Target ID VCO Sound

[rolling rock]

It struck me one day that the IB detector I was trying to develop really could benefit from having the VCO audio where the higher tones correspond to higher conductivity, instead of having higher tones indicating increasing target strength. To me, it's a no-brainer, why did I not think of it sooner? So I did a search here, and didn't find much except for folks wanting visual target ID indication. So I decided to revisit some old ideas regarding the RC4200 analog multiplier chip that is largely obsolete, but can still be had. Or maybe find a suitable substitute, or build a replica from scratch.

RC4200 was used in some Tesoro detectors for target ID meter indication. I imagine the principles are pretty much the same whether driving a meter or driving an audio VCO. Wouldn't it be nice to run some circuit simulations to try and get a working circuit.

I was poking around through some old MAT02 datasheet and realized that their circuit for a one quadrant muliplier-divider corresponds closely to the RC4200 architecture. It might not be so obvious from the RC4200 datasheet, because their drawing is reversed, but the NJM4200 datasheet turns it around.

The Analog Devices circuit reverses the X and Y inputs, but everything else is pretty much the same. The RC4200 looked to me like booger hell to try to figure out, but you simply replace the current sources with resistors, and pretend a positive power supply was there all along, and Za-POW, now a LTspice simulation runs like a top.

So anyway, I would just like to throw this out for what it's worth. You should recognize this as a Tesoro type sound system, except that I am not using a wired-and CD4024 output to get the rectangular pulses. But also be aware that the 555 output duty cycle requires that Q5 invert the sound impulses or your battery will get worn down FAST.

[Op04]

Is this works? Please put more understanding schematic so I can draw PCB and put here too

[rolling rock]

This could be a stand alone circuit to replace something already existing but I am trying to work it into an IB detector of my own making, where it will be featured on the main PCB. Notice I did not say "my own design", because I stole most of the bits and pieces from somebody else's design.

One of my complaints about the existing circuit I have is that the VCO output signal has a fixed LOW time, meaning that duty cycle varies as the frequency changes. So at high frequencies the higher DC consumes more power than necessary, and at low frequencies the low DC looks like it would barely activate the speaker. So, we borrow the excitation circuit from the Delta Pulse and now we have two timers, with one controlling frequency, and the other increasing DC for lower frequencies and vice-versa.

But I must remove what was the second transistor in the speaker amplifier, to UN-invert the signal because duty cycle is now "right-side up", and the volume control also needs removed, or maybe "adjusted" (some other time).

It is my belief that less power consumption can be realized by manipulating the audio duty cycle, while at the same time giving a more listenable sound.

[Op04]

Ah...IB detector I get it now. Thanks for sharing Sir

[Carl-NC]

When tone ID first came out (Teknetics 8500?) detectors were still all-analog (no micros) and tone ID was done discretely using comparators. By the time there was a "linear VDI audio" mode detectors were using micros. I can't think of a design that ever implemented linear VDI audio using a non-micro approach. Which begs the question, why not just do it with a micro? Then you get total control over how it behaves, and can even implement other modes or oddities.

[rolling rock]

My last circuit was BOGUS and I apologize to everyone. Sorry folks. Doh, I inverted the original waveform by using that second timer's output, and I neglected go back and check that the speaker output follows target response envelope, V4. Today I see that it sure enough does not. Good idea, bad implementation.

Well, Carl I guess that was just too easy, and in answer to your question: if it had worked, I think it would be nice not to have to worry about programming a darn micro, and still home-brew a target-ID sound Mo-chine! But it ain't working yet. I'll try some more, just not today. Grrr. Because today I eat crow! Oh, yummy.

Still, there might be a way I can hack this 556 idea to where it works. I'll try some more.

[rolling rock]

Improved TGT ID sound channel

I should not have implied that the circuit of my last post was a VCO (and annunciator system) which gives a medium power, relatively constant duty-cycle throughout its range (thus to hopefully optimize battery life vs sound "punch"), as frequency goes up and down in proportion to target conductivity - while also giving the operator control over threshold of detection.

These last criteria absolutely were not adequately addressed.

Here is a correction which marries the 556 output pulses and target detection envelope a lot better.

It's really not so complicated as it looks. Remember that about half of those components in my circuit are going to be needed no matter which approach is used (well, except for the PIC approach to which I say "not today").

So I'm hoping this is the final PCB section I'll have to worry about, and now I look for a coil.

[rolling rock]

Better Late Than Never?

After fiddling with this circuit for only a short time it became apparent why the various manufacturers gravitated towards a using a PIC (or similar uP) for target ID audio, or else them going with "beep" only, and nevermind the VCO target ID. There are way many trim pots to preset. Somebody would have to write detailed setup instructions for the technicians, when all the time, management would actually be wanting to eliminate those particular tech positions. So, instead they pay somebody to write and test/debug the code, and save big in the long run.

Of course that sucks if you're the (jobless and poor) technician who just happened to love twiddling potentiometers... doh.

I don't code very well, and I've got the time, so I went ahead and drew up a circuit based loosely on Tesoro Lobo, with Cibola phase network, and ANALOG target ID audio that's sort of like the proverbial red-headed step child. ahem

The transmit frequency is derived from a crystal / divider chain from which I/Q clocks are available (WHICH CAME IN HANDY), but transmit frequency is set in stone and I will need to match the transmitter to an appropriate coil/cap.

Eh, the circuit in my last post was a mess because I inadvertently omitted a base resistor, without which, something would surely smoke. But when properly constructed, it should give a nice response. I could have corrected that earlier, but I figured nobody is probably paying attention anyway.

Tentatively, I think I would like to hear vco audio between ~300 to ~1kHz, and have audio pulse duty cycle be approximately equal throughout that range. Based on LTspice results, the available adjustments to the 556 provide should provide adequate control over frequency range and duty cycle.

I have the DC set to where target detection audio draws about 20ma using a 32Ω speaker. It can easily be set to about half that, but I am not sure I would hear that as well as I should. Time will tell? There is an optional volume control (which I probably won't do without).

My layout is largely SMD but the RC4200/NJM4200 will be DIP so I can conduct incoming electrical tests on the part, due to the prevalence of CHEAP, suspected counterfeit parts out there. I mean really, really cheap. So, will they work?