Question about tgsl

[waltr]

Oki this what I did, first I removed some turns from Tx because the inductance wasn't correct, then while I was measuring U101a-pin7 I gradually removed turns from bucking coil until I reached 15mv ~ ,but I don't know why Rx waveform looks ugly, anyway even though that I'm so close from nulling and Tx is almost 14.3khz but still there is no reaction from the device at all when I move a target or touch anything.

The wave forms look fine. RX is noisy due to electrical noise in you shop. My RX signal looks worse than that but TGSL works great.
I have read that with a concentric the RX phase shift is about 200 degrees.
Again use both channels of your scope to compare when the FET switches turn on relative to the TX and/or RX signal phase.

Next is check the outputs of the next two amps, U103 pins 1 & 7, then U104-6 & U105-7.
These should be very near to Zero Voltages.

Did you try All Metal Mode?
I am guessing it still beeps when powered on.

It did take me a while of fiddling to get my TGSL working. Keep at it.

[waleed]

Oki this what I did, first I removed some turns from Tx because the inductance wasn't correct, then while I was measuring U101a-pin7 I gradually removed turns from bucking coil until I reached 15mv ~ ,but I don't know why Rx waveform looks ugly, anyway even though that I'm so close from nulling and Tx is almost 14.3khz but still there is no reaction from the device at all when I move a target or touch anything.

https://i.ibb.co/pj7n6WB/20191231-153817.jpg

https://i.ibb.co/Pt4qmSd/20191231-171812.jpg

https://i.ibb.co/nBYrWtK/20191231-171425.jpg

https://i.ibb.co/qxgz23R/20191231-173050.jpg

https://i.ibb.co/JRjJYb9/20191231-173137.jpg

[waltr]

Hey waltr, well I'm aware that the coils are not nulled and I know how to do that but the thing got me is the square wave maybe as you said the amplitude of rx is high, but if the coils are not nulled I'll not get any reaction right?, and I'm sure that Tx is 6mh but I'll check the caps, oki I'll null the Rx coil then I'll check the waveform of pin7

Correct, pre-amp is Overloaded and so are the following stages so the slight amplitude change with a target passing the coil can not be detected. Once you have a better null and the pre-amp out is a sine then it will start detecting a target.

You also want to put the scope Ch1 of TX and scope ch2 on the RX pre-amp out. This is to compare the amplitude and PHASE of the RX relative to the TX. This is more important than have an exact TX frequency.
Also as I have said: checking the inputs, gates, to the FET switches in relation to the TX. dbowers shows these scope pictures in the TGSL 101 pdf.

[waleed]

1- The TX wave forms look good but the Frequency is only 13kHz. Should be 14.5kHz. This means either the TX coil's inductance is too high or the TX oscillator caps are to large. More likely the coil. Note: the TGSL will work at a different TX frequency.

2- The RX waveform is too large in Amplitude which is why (3) is a square wave.
The pre-amp has a gain of 220k/4.7k = 4.68.
The RX waveform is nearly 3Vpp and the pre-amp gains by 4.68 = 14V. But the per-amp's rails (power) is only 13V and can not output to the rails so it Clips the peaks.

The problem is that the Coils are not NULLED. You need to adjust the TX Buck Coil to get near the smallest RX p-p Voltage (and about a 20 degree phase shift relative to the TX).
Watch the RX input (J2-1) and play with the extra wire on the Buck coil. At some point the RX should reduce in amplitude.

Re-read the 101 pdf and other TGSL coil building docs.

Attached is one stuff:
[ATTACH]48942[/ATTACH]

Hey waltr, well I'm aware that the coils are not nulled and I know how to do that but the thing got me is the square wave maybe as you said the amplitude of rx is high, but if the coils are not nulled I'll not get any reaction right?, and I'm sure that Tx is 6mh but I'll check the caps, oki I'll null the Rx coil then I'll check the waveform of pin7

[waltr]

1- The TX wave forms look good but the Frequency is only 13kHz. Should be 14.5kHz. This means either the TX coil's inductance is too high or the TX oscillator caps are to large. More likely the coil. Note: the TGSL will work at a different TX frequency.

2- The RX waveform is too large in Amplitude which is why (3) is a square wave.
The pre-amp has a gain of 220k/4.7k = 4.68.
The RX waveform is nearly 3Vpp and the pre-amp gains by 4.68 = 14V. But the per-amp's rails (power) is only 13V and can not output to the rails so it Clips the peaks.

The problem is that the Coils are not NULLED. You need to adjust the TX Buck Coil to get near the smallest RX p-p Voltage (and about a 20 degree phase shift relative to the TX).
Watch the RX input (J2-1) and play with the extra wire on the Buck coil. At some point the RX should reduce in amplitude.

Re-read the 101 pdf and other TGSL coil building docs.

