What are the advantages of using a programmable clock oscillator as a timing reference for an FPGA system?
Today's complex FPGAs contain a number of functional blocks for implementing various circuits and systems, such as logic arrays, memory, DSP blocks, processors, phase-locked loops (PLLs) for timing generation, and delay-locked loops (DLLs), standards. I/O, high-speed digital transceivers, and parallel interfaces (PCI, DDR, etc.). These different functional blocks are usually driven by multiple clocks. FPGAs typically use an external oscillator and an internal PLL and DLL to generate the clock. System designers must decide how to combine external and internal resources to achieve the best clock tree design. The programmable clock oscillator is used as a timing reference for FPGA systems and offers a number of advantages. The primary advantage is the design flexibility of high-resolution frequency selection for clock tree optimization. Another great advantage is the spread spectrum modulation that reduces electromagnetic interference (EMI).
The intrinsically programmable silicon MEMS clock oscillator architecture can help system designers using FPGAs solve many challenges. This MEMS architecture easily integrates other features such as spread-spectrum clocks for EMI reduction, digitally controlled oscillators for jitter removal, and fail-safe features in high-speed applications.
Frequency selection
A typical system requires a series of clock frequencies. Some of these are standard frequencies, which may be due to industry-enforced requirements (eg, 100MHz required by PCI Express) or may be due to a wide range of applications (eg 75 MHz for SATA or 33.333 MHz for PCITM). These frequencies are associated with the I/O interface to ensure interoperability because the interfaces may not belong to the same system. In contrast, the user can select the clock frequency used to drive the processor, DSP, and state machine engine to optimize speed, power, or resource usage.
When speed optimization is performed, the processing engine should be driven at the highest clock frequency to maximize the number of operations per second. However, the clock period jitter must be low enough to ensure that the minimum clock period is greater than the critical timing path of the design, otherwise logic errors may occur. A common method of frequency selection is to use an internal FPGA PLL to synthesize the high frequency clock from a standard external reference oscillator. This method is only effective if the internal PLL has high frequency resolution and low jitter.
Some FPGAs integrate an internal low noise fraction PLL to meet all of these requirements. In this case, a simple external oscillator reference can be used. However, in many cases the FPGA uses a PLL with a ring VCO and an integer feedback divider to synthesize different frequencies. This PLL is small, flexible, easy to design and control, and consumes very little power. However, it is difficult to achieve both high resolution and low jitter when using such an internal PLL.
Figure 1 shows the general architecture of an integer PLL. Programming the PLL output frequency is done using a prescaler (P), a feedback divider (M), and a postscaler (N), as shown in the following equation:
The PLL feedback loop forms a band limited control system. The output period jitter is mainly determined by the reference clock phase noise (PNin) and the internal VCO phase noise (PNVCO) as shown in the following equation:
The input reference clock phase noise and VCO phase noise are closely related to the output phase noise and are reflected by the low pass filter and the high pass filter response, respectively, such as Hin and HVCO in the expression. HVCO is directly related to the cutoff frequency of Hin. Figure 2 illustrates the relationship between Hin and HVCO in a typical second-order PLL. The highest PLL bandwidth depends on the update rate of the phase detector. The maximum actual bandwidth limit for most actual PLLs is as follows:
For example, if the PLL input frequency is 40MHz and P=40, the highest actual PLL bandwidth is 100kHz.
The period jitter is correlated with the phase noise by the sinusoidal filter response, as shown in Figure 4. [1] It can be seen that the period jitter is more sensitive to the overall PLL output phase noise at a frequency offset position close to fout /2. Since the PLL bandwidth is much lower than fout /2, the reference clock generally has less impact on the period jitter, while the internal VCO phase noise has a greater impact.
Higher PLL bandwidth reduces the impact of the internal VCO on output cycle jitter and reduces overall cycle jitter. In most cases, you can reduce internal VCO noise and improve jitter by setting a higher bandwidth. On the other hand, achieving a higher frequency resolution requires a larger divider P value, which limits the maximum PLL bandwidth. This contradiction requires a trade-off between high resolution and low jitter. The use of an external high-resolution oscillator can alleviate this problem because high resolution can be achieved by external reference.
High-performance programmable oscillators (such as those provided by SiTIme) can be used as external high-resolution oscillators. With this type of oscillator, the internal PLL only needs to support very limited frequency synthesis to increase bandwidth and reduce jitter.
CCTV pole is a type of pole used to mount a closed circuit television camera for surveillance and monitoring purposes. These poles are typically made of durable materials, such as steel or aluminum, and are designed to withstand a variety of weather conditions.
JIANGSU HONGGUANG STEEL POLE CO., LTD. , https://www1.hgsteelpole.com