TGSL Tuning & troubleshooting

[simonbaker]

[ATTACH]25184[/ATTACH]
graph TX and pin 1 and pin 7 of 353
[ATTACH]25185[/ATTACH]

That's a start.

Are you measuring the TX or RX signal as reference (your scope screen says RX)? Most useful to start with is measurements vs the TX signal.

Have you tried moving the cable in different configurations, touching it with your hand, etc. to see effect on beeping?

Do you have some RCA audio cables you can try as a substitute?

Probably not easy to help you troubleshoot without good equipment. Obvious ideas are to double-check your solder joints; make sure no solder across traces; clean resin off PCB.

I would still want to see all signals I mentioned compared to TX, to get phase relationships, although the PC scope may be interfering somewhat.

Regards,

-SB

[Davor]

android oscilloscope for 0$
http://retronics.wordpress.com/2012/...-oscilloscope/

A true sillyscope

Next time someone complains about not having an oscilloscope...

[sdesic]

android oscilloscope for 0$

http://retronics.wordpress.com/2012/07/14/the-0-android-oscilloscope/

or bluetooth android oscilloscope

http://projectproto.blogspot.com/2010/09/android-bluetooth-oscilloscope.html

[Davor]

I think that this soundcard ocilloscope can't be used for measuring exact measures
maybe Davor can explain as he used both ...

Soundcard measurements are pure AC, and unfortunately uncalibrated. You can calibrate measurements, and these softwares do support calibration, however you usually have no calibration source at hand, and such calibrations are usually only temporary. Some of the unused 4024 outputs could be used for calibration, at least I did so.

It is possible to perform DC measurements with AC instrument by chopping the input signal. This is a simple way to do it: http://lea.hamradio.si/~s57uuu/scdsp.../cheapchop.htm

With all shortcomings, I'm much happier with some oscilloscope than none at all. I do own a CRO now, but I'll also continue my familiarity with soundcard oscilloscopes. I wonder if there are any similar solutions for Android phones, as in fact they too have some sort of "soundcard" built in.

[sdesic]

I am using this proble to connect to PC audio card

[sdesic]

graph TX and pin 1 and pin 7 of 353

[sdesic]

I think that this soundcard ocilloscope can't be used for measuring exact measures
maybe Davor can explain as he used both ...


tx levels and settings of software


when I put jumper - short circuit RX coil - beeping stops

[simonbaker]

>5. TX signal voltage.
22mV

Thanks for your info.

1. For TX signal I mean voltage (measured with Oscilloscope) across the TX coil leads. It should be at least 12 V peak-to-peak, or something like that. Usually about 16 V p-p, I think.

2. When you hear beeping, can you try putting a jumper wire across the RX coil leads and tell us what happens?

-SB

[sdesic]

I think this is schematics... close


>1. TX coil inductance and resistance.
6,05mH 24,4 ohm
measured at PCB

>2. RX coil inductance and resistance.
6,71 mH 26 ohm
measured at PCB

>3. Capacitor (C6 on some schematics) across the RX coil.
was 15nF as schematics ... replaced with 1nF and it is now 10nF (103 ceramic)

>4. TX signal frequency.
sorry ... no freq meter

>5. TX signal voltage.
22mV

I will do those measurements and report tommorow.

>Usually it is possible to use unglued coils like you have without beeping problems, but it is also possible that the coils could be moving >slightly, since the TX coil is an electromagnet. But I think you had the same problem with glued coils, is that right?
that is right ... glued and no ... same thing... glued most of coils - not 6 wires and connecting cable wires

[simonbaker]

Two gates TR4 and TR5 on two channels ... total silence - no beeps ....
(only if sens pot is on 3/4 smaller beeps as it should be)
coils close to nulling point but no beeps ... only difference is connection with scope ... ??
please explain this to me if you can
[ATTACH]25171[/ATTACH]
gates of silence


and waving metal over coils
[ATTACH]25172[/ATTACH]

1. Can you use a higher resistance resistor to measure the TR4, TR5 gate voltages, and find what resistance is required to cause the beeping again?

2. In these pictures (pictures 6 and 7) why do we not see the 14.5 kHz square wave signal at the gates of TR4 and TR5? Or am I not reading the time scale correctly? It looks like 20 Hz signals. Can you confirm what you are measuring?

Regards,

-SB

[simonbaker]

The first thing that would help me is to find the schematic for the Silverdog version you built and post or link to it here so we can refer to it. Then we need to identify, on the schematic, which channel is the one that is beeping in picture 1 (the GB or the DISC channel).

