TGSL nulling/RX-problem

[fatAngel]

hi *,

as my TGSL from ivconic doesn't work as expected, a few question came up.
first of all, a description of what i have:

** 2 homebrew coils (TX: 14.42kHz, RX: 13.92kHz - both measured at J1)
** power-supply is fine (12V, 7.98V, -6.2V)
** block 9 and 10 (GB as described in TGSL_101_PART1) are ok
** there is a short beep during power up/down
** replaced BF245C with BF246A; replaced LM308 with TL071

so i tried to null my coils. with RX connected to my DVM i managed to get as low as 15mV.
but the output of the preamp U101a (pin 7) is clipping. the voltage is nearly 14V!!
i double-checked any resistor and capacitor. replaced the LF353 with a new one.

my questions:
i often read about 16.1kHz in the RX-circuit. but my RX-frequency is the same as the TX-freq. (14.5kHz)
is this a normal behavior?

the osci-graph in the complete-details at test point 4 (U101a, pin 7) shows a voltage of about of 0.02V (.01V/DIV * 2DIV) for the RX-signal.
is this correct? shouldn't that be the _input_ of the preamp?
what is the gain of this preamp?
i expected the signal to be much higher (of course without clipping)

when i move some metals/ferrit across my coils i can't see any changes in my RX-signal. neither changes in amount nor phase.
is this a consequence of my problem or another mistake (maybe how i built my coils)?

btw: i have an oscilloscope and could post some images if necessary

any hints? what could i try?

many thx in advance

[Davor]

I think you should replace a Rx capacitor to obtain resonance above the Tx resonance. Point is that Rx coil with the capacitor actually makes a low pass filter with steep slope above the resonance. You may observe the red curve on a diagram here: http://www.geotech1.com/forums/showp...&postcount=136 to see what I am referring to.

[dfbowers]

hi *,

as my TGSL from ivconic doesn't work as expected, a few question came up.
first of all, a description of what i have:

** 2 homebrew coils (TX: 14.42kHz, RX: 13.92kHz - both measured at J1)
** power-supply is fine (12V, 7.98V, -6.2V)
** block 9 and 10 (GB as described in TGSL_101_PART1) are ok
** there is a short beep during power up/down
** replaced BF245C with BF246A; replaced LM308 with TL071

so i tried to null my coils. with RX connected to my DVM i managed to get as low as 15mV.
but the output of the preamp U101a (pin 7) is clipping. the voltage is nearly 14V!!
i double-checked any resistor and capacitor. replaced the LF353 with a new one.

my questions:
i often read about 16.1kHz in the RX-circuit. but my RX-frequency is the same as the TX-freq. (14.5kHz)
is this a normal behavior?

the osci-graph in the complete-details at test point 4 (U101a, pin 7) shows a voltage of about of 0.02V (.01V/DIV * 2DIV) for the RX-signal.
is this correct? shouldn't that be the _input_ of the preamp?
what is the gain of this preamp?
i expected the signal to be much higher (of course without clipping)

when i move some metals/ferrit across my coils i can't see any changes in my RX-signal. neither changes in amount nor phase.
is this a consequence of my problem or another mistake (maybe how i built my coils)?

btw: i have an oscilloscope and could post some images if necessary

any hints? what could i try?

many thx in advance

i often read about 16.1kHz in the RX-circuit. but my RX-frequency is the same as the TX-freq. (14.5kHz)
is this a normal behavior?

Yes.. The RX frequency will be the same as the TX, but the phase will be different. The closer the RX coil gets to resonance, the higher the amplitude will be and the phase difference should be less... Remember, the RX SIGNAL is being driven by the TX coil. It's just a little out of "tune".

the osci-graph in the complete-details at test point 4 (U101a, pin 7) shows a voltage of about of 0.02V (.01V/DIV * 2DIV) for the RX-signal.
is this correct? shouldn't that be the _input_ of the preamp?
what is the gain of this preamp?
i expected the signal to be much higher (of course without clipping)

When attempting to null your coils, keep things simple to start with. Just measure U101a pin 7 with a DVM and shoot for 4mV. The TGSL will work if you can get things in the ball park, but the output should definitely not be in saturation.


when i move some metals/ferrit across my coils i can't see any changes in my RX-signal. neither changes in amount nor phase.
is this a consequence of my problem or another mistake (maybe how i built my coils)?


Phase and amplitude changes are difficult to see. You are better off monitoring the changes further down the line.. like at the inputs to the LM308s.

Hope this help!
Don

[fatAngel]

many thx for your quick and helpful responses.

@davor
i understand what u mean. when i replace the original 15n RX capacitor with 10n everything look a bit better. nulling is easier and clipping is gone.

@dfbowers
at the moment its impossible to go for 4mV. i have 400mV!

there seems to be another mayor problem within my circuit.

what i wonder:
in the chapter "coil making" it is always stated as very important to build both coils in the same manner. (direction of winding etc.)
so it should make a difference when i exchange RX-start and RX-end.
i have mounted my TX coil on a plate and have 2 options, how to put my RX above it. lets say bottomside-up or topside-up.

but ... no difference. right?

thx again

[dfbowers]

many thx for your quick and helpful responses.

@davor
i understand what u mean. when i replace the original 15n RX capacitor with 10n everything look a bit better. nulling is easier and clipping is gone.

