Reg...Big Nuggets?

[Anonymous]

Hi Reg,
I notice that my GQ is very sensitive to small nuggets at depth, but as the nugget size increases the expected detection depth does not appear to be relative.
From what I’ve learned from other posts, the whole mass of the target needs to be excited by the pulse, not just the skin, so as to produce persistent eddy currents. The larger the nugget, the greater the inherent time constant, (ignoring constituents) and the longer the pulse required to saturate, the core structure.
My intention is to replace the clock and decade driver with a PIC, and use alternating short and long pulse widths, and have the reject control independent of frequency.
I’m wondering what value delay times have you been able to achieve on the HH shark at longer pulse widths?
A 30us pulse on the GQ gives a 10us delay. At 30us the current has risen to about 94% of steady state, and a delay of less than 10us is achievable. After 100us the current will have only risen another 6% or so, is a delay of less than 10 us still possible?
I plan to build a HH someday, as I want to observe what you said about short and long pulses with metal/mineral targets. Sounded very interesting…….?
I would also be able to answer my own questions.
Thanks Reg for generously sharing your knowledge.
Ta
Kev.

 

[Anonymous]

Hi Kev,
First, what do you mean when referring to large nuggets? Over here, I consider a 1/2 oz, to be large since they are so difficult to find. Larger ones are even much more difficult.
Now, as for the Hammerhead (HH)I have been able to get down to 10 usec delay with a pulse duration of about 100 usec. I briefly tried longer pulses but saw little advantage. At the time, I wasn't too interested in the keeping the delay that short with the longer pulses.
Part of problem of trying to compare the results between long and short pulses could be because of interaction due to the size of the batteries used. For testing purposes, I do recommend larger batteries and maybe a regulator type circuit, depending upon the accuracy of the testing you desire.
So many things interact it is hard to tell just what happens as the pulse width is increased. On the HH, one has try to do their best to maintain the same threshold when testing because of the non-linearity of the signal at low levels. As the pulse width is changed, the threshold changes just slilghtly which alters the results.
Overall, the HH is a really great for experimenting. For those who are wanting to experiment with different designs with various pulse widths, etc, such as what you described, I think the HH is as good of a design as you will find in kit form. One just has to be careful to try to keep all other factors equal when adjusting one item such as pulse width.
BTW, it is interesting to see what happens to the decay curve as the pulse width is increased.
Now, I think you will not see a dramatic increase in depth on larger nuggets as only the pulse width is increased. At least, I didn't, but I really haven't experimented that much testing very large low conductive targets. Anyway, it certainly is worth building the HH for the testing you desire. I experimented using much larger NiMh batteries and found things to be much more stable. Personally, I found some 4/3 A 3800 mah batteries at an extremely good price and used them.
Reg

 

[Anonymous]

Hi Reg,
Where I'm hunting, an ounce nugget is big, it's geologically similar to Arizona. Really big nuggets in the tens of ounces can be found in areas having the same geology as the golden triangle in Australia, and I hope to take some holidays at a couple of these places over the summer months.
It's interesting that you mention the batteries. I bought a couple of sets of fairly low capacity NiCads for about half the price of a NiMH pack. Maybe they can't deliver the goods when the going gets tough. When one thinks about it, Lenz said that coils don't like any changes to the status quo, so if the battery potential isn't maintained sufficiently strong enough through the Tx cycle, it's got to cause decreased detection range.
I must remember that it's often the simple things that make a difference. I'll try some different combinations and see if it helps.
Cheers Reg.
Kev.

 

[Anonymous]

Hi Kev,
Once you build the HH PI, you might want to observe the voltage at the FET when the audio is activated. Then adjust the various associated audio controls to allow for a signficant audio interaction and observe the difference. I think you will be amazed at just how much of an influence it can have.
On a different note, I think I read where you installed a DD coil on your GQ. If so, I was wondering if you have had the opportunity to try it in some of the more difficult ground in Australia, or bad ground in NZ.
I am curious as to how much it helps with the ground conditions over there.
BTW, did you modify your GQ to allow the use of ML compatible coils?
One other note, have you tried reducing the value of the resistance in series with the FET on your GQ? This should allow a little different TC for test purposes.
Cheers,
Reg

 

[Anonymous]

Hi Kev,
The large reservoir capacitor across the TX is there to buffer the coil from the battery during the TX on period. I have found it useful in development to insert a small resistor (0.1 ohms) in series with the ground end of the coil. Using a scope, you can look at the coil current waveform and see how it changes with different pulse widths and voltages.
Eric.

 

[Anonymous]

Thanks Eric,
That's good to know. I had been wondering how I could unobtrusively observe battery performance at the coalface.
This would be the method you would use to ensure that the charge storage capacity is sufficient?
Cheers
Kev.

 

[Anonymous]

Hi Kev,
Yes, you obviously get an exponential current rise which flattens off when the TX pulse width exceeds the total rise time of the coil. If the pulse then starts to droop, either the capacitor is not big enough, or the battery impedance too high.
Eric.

