TGSL Experiments

[ivconic]

Nice work U.Haze. But please try to keep photos in a bit lower resolution. Such huge photos are slowing down page to much. That is the case with so many other posts on other threads. People do not realise that and post as higher resolution as possible. So some pages are very slow here on this forums. Again, nice work. But you should post those in TGS thread not here...

[Ultimate_Haze]

Hi,
The lowest R that I could find now is 20ohms.When I add it to TX winding frequency is 15,1.Measure from TX,G TR4 and GTR5.
Without this 20ohms R I get same picture as before:TX 14,9;G TR4 30,1;G TR5 14,9.
So it must be wire after all!I will have to buy smaller wire,so I can get some resistance.)))But this wire is 0,25?

Nice work U.Haze
Thanks but this is nothing compared with your masterpieces.

keep photos in a bit lower resolution
I have used lowest resolution.My camera make a lot better pics.
Will shoot with other lower res. device next time.I agree.

you should post those in TGS thread not here...
Its not TGSLF yet.Not working like one.Have posted this pictures just for comparation.When its working,cleaned,and pack I will post pics in TGS
thread.

Thanks

[simonbaker]

No, this will not be a case here. If coil acts bad in air (in workshop) than it will act bad on real soil for sure. No doubts. Good and proper coils are acting good in both case.
Yet, your way of thinking is right to me. I tend to agree. Stabilisaton part (if that was its original role after all) is also very interesting to me. But if you analyze it a bit detailed, you will found out that base of 2N2907 is always coupled to ground via 47K. Fet's S and D are "shorted" in 99% of cases. Just when something happens strange, fet gets a signal to G and than S and D "opens" and switch 2N2907 base from 47K to 330K+47K to ground. So it acts actually as simple, ordinary switch ; but conditioned with G components (48,749,100K,0.22). So...i simply removed those and left only 47K to ground.
I guess all your "problems" are coming from inproper wire.
For a test...you can add resistor in serie with TX coil and see what will happen...

I think referring to JFet in oscillator... it depends on JFet. JFets have very different cutoff voltage, completely changes behavior in TGS oscillator.

I made prototype TGSL with JFet in oscillator -- it has perfect sine wave but very low voltage, about 10V p-p. I don't think JFet works as switch there, more like variable resistor. I can change the oscillator voltage gradually by changing the zener diode. I'll bet a JFet with different cutoff voltage would change it also.

Now, that circuit has very nice waveform... but terrible depth! -- so removing JFet is good idea for depth, get maximum oscillation voltage/current. But everything must be just right I think to keep it under control -- because that oscillator is now crazy beast, bigger than Tesoro expected. So one wrong step and... you get hoisted on the horns.

My philosophy is maybe a little mod will keep the beast in chains, make it easier to build coils, not need to be so perfect. I have not found the perfect mod, but I think maybe if we use only fraction of the voltage to drive the discrimination and ground balance, then we don't worry about coil and oscillator so much. We still need good null of course, but not worry so much about oscillator.

The little blip in the sine wave doesn't seem like big deal but there is one case where I think it may cause some trouble. If the blip moves further down the wave (instead of being near top), then you will see some worry at the output of the LF353 pin 1. It makes a "level" section near 0 volts, like a step. This is not good for stability of discriminator -- you want the waveform to cross zero as vertical as possible for stability.

I found with very low resistance coils (7 ohms in my case), the blip does move down the sine wave and creates this worry. I'm not sure yet how important, but something for crazy people who make wild coils to think about.

So TGSL is very good if you make coil just right, no doubt. I am also looking for small mod that allows some more extreme coils to be used also. Maybe will be successful... or not.

Regards,

-SB

[simonbaker]

Hi,
The lowest R that I could find now is 20ohms.When I add it to TX winding frequency is 15,1.Measure from TX,G TR4 and GTR5.
Without this 20ohms R I get same picture as before:TX 14,9;G TR4 30,1;G TR5 14,9.
So it must be wire after all!I will have to buy smaller wire,so I can get some resistance.)))But this wire is 0,25?

That makes me think it is high oscillator voltage causing trouble as suspected, but need oscilloscope to confirm.

Q - What is oscillator voltage without resistor and with resistor for comparison?

I think you want smallest resistor that does the trick, because adding resistance reduces magnetic field. So maybe smart to learn what that is before choosing wire to make coil.

