Clock Circuit - The Heart of the Computer

Publisher:NatureLoverLatest update time:2013-10-21 Source: eefocus Reading articles on mobile phones Scan QR code
Read articles on your mobile phone anytime, anywhere
All digital circuits need to rely on clock signals to synchronize the operation of components. The number of times the circuit can operate per unit time depends on the frequency of the clock. Therefore, the frequency of the clock operation is regarded as a performance indicator of the system operation.

The demand for the motherboard clock circuit

. Readers familiar with hardware should know that the main components on the motherboard, such as the processor, chipset and main memory, each have their own working clock. The external frequency of the central processing unit CPU continues to increase according to Moore's Law. With the recent launch of many new processors by Intel and AMD, the era of 200MHz external frequency has officially arrived (the working speed marked on the CPU is the internal frequency of the processor, which is generated by multiplying the external frequency by the multiplier, and is not directly provided by the motherboard clock circuit). The processor and the north bridge chip are connected by the front-end bus (FSB). Based on the CPU's external frequency, data is transmitted twice or four times per cycle, so 200MHz external frequency multiplied by four times the frequency can get an FSB speed of 800MHz. The memory also follows the footsteps of the CPU, and the working frequency is rapidly promoted to the DDR400 PC3200 specification of 200MHz. The rest of the south bridge chips and AGP, PCI, USB and other buses have their own industry-defined working clock standards, such as 33MHz for PCI, 66MHz for AGP, etc.

Therefore, the clock circuit of the motherboard must provide various working frequencies for many components. In the past, old motherboards used quartz oscillators to handle this, but quartz oscillators can only output one frequency at a time, which is obviously not enough for new motherboards that require multiple clock outputs. Therefore, some manufacturers integrate these oscillation circuits that were originally scattered throughout the motherboard into a chip that can output various frequencies. The motherboard uses such clock generation chips to save costs and space.

The basic structure of the clock generator

Phase Locked Loop (PLL) is the core technology of the clock generator. Modern clock generators only need to provide a reference frequency from a quartz crystal, and use more than one PLL, with different ratios of frequency division circuits, to generate clock outputs of various frequencies, replacing multiple quartz crystals in traditional systems.

The PLL part has two input terminals, namely the reference frequency (Fref) and the feedback frequency (Fvco), and an output terminal (Fout). The relationship between the three can be expressed by the following formula.

Fout=(Fref·P)/(Q·N)

PLL is basically a negative feedback system. The feedback signal is used in the loop to lock the signal frequency and phase of the output terminal to the frequency and phase of the reference signal at the input terminal. The Phase Frequency Detector (PFD) compares the phase relationship and frequency difference between the reference frequency (Fref) and the feedback frequency (Fvco), and detects the phase difference and frequency difference between the two phases to affect the frequency output of the Voltage Controlled Oscillator (VCO). When Fref/Q is ahead of Fvco/P, the UP high-voltage output speeds up the Fout frequency; on the contrary, when Fref/Q lags behind Fvco/P, the DN high-voltage output slows down the Fout frequency, and finally reaches the stable output state as shown in the formula. Therefore, it is only necessary to adjust the ratio between the P, Q, and R values ​​of the PLL external frequency division circuit to obtain the required output frequency.

The relationship between PC overclocking and clock circuit

Overclocking is the most popular topic for computer enthusiasts. The so-called overclocking is to force the system's working clock to a higher frequency than the marked frequency, so as to achieve the purpose of improving performance. The

basic overclocking method is to increase the operating frequency of the central processing unit to a standard operating frequency by manual adjustment. Generally speaking, in order to ensure the stability and reliability of the CPU, manufacturers of central processing units usually mark it with a lower specification of the actual test results, so that the manufactured computer system works at a speed lower than the CPU limit. Therefore, users have the opportunity to squeeze out the best performance of the system without paying extra costs.

The CPU's operating frequency is equal to the external frequency multiplied by the frequency multiplier. Whether adjusting the external frequency or the frequency multiplier can achieve the purpose of increasing the CPU's operating frequency, but most CPUs currently have the frequency multiplier locked when they leave the factory, so only the external frequency can be adjusted by the user.
In the past, the method of adjusting the external frequency/frequency multiplier required the user to adjust the jumper or DIP switch on the motherboard according to the manual to obtain the desired frequency. The new generation of clock generators are equipped with an SMBus (System Management Bus) interface and can be directly controlled by the BIOS. Therefore, the user does not even need to remove the case. Just sit in front of the computer and adjust the system operating frequency at will through the keyboard and screen. In addition, by controlling the register control bits in the clock generator, the CPU's external frequency (in MHz) can be fine-tuned in very small linear steps. Unlike the previous jumper setting method, which jumps directly from 100MHz to 133MHz, the CPU is likely to exceed its limit and cause a crash.

As mentioned above, each component on the motherboard has its own fixed operating frequency, and the operating frequency of each bus and the system frequency mostly maintain a fixed ratio to work. In other words, traditional clock generators usually use the CPU's external frequency as the base frequency, and generate the clock used by other peripherals through a fixed ratio of frequency division. Therefore, when the user increases the CPU's external frequency, the bus and peripheral clocks will also be increased in proportion. Sometimes the CPU has not exceeded its working limit, but the peripherals cannot withstand the high frequency and go on strike.

