IGSL

[giuhalftrack]

Read topic from a start and you will find answers.
Cheers!

ok I read that you cannot reject ferrite (ideal soil) at both blocks in the same time.. but i cannot understand why..... probably i need to study more

[ivconic]

Hello!

Homemade IGSL from Hungary. Modified head. The built-in amplifier and oscillator head.
1 Hungarian Forint (16 mm) in air: 30 cm
20 Hungarian Forint (26 mm) in air: 40 cm
http://www.youtube.com/watch?v=cxBNNl9d55U

Somehow i missed this post!?
(my ages and my focus lately)
Congratulations Mateatek!
Excellent!
You made it to work exactly as it should work!
Regards!




P.S. Maybe if you pick next lower outputs at 4024; you'll get a more pleasant tone combination. Higher tone you picked is too high and not so pleasant.

[Davor]

I may have a quick fix for the Fe channel shortcomings in the existing builds that is really dead simple. In fact I already implemented it in my other mods. It is not competing with the other mods or in functioning of the rig.

Short explanation...

There are GEB and Disc channels. In TGSL schematic both are compared against a Disc. sens. voltage, so that the noise does not pass through in the form of chatter. Fine. But the point is that the Disc channel is merely determining the quadrant signal is coming from, hence the perfect reference is not Disc.Sens. but exactly 0V, and it's ideal mode of operation is binary. As I already noticed the need for better discrimination criterion at low signals, I already pumped up the gain of the second stage of both Disc. channels in my mods, but failed to notice that it also cures the faulty behaviour of the Fe channel in the original build.
The comparator pins are swapped in the Fe disc channel and the correct operation would be assured if they are corrected, but this quick fix will improve things a lot, and will also improve discrimination over a whole target distance range.

How to do it? Locate the diodes in the gain stages and identify them. Do the same for both Fe and non-Fe channels. Extract the diodes in the first stages, extract the capacitors in the second stages, also extract the 1M resistors in the second stages, place the antiparallel diodes in place of the extracted capacitors (indicated by a circle). Done. Enjoy!

So in essence you don't have to swap the comparator pins in Fe Disc channel if you do this mod

See picture

[Davor]

I ported phase control for both GEB channels together so they both do the same thing. Ivconic had some interesting ideas of what to do with separate GEB, but I use in fact a single GEB because it is more straightforward this way.

[ivconic]

Hi today i have a question about IGSL TGSL..

I know that igsl tgsl is made by 2 tgsl, but the question is.. why we need 2 ground balance channels ? is worth had only 1 channell (ground balance) for 2 TGSL ???

the soil is the same for the 2 detector.....

please explain.

Read topic from a start and you will find answers.
Cheers!

[giuhalftrack]

Hi today i have a question about IGSL TGSL..

I know that igsl tgsl is made by 2 tgsl, but the question is.. why we need 2 ground balance channels ? is worth had only 1 channell (ground balance) for 2 TGSL ???

the soil is the same for the 2 detector.....

please explain.

[Davor]

I think you found the source of your detector problem. Having -0.2V offset is far too much, and I suspect the capacitor C17 at 470nF to be leaky, or some problem must have happened at the PCB around there. U6b and the other second stage gain blocks are working at fairly high impedance so if anything leaks it may affect the offset. In unlikely case that you use some acid flux for soldering, make sure you wash your PCB with alcohol prior to any further testing because these leak current, and in a long run your PCB would corrode.

Supply voltages are OK. You could check the current consumption, simply for future reference.

As for the oscillator, you may replace the 2N2907 with some more lively transistor such as BC327

[giuhalftrack]

hi it's me again

I wonder if this values on my igsl - tgsl are correct.
(my problem is still the same very poor detection depth and instability).

on pin 3 in LM7808 i had 7.93 V (value on schematics is 8V ) i feed the IGSL with 15 V DC.
on D4 i had -5.90 V (value on schematics is -6.2V)

on pin 7 of U6 (LF347 in schematic) i had -0.2 -0.3V when coin passing on coil, always a negative value ? in TGSL101 document this test point must swing from + or - 1V as target pass the coil.

now my last obsession on TR3 (100k trimmer) to tune TX sinus i get a good oscillation (the detector works) only for resistence value under 600 ohms but in the TGSL design the resistence value in this point (was a fixed resistor) is about 47 K if i put here a 47K resistor i had no oscillation at all.

can anyone measure this value on a working IGSL please ?

