10G network cable Testing

In recent years, some new network application types have emerged in the Integrated Wiring field. bit streams are 10 times larger than existing networks, which are commonly known as 10G Ethernet ). Earlier, the application of IEEE 802.3ae as the backbone transmission was very mature, so now we are going to increase this type of transmission to a new height.

Today, we are talking about a brand new network type. This means to introduce the new transmission bandwidth standard. This summer, the IEEE 802.3an transmission standard will support 10 Gbit/s data streams over copper. Obviously, this means that 1g gigabit network) data will be transferred to personal work points. This will be an epoch of transmission technology and standards.

However, in theory, it is very important to introduce a new standard and whether it can be applied to existing cabling systems. Now we can see that 10 gbaset data streams are not necessarily supported in systems with Class E/6 class MHz. For the 10GbaseT standard, it will surpass many testing projects of the old standard 1000BaseT. Some standards for these cabling systems will be redefined. The current Recommended Practice is to redevelop new iso ea and iso fa standards in Europe. North American countries argue most about the new TIA/EIA 6a (enhanced 6 categories) standard. The existing ISO Class E and TIA/EIA 6 standards have been applied for more than four years. The introduction of new standards will greatly reduce the lifecycles of these two standards. The on-site testing products are naturally affected by these new changes. The following are some important issues.

New challenges

The most obvious change is that the new standard needs to increase the testing frequency. After careful theoretical analysis, we can see that the 10 Gbps Baset transmission rate is no longer supported at a MHz frequency. The originally estimated limit frequency of MHz has been replaced by the current limit of MHz. For example, you can use inner interpolation to replace the limit values of Class E/6 with slight adjustments, to adapt to new indicators. As a connection component, RJ45 can maintain its original functions when upgraded to MHz. The new FA-level standard uses Level F as a model, and the frequency range will be increased from the original 1000 MHz to the current MHz. From the perspective of the existing middleware standard and GG45 connector, these have higher transmission requirements in the above frequency range. However, this is only the application of the high-bandwidth shielding system. Here we will focus on the changes to the class-6 unshielded e-class standards.

New test parameters

The new standards have been described above, and challenges for field testing technology have also been raised. All friends with cable testers have to face a problem, that is, do they need to invest in these new problems? The answer is: "not necessarily ". However, we must make it clear that a third-level precision 6-channel testing device will no longer be suitable for this competition.

As we have already said, the bigger challenge comes from the ever-increasing field testing frequency range. The frequencies for category 6a and EA have increased from 500 MHz to MHz. The FA-Level Test Range has increased from 1000 MHz to an unprecedented MHz. The frequency range is increased. It is obviously inappropriate to use the previous test accuracy. Because the three-level precision testing standard is for the previous definition of the test bandwidth to the existing 250MHz frequency. The new 6a/EA-Level test requires an updated precision category to be differentiated, that is, enhanced Level IIIa ). Of course, for ISO Class FA, there is an enhanced Level 4 Precision Level IVa. Currently, the four-level precision category is less than 600MHz. At present, the four-level precision and the enhanced four-level precision have covered the enhanced three-level precision in the performance comparison. In fact, this means that each device with Level 4 accuracy has the ability to test both category 6a and EA.

Is there any news that is not conducive to the field tester? The answer is very simple: No! However, it may be a little bit of trouble for you to design some testing projects for the 10G network in the future. But it is only reflected in software programming and drawing the limit curve. The tester manufacturer will formulate the corresponding limit value curve based on the new situation to attach the new standard. On-site testing equipment manufacturers will also update the firmware version of their testers in a timely manner based on standard changes.

Don't worry about external interference

For field test equipment, it is difficult to predict which new test parameters will appear in Channel Test Standards of 6a and EA. In the United States, this is a very hot topic, and even some fierce disputes. Tracing is specific to some unshielded systems. These problems are concentrated in sensitive areas near the MHz range. We know that a single transmission channel will interfere with other adjacent transmission channels. This effect is called "external crosstalk" AXT). This crosstalk not only interferes with adjacent line pairs, but also interferes with signals transmitted externally from the cable. The twisting rate of the crosstalk signal and the cable ry Correlation Line ). In addition, various types of data frames, bit streams, and route loop signals all run on cables. These new parameters are derived from the external near-end crosstalk Alien-NEXT and the external remote crosstalk (Alien-FEXT), abbreviated as ANEXT and AFEXT. There are also integrated ANEXT and integrated AFEXT (Power Sum ANEXT & Power Sum ANEXT ). These parameters are defined based on the number of crosstalk decibels in adjacent data cables, which is of great significance for 10 Gbit/s unshielded cables. For simplicity, the following describes the term ANEXT and why it serves as a reference for other "external interference" parameters.

