TESORO GOLDEN SABRE

[Qiaozhi]

Maybe there is some quirk in the TGS design that causes this problem.

I've just been studying the TGSL schematic. It is the same double-differentiating architecture that I'm using. However, there is one interesting thing about the TGSL design that I think could make the TGSL easier to ground balance, which is even more of a mystery why you're having problems. When it's switched to All-Metal mode the DISC channel is still active, but instead of being sampled with the DISC sample pulse it is sampled by a slightly delayed version of the GEB sample pulse. This has the effect of creating a small dead zone that should make it easier to locate the GEB sample pulse over the zero-crossing, with the result that there is also some tolerance to changing ground conditions built into the design.

Here's an interesting thing you could try (see attached diagram).
Disconnect pin 1 of U106a, which will effectively disable the DISC channel, and let us know how this affects the ground balance.

[Max]

You are correct. Since the GEB sample pulse is positioned at the zero-crossing of the RX signal, then the output from the GEB analog switch (which is part of a sample & hold circuit) will be zero volts. Then only targets that produce a phase-shift (so that the pulse is not longer sampling at the zero-crossing) can cause a voltage to appear at the output of the gate. Neutral ground does not produce a phase-shift, only a decrease in amplitude. In this case the signal in the ground channel does not change.

The result is that 1mV, 15mV, 100mV ... it doesn't really matter that much how much residual voltage is there. The only disadvantages with a high residual voltage are worse discrimination for targets far from the coil, and more difficult (but not impossible) to ground balance.

I do not have a 20-turn pot on my detector, and ground balance is quite easy to achieve. Variation in ground conditions will always cause some falsing in a detector with manual ground balance, and fragments of ferrous brick can also cause some chatter. You should be able to eliminate this by increasing the level of discimination. By the way, the detector I'm using is my own design, but essentially it has the same classic 2-channel architecture used in many other analog detectors (with some refinements). The TX frequency is 10KHz and it uses the same capacitor values for the TX and RX coils on the Bandido. I've also checked this with an original Bandido and the results are the same. Maybe there is some quirk in the TGS design that causes this problem.

One thing you can be sure of is that Tesoro do not spend many hours trimming each coil to reduce the residual voltage. It just wouldn't make economic sense.

Hi,
yes I think too they don't spend much time on that...also cause they use standard housings...all the same...same turns for every model of coil they made... and just move the last half-turn maybe to get the "null".

I think that they null within 5mVpp tollerance... but then something could happen (relative movements of temporary fixings) during sealing process.

Just my thoughts.

Best regards,
Max

[Max]

I make several test with 6 DD coils with 10mVpp and 20mVpp in my IB detectors (they aren't TGSL, and work in 8KHz).
I tuned these voltages over the table (height=0.7m) in my laboratory, when I made this coils.
When, I put the coil of the flooring in my laboratory, the voltage is changes (decrease with ~8-15 mVpp), and the balanse are: ~2mVpp (-5mVpp?) and 12mVpp (5mVpp).
Also, when I put the coil over the earth of 2-3cm, the balanse is changed.
The detectors with 20mVpp have better depth and stability in laboratory and in the ground.
I think, when the coil is tuned with 0mVpp in the room, she in the ground will have -15mVpp and detector will have less depth and stability. It's a my supposition of this mystery.

Sorry for my bad english.

Hi,
yes you are right on conditions change: not the same nulling the coil on lab (e.g. suspended in air or over a wood/glued structure or plastic) and then on real ground. Of course, it's true that on the real ground you get the soil mineralization (iron oxides and the like) that put/increase magnetic histeresys of the whole stuff... then also the conductive part from ground (due e.g. to salts and wet)... then is the coil is not properly shielded also static charges accumulation and capacitive effects that change your TX freq. or your RX resonance... all true, all real and effects could be really heavy if some care e.g. on shielding aren't took.

Now about the lowering to floor of lab and increase... well... you have to know which kind of stuff there is in the floor layer... I mean some concrete e.g. panels are reinforced with iron-bars... some insulating stuff could be present (in some cases even al-foil were used in constructions)... there could be copper wires running over or below or in the floor itself...iron and other oxides and compounds could be present in the structure (that's really easy for concrete materials having aluminium and iron compounds traces but also other stuff like titanium oxides etc etc)

It's not a good test for nulling cause you probably don't know exactly what you actually have under your feet!

