TESORO GOLDEN SABRE

[Molzar]

"...Problems are some erratic noises..."

I heard this from many people recently. Among 22 devices made so far, i also had few cases, where some ocassionall chatters and noises occurred.
Really...I dont have universal solution to solve this problem. Usually it is about bad coil.
I posted many posts arguing and explaining ONLY one proper way to null specially this DD coil for this TGS. Yet..very few memebers accepted my
advices.
Do not null coil using RX stage outputs (LF353)...WRONG!
Null exact coil for exact TGS using only its TX output connected to TX coil and RX coil leads connect directly to milivoltmeter, BUT NOT TO TGS!
I know...few experts here will again deny my claims and method, explaining theoretically why it is not proper.
Do i give a sh*t? Of course not.
My TGSL's and coils works excellent,calm and very "deep", despite "wrong" method i used to null them.
1e coin at 34cm in the air and 28cm in the ground. Rock stabile, no chatters, no falses. Perfect discrimination.
Only problem i do have with almost all TGSL's is weak audio. Not so loud.
Everything else is perfect.
I wish you success at the end.
After all....try my method and you will see.
Best regards!

Hi Ivconic,

I think noise could also be "bad coil" or other "xtalk" on pcb. This will be focus after solve coil / depth problems, maybe solve both with good coil!!!. I will try your method for nulling Rx coil.

When you say "not connected to TGS" this means both Rx leads are isolated from PCB correct? What about shield? I would think shield of both (Tx and Rx) should be connected to TGS pcb.

About coil making methods, is it important to keep inner winds far from outer winds or is "scamble" wound OK? I use "fixture" that winds coils fast so keeping turns in relative position is impossible. Also I use poly spiral wrap to bind coils, then put sheilding. But I think you post here you put shielding directly over coil wire. Any ideas or expert advise would be helpful here.

I have been following this thread for much time and have much repect for your ideas and advice.
Thanks,

Molzar

[Max]

Hi Ivconic,

I think noise could also be "bad coil" or other "xtalk" on pcb. This will be focus after solve coil / depth problems, maybe solve both with good coil!!!. I will try your method for nulling Rx coil.

When you say "not connected to TGS" this means both Rx leads are isolated from PCB correct? What about shield? I would think shield of both (Tx and Rx) should be connected to TGS pcb.

About coil making methods, is it important to keep inner winds far from outer winds or is "scamble" wound OK? I use "fixture" that winds coils fast so keeping turns in relative position is impossible. Also I use poly spiral wrap to bind coils, then put sheilding. But I think you post here you put shielding directly over coil wire. Any ideas or expert advise would be helpful here.

I have been following this thread for much time and have much repect for your ideas and advice.
Thanks,

Molzar

Hi Molzar,
you use the spiral there ??? Like in PI coils ? Why ? Don't need it...

I use too spiral but only with PI coils not VLF.

Try this way if you can: wound coils, then wrap plastic tape on them; if needed you can wound two times with tape to make some spacing more but few/thin.

The important is you have at the end the coils really thight fixed: use other stuff if needed like cotton cords etc etc... and over you put the tape.

Often you can just use the tape only to fix thight the coils... then wound above the shield strip: one UNIQUE strip of foil, cut long as you need (1m or more).

Do not overlap too much, e.g. if strip is 2cm large just try to overlap 7-8mm at each turn (less is better) BUT remember that you must cover all coil apart gap space and overlap all turns from first to end.

Then after shield is wound, wound again with tape to keep it well fixed.

That's all. Then epoxy everything after you positioned coils in container.

This way you'll get very nice VLF coils ideal for TGS, though there are many variants ... like using mylar etc etc

Kind regards,
Max

[ivconic]

"...When you say "not connected to TGS" this means both Rx leads are isolated from PCB correct? What about shield? I would think shield of both (Tx and Rx) should be connected to TGS pcb...."

You better search for wiring diagram on previous pages.
I sketched pretty "handy" coil-cable-device wirements.
Yes,Faraday cage is connected permanently in search head with both;TX and RX ends. Yet,RX end is separated through cable with standalone wire(pvc coated) and goes to RX along with "hot" RX, on pcb.
But in case of nulling,you better disconnect both RX from pcb and connect them (in air) with 200mV AC milivoltmeter. TX remains connected on pcb.
Switch ON device and meter and observe induced voltage.
So called residual voltage. With caution, move one of coil over other, and try to achieve least possible voltage readings on meter.
Usually takes long time for me to finish this and achieve almost zero induced voltage on RX coil. ALMOST zero....not absolute zero, cose,simply, it is not possible.
Dont know (dont mind) about exact resonant frequency on RX coil. Really, showed less important. 24 excellent coils made so far are proving my words.

Last comparator hysterezis already solving most of selfoscilations.Nothing more could be done there, so dont bother.
Fets comutation....yes,Max is right there.Proper choice of Fets could solve this problem also. I dont think BF245 are best solution. I am using J107.

