Ralph Bryant
New member
n/t
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
Changed email? Forgot to update your account with new email address? Need assistance with something else?, click here to go to Find's Support Form and fill out the form.
Are the Xterras 2, 3, or 4 filter designs ?
- Thread starter Ralph Bryant
- Start date
I don't know what the filter set up is in the X-Terra. It seems to respond faster than most of my 2-filter detectors. But a bit slower than my 4-filter XL Pro. Seems logical that there would be some, with manual GB on the X-50. But you got me thinking! Could it be possible that it does not have a filter system as we know it? If you have a micro-controller in the coil communicating digitally with the micro-processor in the control box, would it still require specific filters? Why couldn't it "self compensate" in an effort to maintain software driven parameters? And, if the digital electronics could accomplish that, why wouldn't it have auto-tracking? Hmmmmm??? Funny how one question leads to 4 more!! HH Randy
Ralph Bryant
New member
The microprocessor has to have the basic underlying electronics to run in the first place. It's analogous to the computer that tells the engine in your car what to do. The engine has to be there first. It doesn't replace all of the analog components, but rather tells them what to do and provides digital conversion and transmission where doing so will provide some benefit, either in reducing the physical size of the total components needed or in providing cleaner information transmission within the system. Where detectors are concerned, you can't send 0 and 1 digits into the ground to replace the signal transmission that produces the electromagnetic field at the coil. But you can digitize the returning information received back into the RX circuits, either to clean them up or to transmit them in a cleaner form.
It's been my experience that the lower filter machines actually have the quicker target response times and are more forgiving on sweep speed as well. The more filters involved, the more consistent your coil sweep is required to be. If I was just guessing, based on what I understand of other designs, I would think the Xterras are most likely 3 filter designs, but I don't know that for sure, and that was why I was asking. They still have a slight response delay compared to some of the true 2-filter machines like the Tesoros, not to confuse target response speed with target separation or recovery speed, though they do go hand in hand. We're actually looking at two sides of the target, with response speed being how quickly the machine acquires the target signal when you first sweep the coil over it, and recovery is how quickly the signal re-sets in preparation of acquiring the next target. I think alot of people tend to get the two terms confused.
Ralph
It's been my experience that the lower filter machines actually have the quicker target response times and are more forgiving on sweep speed as well. The more filters involved, the more consistent your coil sweep is required to be. If I was just guessing, based on what I understand of other designs, I would think the Xterras are most likely 3 filter designs, but I don't know that for sure, and that was why I was asking. They still have a slight response delay compared to some of the true 2-filter machines like the Tesoros, not to confuse target response speed with target separation or recovery speed, though they do go hand in hand. We're actually looking at two sides of the target, with response speed being how quickly the machine acquires the target signal when you first sweep the coil over it, and recovery is how quickly the signal re-sets in preparation of acquiring the next target. I think alot of people tend to get the two terms confused.
Ralph
Garrett 1350. They don't use specific filters, but utilize Frequency Domain Analysis. Reportedly, this DSP based technology produces a near infinite number of filters to remove virtually all of the ground mineralization and other "external" noises that standard 2 and 3 filter detectors miss. In essence, the digital signal processing drives the filter to the level required.
And, I agree with you on response speed, target separation and recovery speed. I call the time it takes for a detector to resume hunting mode after passing over (and nulling) a rejected target, the reset speed.
Apparently, I need to keep my thoughts about VFLEX to myself so I don't "muddy up the waters" for those who truly understand it, or further confuse those who don't. Oh well, I gave it a shot! HH Randy
And, I agree with you on response speed, target separation and recovery speed. I call the time it takes for a detector to resume hunting mode after passing over (and nulling) a rejected target, the reset speed.
Apparently, I need to keep my thoughts about VFLEX to myself so I don't "muddy up the waters" for those who truly understand it, or further confuse those who don't. Oh well, I gave it a shot! HH Randy
Ralph Bryant
New member
Randy,
No problem here with your thoughts and ideas, as I am always in the "learning mode" and soak it all in, then try my best to decypher it all into something useful and that makes sense. There is still ALOT about these machines that I don't know or understand completely.
I'm not familiar with the Garrett machines, and have only used a couple of them over the years, including an old BFO. I do know that Fisher uses Frequency Domain Analysis (better known as Fourier Domain) in their CZ designs. Since the DSP or digital signal processing occurs in the receiver circuits, just as the various filtering processes of analog, I'm sure what you are implying may be possible. Cody might jump in here and give us his thoughts, as he is more up to speed on digital technology I think.
