GPZ 7000 Tear-down

[jasong]

Ok, he has to be reading these posts. 😄

His latest video shows some similar things about the 6000 as I was saying in prior posts about desensitizing and staying desensitized. His just caused by EMI, but I know big iron also affects it too and I think he'll discover that soon with more experiments as well. The problem is when the auto adjustments are sticking around rather than being auto readjusted back up when the EMI/iron goes away, and you never know if you're running dumbed down (mine were 3-4" reduced in my brief tests) all day from an auto adjustment event potentially hours prior, because there is no way to check. Trust in a detector is paramount to do good work, this is bad.

One EMI/iron event might cause you to run desensitized for the rest of the day? Do we have to reboot every 5 minutes to avoid that? Even after rebooting I was still desensitized sometimes in my brief tests, but that too was unpredictable. It was happening in "manual" too, more problematic. 

* Bleh, deleted paragraphs of me ranting. It's Minelab. It is what it is. I'm not going to go on another single man crusade to fix yet another thing for a detector which I am just going to retire the first chance I get to get something better anyways. I don't get paid or free detectors, not happy about doing free work for them. People can draw their own conclusions, but I know what I've seen in the field and the 6000 definitely misses a lot of gold well within it's technical range to hit, due to unpredictable adjustments, even when operated by experienced detectorists.

[jasong]

He's posted the results of the modded 7000, first one I've ever heard about. 

He's getting like 3" more depth on a AUS 20 cent coin, looks about like a US quarter? That would be exceptional if so. A big stress on the "if". 

I'm skeptical though, because the only mod he said he did was reduce the self generated power supply noise. Less noise should allow you to run more sensitivity, but he tested a stock and modded GPZ both on 16 gain, so no sensitivity increase.

So where is the depth gain coming from? The stock machine sounded pretty similar in terms of EMI to the modded machine too. Anyone got a better idea of what is happening here? I'm not quite understanding where the depth gain is coming from.

 

[phrunt]

Could it be the self generated EMI hindering performance?  They may have sounded about the same but doesn't mean they are.

I always felt my GPX 6000 did better after it's EMI mod fix.

It really makes you wonder though, some people like Woodys GPX 4500/5000 mods, others think they don't do much at all if anything.  JW has a GPX 4500 with woody mods and he seemed to like it over a stock one from memory.

I just can't understand how smart people that are capable of making a GPZ 7000 could make such a simple mistake that a minor shielding change alters performance so much so I have my doubts about the mod so far without more information.

The 6000 was different, it was a part that for some worked fine, others not, so inconsistency in components and all those testing it perhaps had good ones and it's when hundreds get out into the public's hands the dodgy ones showed their ugly faces.  Minelab should have done all they could to get their hands on a noisy one being returned under warranty and troubleshooting the cause though, it shouldn't have taken a year or more to work it out other than they didn't want to and were hoping the problem would just go away.  Maybe they just thought it was a bad run of components that would sort itself out with no change to anything without even working out what is the cause as a new batch comes through with the problem solved and those with the bad ones will think it's normal, pretend the problem doesn't exist and it will go away.

Maybe they're just always leaving some fuel in the tank for the next model, I can see it now.... GPX 8000, 3" more depth on a test nugget (20 cent coin in this case) over the GPZ 7000. 🙂

[jasong]

9 hours ago, phrunt said:

Could it be the self generated EMI hindering performance?  They may have sounded about the same but doesn't mean they are.

As far as I know the 7000 doesn't auto adjust sensitivity down due to EMI though. So if he had the sensitivity set equal on both machines, one should just be quieter than the other. I'm not sure how one machine is that much deeper than the other just by making it slightly quieter though, I can't see the mechanism for which that might happen. The purpose of making machines quieter is being able to boost sensitivity (and to reduce target masking), but he doesn't appear to be boosting sensitivity.

Target masking can occur with too much noise, but that doesn't appear to be happening here as it's the same target in open air, with similar noise levels, and on one machine it's simply being detected much deeper with very similar noise levels. Dunno... I'm a bit skeptical though, maybe he'll explain it all further?

I personally don't believe the 6000 was inconsistent among different owners. My observation has been some people simply didn't recognize it. The problem with it existed on literally every single 6000 I've ever come into contact with. In many cases the owners themselves were unaware, or saying they didn't see the issue - until I used it and showed them. Most of these people were primarily headphone users, or people who insisted excessive noise on detectors is "part of the game", etc.  I haven't encountered a single 6000 that didn't have that issue, so I doubt it was unreliable components, I think it was a design flaw with unshielded components on every 6000 made up until a certain point. At least, I haven't yet seen any evidence to prove otherwise, but I'd be pretty curious to see a 6000 in person that was bought before the cut-off date which doesn't have this issue when in my own hands, if anyone in the US wants to show me one in the field here. 

