NOT A PI DETECTOR

[Anonymous]

The Minelab multi-frequency machines are NOT PI's. What is now called BBS is also NOT a Minelab invention. Fisher came up with using square waves which provide strong odd harmonics. The receiver seperately demodulates the fundamantal and harmonic frequencies. Each frequency is demodulated by two balanced modulators which provide the resistive "R" and the Reactive "X" signal for each frequency. Different targets have different ratios of "R" and "X" for each frequency. This provides a really good discrimination ability which is especially useful against iron. The method also requires a balanced search-coil
The patent quoted from was for a Minelab PI detector and NOT for their Multi-frequency BBS detectors. The word "Discrimination" seems to refer more to the ability of the method to discriminate against hot rocks etc from metallic targets. I don't blame anyone for becoming confused. Bruce Candy obviously writes the body copy of his patent applications in a confusing way. I deal with patents every day of my working life. I also have numerous patents of my own. I can tell you that I have always been disgusted by the way that Candy's patents were written. No self respecting engineer should ever write such a crock of rubbish for any reason, Dave. * * *

 

[Anonymous]

HI Dave & all
I suppose the only way to actually settle this question is to put a CRO on to the output and see if there is a continuous square wave pulse emitted or if indeed the pulse is interuped for a listening pause after each pulse.?????? I believe that it was while developing this detector that Bruce Candy came up with the idea for the SD series detectors.
It is a pitty that Minelab dont do something aabout making the Sov/Exp system into a true discriminating gold detector. Running frequencies as high as 100khz would add considerably to the ground cancelling problems. If they were to make the machine drop off the higher frequencies as the ground conditions got worse they might be able to come up with something interesting. Perhaps they could make a detector that selected the best frequencies for the prevailing ground conditions. they would also have to up the sensitivity to small gold to be viable. A detector such as this, if it could be made to work, would be ideal for high trash areas but would never replace the PI detectors for deep targets.
Cheers
Steve D

 

[Anonymous]

Dave
Thanks for answering this question I once wondered about (exactly what is this BBS was refering too).
Certainly the x and r response changes with frequency. So do these guys run a pair of demodulators (x and r channels) at the fundamental, a pair at the third harmonic, a pair at the fifth, .... ? And what great things can you do with this information? The signal is going to be weaker at the higher frequencies, I assume, since it is a harmonic (weaker than the fundamental) and the ground may absorb (dissapate) more. The front end amplifier (amplifiers)will have to be narrow band in order to keep the noise down and will have to be tuned to each frequency. Sounds like one would need one narrow bandpass for each pair of demolulators.
Can't tune the transmit coil for the transmit frequency, since need to do square wave, so this will cost some battery power.
You have tweaked my interest in learning more about this scheme. Course if there is no clear reason for it, that would be interesting too.
JC

 

[Anonymous]

Hi Steve,
Ok, you got me, what's a CRO? Is it a cathode ray oscilloscope?
I haven't looked at the output of the BBS machines, but I have looked at the output of the SD's and they are a multi-period device, if you want to call two different pulse widths a multi-period machine. In fact, somewhere have a pic of what the SD output looks like.
Reg

 

[Anonymous]

I don't think they are really using all 17 or 28 frequencies. From what I can tell, there are only 3 I/Q demodulators, meaning they only use up to 3 frequencies. If you read their literature & ads, they only talk about <I>transmitting</I> 28 frequencies... nowhere (that I've seen) do they ever claim to analyze 28 frequencies!
White's DFX is similar... lot's of harmonics transmitted, but they state up front that they only use 3kHz and 15kHz, the fundamental and the 5th harmonic.
Obviously, for this scheme to work the TX has to be wideband. The RX can be narrowband at each demod if you limit it to a particular frequency (can't hop it to different harmonics). If you want to hop a demod pair to another frequency, you either need a WB frontend or switch in/out tuning caps. I think White's has a patent on switching the tuning caps.
I just got a DFX, and I have an Explorer on the way. Hope to rig up the oscope to take a look at these beasties...
- Carl

 

[Anonymous]

Carl
Thanks for the response, I'm not sure how they could be using all the frequencies either, or that the higher harmonics could provide any information past 0.5 inch from the coil.
Certainly it is easy electronically to generate the square wave and having reactive and resistive information at two frequencies could be used in some type of discrimination/identification schemes or ground balancing.
Have to admit I don't know much about this one. Thanks again.
JC

 

[Anonymous]

The phase responses from different metals varies with frequency. So if you use several frequencies, the different phase returns might help ID the target.
There are trade-offs with frequencies, higher ones give better response to low-conductive targets but don't penetrate soil as well. But even 100kHz is perfectly us-able... in fact, 1kHz, 10kHz and 100kHz are evenly spaced (logarithmically) and might make good choices. Hmmmm...
- Carl
PS- had to hyphenate us-able... try using this word in a posting without the hyphen <IMG SRC="/forums/images/wink.gif" BORDER=0 ALT="">

 

[Anonymous]

Hi everyone,
It's great reading posts where people are speculating how something works! What is suprising is how many people guess correctly. The multi-frequency detectors usually use two or three frequencies. As an example a pulse is transmitted with a fundamental frequency of 5KHz. The third harmonic signal will have an amplitude of one third of the fundamental frequency. The amount of power it contains will be one ninth that of the fundamental. Still, you now have a 5KHz - 15KHz detector with synchronous signals. The receiver will bandpass filter the signals to seperate them prior to demodulation. The filters do not have to be too sharp. It is also possible to use the other odd harmonics such as the fifth, seventh, and ninth, harmonic. The amplitude of the ninth harmonic would however be at one ninth the amplitude of the fundamental and contain one eighty first or 1/81 of the fundamental power.
It is also possible to transmit a series of multiple period pulses. To do this we send a series of pulses of for example, 333.3uS which have a fundamental frequency of 3KHz. We use the third harmonic at 9Khz and the fifth harmonic at 15KHz. We now switch to another pulse width of say 250uS and have 4KHz 12KHz and 20KHz. As Carl noted, there is a different ratio of "R" to "X" for the same target (or ground minerals) for different frequencies! The combination of "R" and "X" provide us with the phase angle we see (in air) for different targets. Minelab use a plurality of different frequencies. They store information about different targets and their response to different frequencies in a micro computer. Anyone who has ever used a Minelab Soveriegn or the later Explorer will tell you about the fantastic ability of these detectors to knock out iron. They can also detect a coin right next to a piece of iron. They are not the deepest seeking detectors and don't come near the depth of a PI and are useless for prospecting. They are however a most useful detector to own. Dave. * * *

 

[Anonymous]

Hi Dave
I guess, like alot of people, when I read about all these multiple frequencies I knew they were obviously transmitting a rectangular pulse or square wave, but after that wasn't sure what was going on with these detectors.
Easy to see how someone could get confused about what to call em (pulse or vlf). In fact an arguement could be made for either or better yet neither. Sounds like its own species to me.
Nice to know that this is a useful system when iron is a problem or I guess sharp discrimination is called for. Well I learned a lot!
Thanks again
JC

 

[Anonymous]

What I think would be an interesting project is to weight the harmonics to the same level as the fundamental. Depending on the phases, you can get a funky sine(x)/x looking waveform. Those who have MathCad can try it out.
Why hasn't it been done? I think because the waveforms are difficult to generate, although not so much anymore. There are easy digital methods that would work. Still more complicated that blasting out a squarewave.
- Carl

 

[Anonymous]

nt