LOOKING FOR GOLDSCAN 4 Schematic.

[Thomas]

I did come across GOLDSCAN 4 SCHEMATIC on this forum a while back. Does any one know where it is located.
Thanks

[Esteban]

Re: LOOKING FOR GOLDSCAN 4 Schematic.

See the two keys:


http://www.thunting.com/cgi-bin/geotech/forum/view.cgi?action=message&forum=tech&message=2813


http://www.thunting.com/cgi-bin/geotech/forum/view.cgi?action=message&forum=tech&message=2835

[Hector L. Bombino]

Re: LOOKING FOR GOLDSCAN 4 Schematic.

Is there a board available from someone for this detector? Alex perhaps? Would like to know

Thanks

[kt315]

respect

[Mrand]

Schematic contain error. In tunning circuit, drains of J113 must be connected to GND.
Mrand

[Chris2]

Hi Mrand,

which J113 needs to be connected to GND?

[Unregistered]

I guess the Gnd is comun to all 10K resistors connected to J113 fets

[Mrand]

n/t

[kt315]

power supply

[kt315]

I re-carry a text of Eric from PI Forum. I hope it will be interesting to all.

Goldscans etc.
Eric Foster
Hi all,
Here is the post I made on the Finders.co.au forum. I guess some of the content will be of interest here.
The Goldscan actually started as a gold prospecting detector after a visit to Australia in 1982. So long ago "> It was evident from tests that I did that the ground mineralisation was very much stronger than anything I had encountered elsewhere, and that a straight PI had almost no hope of working. I then devised a multiple sampling and filtering system to get rid of the ground signal. First there was Goldscan 1 in 1983, which was a completely shaft mounted unit. Just a few of these were made and tried in the more mineralised areas of the UK, such as in parts of Scotland. From the production point of view it was easier to make a chest mounted control unit with built in rechargeable battery pack. This also had the advantage that the coil and shaft assembly could be made very light and comfortable. The Goldscan 2 was produced from 1984 and continued until 1992. By further miniaturising the electronics and having a belt mounted battery pack, it again became feasible to have a lightweight, shaft mounted control unit and the Goldscan 4 was born, which was latterly made by Pulse Technology. What about Goldscan 3?. This was a high power unit running from built in 24V lead acid gel batteries. It certainly had performance, but only a dedicated and rugged few braved its weight and the heavy digging tools required for retrieving targets.
All of these detectors had manual ground balance. An auto-ground balance version was designed on paper but never got to production. No real magic to auto-GB; just a feedback system to correct changes in ground mineralisation. Actually, with manual GB, little change in setting is required when going from one area to another, provided the ground signal stays within the linear range of the receiver. Hot rocks have very high concentrations of iron minerals and, particularly when using a mono coil, can cause temporary overloading. This distorts the decay curve and the signal then falls outside the pattern required for the GB filter to work. Hence a false response. This is where the Goldscans fell down in Australia, but having moved on to other detector developments, mainly industrial and underwater, I did not pursue it further. There are various ways of preventing receiver overload, but one good way is the use of a DD coil. SD users will be aware that a DD is quieter on highly mineralised ground than a mono. This is because a DD has a natural cancellation of near surface mineralisation signals, due to its balanced geometry. A mono coil, on the other hand will give a very strong response to mineralised ground due to the same coil being used for both functions. A hot rock passing under the coil edge will give a very strong signal due to the tight TX-rock-RX coupling.

[kt315]

[Cadophobic]

timing error

I had been planning to put together a MD, and I worked on the Magnum IB for a long time - but I am taking (Darth Carl's ?) advice to do a PI first, mainly because a GOOD coil for an IB is a daunting endeavor. Sooo I like this particular project because I want to find gold. But it looks to me like the reverse-engineered timer circuitry has problems.