It's very difficault predicting what null voltage could be on soil(s)... you know that mutual inductance could change cause of external conditions.

From what I know the BIG companies do null using scopes like we do... but who knows what's the real reference voltage for null.

From my experience I've noticed much more stable working on soil with under 1mVpp null at lab.

I still think that the Qiaozhi's 100mVpp coil has something wrong inside.

Just my 5cents.

Best regards,
Max

[Qiaozhi]

Hi Ivconic,
I suppose he mean that if you rotate axes system by the right phase angle you actually null ground signal totally... so have no meaning that coil is nulled at 1mVpp or 100mVpp... cause projection of ground signal is always zero (90° to useful signal axis).

If so... you need to ground balance PERFECTLY to really zero out the ground signal... but I found it almost impossible to do if e.g. you don't use a 20turns pot for geb... of course... on the panel.

In theory it MUST work even at 100mVpp or more... BUT

then even slight variation of ground conditions or some fragments of read clay/bricks etc... can make you crazy with retune! That's my experience.

Am I right ?

Best regards,
Max

You are correct. Since the GEB sample pulse is positioned at the zero-crossing of the RX signal, then the output from the GEB analog switch (which is part of a sample & hold circuit) will be zero volts. Then only targets that produce a phase-shift (so that the pulse is not longer sampling at the zero-crossing) can cause a voltage to appear at the output of the gate. Neutral ground does not produce a phase-shift, only a decrease in amplitude. In this case the signal in the ground channel does not change.

The result is that 1mV, 15mV, 100mV ... it doesn't really matter that much how much residual voltage is there. The only disadvantages with a high residual voltage are worse discrimination for targets far from the coil, and more difficult (but not impossible) to ground balance.

I do not have a 20-turn pot on my detector, and ground balance is quite easy to achieve. Variation in ground conditions will always cause some falsing in a detector with manual ground balance, and fragments of ferrous brick can also cause some chatter. You should be able to eliminate this by increasing the level of discimination. By the way, the detector I'm using is my own design, but essentially it has the same classic 2-channel architecture used in many other analog detectors (with some refinements). The TX frequency is 10KHz and it uses the same capacitor values for the TX and RX coils on the Bandido. I've also checked this with an original Bandido and the results are the same. Maybe there is some quirk in the TGS design that causes this problem.

One thing you can be sure of is that Tesoro do not spend many hours trimming each coil to reduce the residual voltage. It just wouldn't make economic sense.

[rdec4]

I make several test with 6 DD coils with 10mVpp and 20mVpp in my IB detectors (they aren't TGSL, and work in 8KHz).
I tuned these voltages over the table (height=0.7m) in my laboratory, when I made this coils.
When, I put the coil of the flooring in my laboratory, the voltage is changes (decrease with ~8-15 mVpp), and the balanse are: ~2mVpp (-5mVpp?) and 12mVpp (5mVpp).
Also, when I put the coil over the earth of 2-3cm, the balanse is changed.
The detectors with 20mVpp have better depth and stability in laboratory and in the ground.
I think, when the coil is tuned with 0mVpp in the room, she in the ground will have -15mVpp and detector will have less depth and stability. It's a my supposition of this mystery.

Sorry for my bad english.

[ivconic]

"...In theory ..."


If i havent tested theory claims so many times...!!!?? I am dealing practically with things - always!
And my practice says that coil nulled so far from zero is.....unuseable on VLF machines!
OK i can accept totally unbalanced DD coil on PI detector. Never mind in that case...but on VLF???
As i assumed Quiaozhi's coils are more than damaged, that's why produce so much voltage...

Theory...heck!
I respect only things proved in practice. If i cant gain GEB with that coil than its "depth" in air means nothing to me!

...

[Max]

Hah,hah,hah,hah,hah !!!!!! I KNEW IT!!!!!
Damn! Now i cant sleep whole night!


Also...now....NOTHING CLEAR TO ME!
I dont understand this at all!?

Quiaozhi, tell me than; how the heck i will achieve good, stabile GEB???
If this is not dependable of good coil design....well..????
And i thought i know something!?
Regards!

Hi Ivconic,
I suppose he mean that if you rotate axes system by the right phase angle you actually null ground signal totally... so have no meaning that coil is nulled at 1mVpp or 100mVpp... cause projection of ground signal is always zero (90° to useful signal axis).

If so... you need to ground balance PERFECTLY to really zero out the ground signal... but I found it almost impossible to do if e.g. you don't use a 20turns pot for geb... of course... on the panel.

