Cabling test tutorial-site test

Before formal construction, the feasibility of the cabling system should be preliminarily evaluated to ensure the health of the cabling system.

Cabling systems are mainly divided into bare cable, jumper, module, and external crosstalk tests. Due to technical complexity, Module Testing is mainly used in laboratory technical models. It mainly targets the manufacturers of cabling systems, while the other three are currently common incoming testing technologies for cabling systems. So how can we conduct qualitative and quantitative checks on these factors before construction? This article will be divided into Cable Testing, jumper testing and external crosstalk testing for a detailed introduction.

1 pre-construction test: FCL cable test

Currently, most data center cabling systems are composed of optical fiber cables and twisted pair wires. Optical fiber is mainly used to connect storage devices and servers. Twisted Pair wires are mainly used to connect physical devices and servers and interwork with switching devices. The use of twisted pair wires accounts for 70% of the data center cabling system ~ 80%. The quality of the twisted pair directly affects the quality of the data center cabling system. Ensuring the twisted pair Quality is a key point to improve the quality of the cabling system.

Considering the data center transmission rate requirements, most of Party A will require the project implementers to use famous brands or well-known cables on the market. However, the market is filled with cables of various brands. The price gap is very large, and the quality is also very different. When ordering twisted pair wires, customers often encounter such a situation. If the customer requires a line of a certain brand, sellers often ask which brand line they need: average quality, good or best? The best of course is the most expensive and most likely a brand line. The general or good quality may be the imitation of some small factories in China. Although the core technology of twisted pair cables is in the hands of some large manufacturers, it is easy to copy from the basic appearance style, if it is not difficult for a professional to see the difference between products of a big brand and fake lines, this is also one of the reasons for the proliferation of fake lines. A major difference in prices is the use of materials. Many informal manufacturers add some aluminum and iron to the copper core to save material costs. The addition of aluminum and other low-cost metals greatly hinders the transmission of high-frequency signals, especially affecting the transmission rate. There are also some technical differences. For example, in order to reduce the crosstalk between each other and interfere with each other between the 4 pairs of twisted pair wires, the gap between the balanced wires must be inconsistent. The higher the frequency of transmission, the greater the mutual interference between line pairs. Different levels of twisted pair wires have different twisting rates. This technology is also one of the core technologies of well-known cabling brands.

The so-called "General brand", "better brand", and "real brand line" cannot be determined with the naked eye during cable procurement. Due to Party A's lack of understanding of the cabling market or information asymmetry, Party A cannot learn the quality of the cables used before installation. Therefore, some bad Party B will win the bid at a low price and use fake brand-name cables to maximize the benefits. Therefore, Party A needs some effective means to supervise Party B, and Party B also needs to prove the quality of the materials used by Party.

2. How to perform Cable Entry Testing

Fortunately, in the cabling industry, the quality standards for cables are already available, such as TIA 568B and ISO 11801. After selecting the cables, the customer selects several m long cables from the cables delivered by the project team for testing. Because the link density and total number of data centers are relatively large, most of the products purchased in batch are checked by warehouse receiving. The proportion of the tests is from 2% ~ 20%, which is closely related to the total number of purchased products, the contract inspection ratio of Party A, and the degree of attention of Party A on product quality. Of course, the customer is conducting Cable Testing, you must use a test instrument that supports the cable test standards in TIA 568B, such as the "DTX1800 cable analyzer" + "cable test adapter" LABA "pair + DTX-REFMOD module, after selecting a m twisted pair, follow these steps to test the cables before installation.

1) device calibration. Use the calibration module provided by the instrument to calibrate the host and the remote end to eliminate errors between the testing instrument.

2) connect to the dedicated cable test adapter. Insert the laba mn module on the host and far end.

3) connect to the tested cable 1.

4) Select the test standard. Select a dedicated test standard for twisted pair wires in the host. Take the 6A line as an example, select TIA Cat.6A cable 100mLA ).

5) start the test. The Instrument Automatically completes the test item according to the standard. Check the test result after 9 seconds and check whether the twisted pair wires are qualified.