 

[Anonymous]

Hi Reg.
I've been following your posts on the external audio amp, and can see how isolating the headphones from the main circuit is a good idea. I'm going to try this when I've finished playing around with pulse widths and batteries.
I've not taken my GQ to OZ, what I meant was, some of the big nugget areas of NZ have the same types of rocks and gold mineralisation as Victoria. In fact the West Coast of the South Island is on the edge of the Australasian continental plate, which is juxtaposed to the Pacific continental plate on the East coast. The pacific plate is sliding southwards while the Australasian northwards, at a rate of 10mm per year.
Miners who arrived from Oz in the 1860's thought that dust from OZ had blown across the Tasman sea and deposited itself in NZ, and that's why the gold producing rocks looked the same, I don't know how they thought the Gold came over the sea though!
We don't have the laterite like they do in Oz, so the ground is relatively quiet. I mainly built my first DD for beach hunting here in Christchurch, where the beaches have a lot of black sand derived from volcanoes. Some of the places I hunt for native gold are in highly metamorphic ground that does have relatively high Fe mineral content. The DD does reduce this considerably, but I still need to extend my delay too much for my liking. When I can slot the PIC in where the 4017 sits, I hope to do some trickery on the integrator sample times, so that I can reduce this ground noise further.
I'm presently working on a new coil though, expressly for nugget hunting, and one of the reasons is, I can't afford the ML coils (hence I haven’t modded my unit) anyway, with the help I've received from yourself, and others, I believe I‘ve been able to make a comparable product. My wife will tell you, I've spent heaps of time fiddling and experimenting. So it should pay off, since I'm tailoring my coils for the conditions at hand.
I'll be taking another trip soon, with my new coils, and hopefully some mods on the GQ, so I'll let you know how I get on. It must be appreciated though, that NZ is such a diverse geologic hodgepodge within such a small area, I guess it's the result of being on the cutting edge of a continental collision zone. There are tough areas, but as far as I’ve been I can say that the DD does really help.
I also want to build and try what I call the OD. It's a mono with one D. This would enable me to search in quiet creek beds with the sensitivity of the mono, and if I venture onto the benches above, and the ground becomes noisy, a switch will couple the D to the receiver. It wouldn't have the same ground ignorance of the full DD, but in less mineralised areas I think it would be handy. It can be built in a standard DD shell.
My present DD is only a 10-inch shell being a prototype. The coils are effectively only the equivalent of an eight-inch coil. I’ve got some 10 x 14 inch shells so my next ones will be much more effective. I’m also going to splash out and buy some expensive cable. I’ve noticed that poor cable spoils an otherwise well executed build.
I noticed that you had reduced your series resistor. When you posted the info on modding the front end for DD use, I spied the parallel resistor, looked about 540R?
So your series R would be about 36R + FETR + coilR?
I haven’t altered mine at this point. I worked out that I could reduce it a bit before over reaching the FET.
There seems to be a paradox here, where reduced series R increases the time it takes to achieve steady state coil current and visa-versa, it’s something I’ve always had trouble getting my head around.
I’m not completely sure either how these parameters relate; does voltage determine the amplitude of the target response, and current, how quickly this is achieved?
I realise that it all depends upon the target’s time constant, which is a function of L & R, can it also be LI/V?
Can it’s time constant be manipulated by the I & V forming the flux interrogating it, or is it absolute? Trying to remember previous posts, I think Eric would say NO to the former and YES to the later.
Sorry Reg, just yabbering on. Must do some work.
Catch you later.
Kev.

 

[Anonymous]

Hi Kev,
I like your coil idea of a OD type design. I am going to have to try one to see just how much of a difference there is in the ground reduction. If it works like I think it will, it will be more like a lop sided concentric coil. Ground response should be greater than with a regular DD, but the larger detection area should help with detection of smaller objects.
Now, changing the topic again, Eric's new 2 filter design should work very well, especially with a DD or a concentric coil. At least, I would think it would better with them than on a mono considering any target response on a concentric or a DD should be a higher frequency than the typical ground signal. Anyway, it is something to think about as an alternative to full blown ground balancing.
The basic principle of a double derivative type circuit is used in VLF motion discrimination to minimize the ground effects and it works there quite well.
Now on another different note, you are right, I have reduced the resistance in series with the FET, but more signficantly than you noted. With the piggyback installed, I have about 20 ohms total where the 39 ohm once was. The lower resistance makes it a bigger challenge to get the delay down.
I might as well bounce around again. Don't sell your 10" DD coil short. I find I do not lose much depth with the DD when compared to a similar sized mono coil even though the windings are considerably smaller. The less noise and less ground response generally allows one to hear deeper targets. At least, that is what I have found.
Ok, I have avoided your math questions because it has been too many years since I worked with Laplace Transforms or diffy Q's. I am not sure I can even spell them right,let alone do them.
As a SWAG (scientific wild asss guess) I would say you can subsitute I/V for 1/R. However, solving for I and V will be dynamic, I would think.
BTW, I sent a couple of emails your way. I was just wondering if your got them ok.
Reg