XP deus coil modification

[Henrikras]

Thanks Skippy.... You just explain it Very good.
I would make a list to Remember all this, and start the work this weekend.
The coil I was dreaming about, was a Very small one.. DD butterfly model like you say. I got the a coil cover to a Vista 5x8 and that could Be helpsome in this project.
Ill Be back with some more data and measurement.
Henrik.

[golfnut]

some guesses..

[Henrikras]

It not guesses... Seem close to spot on. Sorry about the silicone, but I'm afraid to damage something.

[sinclairuser]

white vinegar and a soft toothbrush will help remove that silicone, it takes time but works.
also the shielding (carbon/graphite) may be mixed into the potting compound, designed to make contact with the negative coil wire in the potting compound itself, a have opened a coil and repaired the wire for a customer but it was ages ago and cant remember what it was off, but the shield was connected to the - rx terminal with a small piece of wire embedded into the pot compound.
the positive tx coil wire was detached from the cable, i removed just enough of this compound to reconnect it and filled it up with epoxy, then i glued the shell back together, it worked after but if it had been the shielded rx wire i doubt it would have worked.
this was not a hobby detector but the principle is the same.

[ivconic]

white vinegar and a soft toothbrush will help remove that silicone, it takes time but works.
also the shielding (carbon/graphite) may be mixed into the potting compound, designed to make contact with the negative coil wire in the potting compound itself, a have opened a coil and repaired the wire for a customer but it was ages ago and cant remember what it was off, but the shield was connected to the - rx terminal with a small piece of wire embedded into the pot compound.
the positive tx coil wire was detached from the cable, i removed just enough of this compound to reconnect it and filled it up with epoxy, then i glued the shell back together, it worked after but if it had been the shielded rx wire i doubt it would have worked.
this was not a hobby detector but the principle is the same.

Yes white vinegar will slowly dissolve silicone, i also tried it several times. That's good advice.
...
"...money is the root of all evil..."
Finally one honest and decent man from "western" society!
My personal solace is that i am too far away from any evil lately!
...
I was "against" taking a part Deus coil from very simple (and now obvious) reason: it's damn SMT with very small, flea-look-alike components, most of those modern and unknown for common diy'er, most of those impossible to obtain at local suppliers...etc
It's based on some PIC with "million" pins, all around, protected code, very tricky wifi tranciever which may easily burn for no obvious reason (have seen few burned so far).
Apart from that; coil is filled with hardest epoxy you can imagine, some of the wires are literally "floating" somewhere in epoxy mass, we don't know if there are some "hidden" components in epoxy too... etc...etc
Many sane reasons for not to open it. Even if opened; nothing we can do with it else than keep it in good shape and use it for it's primary purpose.
Main Deus advantage lays in it's design, shortest possible path from coil to A/D uPc inputs. Direct conversion and most accurate phase analyze seen so far in this category.
I am afraid that playing with custom coils, longer signal paths, eventual ground loops; will significantly distort A/D conversion accuracy and will cause improper and inaccurate behavior.
Now when is done; best possible advice would be to keep up the shortest possible paths from coils to pcb, to balance custom coil directly on live Deus (observing all the reports on it's LCD), and to take very special care around it's wifi because it is very easy to burn it.
I must admit that i didn't expected such high inductance at TX !?
Anyway; thanks for the effort and sharing the photos!
Cheers!

[Tepco]

My greatest concern about Deus coil is that it uses some form of electronic\software coil balance, either directly (direct software subtraction) or by substitution with internally generated signal. Not sure if coil balance can be established purely by mechanical adjustment with all the stuff inside. Also, some other internal calibration data, specific for each coil, can be stored in memory. If so, these routines are probably well protected and hard to hack, this can make any coil modification nearly impossible. BTW what MCU and wireless chip type is inside? Dissecting it just for curiosity can be interesting.

[ivconic]

...BTW what MCU and wireless chip type is inside? ...

It's a PIC.
It's code looks like:

":80f99c590c0ff9:b359f77f77edd928fd31a66fdb
:8ae39f6af42c11
:798a6261337a28025a4b0c4d4c3c1e9402103b4417
:19b80de2485c4367daf8261ef88c60256ef0ccfb05
............................"

