MY NULLING PICS TGSL

[ashefr10]

I provide a picture of my nulling, thank you to give your comment on the good and less good. Output of U101 I can not have low voltage, and phase is 16 micro seconds instead of 4

http://imageshack.us/photo/my-images/196/20120411098.jpg/

[porkluvr]

Can you tell us what the inductance of your rx coil is?

Either measure it directly with LCR meter or else temporarily connect it to the transmitter and measure transmit frequency.

[ashefr10]

my value Rx 6,52 mH and my value Tx 6,22 mH .

[getsa]

The shape of Tx signal isn`t good - shouldn`t have any curves!

[ashefr10]

The shape of Tx signal isn`t good - shouldn`t have any curves!

I tried several coil I toujour this oscillogram. a friend has also TGSL and also has the same signal.

[getsa]

The working mode of transistor isn`t selected properly - it is necessary to select suitable values of the passive elements(resistors), determining working point of transistor for the concrete coil.

[ashefr10]

ok , thanks , but it does not prevent the proper functioning of the whole has...

[porkluvr]

my value Rx 6,52 mH and my value Tx 6,22 mH .

I'm surprised - I expected that your rx coil would have a much higher L value to explain the large phase difference. It is hard to be sure because of the tx wave distortion but that looks like around 60 degrees to me.

Your null could probably stand to be a little better but that might be usable.

For your phase setup, maybe either C6 or C7 is too large? Did you measure and are they 15nF and 5pF? Either one being too high will delay the rx signal. Maybe C6 or C7 is the wrong value or out of tolerance.

Use a piece of ferrite to simulate the ground and see if you can zero the rx signal as you wave it around. If so, then see if the discriminate circuit is working for different types of metal.

There is not really a lot you can do to improve the transmitted waveform unless you used a different oscillator. Although having the distortion is not the ideal situation, it is not unusual and you should not worry too much about it.

[ashefr10]

I'm surprised - I expected that your rx coil would have a much higher L value to explain the large phase difference. It is hard to be sure because of the tx wave distortion but that looks like around 60 degrees to me.

Your null could probably stand to be a little better but that might be usable.

For your phase setup, maybe either C6 or C7 is too large? Did you measure and are they 15nF and 5pF? Either one being too high will delay the rx signal. Maybe C6 or C7 is the wrong value or out of tolerance.

Use a piece of ferrite to simulate the ground and see if you can zero the rx signal as you wave it around. If so, then see if the discriminate circuit is working for different types of metal.

There is not really a lot you can do to improve the transmitted waveform unless you used a different oscillator. Although having the distortion is not the ideal situation, it is not unusual and you should not worry too much about it.

hello , c6 and C7 are 15 nF and 10 pF

[porkluvr]

According to my schematic C7 should be 5pF. So, 4.7~5.1pF would be pretty reasonable but 10pF is not what you want.

Changing to 5pF should advance the rx signal about 6.5 degrees.

edit: Well, I was looking at the schematic for a different model. So your 10pF should be right but what I said about advancing by 6.5 degrees should still be valid. But I do not understand why your rx signal phase looks so wrong.

Just please see if you can ground balance using a piece of ferrite for the target.

[ashefr10]

I used version 3 of TGSL made available in PDF on the site. So the oscillator is the same. I can be a problem on my card. in all I manage to reject the ferrite bar and keep a silver coin

[ashefr10]

According to my schematic C7 should be 5pF. So, 4.7~5.1pF would be pretty reasonable but 10pF is not what you want.

Changing to 5pF should advance the rx signal about 6.5 degrees.

ok, I will try to change c7 but are what I really need to have the 20 degrees?

[ashefr10]

we also see on the oscillogram that the frequencies of two signals are very close, are what this is correct? because I felt it was necessary that one is a little different from the other.

[porkluvr]

I made a mistake looking at a different schematic. Your C7 being 10pF is correct... but what I said about advancing by 6.5 degrees is still valid, not that you should need to go that route.

Your tx and rx frequencies are going to be exactly the same. When you tune your rx tank slightly higher or lower, what that does is affect the signal phase and amplitude but the rx frequency will be the same as what was transmitted.

... in all I manage to reject the ferrite bar and keep a silver coin

Then what is wrong? Is your discrimination working? Test it for different materials.

[ashefr10]

I have no particular problem but I read that I must have a phase of 20 degrees and the lowest voltage possible output waveform on my U101 and it's not the case