Help someone

[Reg]

Re: Help someone (sorry its long)

Hi Wyndham,

Your idea of finding a system when iron no longer dominates is interesting. Now, on a VLF, the lower the frequency, the more iron seems to dominate. The first VLF's operated way down in the 1 to 2K hz if I remember and they loved iron. So, raising the frequency would or should help some. Higher frequencies such as something like that of the Gold Bug 2 could make it much easier to separate iron from non-iron when compared to a typical coin machine operating at 10 to 15khz. I know from testing that the GB 2 does a much better job of allowing a gold target to be seen "through" a piece of magnetite or iron target. At least, it did in my testing.

The down side of using this higher frequency of something like 70Khz or so is target separation as used in visual TID machines would be very difficult. That becomes a non issue if only iron/non iron ID were the objective.

I have thought of "off resonance" circuits as a possibility since they do have a very good level of discrimination simply due to the technique involved. Now, one thing I have considered doing is building a prototype of AH Pro's ground canceling "off resonance" circuit to see how well it worked.

I will have to give a lot more thought on how such a technique could be applied to PI's.

Reg

[Ray Huffman]

Re: Help someone (sorry its long)

hello all
I use a whites GM4b for nugget hunting. I have found that for large pieces of iron the id works ok but once you get down to tiny pieces then you have to dig it to find out. When I work on bedrock it will find gold down to a 1/10 gr, so those tiny pieces of iron fool me now and then. I think for me I mainly listen for the sounds the target makes and on the whites the sound gold makes is different and almost unique compared to what iron sounds like.
Ray

[Carl-NC]

Re: Help someone

Just now catching up...

I agree that no iron disc will be foolproof, but even 50% effective is better than 0%. I suspect that even a simple burst-mode phase detector that runs between pulses will be decent, especially if some digital ground filtering can be applied.

Reg said: "Now, one thing I have considered doing is building a prototype of AH Pro's ground canceling "off resonance" circuit to see how well it worked."

Reg, Allen Hametta is still in business, and might be willing to help out. Last I spoke with him, he was messing with magnetometers.

--Carl

[Reg]

Re: Help someone

Hi Carl,

Thanks for the info. I wasn't aware that Hametta was still in the basic business. The fundamental design of his ground cancelling off resonance circuit is interesting. I suspect the depth would be limited but I would think the discrimination would be very good.

Before I forget, did you ever see the inside of one of Hametta's coils? He used to build his shielding by simply etching the bottom side of a pc board with very short lines close together. It worked quite well. The top side was etched to build the coil windings.

Reg

[Carl-NC]

Re: Help someone

>Before I forget, did you ever see the inside of one of Hametta's coils? He used to build his shielding by simply etching the bottom side of a pc board with very short lines close together. It worked quite well. The top side was etched to build the coil windings.

No, in fact I don't think I was aware that he used etched coils. In chip design, etched coil shields are often made with radial metal strips. Makes sense to do the same on a PCB.

--Carl

[Brian King]

Re: Help someone (sorry its long)

Hi Reg,
I am currently working on some different coil configurations and front end electronics for PI and the discussion on disc is very interesting.
My experiments show that the coil/transmit pulse/amplifier stages in themselves can definitely be improved but there is always a trade-off in price,weight,ease of use versus actual benifits. Down here in Australia gold bearing country usually abounds with 'hot rocks' and in the UK the main areas of interest abound with Iron artifacts. Point here being that a machine to handle both areas of interest would be quite expensive based on current PI designs which are excellent but not perfect.
On a slight tangent if I may,
It is interesting to note that UXO technology trends heavily to multiple receive/single transmit coil arrangements and that the transmit pulses are bipolar. Some of these mine detectors are completely mobile while some have a stationary transmit loop. Both allow a 'view' to the target type which in itself, is discrimination. Of course there is usually GPR involved for detailed mapping of UXOs.
However, even without GPR an array of receive coils can give useful information such as shape/posible orientation/depth.
The coil arrays are kept at a constant distance above the ground and there is some heavy duty computations going on with the signals but it works.
Thinking down these lines (and not an orphan here I suspect!) an experiment was set up to test the idea on a small hand held scale.
Results were encouraging enough to keep me focused on this design method. Once I have some meaningful results I will post them.
The main tasks for my current project are;
1) Low noise stable front end electronics which includes flash type ADC buffered directly into memory using DMA control.
2) Coil array physics, size, shape, orientaion, weight.(have experimented with multiple tapped and single coils numbering from 3 to
3)Proccessor board interface providing both graphical and audible features.
I expect this to take another 12-18 months to complete as I am still currently employed in electronics engineering and time can slip by quickly with my work load.
I have had some exchange of emails from Willy B on this site and I would really like to finish off my front end design first as Willy sure sounds like he is doing some excellent work on target discrimination too.
Sorry if I rambled on here a bit but it is a great forum for ideas to bounce back and forth and could'nt help myself, just had to enter this discussion.
Cheers for now

Brian K

[Brian King]

Re: Help someone (sorry its long)

Hi Reg,
One thing I meant to say but missed is that I agreed with your view regarding how difficult it actually is to have a simple fix to suit all conditions,therein lays the challenge for future PI designs!

Regards
Brian K