In theory it MUST work even at 100mVpp or more... BUT

then even slight variation of ground conditions or some fragments of read clay/bricks etc... can make you crazy with retune! That's my experience.

Am I right ?

Best regards,
Max

[Max]

The Tesoro coil that I tested is a 9x8 web coil, and it's reasonably new. The residual voltage is actually 13mVpp. I also tested a Laser concentric coil (same as Cortes) and this is 100mVpp! Both of them work absolutely fine. Unfortunately I cannot test my 10" concentric coil, as a wire is currently broken in the plug. Just need to get enough enthusiasm to re-attach it.
I do not own a tesoro widescan DD coil.

Personally I think you're taking this zeroing thing too seriously. The only difference you're likely to notice, if the residual voltage is too high, is worse discrimination for deep targets.

I'm sure that statement is going to spark off some interesting discussions.

Hi,
now I'm really confused too ! I belive your words ... so if you say that 13mVpp and 100mVpp for coils are good about working... OK.

But I cannot explain myself why I get so worse results with e.g. 20mVpp homemade coil... or why I get 3.5mVpp from my 9x8 and you get 13mVpp that's much more ...4 times what I have. Are Tesoro's procedures so different from coil to coil ? Or just your coil (or mine) is damaged ?

The 100mVpp coil I really don't understand how can it work on soil without giving much falsing... this is a real mistery for me.

So... what I read above means that 1 or 100mVpp change nothing about ground balancing ??? Which detector have you used to test them ? Has it the manual GEB pot on panel ?

I've noticed that I can compensate manually by GEB pot a higher "nulled" coil ... nulling ground influence by the axes system rotation but that is very critical do so with e.g. 20mVpp coil... and results are not so good as with 1mVpp coils.

Please explain... I'm really interested in that topic.

Best regards,
Max

[Geo]

Hi
I conceive that as long as least it affects the Tx signal in the coil of the receiver we so much more can strengthen the Rx input and gain depth. Simply the subject is "up to who point it is in effect in the practice". From who point and beyond we do not have essential improvement?? And naturally this only in the practice will appear, whenever Ivconic because you make a lot of Coils you can make some trial on this and say us the results
regards

[ivconic]

The Tesoro coil that I tested is a 9x8 web coil, and it's reasonably new. The residual voltage is actually 13mVpp. I also tested a Laser concentric coil (same as Cortes) and this is 100mVpp! Both of them work absolutely fine. Unfortunately I cannot test my 10" concentric coil, as a wire is currently broken in the plug. Just need to get enough enthusiasm to re-attach it.
I do not own a tesoro widescan DD coil.

Personally I think you're taking this zeroing thing too seriously. The only difference you're likely to notice, if the residual voltage is too high, is worse discrimination for deep targets.

I'm sure that statement is going to spark off some interesting discussions.

Hah,hah,hah,hah,hah !!!!!! I KNEW IT!!!!!
Damn! Now i cant sleep whole night!


Also...now....NOTHING CLEAR TO ME!
I dont understand this at all!?

Quiaozhi, tell me than; how the heck i will achieve good, stabile GEB???
If this is not dependable of good coil design....well..????
And i thought i know something!?
Regards!

[Qiaozhi]

Now, this is strange? Also could be very interesting! Could be solution to my last problem i havent solved; mentioned ocassionall bleeps. Although those are very,very rare with my latest coils...Maybe Tesoro need coil not nulled absolute to zero? Is it possible?
Also other explanation is possible; coil which gave you 15mV might be old,many times used and a bit damaged, unballanced due pass of time,moisture and relative movements so far....Could this be a case?
I need to be perfectly sure about that! Can you measure another Tesoro coil?

The Tesoro coil that I tested is a 9x8 web coil, and it's reasonably new. The residual voltage is actually 13mVpp. I also tested a Laser concentric coil (same as Cortes) and this is 100mVpp! Both of them work absolutely fine. Unfortunately I cannot test my 10" concentric coil, as a wire is currently broken in the plug. Just need to get enough enthusiasm to re-attach it.
I do not own a tesoro widescan DD coil.

Personally I think you're taking this zeroing thing too seriously. The only difference you're likely to notice, if the residual voltage is too high, is worse discrimination for deep targets.

I'm sure that statement is going to spark off some interesting discussions.

[ivconic]

"But my guesses are related to the RX part... would a very good null at 10KHz be very good at 14KHz too ??? How e.g. resonance will play a role disturbing our results ?"

