TESORO GOLDEN SABRE

[simonbaker]

Could the method described in this patent http://www.geotech1.com/pages/metdet.../US4486713.pdf be same as TGS uses?

It's pretty close, although the TGS does not directly compare amplitudes of the two detected channels in a single comparator like that. Rather, it uses two separate comparators, each having a reference threshold, and the two outputs are logically "AND"ed together, creating a detection logic circuit where each channel must detect the target independently for final audible detection. Also, the patent diagram has an "Absolute Value" block in each channel which I'm not sure of the function, but does not exist in the TGS.

I don't pretend to really understand the details of this patent (or others), the details are hard for me to be really clear on without exhaustive study, and patent writers always try to hedge their ideas to include myriad variations by implication.

Regards,

-SB

[habitbraker]

Could the method described in this patent http://www.geotech1.com/pages/metdet.../US4486713.pdf be same as TGS uses?

[habitbraker]

Thanks Gwil!
This is the explanation I needed!!
Now it makes sense

[Gwil]

LM393 Output

Hi Habitbraker

The LM393 doesn't have the same kind of output circuit as an op amp, which can swing close to the positive and negative supply rails. Its output is the collector of an npn transistor whose emitter is connected to the negative rail - an "open collector" output. Therefore the output is either pulled down to the neg. rail or open circuit, when it is pulled up by whatever is connected to it. This feature is shown in the LM393 data sheet.

For all sorts of semiconductor data sheets, try http://www.digchip.com/

Gwil

[habitbraker]

Thaks for response!
But that was my question - why output voltage of com. doesnt go above ground?

In other words - voltage on inputs of comparators is bipolar(Tx signal). Then why the output voltage is only below ground instead of -5 <--0--> +8 ??

[simonbaker]

Hello!
I have one question regarding to operation of tgsl.
Why isnt there full voltage swing (from +8 to -5V) on the outputs of comparators(U102a/b)?
But instead of that only negative referenced to ground?
I understand that it is neeeded for n-chan jfet to work corectly(not to forward bias).
But input to comparator is +/- isnt it?

Hi habitbraker:

It seems you answered your own question well I think. Output does not go above ground so the JFets work correctly.

Maybe the answer is that the LM393 comparator requires a "pull-up" voltage for the output which is different from the supply and input voltages? So we can make the max output voltage anything we want.

The purpose is simply to make a square waves with voltage range from V- to ground from the TX sine wave input. We also put in some phase shifts with the discrimination pot and the ground balance pot before making the square waves.

Regards,

-SB

[simonbaker]

Hi Simon ,

In this YouTube link : http://www.youtube.com/watch?v=TVrISTKtxAw

I've recorded some of the 'coil nulling ' : Tx vs. Rx on output of LF353 .

With this set of coils ( shielded ) I don't get any closer as 35-40 degrees ...

For the detection in All Metal Mode the shift doesn't seem to care that much...

I didn't experiment that much in Disc mode or Notch Mode because it didn't discriminate that much so I had my (wrong ? ) focus in the All Metal


best regards,

Dennis the Mennis

Quite excellent DtM! Nice way to show your new scope.

I'm interested also in the LF353 output that drives the sync pulse, what does it look like?

I get those little spikes on the RX signal also. I suspect it is caused by the negative power rail circuitry, but not sure.

Your TX signal looks more sinusoidal than mine. Do you have the JFet in the oscillator or not?

You mentioned Notch -- did you build the full TGS circuit with the notch and battery check and everything?

I think next you should start using the discriminator mode -- that's where the fun begins!

Regards,

-SB

[habitbraker]

Hello!
I have one question regarding to operation of tgsl.
Why isnt there full voltage swing (from +8 to -5V) on the outputs of comparators(U102a/b)?
But instead of that only negative referenced to ground?
I understand that it is neeeded for n-chan jfet to work corectly(not to forward bias).
But input to comparator is +/- isnt it?

[Dennis the Mennis]

Output LF353

Hi DtM -

Thanks for extra info.

W and M is a nice way to explain the detection waveform due to coil polarity. When coil is backwards, you can see why we can get a double-beep easily.

I was wondering if you have display options for displaying Multimeter type of measurements at the same time as the graph - for example, the RMS and peak voltage values?

Can you show us the outputs of your LF353 for fun?

