Tone ID for TGSL

[dfbowers]

This may defeat the whole concept for the TGSL but could be applied to many other circuits. I was looking at the post on Lissajous patterns posted by Ivconic where we plot one signal against another on a scope, instead of plotting at single function against time.

One thing that becomes obvious (at least it's something new for me) is the relationship of signal strength for each channel vs target conductivity. The signal strength for a low conductor is much stronger on the disc channel as opposed to the AM channel and vice versa.. A ferrous target never makes it into the upper right quadrant and that's why we don't hear it when in the disc mode.

Now the case of using a Nautilus DMCIIb with dual channels we can use sound to our advantage in that the conductivity of an object can be judged my the signal intensity in ether the AM channel or the DISC channel.

So why not a simple circuit to add tone ID to The TGS or TGSL? If we compare signal strength between channels and send it a VCO seems to me it would work. The only drawback would be that it would be highly dependant upon the initial setting of the disc and GB pots.

http://www.youtube.com/watch?v=m9DuMekVwic

[dfbowers]

This may defeat the whole concept for the TGSL but could be applied to many other circuits. I was looking at the post on Lissajous patterns posted by Ivconic where we plot one signal against another on a scope, instead of plotting at single function against time.

One thing that becomes obvious (at least it's something new for me) is the relationship of signal strength for each channel vs target conductivity. The signal strength for a low conductor is much stronger on the disc channel as opposed to the AM channel and vice versa.. A ferrous target never makes it into the upper right quadrant and that's why we don't hear it when in the disc mode.

Now the case of using a Nautilus DMCIIb with dual channels we can use sound to our advantage in that the conductivity of an object can be judged my the signal intensity in ether the AM channel or the DISC channel.

So why not a simple circuit to add tone ID to The TGS or TGSL? If we compare signal strength between channels and send it to a VCO, it seems to me that it would work. The only drawback would be that it would be highly dependant upon the initial setting of the disc and GB pots.

http://www.youtube.com/watch?v=m9DuMekVwic

Sorry.. Missed component.. I have not tested or simulated this yet. Plus, LMC555 might need duty cycle reduced.. I just posted for discussion or comments.


Don

[pebe]

I think you will have to rewire U107a to make it run in inverter mode.

The outputs of U106a have open collectors, so are open circuit in the absence of enough signal. With a signal, one of them will conduct and D12 will conduct taking the +ve input of U107a low. So its output goes low. To make D12 active, the sensitivity control must have been set up so that in the absence of sign, U107a input and output are high.

With that configuration, there will be drive to TR8 with no signal and the 555 will modulate it with audio. You need the opposite.

[simonbaker]

A frosty feature!

Practically, we wouldn't want to introduce a new oscillator onto the PCB that isn't synced to the TX oscillator -- worries about phase noise.

But if you can select tones from the CD4024 circuit it might be benign. It would take a clever circuit to select monotonically rising tones as needed. Also, some kind of normalization of the channel voltages would probably be needed -- I think we really want division, not differencing. And as you mentioned, the position of the sync pulses is really part of the equation.

A microprocessor-based MD could probably do this nicely with a simple software routine. I can see why manufacturers love u-processors -- they can go crazy with features (useful or not).

-SB

[dfbowers]

A frosty feature!

Practically, we wouldn't want to introduce a new oscillator onto the PCB that isn't synced to the TX oscillator -- worries about phase noise.

But if you can select tones from the CD4024 circuit it might be benign. It would take a clever circuit to select monotonically rising tones as needed. Also, some kind of normalization of the channel voltages would probably be needed -- I think we really want division, not differencing. And as you mentioned, the position of the sync pulses is really part of the equation.

A microprocessor-based MD could probably do this nicely with a simple software routine. I can see why manufacturers love u-processors -- they can go crazy with features (useful or not).

-SB

I don't think what I proposed would be the cleanest design, like you said and utilizing the outputs of the CD4024 would make the most sense. I can see your point in division because the derived signal would be the same regardless of the amplitude of both. The big downfall of the whole idea would be that adjusting the sync pulse would screw with everything .

[simonbaker]

I don't think what I proposed would be the cleanest design, like you said and utilizing the outputs of the CD4024 would make the most sense. I can see your point in division because the derived signal would be the same regardless of the amplitude of both. The big downfall of the whole idea would be that adjusting the sync pulse would screw with everything .

It's one of those problems that seems simple at first, then gets more complicated the more you think about it. Then some day someone finds an ingenious simple solution that works well enough. A good puzzle to work on by the fireplace with brandy and spice (LTSpice that is).

-SB