Minelab GPZ 7000

[greylourie]

So far I've seen feedback loops to compensate for the TX drive tilt. But mineralized ground alters the loop inductance and changes the overall drive level. Minelab uses a feedback loop to detect this effect and maintain a constant drive level. That's what their patent is about.

Hello Carl,

What is the difference between constant current and constant drive level ?
There was a thread over here, that fizzled out : http://www.geotech1.com/forums/showt...urrent-control
I get confused with everything.

[greylourie]

And here I had shown how to make a square wave with a constant current intervals -

http://www.geotech1.com/forums/showthread.php?20140-Square-wave-Pulse-Induction

And you see , that circuit has a regulated energy loss compensation vith a simple voltage control ( kick circuit ) , so it can set the current line horizontal ( make a perfect constant current ) either in manual or in automatic modes , it's not a problem at all ...

Deemon, does your system automatically compensate for changes in inductance caused by ground changes ?

[Carl-NC]

So far I've seen feedback loops to compensate for the TX drive tilt. But mineralized ground alters the loop inductance and changes the overall drive level. Minelab uses a feedback loop to detect this effect and maintain a constant drive level. That's what their patent is about.

[greylourie]

Here is a good post from PiTec. It has a lot of links.

http://www.geotech1.com/forums/showthread.php?20038-Triangular-Wave-Technology&p=164732#post164732

[Davor]

... and PiTec of this forum under a different name had designed an unipolar constant current drive with simple feedback control around 2002 at PI forum, and it can be seen at findmall forum even today. We also played with variants of that circuit on this forum, I'll post links when I find them. And yes, it is obvious. We already know what constant current drive does.

Now, maybe you do not have to list the forum references, but the fact is that the "obvious" category is formed by the opinion of the technicians versed in the art, and those tend to frequent the technical fora, such as this one. E.g. if you ask about the new GPZ coil, many people will instantly recognise the design mikebg promoted here.

As for using non-patent references, the manufacturer-who-must-not-be-named had no problem bashing the Corbyn 1980 Wireless World article in a US5576624 patent as they found it fit, and some, and for a reason of justifying their meaningless Tx timing scheme backed up by silly math. Now the paradigm has shifted, and the best timing scheme is the one without frills with more of the same math, mostly disfigured and not referenced Das 2004 paper ones. Go figure.

IMHO it is up to the copycats to insist on manufacturer-who-must-not-be-named timing schemes to get caught red-handed, and not so much about this being better than that. A kind of fly glue for the half-witted. Just about anyone interested in similar solutions may find references from before the priority date.

[deemon]

Minelab has previously implemented a TX feedback methods, and other people have previously implemented constant current transmitters, so whether this is innovative or an "obvious step" is debatable. I know a few people who will insist it's obvious and non-inventive, and they are welcomed to show prior art of a constant-current transmitter with a feedback loop to compensate for ground effects.

As for me , I implemented even 2 feedback loops in one transmitter , to maintain the perfect constant current interval -

http://www.geotech1.com/forums/showthread.php?19273-PI-metal-detector-with-energy-recuperation

And here I had shown how to make a square wave with a constant current intervals -

http://www.geotech1.com/forums/showthread.php?20140-Square-wave-Pulse-Induction

And you see , that circuit has a regulated energy loss compensation vith a simple voltage control ( kick circuit ) , so it can set the current line horizontal ( make a perfect constant current ) either in manual or in automatic modes , it's not a problem at all ...

[Carl-NC]

The US patent office is overloaded and undermanned, so most patents don't get the scrutiny they need, and most examiners don't have the experience to know what to scrutinize. And, no, searching obscure forums for past discussions on similar methods is not part of their prior art research.

I haven't read the patent in detail, but Claim 1 (the single major independent claim) focuses on a feedback loop to maintain proper coil drive in the presence of severe mineralization. The rest of the claims encompass how you could use this in a square wave current transmitter. Minelab has previously implemented a TX feedback methods, and other people have previously implemented constant current transmitters, so whether this is innovative or an "obvious step" is debatable. I know a few people who will insist it's obvious and non-inventive, and they are welcomed to show prior art of a constant-current transmitter with a feedback loop to compensate for ground effects.

[Davor]

Against my better judgement and justified grudge against the manufacturer-who-must-not-be-named, this patent describes a workable device that makes sense. However, most of the solutions inside were discussed on technical fora for the past 15 years or so, others are public domain, and I can't seem to find any of those mentioned in prior art section. I've seen crazy things patented in OZ, including the wheel (!!!), but this is an US patent, and it means that US also succumb to the immense pressure of bull****.

Besides, the actual detector is perhaps 1/20 of all the gremlins they crammed inside. It could as well continue running in simulation mode for weeks before anyone realises that the detector part is broken.

This may be translated in 3 ways.
1) being aware of the prior art, both in patents and fora, you may build a machine that resembles the action of this one with virtually no consequences,
2) your machine from "1)" will be better simply because of less gee-gahs and doohickeys,
3) you may build an even better machine by following your own recipe, and leave the claims of GPZ to the marketing la la land where it belongs.

In case you as a hobbyist pick the path "3)" - try hard not to get it taken away from you by the manufacturer-who-must-not-be-named, as it happened with bugwhiskers' qed, by accounts of people who actually tested it in real nature - a far superior machine.

I was positively shocked when I first saw the patent and a complete absence of ridiculous timing schemes so typical of the manufacturer-who-must-not-be-named. Maybe this is a start of a new era for them, but unless a radical change happens in their attitude against the hobbyists, I'll continue to see only their shortcomings.

[Ronk]

god help anyone trying to enforce this . patent it just shows if you put enough bs together it might be taken seriously

[Qiaozhi]

Here's a link to the patent that atd tried to post.
atd - When the text is extremely long, please either provide a link, or attach the file for downloading.

http://minelabgpz7000.com/wp-content/uploads/2015/02/GPZ-7000-patent.pdf

[greylourie]

Maybe reading these posts would be easier.

Perhaps citing pat number would be enough. Above post, with unusual text arrangement makes me go cross-eyed.

Early posts in Findmalls Technology section have some info to help understand what is going on. And offer inspiration for experimenters.

I found some posts by Dave Johnson, whilst trawling yesterday.

http://www.findmall.com/read.php?34,129490 power-efficient PI .....Invention?
http://www.findmall.com/read.php?34,129533 Unipolar CCPI
http://www.findmall.com/read.php?34,129535 Low-conductivity targets and CCPI
http://www.findmall.com/read.php?34,129571 Dave J- Question on CCPI

[Ronk]

when Bruce candy shows the magnet rotating he is simulating his coil in action so he is reversing the fields between samples and presumably adding them or is he just pulsing as std and reading the 2 outside coils in reverse of each other ?

[greylourie]

I wonder how they hold that bundle together ? Must be bonded somehow ?
I have seen posts by Stefan (may have spelling wrong), he gives a lot of tips for building fast low capacitance type coils. And in one of his posts he describes this square/recatangular profile, or rather, describes winding a layer, followed by insulation material, repeated until the bundle contains the requisite number of total turns.

That screenshot of their engineering example doesnt appear to have inter-layer insulation. Still a very interesting image to look at. There is a video tour of Garretts workshops online. They also use similar black plastic bundle spools in some of their coils. Guess this offers consistency and repeatability.

[golfnut]

thats a serious wad of litz in the middle - probably not resonant Tx

[deemon]

And square waves with constant current intervals , as I proposed here too ...