Question about tgsl

[Boadil]

Thanks!!!!!

[waleed]

Hello everyone..well I got some time this week and finished the board. But before I make the coils I would ask if I can make some tests just to verify that everything is working well or at least the board doesn't contain any broken components

[kt315]

no, you can not. you must connect a coil and then do a test. without coil the circuit does not generate negative voltage needed for the working.
thats specific feature of this circuit.

[waleed]

no, you can not. you must connect a coil and then do a test. without coil the circuit does not generate negative voltage needed for the working.
thats specific feature of this circuit.

Yeah I read before that I should connect the coils to get the -5v but I thought I can do another tests rather than that..but anyway I'll go ahead and make the coils I hope today

[waltr]

Correct, you need at least a TX coil to have the TX oscillator running.
Then check the -5 (-6.2V) power.
Next check the outputs to the JFET switches.

I wound a 100mm diameter coil for bench testing these parts.
Use the coil calculator for number of turns to be close to 6mH.
This works for the initial testing.

[waleed]

Hello there..well I connected the tx coil only (I don't know its inductance exactly but same n of turns and 24 ohm) but when I turned the device on nothing happened I mean there is no sound at all..I checked the 8v it was 7.9 I think no problem, but I don't have the negative 5v it was 0.79v. Also I checked the output of tx coil on the ac settings but it didn't give any output on the oscilloscope.
I think the problem with the tx coil Idk..and how do I measure its resonance frequency on the oscilloscope

[waleed]

Note: I used PN2222 instead of 2N2222 but I think that i have placed the two transistors on the tx coil wrong maybe i have reversed emitter with collector idk

[dbanner]

Check all transistors for proper orientation. You have to also test them for integrity, as putting them in the wrong way could potentially cause damage.
Check the transistors for the voltage inverter.

[waltr]

Note: I used PN2222 instead of 2N2222 but I think that i have placed the two transistors on the tx coil wrong maybe i have reversed emitter with collector idk

Check all transistors for proper orientation. You have to also test them for integrity, as putting them in the wrong way could potentially cause damage.
Check the transistors for the voltage inverter.

Yes, different make transistors do have different pin outs.
This is a huge problem for beginners and persons with lots of experience. This has gotten me a few times.

A transistor tester, many DMM have this feature, or the diode scale on your Ohmmeter can determine the pin out. Look for the emitter to base diode.

The TGSL's oscillator will run with a wide range of coil inductances.
So issue is most likely the transistor pin out.

7.9V instead of exactly 8V is no problem as this is simply the tolerance of the 8V regulator's output.
The negative 5V MUST have the oscillator running.

First step is getting the oscillator running like you are trying to do.

[waleed]

Thanks guys for the reply I really appreciated it.. yea now I'm sure that I have placed them wrong so I'll test them first for integrity then I'll placed them the right way and I'll see what will happen

[waleed]

Hello there..well I have replaced the 4 transistors (tr1 try tr7 tr8 ) the right way (I'm not sure if their integrity is good or not tbh) and when i turned the device on it gave a single beeping and stopped (this happen every time I turn the device on) and I felt it will work well but when I pass a target on the coils nothing happen, I measured the tx oscillation and there was ac output on the oscilloscope but I think the frequency on tx is not correct (idk if this will cause the problem) , also I measured the negative voltage and it was -5.7v it's good I think, also I measured U100 pin6 and there was a clocking signal output on the oscilloscope but I think the period is not same as in the document one, and finally the thing which took my attention is that when I measured TR2 (base) the voltage was -5.6 and i think it should be almost 6 volt normally (when no targets on the coils).

I'll say it again I don't know how much exactly the coils inductance so maybe the resonance frequency of both is not correct, let's say tx coil resonance isn't 14.5khz (less or more) is this will make the device have no sound when target pass the coils? Or at least there should be a sound even if the resonance is not correct and it will effect the depth or the disc, right?

I have two doubts maybe the coils isn't made properly or I have an damaged transistor (maybe TR2 because of the -5.6v on the base).idk

[waltr]

Ok, progress.
TX frequency does not seem correct but the TX wave form shape does.
A single beep on power up is normal so seems circuits are now working.
The TX & RX coils plus GB adjustment must be correct to detect metal.
Do not worry about the audio section yet. You need to start at TX and work through the first circuits.

Next is to look at the gates of TR4 & TR5. Use the scope ch1 on TX oscillator look at the gates with scope channel 2.
These should be switching on in sync with TX and about 90 degrees over lap.

