LM394 on my Minelab PCB

[mario]

1.176MHZ?

K Mick,

I went through my figures at least 4 times and I'm getting the same Frequency across all components (including caps) of yellow AND purple rail when referenced to coil ground/ blue wire : 1.176 MHz. I reproduced that frequency in both 1x and 10x settings, if worth mentioning.

I have an old analog scope, but I also have a digital function generator where I was able to reproduce my calculations to verify I was doing them accurately. I'm not sure if there's anybody out there with a 2100V2 that may confirm this Frequency.

I do agree with you in that it's a bit low. I was also expecting a number between 2-4 Mhz also. I don't know why I'm getting that Frequency. Right now my earthfield trimmers are all out of whack, but these should not change that frequency by a huge measure. Perhaps, I should have reset them back to factory settings before doing frequency calculations?

Also, when referenced to coil ground, the voltages are:
yellow rail: -4.94
Purple rail: -2.28

Now, before I start playing with the caps, what type caps should I get? I know you mentioned 1uF caps, but what material? Higher frequency is better for the smaller objects right? If that's the case, I'm going to get that frequency up to 3, 4 or maybe 6Mhz!

Also, those jumpers don't do a damn thing!

At least, when frequency is concerned. Maybe, I should have done more testing...

[Mechanic]

Hi Mario,

Your blue trace must be ground, and the green trace +5v.

Your earth field trimmers will have no effect on the operating frequency of the detector. They just adjust the amount of one sample that is subtracted from the next sample which when adjusted correctly will cancel low frequency noise and earth field voltages as the coil is swung and tilted.

Now if that's the frequency, well then that's the frequency! Does adjusting the frequency trimmer knob change it slightly? Maybe not enough to notice on a scope. That was 1nf, not uf! Use some polyester caps(the yellow ones) to experiment with and then if your ever happy replace them with polystyrene or polypropylene caps.

Now maybe the jumpers don't effect the tx timing but what about the sample timings? If you probe the control pins on the 4066's there may be something different there. Compare it to your transmit trace, I hope you have at least a dual trace scope. Also you will be able to draw yourself a diagram of the transmit and receive timings this way. You will also be able to look at the eeprom pins and see which pins are outputs, as you should see the same wave(eg samples pulses and transmit pulses), perhaps just a lower voltage.

Any chance you could post a pic of this 1.176mhz wave? Does this wave also go to your eeprom to the pin you have marked 7?

Have I mentioned before, don't blow it up! Might be a good idea to do that scope shot catalogue of all of the pulses going on in there use the transmit pulses for a timing referance. Could save a lot of time later....

Cheers Mick

[mario]

Good job gentlemen. Give me the remaining ICs, and I promise you the hex!
High Res here

[mario]

Nice drop in upgrades:

LM74HC273D with MM74HC273WMX
http://www.fairchildsemi.com/ds/MM/MM74HC273.pdf

74HC4040 with TC74HC4040AFN

http://www.toshiba.com/taec/components2/Datasheet_Sync//151/353.pdf

The replacement for the 74HC273D is note worthy in that the chip will be able to handle frequencies up to 30Mhz (up from the current limit of 7-10Mhz), while clock output delay times are now reduced to 18ns.

But a more noteworthy upgrade is the TC74HC4040AFN
Those specs are simply amazing! I'll probably end up dumping the hex before I replace those chips...although, to be honest, I can't wait....

[Mechanic]

Hi Mario,

Personally I wouldn't worry too much about changing either of those components. Faster turn on times means more supply current used especially where long traces are used. I don't see any advantage in this. The best thing for you to do is to get control over the clock, figure out the eeprom and how to program your own timings in and find sources of internal noise and reduce them.

Your yellow trace is -5v your blue trace is gnd and your purple trace is +5v These will go to just about all of the ic's. It might be an idea to just mark the pins at the ic and bypass cap with the color and not run the traces all over the place as it will very quickly get messy!

Ic6 must be a level shifter and the one above probably is too.

Cheers Mick

[Rov]

Is there a dump from the prom! for this yet anywhere,an what format to read?

[Rov]

Maybe just a txt file to re edit and reprogram.

[mario]

Maybe just a txt file to re edit and reprogram.

If things were only that easy!

I have not dumped the file just yet, but its not going to be a txt file, its going to be a binary editable with a hex editor. The good thing is that this eeprom has wide adoption so the software to extract it and modify it is not going to be an issue.

The hard part is the actual removal and relocating it. From prior experience, I can attest to the risks involved in the damage to data integrety when handling these eeproms.

