Coil shielding problems

[jackdetect]

Re: Differential coils!

Hi Eric,
But I think, they are using shorter pulses, especially for mine and UXO detection. Of course, they also use “heavy artillery”- DSP, data storing and PC analysis (imaging). With such process, they can easily avoid any noise.

I was thinking a lot about one question, and I think, you are best person to answer.
Why some designer of large loop Time domain locators using a bi pulse? What they get from it? Is any benefit if we try to design classic Pi detectors with bi pulse?
Thanks,
Jackdetect.

[Dave Emery]

Re: Differential coils!

Bipolar Pulse detectors are used in demining operations. This is because the average field from a bipolar pulse detector cancels to zero. There are some types of land mine which use a magnetic trigger. These mines will explode as a truck goes by them. A regular metal detector which does not use bipolar pulsing will cause these mines to explode and kill the operator!

[Eric Foster]

Bipolar Pulses.

Hi again,

One reason that many geophysical TD units use bipolar pulses, is to avoid induced polarisation signals. These are caused by electrolytic effects in the ground which are driven by the induced eddy currents. Because the excited volume of ground is large, it appears as a capacitor in series with a large resistor. A unipolar pulse train will charge this up over many pulses and there will also be a discharge signal with a long time constant. This will appear superimposed on the conductive response they are looking for, from ore bodies. With a bipolar pulse, the induced polarisation never builds up, as the effect caused by one pulse is cancelled by the next. There has never been any evidence that induced polarisation is noticeable on relatively low power, small loop treasure hunting PI's, although the bipolar pulse does have the benefit that Dave Emery mentioned in mine detection.

Eric.

[Reg]

Re: Open loop is not fully faraday shield!

Hi Jackdetect,

You are correct, I have mentioned the figure 8 coil as a means to reduce noise. I have built a couple of them as an experiment just to see how well they worked. They do work but there is a loss of sensitivity when using this type of coil and comparing it to a similar size mono coil or DD.

One other design is to use a large coil for the transmit and a figure 8 receive. This provides the benefits of noise reduction and ground signal reduction. I think Coiltek calls them Salt coils.

Reg

[Eric Foster]

Re: Differential coils - additional info

I made some rectangular figure of 8 coils for an industrial application on a double sided pc. These were considerably better at cancelling noise at higher frequencies, as the two halves were obviously very much better matched. The same would no doubt apply to two coils in a gradiometer configuration.

Eric.

[jackdetect]

Re: Serious dilemma about Pi! Eric, please help !

Hi Eric,
Maybe is the right moment, to explain my problems with metal detector, which probably the same for lot of my colleagues. After 20 years of exploitation pretty good commercial metal detectors, overhead strata of ground (about 50cm) become empty of artifacts.
I was many years work with lot of my VLF design, relatively large coils etc. But they simple can not go deeply in ground.
Now, I must confront with Pi detectors. But, for my, and many other peoples, I think, coils about 50cm can not satisfied. Meteorites hunt is the same theme.
BUT, Pi detectors with large coils are very poor covered in many projects of Pi detectors.
They somehow mystify. We are exchange lot of best VLF design schematics, but Lorenz, or Puls s II or III and similar, they are unattainable.
I think, they not so well done, and designers have a lot of space for better designs. But we have not a start position.
Latest argument on forum, is move more importance points.
Help us to find right way in design Pi with large coils!
We need good sensitivity for small metal object, and I suppose we need about 400-500Hz. On such frequency, capacitance according to ground probably will be a problem.
Induced polarization is also can be a problem if we using high transmit power.
Large coils usually pick more noise then small.
And maybe for my self the most problem is a how to avoid minerals in ground, and also, how to get good sensitivity for bronze and cooper.
Of course, I not ask schematic of such detector, but is there any possibility to give us some ideas. What we need? Two channels, shielding, differential coils, basket coils, what frequency, bi pulse, large Tx, or something different. I simply can not deciding, which concept is the best for deep search Pi with large coils.
Thanks,
Jackdetect.