Now get the coils to the correct inductance. TX coil = 6.0mH and RX coil = 6.5mH so TX frequency is correct.
Go back to dbower's TGSL 101 pdf for coil building. His instructions work using the TX oscillator to adjust coil turns to get 14.5kHz on TX coil then 13.9kHz with the RX coil connected to the TX oscillator. It is this slight difference that is required for best detection and for the discrimination to work correctly.

Then with scope channel 1 on TX oscillator output look at the RX pre-amp output (U101 pin 7) with channel 2 of scope.
Adjust coil over lap to obtain a null at Pre-amp out and a phase shift of about 20 degrees.

This is a VERY Fiddly operation as the very slightest movement makes a huge difference in the RX null.

[waleed]

Ok, progress.
TX frequency does not seem correct but the TX wave form shape does.
A single beep on power up is normal so seems circuits are now working.
The TX & RX coils plus GB adjustment must be correct to detect metal.
Do not worry about the audio section yet. You need to start at TX and work through the first circuits.

Next is to look at the gates of TR4 & TR5. Use the scope ch1 on TX oscillator look at the gates with scope channel 2.
These should be switching on in sync with TX and about 90 degrees over lap.

Now get the coils to the correct inductance. TX coil = 6.0mH and RX coil = 6.5mH so TX frequency is correct.
Go back to dbower's TGSL 101 pdf for coil building. His instructions work using the TX oscillator to adjust coil turns to get 14.5kHz on TX coil then 13.9kHz with the RX coil connected to the TX oscillator. It is this slight difference that is required for best detection and for the discrimination to work correctly.

Then with scope channel 1 on TX oscillator output look at the RX pre-amp output (U101 pin 7) with channel 2 of scope.
Adjust coil over lap to obtain a null at Pre-amp out and a phase shift of about 20 degrees.

This is a VERY Fiddly operation as the very slightest movement makes a huge difference in the RX null.

Thanks waltr for those useful informations,now I will build a proper coils as in the document maybe I'll build the concentric one I'll make sure that they have the correct inductance, but I would ask what about the -5.6v one the base of TR2 it is because of the coils? Because I saw that it should be almost 6v when no target on the coils I'm right?

[dbanner]

Well, I can tell you that winding coils to precise inductances requires some effort, and a jig to make things easier to keep count of those turns.
But knowing the effective parallel capacitance of the Colpitts oscillator, you can just bodge in a small value cap to tweak the resonance up or down as appropriate.
A better way of getting the resonances right is to start with Tx resonance as close to 14.5kHz, it doesn't have to be spot on, then adjusting the Rx resonance for the same spread (16.1-14.5=1.6kHz). The Rx resonance is easier to manipulate because it is just a parallel capacitor which can be adjusted up or down in value to get the right 1.6kHz spread. So if you end up with a Tx of say 14.4kHz then set the Rx resonance for 16kHz. The detector will function the same.

I have found that constructing a universal oscillator (coil tester circuit found here on forum) is very helpful when trying to set the Rx resonance. Just hook up the Rx coil to the coil tester circuit and scope the waveform while adjusting the Tank capacitance for the desired resonance. Having gotten the correct resonance, it is simply a matter of reconnecting the Rx coil to the metal detector circuit and putting in the correct capacitor value. This sort of approach obviates the need to keep removing the Tx coil from its oscillator.

[waleed]

Well, I can tell you that winding coils to precise inductances requires some effort, and a jig to make things easier to keep count of those turns.
But knowing the effective parallel capacitance of the Colpitts oscillator, you can just bodge in a small value cap to tweak the resonance up or down as appropriate.
A better way of getting the resonances right is to start with Tx resonance as close to 14.5kHz, it doesn't have to be spot on, then adjusting the Rx resonance for the same spread (16.1-14.5=1.6kHz). The Rx resonance is easier to manipulate because it is just a parallel capacitor which can be adjusted up or down in value to get the right 1.6kHz spread. So if you end up with a Tx of say 14.4kHz then set the Rx resonance for 16kHz. The detector will function the same.

I have found that constructing a universal oscillator (coil tester circuit found here on forum) is very helpful when trying to set the Rx resonance. Just hook up the Rx coil to the coil tester circuit and scope the waveform while adjusting the Tank capacitance for the desired resonance. Having gotten the correct resonance, it is simply a matter of reconnecting the Rx coil to the metal detector circuit and putting in the correct capacitor value. This sort of approach obviates the need to keep removing the Tx coil from its oscillator.

Yea you are right I know that the difference between tx and rx resonance should be 1.6khz,I don't have any problem on how to get the right resonance for both coils. But before I go ahead and make the correct coils I'm wondering if the -5.6v on TR2 (B) is normal or not,it is related to the coils?

Also I did not mention that I didn't put diode (D5) on the board cuz I thought it will give some depth and I wanted to try that but I don't think this have to do with the no sound problem right?