As far as VFLEX goes, I don't think any of us totally have it figured out yet, though I've seen some interesting possibilities. It would be nice to have a Minelab Tech Forum where we could actually ask questions and learn about such matters without the fear of the competition getting a few ideas of their own.....
Ralph
No problem here with your thoughts and ideas, as I am always in the "learning mode" and soak it all in, then try my best to decypher it all into something useful and that makes sense. There is still ALOT about these machines that I don't know or understand completely.
I'm not familiar with the Garrett machines, and have only used a couple of them over the years, including an old BFO. I do know that Fisher uses Frequency Domain Analysis (better known as Fourier Domain) in their CZ designs. Since the DSP or digital signal processing occurs in the receiver circuits, just as the various filtering processes of analog, I'm sure what you are implying may be possible. Cody might jump in here and give us his thoughts, as he is more up to speed on digital technology I think.
As far as VFLEX goes, I don't think any of us totally have it figured out yet, though I've seen some interesting possibilities. It would be nice to have a Minelab Tech Forum where we could actually ask questions and learn about such matters without the fear of the competition getting a few ideas of their own.....
Ralph
Ralph, this is one similar to the halo question. Design engineers kind of agreed that users don't understand the number of poles of a capacitor so settled on filters as an explanation. I think the coil and a two pole capacitor would be two filter. A capacitor is XC=1/1/2llFC so we see the frequency component in the equation. This consumes energy and also causes the snappy or slow response. Filters will help to eliminate the noise from magnetic minerals and stray rf etc, but then slow down the response. My guess is going to be 2 or 3 filter since they are using silent search. A major reason for silent search is so users do not hear the snap crackle and pop of the noise that more filters would eliminate. Three seem to be a good number for depth in heavier minerals, a good snappy response, and when used with silent search is a pretty good combination. I think it boils down to less filters, silent search and then being able to switch between channels for good depth on heavy minerals and a snappy response.
As you know a capacitor can be constructed with a single negative and positive lead such as the old common paper and wax type. Capacitors can are also constructed to have multiple negative and positive leads. Now days they would look like a chip with the diagram looking like several capacitors in a chip. I know you have used capacitors that were the old larger type that have several leads instead of the standard two. The capacitor can be used with coils or resistors to construct filters in different configurations such a phi, LR,LC, etc. A three pole capacitor can be used with a single input and two outputs so each output is considered to be a channel. This would be a two filter. mechanical I think, since the RX coil is part of the design. I believe afte the standard two mechanical then it is more likely to be digital filters if we go to 3 to 6.
I don't pay much attention to this but this is the general idea so I am filling in the blanks. I really simple go by how snappy the detector responds and how does it take care of noise and heavy minerals, and how forgiving is it for sweep speed variations.
I don't pay much attention to this but this is the general idea so I am filling in the blanks. I really simple go by how snappy the detector responds and how does it take care of noise and heavy minerals, and how forgiving is it for sweep speed variations.
Frequency domain, Fourier Domain, Time Domain, is a way of saying pulse induction (square wave drive) or sine wave driven. That in conjunction with how the signals are processed. As an example the DFX is square wave driven but used FD to process the signals. The Explorer uses a rectangular wave to drive the coil but time domain to process the signals. So the DFX is very interested in phase angles while the Explorer is based on temporal evolution of signals so has little interest in phase angles. I think Fourier Domain is another way for a manufacture to say they use square wave drive and frequency domain. I don't think I know of any detector that use square wave drive and time domain with discrimination other than Minelab. I believe the other use square wave drive and frequency domain.
Fourier in this sense expresses that we cannot view the frequencies in a square wave with a time domain device such as an oscilloscope. We demodulate the the signal into the frequency components which are multiples of the fundamental frequency of the contained in the square wave. Fourier is a PR way of saying multiple frequency and making it sound good like there is some deep mystery to how this is done. IMHO
Fourier in this sense expresses that we cannot view the frequencies in a square wave with a time domain device such as an oscilloscope. We demodulate the the signal into the frequency components which are multiples of the fundamental frequency of the contained in the square wave. Fourier is a PR way of saying multiple frequency and making it sound good like there is some deep mystery to how this is done. IMHO
Good point. As an example when I was a system engineers working with computers in space systems we used batteries to filter the AC. No better DC that what we get from a batters. My guess is the coil and input capacitor, if there is one, is the first two filters so there are at least two channels to work with for discrimination. Heck, I look at surface mount technology and wonder what I am looking at.
