RELIC HAWK

[Max]

Hi Max,

Yes we can say for sure that it goes to pin 1 of 393, so there was a mistake when shematic was traced pin 5 and 6 were mixed, wich is very easy to happen.

Nice team work, an congrats to Max who found this before.
In fact the "problem" with ferrite looks like a proof that it really works,thanks to the manual.
Regards,
Fred,still waiting for parts

Hi Fred,
well, yes I like all this team working here... very nice and productive!

Now I have to focus about solving the geb stranges... using a better... 10turns pot... and see what happens... if I can null with it on ferrite we'll be sure GEB works also in all-metal, so that it works REALLY.

Also the other problem is at disc phase shift... I see different on scope vs simulation... must understand why.

Kind regards,
Max

[Fred]

Hi Max,

Yes we can say for sure that it goes to pin 1 of 393, so there was a mistake when shematic was traced pin 5 and 6 were mixed, wich is very easy to happen.

Nice team work, an congrats to Max who found this before.
In fact the "problem" with ferrite looks like a proof that it really works,thanks to the manual.
Regards,
Fred,still waiting for parts

[KingJL]

I checked your simulation and seems all fit about geb ... about in expected ranges, but disc I don't understand...I read from 124 to 248° as start at one geb extreme...I cannot see that in simulation.

Check both ends of R5 in the asc schematic. That is equivalent to min/max (or max/min) of disc. Actually it is pretty slick ,the way they arrive at DISC delay.

Is there maybe some mistake in disc section ?

I don't think so. Remember there is really a 180 degree phase shift from the signals at R5 vs GEB as seen by the phase detector (GEB to inverting, DISC to non-inverting). I think that results at ~(-42 to -314) /~(-67 to -297). The 2 ranges on each side of '/' are for the 2 extreme settings of GEB.

Regards,
J. L. King

[Max]

Hi,
I think the mistery is solved now.
The "f" goes to pin5 of an LM393, then the pin6 ...where's inverting input goes to a 18K resistor: it's GEB control line, cause just GEB comparator take input at that 18K = out pin of previous op amp (U6 on our schematic).

So Minelab bind "f" line at GEB control signal, not on DISC as we have in our schematic.

Do you agree ?

If ok and no questions I will put in next revision.

Kind regards,
Max

[nakky]

Max,look at this.

[Max]

Eclipse,can you make two better pictures from the total board,top and bottum side or can you trace the F line ?It is hard to do with this pictures.Grt Nakky.

Hi Nakky,
I think you found it.

I was following it... just under the smd resistor there and found your post that let me think you found it already.

Now we have to consider the 2nd comparator inside there... and look about where signal came from.

I mean... we have to look for disc or geb related control signal.

Kind regards,
Max

[Max]

I belive this design is meant to have the coil slightly off of null. This is forced by the feed of the TX through the 4 Meg resistor. I believe (but not sure) that the RX coil is supposed to be nulled by viewing the output of the preamp. This would slightly bias the RX coil signal to provide a better S/N from the coil (never let it really be nulled to 0). This may ever so slightly also increase the area of the DD overlap, increasing the width RX field.

Regards,
J. L. King

Hi JL,
probably you're right... I think the same.

Now my readings about signals from the circuit running at 5Khz, considering TX as reference.

With geb knob fully counter-clock-wise GEB control gate pulse (positive) start after 79.2° from TX start. It ends at about 260.4°.
With geb knob full clock-wise I get pulse (positive) start at 97.2° and ends at 277.2°.

About disc, with knob fully CCW I get 124.4° as start and 303.6° at end.
With knob fully CW I get 248° at start and end is at 428°. This at one extreme of geb pot... is different from simulation ?

I checked your simulation and seems all fit about geb ... about in expected ranges, but disc I don't understand...I read from 124 to 248° as start at one geb extreme...I cannot see that in simulation.

Is there maybe some mistake in disc section ?

Kind regards,
Max

[nakky]

Eclipse,can you make two better pictures from the total board,top and bottum side or can you trace the F line ?It is hard to do with this pictures.Grt Nakky.

[KingJL]

Hi,
I see device is not so critical about nulled voltage... could be also some volts at output of preamp... and it still work (on the bench) like when there's 100mV there but I don't know about exact phase shift required.

I belive this design is meant to have the coil slightly off of null. This is forced by the feed of the TX through the 4 Meg resistor. I believe (but not sure) that the RX coil is supposed to be nulled by viewing the output of the preamp. This would slightly bias the RX coil signal to provide a better S/N from the coil (never let it really be nulled to 0). This may ever so slightly also increase the area of the DD overlap, increasing the width RX field.

Regards,
J. L. King

[Max]

Hi,
that's what I'm looking for... in page 18 of manual you can read this...

so it was not my imagination: in all-metal they know was really hard to compensate for GEB!

I'll do like explained... lowering sens... but I see it hard to find the spot before mounting some real multiturn stuff...I have a common pot there at now...

Kind regards,
Max

[Max]

Here is zip containing the asc file and screenshots of run with GB set at 10K and 0.1 ohm.

Regards,
J. L. King

Hi JL,
many thanks... this save me lot of work. I wanna check for mistakes on timing... but so far I haven't spotted anything bad.

I will compare simulations with read values and let all you know.

Kind regards,
Max

[Max]

Hi Max,I read in the manual that the manual gb trimpot must by 10 turns.Grt Nakky.

Hi nakky,
yes, right... I suspected that but someone told me before it was multiturn stuff.

As I said... probably less than 1° difference and GEB will not work at all... I see that on my clone... change to phase is so critical that anything apart a multiturn pot will solve that when on soil.

Kind regards,
Max

[Max]

Max don't forget that this is musketeer, and RH should have some differences (i think), but overall is no less than 95.99 % equal schematic. Probably the bigger coil gives extra boost in depth and also the musket misses the LSM switch.
I sold the device, i would be more helpful if i had it but ...
There was a bidding on ML RH on ebay few days ago but i didnt win cos i had a vacation and was away from the internet
the bid was too low.

Hi,
dang... you miss the RH on ebay! I saw the last offer was low for the value of machine. You lose that bargain... anyway... we are doing good here... probably you'll get enough informations to build one at the end.

Yes, they are not the same stuff... I think the 910K resistors are employed to lower the gain of some particular stages that have the bad tendency to saturate easy. But probably more than 95% is the same schematic... I see many similar pieces in the picture.

The brown smt with no labels are the decoupling caps 100nF at op amps.

Thanks for posting that pictures... everything is useful.

Kind regards,
Max

[Max]

Relic Hawk Manual (English) (2822 KB):

http://rapidshare.com/files/127811179/RelicHawkManualEng.pdf.html

One of the few manuals not available on Minelab's website. Finally got around to scanning it in. No, I don't have a Relic Hawk Metal Detector but I do have the manual.

Rufus

Hi Rufus,
very good ! Now we got the manual...

It's important about GEB tuning and the like!

Thanks,
Max

[KingJL]

Relic Hawk Manual (English) (2822 KB):

http://rapidshare.com/files/127811179/RelicHawkManualEng.pdf.html

One of the few manuals not available on Minelab's website. Finally got around to scanning it in. No, I don't have a Relic Hawk Metal Detector but I do have the manual.

Rufus

Rufus,
Many, many, many thanks!!