MINE_LAB -F1A4 -A/OK- IMOP..

[sinclairuser]

a tip you can try once paint is removed, is scribble all over the chip top with a UV security marker, wipe it back off quickly, then try a uv light source over it, sometimes that brings out the numbers, you can try another application of the marker if the first does not help.
i was told a long time ago that trick, originally it said just scribble on the chip and uv it, but i found wiping off most of it helped.
sometimes it does not help, sometimes it makes numbers readable, it depends how deep it has been scoured.
and of course, eeproms that use uv to erase/flash should not be done this way, only electronically flashed stuff(though i doubt that will matter here).

[Kev]

That's an excellent tip, thanks Sinclairuser.

Here's the 2100 standard timing (Thankyou Mario et al.) that's located at the first page of the PROM. The sequence is 255 segments wide and if the long pulse is 240us, each segment is 3.75us.
I think trace 3 is the TX, 5 and 6 appear to be sync pulses, 0,1,2,7 samplers, and perhaps 4 is for the power supply, not sure switching right before sampling the long pulse doesn't seem to add up.

What this all shows is that there are two potential ways to skin the cat(fish thanks George and Carl)
The PROM has the capacity for dozens of timings, easily selectable toggling the upper address bits with a small micro, alternatively using a micro to alter the rate at which the data is clocked out of the PROM allows for very minor changes to be made to delays and pulse widths to target different sized targets. The former would be more powerful and allow �tunning� the machine to the dirt and atmospheric conditions.
Hopefully the F1A4 is like the 2100

Cheers

[hoadlies]

HI TO ALL from A Bl&$*?y cold Australia ,,

I am afraid I cannot contribute to the above its way passed my understanding!!!!
My F1A4 with the 14''-mono-coilteck Coil is fairly Good ,& the Whites GMT is deadly on very small Targets ,Plus the IDX so I guess that will do me for now ,I am yet to find even a small piece of GOLD HA.Haar..

MEAN-WHILE I am Busy getting this Toyota 4-runner 4WD to use in my nearby Bush to get-out detecting,,

I have Made a TOWBAR & almost finished a Bull Bar to protect against Kangaroo's in the Area where I will be going..THESE-Jobs come easy for me as an old Mechanic welder & I still have plenty gear to do this work & a good size W/Shop [ Shed 40x30 ft ]
This1993, Toyota was my Sons & I am buying it from him as he has Upgraded to later model ..
It will be much better than using my Ford Falcon wagon [Too-low]..

CHEERS GUYS,,,,,,,,,,,,,,,BARRY..

[sinclairuser]

dont worry barry, if these mods are doable i'm sure kev others and i can make them simpler, as i said before its the first time you do it thats hard, once sorted out you can make it easier and sometimes even fool proof.

[Kev]

Hi Barry,
In addition to what sinclairuser was saying, if you made the IDX and got it working then you will be able to do what I'm intending, but I can't be sure that what I'm intending is possible until I get my unit.
I propose a couple of levels of complexity.
1) simple switches and a knob that are to be soldered to point X. Similar to whats being offered for many $100s. It will still entail lifting a few pins slightly and sliding a thin insulator under them, but if I do a HD video, a child should be able to mimic.
2) a micro, probably a 30F3012 or 14 just because that's what I've got lots of, which will allow much more flexibility with respect to timings, gain and noise immunity, which adds up to being able to target all sizes of gold more effectively (get rid of a few of the compromises).

It's going to take a bit of time to sort it out, but there's some gold out there that's waited a long time for you to come along Barry, so a bit longer wont be a problem, in the mean time though concentrate on these few things.

a)Keep your coil sweep speed about 1/2 a metre per second with the F1A4 this was the best rate determined by the UXO detector test people.
b)Keep your coil as close to the ground as possible throughout the whole sweep especially the ends.
c)Listen carefully for small changes in the audio, run the coil back over the noise to see if you can brighten it up a bit, try sweeping the coil at 90 degrees, if the "target" appears to be moving about a bit, forget it, move on. If it brightens scrape and try again.
d)Keep you eyes on the ground in front of you, don't be looking ahead trying to judge whether that spot ahead of you looks better or not.
e)Look for areas that are difficult to work, i.e. under bushes, close to the base of trees, or large rocks, along the base of a rock face or crevice, rather than going round and round a stacking, go from the base to the top and back down again, working your way around this way. Traversing the heap this way especially when there are larger rocks in it allows you to get your coil closer to the dirt on the lower side of these rocks, a perfect place for gold working its way down the heap to come to rest.
f)If there are hills where you're detecting, carefully detect the Roo trails around the side of the hills, gold has a habit of coming to rest in the earth on the downward side. If you find a piece work carefully up and down slope from this spot.

I hope I'm not telling you how to suck eggs, that's not my intention, I just like to help people find their first piece, that one is the hardest, it's easier after that.
If these tips lead you to a big patch, all I ask is you let me step across the ditch and help you clean it up

This is a good thread, sadly they a few and far between on Finders these days.
http://www.finders.com.au/forum/view...t=1615&start=0
especially the post by Reg Wilson on the second page, a bit of inspiration for you.

Cheers
Kev.

[hoadlies]

HI,,,, S/USER,KEV,, Thanks for the encouragement ,My area is not noted for finds by detecting ,But it has not been detected much from info I have gained so Far ,So as I am a Pensioner I can spend time prospecting [After gaining permission from property owners ]its mainly sheep-station country with a few old mines dotted around over 200 odd km's,in hilly country..

