Generally, the clock circuit of a microcontroller uses an external passive crystal oscillator in combination with load capacitors to connect to the Xin and Xout pins of the microcontroller. The passive crystal oscillator cannot oscillate on its own, so an external resonant circuit is required to produce the oscillation signal.
However, in practical circuit design, a resistor is often connected in parallel across the crystal, and this resistor is known as the feedback resistor.
First, let's take a look at the basic principles of clock circuits. Generally, clock circuits are also known as Pierce oscillator circuits because they are simple, efficient, and stable, superior to other types of quartz crystal oscillator circuits. The Pierce oscillator requires very few components: an inverter, a resistor, a quartz crystal, and two small capacitors.
For microcontrollers, most chips typically include an inverter internally, and some may also have a parallel feedback resistor built in. This information can be found in the specific chip's datasheet. If not, you may consider adding an external feedback resistor across the crystal terminals.
Enhance stability and facilitate easier oscillation
The internal inverter in a microcontroller is a Schmitt trigger inverter, which is incapable of driving the crystal oscillator's oscillation. Hence, a resistor is connected in parallel at the inverter's two ends to complete the output signal, feeding back 180 degrees out of phase to the input for negative feedback, forming a negative feedback amplification circuit to achieve signal inversion. Additionally, the parallel resistor lowers the resonant impedance, making it easier for the oscillator to start.
At the same time, adding negative feedback resistors increases stability. In other words, even with temperature variations, adding resistors can reduce the oscillator's frequency drift. Without the feedback resistor, the crystal oscillator circuit may oscillate, but there is a risk of non-oscillation or stoppage.
Operates the inverter within its linear region
This resistor causes the inverter to operate within its linear region, turning it into a high-gain inverting amplifier and ensuring oscillation occurs. Assuming this inverter is an ideal inverter with infinite input impedance and zero output impedance, the parallel resistor ensures that the input voltage equals the output voltage, preventing the transistors within the inverter from operating in a fully saturated or fully cutoff state. Instead, they operate in the intermediate transition region with gain. Operating in the fully saturated or fully cutoff states corresponds to saturation, which has no gain, and without gain, oscillation cannot occur. The linear region of the inverter is within the shaded area in the diagram below, ensuring the working point voltage at the inverter's input is at VDD/2. This guarantees that when the oscillation signal is fed back to the input, the inverter operates in the appropriate working region.
The impedance of the oscillation circuit changes in high and low-temperature environments, and when the impedance increases to a certain extent, the crystal oscillator may experience difficulties in oscillation or non-oscillation. In such cases, it's necessary to check if the parallel feedback resistor's value is appropriate. Generally, in a clock circuit, the relationship between the crystal oscillator frequency and the feedback resistor size is as follows:
