A Bit About How Sensitivity Actually Works

The Sovereign has a unique chip in all it's coils not found on the FBS machines. It's a pre-amp put on the RX coil which does an initial boost to the signal before it has a chance to pick up noise or become distorted. Then it's sent up the cable in a more strong or "robust" condition to be less prone to degrading along the way, at which point in the control box your sensitivity control further increases the gain on the signal to desired adjusted levels. Far as I know only the Xterra machines have a pre-amp in the coil like this, but there may be one or two other machines on the market that I'm not aware of. Perhaps the Nautelists (I know I spelled that wrong)?

A lot of people think the sensitivity control on most detectors is adjusting the transmit power. Not so, that is static on most machines, as there isn't much in way of gains that can usualy be had by increasing that beyond an optimized setting they are calibrated to. There is an amp for the Sovereign that increases the voltage to the TX winding but it's had mixed reviews. Many found it was useless in even moderate mineralization. That's why TX output power it usualy set staticly on most detectors, as there's a point of no return beyond a certain level where gains usualy aren't made in depth from it in most soils.

It's the RX signal that is usualy amplified (your sensitivity control), and that's why there are limits to how much you can use. If you increase the gain too much, too much EMI noise can be strong enough now for the machine to "notice" in it's processing, or even the ground glare can increase in strength enough to also be drawn to the attention of the processor. Think of it as using a hearing aid in a crowded room. You want to turn it up just enough to hear the guy you're talking to, but not so high as to amplify the rest of the conversations in the room so loud that they become a destraction. Have you ever turned on an amplifier for an old stereo system and heard that type of "white noise" or "hiss" in the speakers? That's the problem with too much amplification. It not only picks up more external EMI, it also picks up more internal EMI being generated by the electronics inside the box. Those two problems, along with making the ground signal too "hot", are what limits how high sensitivity can go and still be useful. The interesting question going on in the other thread right now, is that is the useful limit to where the machine begins to false, or is the limit actually somewhere much lower than that in terms of getting the best depth or at least ID at depth.

I'm surprised the coil-with-preamp solution isn't more commonplace. It's easy and cheap to do, in principle just one extra power wire is needed for the electronics, or in the case of FBS/BBS, the power is robbed from the transmit coil drive.
It would be an interesting, though difficult, experiment to see what differences are made with modest increases/decreases in transmit power (with corresponding RX gain changes to keep overall sensitivity unchanged). EMI issues would be improved with more TX signal and worsened with less TX, but I'm skeptical about 'ground saturation' effects.

Yes....... this has been stated on this forum for many years; yet, is well worth the repeat.

Also........... most EMI enters the detector through the antenna (coil). To preamp.... or not to preamp.... has attributes/detriments.

Some machine's on the market have the transmit and recieve maxed out and the control that says sensitivity on the control panel actually controls the target volume....

they run wide open and you blank out the audio response with the sens control ..even on zero sens louder targets will still report......these are your vlf deep seeker's..Like Musketeers ..1270's..turn the sens up high enough and you will hear a thresh come in...

very deep machine's in the right conditon's..

Keith

Pimento Wrote:
-------------------------------------------------------
> I'm surprised the coil-with-preamp solution isn't
> more commonplace. It's easy and cheap to do, in
> principle just one extra power wire is needed for
> the electronics, or in the case of FBS/BBS, the
> power is robbed from the transmit coil drive.
> It would be an interesting, though difficult,
> experiment to see what differences are made with
> modest increases/decreases in transmit power (with
> corresponding RX gain changes to keep overall
> sensitivity unchanged). EMI issues would be
> improved with more TX signal and worsened with
> less TX, but I'm skeptical about 'ground
> saturation' effects.

Yes, a pre-amp circuit is a very simple thing, with just one IC and a few resistors or capacitors. You can find a pre-amp schematic for the Sovereign in the Geotech forums as some of those guys make their own Sovereign coils. Far as I know the FBS machines don't have a pre-amp, just the BBS Sovereign/Excals do. At least I've never heard that, nor does Minelab say anything about it on their website where they discuss their various patented technologies, including the pre-amp used in the Sovereign as well as the Xterra line. The Xterras also communicate the frequency the coil is using to the machine, and I believe have an analog to digital conversion chip in the coil as well, which has it's merits but also it's minus IMHO.

Yes, on the BBS machines power is derived for the pre-amp on the RX coil via the voltage being sent to the TX windings from the control box, but they are wired in parallel to each other. Meaning, the TX coil is still seeing the same amount of voltage that it would without the pre-amp feeding off the same power source. Far as I know the voltage is more critical in strength of the field being generated by the TX coil, and not the current, which I'm sure Minelab made more than enough available for both the pre-amp and the TX coil anyway...VLF detectors are very low in current draw. The Sovereign from memory draws around 50ma which is almost nothing.