[Mechanic]

Hi Mario,

The best way I can think of to extract the binary would be to remove the 4040 and hookup you address lines from an eeprom programmer there and then also hook up your i/o lines to your programmer to read from. But you will not get the full eeprom this way but you will get enough to revert back to the original timings. To do a full dump you would need to also gain control over the other address lines too. If there were any secrets in there they will be revealed this way and you would also be able to figure out how to access them.

The best way I can think of to remove the eeprom would be using a hot air rework station or an even better way may be to use an infared soldering station. A mobile phone repair place should have one of these and probably places that repair playstations and x-box's too. But I would not recommend you do it because if it fails your detector will be kaput! If you were going to do it however, I would make sure you had all of the timings (i/o pins)figured out so you could replicate them if you had to start from scratch programming a new eeprom. The other thing we don't know is if the addresses are being accessed in their logical order. Someone more cluey on the 4040 may be able to figure that out.


Cheers Mick

[mario]

scope out put of the different caps

No caps, Stock
50mv/Div 2uS/Div x10 settings
Purple trace between caps








Pin 10,IC 4040
.1V/Div 2 uS, 10X


====================================
====================================
5.8nF cap

50mv/Div 2uS/Div x10 settings
Purple trace between caps


Pin 10,IC 4040
.1V/Div 2 uS, 10X


====================================
====================================
14nF cap

50mv/Div 2uS/Div x10 settings
Purple trace between caps





Pin 10,IC 4040
.1V/Div 2 uS, 10X



===================

Hi Mick,

I removed one of those three caps ( the fat green one next to IC 7 ) and installed some resistor pegs to the soldering points. From those pegs I soldered my caps.

Those green, larger caps are adjustable caps, as they are only rated at 1 nF! It is the small purple one (the one next to the other fat cap) that is rated at around 86nF...

Anyways, so much to do so little time...I don't have have caps rated at >15pF, so I am hold up on the order...

Can you help me understand what is going on with my scope readings? I don't think the frequency is changing by much ( just the voltage), by increasing the size of those caps? Am I wrong?


Also, I have thought about it, and you're probably right. I will have to have some one else with the proper station to remove that eeprom. I don't mind at all if it helps me preserve the contents of that IC...The only problem that I see is time, I have so much to get done, and this will come at a later time.

[Mechanic]

Hi Mario,

It does not look like the frequency is changing at all. You may have to try the 86nf one and see what happens then. Also you may have to try changing resistor values in the area. Use some variable resistors. Make sure you record the value of the original resistors along the way too.

Cheers Mick

[mario]

Hi Mick,

Is it possible to calculate the frequecy from the pictures? or is it best to set up the scope for that?

[Mechanic]

Hi Mario,

I just had a second look at the scope shots and the frequency is changing with the different caps! Bigger cap = longer tx and longer sample delay and smaller = shorter tx and shorter sample delay. The frequency does look to be about 1.2mhz standard and the frequency with the different cap is about 1mhz. If you can get the sine wave to complete in about 500 to 600ns you should notice a definite improvement in small target response, but ground noise may suffer, but then again the ground you have over there is nothing like the varaible ground here so it may not be a problem.

Now that oscillator I think would be a VCO so if you can change the control voltage by changing some resistor values this may also help control the output frequency. I would start be removing a resistor and measure its value and then put in a small trimmer potentiometer with the original value about midway and then just carefully move it a little bit to see if the frequency changes.

Cheers Mick

[Mechanic]

Hi Mario,

How long is one full cycle? (start of long tx back to the start of the long tx)

Cheers Mick

[mario]

man is this thing kicking me in the balls!

I doubt It'll ever be the same after I'm done with this. I'm melting the plastic shielding of a lot of wires from all over the board. I just don't have much room to maneuver the soldering iron...

I've removed many resistors from around the oscillator area. I'm finding it difficult soldering those 1/8 watt resistors back without negatively impacting the shape of the sine wave. Its taking me a lot of time through trial and error getting that sine wave back to its proper shape (I have been reinstalling every single resistor I remove until that sine wave starts looking nice. That white gunk on the resistors is having some impact on the soldering ). So far, from the resistors I've replaced, I have yet to find one that will change the frequency.

Also, if you're going to upgrade those darn BYV28-200 diodes, make sure you install them properly (as shown on my diagram)! I might end up replacing many more parts simply because I improperly installed one of them.

I should get all my caps tomorrow, sorry for the lack of updates...