These beeping problems sometimes are very subtle and difficult to troubleshoot without having the actual circuit in front of you, so your own experiments will probably be the most useful. But we will try to suggests things to try.

It could possibly have to do with the phase of the "null" signal and some modulation of the TX signal.

What again are the most accurate measurements you have for:

1. TX coil inductance and resistance.
2. RX coil inductance and resistance.
3. Capacitor (C6 on some schematics) across the RX coil.
4. TX signal frequency.
5. TX signal voltage.

There are many phase and voltage measurements I would like to see if I had the circuit. With dual trace scope, I would like to compare the TX signal to signals at output pins 1 and 7 of the LF353 op amp and the output pin 7 of the LM393 U102b and the gate of TR4, TR5 for Disc and ALL METAL modes.

I would also look at the signal that is input to LM393 U102a pin 2 while rotating the DISC pot to make sure it shifts correctly. Also look at the output of LM393 U102b pin 7 while rotating the GB pot.

That's a lot of measurements, but it is probably what I would do. We mainly need to see that the phase shifts are all correct.

Hopefully we can find something to fix.

Usually it is possible to use unglued coils like you have without beeping problems, but it is also possible that the coils could be moving slightly, since the TX coil is an electromagnet. But I think you had the same problem with glued coils, is that right?

-SB

[sdesic]

>1. What does it mean that "Amplitude Channel 2" is grayed out / disabled? Are both channels the same volts per div?
it grayed out because channel 1 and two are in sync ... same aplitude is applied to both graphs

>2. What does the amplitude dial actually determine? For example, a label says "[1/Div]", but there is also an amplitude "value" that >reads "60m" or "500m" for example. What exactly do these refer to?
aplitude of signal ... like zoom function ... to me I am just using it for a few days

>If we label your pictures 1 through 7, my first observation in picture 1 is that I see waveforms with a period of approx .25 sec, or 4 Hz. >However in picture 3, the same waveforms seem to have doubled frequency to about 8 Hz. Is that consistent with what you hear in >the beeping?
this scope is on automatic scaling of signal amplitude so can fit on screen so it is on time that size and other time ...
maybe I had ajusted it to fit screen manually or I restarted program
I hear same frequency all the time, no change

>In pictures 6 and 7, we see waveforms with periods of about 50 mS, or about 20 Hz. Is this what you hear, beeping that varies from 4 >Hz to 20 Hz?
same as above ... same freq

>3. In picture 4, would you explain exactly what points you are measuring in both graphs and what you mean by "nothing connected"?
first channel - red is not connected ... gnd is on gnd and + is not connected
lower graph - green - is connected on audio - connected one + to speaker and other one is gnd

>3. The most interesting thing is that the oscillations are present on the lower channel even with big overlap of your coils. I think we >need to focus on that channel in that case and see what we can find.
what can I do ...

thank you for you time Simon

[simonbaker]

Thanks for all the pictures, these will be helpful.

It is always difficult to read an unfamiliar oscilloscope, so I may need some help.

1. What does it mean that "Amplitude Channel 2" is grayed out / disabled? Are both channels the same volts per div?

2. What does the amplitude dial actually determine? For example, a label says "[1/Div]", but there is also an amplitude "value" that reads "60m" or "500m" for example. What exactly do these refer to?

If we label your pictures 1 through 7, my first observation in picture 1 is that I see waveforms with a period of approx .25 sec, or 4 Hz. However in picture 3, the same waveforms seem to have doubled frequency to about 8 Hz. Is that consistent with what you hear in the beeping?

In pictures 6 and 7, we see waveforms with periods of about 50 mS, or about 20 Hz. Is this what you hear, beeping that varies from 4 Hz to 20 Hz?

3. In picture 4, would you explain exactly what points you are measuring in both graphs and what you mean by "nothing connected"?

Observations:

1. From picture 1, we see "beep" signals in one channel, but no audio. This should mean the speaker is not causing feedback. The speaker is the main component that might drag down the supply voltage.

2. Small short spikes on the supply voltage rails are not necessarily a problem -- they are often present in a normal working TGSL and you would expect them when the speaker sounds.

3. The most interesting thing is that the oscillations are present on the lower channel even with big overlap of your coils. I think we need to focus on that channel in that case and see what we can find.

-SB

[sdesic]

Two gates TR4 and TR5 on two channels ... total silence - no beeps ....
(only if sens pot is on 3/4 smaller beeps as it should be)
coils close to nulling point but no beeps ... only difference is connection with scope ... ??
please explain this to me if you can

gates of silence


and waving metal over coils

[sdesic]

graph 12V and sound output with beeps
magnified disturbance so you can see what it is