@dfbowers
at the moment its impossible to go for 4mV. i have 400mV!

there seems to be another mayor problem within my circuit.

what i wonder:
in the chapter "coil making" it is always stated as very important to build both coils in the same manner. (direction of winding etc.)
so it should make a difference when i exchange RX-start and RX-end.
i have mounted my TX coil on a plate and have 2 options, how to put my RX above it. lets say bottomside-up or topside-up.

but ... no difference. right?

thx again

I don't worry about coil direction, or coil placement (top or bottom for RX). Just reverse two leads of either RX or TX coil if you get a double beep. But keep in mind where you are grounding the sheilds to. I prefer a slight over-coupling, and that should give you an approximate 20 degree phase shift between RX and TX.

To get the best possible performance, it's nice to shoot for 4mV, but that's probably ideal. I rarely get mine that well nulled and things still work OK.

Most of all, make sure that GB and DISC work correctly.

don

[Mullihaka]

To get the best possible performance, it's nice to shoot for 4mV, but that's probably ideal. I rarely get mine that well nulled and things still work OK.

Hi dfbowers,

Don't get mad, I just want to double(triple)-check the residual voltage. Do you really get 4mV at the output of the U101a, not at the input?

I can get 527mV at the output, measured with an oscilloscope. That divided by the gain of the U101a (220/4.7 = 46. gives 11.3 mV. The DVM measures rms, so 11.2mV divided by (2*sqrt 2) gives 4.0mV. So, my 527mV with an oscilloscope at the output corresponds 4.0mV rms with a DVM at the input. Coincidence?

[Mullihaka]

I meant " That divided by the gain of the U101a (220/4.7 = 46. gives 11.3 mV."

[dfbowers]

Hi dfbowers,

Don't get mad, I just want to double(triple)-check the residual voltage. Do you really get 4mV at the output of the U101a, not at the input?

I can get 527mV at the output, measured with an oscilloscope. That divided by the gain of the U101a (220/4.7 = 46. gives 11.3 mV. The DVM measures rms, so 11.2mV divided by (2*sqrt 2) gives 4.0mV. So, my 527mV with an oscilloscope at the output corresponds 4.0mV rms with a DVM at the input. Coincidence?

Sorry!! I am mistaken. You are absolutely correct. I read Ivconics TGSL coil making and he suggests measuring the null signal with the Rx coil disconnected from the Rx circuit. That would account for a 4mV reading, so no coincidence. That's assuming the frequency response of your DVM includes the 14KhZ range.

That is a big mistake in my TGSL_101 part 1 document!! I'm surprised someone did not comment on that earlier.

The coil I used to make measurements here shows about 40mv p-p at the output of U101a. I think that's pretty typical on what I see. Again, thanks for catching that!!

Update - I just measured one of my best coils. I have 6mV p-p at the raw ends of Rx leads and around 300mV p-p at u101a pin 7. Hope this puts things in perspective!!

Don

[Mullihaka]

Thanks!

This releases me from trying to null the coils on the microvolt range!

So, I'm already on the correct decade (your 300mV versus my 500mV).

I have been wondering this from the very beginning when Ivconic wrote that he usually gets the value to 0.02 on 200mV scale on a DVM. That would mean 20 microvolts. Boy, steady hands needed!

Thanks for the good work, your manual has been of a big help!

[dfbowers]

Thanks!

This releases me from trying to null the coils on the microvolt range!

So, I'm already on the correct decade (your 300mV versus my 500mV).

I have been wondering this from the very beginning when Ivconic wrote that he usually gets the value to 0.02 on 200mV scale on a DVM. That would mean 20 microvolts. Boy, steady hands needed!

Thanks for the good work, your manual has been of a big help!

I think I found another mistake while reading TGSL101a Part 1 on page 6 where I have the output of u101a pin 7 displayed, I was actually using a x10 probe, so it would have been .1v/div.. we can always make the document better with the help of everyone!!.

[Jerry]

Hi Don,

Regardless of any typo's or errors it is still a valuable document. I printed mine out and put it in a loose leaf folder and have all kinds of notes written on the back sides of the pages and also extra pages of notes added in. I recommend it highly for anyone building the TGSL.

Jerry

[dfbowers]

Hi Don,

Regardless of any typo's or errors it is still a valuable document. I printed mine out and put it in a loose leaf folder and have all kinds of notes written on the back sides of the pages and also extra pages of notes added in. I recommend it highly for anyone building the TGSL.

Jerry

Hi Jerry,

If you have any suggestions on what to change or add, please let me know.

Thanks!

[Jerry]

Hi Jerry,

If you have any suggestions on what to change or add, please let me know.

Thanks!

I have a few on the screen shots of the wave forms, but rather than suggest I can take some next time I have it set up and send them to you and see what you think.

Jerry

[fatAngel]

after i took a break for a few days, i again checked my coils and circuit.
my problem: none!

as mentioned: i stuck too deep into details and missed the important parts. my TGSL always responded to ferrit and bigger iron-objects. but i didn't realise this, because the sensivity is very weak.

as pointed out, i also misunderstood the description of nulling and the voltages recommended. my values are in a suitable range now.

thanks all for your hints!

still a little suggestion: i would really like to unterstand how to increase and fine-tune the TGSL sens. a little how-to would help a lot.
maybe someone knows of a procedure to improve performance?

mart;n
ps: a view pics of my project (osci-graphs) will follow