LM394 on my Minelab PCB

[mario]

Rub it in why don't you!

It's been a while since I've been out there digging lead!

Let me know if you find one like this one. I will have no qualms whatsoever booking a direct flight to help you dig it.

I think I've had more luck finding meteorites in my part of town. Then again, I have always been too lazy driving north to the richer gold sites.

Anyways, I took a closer look at that IC, and guess what, it's looking like its not 74AHCT00. In fact, it's going to be 74HC00. Similar number, different datasheets with different specs. Unbelievable, no doubt. The difference is in (you guessed it) the output voltages. I think we've closed the book on that one.

Now that we have identified most chips, I think I'm going to focus more on the mods! We got our bases covered, and anything that blows can be replaced -Yay!.

...That is, unless we blow that remaining IC...

[mario]

update photo

Google sucks, so many screwed up links...Anyways, here's my updated high res photo. I'll confirm part 74HC00 as soon as it arrives.

Some other ICs worth replacing: LM393.

Faster parts are now available. I'm not sure if Minelab used those faster parts in 99 (or 23/8/2002 in my particular case), but even if they were available back then, i'm sure more power efficient alternatives do exist.

Or maybe you may want to replace them because you have your detector all butchered on your working station, you are bored with nothing better to do...those parts can't arrive soon enough

[mario]

LT1057S8 are incorrectly identified

I'm trying to wrap this project gentlemen. But it seems like there's always a set back. I got a hold of OP249s to upgrade the LT1057S8s after successfully identifying the 74HC00D. However, after replacing the LT1057S8s, I eliminated the threshold and both chips got hot quickly.

I focused my trace on that part of the PCB and after comparing the trace to the datasheets (here), It seems like LT1057S8 are incorrectly identified. Google should have updated my trace. Otherwise, you can find it here.

Also, I removed 2 NE5534s and replaced them with AD797s. I tried replacing the NE5534 underneath the original AD797, and It did not work. Because any other NE5534 will work, I'm assuming several resistors there are making the last AD797 incompatible.

Recent ICs that I have confirmed recently:CD40106, TLC271, LM393s, LM398, and 74HC4066.

Anyways, any suggestions on the LT1057S8 (or CHIP 9 for that matter )? I'll try to proper identify the LT1057S8 as soon as possible. Currently, I'm concentrating on interfacing with the eprom and replacing it with a micro -My next stage in the project. My preliminary research so far has been positive, but very time consuming. Its been a long long time since I did any programming.


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

Replacements for the BYV28-200s: SBYV27-200-E3/2: They're lower capacitance and faster diodes (15ns!).

[Mechanic]

Hi Mario,

Those diodes may look good but their current handling ability is not so good. This will generate heat in the diode which will cause the diode characteristics to change. It may make the detector more noisy after it starts and has been running for 20 seconds or so.

Those ic's we thought were lt1057 are not! have a look at the attahed image to see the pinout you need to look for and package. Also make sure it has jfet inputs and similar or better characteristics than the lt1057. A high slew rate is not important, take note of the slew rate of the 1057. The 10nf caps in the integrator feedback loop limits the slew rate of this device.
Theres' every chance it could be a texas instruments device.

Cheers Mick

[Morgan]

SD 2200 D Minelab

Also looks like there is 4 timings in there. Put 2 toggle switches on the front panel hooked up to those jumpers, compare the sensitivity between the different timings. When I have figured out what all of the bits do we will have a pretty good picture of what the different timings do.

Mick

Hi

Do you have imformation about how to build BIG coil for MINELAB SD ?

Regards

[Mechanic]

Hi Morgan,

Not as such, but I can give you some specs to aim for.
300 to 310uh, 0.4ohm, capacitance as low as you can get it. Make sure you space the shielding out from the windings as much as possible to reduce capacitance.

I think if you had an old coil such as the old green 18" coil, you could salvage the coil wire from that and measure it for strand size and no. of strands and wind it on a bigger form and still end up with the correct inductance.

Litz wire is preferable but hard to get in the specs. I believe there is some tin plated wire about that has been reported to work quite well too. Make sure when you are selecting the type of wire that it is insulated with a material that has a low dielectric constant such as teflon.