6) generate a test report. Because DTX series instruments support digital signal processing technology, the frequency domain signal is converted into a time domain signal through Fourier inverse transformation and transformation. After the strong interference signal received from the two ends of the M line is removed from the time domain, reconvert to a frequency-domain curve. In this way, the standard test is completed accurately to identify the performance of the cable.

3 pre-construction test: jumper Test

In the cabling system of the data center, a large number of jumpers are used to achieve fast interconnection between switch devices or server connections. Frequent plugging and repeated use, patch cords with poor quality often make the jumper a fault point in the data center. Experienced data center maintenance personnel discard some patch cords that are repeatedly used multiple times and replace them with new ones. In the past, maintenance personnel often used replacement methods to ensure the quality of jumpers.

With the increase in data volumes in the data center, the requirements for the transmission bandwidth of the cabling system have also increased. The use of cat6 and cat6 links has become increasingly popular in data centers. Maintenance personnel found that patch cords are increasingly becoming fault points. Sometimes replacing a new 6-category patch cord will also be the bottleneck of the entire Link, and 80% of the Jumpers made by experienced maintenance personnel will encounter various problems. Jumper itself is short, most of the Jumpers do not exceed 5 m, more than 10 m is a long jumper. Therefore, the cable itself has little impact on the jumper quality, and the quality of the crystal head and its end connection is the key to Jumper testing. As the transmission frequency increases, MHz or even higher, there is a higher requirement on the end connection when the cable supports the transmission rate. In short, when the twisted pair is inserted into the crystal head, a large deformation occurs, for example, the twisted pair is separated, which will cause a sudden change in the characteristics of the cable. Such as ripple loss, the crystal head becomes the bottleneck of High-speed transmission. Needless to say, there is a high percentage of unqualified jumpers directly pressed by many famous manufacturers by machines. I once had the opportunity to test 10 categories of 6 jumpers from a major manufacturer, one of which is unqualified, and the other is just stuck with a qualified line, leaving a very small margin. Therefore, both the patch cord from the manufacturer and the manual patch cord must be tested before installation to ensure the quality of the data center cabling system.

4. Misunderstanding of jumper test: replacing jumper test with Channel Test

Currently, in the industry, many integrators or engineering construction parties use Channel Test Modules to detect jumpers. The test results were satisfactory and were hardly rejected. In fact, the fault point of jumper is usually on the crystal head rather than the cable. However, the standard stipulates that the channel test model does not include the crystal headers at both ends of the link. That is to say, the effect of the crystal headers at both ends of the link will be automatically deducted during the channel test. Therefore, a cable with a length of 5 m and 10 m can easily pass the test, and a real Jumper test contains the quality of the crystal headers at both ends of the test jumper.

5. How to perform jumper Testing

5.1 cat6 patch cord Testing

In fact, for the 6 types of jumpers, TIA and the ISO Standards Committee have already had relevant parameters and test equipment requirements. After 6 categories of cables have just entered the market, various manufacturers of crystal headers and modules of 6 categories of jumpers have their own practices. In consideration of fairness, the Standards Board finds a third-party module, an SMP production module, which can be better compatible with the crystal headers of various manufacturers. This Standards Board puts third-party module SMP) into the test specification. Therefore, the SMP module adapter must be used for 6 type jumper tests. Use the cooperation of the DTX1800 + SMP patch cord adapter to complete the test of the six types of jumpers through the following steps.

1) device calibration. Use the calibration module provided by the instrument to calibrate the host and the remote end and Its adapter to eliminate errors between the test instrument.

2) access a dedicated 6-class jumper test adapter. Insert the SMP module into the host and the far end.

3) connect to the tested cable.

4) Select the test standard. Select a dedicated test standard for twisted pair wires in the host. Take the 6 categories of jumpers with 3 m length as an example, and select the TIA Patch Cord cat.6 3.0 m.

5) start the test. The Instrument Automatically completes the test item according to the standard. Check the test result after 9 seconds and check whether the jumper is qualified.

6) generate a test report.

