Menu
What's new
Find's Treasure Forums

Welcome to Find's Treasure Forums, Guests!

You are viewing this forums as a guest which limits you to read only status.

Only registered members may post stories, questions, classifieds, reply to other posts, contact other members using built in messaging and use many other features found on these forums.

Why not register and join us today? It's free! (We don't share your email addresses with anyone.) We keep email addresses of our users to protect them and others from bad people posting things they shouldn't.

Click here to register!

Need Support Help?

Cannot log in?, click here to have new password emailed to you

PI's and Target ID

A

Anonymous

Guest
It's not good when I wake up at 3am because my mind tends to make strange leaps of logic but here goes anyway <img src="/metal/html/smile.gif" border=0 width=15 height=15 alt=":)">
Recently Joe Patrick came out with a new TID Meter for the Minelab Sov and Elite. He took a very innovative approach to TID with his new meter. He used a target signature lookup table that he created after measuring the response to hundreds if not thousands of targets and then programmed them into the meters microprocessor for instant comparison when a target is detected.
Now for the leap in logic <img src="/metal/html/smile.gif" border=0 width=15 height=15 alt=":)"> For a long time we have known that there is a wealth of TID Information contained in the recieved signal of a PI detector but had no effective way to put this information to use. Each target has it's own target signature. So why couldn't a target ID system based on Joe Patricks lookup table concept be adapted for use on PI's. If hundreds or even thousands of target signatures from say a Goldquest SS were programmed into the lookup table you might have a viable TID System for it. Obviously due to design differences this meter would not be able to be placed on the coil wire as with a Minelab but that is not what I was thinking anyway. I was thinking more of it being an integral part of the detector's electronics. Just some early morning thoughts.
For those not familiar with Joe Patricks meter/concept there is a link to his website where you can read in detail about the concept that I am talking about.
Any and all responses welcome especially from the design engineers out there!!!!
HH
Beachcomber
Mr. Bill I hope that you will not delete this POST because of the link to Joe's Patrick's website. I know we are not supposed to post links to dealers websites but for people unfamiliar with what I am talking about there is a wealth of information on the concept and design of the TID Meter there. Thanks
 
the subject is PI related. Those answering may need to look at what your referring to.
Bill
 
Interesting, but the two s <img src="/metal/html/smile.gif" border=0 width=15 height=15 alt=":)"> ystems VLF and PI are completely different, don't know how you would wire it, also if it was that easy to add discrimination to a PI, it would have been done long ago, it is basically an ohm meter, if you look on Mod forum of Finders it shows how for a Soveriegn.
 
What is on Finders has nothing to do with the method Joe Patrick has used. Yes the front end measuring device is just a voltmeter. That part would be different on a PI. It's what he is doing with the information gathered from the voltmeter that is different and what should be able to be applied to target signature analysis of the recieved PI signal. You would need to read the in depth explantion on Joe's website and read the concept and development of his meter to get a better idea of what he has done. Today's technology is what could make this possible for a PI. Not long ago the processors that would be required for this were not available so it really couldn't have been done at least not economically until now. I am still learning about PI's and thier receive signals characteristics. But hopefully there is a way the this type of signal analysis can be applied to PI's. If there is it could be the next advance in PI design.
HH
Beachcomber
 
I wish you were right, but I would think Eric Foster would have done this by now, so don't think it is that simple, the PI is completely different and cannot be read the same way a VLF can, but here's hoping maybe some of the new devives or chips may help unlock a way to. HH
 
Hopefully we'll hear from Eric or some of the other people who do know a lot about PI signals and thier thoughts on what I have posted. I wasn't thinking about any type of discrimination just a way to quantify and identify the received signal which it seems is what Joe has done with microprocessor controlled meter that he came up with. Kind of thinking outside the box as it were <img src="/metal/html/smile.gif" border=0 width=15 height=15 alt=":)"> <img src="/metal/html/smile.gif" border=0 width=15 height=15 alt=":)"> <img src="/metal/html/smile.gif" border=0 width=15 height=15 alt=":)"> <img src="/metal/html/smile.gif" border=0 width=15 height=15 alt=":)"> To use this method with the PI you would have to multiple samples of the received signals waveform and convert it to ones and zeros creating a unique code for that particular signal. The more samples you could take the better your target resolution would be. This would need to be done for hundreds of different targets under a variety of conditions and then all these signatures would need to be programmed into the microprocessor's memory as a lookup table. from then on when the detector detected a target it could compare what signal with teh signals already stored in memory and determine what the target is most likely to be. As we all know from VLF TID no TID is perfect as every possible combination can never really be taken into consideration. There will always be a margin of error. Still if it could work it would be a step forward.
HH
Beachcomber
 