I have purchased the Litz-wire etc to build a new coil for ,not shure what to build as yet ??? ..

We have unusual rain here that would make the creeks run so creek beds will be worth investigating..[too wet at the moment but with a 4wd [once ready will be the go..

I will report my results from time to time ..

Keep up the good work on the F1A4 & investigate that'' Programming port'' that/s incorporated in the unit I feel its part of the Multi-pin power connector as only 2x pins deliver the 6volt supply..leaving many other options ..M/lab could explain if they would be forthcoming?????..

CHEERS ,,,,,,,,,,,,,,,BARRY..

[Kev]

Hi Barry,
I think the port is just an output, but it could be worth investigating. If it gives out responses in real time before signal processing, then loading the logged data and GPS tags into software such as magnetometers use could show up a reef or such like not evident on the surface.

I was going to say detect the sheep tracks, cause that's what we have here, but I thought you'd have more Roos. Sheep tracks on hill sides are worth following too.
If very few people have been out your way you might be first to get to a patch or two. As Reg says research is key, and winter's the best time to do it.

All the best
Kev.

P.S. speaking of magnetometers, I recently loaded a mag app on my smartphone that's quite sensitive, I've yet to try it in the hills, but it may just help in extending the line of a reef!

[Kev]

The 2100 has 3 hidden timings in its PROM.
It's interesting to compare the differences between them. What is plainly apparent is how little adjustment there needs to be to effect a change in operational characteristics, I assume.



This is the timing on the 3rd 256 byte page, named X07set3b,
I assume it is a test timing of some sort.
It is the one that differs most from all the rest. Note trace 0 is missing two of the short pulse medium frequency demod time spaces.
It would be good know how each of these small and large changes between timings affect the mathematical expressions that govern their function.

Possibly a sweep algorithm could be employed to cover the boundaries of any given piece of ground. I imagined this would've been produced already given the emmergence of SETA but I suspect Australian soil is so variable within such a short time frame that this is not feasible.
For countries not affected by such lateritic dissolution, it could be feasible though?

I have simulated some code stubs but the resolution so far is 100ns, which will be too wide I suspect to allow stepping without triggering a target response in the processing circuits.

I hope to dump the F1A4 timing/s as these will no doubt be different to any of the SD/GPs and post them here, since this is a discontinued product and the patents will be nearing or will have reached their end.

Customs still have my unit.
Kev.

[Mechanic]

Hi Kev,

Check your pm's....

Cheers Mick

[Kev]

This is a warning example to anyone considering editing their SD2100/2200/F1A4 timings.
I must've dropped a couple of bits off the hex dump but after decoding, I still had what looked for all intents and purposes valid data sequences, but after rechecking to see why I only had 255 bits instead of 256 in each bit stream, I found that just one or two bit shifts and you could fry your transmitter.

This I believe is the correct base timing.

The one I posted the other day had the transmitter stream on data line 3 whereas on this one it is 7.

Fiddlers beware
Kev.

[Mechanic]

Hi Kev,

Nice picture, mind if I ask what program you used to generate that?

Cheers Mick

[Kev]

Hi Mick, pretty crude really, a spreadsheet screenshot. Crook in bed so got heaps of time. Hopefully I'll get my F1 tomorrow, l'Il have to make sure I don't make a too speedy a recovery, or me misses will have me back on painting the house.
Cheers
Kev

[downhalo]

So this is where all the F1 action is happening.. "or not"... hope someone posts some DIY info one of these days..

seriously, do it!

[Kev]

Hi,
This is one of the "hidden" timings on the F1. It has a single pulse (D7) and the potential to go very deep on real nuggetty stuff.

It requires a clock adjustment to run properly on the base unit.
I'm still working on a solution to that but have the hardware pretty well sorted and many stubs running on the development board which just need to be streamed together.

I'm using two 16bit PICs a master and slave, both running around 100MHz, this reduces any latency to naught and ensures glitch-less integration.

The slave produces a PWM system logic clock that is injected where the LC oscillator was. The 4040 logic clock divide by 8 causes an interrupt on the slave which outputs the timing sequence to a 573. The 4040 / 8 clock also goes to the 573 and latches the previous sequence out to the demodulators.

The master PIC has an 8x2 LCD attached and menu arrangement for gain, frequency and timing selection, it also controls an LTC6903 DDS chip which produces the master clock for the slave, so minor adjustment of the 6903 master clock produces a corresponding frequency change in the slaves PWM logic clock.
A 2 wire interface between master and slave allows timings stored in the slave to be changed. Sound is muted during timing and frequency changes. I may add a volume and tone adjustment later.

The F1A4 may be a sleeping tiger, I have seen an excellent potential during bench testing. I'm certain that with the right combination of timing, frequency and coil, the F1 will find gold that all previous Minelabs have missed. Jack Lange's video comparing his modified F1 to the 3500 is impressive, and underscores this units potential.

This has gone form being a simple mod to quite an intensive operation, and I'm not certain at this stage if I should make the artwork and code public, for one thing I need to try the unit out in the field first, and my patches are mostly under snow and ice at this time..............we'll see.

Kev.

[downhalo]

Wow, Kev yeah I wouldn't attempt something like that myself..

I'm more looking for the simple mods, Tone/Threshold.. maybe an extra frequency or 2... mostly just the Tone mod, as the default threshold tone makes me want to rip my ears off.. lol