Cheers,

-SB

Hi Simon ,

In this YouTube link : http://www.youtube.com/watch?v=TVrISTKtxAw

I've recorded some of the 'coil nulling ' : Tx vs. Rx on output of LF353 .

With this set of coils ( shielded ) I don't get any closer as 35-40 degrees ...

For the detection in All Metal Mode the shift doesn't seem to care that much...

I didn't experiment that much in Disc mode or Notch Mode because it didn't discriminate that much so I had my (wrong ? ) focus in the All Metal


best regards,

Dennis the Mennis

[simonbaker]

Hi DtM -

Thanks for extra info.

W and M is a nice way to explain the detection waveform due to coil polarity. When coil is backwards, you can see why we can get a double-beep easily.

I was wondering if you have display options for displaying Multimeter type of measurements at the same time as the graph - for example, the RMS and peak voltage values?

Can you show us the outputs of your LF353 for fun?

Cheers,

-SB


Hi -SB ,

The USB scope I'm using ( for my job ) is a PicoScope 3205

http://www.picotech.com/picoscope-specifications.html

( and no , I'm not a shareholder of PicoTech )

The software I'm using is version 6 and can be downloaded :

http://www.picotech.com/picoscope-oscilloscope-software.html

If I'm right this version is used for all type of scopes and this also the version I'm

using .

When downloaded you've got full functionality , no license key or what so ever needed ,

only it won't recognize the USB device so it will ask if you want to run it in ' demo ' mode

and there you can play around with several pre installed measurements and cases .



I still have to get used to the controls , If I may say so I could 'dream' the ordinary scope ... I guess time will do rest



I think the min voltage is currently 20 mV/div but you can zoom in a lot , haven't tried

that yet.

The max voltage is on both input and ext triggering 20 V ( big disadvantage because

collegeas blew up the input , they used during measurements with a pick up coil near a

transmittercoil : big induction , bye bye input )

If you don't know what voltage to expect use a x100 probe ?

The measurements you can do are to many to tell , RMS ? yes no problem

Try the demo mode , the two( or more ) inputs can be displayed in several modes : in scope / analyzer / together / separate etc etc






No , I'm not sure , in fact I used the wrong graph overthere , sharp observation though , the zoom-in graph is not correct there .

But is shows prefectly why the TGS(L) detects a coin 'twice' when it passes the coil one time and the leads are twisted or the coil is flipped compared to 'normal ' operation :

In normal use the Target Signal has the W shape : first negative and the positive where it triggers the positive alarm treshhold of +/- 50/60 mv (?) once.

When the coil is flipped or the leads are reversed the Target Signal seems to have a M shape : First positive ( also first alarm on positive alarm treshhold ) , next negative sinus ( no alarm) , last of M shape Target Signal is positive again ( and alarms for the second time ? ) see attachment


hope to post more findings soon

best regards ,

Dennis the Mennis[/quote]

[Dennis the Mennis]

[/quote]

Hi DtM!

Very nice graphs - it makes me very interested in a USB scope.

One question - in your PDF, you show a graph labeled:

Next you show a graph labeled

However, the close-up graph does not look like the same signal -- it does not seem to go negative first when detecting the coin. Are you sure you have the correct graph there?

Can you tell us some more about your USB scope? For example, is it easy to display other measurements along with the signal, such as RMS value of the sample, Max voltage, Min voltage, etc?

Thanks for this data. It gives a good example of what to expect with the RX coil shorted and also the typical noise in someone's workshop. We can compare our typical noise levels and perhaps help explain why we obtain such different air tests in our workshops!

Nice work!

-SB

Hi DtM!

Very nice graphs - it makes me very interested in a USB scope.

Hi -SB ,

The USB scope I'm using ( for my job ) is a PicoScope 3205

http://www.picotech.com/picoscope-specifications.html

( and no , I'm not a shareholder of PicoTech )

The software I'm using is version 6 and can be downloaded :

http://www.picotech.com/picoscope-oscilloscope-software.html

If I'm right this version is used for all type of scopes and this also the version I'm

using .

When downloaded you've got full functionality , no license key or what so ever needed ,

only it won't recognize the USB device so it will ask if you want to run it in ' demo ' mode

and there you can play around with several pre installed measurements and cases .

