Decoupling Capacitor Overview
In some literature, decoupling capacitors are considered to be bypass capacitors. In other literature, the difference between decoupling capacitors and bypass capacitors is that: “Bypass capacitors take the interference in the input signal as the filtering object while decoupling capacitors take the interference of the output signal as the filtering object to prevent interference signal from returning to power.”
From the name, decoupling is the effect of removing coupling. So what is coupling, and why does coupling occur?
The coupling here refers to the coupling between the output of the previous stage and the input of the subsequent stage. The so-called coupling refers to “in a digital circuit, when the circuit is converted from one state to another state, a large peak current will be generated on the power line, forming a transient noise voltage, which will affect the normal operation of the front stage. This is the coupling.” In this understanding, it is very close to the second cause of power supply voltage fluctuations mentioned earlier, which is caused by the increase in the current drawn by the subsequent stage device, affecting other devices.
The role of the decoupling capacitor to solve this problem is to act as an energy storage battery to meet the current change of the drive circuit, thereby avoiding mutual coupling interference.
In summary, decoupling capacitors have two roles. The first is a function similar to the bypass capacitor, which bypasses the high-frequency noise output by the device; the second is to act as a storage capacitor to provide power when the current required by the load suddenly increases to meet the current change of the drive circuit. This is very similar to the function of the bulk capacitor, and the difference between the two will be discussed later.
Decoupling Capacitor VS Bypass Capacitor
An important electrical characteristic of a capacitor is “passing AC, blocking DC”, and the formula for calculating its reactance is as follows:
Z=1/2ΠfC, where f is the signal frequency
The function of the bypass capacitor is to bypass the high-frequency noise in the system to GND. Generally, some capacitors with small capacitance (typical value 0.1uF) are connected in parallel between the power supply pin and GND, as shown in the figure below, to short-circuit high-frequency noise to GND, so as to prevent noise from entering the power supply pin of the device.
Filtering out high-frequency noise is the most important function of the bypass capacitor, but if you consider how the capacitor accomplishes this function, or how the capacitor passes AC and blocks DC. In fact, the essence is that the capacitor is an energy storage device for electrical energy. When the voltage difference between the two poles increases rapidly, the capacitor is charged; when the voltage difference decreases, the capacitor discharges. From this point of view, the bypass capacitor can also be regarded as a small energy reservoir, which is similar to the decoupling capacitors and bulk capacitors introduced later.
Decoupling Capacitor VS Bulk Capacitor
The purpose for bulk capacitors is quite clear. bulk capacitor is obvious, that is it is when the load that is powering it requires a significant current immediately it will supply enough flow to the circuit in order to ensure that the voltage is stable. voltage that powers the supply which is the same as that of the reservoir for energy. Thus, the bulk capacitor usually selects the polar electrolytic capacitor which has greater capacitance. It is typically arranged parallel to the output pins in the controller.
The energy storage capacity of bulk capacitors is like that of decoupling capacitors. So what is the distinction between them? There isn’t a significant distinction between them in terms of their function, however the bulk capacitor is larger in space, typically covering a large area that is able to store and have additional capabilities. The decoupling capacitor could be called a local device and every chip has its own capacitor for decoupling. In comparison to bulk capacitors, decoupling capacitors are smaller energy storage, but they respond quicker. Decoupling capacitors have to be placed near to signals with high frequency, and their distance should be adequate. This is something bulk capacitors are unable to accomplish due to their larger dimensions. If these pins move rapidly the capacitors that decouple them next to them will provide enough power.
Schematic of bypass capacitor filter capacitor and decoupling capacitor
- The bypass capacitor is primarily used to filter the input signal. Its primary function is to decrease the magnitude of the ripples in the circuit, in order to guarantee the smooth functioning in the circuit.
- The filter capacitor is used to filter the power supply. Its purpose is to minimize the magnitude of the ripples from the power supply and to help ensure the proper operation of the circuit.
- A decoupling capacitor principally blocks the signal interference of the output signal, thereby acting as the filtering device; it has two primary purposes; 1. Energy storage. If the current transient of the load changes the capacitor discharges the load and functions like a power source 2. Impedance. It is mostly utilized to lower the AC impedance of the power system.
How to Choose the Right Decoupling Capacitor?
In the following part, Easybom will elaborate on decoupling capacitor selection.
The decoupling capacitor is a capacitor installed at the power supply end of the component in the circuit. This capacitor can provide a relatively stable power supply, and at the same time, it can reduce the noise coupled to the power supply end of the component, and indirectly reduce the noise impact on other components.
There are many types of decoupling capacitors on the market, each with different electrical characteristics, polarity, and cost. Below is some common decoupling capacitor information to help you choose the right decoupling capacitor for your practical application.
- Small size, low cost
- Low ESR (equivalent series resistance)
- Limited capacitance range
- Poor temperature stability, capacitance value varies with temperature and voltage
- High-frequency products
- Wider range of values
- lower cost
- Leakage current increases with the increase of temperature and voltage
- Short life
- Consumer Products
3. Aluminum-polymer capacitors
- Very low ESR
- Relatively small enclosure
- Performance deteriorates rapidly when the temperature rises to 105°C or above
- Laptops, flat panel monitors, digital switches
- Low ESR
- Very stable and accurate
- Typically limited to 50 V or less
- Risk of fire due to reverse voltage connections
Military communications, aerospace
Placement of Decoupling Capacitors in PCB Layout
Decoupling capacitor placement will be discussed below.
What happens if you skip decoupling capacitors in your design?
Without decoupling capacitors, the onboard microcontroller will not function properly because voltage fluctuations send it into the power-down mode, which resets it. Any attempt to get reliable ADC conversions is futile, as the analog voltage source is barely stable. If you were to send a PCB to the field without the decoupling capacitors installed, you would run into a lot of weird issues due to the greater electrical noise.
