LOOKING FOR GOLDSCAN 4 Schematic.

[kadir6141]

mr. İvconic I am used whites dual field coil and there was no change in sensitivity or depth. dual field coil 300uh and 2.5R

[eduardo1979]

Clearly.
I didn't even expect that proper damping will solve the "depth" problem.
I just wanted to check how much difference there is between the provisional value of 470R that I randomly chose (because is 5W) and the real value that I will get with the "jig".
As time goes on, and as my experiments pile up; I start to suspect more and more that there is something else wrong with the pcb design.
I have already found 5-6 errors which I corrected and posted here on one of the pages.
But it is possible that there is another one. I don't know who drew it, but it definitely raises suspicions.
I hate wasting time and looking further for mistakes. I have another pcb design that is (supposedly) 101% verified.
And I plan to make a detector on that pcb.
But even this pcb with errors is good enough for exercises and experiments.
All the time the serial 2R2 is on the pcb. Someone suggested that I remove it, but I don't think it will significantly affect the problem.
Someone else made a comment about the J112 and how that fet "does not work good".
And then someone else says their detector works great with the BF245.
I'm almost certain that has nothing to do with the problem. The J112 will work perfectly fine there as well as the J113.
It's just... everything mentioned so far doesn't make sense. And it's not true reason for the problem.
Barracuda has a range on the same targets 5 times even more!
Of course, with the Barracuda there are no two channels and no channel mixing due to GEB.
And this GS4 pcb, in "Off" mode, has noticeably better "depths"... but not half as much as the Barracuda.
It's rude to say it... but this whole thing about the GS4 is one big horseschmitt!
But I'm satisfied, I had very extensive exercises, experiments and I "returned" to the topic a lot.
...
Carl you didn't suggest damper values for the given coils?
Nor did you choose the moment on the video that you think is the best damping?

Hello Ivconic.
I attach the functional diagram of the GS-4 that I made many years ago. The PCB has no errors.
Enjoy.

[ivconic]

Hello mr. İvconic
Did you use lm709 or Ne5534? What did you set their output voltage to?

There is DIL socket on the pcb and I tested with both, pretty much the same outcome.
Of course I set the offset voltage to 0 volts.

[ivconic]

Hello Ivconic.
I attach the functional diagram of the GS-4 that I made many years ago. The PCB has no errors.
Enjoy.

Thanks, I will also order that pcb at JLPcb in the next "iteration"... I have another one that is already finished, I will do the detector with it first.
But I like your version better because of the better audio.

[ivconic]

mr. İvconic I am used whites dual field coil and there was no change in sensitivity or depth. dual field coil 300uh and 2.5R

What I especially like about the GS4 is the great tolerance toward coils.
I took a discarded White's "something Max" VLF I/B coil and took out only the TX part, unwound part of the coil and got 0.3mH and some R... (I have it on the wall, I hate measuring it now).
TX is wound with Alu wire!
It works almost the same as the others.

[ivconic]

The concept of channel mixing (subtraction) like this is very bad!
The GEB works well, but such a detector with the GEB set is in no way separated from similar VLF I/B detectors around the main drawback they have.
Because GEB "rejects" the influence of soil and mineralization successfully... but at the same time it also rejects all good signals in a certain range.
I will shorten the story to a simple example that each of you can check.
Take a good coin or a larger piece of gold, or a smaller piece; bring it close to the coil and detection will occur.
And then bring a coil nail, some iron, similar... and the detection will decrease and stop, the detector will be silent!
Identical situation as with many VLF I/Bs that suffer terribly from "masking" problems. They are unusable on real terrain.
So... miss!
It is for these reasons that I came up with the idea of IGSL once upon a time.
So why am I wasting my time here making several variants of GS?
The idea is to first completely return to the problem, to recall all the details (it was forgotten, I haven't dealt with PI detectors in this way for years).
And my age and memory gaps have caused me to easily forget even things I've personally done.
The next step is to modify GS as a platform to overcome this problem.
Much the same way it was done at IGSL.
4 maybe 5 channels, each giving a separate audio(s) and thus covering all possible holes.
It's a long journey there. But I'm not in a hurry and I enjoy journey the most.
We'll see... a lot still needs to be done and verified in practice.
I have long since given up on the faster and easier way with the help of a simulator.
Because simulators are not good enough. As far as I'm concerned, they have their role only in academic thinking.
I regret now that I didn't check this with the couple of ML PI machines I've had so far. It just didn't occur to me.
I wish everyone else who is engaged in high-minded thinking about a better PI; instead of one target they try to check the detection of their works on mixed targets.
Coin next to iron etc.
The condition is that the detector has the GEB set to a very restrictive level, as if working on highly mineralized soil.
The point is; subtracting the GEB channel from the signal leads directly to "masking" the good target.
That is, if the PI is set to better conductors; it will remain totally mute if a worse conductor is placed next to a better conductor.

[Jose]

I suggest that you check the operation of the autozero circuit, which may be the cause of the low sensitivity.
I make the adjustments with the reset activated and the coil disconnected.
In normal operation, the voltage at terminal 8 should be approximately -4.2 volts so that the fets do not conduct and the gain is maximum.
The 220k resistor on terminal 8, I replaced it with 22k and it works better.
The Gs4 in normal operation behaves the same as other PI detectors in terms of ambient noise and instability.

[daluka]

Ivconic
Did you connect this correctly? 2N3906 = 1E - 2B - 3C

[ivconic]

I suggest that you check the operation of the autozero circuit, which may be the cause of the low sensitivity.
I make the adjustments with the reset activated and the coil disconnected.
In normal operation, the voltage at terminal 8 should be approximately -4.2 volts so that the fets do not conduct and the gain is maximum.
The 220k resistor on terminal 8, I replaced it with 22k and it works better.
The Gs4 in normal operation behaves the same as other PI detectors in terms of ambient noise and instability.

Done that at the very first attempt, several weeks ago. I followed precise procedure of adjusting all the trimers properly.
Also replaced 220k with 22k, haven't noticed any improvements.

[ivconic]

Ivconic
Did you connect this correctly? 2N3906 = 1E - 2B - 3C

Of course.
It is basic thing.
I don't make mistakes on that level...

[daluka]

[daluka]

[ivconic]

...

[ivconic]

Of course.
It is basic thing.
I don't make mistakes on that level...

YES I DO !!!

[ivconic]

Ok, let me correct that and see what happens!