As we know, sources of interference sources are very complex. To effectively test these parameters in field tests, even with reasonable efforts, it is unlikely. The problem must be solved by other means. Some of the parameters and judgment data we described earlier are obtained from the laboratory. There is a plan to establish a standard test mode, known as the "6 packet 1" test method, including creating a test link containing six sets of interfaces. In other words, a total of seven other long chains are connected at the specified distance. Each line must be tested against other cables, and a total of 96 independent tests are performed.

(: 6 packet 1 crosstalk test method-different effects of shielded and unshielded cable links)

This test method is actually the worst case for simulating a repeatable experiment. Here, the keyword is "repeatable test", because ANEXT only appears for a short time in extreme cases. Slight changes in the cable geometric structure during the test will affect the ANEXT value. For example, slight changes in the position of cables, connectors, jumpers, and so on can cause some changes in the test results. The above test method is an excellent way to determine the parameter value in the worst case. It can accurately reflect the maximum interference between links and cables. Any impact that can occur in practical applications is much more minor than this.

In fact, the permanent installation of the wiring system does not conform to the "6 pack 1" principle. For field testing, the above test method is actually useless.

In the above test method, we only consider the mutual interference between links. In practical applications, we can always find other adverse factors, such as external electromagnetic waves. The interference sources in buildings are not limited to data cables, but also hundreds of electromagnetic waves inside and outside the building, interfering with other cable pairs in different directions. This includes the atmosphere of the electron, motor, cable, antenna, microwave, telephone, microcomputer, and so on. Even thunderstorms have an impact. Therefore, the on-site test is also very dependent on the on-site and surrounding environment. It is difficult to determine how many negative effects affect the final test results within a period of time. For example, if you test in the same office on weekends or on weekdays, different results will be obtained. Why? Most office appliances are shut down on holidays and weekends, and even air conditioners are shut down or run at low speed.

If the interaction of all cables is considered during field tests, a series of complex and time-consuming tests are used. The complexity of a series of tests for ANEXT is several times that of a single class E/6 or Class F cable using handheld instruments. The construction party cannot accept additional time and capital investment under excessive cost pressure.

(: Configuration Requirements for reliable "out-of-Line crosstalk" tests on-site installed cabling systems)

How to deal with "out-of-Line crosstalk" on site "?

The standard organization provides technical support for how to handle out-of-Line crosstalk. For example, some "additional" Proposals for 6a cabling products have been appropriately defined to address the ANEXT problem, at least as defined on ANEXT and related issues, in order to ensure that the 10G network can run smoothly. When cables, distribution frames, jumpers, and jumpers are designed and manufactured to prevent ANEXT, there is no need to verify these effects at the site.

For existing unshielded systems, the standards will be defined as special cases. For the existing 6 types of unshielded systems, there are currently two standards for judging the effect of running 10g applications on them. The Technical Bulletin TSB155 (Technical Service Bulletin) developed by TIA/EIA defines how to test the 10g application performance of the six types of cables that have been installed. Similarly, the ISO/IEC Technical Report TR24750Technical Report is used to test 10 Gb performance for installed e-level cables. In addition, special provisions: due to the length of the cable, no longer testing involved in off-line crosstalk A-XT) parameters. However, the Rules also show that the special situations in theory are still meaningful in practice. As of the standard promulgated, no regulations have been set for possible on-site tests so far.

It is also worth noting that, in the current state, cable manufacturers can only ensure 10 Gbps network performance on the entire channel. One simple reason is that the standards for components and permanent links are still in the development stage. Combinations and matching methods based on six types of standards are not competent yet. We will pay close attention to the indicators of a single component provided by the vendor. The shielding system is easier to use in this case because it is capable of resisting ANEXT. Because the unshielded system can effectively prevent off-line crosstalk, some manufacturers have already added their unshielded systems to the competition for 10G network applications, this is especially true in countries that only want to use unshielded solutions.

Conclusion:

6a, EA, and FA standards have been established and are under development. With the improvement of these standards, at least in the field test standards, there will soon be a set of clear provisions on "what should be tested" and "what should not be tested. One thing is obvious, that is, testing instruments that meet the level IV precision indicators and reach a frequency of MHz or higher will be a hot spot.