For one, a CD4538 device needs grounds on pins 1 and 15. CD4538 timer grounds are not supplied internally , unlike their TTL counterpart 74x123?). ( For that matter, I don't know why those pins arent named GndA and GndB instead of the usual CxA and CxB, or something-or-other.) I may be wrong on this point, haven't breadboarded, and in fact similation using (the only model I have available) (and) LTspice seems to work whether or not I add an explicit ground on those two pins!
Of course the model I use is only a spice MODEL and until any circuit is hard-wired (and tested) there can be no knowing for sure.

On another note, look at -any- CD4538 datasheet. You can verify that so long as the +in pin of the device is tied to the Q output while it is in a positive state, the output logic output level can never change, no matter how many times the (-) input is toggled. !! The TC4538 #C, and TC4538#D " are wired in that fashion, according to the schematic. That must be incorrect.

Anybody buying into this conundrum?

download SwitcherCad (aka) LTspice from http://www.linear.com/company/software.jsp

I've had enough for today..
bye

[Cadophobic]

circuit update

If all goes well there are graphics of the timing circuit as I have developed through process of elimination. I have no access to the original Goldscan_IV except through what's provided by previous posts. I am open to criticism or advice on the accuracy of this post.

[Unregistered]

Gldscan schematic

Hello forum, I am who make the reverse schematic of the Goldscan many years ago , don' t waste your time to tell us it's not working because the complete seccion of timing is 100% OKKKK, for you it is much better to take your solder pen with fumes of flux (better that playing with spice) and realize the complete Goldscan4 and you would see for yourself it work perfectly....please do not spread doubt for the people who want to construct the GS4, it is a very, very good pulse detector with a beautiful engineering design (thanks a gran part to Eric)...I have realize at this time many version of GS with different variation design (audio, autotrack,etc...) but the main seccion for exemple the timing are still not changed ....Very soon I will have a web site with all the infos about and the printed circuit in semi-kit ...
have a good time and goods constructives researchs all guys of the forum.
I do not write many time in the forum because in this moment i am very busy in working.
Alexis.

[Cadophobic]

dual channel timing circuit, CD4538

I would love sucking in some solder fumes, lead and all, but I don't have a facility to hard-wire circuitry as much as I'd like to. But there's the beauty of SPICE, it helps prevent our falling victim to subtle errors of logic - which brings me to my point: I stand by the relevance and accuracy of my previous post.

In the reverse engineered schematic drawing of the Goldscan, there appear to be errors in the timing circuit. Refer to the attached image. I won't say exactly what CD4538 datasheet that I clipped, because Fairchild might come after me. I already have enough legal problems without begging for more. But it should be obvious that, if CD4538 pins 4 and 6 are connected, after the Q output (pin 6) assumes a logic one state, it will stay there. My SPICE model may not be entirely accurate, but please do not expect me to ignore the datasheet.

Toggle the input on pin 5 until the cows come home, but Q should not change state because, having put logic 1 on the or gate positive logic input, the output will continually be forced high. A latch-up happens. I don't even have to melt solder to know that. It's a natural fact. If not, then I am in the wrong line of work (which could explain why I don't have a job!) I do not have a Goldscan IV, and even if I did, I doubt that I would have the energy or the resolve to extract a schematic by analyzing the circuit board. And so, I credit you for your accomplishment by reverse engineering the schematic.

However, can the timing circuit, as it is drawn, be logically rationalized? No. When you built your functioning units, did you go by the schematic drawing as presented here?? I wish the rest of the circuitry was as simple to analyse as the timer, because I am also having a problem with the sample and hold. It's principle has thus far escaped me. I am not a SPICE expert, nor an engineer. Just a wannabe, and thus far I can't rationalize the S/H, although I will continue to try.

By the way, I grabbed the CD4538 spice model from LTspice user's group at http://groups.yahoo.com/group/LTspice.
I made a minor modification to the available part drawing, but in no way did I modify the model file. I am not so brave.

Anyway, have a nice day, Y'all. You don't have to believe the rantings of a fifty-something year old wannabe, but you should believe the datasheet.