Ralph Bryant
New member
I still remember all the giddiness in the detector world when the CZ was first introduced, with the general hype and belief that "FOURier domain" had to mean it operated on FOUR frequencies. 
Ralph
Ralph
Captain Kirk
New member
Cody,
I was also confused about "multiple pole" capacitors. This confusion came about because I normally think of poles and zeros in the PZ plane. In this case a single element does not have any poles or zeros if there are no parasitics. The poles and zeros only occurs when multiple elements are connected together.
It is my understanding that multiple lead capacitors (and inductors and resistors) are used particularly for high frequencies (and/or high currents) where parasitic inductance and resistance of the leads will introduce errors. This allows one set of leads (that carries no current) to be connected right across the actual element to eliminate these errors. Is that your understanding?
HH,
Glenn
I was also confused about "multiple pole" capacitors. This confusion came about because I normally think of poles and zeros in the PZ plane. In this case a single element does not have any poles or zeros if there are no parasitics. The poles and zeros only occurs when multiple elements are connected together.
It is my understanding that multiple lead capacitors (and inductors and resistors) are used particularly for high frequencies (and/or high currents) where parasitic inductance and resistance of the leads will introduce errors. This allows one set of leads (that carries no current) to be connected right across the actual element to eliminate these errors. Is that your understanding?
HH,
Glenn
Yes, and I should have been more specific in making it clear I was using the term in relationship to poles of a filter consisting of a multiple pole capacitor and coil as an example. My guess is filters are digital or RC and in chips. I thought that would only confuse it further so went with a basic LC filter explanation of the coil and multiple pole capacitor being "called 2 pole".
I have almost forgot the point but think it was I would test a detector and go by how it responds more so than by how many filters it is "reported" to have. In saying this I tried to imply that if the X-terra respond as the user likes then not to get caught up in how many filters it has.
In general a two filter design is fast and has a forgiving sweep speed. However, soil minerals come into play so I am a firm believer in testing a machine before we buy in our neck of the woods. We all know how different reports are for the very same machine from one geographical region to another. If we select a machine based on how many filters and what we think that means we may be disappointed.
I have almost forgot the point but think it was I would test a detector and go by how it responds more so than by how many filters it is "reported" to have. In saying this I tried to imply that if the X-terra respond as the user likes then not to get caught up in how many filters it has.
In general a two filter design is fast and has a forgiving sweep speed. However, soil minerals come into play so I am a firm believer in testing a machine before we buy in our neck of the woods. We all know how different reports are for the very same machine from one geographical region to another. If we select a machine based on how many filters and what we think that means we may be disappointed.
Ralph Bryant
New member
Hi Cody,
What other design or tuning factors do you think should be considered in how "quick" of a target response time we should expect from any particular machine. I'm talking specifically about the initial target response rather than the retune or recovery speed. I've noticed a slight lag in the target response speed with the Xterra-30, but the retune or recovery speed and secondary target acquisition speed appears much faster in that I can waive a coin under the coil at breakneck speed back and forth, and get a good separation of the signal in both directions. This one has me a bit puzzled. Still hoping for a smaller coil to see how much of a difference that makes on this particular machine.
Ralph
What other design or tuning factors do you think should be considered in how "quick" of a target response time we should expect from any particular machine. I'm talking specifically about the initial target response rather than the retune or recovery speed. I've noticed a slight lag in the target response speed with the Xterra-30, but the retune or recovery speed and secondary target acquisition speed appears much faster in that I can waive a coin under the coil at breakneck speed back and forth, and get a good separation of the signal in both directions. This one has me a bit puzzled. Still hoping for a smaller coil to see how much of a difference that makes on this particular machine.
Ralph
Ralph, any circuit can be reduced to a simple circuit comprised of inductance, capacitance and resistance. That RCL circuit can then have a signal IN and OUT. The filters, discrimination, sound generation, setting, speed of operation of the microprocessor, lookup tables if used, are all part of the problem. The way we mess with the RCL is the problem or blessing when it comes to very quick reports of a target, delays, and all that.