So, would put a few decoupling capacitors at random on the PCB do the trick?
The placement of decoupling capacitors is critical to mitigating voltage fluctuations. The effect will be minimal if the capacitors are not placed in the correct position. In some cases, the wrong placement of decoupling capacitors can be a problem in itself, as it can absorb EMI coupled to copper traces.
What should the decoupling capacitors be placed in the right place?
decoupling radius of Capacitors
A major issue in capacitive coupling involves the radius of decoupling that the capacitor has. The majority of data say that the capacitor must be placed close to the circuit as it is however, the majority of information is discussing the distance of the placement from the standpoint of reducing the inductance of the loop. In fact, reducing the inductance is a major reason but there’s an significant reason that the majority of the information does not discuss, and that is the issue with its decoupling radius. In the event that the capacitor gets located too far away from the chip, or beyond it’s decoupling range, it will lose its decoupling power.
The most effective way to comprehend how the distance of the decoupling will affect you is to examine the phase relationship between sources of noise and capacitive compensation. If the demand for current on the chip fluctuates the voltage disturbance is created in a specific area within the power plane. In order to compensate for this change in voltage (or voltage) it is necessary for the capacitor to initially detect this voltage disturbance. There is a time limit for signals to travel through the medium. Therefore, there is a gap between the occurrence of an isolated electrical disturbance as well as the capacitor’s sense of disturbance. In the same way, a delay needed to allow the compensation current of that capacitor to travel to the affected area. This is why the inconsistency between the source of noise and capacitive compensatory current has to be the result of.
A specific capacitor is the best in absorbing noise that is at exactly the same that is the frequency that we are able to determine this phase relationship. If the frequency of self-resonance is F and the wavelength be l. The compensation current equation can be expressed as:
Of these A is the current’s amplitude and R is the distance between the area that is to be compensated by the capacitor while C represents the speed of signal propagation.
If the distance between the area of disturbance to the capacitor exceeds 4 and the timing of the current compensation will be p which is precisely 180 degrees from that of the source of noise, which is, totally out of the phase. When this happens the compensation current does not work as the decoupling function is ineffective which means that the compensation energy can’t be transferred in a timely manner. To effectively transfer the energy of compensation there should be a phase gap between source of noise as well as the current used to compensate must be as minimal as is possible as possible, and should be within exactly the same direction. The further away the distance and the lower the difference in phase and the greater compensated energy will be transferred. When the distance falls to zero 100 percent of the compensated energy goes to the area of disturbance. This means that the source of noise to have the closest distance is possible in proximity to the capacitor and much smaller than l/4. In real-world applications the best distance to be managed between l/40 and l/50. This is an empirical measurement.
For instance, 0.001uF ceramic capacitor, If the total parasitic capacitance after it is installed onto the board 1.6nH The frequency of resonant after installation is 125.8MHz and the resonant frequency is 7.95ps. If the signal travels at 166ps/inch over the board the length of the signal of the signal is 47.9 inches. Its capacitive radius of decoupling is 47.9/50=0.958 inches that is roughly equivalent the length of 2.4 cm.
In this instance will only be able to be able to compensate for noise from the power supply within 2.4 centimeters of the capacitor, meaning that the decoupling radius of 2.4 centimeters. Different capacitors come with different resonant frequencies and decoupling radiuses that differ. Large capacitors are a good choice because their resonant frequency is extremely small and their wavelength is extremely long, their decoupling radius is huge this is why we don’t pay any focus on the location for large-sized capacitors circuit boards. Small capacitors, because of their small distance to decouple, they must be placed as close as they can to the chip that has to be separated from the chip. This is a point that is frequently repeated in all materials. The capacitor should be put near to chips as they can.
To summarize in determining an appropriate decoupling capacitor, elements to be considered are factors like the capacity of the capacitor, its ESR, ESL value, and the resonant frequency. While layout, you must determine the amount of decoupling capacitors depending on the amount of IC power pins as well as the space around the layout. A coupling radius is determined by which layout location.
How do I Determine the Decoupling Capacitor Value?
The primary purpose behind decoupling is to keep an acceptable voltage range within the acceptable error limit, in spite of the circuit’s rules and the requirements for fluctuations in current.
In this way, the capacitor C in the capacitor decoupling needed by an IC is calculated.
U is the maximum reduction that can be made in the power bus’s actual voltage, which is V.
I is the maximum amount of amount of current that can be achieved in the A (amps);
It’s the time it takes to keep it running.
Method for calculating capacitance of capacitors decoupled: It is suggested to choose a capacitance greater than 1/m times the equivalent open circuit capacitance.
This is the highest percentage of power bus voltage changes that are allowed for those power lines of IC. In general, the datasheet of the IC will provide specific values for parameters.
The equivalent open-circuit capacitance can be described as C=P/(fU^2)
In the formula In the formula: In the formula: P —-the total power dissipated by the IC In the formula: Then, U —-the highest DC power supply of the IC as well as F —-the speed of clocking for the IC.
When the equivalent capacitance of the switched is established, multiply it by a figure larger than 1/m, to calculate the total decoupling capacity that is required by IC. Divide it by the power output pins linked on the same bus and then determine the value of the capacitor that is installed close to all power pins that are connected to the bus power source.
The reasons for choosing various configurations of capacitance capacitors decoupling:
When designing capacitors for decoupling, a variety of capacitances are typically employed (usually that differ in two-to-3 orders of magnitudes for example, 0.1uF as well as 10uF). The primary goal is to disperse the series resonance in order to achieve a lower impedance over greater frequency range.