# Oscillator circuit

An oscillator circuit is an electronically implemented oscillator (hence also called an oscillator for short ) for generating a sinusoidal alternating voltage .

Basic circuit of a phase shift generator for almost sinusoidal output voltages

There are, among others, the following options for setting up such a circuit:

## Vibration condition

### Feedback amplifier

Collector voltage of a short-wave oscillator immediately after switching on until the amplitude limit is activated

Amplifiers can become an oscillator with appropriate feedback . The vibration conditions, which were initially formulated in the stability criterion by Barkhausen , lead to permanent vibration on a linear feedback amplifier with a certain frequency . They clearly read:

1. The loop gain must be exactly 1 for a stable oscillation.
2. The phase shift of the feedback loop must be an integer multiple of 360 ° at this frequency.

In the theoretical model, a loop gain that even slightly exceeds the value 1 would lead to an infinite increase in the oscillation. So that the oscillator starts to oscillate independently ("by itself"), the value 1 must first be exceeded (see picture). This contradiction leads to studies of the oscillation behavior and the breakdown of the vibrations and is represented by corresponding characteristic curves. One also speaks of a hard or soft vibration insert. The physical background is essentially formed by two phenomena:

1. The gain depends on the operating point and can increase (hard vibration application) or decrease (soft vibration application) with increasing control of the amplifier.
2. If the modulation is high, the amplified signal (the oscillation) is limited.

In reality, the amplitude is limited (the power supplied to the amplifier from the power supply is finite).

### Description in detail

Principle of the feedback oscillator

Feedback oscillators consist of an amplifier and a passive, frequency-dependent network. The amplifier output feeds the input of the network. The output of the network is connected to the amplifier input (feedback). You can mentally disconnect the feedback line and instead of the closed circuit you get a transmission chain with input (E) and output (A). For the same conditions as in the closed circuit, the phase position of the output oscillation (φ3) must match the phase position of the input oscillation (φ1) (phase condition).

If the amplifier itself causes a phase shift of 180 ° and the signal propagation time is zero, the network must cause a further phase shift by 180 ° for at least one frequency in order to achieve a total phase shift of 360 ° = 0 °. The frequency-dependent phase shift is shown in a phase response diagram.

In the case of the phase shift oscillator, the network consists of (at least) three RC elements connected in series (low or high passes). If each of these elements causes a phase rotation of 60 °, three RC elements are sufficient for a total phase rotation of 180 °. If the amplifier is not overdriven, the generated alternating voltage is sinusoidal.

### Negative resistance

A lossy resonant circuit can be undamped by a component with negative differential resistance , for example a tunnel diode or lambda diode , and then generates alternating voltage. The condition is that the total resistance is zero. The energy required for operation is supplied by an external power supply unit or a battery.

In the case of other oscillator topologies, which are based, for example, on the negative characteristic curve such as the relaxation oscillator (see below), the stability criterion has no direct reference.

## quality

Phase noise of a PLL oscillator in the SW range. In the case of a quartz oscillator, the corresponding characteristic would be an almost vertical line on the left edge of the picture.

The quality of an oscillator is generally assessed according to the stability of amplitude , frequency and phase . If the fluctuations can only be described statistically, they are referred to as noise . Phase noise ( jitter ), which characterizes the sensitivity of a heterodyne receiver in the immediate vicinity of a strong signal, is the only common term used here . Stability against fluctuations in temperature and supply voltage is also important , although there are significant differences: The frequency of relaxation and ring oscillators is very sensitive to changes in the operating voltage. In the case of oscillators with a resonance circuit, this dependency is very low and in the case of crystal oscillators it is negligible.

Another important criterion for measuring devices, for example, is the accuracy with which the desired curve shape is generated. In the case of sine oscillators , this can be described quite simply by the distortion factor . Although primarily used for sinusoidal oscillations, this criterion also applies to other signal forms.

## application

Unmodulated oscillators are used to generate the clock frequency of computers or electrical clocks.

With modulated oscillators, amplitude, frequency or phase are influenced within certain limits by additional components. This allows messages to be transmitted through modulation . This is used to

## Categorization

### Resonance oscillators

In the case of a resonance oscillator , the frequency generated is determined by an oscillating circuit , an oscillating crystal or a ceramic resonator . The resonance oscillator usually delivers a frequency-stable sinusoidal oscillation .