The more we add to the RCL the more sluggish the detector are compared to the real quick simple RCL box. I am including the coil, wires, circuits cards, the entire nine yards in that RCL. I don't think there is a way to get the response you like with the complexity of some of the detectors we have discussed.
Think of a detector operating at 15,000 cycles per second trying to ID a tiny signal at 8 inches through all that magnetic soil. So the engineers says well if we hit get 60,000 samples then we can have a much better discriminator. Or lets use 6 filters to isolate the tiny signal from the magnetic noise then we go trouble right here in river city. We just cannot load our truck down with a great big V8 and all the accessories and want 30 miles per gallon. I see the RCL box growing by leaps and bounds just like the truck sucking up that gas. You know how that is "if you got to worry about the price of gas then you need a smaller truck". Quick snappy responses and all that extra don't go together.
I guess I am rambling on this but that is where I am at least today. I think I am over new truck fever. Ha
The more we add to the RCL the more sluggish the detector are compared to the real quick simple RCL box. I am including the coil, wires, circuits cards, the entire nine yards in that RCL. I don't think there is a way to get the response you like with the complexity of some of the detectors we have discussed.
Think of a detector operating at 15,000 cycles per second trying to ID a tiny signal at 8 inches through all that magnetic soil. So the engineers says well if we hit get 60,000 samples then we can have a much better discriminator. Or lets use 6 filters to isolate the tiny signal from the magnetic noise then we go trouble right here in river city. We just cannot load our truck down with a great big V8 and all the accessories and want 30 miles per gallon. I see the RCL box growing by leaps and bounds just like the truck sucking up that gas. You know how that is "if you got to worry about the price of gas then you need a smaller truck". Quick snappy responses and all that extra don't go together.
I guess I am rambling on this but that is where I am at least today. I think I am over new truck fever. Ha
Captain Kirk
New member
Cody,
Do you think that the detector uses physical filter components or is the filtering done using digital filtering techniques? Digital filtering uses no components. Of course, digital filtering would place an extra burden on the processor.
Even at a fundamental stimulus of 1KHz we get a target sample every 1mS. Moving the coil at 1 foot/second then gives a sample every 0.012". Ten samples should be more than sufficient to very precisely filter the received signal. This means we should be able to qualify the target in 0.12". That would be plenty quick enough don't you think?
Keep in mind that the faster we run the processor, then the more drain on the battery. What do you suppose the clock speed is on the processor?
Glenn
Do you think that the detector uses physical filter components or is the filtering done using digital filtering techniques? Digital filtering uses no components. Of course, digital filtering would place an extra burden on the processor.
Even at a fundamental stimulus of 1KHz we get a target sample every 1mS. Moving the coil at 1 foot/second then gives a sample every 0.012". Ten samples should be more than sufficient to very precisely filter the received signal. This means we should be able to qualify the target in 0.12". That would be plenty quick enough don't you think?
Keep in mind that the faster we run the processor, then the more drain on the battery. What do you suppose the clock speed is on the processor?
Glenn
I would be amazed to see anything that remotely looks like a filter which is the point of an RCL box as an example. That takes me back to my original suggestion that a user should try one in his neck of the woods and see if it meets his/her idea of quick or slow. That is the whole point in that when we talk of filters, audio, etc, we are talking of days gone by. We might as well be talking about electron tube or transistor amplifiers.
Glenn, sorry I did not end up answering your question about the clock speed. In these posts I realize we really cannot talk about surface mount technology so try to keep that in mind in my posts and also your good advice a year or so ago about the problems with technical posts.
I have wondered about the clock speed and in particular where the Explorer is concerned. It appears to me that it is fairly slow in electronics terms from what I see on the display and hear in the audio. As you well know when we are talking of something below 100khz we are at a snails pace in electronics even for the frequencies we use in these machines. However, it appears to me that if anything they have to slow things down such as when using Audio1, 2, 3 on the Explorer.
I have no hope of more data than what can be extracted from the patents and I have kind of burn out on that at least for now.
I have wondered about the clock speed and in particular where the Explorer is concerned. It appears to me that it is fairly slow in electronics terms from what I see on the display and hear in the audio. As you well know when we are talking of something below 100khz we are at a snails pace in electronics even for the frequencies we use in these machines. However, it appears to me that if anything they have to slow things down such as when using Audio1, 2, 3 on the Explorer.
I have no hope of more data than what can be extracted from the patents and I have kind of burn out on that at least for now.