[Henrikras]

Hey again...
First .. Its to much of a risk to try vinegar. I got no exsperience and the silicone is Very soft.. Sorry but that wouldn't Be tryed.
I've just put the board and Battery back in place and turned on.. Its working. (Here In lab). But I must take it out to the test bed outside to see Maybee lose or chattering.
But so fare it seem to Be okay.
Ive tryed hard to figure out the shielding.... At the picture you see this Grey resin or epoxy. Its got NO resistance at all.. Even mega ohm none. And at the botton side of coil (Grey resin) allso no resistance can be measured. With this observation and that there is no fifth wire coming out, I came to the conclusion that there is NO normal known Sheilding. If they had a connecting hide inside the coil, I would say it would difficult in production, so that theory I dont think either.
I think XP found out a way to make this, where they dont need Sheilding...
I could measure at lot (tx power and so) but I think I have to keep this simple, I not an Expert like you.. Therefore If there is anything very important you can live without, please ask, and I will try.
I still can't see any advance to keep the board and Battery close to the searchcoil.... But Maybee you right. Time will tell.
It seem more to me that this XP system isn't so sensitive.. (By mistake Ive change the RX in the TX place and TX in the RX place, and turned it on.) it did start up, and the control box and the board found out that the coils wasn't right... The control box did a scan at the Freq. and at tweaked Freq. 11.416 came up and it began to work, of course not good, but it reacted and gave signals...
The 17 khz area didn't work, but 11 and 7 and 4 khz, did. Its Saying to me that this project wouldn't Be so difficult.. (Hope so)
Next step is scope and phase measureing.
Henrik

[Skippy]

While it is true that the micro on the board could store calibration data unique to each search-coil, there's not a need. If the induction balance is good enough that the analogue electronics can cope with the levels (eg. within +/- half of full-scale range) then the power-on auto-calibration should be enough. Coils all drift a bit with temperature, long-term aging etc, so there is no point in storing fine details of residual signals.
One potential problem with this small-coil conversion: The coil might identify itself as a 9 inch coil to the control box/headphones, and they may operate in a different manner - eg. depth gauge readout, filter time-constants, more. Changing the coil to a smaller one may result in slightly less than optimal operation. It won't be serious, after all, just about every other machine on the market is dumb and can't tell what size coil is on it.

[ivconic]

Hey again...
...I think XP found out a way to make this, where they dont need Sheilding...
...
Henrik

Our dear fellow Davor here on forum already elaborated such approach.
And i think he is on right way.

[Tepco]

It's a PIC.
It's code looks like:

":80f99c590c0ff9:b359f77f77edd928fd31a66fdb
:8ae39f6af42c11
:798a6261337a28025a4b0c4d4c3c1e9402103b4417
:19b80de2485c4367daf8261ef88c60256ef0ccfb05
............................"



[ATTACH]28528[/ATTACH]

Did you tried to disassemble it? What type PIC is inside?

[ivconic]

Did you tried to disassemble it? What type PIC is inside?

My goal is to become friendly with Mr.Alain Loube; not to be put on their black list!
Sorry!

[Tepco]

Then I have to resort to my Russian friends, they are not friendly with anyone. Actually someone managed to steal entire source code. Only, this is not a priority at this moment, I even don't have Deus.

[ivconic]

Then I have to resort to my Russian friends, they are not friendly with anyone. Actually someone managed to steal entire source code. Only, this is not a priority at this moment, I even don't have Deus.

It's pointless.
It's much cheaper to legally buy it and use it.
For hacking up the protected mcu you will need original device, send it to "the guys" and wait infinite time.
There is always good chance device to be busted, you will receive no guarantees from "the guys".
Even if they make success (slim chances in this particular case), they will ask you least $5000 in advance, before you receive back your device with extracted code.
So as you see; much much cheaper to legally buy it, use it and stay happy with it.
...
As i suggested at the start of this topic; we better gather here, vote and form public appeal to Mr. Alain Loube, asking him to reconsider production of smaller coils, more suitable for prospecting.
Because, so far as i saw; new Gold Field program is just SPLENDID for prospecting! Easily beats up some specialized and much more expensive models (5000e ... ehemmm...)!!!
Cheers!

[Skippy]

It's certainly possible for XP to make a small coil - talk of not being able to fit the electronics/battery inside is not really true. You only need 18 &12KHz with a small coil, and users generally report that there is hardly any difference between the two, anyway. So you could just have one frequency. 12, 15,18 no real difference, then you can do away with the relays. The circuitry could be shrunk considerably, the LiPo cell could be half the size, probably smaller, as capacity improvements have been made, and you likely don't need the 'high power' setting anyway, as small coils pick up less EMI etc. If I was the designer, I would go for 2 freqs in a 2:1 ratio, like 8KHz & 16KHz. You can get smaller relays, those [durned expensive] TO-5 can type, I'm pretty sure I've seen latching types.
It's interesting that the TX winding in Hen's coil is just a plain single winding. When you consider it's likely to be driven from 3 Volts, you would think that a centre-tapped coil would be a good choice. They do that in the Tek T2, which has 5 Volt [+ maybe 6 V] internal supplies.