I hope that is of some help to you.

Cheers Mick

[mario]

Thanks Mick,

I think you might be right on the ID of those op amps. I have a strong impression the original parts might be TL072s.

If so, I found awesome drop-in replacements in the OPA2140. Very expensive, but a much needed upgrade. I have them on order and will be confirming them as soon as I get them.

Also, thanks about the heads up on those diodes.

OPA2140

Datasheet: http://focus.ti.com/lit/ds/symlink/opa2140.pdf


Bandwidth 11MHz
Low Noise: 5.1nV/√Hz
Slew Rate: 20V/ms
Settling time: 880ns


Morgan wrote:

Hi

Do you have imformation about how to build BIG coil for MINELAB SD ?

Regards

Hi Morgan,

lots of information can be found in the coils section of the forum. You're going to need a LCR meter, an online inductance calculator, and a lot of patience. Just make sure your inductance is as close as possible to 300uH and Resistance is around .5 OHM. I've gotten away with 1.2 OHM resistance, but your coil will not work if you stray too far from the 300uH inductance. Also, if your resistance is too high, it can potentially damage your detector, so don't stray too far from the parameters. As you create larger coils, resistance is the one issue that becomes more difficult to manage, and it depends much more on wire thickness. The thicker the wire the less resistance. Shielding is also important. At some later point I will be updating my threads to include my new methods for proper shielding. Still, I think I've posted enough information on how to do my two coils where you don't really need special equipment. .... good luck

[serci78]

ZVP0530A MOSFET 0.55A 300V P CHANNEL TO92 ZETEX
ZVP1306A MOSFET 0.2A 60V P CHANNEL TO92 ZETEX
ZVP2120A MOSFET 0.12A 200V P CHANNEL TO92 ZETEX
ZVP1308A MOSFET 0.2A 80V P CHANNEL TO92 ZETEX
ZVP3310F MOSFET 0.075A 100V P CHANNEL TO92 ZETEX
ZVP1306A MOSFET 0.2A 60V P CHANNEL TO92 ZETEX
ZVP1308A MOSFET 0.2A 80V P CHANNEL TO92 ZETEX

ZVN0120A MOSFET 0.16A 200V N CHANNEL TO92 ZETEX
BSP149 MOSFET 0.48A 200V N CHANNEL SOT223 SMD SIEMENS
BS107G ROHS MOSFET 0.16A 200V N CHANNEL TO92 ON
BSS92 MOSFET 0.15A 240V P CHANNEL TO92 PHILIPS

byv28 = use BYW29-200=DSEP8-02A DIODE HiperFRED 8A 200V 25ns

lt 1057=other opamp use ???====use instead ??
TLC27M7 =instead=TLC272 ???How many amps drain source current of MOSFETs?????

[mario]

Serci,
my friend. Welcome to the forum.

Regarding the p fet, you want to find one with better capacitance values than the original part. All that info is found in the thread (you may also google the datasheets)...now, regarding the tl072, we are currently working to ID this op amp. I will let you know in a few days when i get them. Good luck and please do read the thread in its entirety before posting. you will find a lot of useful info that will help you answer a lot of questions that you may have.

[Morgan]

SD 2200 D Minelab

Thanks Mick,

I think you might be right on the ID of those op amps. I have a strong impression the original parts might be TL072s.

If so, I found awesome drop-in replacements in the OPA2140. Very expensive, but a much needed upgrade. I have them on order and will be confirming them as soon as I get them.

Also, thanks about the heads up on those diodes.

OPA2140

Datasheet: http://focus.ti.com/lit/ds/symlink/opa2140.pdf


Bandwidth 11MHz
Low Noise: 5.1nV/√Hz
Slew Rate: 20V/ms
Settling time: 880ns


Morgan wrote:
Hi Morgan,

lots of information can be found in the coils section of the forum. You're going to need a LCR meter, an online inductance calculator, and a lot of patience. Just make sure your inductance is as close as possible to 300uH and Resistance is around .5 OHM. I've gotten away with 1.2 OHM resistance, but your coil will not work if you stray too far from the 300uH inductance. Also, if your resistance is too high, it can potentially damage your detector, so don't stray too far from the parameters. As you create larger coils, resistance is the one issue that becomes more difficult to manage, and it depends much more on wire thickness. The thicker the wire the less resistance. Shielding is also important. At some later point I will be updating my threads to include my new methods for proper shielding. Still, I think I've posted enough information on how to do my two coils where you don't really need special equipment. .... good luck

Thanks
I will be very careful to not damage this detector.
Do you know how to connect the coil to the five pin in the cable ???

