TESORO GOLDEN SABRE

[Dennis the Mennis]

Dicrimination made visible

Hi habitbraker -

Normally you don't center DISC sync pulse on zero crossing of null signal.

DISC sync pulse is typically centered on zero crossing of target signal you want to reject. However, we normally never see the target signals, they are too small! We just trust they are there and that they have a certain phase relative to the TX signal.

Yes, they are very , very small .
This graph is made on C12 direct output of the FET vs the TX signal, but I had to drag the coin over the coils to see this respons :



I think it is (almost) impossible to see phase relative to the Tx , can't see which one triggers the target signal .

Someone any idea for this : I can ad a mathimatical channel to the scope ( for example A + sqr B ) ?

This next graphs show the results on the outputs of the LM308's
( Green = U105 GB channel and Red = U104 DISC channel)

The phase of the target signals is completely independent of the null signal phase. You can change your null signal phase by balancing your coils, but it does not affect the target signal phases.

So if iron has phase1 and aluminum has phase2, we use DISC pot to center the sync pulse over the iron signal zero crossing. Then iron will have zero response and other metals with greater phase shift will make positive pulse.

Detection on piece of Al foil in AM mode :

If we use DISC pot to center the sync pulse over the aluminum signal zero crossing, then metals with phase less than aluminum will make a negative response, and metals with phase greater than aluminum will make a positive response.

A nickel Disced out / rejected or 'nulled'





Same nickel at the point to be rejected :





Steel Bottle Cap rejected (as replacement 2 steel washers ) :





Iron discriminated out :

In the GB channel, it should work out that all metals make a positive response if you set it correctly.
As you know, only if both channels make a positive response will we hear audio beep.

Detection of an 1 Eurocoin in AM mode :

In the ground channel, we set the pot so the null signal makes a zero response ideally if it is modulated by the ground. Then we don't care too much if the null signal makes a positive response or not in the DISC channel. However, I believe the ground can still modulate the DISC channel response in some cases, which is worth thinking about.

Ferrite slug / bar 10 cm with GB pot set correctly in am mode

That's my basic take on it. There are some subtle questions on how much you can control the "true" null signal phase during your nulling to optimize your MD.

Regards,

-SB

Just a small graphical support for anyone's interrest

kind regards ,

Danny

Dennis the Mennis

[taliesin]

can someone help please?

i've just built tgsl v 3.on most of the tesoro designs (that i've anyhow) have the MR pin 2 on 4024 tied down to ground.on the v3 there's a 1k resistor from pin 2 via one of the fet's .is this paramount to the design or am i able to just lift the 1k.?

[habitbraker]

Thanks SB!

[simonbaker]

Actually i think I finally get it. It is like that, I checked. At exact point where foil is disced out Vout of SD doesnt change when i wave foil over the coil- explines zero cross.

If disc is increased, Vout starts to decrease when I wave it and when decreased below al zero cross point, Vout starts to go up.

And because of AC coupled amplifiers only changes are detecded...

Yes!

-SB

[habitbraker]

Actually i think I finally get it. It is like that, I checked. At exact point where foil is disced out Vout of SD doesnt change when i wave foil over the coil- explines zero cross.

If disc is increased, Vout starts to decrease when I wave it and when decreased below al zero cross point, Vout starts to go up.

And because of AC coupled amplifiers only changes are detecded...

[habitbraker]

Thanks for the answer!
Yes I understand further signal processing (filters, opamp-AND etc)

But why received signal changes phase and A when I move metal objects near the coil?
Isnt those target signals which I can see?
So more conductive metal - more phase shift. If i disc out foil, i just center disc sample on al foil signal zerro crossing? So in that case al foil will have zero response and iron negative?

Is there some proof what i can see on scope?

[simonbaker]

Long time i thought that i understand how discrimination on tgsl works, but something doest match with my waveforms and my theory. Before I play with modifications I must understand it.