Do not take original Tesoro coil and according to it do not make any conclusions - wrong!
Make your own coil.
Whenever i listen some - same old "experts" here i fail to achieve best results.
Once i started to use my own brain - all started to work excellent.
Cut the crap,stop "scientific" B.S. and start thinking properly.
Many years of experience in REAL making of REAL,WORKABLE and EXCELL metal detectors made me think this way.
To many phylosphes here...without at east hour in practice.
Ask those how many workable detectors they made so far! Neither one!
Yet they are very ready to come here and spread "wisdom" and "science" with lot calculations,formulas and blah,blah....
Listen Max,Nakky,Ap Berg and me if you want to achieve something.
The rest......forget those!

[Qiaozhi]

Hi Qiaozhi,

This should be easy to prove by swapping the coils so the Tx is the lower frequency (higher inductance). Also if you take a look at Carl's coil data http://geotech.thunting.com/cgi-bin/...oils/index.dat the one Tesoro DD has the Tx as the higher inductance and all the concentrics are the opposite, Tx lower inductance.

I've read many posts on Tx frequency of ~14.6 KHz here, but where does the Rx frequency (16.12 KHz) you stated above come from?

Will try swapping coils tonight...

Thanks,

Molzar

Hi Molzar,

Swapping the coils will probably make very little difference. In fact, the resonant frequency of the RX coil, plus its associated capacitor, should still be above the TX frequency. However, at the moment, I can only think that this could be the reason for the discrepancy. When I get some free time I'll make some tests and let you know the results - using my own "working" detector.

You can easily calculate the TX and RX frequencies using:



I measured the Tesoro coil inductances as 5.9mH (TX) and 6.5mH (RX), which are slightly different to the coil data chart

[Molzar]

[quote=Max;72076]Hi Molzar,
you use the spiral there ??? Like in PI coils ? Why ? Don't need it...

I use too spiral but only with PI coils not VLF.

Max,

I use spiral wrap cause is good (tight) constraint on coil wires. Thought this way would apply to VLF as well as PI. I will use your silimlar method on next coils to see any difference.

I guess since no one answer on care of windings, this doesn't matter?

Ivconic,

Yes, I am using your wirement drawings for coil connections, very helpful. I guess what I really whated to know was if "when nulling" if the one Rx (not Hot) is still connected to gnd in search head? Also Do you put tape (like Max) on coil winding or shielding first?

Tried nulling using your methods (I left the Rx lead connected to gnd in search head, mainly cause this is soldered and to change is PIA (Pain In A**) other connections have screw terminals for easy experimentation. I'm using 2 pair shielded Belden cable, should be good for this app.

Nulled down to ~1.5mV AC with multimeter...... But still only 19 cm on 2 cm coin..., maybe is not coil? Or maybe spiral wrap is coil killer? What you think? I need big BEER now...

Maybe I try your other way nulling by swing coin over coils while moving coils, very simple, effective method!!!!!(If you have tuned TGS)

I am very "practical" person, I too know from experience that "theory" does usually not work and should always use our first hand knowledge to solve problem in real world. But theory has its place, you just can't let it limit your thinking and ideas.


Qiaozhi,

Swapped coils, Rx still shifts to right when triggering on Tx. Let me know if you figure this out.




Many thanks to all,

Molzar

[Aziz]

Nulled down to ~1.5mV AC with multimeter...... But still only 19 cm on 2 cm coin..., maybe is not coil? Or maybe spiral wrap is coil killer? What you think?

Hi Molzar,

Multimeters have mostly a cut-off-frequency of about 1 kHz. This means, you can measure AC voltages only up to this frequency (-3dB).

Anyway, with multimeters balancing can be still made by minimizing the voltage on the RX coil.

Aziz

[erolunall]

Hello
My friend asked help for not working TGSL works that he attach his works below links.
it is not giving sound.
İf anyone find any error please write.
Regards
Erol
http://www.postimage.org/image.php?v=PqS6M3i
http://www.postimage.org/image.php?v=aV21O9vi
http://www.postimage.org/image.php?v=aV21OjtS

[ivconic]

"...Ivconic,

Yes, I am using your wirement drawings for coil connections, very helpful. I guess what I really whated to know was if "when nulling" if the one Rx (not Hot) is still connected to gnd in search head?

YES, it remains connected in search head with F.Cage and TX end.




Also Do you put tape (like Max) on coil winding or shielding first?

I put selfadhesive Al foil tape over windings. Than scotch tape over it.

Tried nulling using your methods (I left the Rx lead connected to gnd in search head, mainly cause this is soldered and to change is PIA (Pain In A**)

Indeed, i agree.


other connections have screw terminals for easy experimentation. I'm using 2 pair shielded Belden cable, should be good for this app.

Belden cable is superb!

Nulled down to ~1.5mV AC with multimeter...... But still only 19 cm on 2 cm coin..., maybe is not coil? Or maybe spiral wrap is coil killer? What you think?

??? I am confused!? How can i pronounce my readings??? At 200mV AC range it says 0.07-0.09 when coils are nulled perfectly..First excellent working coil i made has that reading so i adopted it as reference for future coils.Ever since i nulled all my coils to achieve same readings...No mistake so far!