Regards

[Morgan]

SD 2200 D Minelab

Hi Morgan,

Not as such, but I can give you some specs to aim for.
300 to 310uh, 0.4ohm, capacitance as low as you can get it. Make sure you space the shielding out from the windings as much as possible to reduce capacitance.

I think if you had an old coil such as the old green 18" coil, you could salvage the coil wire from that and measure it for strand size and no. of strands and wind it on a bigger form and still end up with the correct inductance.

Litz wire is preferable but hard to get in the specs. I believe there is some tin plated wire about that has been reported to work quite well too. Make sure when you are selecting the type of wire that it is insulated with a material that has a low dielectric constant such as teflon.

I hope that is of some help to you.

Cheers Mick

Thanks a lot for all the imformation.

My experience is not good in building coils,only experience with BFO coils ,very simple...

Regards

[mario]

250V cap!

Hi Mick,

I continued working on my trace, and I made a discovery: That large cap on the top right corner is rated at 250V,100uF. If it's rated at that high voltage, those diodes should be rated at least 250V too, no? Otherwise those diodes must be operating at the height of their limit assuming they're BYV28-200s. Unfortunately, when I first removed them, I didn't pay much attention to part numbers, so I don't know if they're BYV28-200s.

Also, Its no wonder this machine uses IRF640s instead of BUK455s, and IRF730 instead of IRF710? The higher voltages?

I'm guessing it's going for depth at the expense of sensitivity? Maybe that's why the signal length is at a long 240uS (to stabilize the threshold )?

Should it be better to replace that 250 cap with a 200V low ESR part? That way the threshold stabilizes at lower than 100us and be even more sensitive? What do you think?



Also, the OPA2140 arrived in the wrong package so I was not able to confirm TL072s. I reordered them in the correct package and should be getting them in a few days.

[Mechanic]

Hi Mario,

Yes the cap is rated to 250v, but this is all it needs to be. It only ever sees 180v. The irf730 is used to keep this cap at this voltage by shorting across the cap in series with the 68 ohm resistors. The tlc271 senses the voltage across the cap and feeds the gate of the irf730 to control the voltage. It senses by means of a voltage divider formed by the 2 110k resistors and the 3k resistor which is hooked up to -2.5v fed from the lm336z2.5 in the power supply section. I don't see how tinkering with this area will improve your detectors performance. You could lower or raise the flyback voltage by adjusting the value of the 110k resistors, but I have been there and it makes it harder to ground balance and I could not see any depth improvements. Also If you do fiddle just be warned that you will end up frying the tlc271 and then end up with a flyback voltage of 12v that lasts 200us! so your detector will not work. The other direction things can go wrong is if the flyback voltage goes too high. The voltage will avalanche across the tx fets at about 250v. The threshold will become unstable and you will have problems ground balancing.

Cheers Mick

[mario]

...Also If you do fiddle just be warned that you will end up frying the tlc271 and then end up with a flyback voltage of 12v that lasts 200us! so your detector will not work. ...

I solved that problem a long time ago!

[mario]

I have a question that has bothered me for some time: has any one discovered the origin of that sine wave we see in the oscillator section?

Here's a possible clue: When you remove all 3 caps ( 1nF,1nF and 6.8nF), the detector will not boot. However, one day I removed CD40106 in addition to the 3 caps and discovered that the detector boots! All you get is a perfectly smooth null sound.

With the CD40106 removed, I was able to detect the sine wave very, very faintly -too small to resolve on my scope, but it was there. I found it in the purple trace between the two 1nF caps -the purple trace north of that so23 small n fet looking component.

Maybe it originates as noise from the 4066 or MC14504 (though this does not make any sense), and those caps amplify it?

I ask because none of the components seem to be a dedicated wave generator.