[mateatek]

Hello!

Homemade IGSL from Hungary. Modified head. The built-in amplifier and oscillator head.
1 Hungarian Forint (16 mm) in air: 30 cm
20 Hungarian Forint (26 mm) in air: 40 cm

http://www.youtube.com/watch?v=cxBNNl9d55U

[Davor]

It is worth it

[golfnut]

That a very good advice, in proper checklist
You are right on the spot, I'm concern myself with unnecessary thing at the moment.
I'm a beginner in electronic stuff and new here too. Just finish reading the whole IGSL theread recently. I do read TGS thread too, in random order thanks to search function. There excellent guide to make TGSL coil and musketeer coil seem a bit more difficult.
Oh wait, 1st, a suitable housing. I got a aluminum box from external hard disk. Should be fit.
I should done with getting part and start soldering

Good luck.

The kit on here that I find easiest in order ...

Surf Pi easiest
IDX
TGSL
IGSL least easy

[Hahhaa]

You will not need to worry about noise and swapping chips around.

You will need to concern yourself with..

A suitable housing
soldering all the parts correctly
Wiring all the pots the right way
Obtaining a 9" DD coil shell
A shaft plus fittings and handle/grip
Appropriate battery technology
Coil shielding techniques
Coil Nulling process
Sourcing quality screened cable to coil
Trouble shooting if things go bad
Setting up
Finally learning how to use it properly - to get something out of it.

In 3 months time when you are up to snuff, start swapping chips around.

Steve

That a very good advice, in proper checklist
You are right on the spot, I'm concern myself with unnecessary thing at the moment.
I'm a beginner in electronic stuff and new here too. Just finish reading the whole IGSL theread recently. I do read TGS thread too, in random order thanks to search function. There excellent guide to make TGSL coil and musketeer coil seem a bit more difficult.
Oh wait, 1st, a suitable housing. I got a aluminum box from external hard disk. Should be fit.
I should done with getting part and start soldering

[giuhalftrack]

LF444 should do too. I guess it is easier to obtain.

I had a pair of this in a chinese MD named MD318B now i'm going to try.

You will need to concern yourself with..

A suitable housing
Wiring all the pots the right way
Obtaining a 9" DD coil shell
A shaft plus fittings and handle/grip
Appropriate battery technology
Coil shielding techniques
Coil Nulling process
Sourcing quality screened cable to coil
Trouble shooting if things go bad
Setting up
Finally learning how to use it properly - to get something out of it.

In 3 months time when you are up to snuff, start swapping chips around.

Steve

Steve my tgsl is in an alluminum grounded box, power supply come from house, constant 15V dc, my coil is the original Tesoro 8" "donut" 5pin, with original cable and connector. Really i don't want to change chips i wont go out searching my favorite zone with igsl - tgsl, but had trouble to get an acceptable detection depth.

[golfnut]

You will not need to worry about noise and swapping chips around.

You will need to concern yourself with..

A suitable housing
soldering all the parts correctly
Wiring all the pots the right way
Obtaining a 9" DD coil shell
A shaft plus fittings and handle/grip
Appropriate battery technology
Coil shielding techniques
Coil Nulling process
Sourcing quality screened cable to coil
Trouble shooting if things go bad
Setting up
Finally learning how to use it properly - to get something out of it.

In 3 months time when you are up to snuff, start swapping chips around.

Steve

[giuhalftrack]

No, this is not a stupid question at all. It deals with offset problems of various families of opamps.
In general you have two flavours of op amps, those with FETs and those with BJTs on their input. BJT input opamps have transistor bases directly at their inputs, and those require some base current.
In case of a TGSL (and IGSL) configuration, and using LM324 as PNP BJT input opamp, the final gain stage has a non-inverting input nailed to the signal ground, while the inverting one passes the base current through a considerably high resistance, which results in about -15mV offset at the output. With Disc Sens control starting at 3mV, you need a signal that is 18mV instead of 3mV to give you a beep, and that accounts for some serious depth loss.

FET opamps have negligible bias current and their offset is typically not related to input resistance mismatch and you don't have such loss.

In case you insist on using LM324, because of low power consumption, voltage flexibility etc. - you'd have to do something about the offset problem. But frankly, the easiest thing is to just replace the opamp with a FET input one.

ok very interesting !!! I was disappointed when the guy at shop said: "they are equal".