[Eric Foster]

Emissions

Hi FJ,

I can't speak for other manufacturers, but it is not too difficult to get the emissions within spec. Radiated emissions testing starts at 30MHz and extends to 1000MHz. Large loop detector TX circuits tend to be overdamped and the pulse switch off rate limited by this and other factors. TX harmonics should not extend beyond a few MHz. Often the problem is not the transmitter itself, but other devices in the detector which have faster switching speeds. A 555 clock generator can produce harmonics to over 30MHz, as can fast logic circuits. These harmonics can be coupled via internal wiring, inappropriately laid pcb tracks etc, so that they are superimposed on the transmitter output. I have had a design fail because of this, and the cure was to shorten the output track from the clock to the rest of the pulse generator circuitry. Also, reduce rise and fall times of any pulses to no more than is necessary to achieve the desired result, i.e. use the slowest logic that does the job. It does appear that a large proportion of emissions problems with electronic equipment, is related to pcb layout.

Eric.

[Stefano]

Re: A Question

Hi Eric,
Yes, actually the foil was grounded via coax braid at that time and not as I learnt later separately.
I'm still using at the moment rg58 coax for tests beacause it's the lowest capacity cable I've found till now, anyway as soon as I'll find an acceptable shielding material I'll look for a 2 wire shielded cable connecting the coil shield separately and directly to pcb gnd trough the cable screen and leaving the 2 inner wire for the coil as you and Reg suggested in past.
I'm doing tests with poliester film taken from capacitor at the moment which is extremely thin and seems to work much better, and maybe some Easter chocolate egg silvered paper which I found to be conductive !!!
About italian beaches condition, I don't know exactly, but i don't think they're very mineralized and that the problem was due to mineralization. I had in fact almost the same false signal lowering the coil on the grass in a courtyard. Problem is our beaches themselves , or at least the Adriatic side ones, full of trashes and junk...
Thanks for your reply and your kind advises.

Stefano

[Reg]

Re: A Question

Hi Stefano,

I am not sure just what you are saying when you say the faraday shielding should be connected separately.

If the faraday shielding is the type where the shield is wrapped around the windings then I have found that the faraday shielding can be electrically connected to the coax shield in the coil housing and doesn't need to be taken back separately to the pc board. So, the shield does not need a separate wire back to the pc board.

In other words, RG 58 will work just fine and no extra or special wire is needed for the faraday shield.

I use a little different coax than the RG 58U simply because it has a lower capacitance, and is not readily detected if it moves. I use an S Video cable made by Mogami. This cable has two small diameter coax's bonded together. What makes the coax a little different is the coax shielding is not a braid, but is a layer of very fine wire wrapped around the inner insulation. This type of coax shielding, which is called "served", is not detected as easily as a typical braid shielding.

When I build a mono coil, I simply separate the two coax's of the S Video cable and simply use one of the very small coax's.

The confusion may have come because of how I explained the building of a DD coil. On a DD coil, I keep the transmit winding and shield fully isolated and insulated from the receive coil and its shield.

On a DD coil, the transmit coil faraday shield is electrically connected to the transmit coax shield while the receive coil shield electrically connects to the receive coax shield. The two coils and/or their faraday shields do not physically touch or connect electrically in the coil.

The two coax shields can tie together in the coil plug or use separate pins depending upon the plug used, but not in the coil housing.

I have found that when the transmit and receive coax electrically connect in the coil, there is a significant increase in the decay time because the two coax's are in parallel, thus increasing the capacitance. There are other combinations that will work fine, but to explain them would further complicate things. One can and probably should experiment with different connection setups to see just how much of an effect it has.

Reg

[jackdetect]

Re: Questions about gold nugget hunt!

Hi Reg,
I was reed you article about gold nugget hunt, and I think it is excellent. I learn a lot.
What you think about Fisher Gold Bug, about low frequency for nugget, I think, and about his work without disc?
What about practice experience on the field?
Thanks,
Jackdetect.

[Reg]

Re: Questions about gold nugget hunt!

Hi Jackdetect,

I am not sure which article you read since I have written a lot of them about gold hunting. Most of the articles came out in Lost Treasure magazine a few years ago.

As for the Gold Bug, it still is a great metal detector for gold as are some of the other older VLF's.

