VoyForums
[ Show ]
Support VoyForums
[ Shrink ]
VoyForums Announcement: Programming and providing support for this service has been a labor of love since 1997. We are one of the few services online who values our users' privacy, and have never sold your information. We have even fought hard to defend your privacy in legal cases; however, we've done it with almost no financial support -- paying out of pocket to continue providing the service. Due to the issues imposed on us by advertisers, we also stopped hosting most ads on the forums many years ago. We hope you appreciate our efforts.

Show your support by donating any amount. (Note: We are still technically a for-profit company, so your contribution is not tax-deductible.) PayPal Acct: Feedback:

Donate to VoyForums (PayPal):

[ Login ] [ Contact Forum Admin ] [ Main index ] [ Post a new message ] [ Search | Check update time | Archives: 1, 2, 3, 4, 5, [6], 7, 8, 9 ]

[ Next Thread | Previous Thread | Next Message | Previous Message ]

Date Posted: 12:54:58 08/28/02 Wed
Author: Reg
Subject: Re: PI's and small gold - Questions (Very long post)
In reply to: Steve Herschbach 's message, "Re: PI's and small gold - Questions" on 11:19:47 08/28/02 Wed

Hi Steve,

I will try to take your questions one at a time.

First, lets take the idea of having two different channels. Minelab actually pulses the coil with two different pulse rates and pulse lengths. It is my opinion they do so to overcome one of the problems of ground balancing and nothing else.

When you ground balance a PI, about the only way to do it is to take two different samples, lets say at 10 usec and then at 30 usec. Both samples will contain ground signal, so the trick then is to amplify the later and subtract it from the first. If the amplification is correct, then the ground signal in the first sample is eliminated, thus you do not hear any ground response.

One of the negative side effects of this type of ground balance is, unfortunately, some nuggets have very similar characteristics as the ground. Other nuggets, usually the larger ones, will have a response, nomally called a time constant, that is longer than the ground. All targets have a time constant which determines how fast the signal decays. The longer the TC the longer it takes to decay.

What does all this mean? Well, in simple terms, when the first sample is ground balanced, some nuggets will respond with a loud signal, some with a little response or maybe no response at all, still others will respond with a negative response, much like a hotrock does on a VLF. It is strange, but it does happen and there is no direct correlation between nugget size and the ground.

To overcome this condition, the negative responding signals can be rectified and converted to positive responses. That takes care of the negative signals but still leaves those nuggets that now give little or no response.

To be able to detect these, a second pulse is generated having a longer pulse on time. A longer pulse will change the ground balance setting, meaning it will now take a different amount of amplification of the second signal to match the ground signl of the primary or first sample. Again, the second sample will be subtracted from the first and again there will be nuggets that give a negative response and some that will give no response.

Fortunately, the ground response is dependent upon the duration of the pulse applied so the two samples create two different ground responses. Fortunately, gold and other non ferrous objects will not change as the result of different pulse lengths, so their responses will be the same whether a long or a short pulse is used.

In a nutshell, this means that nuggets that produce no response when using a short pulse will produce a nice audible response on the long pulse, and visa versa. So, if both "channels" are now added together almost all nuggets will give a nice response. I say almost all, because there is always something strange about some nuggets. It could be the matrix they are attached to, etc.

Now, lets tackle the question of a knob for adjusting the sampling rate. Most manufacturers are using microprocessors for generating the pulse and sample timing. A pot could be used as is the case on the C Scope CS 7 which uses a microprocessor also. A simplier way would be to use a toggle switch to increase or decrease the sampling time. Regardless of how it is done, it could be accomplished within the limits of the circuitry and the coil design, with the coil most likely being the determining factor.

The signal from two or more different delays could be used to change the tone of the audio signal produced. This would be the logical technique to use on a nugget hunting detector.

Lets tackle minimum delay requirements next. All coils have to match the circuitry used regardless of their size. There are certain factors that require the signal from the coil with no target present look the same regardless of the coil size. In other words, as far as the circuit is concerned, a 10" coil has to appear identical to an 8" or even a 14" coil. If it doesn't, then strange things happen that do not allow the circuitry to work correctly. In the worst case, the detector may appear to not work at all or have an extremely weak output.

Matching coils to work at 15 usec isn't that difficult, but it does get tougher at 10 usec. One of the critical factors is the coil cable. As the cable gets longer as is necessary on a hip or chest mount detector, the more difficult it becomes to obtain a 10 usec delay.

The coil windings, the coil cable, and something called the damping resistor all interact. Lengthen the cable and the damping resistor may have to be reduced in value. But, if the damping resistor is reduced, then the minimum delay will most likely have to be increased.

The easiest way to address this is to select a delay where things are more forgiving. By selecting the damping resistor for the worst case conditions, then coils and their cables can be less precise. This makes it easier for mass production.

Now to tackle your question about discimination and dual sampling. Actually, it is possible to take many samples of the received signal. The limit would probably be set by the interaction of the samples themselves.

Could a multitude of samples be taken? Yep, but the only logical way to implement this concept would be to use a microprocessor type circuit. Most manufacturers use a specialized one called a DSP. It would allow for very fast sampling.

Would the various samples be of value? Could be except for the ground signal. Because the ground response can be very strong, weaker target responses can easily be masked. Utilizing the ground balance technique mentioned above, would produce some gold signals that are positive, some negative and some would be nil. If multiple sample combinations were to be taken, and then anlayzed, a better analysis could be done, but it would be a maze of different responses.

If you look at a VLF having a meter displaying where gold might fall, you will see that it covers the entire range from foil to a $1. It really gets worse since the ground can influence the signal enough that small gold can fall in the iron range if the mineralization is bad enough.

Now, on a PI, two problems exist. First is the wide range of responses gold can give like on a VLF. Most small nuggets will quickly die out signal wise in 50 usec or less. So, if a sample is taken at 50 usec, then most small gold will produce no signal. However, there will be nuggets that will produce a weak signal, some a stronger signal, and a few with a really stong signal. That means then a later sample has to be taken to see if it is larger gold or not.

In the meantime, iron objects act just the opposite. Most have a very long time constant, some objects have a shorter time constant, and a few objects have a very short time constant.

Now, if the delay were to be increased, then both nuggets and more and more iron objects will not give a signal. The longer the delay, the more iron that will fall into this category. At some point it would be impossible to determine if the object were a nugget or a certain type of iron object.

As you can see, the concept of disc on a PI becomes quite complex.

Well, I have already taken up too much room at this time. Sorry for the long post. Hope it helps.

Reg

[ Next Thread | Previous Thread | Next Message | Previous Message ]

Replies:

[ Contact Forum Admin ]

Forum timezone: GMT-6
VF Version: 3.00b, ConfDB:
Before posting please read our privacy policy.
VoyForums(tm) is a Free Service from Voyager Info-Systems.
Copyright © 1998-2019 Voyager Info-Systems. All Rights Reserved.