This also includes the magnetron , although it is also a time-of- flight oscillator .

The main area of ​​application of the sine wave oscillators is radio technology.

It is always ensured that the resonator has a sufficiently high quality factor so that the bandwidth of the signal generated is limited to the close vicinity of the resonance frequency. This reduces the proportion of harmonics in the output signal, even if the amplifying element, for example a transistor, is overdriven and actually generates strong harmonics. In contrast to a Wien-Robinson oscillator  , for example, resonance oscillators deliver a well sinusoidal signal even without amplitude stabilization .

### Runtime oscillators

The Gunn diode oscillator uses the transit time of electrons in a crystal

In the case of transit time oscillators, the transit time of pulses in certain circuit parts determines the period of oscillation and thus the frequency. The ring oscillator with its inverter chain is used here as an example . But oscillators such as the reflex klystron and a Gunn diode oscillator also belong to this category, although both have oscillating circuits. In phase shift oscillators, the signal propagation time is generated by RC elements. There is an area of ​​overlap with the relaxation oscillators because the time-determining RC elements there can also be viewed as delay elements. The frequency stability is generally rather mediocre.

### Relaxation oscillator

A relaxation oscillator is a tilting oscillator . It is not an oscillator in the strict sense, as it normally does not generate a sine wave. The frequency is typically determined by the discharge processes of a capacitor in an RC element. When the capacitor voltage reaches a certain value, the output voltage is switched (it "flips") and the capacitor is recharged. The best-known circuits are multivibrator and tilting oscillator. Square or triangular waves can be picked up at suitable points in the circuit . Since, in addition to an RC element, the threshold voltage of the trigger stage involved also influences the stability, relaxation oscillators are much less stable than resonance oscillators and can therefore not be used in radio technology. This easy ability to be influenced is used in electronic sirens or in digital measurement technology, for example in voltage-frequency converters.

Classification of oscillator circuits
according to principle according to the waveform by name of the inventor according to purpose

## Oscillator circuits with differential amplifiers

Modern oscillators avoid the disadvantages of the classic oscillator circuits invented about 100 years ago (such as the Meißner circuit , Hartley circuit , Colpitts circuit ), which can generate undesirable parasitic oscillations of a few gigahertz and tend to low-frequency breakover oscillations if the components are not dimensioned well , or have a noticeably different waveform from the sinusoidal shape.

One possible circuit uses a differential amplifier with two transistors and is characterized by its very good-natured behavior (see differential amplifier oscillator ). The pictures below show a variant with NPN transistors with which - depending on the data of the resonant circuit - frequencies in the range 0.05 MHz to 40 MHz can be generated without changing other components. In the other circuit, PNP transistors were used and the values ​​of the components were dimensioned for frequencies in the range from 1 Hz to 500 kHz. In this circuit, the resonant circuit is at zero potential, which is advantageous for some applications (as a rule, the negative pole is the reference point for all measurements).

The spectral purity of the generated oscillation is better if the feedback is so weak that it is just sufficient for reliable oscillation. With differential amplifiers, the amplitude limitation also sets in more gently than with other oscillator circuits. This reduces the harmonic content.

Commons : Electronic Oscillators  - Collection of pictures, videos and audio files

## Remarks

1. In "Fachwissen Elektrotechnik" Verlag Handwerk und Technik, as well as Tietze / Schenk, oscillators are consistently referred to as sine wave generators, while astable multivibrators are called signal generators. Since "generate" means to produce, oscillare to rock, the appropriate generic term would be "signal generators". In the absence of clear clarification, Wikipedia is currently using "electrical oscillator" as a generic term. Square-wave generators and the like could therefore also be thematically separated from the oscillators.
2. The difficulty here lies in the word "suitable". With clear exaggeration, practitioners put it: "An oscillator never vibrates, an amplifier always."
3. Non-linear properties of the amplifier can lead to a voltage building up on a capacitor that shifts the operating point so far that the oscillations break off. In such a case (sinusoidal) vibrations can arise that are modulated with tilting vibrations (relaxation vibrations). The simple linear model is therefore not sufficient in practice to adequately describe an oscillator.

## Individual evidence

1. Vienna Bridge Oscillator ( Memento from June 10, 2007 in the Internet Archive ) (PDF; 65 kB)
2. Kent H. Lundberg: Barkhausen Stability Criterion . , November 14, 2002 (English)