Hi, folks, I have an interesting historical question that requires some technical expertise in electronics to answer. I am attempting to find out what would qualify as the first truly electronic musical instrument (or even just sound producing device in general). In order to do this, I would need to identify the first properly electronic oscillator.

As I understand it, a circuit must contain at least one active device to qualify as truly electronic. Electromechanical oscillators (like the string-pickup system in an electric guitar or the tonewheel-pickup system in the Hammond organ) use passive components at the oscillator stage. They require electronic amplification in order to be heard, true, but the initial vibrations are produced electromechanically, not electronically.

So what was the first electronic oscillator? Was it tube-based or does that honor go to some other component? I ask because I don't know if the electric arc qualifies as active or not...

In 1899, more than a decade before the invention of the vacuum tube, William Duddell discovered that if he placed an LC circuit across the electric arc of a carbon lamp, a musical tone would be produced. The arc's negative resistance excited audible oscillations in the tuned circuit at its resonant frequency, which could be heard as a musical tone coming from the arc itself.

Was the arc acting as an active device in this case? Or was it still passive?

I appreciate any insight here! Thanks!

 

I would say the arc was an active device since it was consuming power to produce audible power.

Preceding that by a few years was the ELECTROMAGNETIC/CARBON AMPLIFIER which had many patents by 1896.
This was used for several years to amplify long distance telephone signals before valve amps were developed.
But its limited frequency response would have made it a poor candidate for music amplification.

 

Also an early sound producing device came after the discovery of the first solid state device/rectifier, the Galena Diode in 1898, shortly after this the Cat's whisker radio became popular.
Max.

 

What about the sounds that have been produced by piezo electric effect around earthquakes? I heard it only once myself in Mexico, the locals said they often saw strange glowing lights close to the ground at the same time so probably was an electric field. It certainly didn't sound "mechanical" like rock moving or breaking, more like a whistle that varied in pitch for about 6 seconds. So in a roundabout way, maybe the planet is its own music box.

 

not too sure when it was invented, but putting a high voltage modulated with sound across a candle flame will also make sounds.

 

I would say it was the spark (or charge) of static electricity. Well studied by many in the late 1600s to early to mid-1700s.

https://en.wikipedia.org/wiki/Stephen_Gray_(scientist)

 

Hi, folks, I have an interesting historical question that requires some technical expertise in electronics to answer. I am attempting to find out what would qualify as the first truly electronic musical instrument (or even just sound producing device in general). In order to do this, I would need to identify the first properly electronic oscillator.

As I understand it, a circuit must contain at least one active device to qualify as truly electronic. Electromechanical oscillators (like the string-pickup system in an electric guitar or the tonewheel-pickup system in the Hammond organ) use passive components at the oscillator stage. They require electronic amplification in order to be heard, true, but the initial vibrations are produced electromechanically, not electronically.

So what was the first electronic oscillator? Was it tube-based or does that honor go to some other component? I ask because I don't know if the electric arc qualifies as active or not...

In 1899, more than a decade before the invention of the vacuum tube, William Duddell discovered that if he placed an LC circuit across the electric arc of a carbon lamp, a musical tone would be produced. The arc's negative resistance excited audible oscillations in the tuned circuit at its resonant frequency, which could be heard as a musical tone coming from the arc itself.

Was the arc acting as an active device in this case? Or was it still passive?

I appreciate any insight here! Thanks!

In the pioneering days of wireless, they eventually developed on from the spark transmitter with a rotating wheel that had "teeth", these passed a charged contact at a rate of a few hundred times per second and produced an intermittent arc.

It was about as close as they would get to true CW before electronic means made oscillators possible.

Depending how fast the motor, and how many teeth - some produced a decidedly "musical" note.

 

Thanks for the interesting responses, guys.

I guess I'm just trying to figure out whether electric arcs (and spark gaps, for that matter) are considered active devices when paired with a tuned circuit. It would seem they consume power, but the literature I've seen is unclear about their passivity status. Is the spark-gap in a resonant transformer like a Tesla coil an active device? The wikipedia article on electronic components (https://en.m.wikipedia.org/wiki/Electronic_component), for instance, lists carbon arcs and spark gaps as passive. It's a bit confusing for an electronics novice like myself...

 

You can describe (somewhat indirectly) how Spark gap (electric arc) devices operate using dynamic negative resistance. The energy from the electrical source is absorbed by the negative resistance and a portion of this energy is released to the LC circuit to sustain oscillation by canceling the positive dissipative resistance of the LC components. If the arc circuit is operated in this negative resistance region it can be considered to be a active device (like a Gunn or IMPATT diode) as negative resistance can amplify signals. Whether it's active or passive depends on the circuit.