But yeah...as to Woody mods, I'm skeptical. If he is reading here, it'd be interesting for him to post and answer some questions maybe. I too find it hard to believe ML would make such basic, correctable errors... But they did seem to do exactly that with the 6000 unshielded components, so...

That Woody hasn't found the EMI/speaker/whatever issue on the 6000 himself, when this appears to be his primary job, makes me wonder too. Or maybe he has a newer 6000 with the fix already in place?

Also thought similar re: your last sentence - I think they probably did leave some gas in the tank for whatever the 7000 successor will be, easy gains in combination with lighter weight. But with this long between models, I'd sure hope for something else good in terms of advancements. 8 years is an eon in the tech world. 

[bklein]

I asked him if he did any CTX mods - he said no. Boy he has guts to get into a GPZ.  After my episode with my GPZ I’m keeping it stock. Very expensive to replace anything in it.  The CTX just feels like it has more to offer with tweaks. A gold mode for example. I mean if he tweaked it to obviously outdo a Manticore wouldn’t he get business?

[D&P-OR]

On 4/11/2023 at 3:29 PM, jasong said:

Ever seen Star Trek? In it is a race of beings called the Borg that are half computer and half alien. They act like computers insomuch as they won't attack an actual invader on their own ship unless they calculate them to be a threat, they'll literally let trespassers walk around among them and prefer to spend time by paying attention only to relevent matters at hand, not seeing the potential problem it might cause in the future having intruders there. But as soon as they change their minds and see you as a threat, you will be rapidly assimilated or destroyed, but often it's too late. And they pretty well assimilate all technology around them and make it into super Borg tech. No one debates they don't do it as well or better than anyone else in the universe though.

In a way, Minelab reminds me of the Borg. It seems like there are a number of very obvious things they simply choose to ignore because they simply aren't viewed as causing problems...yet. But when it's finally made obvious to them a problem exists, then they dedicate all sorts of resources to fixing it, but often much too late. They sure do like some good ole assimilation via patents from what I can tell too. 

"An ounce of prevention is worth a pound of cure" comes to mind!

[Chet]

A few added thoughts related to the above conversation and my thoughts on how a pulse induction detector works.

The coil is not a tuned circuit like a radio receiver antenna coil which has a variable capacitor to select different radio Channels/Stations. The narrow bandwidth of a tuned radio antenna circuit eliminates all broadband EMI frequencies that are above or below the tuned in station frequency. If the radio is tuned to 1590000 Hz at the top of the AM radio band it will allow a continuous flow of sine like waves to pass through to be converted into voice and music.  If the station is at a long distance only EMI that is within the narrow bandwidth of the tuned circuit will cause static and distortion of the station’s signal.

A pulse induction detector is not a radio frequency transmitter and receiver. It is a timed device that gates the receiver open at a precise period at the end of each transmit pulse. All EMI and noise that occurs when the receiver is off is eliminated. Unfortunately there is still a lot of EMI and noise that is also gated into the receiver during the gated period. The transmitter induces a high energy collapsing magnetic field into the earth. At the tail end of the collapsed field the receiver is gated on to allow measurement of a feeble distortion of normal ground return and background noise signals caused by a small target’s magnetic signal.  The detector coil is a broadband circuit allowing many Channels/Stations/EMI in simultaneously. Each time it gates the receiver on it is trying to detect a slight magnetic induced change in one single received slope which is the equivalent of one of the 1590000 sine waves of the AM radio station signal. 

The frequency control on a pulse induction detector changes the number (hundreds to a few thousands) of transmit pulses transmitted each second. The pulses and target return signals are not tunable sine waves. The transmit pulse is a very high current change in energy switched into the coil. The number of energy changes, grouping, widths, energy and spacing varies with different modes. The automatic frequency (pulse rate) control tries to select a pulse rate so that the very weak received signal slope is not synchronized with predominate EMI rates. If a 50 or 60 cycle power line is nearby it will choose a pulse frequency that does not synchronize with that frequency or any strong harmonics of the primary frequency. It cannot handle all of the random power line EMI that comes from computers, TV distant ARC welders and other heavy changing load loads on the power grid. Pulses and harmonics from other metal detectors will also be avoided by the automatic frequency (pulse rate) control. Random EMI noise/pulses from distant electrical storms, aircraft and other sources that vary in repetition rates and signal strengths will fade in and out and require more frequent automatic frequency (pulse rate) actions.

