industry to create predictable worst-case transmission performance, while also meeting the requirements of the IEEE standards for each successive generation of Ethernet. This allowed network designers to create cost-effective networks and for BASE-T Ethernet to dominate the LAN market. Going beyond 100 m, then, can be a risky proposition, as most networking equipment is not designed to support greater distances. While certain grades of copper cabling are crafted to support quality performance beyond the TIA transmission performance standards, most cabling infrastructure is created to operate optimally within the 100 meter standard. It may be possible to extend a twisted-pair channel further than 100 meters, but the system may not support the same level of performance. Altering the construction of a typical structured cabling solution to extend beyond the standards- defined distances often results in unintended consequences for transmission quality. As the channel length increases, these effects become more pronounced: • Insertion Loss Highly length-dependent, insertion loss is defined as the portion of a signal that does not reach the end of the channel when compared to the signal sent from the transmitter. The longer the channel, the more the signal will degrade due to insertion loss. • Propagation Delay Delay skew is the time difference of signals arriving on different pairs. Data packets are split up between the pairs in the cable for 1000 Mbps and faster speeds, so the timing of signal delivery on each pair becomes crucial to successfully transmitting data. • Power Over Ethernet (POE) Loss 100 m is also the supported distance between power sourcing equipment (PSE) and a powered device (PD). Going beyond that limit can significantly increase power losses, resulting in insufficient power levels, interruptions, and outages.
Not all IEEE-compliant Ethernet devices have the same level of tolerance for cabling channels exceeding standards-compliant lengths. Even if a specific combination of devices is able to establish a connection over a 170 m (558 ft) channel, if just one of those devices were replaced, there is a chance the new combination would not be able to function over that same 170 m channel. Another concern is equipment aging. While it may work now, as the equipment ages, that tolerance for going beyond the standard distance may shrink. The same equipment may not work as expected four years from now, for example, depending on the components used. However, a lack of standardized support for extended
hour at lengths up to 210 m (689 ft). 100BASE-TX also consistently performed well, with all channels supporting relatively error-free transmission at distances up to 180 m (590 ft); a shorter range than 10BASE-T, but given the increase in speed, still generally positive results. At lengths beyond 180 m, including 210 m, channels with both 22 AWG and 23 AWG conductors exhibited a significant number of dropped frames (errors) during the test. 1000BASE-T transmission exhibited very different performance as the quality of the transmission was
significantly compromised at channel lengths longer than 150 m (492 ft), regardless of the cable’s conductor size (Figure 1). The study showed that most 1000BASE-T applications past 150 m can suffer significant frame errors, and the error rate can be significantly detrimental to the proper operation of the connected devices. At 210 m, all links tested dropped so many frames that the connected device would be non-functional. Length With copper cabling, insertion loss is heavily dependent on the length of the channel. The longer the channel, the more signal strength is lost before it reaches its intended destination, the greater the signal delay, and the more the original signal is degraded. Additional losses, like return loss reflections from connectorization can compound this effect, weakening the signal even further. It is possible to mitigate the effects of additional length, but it is not currently possible to eliminate it as a factor entirely. In another study, several thousand unique device connections were made using five different cables, all from different manufacturers, with different constructions, performance ratings, and conductor gauges, using three different network speeds. Figure 2 from the study, titled “Twisted Pair Ethernet Applications at Lengths Greater than 100 m,” shows that the greater the cable length, the higher the insertion loss, and the lower the probability a link could be established. Active Equipment Active equipment also plays a significant role in successfully transmitting error-free 1000BASE-T data at extended distances. The study, “Quality of Twisted Pair Ethernet Applications at Lengths Greater Than 100 Meters”, showed that different switch port combinations produced significantly different results at lengths over 100 m. Switch combination A (Figure 3), previously found to be the most capable combination to establish a link at extended distances, exhibited error-free transmission at 150 m for one hour using different cables with different conductor gauges. Switch combinations B and C, previously shown to be
Switch Combination A B C
Length Cable Type Channels
x x x x x x x x x x x
x x x x x x x x x x x x x x x x x x x x x x
x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x
1 2 3 4 5 6 7 8 9
22 AWG Cat 5e 22 AWG Cat 6/6A 23 AWG Cat 6
distance channels has not precluded many organizations from deploying them. Cable
manufacturers have offered products that exceed the TIA and ISO standards minimum requirements since the inception of structured cabling standards. In some cases, manufacturers even offered performance guarantees for channels of lengths up to 116 m (380 ft). Extensive research has been conducted over the years on this subject, testing risk factors to uncover the causes of transmission degradation and understand how to secure reliable connectivity beyond 100 m. FACTORS THAT IMPACT TRANSMISSION QUALITY Transmission Speed In a study conducted by Leviton and presented at the International Wire and Cable Symposium (IWCS) titled “Quality of Twisted Pair Ethernet Applications at Lengths Greater Than 100 Meters,” the transmission quality was tested for 11 different channels longer than 100 m consisting of cables with various gauge sizes and using different combinations of network switches. A total of 396 port combinations were set up for one hour each, while measuring the channel’s frame error rate (FER) to determine the quality of each channel’s transmission performance at different speeds and lengths beyond 100m. Three different transmission speeds were used: 10BASE-T, 100BASE-TX, and 1000BASE-T. Of the three, 10BASE-T transmission was the most reliable, as each channel supported error-free transmission for one
210 meters
23 AWG Cat 6A
10 11
1 2 3 4 5 6 7 8 9
22 AWG Cat 5e 22 AWG Cat 6/6A 23 AWG Cat 6
x x x x x x x x x x
180 meters
23 AWG Cat 6A
10 11
1 2 3 4 5 6 7 8 9
22 AWG Cat 5e 22 AWG Cat 6/6A 23 AWG Cat 6
x x x
150 meters
x
23 AWG Cat 6A
x x x
10 11
1 2 3 4 5 6 7 8 9
22 AWG Cat 5e 22 AWG Cat 6/6A 23 AWG Cat 6
x x
x x
120 meters
x
23 AWG Cat 6A
x x
x x x
10 11
Acceptable FER
x
Unacceptable FER
FIGURE 1 : 1000BASE-T Link Quality Test Results. Source: Leviton
I
I
42
ICT TODAY
July/August/September 2025
43
Made with FlippingBook - Online catalogs