An electric arc has a non-linear relationship between current and voltage. Once the arc is established (either by progression from a glow discharge [9] or by momentarily touching the electrodes then separating them), increased current results in a lower voltage between the arc terminals. This negative resistance effect requires that some positive form of impedance—an electrical ballast—be placed in the circuit to maintain a stable arc. This property is the reason uncontrolled electrical arcs in apparatus become so destructive, since once initiated, an arc will draw more and more current from a fixed-voltage supply until the apparatus is destroyed.

https://en.wikipedia.org/wiki/Electric_arc

 

In 1899, more than a decade before the invention of the vacuum tube, William Duddell discovered that if he placed an LC circuit across the electric arc of a carbon lamp, a musical tone would be produced. The arc's negative resistance excited audible oscillations in the tuned circuit at its resonant frequency, which could be heard as a musical tone coming from the arc itself.

Was the arc acting as an active device in this case? Or was it still passive?

I would call this passive. Like ringing a bell, there was an external energy source containing many harmonics that was bandpass filtered to a narrow range of frequencies or single frequency by the resonant properties of the LC tank (or the bell). In both cases, sustained output is not possible without energy input already at the output frequency. This is completely different from an oscillator circuit that runs on pure DC and makes what might appear to be a similar output in some situations but is actually significantly different.

ak

 

Thanks for the interesting responses, guys.

I guess I'm just trying to figure out whether electric arcs (and spark gaps, for that matter) are considered active devices when paired with a tuned circuit. It would seem they consume power, but the literature I've seen is unclear about their passivity status. Is the spark-gap in a resonant transformer like a Tesla coil an active device? The wikipedia article on electronic components (https://en.m.wikipedia.org/wiki/Electronic_component), for instance, lists carbon arcs and spark gaps as passive. It's a bit confusing for an electronics novice like myself...

With the rotating spark gap I mentioned, they found they could transmit further without increasing power.

The original spark transmitter just produced a big back emf, the arc drawn by the rotating gap was a very crude approximation to resonating what eventually got implemented as a tuned circuit.

 

I would call this passive. Like ringing a bell, there was an external energy source containing many harmonics that was bandpass filtered to a narrow range of frequencies or single frequency by the resonant properties of the LC tank (or the bell). In both cases, sustained output is not possible without energy input already at the output frequency. This is completely different from an oscillator circuit that runs on pure DC and makes what might appear to be a similar output in some situations but is actually significantly different.

I disagree.
To me a bell ringer is an active device, not passive. It converts electricity to sound. How can that be called passive?
Just because it's different from an oscillator circuit using an electronic amplifier doesn't make it passive.

 

If it needs an outside stimulus to keep producing sound it is passive.

 

I disagree.
To me a bell ringer is an active device, not passive. It converts electricity to sound. How can that be called passive?
Just because it's different from an oscillator circuit using an electronic amplifier doesn't make it passive.

Also the argument that the bell ringer hammer (or the electric arc) just transfers already existing energy at the frequency of the bell (or LC circuit) shows the problem of looking at circuit elements in isolation instead of the entire physical system.

 

If it needs an outside stimulus to keep producing sound it is passive.

That's a pretty wide description that would make everything but perpetual motion machines passive.

 

If it needs an outside stimulus to keep producing sound it is passive.

But I don't see how a solenoid used to ring a bell is fundamentally different from an oscillator electronics and speaker use to generate the same sound.
At a system level they both require an external electrical energy source to generate the sound.
If you put them both in a black box you couldn't determine which was which from their input and output.

 

But I don't see how a solenoid used to ring a bell is fundamentally different from an oscillator electronics and speaker use to generate the same sound.
At a system level they both require an external electrical energy source to generate the sound.
If you put them both in a black box you couldn't determine which was which from their input and output.

I think things have wandered a bit. I was speaking of the bell itself, not the bell-ringing system. The bell is a passive resonant bandpass filter just like an LC tank, the genesis of this thread. When excited by external energy in the form of an impulse function, it responds only to its frequencies. I think this is completely different from an oscillator circuit that runs on DC. If you stand next to a bell and push on it with constant force, the mechanical equivalent of DC, there is no sound.

ak

 

FWIW saturable reactors achieved amplification prior to the advent of electron tubes and semiconductors -- though I'm uncertain as to the applicability of said observation to this context

 

Thank you all for the interesting answers...

AK, since you argue the components are passive, would you consider the singing arc, Tesla coils, spark gap transmitters, etc. to be electronic? If there isn't an active device in the circuit, then it's just an electrical device-- not properly electronic, no?

It would seem a lot of this hinges on different conceptions about what is active. Some people say gain, and if arcs and spark gaps can't provide that, then it makes sense that they would be considered passive. Some people say active means the component requires power to operate properly, some people say active means the component can control electricity with electricity... I guess it just depends.

 

FWIW saturable reactors achieved amplification prior to the advent of electron tubes and semiconductors -- though I'm uncertain as to the applicability of said observation to this context

I've seen a saturable reactor used to regulate the B+ rail in an early VGA multisync monitor.

They were also used in the dynamic convergence unit on the old delta-gun CTVs. Some sets also had one in the pincushion correction circuitry.