and really have spend a number of hours deep in the posts of all on that thread and especially the post of you and Mr. Foster, I sure want to understand more.
Really good and interesting reading and studing...I have much to learn and sure am glad there are fellows like you guys out there... WE can only benefit from your dedication. Frank

 

[Anonymous]

Hi Frank,
I apologize about the tech stuff. When one works with terms like usec, it becomes second nature to use it when writing and I have a tendency to forget other people might not fully realize some of the terms. For example, the letter u usually stands for micro, so a usec means microsecond or 1 millionth of a second. Even then it is hard to conceive what one is talking about.
To try to put just this single term, usec, in perspective, a typical blink of an eye is considered to take 300 milliseconds or so (a millisecond is 1 thousanth of a second). So, 300,000 usec's pass in a single blink. Even with working with equipment that can measure such short times, it is hard to conceive something like it. It least it is for me and I work with equipment used to measure such short times.
Fortunately, I have enough of a technical background that a schematic of a PI makes sense. My first impression of them was, they aren't so complicated. Boy what a rude awakening.
They may seem simple but getting one to work, especially at the delays Eric does is a real challenge to say the least. I think several people who have tried to build a PI would agree.
In a nutshell, to get a PI to work at extremely short delays, one really needs to know a whole lot about each and every component used. Parts such as the preamp and associated circuitry become very critical, as does the filtering of all noise. It doesn't stop there either.
The reason noise is so critical is the detector is trying to amplify microvolt type signals. Yes, that signals in the millionth of a volt range.
In the future if you have a question about the technical stuff including the wording, I will try my best to explain if I can.
Reg

 

[Anonymous]

Hi Reg,
I am glad to know it wasn't only me <img src="/metal/html/smile.gif" border=0 width=15 height=15 alt=""> I felt the same as you when I looked at my 1st PI schematic. I thought geez these are pretty simple and straightforward. But as with most things in life the more you learn about something the more you find out there is to learn about it <img src="/metal/html/smile.gif" border=0 width=15 height=15 alt="">
Happy Holidays to everyone!!!!!!!!!!!
Beachcomber

 

[Anonymous]

Hi all
Question. What is the voltage range for a "weak", "medium" or "good/strong" target signal (voltage level in mV) at the receiver stage befor amplified.
Chris

 

[Anonymous]

Eric and Reg, do these undetectables have higher than normal readings on ohm-meter or diode test? They might be like a dry joint.

 

[Anonymous]

Hi Chris,
I have never been able to measure the voltage at the input to the preamp and would have to do the calculations. I will when I open my PI again shortly to try another coil design.
Instead, I always look at the signal at the output of the preamp where it can be seen and measured with greater accuracy. That is where I would recommend anyone to take their measurements, rather than trying to measure a signal at the coil or input to the preamp.
Now, with that in mind, one can look at Eric's pics above and get somewhat of an idea on what is involved in trying to answer your question.
If you look at the two scope pics, but mainly the second one, you will notice the signal from more solid nugget almost decays by 20 usec but is quite measureable at 5 or 10 usec.
I am not sure of the Vertical scope settings on Eric's pic, but it gives you a general idea of the difficulty of trying to answer your question with a specific answer.
First, the signal is decaying so a point or delay time should be referenced for signal amplitudes measurements and then it is difficult because a signal that appears to have decayed will still produce a reasonable audio signal.
Hopefully, Eric will provide the verticle scale settings for the pics involved. If I had to guess, I would say the scope was on the 50mv to 100 mv scale (where mv means millivolt), which would mean the more solid nugget was generating something like a 25 to 50 mv reading at 10 usec.
Now, in the most general terms, a very weak signal generally will not be able to be measured easily even at the output of the preamp. The signal will be so weak it will be lost in the hash. At least it is where I work on my PI. I don't have good control over the noise level.
What I would call a medium type signal on my detector is a target that will only provide a several millivolt signal, maybe 10 to 20 millivolt or so.
A large signal will be anything greater than that.
Working backwards, this would mean a medium signal might be in the 20 to 40 microvolt range.
Now, most of this info is done from memory which, in my case, is fading fast, so I could be off by a fair amount.
BTW, these two pics posted by Eric are great examples of just why the sample time is extremely important when trying to detect small gold. The first pic is expanded horizontally by a factor of two when compared to the second, so if the signal in the first pic were to be superimposed on to the second pic, the entire signal would fit in the left half of the screen of the second pic. This would basically mean that the more solid nugget would produce a signal at 10 usec that could be detected, but the first nugget signal would have already decayed to 0V and produce no signal.
Reg

 

[Anonymous]

Hi Allan,
I haven't tried an ohmmeter test because the nuggets don't look too solid and the fine crystaline structure looks too delicated. I am afraid of destroying part of them. They are really hard to describe as to what I see.
When I look closely at the nuggets with a magnifying glass I can see rock in several places and it almost seems like the gold has just been thinly deposited on this rock, rather than laced through it.
I guess the best way to explain what I see is to say that the extremely small thin crystalline gold appears to have "grown" on a host rock. So, I am not sure just what type of probing could be done without negatively impacting the nuggets.
Since they do not belong to me, I am hesitant to do anything to them to hurt them.
Since I can see rock in several places it is safe to assume that there would be poor connections across the entire surface.
Reg

