TGSL Tuning & troubleshooting

[jroux]

Hello, I just read the 53 pages of this post and I have a similar problem with my kit purchased from Siverdog with TL071, my coil works fine on another detector, my RX and TX 6.55 mh 6 02 mh which seems correct,i use Belden cable 8723, I detect a 1 euro to 4 cm in all metal or disk, sometimes the detector starts ringing continuously alone, I tried to change the 15NF withe a 10nf still the same problem, I have a frequency meter and a multimeter, I am desperate because this is my third md I built and that gives me trouble, if someone could help me , thank you.

[Qiaozhi]

Hello, I just read the 53 pages of this post and I have a similar problem with my kit purchased from Siverdog with TL071, my coil works fine on another detector, my RX and TX 6.55 mh 6 02 mh which seems correct,i use Belden cable 8723, I detect a 1 euro to 4 cm in all metal or disk, sometimes the detector starts ringing continuously alone, I tried to change the 15NF withe a 10nf still the same problem, I have a frequency meter and a multimeter, I am desperate because this is my third md I built and that gives me trouble, if someone could help me , thank you.

The only real way to debug your problem is with an oscilloscope. With only a frequency meter and a multimeter, you are stabbing in the dark.

[jroux]

Ican't buy an oscilloscope much money for me, but i think we can troubleshoot with other solutions, i detect 1 euro from 10cm in all metal in disk mode 5 cm, the dicrimination doesn't work, i've 1.66mv on pin 7, in disk mode the md sometimes sounds just salt, and must be cut so that it works again, an idea can be ?

[paku]

Ican't buy an oscilloscope much money for me, but i think we can troubleshoot with other solutions, i detect 1 euro from 10cm in all metal in disk mode 5 cm, the dicrimination doesn't work, i've 1.66mv on pin 7, in disk mode the md sometimes sounds just salt, and must be cut so that it works again, an idea can be ?

Bro PM me I will Try To Help You

[zoomix]

Hello,

I would like to make a récapilatif my tests smd TGSL.

My coils are made with 0.28mm and I precisely these values:
6.5mh and 6mH

The capacitor is 21.8nf TX and RX 16NF. I have a frequency 14.61khz

I had not yet oscilloscope and I am the nulling coils in AC voltmeter, I have the best result 450mv AC pin 7 of U101

I get the 32 cm and 1 euro 5centime

Potentiometer in the middle of the GB and I eliminate a ferrite toroid. (I do not have a cylinder)
With these results, the steel is discirminé with the potentiometer to the minimum. So I do not have control of the metal with the discirmination but aluminum rings and can be discirminer.

Today I had a two oscilloscope traces and I checked the phase in this configuration and I get + - 60 ° phase shift. (RX behind Tx)

So I tried to set at 20 ° and I get better at 28 ° which gives me 34mV AC on pin 7 of U101.

With this setting, the steel is discriminated with the potentiometer discir +-in the middle .. I think the result is better. I keep + - 30cm 1 euro and 5cent.

Here is a picture + - 28 °

a beginner question: The RX voltage is not symetric (-75 +130mv)

Do I have décallé track of RX to TX or focus on is what I have to align the 0V lines?

Also I noticed that I did not -5volts but -6.2volts?

I ask these questions because I'd like your advice before filling the coils.

For the moment these tests are done with the coils off lid but I think I can improve
the nulling with fit snugly in the receptacle coils.

Thank you for your advice

[Davor]

Voltages are fine as they are. TGSL is a motion compensated detector, and as such it discards all offsets of static nature.
Try improving the Tx signal, as apparently there is something causing impurities.

[jroux]

It's possible to have an image of this error on pcb.
Regards.

[jroux]

an image of this error possible

[zoomix]

Hello,

Information for the image of my previous message is a complete measurement error.

I remade test the result:

I have very difficult or impossible to achieve 20 ° phase shift while maintaining the correct amplitude RX

When I am approaching this value, the performance drop.

I get very good results with ± 450mv AC on pin 7 of U101 but the phase shift is 60 °. GEB pot in the middle and functional discrimination.

With these settings the voltmeter and ferrite I 5cent euro 32 cm 34cm ditto 1 euro and 1napoléon to 35cm

I tested with different capacitor RX and TX but the variation is too low.

The lateral displacement of TX and RX low there is the amplitude varies.

My coils are of 0.28mm 6.5mh specifically for RX and TX for 6mH.

Will you tell me why I do not reaches the 20 ° phase shift?


Thanks

[jroux]

Many thanks, i've the same problem 460mv on pin7 but the depht is maximum 15cm, geb to 40% impossible to discriminate ferrite.

[zoomix]

Hello JROUX ,

I measure AC on pin 7 of U101 in DC because I at least 30mv while in AC I have more precision.

I think your problem is different . Do you have a coil position or you get good results ?