Attached is one stuff:
My concentric eccentric adventure.pdf

[waleed]

Hello guys, after I finished my exams I start to continue working on my project ,anyway something weird happened well the last time that I said the device started to detect metals now when I tried it it didn't give any reaction, then I made the 10" concentric coil (Rx:6.5mh & Tx:6mh) and I attached it to the device and when I turned the device on it gave a single beep and when I moved a target on the coil nothing happened no reaction at all, I checked everything and all tests looks good except that I noticed something that I felt is wrong, when I test the point J2-1 the waveform looks normal but when I test point U101a-pin7 the waveform is square wave,is that normal?, it is possible that I broke the njm2068 IC (I'm using it instead of LF353).

[dbanner]

Good luck!

[waleed]

Thanks dbanner and waltr also kt,now I have enough knowledge to build a good concentric coil with any size I want , I'll give feedback when I finish the 10" coil and attach it to the circuit

[dbanner]

I also did not ground the RX coil. Works great without falsing on wet grass.
In evaluating the design I see no reason to ground the Differential RX in.

I need to play with that Nautilus circuit. I have the question of why connect the RX coil to ground and will this circuit still work without that ground connection. Have a lot of other circuits to work on so won't be getting to this one soon.

The original tesoro coils, Rx is grounded, and there is no falsing on wet grass. So it stands to reason that there is something missing in the self made construction. I suspect strongly that it is in the shielding and the cable hookup configuration, but I can't be too bothered for now.

Grounding the Rx is supposed to make common mode between the shield and Rx, ironically to eliminate falsing.
Nautilus circuit Rx must be grounded I think.

[dbanner]

I tried to make the Dbowers TGSL concentric coil, I followed all his instructions perfectly, using the correct size wire and everything, and guess what, The inductances were close, but I still needed to tinker endlessly to get the coil to its peak performance.
These coils are so finely balanced that if you build ten identical coils using the same size wire and all the rest, they will still end up being slightly different, requiring "personalized" treatment.
Precision requires a coil winder that stacks the wires side by side for a rectangular(square) cross section that comes out perfect all the time, while glueing the turns together in the process of winding at the same time, so when it is dry, the coil is perfect, one after the next.
Then the shielding process must be applied to the housing and NOT the coil so as not to interfere with inductances.

[dbanner]

Read this article, it will certainly add to your knowledge.
Generally speaking, use thicker wire for Tx and thinner wire for Rx. The wire gauges for the Tx and Rx do matter, and not just because of resistance, but these IB concentric coils have a lot of design parameters that are not quite transparent to the new players. Like Q factor and coupling efficiency (tranformer action), temperature stability ( sensitivity in the linear region using 25 degrees C as a baseline) and ground effects in relation to the size and geometry of the coil. There is no getting around this, so follow the advice of george payne and use his guidelines as a reference when making these concentric type coils.
As for the feeback coil, I recommend using a double spool as Cscope has done with their concentric coils. This makes it very easy to add/remove turns on the fly, to bring the Rx into balance(null). Remember its about cancelling the Tx that the Rx coil sees. Overwind the buck, then just keep removing turns until you cancel the Tx field. Then you can reconcile the Tx frequency by capacitance. The double spool gives the buck coil its own groove, so the Rx stays in place and is not disturbed.

[waleed]

try to learn my pics done much years ago.

Thanks for sharing,but what would I achieve by doing this circuit?, it look to me that you can adjust the resonance and the resistance of the wire right?

That article is about all we can tell you without actually have made to specific coil ourselves.

20ga (0.8mm) wire sounds a bit large for the TGSL. The resistance will be lower.
If you read the TGSL threads (hundreds of pages) some one tried thicker wire, thinking higher current = more distance, and the TGSL did not work correctly.

Best if first time building is copy what has been proven to work. Then experiment with mods, etc.

Yea you are completely right,but I think that I can put a resistor series to the coil to fit the rage of 18-25 ohm maybe

[waltr]

Yea I read that,but it still not enough clear because Bx turns not only depend on the area enclosed for both TX and RX but also on the number of turns,anyway I'll figure out how to do that for my 0.85mm wire np.

xD I always like your comments even though they are tough sometimes but actually it's the truth ��

That article is about all we can tell you without actually have made to specific coil ourselves.

20ga (0.8mm) wire sounds a bit large for the TGSL. The resistance will be lower.
If you read the TGSL threads (hundreds of pages) some one tried thicker wire, thinking higher current = more distance, and the TGSL did not work correctly.

Best if first time building is copy what has been proven to work. Then experiment with mods, etc.

[kt315]

try to learn my pics done much years ago.

[waleed]

How you read this:
https://www.geotech1.com/cgi-bin/pag...anar/index.dat

Yea I read that,but it still not enough clear because Bx turns not only depend on the area enclosed for both TX and RX but also on the number of turns,anyway I'll figure out how to do that for my 0.85mm wire np.

there is NOT a manual how-to-balance-coil, NOT EXISTS, just manuals how-to-make. you have to do balance yourself connecting your mind.
the forum has not possibility to keep a staff of professional consultants helping to everybody in coil bulding. no finance, we are just money-free
peacemakers.

xD I always like your comments even though they are tough sometimes but actually it's the truth ��