 

[Anonymous]

Hi Reg and Chris,
The vertical scale on my scope pictures is 0.5V/div. The 3.5V total span is the saturation level of the amplifier, hence the flat top on the top left. The amplifier gain is 245x. This is lower than usual so as to get the bandwidth. These traces are not representative of the signal you would get from a nugget at a distance from say an 8in coil. You cannot even see the decay curve under normal conditions, so the nugget was sat right on a small coil giving a very concentrated field to get the amplitude necessary to show the whole curve. If you turn up the scope sensitivity to 50mV/div and look at the 5uS level, the coil signal works out to about 50uV. This is still large compared to what you would get in practice. You can see what you are missing by sampling as late at 5uS in this case! By the way the figures above are for the 1.4gm nugget.
Eric.

 

[Anonymous]

Hi Eric,
Thanks for the update Eric. Boy, do I feel dumb now. If I had any common sense or been thinking clearly I would have looked closely at the amp saturation voltage and that would have told me the scale used. I was just concentrating on the blip markers indicating the sample period and what they looked like on mine by trying to rely on my memory.
This discussion has pointed out the difficulty of trying to relate definitive measurements to something like what would constitute a weak or a strong signal. I have a tendency to measure such signals as weak or strong by the audio signal level they generate and not the actual measurements throughout the various stages.
However, what might be a weak signal on one detector may be a strong signal another using my approach just due to the different amplification factors involved in subsequent stages.
Also, I have never really tried to calculate backwards just what the signal level, in microvolts might be at the coil for signals from different targets or the same target at different distances.
In any case, it is obvious the signal levels are going to be in the microvolt range for small targets such as small gold nuggets. Whether such signals create a weak or loud audio will be the result of the amplification of the following stages and the time of sampling.
I can't sample as fast as you have displayed so I can't really duplicate your results. However, my sample specimens will generate what I would call a very loud or strong audio signal when the nuggets are sitting on top of my coil, yet the signal level is in the millivolt range at the output of the preamp. At 2" the signal is still noticeable audibly, but not really noticeable on the output of the preamp.
Looking at large or strong from another extreme, placing a large piece of metal on the coil can generate a couple of volt signal at the output of my preamp when measured at short delays.
Reg

 

[Anonymous]

Hi Reg and Chris,
With my ultra sensitive PI, 30mV at the integrator output sends the meter on the dc amp full scale. To get a signal I can see on the front end amplifier, I increased the signal from the target to give 1V at the integrator output. This gives me a deflection of 20mV at the amplifier output. Divide by the gain and you get 78uV from the coil. The signal is decaying fast during the sample period, so this is approximate. For the full scale meter deflection, where there is only 30mV at the integrator, the coil voltage works out to 2.3uV. For 10% deflection, which is a repeatable, not to be missed signal, the target voltage from the coil is 0.23uV.
Hope I have done the sums right, but you can see the order of signals we are dealing with.
Eric.

 

[Anonymous]

Hi Reg & Eric
Thanks for your explanation & time. My Question was not really precise. I know every detector and coil is different to make a common general statement. A medium signal for one detector could be for another detector a strong signal depending on the receiver electronic and the amplifier gain.
For a more general statement, regardless of the detector model or the electronic, lets say we want to measure the target signal voltage level for Gold at 10us at the receiver input.
To make the measurements easier and relay able we can still do the measurement at the receiver output and calculate the voltage level (at the input) backwards with the known gain factor of your receiver electronic.
What would be than the voltage level range for a week, medium and strong target signal?
My guess....the voltage level range for Gold at 10us (receiver input) for week target signal is maybe between 2 to 5mV, for a medium signal 6 to 15mV and for a strong signal 16mV and above?
Chris

 

[Anonymous]

Hi Chris,
You might want to go back and read Eric's post just above yours again. Basically, he did the calculations and they come out in the uV (microvolt range), not mv.
It appears a weak signal could be less than 1 uV change at the coil, maybe a medium signal in the few uV range and maybe a strong signal will have less than 100 uV at the coil.
Again, this is all relative depending upon the gain of the following stages. In any case, the signals are extremely small.
Reg

 

[Anonymous]

Hi Reg,
Oh, I missed Erics post on that.
Chris

 

[Anonymous]

Hi Reg, thank you and others too for those lovely wiews of nuggets. We also have kind of invisible gold, tribal celtic coins. Everyone assured us that is impossible to detect such a stuff, until we used our selfmade detectors. PI detector get them, in virgin ground at three inches. In trashy ground, with iron fragments, we must take a special VLF, two inches. There are no limits for detektors, except of unpickable targets. Sid and Wen