Can you tell us some more about your USB scope? For example, is it easy to display other measurements along with the signal, such as RMS value of the sample, Max voltage, Min voltage, etc

I still have to get used to the controls , If I may say so I could 'dream' the ordinary scope ... I guess time will do rest

Can you tell us some more about your USB scope? For example, is it easy to display other measurements along with the signal, such as RMS value of the sample, Max voltage, Min voltage, etc?

I think the min voltage is currently 20 mV/div but you can zoom in a lot , haven't tried

that yet.

The max voltage is on both input and ext triggering 20 V ( big disadvantage because

collegeas blew up the input , they used during measurements with a pick up coil near a

transmittercoil : big induction , bye bye input )

If you don't know what voltage to expect use a x100 probe ?

The measurements you can do are to many to tell , RMS ? yes no problem

Try the demo mode , the two( or more ) inputs can be displayed in several modes : in scope / analyzer / together / separate etc etc

One question - in your PDF, you show a graph labeled:

Next you show a graph labeled

However, the close-up graph does not look like the same signal -- it does not seem to go negative first when detecting the coin. Are you sure you have the correct graph there?

No , I'm not sure , in fact I used the wrong graph overthere , sharp observation though , the zoom-in graph is not correct there .

But is shows prefectly why the TGS(L) detects a coin 'twice' when it passes the coil one time and the leads are twisted or the coil is flipped compared to 'normal ' operation :

In normal use the Target Signal has the W shape : first negative and the positive where it triggers the positive alarm treshhold of +/- 50/60 mv (?) once.

When the coil is flipped or the leads are reversed the Target Signal seems to have a M shape : First positive ( also first alarm on positive alarm treshhold ) , next negative sinus ( no alarm) , last of M shape Target Signal is positive again ( and alarms for the second time ? ) see attachment


hope to post more findings soon

best regards ,

Dennis the Mennis

[simonbaker]

Hi ,

This is a small impression / test of the Target Signal and Noise Signal of the TGS measured on the output of the LM308 with an USB scope .
With this USB scope I'm using right now these signals can be made more visible compared to my old scope.

A man once asked on this forum if the Target Signal changes if the Rx coil ( or its leads ) is reversed : the answer in attached PDF.

It seems that this man is using his coils nowadays to hold his papers on his desktop .

See the PDF for all pictures !

Whenever I find time I will post some pictures of the Spectrum Analyzer under workshop conditions ... very interresting.


kind regards

Dennis the Mennis

Hi DtM!

Very nice graphs - it makes me very interested in a USB scope.

One question - in your PDF, you show a graph labeled:

"Another Target Signal : A 1 euro coin passes the coil two times on 10-15 cm height , first part of signal is negative"

Next you show a graph labeled

"Close up / zoom in of Target Signal"

However, the close-up graph does not look like the same signal -- it does not seem to go negative first when detecting the coin. Are you sure you have the correct graph there?

Can you tell us some more about your USB scope? For example, is it easy to display other measurements along with the signal, such as RMS value of the sample, Max voltage, Min voltage, etc?

Thanks for this data. It gives a good example of what to expect with the RX coil shorted and also the typical noise in someone's workshop. We can compare our typical noise levels and perhaps help explain why we obtain such different air tests in our workshops!

Nice work!

-SB

[Dennis the Mennis]

Target Signal and Noise Signal on TGS made more visible

Hi ,

This is a small impression / test of the Target Signal and Noise Signal of the TGS measured on the output of the LM308 with an USB scope .
With this USB scope I'm using right now these signals can be made more visible compared to my old scope.

A man once asked on this forum if the Target Signal changes if the Rx coil ( or its leads ) is reversed : the answer in attached PDF.

It seems that this man is using his coils nowadays to hold his papers on his desktop .

See the PDF for all pictures !

Whenever I find time I will post some pictures of the Spectrum Analyzer under workshop conditions ... very interresting.


kind regards

Dennis the Mennis

[eduardo1979]

Thanks José!

[Jose]

Good job, Eduardo
At TGS, I experiment with the design of coils Ivconic.
I changed the TX coil with fewer turns and larger diameter cable.
Stage TX, also modified for more power.
Significantly improves sensitivity, but I could not reliza a proper balance in the RX signal.
Maybe someone will find a solution that would be the ideal.
Jose