You can also experiment this way: instead of adding small resistor in series with TX coil, put a 100K pot in parallel with coil and turn it until the JFet gate frequency is correct. There is a way to calculate the equivalent series resistor, probably Qiaozhi can tell us.

Regards,

-SB

[ivconic]

I think referring to JFet in oscillator... it depends on JFet. JFets have very different cutoff voltage, completely changes behavior in TGS oscillator.

I made prototype TGSL with JFet in oscillator -- it has perfect sine wave but very low voltage, about 10V p-p. I don't think JFet works as switch there, more like variable resistor. I can change the oscillator voltage gradually by changing the zener diode. I'll bet a JFet with different cutoff voltage would change it also.

Now, that circuit has very nice waveform... but terrible depth! -- so removing JFet is good idea for depth, get maximum oscillation voltage/current. But everything must be just right I think to keep it under control -- because that oscillator is now crazy beast, bigger than Tesoro expected. So one wrong step and... you get hoisted on the horns.

My philosophy is maybe a little mod will keep the beast in chains, make it easier to build coils, not need to be so perfect. I have not found the perfect mod, but I think maybe if we use only fraction of the voltage to drive the discrimination and ground balance, then we don't worry about coil and oscillator so much. We still need good null of course, but not worry so much about oscillator.

The little blip in the sine wave doesn't seem like big deal but there is one case where I think it may cause some trouble. If the blip moves further down the wave (instead of being near top), then you will see some worry at the output of the LF353 pin 1. It makes a "level" section near 0 volts, like a step. This is not good for stability of discriminator -- you want the waveform to cross zero as vertical as possible for stability.

I found with very low resistance coils (7 ohms in my case), the blip does move down the sine wave and creates this worry. I'm not sure yet how important, but something for crazy people who make wild coils to think about.

So TGSL is very good if you make coil just right, no doubt. I am also looking for small mod that allows some more extreme coils to be used also. Maybe will be successful... or not.

Regards,

-SB

Yes that could be option too. I can't tell - i never had TIS75 to see the differences. Usually i used J107 and recently BF245c (or B).
What a coincidence! Just 15 minutes ago i received Tesoro Bandido II (old one, not uMax). Wow! This is my very FIRST original Tesoro! I also had chance once to see original Eldorado and original Silver Sabre. But those were here just on servicing. Now, this one is mine! In next few days i will open it and perform serie of measurements. You may suggest me now what are you interested for, to be scoped and measured. It is damn smt pcb (piggy). I dont manage well at those, but i will do my best.
I will post my first impressions later, on new thread..

[Molzar]

Hi Molzar, if you see this:

I'm not sure if asked before, but:

On Tesoro TGS you have, can you look at output of LM393 comparators that get input from LM308. Are there any resistors (like 4k7 resistor) on the output before the outputs come together, or are the outputs directly connected together?

There was one schematic that had resistors, but most don't. I just want to confirm how Tesoro did it.

Regards,

-SB

Simon,

Not sure if you asked this before either, but I did post accurate schematic on TGS thread last year (end November). I have seen that schematic, one of Ivconic's TGSL I think or maybe "Gift Pack"? Short answer is no resistors.......

Has been too hot here to work in shop much lately... but always enjoy your comments and (sometimes nutty) ideas!

Stay COOL

Cheers,

Molzar

[Aiko]

Hello, Simonbaker !

I succeeded to raise the performance on my TGSL, by as reducing of the TX frequency from 16 kHz to 14 and a little kHz, with a 1 nano capacitor parallel to the TX coil. This wrong interpretation on your suggestion gave positive result. Now I succeed to detect 1 euro coin at 28 cm with good signal and at 30 cm with broken signal. This with a coil with the parameters, commended by Ivconic ... but the wire is 0,3 millimeters. Unfortunately I can not obtain a 0.25 mm wire.

I hope that my English is understandable.

Regards!

[simonbaker]

Yes that could be option too. I can't tell - i never had TIS75 to see the differences. Usually i used J107 and recently BF245c (or B).
What a coincidence! Just 15 minutes ago i received Tesoro Bandido II (old one, not uMax). Wow! This is my very FIRST original Tesoro! I also had chance once to see original Eldorado and original Silver Sabre. But those were here just on servicing. Now, this one is mine! In next few days i will open it and perform serie of measurements. You may suggest me now what are you interested for, to be scoped and measured. It is damn smt pcb (piggy). I dont manage well at those, but i will do my best.
I will post my first impressions later, on new thread..