So what i know (at least I think I know) that geb sample is taken at zerocrossing so if changes only amplitude - Vout of SD is zero. Only metals make phase shift so for all metals Vout is positive. Ok in real life ground makes phi too (As Dfbowers describes)

But now disc. Something I dont get. I can sample it at zerocross too and at same time disc out iron.
For example Alu foil makes both positive and negative Vout of SD depending of where disc sample is taken...
What makes it(Vout of disc SD) go negative , positive? How phase shift of metals affect it ? Is it amplitude changes as well?
Maby some examples? Its hard to see exactly it on scope.

Hi habitbraker -

Normally you don't center DISC sync pulse on zero crossing of null signal.

DISC sync pulse is typically centered on zero crossing of target signal you want to reject. However, we normally never see the target signals, they are too small! We just trust they are there and that they have a certain phase relative to the TX signal. The phase of the target signals is completely independent of the null signal phase. You can change your null signal phase by balancing your coils, but it does not affect the target signal phases.

So if iron has phase1 and aluminum has phase2, we use DISC pot to center the sync pulse over the iron signal zero crossing. Then iron will have zero response and other metals with greater phase shift will make positive pulse.

If we use DISC pot to center the sync pulse over the aluminum signal zero crossing, then metals with phase less than aluminum will make a negative response, and metals with phase greater than aluminum will make a positive response.

In the GB channel, it should work out that all metals make a positive response if you set it correctly.

As you know, only if both channels make a positive response will we hear audio beep.

In the ground channel, we set the pot so the null signal makes a zero response ideally if it is modulated by the ground. Then we don't care too much if the null signal makes a positive response or not in the DISC channel. However, I believe the ground can still modulate the DISC channel response in some cases, which is worth thinking about.

That's my basic take on it. There are some subtle questions on how much you can control the "true" null signal phase during your nulling to optimize your MD.

Regards,

-SB

[habitbraker]

tgsl discrimination

Long time i thought that i understand how discrimination on tgsl works, but something doest match with my waveforms and my theory. Before I play with modifications I must understand it.

So what i know (at least I think I know) that geb sample is taken at zerocrossing so if changes only amplitude - Vout of SD is zero. Only metals make phase shift so for all metals Vout is positive. Ok in real life ground makes phi too (As Dfbowers describes)

But now disc. Something I dont get. I can sample it at zerocross too and at same time disc out iron.
For example Alu foil makes both positive and negative Vout of SD depending of where disc sample is taken...
What makes it(Vout of disc SD) go negative , positive? How phase shift of metals affect it ? Is it amplitude changes as well?
Maby some examples? Its hard to see exactly it on scope.

[taliesin]

just built tgsl .got constant tone

i have a constant tone i've nulled the coils at the lowest volume? it happens in the garden as well.when i lift the jumper circled (signal from 4024 ) in edwardo's pcb it stops and i have nothing at all. any ideas ?

[simonbaker]

opps sorry I got confused

I think you are on the right track, and we all will learn from your experience too. Keep us posted how it goes.

Regards,

-SB

[satdaveuk]

COIL

Hi Dave:

I believe the TGSL that dfbowers loaned me has a TX coil inductance of about 6.4 and RX coil inductance of about 6.9, and it works great, so you can see it can vary somewhat. It seems the most important thing is to have about .5 mH more in your RX coil than your TX coil, with the TX coil inductance near 6 mH.

But no guarantees, coils have been known to be subtle and some poeple claim they have good and bad coils with similar inductances. However, I think if you follow the basic practices for building the coils (like Ivconic, dfbowers, Max, etc.) you should be fine.

On the other hand, it really seems to be a good idea to make several coils so you can experiment and have a better chance of finding and fixing any mistakes.

No one I know of has really experimented with how far off you can go with coils and compensate with capacitors -- I'd like to try it, but it requires a good basement and coiling forming equipment and whirly-doodle target sweeper .