5.2 6A, 7 patch cord tests

For Categories 6A and 7 jumpers, as ISO standards member TIA only performed the 6A patch cord standard), the corresponding test standard parameters have been introduced, but the test module has not yet been standardized, therefore, there is no test specification. At present, the industry is more common is the use of FLUKE DTX-1800 + laba set + DTX-REFMOD to complete the test.

1) device calibration. Use the calibration module provided by the instrument to calibrate the host and the remote end to eliminate errors between the testing instrument.

2) connect to the dedicated cable test adapter. Insert the laba mn module on the host and far end.

3) connect to the tested cable.

4) Select the test standard. Select a dedicated test standard for twisted pair wires in the host. Take the 2m7 Jumper as an example and select ISO C7 2 m Patch CordLA ).

5) start the test. The Instrument Automatically completes the test item according to the standard. Check the test result after 9 seconds and check whether the twisted pair wires are qualified.

6) generate a test report.

6 pre-construction test: External crosstalk Test

Due to the large amount of data in the data center, the bandwidth requirements for the cabling system are high. The maximum transmission frequency of Standard 6 lines is 250 MHz, while the maximum transmission frequency is 500 MHz when the 6A line is used, that is, 10 Gb transmission. At such a high frequency, the physical characteristic parameters and construction technology of cables are extremely demanding. Especially when the transmission rate reaches 10 Gb, the quality of the cable itself, the quality of the Jumper connection, and the mutual interference between the same bundled cable 6 packet 1 bundle ), it has become an important factor affecting the link transmission performance. At this time, the interference signal is not only from the interference between various lines, but also from adjacent twisted pair wires, which affects the quality of network transmission. Therefore, an external crosstalk test is added in the cabling system with a transmission rate of 10 Gb or above based on the completion of the corresponding specification Test. In actual cabling systems, especially data centers and high-density cabling systems, external crosstalk between cables is more obvious. Therefore, an indicator that must be verified before the project is implemented.

In a general Integrated Wiring System, the six wiring harness is used as the minimum bundle unit for installation facilities such as the distribution frame. When the cables are bundled, they are 6 packets, 1 package or 11 packets, and 1 package is in a high-density data room, A larger bundle unit may be used. Therefore, consider using 11 Pack 1 for pre-construction testing. Because the external interference test period is long, only five bundle units are selected for the acceptance test after construction. Before construction, you can test a bundle unit for simulation. At present, the "DTX-1800 cable certification tester" + "10 GKit test suite" can be used to complete the external crosstalk test.

7. How to test external Crosstalk

Step 1: perform routine channel and permanent link tests on all cables in the bundle unit, save the test results, connect to the computer through a USB data communication cable, and import the results to the LINKWARE software installed on the computer.

Step 2: Install "AxTalk" on your computer; find the longest or most recent link in each bundle of test cables as the "interference cable "; select a "interference cable", add signal release interference on the "interference cable", add one at a time, rotate all "interference link" test ps anext and PS AACR-F; operation test software, the ps anext test and automatic calculation of the "disturbed cable"; the operation test software, the PS AACR-F test of the "disturbed cable" is automatically calculated.

In fact, the main test of AxTalk external crosstalk is whether the Cat.6A cabling system can achieve 10 Gbit/s transmission performance, which is the basis for 10 Gbit/s Ethernet transmission, before the construction of the 10-Gigabit cabling system, the main function of the test is to test the basic performance parameters of the link equipment of the 6A cabling system, which is the guarantee for the normal installation of the 10-Gigabit cabling system.

8 conclusion

As a professional network cabling Testing Solution Provider, we often receive calls from some engineering engineers asking them to provide them with a qualified test report or modify their test results. Obviously, it must have been the end of the project, but the acceptance failed. The engineer hopes to resort to frauds and cheat Party. This project seems to be a huge loss for the project service provider. All the materials and manpower have to be invested in vain and no money has been paid yet. However, the biggest loss is the waste of Party A, months or even more time. The later network devices will be shelved, and the use of application systems will be delayed. So why did the two sides pass a few simple tests before construction to ensure the integrated wiring system at the source?

1. Design and Test of a gigabit network Integrated Wiring System
2. 10 Gbit/s Ethernet test in twisted pair cabling Environment