I use only J107 so far -- and looking at spec sheet, cutoff voltage can be from -.5 to -4.5 -- big difference for some circuits, especially TGS oscillator. Almost like using different components.

Bandido II original -- super. I wonder how it compares to handmades. We will find out from you!

Regards,

-SB

[simonbaker]

Simon,

Not sure if you asked this before either, but I did post accurate schematic on TGS thread last year (end November). I have seen that schematic, one of Ivconic's TGSL I think or maybe "Gift Pack"? Short answer is no resistors.......

Has been too hot here to work in shop much lately... but always enjoy your comments and (sometimes nutty) ideas!

Stay COOL

Cheers,

Molzar

Thanks, I'll look harder for it. I didn't know if final small corrections were added (single diode on LM308, etc.). But you answered question for now!

I enjoy your and others beautiful MDs and PCBs you make, keeps me going, trying, nutty...

We had such heat my oscilloscope stopped working - scary - but back to life (for now) -- hope it cools down for you, get you back in shop...

Cheers!

-SB

[simonbaker]

Hello, Simonbaker !

I succeeded to raise the performance on my TGSL, by as reducing of the TX frequency from 16 kHz to 14 and a little kHz, with a 1 nano capacitor parallel to the TX coil. This wrong interpretation on your suggestion gave positive result. Now I succeed to detect 1 euro coin at 28 cm with good signal and at 30 cm with broken signal. This with a coil with the parameters, commended by Ivconic ... but the wire is 0,3 millimeters. Unfortunately I can not obtain a 0.25 mm wire.

I hope that my English is understandable.

Regards!

You have a pretty good MD, as good or better than original TGS I think. I believe your thicker wire is better on TX coil if stable for you. On RX coil, depends. You can experiment by adding very low resistors (5 ohms, 10 ohms, 15 ohms) in series with RX coil -- I think should be similar to thinner wire. I doubt it will improve depth (another guess), could affect discriminator phase a little -- sometimes that helps if discriminator phase too far off.

Good work, I'm tinkering on mine still, and will keep reporting...

-SB

[ivconic]

I think this will solve any eventuall doubts....
(So...than, must be the way coil was made and nulled, directly determines importancy of this stabilisation stage)

[simonbaker]

I think this will solve any eventuall doubts....
(So...than, must be the way coil was made and nulled, directly determines importancy of this stabilisation stage)

Yes, I think so -- if coil just right, OK without JFet.

If coil not perfect.... mod that will help???? I'm looking ...

-SB

[ivconic]

Right now i am reconsidering option to put back on pcb that fet and the rest of components... Simonbaker, if i do so, than it would be your "guilt"!

[Ultimate_Haze]

Hello,
Its look like I used wrong wire.I bought 0,25 bare wire.And because its from specialized shop I tough that I can take this as a true.
Now same guy that sold me this wire,take a measure and tell me its 0.21))))
And I bought new wire which have to be 0,25.
Strange now I have higher resistance with same former,but still have to make some extra winds so I can get right inductance.
Hum.....Still some doubts on wire I use.
With new wire I made 2 coils.Data's are:
TX 100w
5,58mH 25,2ohm
RX 105w
5,94mH 26,1ohm
Frequency 14,6 28,9 at G TR4
And
TX 106w
5,95mH 26,7ohm
RX 111w
6,67mH 27,8ohm
Frequency 14,1
Maybe its my former.I use last former posted in TGS thread which is in PDF.And maybe when I print,dimensions are changed.
Damn.This took too many time and efforts.
But I am curies what is that I miss so far.
Just for test I have add the missing TX part with JFet.No big diferance.But coil is balanced much more easy.Dont now if its because of the different wire or JFet.
Same wrong frequency at G TR4.
Sometimes I gain same as TX but sens is still too low.20cm.

[ivconic]

Take care on proper inductance, dont mind resitance right now.
It is because of different wire that you can easilly balance now. Also dont torture yourself with exact former dimensions, never mind.
Must be something in your way of balancing, or your way of connecting coils to cable...or cable to pcb? Or last; could be something at your pcb...
I really can not guess what should be the problem. No way to solve it remotelly.