To find this position without oscilloscope , put the overlooper coil raises the potentiometer in the middle and discir GB minimum .

Then passes a piece of ferite above the coil by reducing the space until there was no more sound. To do this operation , a fixed spool with tape and moves the other coil .

Did you put the pcb in a shielded box ? What is the diameter of your coil wire ? Is that all masses are connected (cable , shielded case) ?

It is very difficult to read 54 pages of thread discution this post . Responses are rare because this project a few years already .
I see that some people have had very good results with improvements and other " animate " the debate with theories without ever having made the TGSL ...

I try to find a summary of the latest changes tested and validated but I difficult to gather the right information.

My discrimination works , the ground balance too. I take a piece of 5cent euro 30 cm and 1 napoleon 35 - 36cm discirminant a piece of aluminum with potetiometre in the middle . all without the 20 ° phase shift but + -60 ° with 450mv AC output pin 7 . (All these tests in the air cavity . )

The sensitivity does not parrait éfficace . The potentiometer does not cause a lot of variation.

My wire is 0.28mm and I 21.8nf on TX and RX on 16NF . frequency 14.61khz ( my parameters with the best results )

My fear is that it becomes unstable on land after completing the coil.

I especially want to know why my sensor operates without a phase shift of 20 ° and why it is impossible for me to reach 20 °.

I saw a post simonbaker speaking precisely the subject. I'll redo the coils with 0.25mm wire but I doubt it changes the phase .

If someone just give me a little insight into the reasons for the problem of phase separation and the limited SENSITIVITY , a big thank you to him ..

[simonbaker]

Your MD appears to work quite well. Does it matter what the phase shift on pin 7 is? Are you talking about the phase of the null signal?

No one has satisfied me that the actual phase shift of the null signal is very important to depth or discrimination by itself. However, it may have some importance regarding immunity to certain noise sources, such as amplitude modulation of the TX signal. It may also be a sign that you did not build your PCB or coil quite right.

But for basic MD operation, you can get a good MD with many different null phases, and your performance is not bad at all.

My tests have indicated that null phase can be very influenced by the cable you use to connect the coil to the PCB, and also the type of shield you use. If the gap in your shield is shorted or low resistance, it could cause a big phase shift, and you would want to correct that. Any metal near your coil or mounted on your coil could influence the null phase (without necessarily causing much harm).

If your TX frequency is too near the RX circuit resonant frequency, you can get an exaggerated phase shift -- but usually your discrimination and ground balance would be off-center and perhaps appear not to work. This problem is usually fixed by adjusting the capacitor in parallel with the RX coil to a lower value. If lowering the capacitor value does not change the null phase much, it probably was not wrong to start with.

It would be best to post actual photos showing the TX signal vs the RX signal at pin 7 as a dual trace, showing 2 or three cycles on the screen to show the phase relationship. Sometimes people have different ways of referring to the phase measurements.

Also, whenever measuring depth, the EMI noise around you affects the depth very much. Sometimes people think their TGSL is bad, but it is not. If you take the MD far away from noise (in the hills for example) and measure the depth there, it may be much better. That is my situation in fact. I often measure about 16 cm depth in my workshop, and 34 cm when miles away from my town in the hills. Such EMI should not affect the discrimination and ground balance much I think if you test close enough to your coil, so you can still test those functions in your workshop.

It is very useful to build many coils and compare them, because there can be surprising differences. Also try different cables.

Regards,

-SB

Hello,

Information for the image of my previous message is a complete measurement error.

I remade test the result:

I have very difficult or impossible to achieve 20 ° phase shift while maintaining the correct amplitude RX

When I am approaching this value, the performance drop.

I get very good results with ± 450mv AC on pin 7 of U101 but the phase shift is 60 °. GEB pot in the middle and functional discrimination.

With these settings the voltmeter and ferrite I 5cent euro 32 cm 34cm ditto 1 euro and 1napoléon to 35cm

I tested with different capacitor RX and TX but the variation is too low.

The lateral displacement of TX and RX low there is the amplitude varies.

My coils are of 0.28mm 6.5mh specifically for RX and TX for 6mH.

Will you tell me why I do not reaches the 20 ° phase shift?
[ATTACH]26139[/ATTACH]

Thanks

[jroux]

Does not work properly for the momment, I expect my tesoro coil I just buy cheap occasion, my pcb is in a plastic box, wire is 30awg, all masses connected, the best result is 15cm for one euro.
My kit comes from silverdog, thank you for your help.

[golfnut]

Try a different value capacitor on the Rx coil.

Tuning this will have a good effect on sens if done well.


s

[jroux]

Not work with original Tesoro concentric coil 5 pin, 7" , 3 cm for a coin, i try all possibilities with wires of tx and rx with different capacitors, I think a component is dead, how to find it.