Either way, just saying that just because there is a pre-amp feeding off the TX power output doesn't mean the TX field strength is suffering from it. Now, some Sovereign meters that don't have their own battery do feed their power from the TX output as well. That's still in parallel, and these meters are just volt meters so they draw very little power at all, but I'd probably still prefer to use one with it's own power source just to have no doubt it's not sapping the TX's strength. Really it shouldn't be an issue as being wired in parallel only the current available would be of potential question and Minelab I'm sure took all that into account when designing the circuit board in the control box, so I don't think the TX coil would be starving for current no matter what. The meters, by the way, don't read the coil. A signal is sent down the coil cable from the control box to the meter for measuring the conductivity of the target via I think a 2 volt scale. People get confused and think that because the coil plugs into the meter the meter must be reading the coil. Nope.

First the only detectors I was aware of that had preamps in the coil was the xterra and XP Deus - did not now the BBS machines had this as well. It all comes down to signal to noise ratio. Antennas, preamps and amps all add noise. In the world of high end audio the quality of the preamp can have a huge effect on the final quality of the sound you hear. The trick is to amplify the good part of the signal more then the bad. The advantage of the preamp in the coil is you eliminate the cable as another source of noise - or at least reduce its noise contribution. Cables are common sources of introducing noise in a system so the stronger the signal in the cable the harder it is for emi to affect the end result. Like most things there is a point of deminising returns and sometimes too much amplification will reduce the signal to noise ratio so I am sure at least for some detectors that sweet spot may be before you get falsing. In fact logic tells you at the point of falsing you have already degraded the signal so the point of max depth is probably some point less then falsing.

The trick is how do you determine that sweet spot? I do not have that answer but maybe someone else does.


Bryanna

The X-Terra's coils have been dissected on Geotech1, too, they only have the pre-amp and a microcontroller for identification/authentication (and maybe to make it a bit harder to copy), there is no ADC, etc.
The noise contribution of the cable is generally minimal, though poor connections caused by worn-out nasty CB-radio microphone type connectors may be more significant. That would be one thing I would like detector manufacturers to use - quality coil connectors, such as the kind that professional audio equipment uses.

When the first tidbits of the CTX were sent out I assumed they were processing the signal on the coil running it through an analog to digital converter (ADC) and sending it to the control box digitally but turns out not to be the case. The advantage of digitizing your signal is noise elimination becomes a lot easier and complex signal analysis is better handled in the digital form. So I was surprised that the CTX did do this. I believe the Deus does this so it is not future technology.

Bryanna

Pimento Wrote:
-------------------------------------------------------
> The X-Terra's coils have been dissected on
> Geotech1, too, they only have the pre-amp and a
> microcontroller for identification/authentication
> (and maybe to make it a bit harder to copy), there
> is no ADC, etc.
> The noise contribution of the cable is generally
> minimal, though poor connections caused by
> worn-out nasty CB-radio microphone type connectors
> may be more significant. That would be one thing I
> would like detector manufacturers to use - quality
> coil connectors, such as the kind that
> professional audio equipment uses.

I thought that on the Xterra models in additon to the pre-amp and frequency identification "talk" they had in the coil, that they also had an analog to digital conversion going on right at the coil as well? I guess it's in the control box then for that part, because it's one of the things they hype about the Xterra line. What I find curious, though, is why there also aren't the same types of conversion chips being talked about in other digital machines. At some point the signal has to be converted to digital I would figure. For instance, the ground "filtering" (probably a bad name for it because conventional ground balancing sets up a filter to ignore the ground while Minelabs do something different) on the BBS machines are described by Minelab as "digital filtering". So, at least that part of the RX signal is being processed digitaly. I suspect the audio might not be, though, or at least is left in analog form as it runs through digital processing. It's audio is too analog like in terms of length and target detail for a pure digital machine IMO. Obviously it has to have some aspects of that being digital because of the numerous tone alerts. Very hard to generate that many in an analog circuit based on conductivity from what I've read. That's why the few machines that I'm aware of that are analog based that do have tone alerts are very limited in the number of tones. For instance, the aftermarket mod on the 6000 to give it alert tones I think is limited to 3 if I remember right. Does anybody know of an analog detector with numerous tone alerts?

Try a Mark 1 from Teknetics it has alot of tones!

Its a 30 year old machine..

Keith

The main problem is defining exactly what digital and analog means in terms of metal detector electronics.
Before Candy got into detectors, he was an audio electronics E. at least that is what I read somewhere.
He really knew how to get the most out of a design, like the gold detectors, then Musky, Sov and Explorers.
IMO they should have stayed with the "analog" designs and made them better, the Musky was one heck of a detector
but instead terminated it and brought out the Xterras.....a big mistake IMO...as they are jack of all trades and master of none.
A Musky with a meter and tones would have been a killer design....?