Garrett Pro Pointer problem

[Eddy71]

Is there any dried up water on the board?
(Sometimes water damages the board underneath the SMD parts)

[Carl-NC]

As Skippy said, it looks like a Chinese fake. They are notorious for poor performance and poor reliability.

[nalag46]

As Skippy said, it looks like a Chinese fake. They are notorious for poor performance and poor reliability.

This isn't a Chinese fake one. You are confusing it with a different one of a previous post, which indeed was a fake of the standard black pointer model.
The one here is a genuine Garrett AT-Pro.

[Carl-NC]

In that case I don't know. Here is a block diagram of the Garrett:


The oscillator runs at 12kHz or so, the ADC (which is inside the micro) reads the peak detector and software does a turn-on calibration for a threshold and then looks to see when the peak detector falls below the threshold. So either the oscillator isn't running properly, the peak detector is wonky, or something is wrong in the micro. I would suspect the oscillator and that requires an oscope,

[nalag46]

In that case I don't know. Here is a block diagram of the Garrett:


The oscillator runs at 12kHz or so, the ADC (which is inside the micro) reads the peak detector and software does a turn-on calibration for a threshold and then looks to see when the peak detector falls below the threshold. So either the oscillator isn't running properly, the peak detector is wonky, or something is wrong in the micro. I would suspect the oscillator and that requires an oscope,

Thanks for that Carl. It is relatively a small circuit but if the problem is within the oscillator I am not able to test it at the moment as my oscilloscope has gone faulty and needs to be repaired. I still believe it is the coil because when I connected a good working one (the one from a full working pointer I borrowed), it worked up to 90/95% on this board and that says it all, really.

[David_1]

when I connected a good working one (the one from a full working pointer I borrowed), it worked .

did you measure inductance and resistance of working coil ?

[nalag46]

did you measure inductance and resistance of working coil ?

I have already posted that information before. On the good working probe, the inductance was 1.225mh. I did not check the resistance but
I believe it is in the 1 to 2 ohms range, but I may be wrong.

[David_1]

I have already posted that information before. On the good working probe, the inductance was 1.225mh. I did not check the resistance but
I believe it is in the 1 to 2 ohms range, but I may be wrong.

I only asked because if the coil is bad, you have the information to make a copy ?

[nalag46]

I only asked because if the coil is bad, you have the information to make a copy ?

All done. Rewound the coil neatly on the ferrite rod, got the inductance as near as it should be and now it has three distinctive levels of sensitivity
to play with.
A slow and patient job but well worth the effort.

[David_1]

so can you share the information of what you did ? pictures ?

[nalag46]

so can you share the information about what you did ? pictures ?

I didn't take any pictures. All I did was unwind ten turns of one of the coil's ends, and (by estimating only) I added a 10 cm length of the exact same gauge cable (roughly four more tight turns) and this got the inductance to the correct value. The windings on the ferrite coil are sealed with tape and there are two thin rubber straps to secure it inside the pointer's tip (this has to be hard pushed tight and carefully with a blunt thin probe, like a long screwdriver?) and all is working perfectly well. I cannot take it out again as it is difficult and by pulling it from the board end too much will easily damage the windings.

[David_1]

I cannot take it out again as it is difficult and by pulling it from the board end too much will easily damage the windings.

LOL no dont break it but thanks for the description of what you did.

[Sean_Goddard]

The micro is a KEA64 by NXP (Philips) with the 2N22J mask program.

It appears from a little research, that this mask has a flaw which allows the program to be read (off you go then and have fun)

https://www.google.com/search?q=8160900+microcontroller&sca_esv=579528709 &sxsrf=AM9HkKmjtz8SpEaLu1YFXC6iNxNZSLoN8w%3A169914 6271216&ei=H-pGZbTwDIPBhbIPwuGL2Ag&ved=0ahUKEwj0-6qP1auCAxWDYEEAHcLwAosQ4dUDCBA&uact=5&oq=8160900+m icrocontroller&gs_lp=Egxnd3Mtd2l6LXNlcnAiFzgxNjA5M DAgbWljcm9jb250cm9sbGVyMgUQABiiBDIFEAAYogQyBRAAGKI ESLMrUNQJWNElcAF4AJABAJgBTaABiAiqAQIxNrgBA8gBAPgBA cICCBAAGKIEGLADwgIEECEYCuIDBBgBIEGIBgGQBgQ&sclient =gws-wiz-serp#vhid=tZ5BQK6vTXEWAM&vssid=l

https://www.nxp.com/search?keyword=2N22J&start=0

https://www.nxp.com/docs/en/data-sheet/S9KEA64P64M20SF0.pdf

Nice try garrett using something obscure lolz