I can't remember if shield increases or decreases inductance... anyone? I think it is a minor effect compared to compressing the coil when you wrap it.

I'm not sure what you mean by capacitance between coil connections... maybe self-resonance of coil due to parasitic capacitance? (you have nice test equipment if so). In any case, I don't think it matters at all, very minor effect I imagine.

I don't think it's worth checking "self-resonant" frequency of coils since it is not the dominant resonance once the coil is in-circuit. Having said that, if two coils act differently even though they have the same resistance and inductance, it would be worth investigating.

If you lived next door we'd be in business because I tried to order five 2n2907's and by mistake ordered 5 bags of them.... but I'm short of 2N2222's....

I'm not the master builder of the TGSL, although I study the circuit a lot. For real building tips, dfbowers is a good one to check with for recent experience, and Ivconic's notes preserved by Qiaozhi in the Projects section.

When it comes to evaluating your TGSL, I've come to the conclusion that it is really important to get far away from all EMI to test maximum air depth. I have to drive 15 minutes into the mountains before dfbowers TGSL registers its maximum air depth.

Also, no one has really tabulated air-depth vs. ground-depth for standard targets with different quality models to see how important it is to get 30 cm air depth vs 20 cm air depth, for example. But we still like to use air depth as a way to evaluate whether we have built our TGSL similar to others.

It's kind of an ongoing project to get your MD working nicely, so allow for some troubleshooting experience. If it works great the first time, nice going.

Regards,
Hi Simon
I got the 2n4444A from a company called Phenoptix on Ebay they were £1.79 for 10 incl p/p, They do post world wide and i recomend them due to the fact they dispatch same day and give follow up on delivery.
I think now after chat with you I shall carry on making this coil and logging measurements on the way as I go through the process for fucture references.
I got a very large reel of magnet wire so I will make many coils and pair up the best matches like often manufactures do when pairing up high end output transistors and valves.
I dont think there is a accurate sience to getting these coils identical everytime because even the room temp when making these coils will make a difference ie exspand and contract of the actual magnet wire and variations follow from there.
You know Simon when I purchase top end graphed LNBs from a manufacture for work, they are not manufactured top end, its just they are picked from a batch of say a thousand for there low noise figures etc, the rest of them go out as standard at half the price.
It also applies to matching pairs of transistors and valves in top end HI- FI equipment.
I think if we were nabours we would be in and out of each others workshops .
I will let you know my results
All the best
Dave

-SB

opps sorry I got confused

[dfbowers]

Hello guys.
Want to know which components if any have to be matched or of high quality for Tesoro detector circuits.
I thought I read in here some time back that one of the caps needs to be a certain type or quality. Which ones??
How critical is matching the resistors and caps in the filter stages etc. ---will it still work if I do not match the components in the two different channels? Or just work better?
Do plain old cheap ceramic caps work?
I ask all this becuase occassionally we do not have the best quality componants on hand and I'd like to know if we need to order and wait for parts before building the circuit.
PHIL

I used plain old ceramic caps for filter and bypass caps, which are not critical at all. Other places you might want to consider polystyrene or polyester. For example, C1,C2 and C6 should be stable over time.

Then, be aware of where electolytics should be used.

As far as matching resistors, probably not that critical either.

I don't know how "correct" my selection is, but you can download the parts list here:

http://www.geotech1.com/forums/showthread.php?t=15710

I have built 3 so far without problems using this parts list.

You might want to see if Silverdog (Andy) can send you his parts selection as well.