Your welcome and here is a link to a website that also has a lot of info on metal detectors and how they work.
HH
Beachcomber
 
Joes does EXACTLY the same as I do in my Sovereign TID meter and have been doing for the last two years.
Read the voltage, this gives you a digital value which is used to point to an entry in a look up table.
Example;
Coin gives a voltage reading (integer) of 1056 (assuming 12 bit PIC A to D used), a look up table programmed into an Eprom. The reading at memory address 1056 is a figure to control the PWM or something. This is then used to drive a precise current to a meter to send the needle to send to a part of a display with a pre-printed item on it.
Nice idea, but...
1) Nothing complicated, or new.
2) Somewhat pricey for what it is.
3) Not very flexible in how it does things.
As for the "Iron indicator", The Sovereign gives a negative voltage out when it blanks iron, a simple LED, transistor and a couple of resisistors can be used to do the same.
Pulsar Electronics meter was miles better, in so many respects, not least because it was user programmable, and multi lingual, also MULTI CURRENCY! The world does not run on cents and dollars...Sorry!
 
So why couldn't this concept also be applied to a PI's recieved signal. Instead of having voltages programmed into the Eprom why couldn't target waveforms be sampled, encoded, and then programmed into the lookup table. It was just a thought. Seems like from what people are saying the technology still doesn't exist to do this effectively or reliably with the PI's received signal.
HH
Beachcomber
 
I'm no expert! Put your PI over a target and slowly bring the coil closer and closer to it. The volume goes up or the frequency (tone) goes higher in pitch. This is because the voltage is increasing. What would a lookup be doing? It would seem to me that it would be all over the map with what it was looking up as the coil moved towards the target and the voltage increased, wouldn't it? Wouldn't this make this method pretty much unusable?
FJ
 
TID is not all that difficult to implement in a PI detector; so why hasn
 
I don't believe so. What you are talking about is signal strength. I am talking about the target waveform as seen on an oscilloscope. Try looking at it this way. If I take an audio signal of 1khz with an amplitude of 1volt peak to peak and look at it on the scope and then I reduce the signal level to 0.5 volts peak to peak I still have the same waveform only smaller. I would think if the TID lookup table were based on the recieved waveform that it could work.
On your detector the reason the audio frequency changes as you get closer and farther from a target is usually a function of the design of the indicating circuitry...a common method is called VCO or voltage controlled oscillator which some companies use. This causes both a change in amplitude and audio frequency as you move closer and a farther from a target. Without the voltage controlled oscillator the sound would just louder or quieter as you become closer or faarther froma target.
HH
Beachcomber
 
Eric,
Has any reseach been done on the specific time delay uS from TX switch off or the number of time delays to analyze in order to extract reliable target ID fingerprints?
I suspect that one could easily start a PI target ID design with two adjustable time delay PI receive circuits that could be examined (meaning combined, summed, or integrated) in a variety of ways. Based on the specific hunting location, a set of time delays and combining methods would allow you cover the most commonly found desireable targets, ground conditions or unwanted junk.
If the above approach proves to reveal enough reliable target details, then additional receive delay circuits could be added to extract even more target information. One key design issue would be to determine the minimum number of delay circuits necessary to extract enough reliable target ID information. What is your feeling about this appoach?
bbsailor
 
Yup, yup, I saw Eric's post and figured out what you all were talking about. Makes sense. Has anyone used FFT methods to see what can be seen in the frequency domain?
FJ
 
I tried a Russian discriminating P.I. last year which lacked in depth and sensitivity but did work.
It had a LCD display providing numbers. What I wanted to ask is why good items came both sides of the number that indicated iron. With other types of machines iron comes first(left) with other metals nicely ranging to the right.
Would the set up you suggest also indicate wanted items above and below the iron number?
 
Is the amplitude of a target the result of signal decay ?
Wont a fast decay target like a thin gold ring at 4 inches display the same rate of signal decay as a slow decay target such as an iron nut which is farther from the coil?
Put another way, will the rate of signal decay be same for a specific metal item as it is moved farther from the coil?
As always, interesting discussion in here.
Rgds, Bing
 
The decay time of the object signal is independant of the signal amplitude. Whether the target is close to the coil, or at the limit of detection, makes no difference. Only thing is, if the targetis far away, the signal disappears into the background noise much quicker.
Eric.
 
You must log in or register to reply here.
Top