In order to improve the system stability during overclocking, the new generation of clock generators use the frequency of AGP/PCI and other buses "asynchronous" with the CPU's external frequency, or add a multi-stage frequency division subsystem, so that users can freely set the AGP/PCI working frequency to meet the working requirements of peripherals.
Currently, software is used to adjust the overclocking frequency. If the frequency setting exceeds the acceptable range of the system, the computer will not work at all. How to adjust the setting back to the original usable state? CYPRESS has added a design called Watchdog Timer to the clock generator. Whenever the BIOS sets a new working frequency for the system, the BIOS is also responsible for setting the countdown time of the Watchdog Timer. After the system restarts at the new operating frequency, the timer counts down according to the set time. If the system starts normally, the BIOS will be responsible for clearing the timer setting through the SMBus, and the system will run at the new operating frequency in the future; if the system cannot start normally, when the timer counts down, the clock generator will send a reset signal to restart the system and restore the frequency setting in the clock generator to the frequency setting that worked normally before. Therefore, when the frequency setting fails, the system will automatically reset to the original state, and the user does not need to intervene to reset the system with hardware.

The clock generator can simplify the motherboard design

. The clock generator designed specifically for the motherboard provides a variety of programmable features to facilitate the motherboard manufacturer to design products. For example, for the user's overclocking needs, the programmable clock frequency can be used to freely set the operating frequency in the BIOS without adding additional control circuits on the motherboard.

In addition to meeting the purpose of overclocking, the programmable clock generator's dynamic frequency adjustment capability can also be used to reduce power consumption. Taking a laptop as an example, the system does not always need all the processor performance when running. At this time, the clock can be reduced to reduce the system's power consumption and extend the battery life.

In addition, the clock generator features that are less relevant to the user include programmable time delay and timing. Motherboard manufacturers can adjust the clock delay between various interface clocks to match various board layouts, so that the components of various related interfaces can keep synchronized (or meet their relative clock delay specifications) and operate. They can also fine-tune the trigger phase of the clock signal according to the different characteristics of various types of memory to facilitate engineers in circuit board design.

Motherboard manufacturers often worry about complying with various electromagnetic interference (EMI) regulations. Products usually have to repeat time-consuming procedures such as testing, rewiring, shielding and isolation, which delays the product's time to market and reduces the product's profitability. The programmable spread spectrum (SST) function in the current clock generator can be used to reduce the EMI of the product.

By using the characteristics of the PLL in the clock generator and making small adjustments with the system clock as the center, the EMI energy can be evenly distributed in a small frequency spectrum range to reduce the peak value of the single frequency EMI. The

programmable spread spectrum ratio can be viewed according to the different layouts of the motherboard's circuits, allowing motherboard engineers to set the spread spectrum ratio parameters that best suit the motherboard design, adjust the best EMI spread spectrum effect, and enable engineers to complete product development in the shortest time.

Clock generators, like CPUs, are also gradually evolving with the times. The multifunctional and programmable characteristics of the current clock generators make users more and more convenient in operation and make manufacturers more flexible in product design.
Reference address:Clock Circuit - The Heart of the Computer

Previous article:Electric guitar based on Atmel microprocessor
Next article:Electromagnetic compatibility technology of embedded systems

Latest Microcontroller Articles
  • Download from the Internet--ARM Getting Started Notes
    A brief introduction: From today on, the ARM notebook of the rookie is open, and it can be regarded as a place to store these notes. Why publish it? Maybe you are interested in it. In fact, the reason for these notes is ...
  • Learn ARM development(22)
    Turning off and on interrupts Interrupts are an efficient dialogue mechanism, but sometimes you don't want to interrupt the program while it is running. For example, when you are printing something, the program suddenly interrupts and another ...
  • Learn ARM development(21)
    First, declare the task pointer, because it will be used later. Task pointer volatile TASK_TCB* volatile g_pCurrentTask = NULL;volatile TASK_TCB* vol ...
  • Learn ARM development(20)
    With the previous Tick interrupt, the basic task switching conditions are ready. However, this "easterly" is also difficult to understand. Only through continuous practice can we understand it. ...
  • Learn ARM development(19)
    After many days of hard work, I finally got the interrupt working. But in order to allow RTOS to use timer interrupts, what kind of interrupts can be implemented in S3C44B0? There are two methods in S3C44B0. ...
  • Learn ARM development(14)
  • Learn ARM development(15)
  • Learn ARM development(16)
  • Learn ARM development(17)
Change More Related Popular Components

EEWorld
subscription
account

EEWorld
service
account

Automotive
development
circle

About Us Customer Service Contact Information Datasheet Sitemap LatestNews


Room 1530, 15th Floor, Building B, No.18 Zhongguancun Street, Haidian District, Beijing, Postal Code: 100190 China Telephone: 008610 8235 0740

Copyright © 2005-2024 EEWORLD.com.cn, Inc. All rights reserved 京ICP证060456号 京ICP备10001474号-1 电信业务审批[2006]字第258号函 京公网安备 11010802033920号