I need big BEER now...

I have extra beer! Join me!

Maybe I try your other way nulling by swing coin over coils while moving coils, very simple, effective method!!!!!(If you have tuned TGS)

Can be used but not accurate...Better do the first method.

I am very "practical" person, I too know from experience that "theory" does usually not work and should always use our first hand knowledge to solve problem in real world. But theory has its place, you just can't let it limit your thinking and ideas...."

I went to schools for 17 years.Of course theory is basic. Problem is that Physics didn't explored and explained all. That's why we are here! Right!

[ivconic]

Hi Molzar,

Multimeters have mostly a cut-off-frequency of about 1 kHz. This means, you can measure AC voltages only up to this frequency (-3dB).

Anyway, with multimeters balancing can be still made by minimizing the voltage on the RX coil.

Aziz

This is true. That's why we cant rely on accurate readings. Yet, in absence of better instruments, AC milivoltmeter can be used. First take some properly nulled coil and measure residual voltage on it. Remember results and than use those as reference for further handmades....
Keep in mind that residual voltage depends of frequency! Most important...or further confusing is possible.

[ivconic]

[quote=erolunall;72099]Hello
My friend asked help for not working TGSL works that he attach his works below links.
it is not giving sound.
İf anyone find any error please write.
Regards
Erol
http://www.postimage.org/image.php?v=PqS6M3i
http://www.postimage.org/image.php?v=aV21O9vi
http://www.postimage.org/image.php?v=aV21OjtS[/quote]


First check voltages. Plus is 8v and minus approx. -6.2V...
Than check oscillator (2N2907). Measure frequency at collector (coil must be connected in all cases)...
If there is approx. 14kHz on collector than trace that frequency up to pin 1. of CD4024. If there is not any frequency reading on 1.pin and there is on collector of 2N2907 than buffer 2N2222 choked it.Change 2N2222.
If there is 14kHz on 1.pin than check for f/2 on 12.pin of CD4024. So must read approx.7 kHz there. If not, CD4024 choked it, replace 4024.
If there is f/2 on 12.pin than check for divided transients on 1N4148 diodes on 5,6,and 9 pins on 4024...Those are forming composite audio.
If everything appears to be good...than just be sure that minus (-6.2V) is present and good. If yes, than chopper works good. If not chooper is choked.Replace 2N2907 and 2N2222 in chopper with BC557 and BC547...(sometimes 2N... transistors do not work there???)...
If chopper is ok and you have proper minus voltage...than check MPSA13 orientation...pins are EBC not CBE...
Sheeeesshhh! I am exhausted for now....
Check this first and let me know...
We will continue later....
Buaaahh....

[Molzar]

Ivconic,

Played around with coil nulling, got down to 0.95mv AC. Was increase in depth by 2-3 cm, so think I'm on right railroad.....

But I don't think my now coils can null lower. I spend big time making small ajusts and will not go lower. Will try making "better" coils using methods described. I think this "key" to good depth.

When I swapped Tx and Rx coils for Rx shift left / right experiment I got extra surprize!!! Noises and beeps all go away! So I keep that way for now.

May not post for few days as there is big party here in few days and have to turn me "workbench" back into Racetrack for childs. I of course will be collecting much cans and bottletops from favorite beverages for MD experiments

Oh...... wanted to ask Max about scope pictures he made with test point maps. Did you use 9x8 Tesoro or Homebrew DD coil for test pictures?

Thanks,
Molzar

[Max]

Ivconic,

Played around with coil nulling, got down to 0.95mv AC. Was increase in depth by 2-3 cm, so think I'm on right railroad.....

But I don't think my now coils can null lower. I spend big time making small ajusts and will not go lower. Will try making "better" coils using methods described. I think this "key" to good depth.

When I swapped Tx and Rx coils for Rx shift left / right experiment I got extra surprize!!! Noises and beeps all go away! So I keep that way for now.

May not post for few days as there is big party here in few days and have to turn me "workbench" back into Racetrack for childs. I of course will be collecting much cans and bottletops from favorite beverages for MD experiments

Oh...... wanted to ask Max about scope pictures he made with test point maps. Did you use 9x8 Tesoro or Homebrew DD coil for test pictures?

Thanks,
Molzar

Hi Molzar,
at that time I used an homemade DD coil... I use often on the bench also for other MDs. It's 100/105 recipe DD, nulled at about 15mV peak-to-peak.

Kind regards,
Max

[Aziz]

Hi,

if anyone is interested on a powerfull LC oscillator, the basic circuit is shown below. The output of the op-amp can be driven with pnp/npn output stage, to increase the drive current.
I have tested this LC oscillator type and it delivers as much power, as you need.

Here the source for this LC oscillator (german):
http://www.home.hs-karlsruhe.de/~ker...us/sinosz4.pdf

Aziz

[tiktak]

Hi,

Hi,
Max.
What is the dimension here ?1,6cm?Because I can see clearly.
Thanks

[tiktak]

Hi,

Hi,
and also this added area exept 12,75.How much it is 1,5cm?2cm?