The higher the frequency of the VLF type machine, the better it is at detecting the real small gold. The original Gold Bug operates at about 20 khz and works very well at detecting gold of all sizes. There is no discrimination mode on this detector but some people got quite well at determining gold from iron just by the sound of the target.

The Gold Bug 2 which operates at something like 70Khz works excellent also. The discrimination on this detector works very well, but no discrimination is perfect when it comes to gold nugget hunting.

I have seen and had the best of detectors fail to properly distinguish junk from small gold. There are too many things that can cause errors.

So, when hunting in gold producing areas, it is best to simply dig everything when possible. Normally, I will not try to rely on any discrimination except when there is just too much junk or I am extremely tired.

As for the frequencies and gold hunting, most VLF type detectors that operate between 10 khz and 100 khz work quite well. The higher frequency machines have a slight advantage in finding real small gold, while the lower frequency machines have a slight depth advantage on larger gold.

PI's do better at detecting larger gold deeper than a VLF but the depth difference is not as much as some people claim. What makes a PI work better for gold hunting is PI's do not react to a lot of the "hotrocks" like the VLF's do. A PI will detect a gold nugget hidden under a hotrock while the VLF, may ignore the nugget.

Reg

[jackdetect]

Re: Questions about gold nugget hunt!

Hi Reg,
It was detailed answer, thanks.
I was reed you article: http://www.gpoc.com/induart22.htm , and one about Pi detectors.
I will be glad to see all you articles, and it will be good, if you can giving to us all links.
Any publicist or designer in world of metal detectors, can become really good, only if its knowledge attest on lot of terrain work.
Best regards,
Jackdetect.

[Reg]

Re: Questions about gold nugget hunt!

Hi Jackdetect,

I am sorry but I have written so many articles over the last 15 to 17 years it would be impossible to add a link to all of them. Many of them are not on the net and even I don't have a copy of most of them. Most of these articles were printed on Lost Treasure magazine. They may have have some of them on their website.

I belong to the GPOC and wrote some articles for their monthly news letter for them well before their website was popular. The articles were later placed on their website. Some of the information written for GPOC really need to be updated when I get a chance because I have learned a lot more about PI's since then.

I wrote one more recent article that is posted on www.nuggetshooter.com which is about PI's and is a little more up to date. It is a general article about PI's written because Bill, the website owner wanted some PI info for his site.

When I write an article I do so to share the knowledge that I have acquired up to at that time. Over the years, it is been difficult for me to find information and I have had to struggle to get the specifics. I just try to make it easier for others.

I don't claim to be an expert, and there are a lot of people with much more knowledge about metal detectors than I have. I learn a lot by experimenting with different techniques on working models. This helps me get a better understanding of just how things work.

I have found that one shouldn't be afraid to try a wide variety of things, even those that may go against what you have learned or been told. Many will fail, but some will succeed and provide unique and valuable information. I don't advocate doing anything unsafe, but do think people should not be afraid to try different things. Nothing we use today was ever developed as perfect at the beginning.

Ideas such as the ones posted by Stefano and Sid mentioning how they used the windings of a capacitor for shielding are great!! They show unique thinking and the willingness to try something really different.

Almost every person who has contributed on this website has come up with at least one really good idea that can help build a perfect detector. The key is to be able to put them all together.

Reg

[jackdetect]

Re: Questions about gold nugget hunt!

Hi Reg,
I agree with everything. I believe, that we all known really capability of present metal detectors. They are so far from perfect. And we must work hard to change it.
On other side, most beginners are getting a quite different picture of metal detectors, creating by seller prospects. I was seen such idealistic views on forums too. I think, we also having a task, to help beginners to see things on right way. Independent and covered with experience point of view, is very deficit in world of metal detectors, I think.
We can not speak only about technical aspects inside electronic box, we also need to speak about real effects on the field. But I was seen a little of such themes on the forums.
Best regards,
Jackdetect.

[Stefano]

Re: A Question

Hi Reg,

helpful clarification!
I was in fact doing my tests with the coax and monoloop though thinking of not being completely right.
I must have misunderstood or mixed something.
DD coil will be the next step.
Thank you again for sharing your experience with us in this forum.

Stefano