Don

[turtlebowl]

Critical component values

Hello guys.
Want to know which components if any have to be matched or of high quality for Tesoro detector circuits.
I thought I read in here some time back that one of the caps needs to be a certain type or quality. Which ones??
How critical is matching the resistors and caps in the filter stages etc. ---will it still work if I do not match the components in the two different channels? Or just work better?
Do plain old cheap ceramic caps work?
I ask all this becuase occassionally we do not have the best quality componants on hand and I'd like to know if we need to order and wait for parts before building the circuit.
PHIL

[simonbaker]

Thanks for getting back to me Simon
In your opinion what would be the highest inductance you would go up to before going any further with this coil ie shielding etc.Am I right thinking that the conductvity will go up when adding a shield or is it just the tightning of the coils that does it? so in a nutshell does the shield its self either up the inductance or lower it, i thought the latter but i maybe wrong. As you said it can be offset with changing capacitor values but there must be a limit, and I dont want to end up with a machine not working to its full spec, I want to end up with the same results as the rest of you.
Also off hand what sort of capacitance reading should I exspect between the coil connections, before I lumed the coil it was reading about 27pf but after luming it jumped to around 70pf or more, but that maybe due to the batterys going flat on my meter and I was very tired.
My circuit is almost compleat, just waiting for the 2n2907 to be delivered from china.
By the way simon can I check the frequency of these coils without the pcb?
I have a very good signal generator with sweep plus 2 very good frequency counters plus a rather old but working dual beam scope.
If its possable I would be greatful if you can tell me in detail what to do because this stuff is new to me.
Many many thanks
Dave

Hi Dave:

I believe the TGSL that dfbowers loaned me has a TX coil inductance of about 6.4 and RX coil inductance of about 6.9, and it works great, so you can see it can vary somewhat. It seems the most important thing is to have about .5 mH more in your RX coil than your TX coil, with the TX coil inductance near 6 mH.

But no guarantees, coils have been known to be subtle and some poeple claim they have good and bad coils with similar inductances. However, I think if you follow the basic practices for building the coils (like Ivconic, dfbowers, Max, etc.) you should be fine.

On the other hand, it really seems to be a good idea to make several coils so you can experiment and have a better chance of finding and fixing any mistakes.

No one I know of has really experimented with how far off you can go with coils and compensate with capacitors -- I'd like to try it, but it requires a good basement and coiling forming equipment and whirly-doodle target sweeper .

I can't remember if shield increases or decreases inductance... anyone? I think it is a minor effect compared to compressing the coil when you wrap it.

I'm not sure what you mean by capacitance between coil connections... maybe self-resonance of coil due to parasitic capacitance? (you have nice test equipment if so). In any case, I don't think it matters at all, very minor effect I imagine.

I don't think it's worth checking "self-resonant" frequency of coils since it is not the dominant resonance once the coil is in-circuit. Having said that, if two coils act differently even though they have the same resistance and inductance, it would be worth investigating.

If you lived next door we'd be in business because I tried to order five 2n2907's and by mistake ordered 5 bags of them.... but I'm short of 2N2222's....

I'm not the master builder of the TGSL, although I study the circuit a lot. For real building tips, dfbowers is a good one to check with for recent experience, and Ivconic's notes preserved by Qiaozhi in the Projects section.

When it comes to evaluating your TGSL, I've come to the conclusion that it is really important to get far away from all EMI to test maximum air depth. I have to drive 15 minutes into the mountains before dfbowers TGSL registers its maximum air depth.

Also, no one has really tabulated air-depth vs. ground-depth for standard targets with different quality models to see how important it is to get 30 cm air depth vs 20 cm air depth, for example. But we still like to use air depth as a way to evaluate whether we have built our TGSL similar to others.

It's kind of an ongoing project to get your MD working nicely, so allow for some troubleshooting experience. If it works great the first time, nice going.

Regards,

-SB

[Qiaozhi]

Hi everyone, i want to build this metal detector but.... i don't know from where i shoud read this thread, from the start or from the end. This thread is full of modifications.
Can someone make a new thread or PDF file with the best (but stabile) version of detector and with all modifications PLEASE !!!!

Greeting from CROATIA !!!

All the information is here ->

http://www.geotech1.com/forums/showthread.php?t=15710