Bad sound on TGSL IGSL

[cristim]

Hello,
I have a problem with my IGSL, the sound is not clear, it sounds like a rustle/noise. The sound is verry faint, and to slightly increase I shorted the R68 and R67
My kit it came with 10uF capacitors instead of 4.7uF, but I think that is not a problem. (and without 10pF and 100pF capacitors, but I subsequently bought them)
I measured all voltages and all are ok!
I put a video here:

Do you have any idea about this problem?
Thank You!

PS. I do not have an oscilloscope!

[kt315]

sounds me like the transformer of Tesla. your main problem you are testing on floor instead you have to hang up the coil being free in air.

[cristim]

Its the same sound down and up above the floor.

[Davor]

Of course it does not work properly - the coil is not nearly balanced as it should be. Your Rx is almost constantly in saturation.
What kind of instruments do you have? You can use a PC soundcard with some oscilloscope program to balance your coil, but you may as well do it with an AC voltmeter capable of measuring in millivolt range and at 8kHz.

Your coil must be balanced to at most 100mV at the preamp output.

[cristim]

I try to balance the coil by ear , moving RX over TX very slowly (milimetric), and I thought that when I get the maximum distance in detection of a coin then the coil is well balanced. In all over positions of the coils, the sound is exactly the same, only distance detection is different. My voltmeter have 200mv range, but in DC only. I will try to use a mic. and a oscilloscope program...thx.
The coil was made with RX = 107 windings and TX = 98, using a 0.25 Cu wire (like in tutorial)
If I fail with virtual oscilloscope, I will buy a LCR meter to make a proper coil.
Untill now I did 2 coil and the result is the same...bad sound!

[Davor]

OK, bad sound can be a consequence of something entirely different, however, achieving a workable induction balance is a very touchy business. You just can't achieve it by chance.
Make a 10/1 resistor divider, say 10k in series with 1k for a PC oscilloscope. I did mine that way. The exact calibration will not be possible, but you can use a negative voltage driver as a pulsed 8V source for not-so-accurate calibration. You must achieve a minimum at Rx coil, so accuracy is not your primary concern.

[hugo]

sounds me like the transformer of Tesla. your main problem you are testing on floor instead you have to hang up the coil being free in air.

Hello Cristim,
I agree with Davor, it's a very, very bad idea to put the coil on the floor because of the iron located into the concrete !
This is the best way to saturate the RX coil.
Try instead, to put it on a plastic/wooden table or chair or anything you want but not metallic stuff at a minimum of 80cm from the floor.
Good luck,
Hugo

[cristim]

I ordered a new multimeter that can measure mV in AC and frequency. I hope I can manage correctly to balance the coil, although I think that could be another cause.
I ordered some CD4024 (U2) and 4066 (U11) to replace them; I received HCF4024 and HCF4066 version from silverdog, but I do not think there will be any difference.
And I buy a radiator for LM7808 because it heats pretty cool. For testing, I use a pwr. sup. 12V DC, 700mA
I will return when I get it all!
Thanks all!

PS. When you say: saturate the RX coil, its mean to nulling the coil? Or something else?

[hugo]

With a good balanced coil, you can get as low as 20-50 mV at the output of the preamp output.

When you say: saturate the RX coil, its mean to nulling the coil? Or something else?

I mean that a metallic device into the coil field increase the output signal.
You can't adjust your coil in this condition.
Take a piece of wood, stick the TX coil in it and play only with the RX coil to acheive a correct balance.
Anyway, you have to re-do the same into your final coil housing.
Hugo

[Davor]

The preamp has a gain of ~50 and you have only ~6V swing. Hence everything over 100mV at coil will surely saturate Rx preamp. You simply must balance your coil.

[cristim]

I think everything is clear now. Expect the multimeter and then I'll post the result.
Thanks all!

[cristim]

I managed to fix the problem. It seems that I forgot to connect the diodes D1 and D2 with 5, 6 or 9 pin of CD4024
I succeed to nulling the coil (aprox. 5mV) and now everything is ok, the sound is clear and distinct between ferrous and nonferrous.
Have to think now about what solution to use for coil encapsulation, and how to build a case for coil. It will be very difficult to keep nulling given that with any slightest touch is no longer retains minimum value.
I saw on a website that someone used a polystyrene plate, I think I'll try this method.
What kind of epoxy should I use? Something that can be found in Europe/Italy.
Thanks!

[Davor]

You could try some ready made shells like these: http://www.geotech1.com/forums/showthread.php?17596
I know for sure that Petravka also has them, but I have not seen him around for some time now.
If possible, try not to commit any of the coil wires to the ground when you make your own coils. It gives you advantage over the commercial coils that can't be un-grounded once they are potted in epoxy. You always may ground wires in a connector or a rig, but not un-ground them from epoxy.
Ungrounded wires also enable balanced operation.

[chuubaba]

My bioard was making faint beeps while Tx and Rx translated beautifully on scope but then i decided to change the 330k resistor in audio driver circuit to 330 ohm. Now its completely silent plus the 330 ohm resistor is getting super hot. Help me restore sound in to my tgsl