Very good question! I dont think so. I noticed differences when tested coils with different oscillators. Classic III oscillator runs at 6.5KHz (my handmade) and testing with it showed some induction more than i expected...Later when tested wit TGSL...no induction.
I think i pictured this...but not awared of whole problematic so i rather not speak yet....dont want to talk rubbish here...Heh,heh,heh...

Regards!

[Max]

Hi Ivica.
10mv is to high and it is too low. We measure at the Rx coil and we give a signal at the Tx coil. So if you connect the Tx coil to the Anker SS60 transmitter (the output is about 140VP-P) then it is very dificult to take an output (Rx signal) of only 5...10 mv. So i think that it is better to have a standrd Tx signal so to can compare the null of any coil and with a standard method (multimeter.... oscilloscope.... parallel resistor... etc)
Regards

Hi Geo,
absolutely right: I suggest we use a 10Vpp oscillator as reference for Rx nulling readings.

If I remember well the BandidoII osc is like this... but runs at 10KHz with standard coil...

So we must also consider a starndard frequency where to test RX response ?

I think that this isn't an absolute need... we can tune coil at 10KHz and then use it e.g. at 14 like in TGS... but will this introduce some additional variation ?

In theory not. If you consider that TGS mount exactly the same coils of e.g. Bandido there aren't problems tuning coil at 10KHz or at 14KHz... and then using them at different TX frequency.

But my guesses are related to the RX part... would a very good null at 10KHz be very good at 14KHz too ??? How e.g. resonance will play a role disturbing our results ?

Anyway... the solution could be using a modified TGS osc with controlled 10Vpp wave at TX... we need just to change voltage supply and remove voltage feedback for stabilization !

Kind regards,
Max

[Max]

One more thing I have try testing with different ithems but its look like my device can't dicriminate rusty horseshoes and etc. only some IRON ont all!
What could be the reason for this?\

Hi,
about OO coils they are really hard to tune at best...so I'm not surprised you get it work but with less depth of e.g. DD.

Also I get good results ONLY on bandidoII-clone... that has modifications by me respect to original Tesoro's design. On TGS I cannot get good results with my "good" OO... and I'm still testing for something better e.g. alignment...so far with no real success.

About horseshoes...yes... Disc works just with small iron... ironwires...and other small trash like nails etc...

When you are near a large iron with your coil... the perturbations it introduces e.g. in fieldlines distribution in the space make you nulled config disappear and so the out-of-balance make your detector sound even if phase is far negative; phase at RX is also a function of target shape, dimensions, soil sorrounding it, orientation etc not only metal kind... so you'll get great phase variations respect to ideal one when a large thing of iron is near coil and that sums with the out-of-null you get from field distortion.

Happen with all VLF detectors expecially with large iron-stuff at few distance from coil bottom.

I can e.g. detect an big iron pole (height 2meters from +50cm away with TGS) : it's normal not having a TOTAL iron disc.

That's ok... you'll dig some horseshoes... fine... but keep attention at sound: you'll notice cracked sound with iron most of the times. Ear can distinguish trash from good target that way... but then you have to trust your ears.

Kind regards,
Max

[Geo]

"5...10mV", is to high! I already made simillar coils in the past. Those were times when i made Classic III. That's why am i talking about good nulling. I had a lot of problems with CL3 and bad nulled coils. A lot of false signals on any kind of soil, everywhere. Totally unuseable.Couldnt distinguish among detection and false signal at all.
Other side, yes i agree, coil can be useable even if it is not nulled to absolute zero. Up to 1mV it can be used with detector like TGS, cose TGS have very good soluted GEB&DISC stage so any eventual loss coming from coil can be easilly compensated inside apparatus. It is not special to null coils good, only need patience and good will...I already explained my method. Usually takes 20 minutes to null coil (due time needed to chill out melted plastic)...

But Quiaozhi confused me with previous claim...!? Wait for his reply on this...

Hi Ivica.
10mv is to high and it is too low. We measure at the Rx coil and we give a signal at the Tx coil. So if you connect the Tx coil to the Anker SS60 transmitter (the output is about 140VP-P) then it is very dificult to take an output (Rx signal) of only 5...10 mv. So i think that it is better to have a standrd Tx signal so to can compare the null of any coil and with a standard method (multimeter.... oscilloscope.... parallel resistor... etc)
Regards