The principle and function of crystal oscillator

Crystal oscillator is a commonly used clock component in circuits. Its full name is crystal oscillator. In the microcontroller system, the crystal oscillator plays a very important role. It combines with the internal circuit of the microcontroller to generate the clock frequency required by the microcontroller. The execution of all instructions of the microcontroller is based on this. The higher the clock frequency provided by the crystal oscillator, the faster the microcontroller runs.

A crystal oscillator uses a crystal that can convert electrical energy and mechanical energy to each other and works in a resonant state to provide stable and accurate single-frequency oscillation. Under normal working conditions, the absolute accuracy of ordinary crystal oscillator frequency can reach 50 parts per million. Advanced ones have higher accuracy. Some crystal oscillators can also adjust the frequency within a certain range by applying an external voltage, which is called a voltage-controlled oscillator (VCO).

The function of a crystal oscillator is to provide a basic clock signal for the system. Usually a system shares a crystal oscillator to facilitate synchronization of various parts. Some communication systems use different crystal oscillators for baseband and radio frequency, and synchronize them by electronically adjusting the frequency.

Crystal oscillators are usually used in conjunction with phase-locked loop circuits to provide the clock frequency required by the system. If different subsystems require clock signals of different frequencies, they can be provided by different phase-locked loops connected to the same crystal oscillator.

Next, I will introduce the role and principle of the crystal oscillator in detail. The crystal oscillator generally adopts the capacitor three-terminal (Colpitts) AC equivalent oscillation circuit as shown in Figure 1a; the actual crystal oscillator AC equivalent circuit is shown in Figure 1b, where Cv is used to adjust the oscillation frequency, which is generally achieved by using a variable capacitance diode plus different reverse bias voltages, which is also the mechanism of voltage control; the equivalent circuit of the crystal after replacing the crystal is shown in Figure 1c. Among them, Co, C1, L1, and RR are the equivalent circuits of the crystal.

Crystal Oscillator Circuit Diagram

Analysis of the entire oscillation tank shows that using Cv to change the frequency is limited: the entire tank capacitance C = Cbe, Cce, Cv that determines the oscillation frequency is connected in series, then connected in parallel with Co and then in series with C1. It can be seen that: the smaller C1 is, the larger Co is, and the smaller the effect of Cv on the entire tank capacitance is when it changes. Therefore, the frequency range that can be "voltage controlled" is also smaller. In fact, since C1 is very small (1E-15 order of magnitude), Co cannot be ignored (1E-12 order of magnitude, a few PF). Therefore, when Cv becomes larger, the effect of reducing the tank frequency becomes smaller and smaller, and when Cv becomes smaller, the effect of increasing the tank frequency becomes larger and larger. On the one hand, this causes the nonlinearity of the voltage control characteristics. The larger the voltage control range, the more severe the nonlinearity; on the other hand, the feedback voltage (the voltage on Cbe) allocated to the oscillation becomes smaller and smaller, and finally causes the oscillation to stop. Through the schematic diagram of the crystal oscillator, you should have a general understanding of the role and working process of the crystal oscillator. The higher the number of overtones of the crystal oscillator, the smaller its equivalent capacitance C1; therefore, the smaller the frequency change range.

The clock sources of microcontrollers can be divided into two categories: clock sources based on mechanical resonant devices, such as crystal oscillators and ceramic resonant tanks; RC (resistance, capacitance) oscillators. One is the Pierce oscillator configuration, which is suitable for crystal oscillators and ceramic resonant tanks. The other is a simple discrete RC oscillator.

The method of using a multimeter to measure whether the crystal oscillator is working is to measure whether the voltage of the two pins is half of the chip's working voltage. For example, if the working voltage is +5V of a 51 microcontroller, then it is about 2.5V. In addition, if you touch the other pin of the crystal with tweezers, the voltage changes significantly, which proves that it is oscillating.

There are two types of crystal oscillators: SMD and DIP, namely, patch and pin types.

Let’s talk about DIP first: commonly used sizes are HC-49U/T, HC-49S, UM-1, UM-5, these are all in MHZ units.

Let's talk about SMD: there are 0705, 0603, 0503, 0302, which are divided into four solder joints and two solder joints. However, the smaller the more expensive, and if it is very small, it is impossible to make a crystal oscillator with a higher frequency.