In order to improve the reliability of serial data transmission, many higher-speed digital interfaces now use a method of encoding data and then performing parallel-to-serial conversion. There are many ways to encode, such as 8b/9b encoding, 8b/10b encoding, 64b/66b encoding, 128b/130b encoding, etc. Below we take the most popular ANSI 8b/10b encoding as an example.
In the ANSI 8b/10b encoding mode, 8-bit data is first converted into 10-bit data by the corresponding encoding rule, and then parallel-serial conversion; after receiving the signal, the receiving end serially converts the serial data to obtain 10 The bit data is then decoded by 10 bits to 8 bits to obtain the original transmitted 8-bit data. Therefore, if the data rate on the parallel side of the transmitting side is 8bit*100Mb/s, the data rate on the serial side after 8b/10b encoding and parallel-serial conversion is 1bit*1Gb/s. The 8b/10b encoding method was first invented by IBM and later became part of the ANSI standard (ANSI X3.230-1994, clause 11) and is widely used on communications and computer buses.
There are several biggest benefits to data after 8b/10b encoding:
There are enough transition edges to recover the clock from the data. Normally transmitted data may have a long continuous 0 or continuous 1, and after 8b/10b encoding, the encoding rules ensure that there are no more than 5 consecutive 0s in the encoded data stream. 1, there will be enough transition edges in the signal, so you can sample the embedded clock mode, that is, the receiver can directly recover the clock from the data stream with the PLL circuit, no special clock transmission channel is needed.
DC balance can be AC ​​coupled. After encoding, there will be no continuous 0 or 1 in the data, but it is still possible that the number of 0 or 1 is more in a certain period of time. From the above coding table, we can see that the same Byte has two sets of positive and negative 10bit codes. The number of 1s in one code is more, and the number of 0s in the other code is more. When the data is transmitted in the 8b/10b encoding of the current Byte, it will select which set of encodings to use according to the number of positive and negative bits in the data transmitted in the previous history, thereby ensuring that the number of positive and negative bits on the bus is basically at any time. It is balanced, that is, the DC point does not change greatly. After the DC point is balanced, we can use AC coupling in the path of signal transmission (the most common method is to connect a DC-blocking capacitor at the transmitting end or the receiving end), so that the signal changes the ground level and common mode noise at the transmitting and receiving end. The resistance is further enhanced and can be transmitted over longer distances. Another advantage of using AC coupling is that the transceiver does not have to consider the interaction of DC bias points when making interconnections. The interconnection becomes very simple and the support for hot swapping is better.
Conducive to signal verification. In order to ensure the reliability of transmission, many high-speed signals are checked for the reliability of the transmission to check whether the received signal is correct. In the 8b/10bit code table, the original 8bit data has a total of 256 combinations. Even considering that each Byte has two positive and negative 10bit codes, only 512 10-bit combinations are needed. The 10 bits of data can have a total of 1024 combinations, so about half of the 10 bit combinations are invalid data, and the receiving end can determine that the data is invalid once such an invalid combination is received. In addition, we have previously introduced that the data should be guaranteed to be DC balanced during the transmission process. Once the data received by the receiving end is found to violate the DC balance, the data can be invalidated. Therefore, after 8b/10b encoding, the data itself can provide a certain signal verification function. However, it should be noted that this verification is not reliable enough, because in theory there may be several bits that have errors in transmission but the results still conform to the 8b/10b coding and rule and DC balance principles. Therefore, many buses that use 8b/10b encoding will also perform corresponding CRC check (cyclic redundancy check) on the upper layer protocol.
Control characters can be inserted. Among the 1024 combinations that 10 bit data can represent, except 512 combinations are used to correspond to the original 8 bit data and some not so good combinations (such as 0b1100000000, there are too many consecutive 0s or 1s in the signal and the number of 0, 1 is obvious. In addition to the imbalance, there are some very special combinations. These special combinations can be used as control character insertion during data transfer. These control characters do not correspond to specific 8-bit data, but can represent some special meaning in some bus applications. For example, the K28.5 pattern, its special pattern combination can help the receiver to easily identify the symbol boundary of the received continuous 10-bit data stream, so it will be sent during the initialization phase of some buses or the header of the data packet. There are also special symbols for link training, marking different packet types, and matching the clock rate at the transceiver.
In summary, to parallel signals transmitted through the serial bus, it is generally necessary to perform parallel and serial conversion of data. In order to further reduce the number of transmission lines and increase the transmission distance, many high-speed data buses use embedded clocks and 8b/10b data encoding. 8b/10b encoding has many advantages in high-speed digital buses such as FiberChannel, PCI-E, SATA, USB3.0 due to DC balance, support for AC coupling, embeddable clock information, strong anti-common-mode interference capability, and relatively simple codec structure. Interfaces such as DisplayPort, XAUI, and RapidIO are widely used.
The On-grid system is connected with utility grid and this system would work only if the grid is available. In case of a power cut, the system will not work and hence the On-grid system is also termed as a Grid-tied system. Grid is required since the inverter needs to be provided a reference voltage and the inverter needs to sync with the grid in order to export energy back into the grid.
On grid systems make sense for locations with no daytime power cut or with less than 2 hrs of daytime power cut. This type of system is apt for reducing your electricity bills since it is cheaper than a Battery based system and there is no recurring cost of replacement of panels.
On Grid Solar Inverter,On Grid Solar System,On Grid Solar Power System,On Grid Solar System Price
Wuxi Sunket New Energy Technology Co.,Ltd , https://www.sunketsolar.com