Transmit coil specifications have not changed for many years. Most coils are wound with a special Litz wire and have a low coil resistance of less than one ohm. The inductance is typically around 290 to 310 microhenrys. The coils are normally shielded for EMI and ground effects with conductive graphite paint. The conductive coating must have considerable resistance to avoid shorting out or severely damping or overloading transmit and receive signals. Any external add-on shielding or metal material will severely alter the performance of the coil. Smaller nuggets will not be detected. As more shielding is added to a coil an overload condition will be reached and hopefully the detector has an automatic shutdown feature that prevents internal damage.

Automatic ground tracking/balance is continuously adjusting the receiver section to compensate for different ground conditions. Some mineralize and/or damp salty ground will tend to bias the ground balance. In these conditions the detector may slowly become desensitized and quiet. And then it goes over a hot rock or tin can and suddenly wakes up to return to a more sensitive and noisy normal condition.

Internal broadband noise and EMI can severely desensitize the detector receiver. Suppressing or shielding internally generated broadband noise is essential. Construction and layout of the circuity, multiple layered printed circuit boards with a continuous copper ground plane sandwiched into the board are some of the methods used. Shielding of critical circuits is most important. The GPX 6000 and GPZ 7000 detectors have a black epoxy covering the entire receiver module. This epoxy may have iron powder mixed into it to provide a very effective shield. Overall sensitivity is determined by a good signal to noise ratio at the first stage of a receiver. It takes very little input noise to desensitize the overall system.

It is a difficult task but some of the manufactures have overcome a lot and provided some very fine detectors. The current problems will be improved upon in future models.

Have a good day,
Chet
 

[jasong]

Good to see you around posting again Chet.

Curious what you think about where Woody is able to throw a solid copper fabric over the GPZ coil to cut out noise in his house during testing, yet is still able to detect targets through the copper mesh without much decrease in sensitivity? Did you see that part of his video? That must mean the GPZ works differently enough that you can use a more conductive shielding? It didn't work when he tried it with the 6000 though.

I'm wondering about it because if a solid copper fabric on top of a GPZ coil works, then should we be able to make coils for the GPZ out of carbon fiber (which is conductive)? At least, on the tops and sides, if not the entire coil. That may lighten and simplify the GPZ type coils if so. It may also be a magnitude better/more effective shielding than high resistance graphite paint is.

I ordered a sheet of carbon fiber to test with out of curiosity, but the mail has been crazy unreliable into Wyoming due to all the snow/ice road closures and this one seems to have gotten lost in transit, so waiting for a replacement.

[Chet]

Jason
    Conductive shielding such as copper mesh can be used. Some of the early BFO and VLF detectors used aluminum foil or copper mesh. One even used a thin wall copper pipe with the windings inside. There must a gap such as a very thin slice of pie or slot removed in the shielding somewhere within the 360 degree circle. The gap prevents the shield from becoming a shorted turn. The entire coil and shielding must encapsulated in epoxy or foam to prevent any movement between the coil and the shield when bumped against anything. In the videos you can see how the slightest movement of the copper mesh causes problems. 
     Since those good old days lol by rule of thumb it was determined that any shielding material that could be detected when passed over the coil should not be used. It was found that a graphite coating with a resistance of 50 to 150 ohms worked well for shielding.
     The GPZ 7000 14" coil has a layer of graphite coated paper on top and bottom of the coil assembly with overlapping fingers around the circumference.  There is a thin bare wire taped to the top and bottom shields that is connected by a blue wire to the printed circuit board ground inside the detector housing. Shields should always be connected to a common circuit ground or ground plane.
      The GPX 6000 should react to the copper mesh similar to the GPZ 7000 if the mesh distance and placement is the same. There will be differences caused by coil types (mono, DD, DOD) and size (11”, 14”). The copper mesh will be a load/damper to both detectors. The distance and amount within view of the coil will determine how much the coil will be loaded down. This is why he is spacing the mesh away from the coil with large plastic boxes. If he wrapped the mesh closer to the coil it would severely desensitize the performance and at some closer distance cause an overload and shut down the detector. 
        Test your carbon sheet for resistance at several places with the probes 1”, 6” and 12” apart. Pass it over the coil to see if it is detected. If the resistance is consistent in the range of 50 to 150 ohms and the sheet is not detected then it should work.
Have a good day,
Chet 
 

[jasong]

Chet, what I'm wondering though is if lower resistance shielding might work on the GPZ since that copper was working for him. And if so, if lower resistance shielding might actually be more effective at cutting more EMI. Woody said it worked on the GPZ but not the 6000. 

He appears to have discovered the 6000 EMI/speaker issue too now. He hasn't quite made the connection yet though. I told him about the fix, I'd be curious to see what he thinks about it and if he can improve it more. 

One thing I'm curious also to try is I'd like to cover my 6000 box with carbon fiber, and see if it lets me carry my phone next to it just as an experiment.