TSB-67 Transmission Performance Specifications for Field Testing of UTP Cabling Systems Prepared by: ANSI/EIA/TIA PN-3287 Task Group on UTP Link Performance
Required Field Test Parameters for TSB-67 Wire Map Length Propagation Delay Delay Skew Attenuation Near-End Crosstalk (NEXT)
The new category 5 wire map same length same attenuation same NEXT same propagation delay already spec d delay skew already spec d Power Sum NEXT new Return Loss new (already in ISO) ELFEXT new Power Sum ELFEXT new
Link Definition Channel (Two transitions at each end) Channel End USER S PATCH CORD WALL OUTLET USER S PATCH CORD Channel End HORIZONTAL CABLE PATCH PANEL WIRING CLOSET Under carpet or furniture connector WORK AREA
Link Definition Basic Link (One transition at each end) Link End Link End PATCH PANEL WALL OUTLET HORIZONTAL CABLE TESTER PATCH CORD TESTER PATCH CORD WORK AREA WIRING CLOSET
Horizontal Cabling Length Limits OVERALL LINK LENGTH LIMIT: 100 Meters Max HUB/SWITCH PATCH CORDS 7m Max* WORK AREA CORDS 3m Max WIRING CLOSET (TC) HORIZONTAL CABLE 90 Meters Max *Total combined length off all patch cords in TC cannot exceed 7m WORK AREA
AutoTest Results Autotest runs all tests required by the choosen standard TSB-67: Wire Map Length NEXT NEXT@Remote Attenuation
Wire Map Continuity end-to-end Shorts between any two or more conductors (or shield) Transposed pairs Reversed pairs Split pairs...and any other miswires
Correct Pair Wiring
Wire Map Results
Reversed Pair
Transposed Pairs Example: mix T568A and T568B wiring standards = Trouble
Pin/Pair Assignment - T568A Pair 2, (Orange) Pair 3, (Green) Pair 1, (Blue) Pair 4, (Brown) 1 2 3 4 5 6 7 8
Pin/Pair Assignment - T568B Pair 3, (Green) Pair 2, (Orange) Pair 1, (Blue) Pair 4, (Brown) 1 2 3 4 5 6 7 8
DSP Series Test Results Transposed Pair
Split Pairs Caused by connecting wires from different pairs to paired pins This effectively results in an untwisted cable which can cause a high level of Near-end Crosstalk (NEXT)
Split Pairs
DSP Series Test Results
Nominal Velocity of Propagation (NVP) The speed at which a signal travels in a cable, expressed as a percentage of the speed of light in vacuum. NVP = speed at which pulse travels in cable speed of light in vacuum X 100% Speed of light in vacuum is 300,000 km/s or 0.3 m/nsec
Length Calculation Length = RT_Prop_delay (nsec) x NVP x Speed_of_Light Example: measurement of Prop_delay: 435 nsec NVP (%) 2 Length (ft) 68.5 293.3 69 295.4 69.5 297.6 70 299.7
Length Results
Length Measurement Reporting Measured Link Length Calculated for shortest electrical delay Includes the length of the two patch cords Test Limit (PASS/FAIL) Maximum allowable link length PLUS 10% Calculated for shortest electrical delay
Propagation Delay 1 2 3 6 4 5 7 8 1 2 3 6 4 5 7 8 481 ns 486 ns 494 ns 484 ns
Propagation Delay
Delay Skew The difference in propagation delay 1 2 3 6 4 5 7 8 1 2 3 6 4 5 7 8 0ns (481 ns) 5ns (486 ns) 13 ns (494 ns) 3ns (484 ns)
Delay Skew
DSP Series Test Results
Attenuation The amount of signal loss in the transmission link (expressed in db) Signal Source Signal Receiver db Loss
Sources/Causes Attenuation The electrical characteristics of the cable materials and its construction Insertion losses due to improper termination (IDC) Reflections due to impedance mismatches Effect Above a certain amount of loss, the cabling link may not reliably transmit network data
Attenuation as a Function of Frequency TEST CABLE (300 ft) 25 Attenuation (db) Pair 1,2 PASS: 30-May-97 01:20:22pm 20 15 10 5 0 0.1 10.1 20.1 30.1 40.1 50.1 60.1 70.1 80.1 90.1 100.1 MHz Mukilteo Cable Co. TIA Cat 5 Basic Link Limit
Attenuation View Plot
DSP Series Test Results
Attenuation Measurement Reporting PASS Highest measured attenuation and frequency at which this occurred Test limit at this frequency FAIL Measured attenuation and frequency at which this occurred Test limit at this frequency
Near End Crosstalk (NEXT) Transmit Receive Crosstalk measures the amount of signal interference from one pair to another pair NEXT measures crosstalk at the signal source (Near End)
Effects of NEXT Similar to noise interference Induced signal may have sufficient amplitude to corrupt the original signal to be falsely detected as valid data Effect: intermittent station lockup complete network attachment failure
NEXT
NEXT Results
NEXT is measured in db NEAR END 1 Tx signal 2 3 Rx dist. 6 FAR END 1 100 2 3 100 6 NEXT (db) = 10 Log Power of measured disturbance Power of transmitted signal
NEXT View Plot Press View Plot Press Next Pair to view other pair combina-tions Press EXIT until you return to the Autotest results screen
NEXT Measurement To be measured from both ends of the link
NEXT@Remote NEXT measurements taken at the far end Highlight NEXT@Remote Press ENTER Press View Plot
Why Do We Need Next Measurement at Both Ends? NEAR END FAR END 1 2 3 6 Tx PAIR Rx PAIR 1 100 2 3 100 6 BAD CONNECTOR Crosstalk of 24 db @ 62.5 MHz
Testing from the Far End Reveals the Problem NEAR END FAR END 1 100 100 2 3 6 Tx PAIR Rx PAIR 1 2 3 6 BAD CONNECTOR Crosstalk of 24 db @ 62.5 MHz
NEXT Measurement Evaluation Worst Case Margin: +4.8dB at 2.7MHz
NEXT - Worst Case Margin Margin: Difference between measured value and pass/fail (minimum performance) value Positive margin means NEXT Loss is better than minimum (Pass) Negative margin means NEXT Loss exceeds the minimum (Fail) Worst case: smallest margin the point closest to the standard s limit across all frequencies
(Pair-to-pair) NEXT B A D C 6 Combinations A B A C A D B C B D C D
NEXT Headroom Worst case NEXT across all 12 pair-topair NEXT tests 6 pair combinations from Main Unit 6 pair combinations from Remote Unit Headroom indicates the quality of the link Positive = PASS Bigger value is better
Calculating Headroom Headroom = worst Worst Case NEXT Margin Headroom = 6.5dB
The traditional two-pair system S/N = ACR Workstation Transmit (Output) Receive (Input) NEXT External noise Signal Signal LAN equipment Receive (Input) Transmit (Output)
ACR Attenuation to Crosstalk Ratio A variant of Signal to Noise Ratio The difference between NEXT (in db) and Attenuation (in db) The best performance indicator to determine useable bandwidth
Link Performance NEXT is used as indicator for quality of components and workmanship. ACR as an indicator of maximum usable bandwidth. 10 db: still usable signal power 0 db: noise power equals signal power you cannot rely on signal energy above the frequency at which the ACR = 0 db
TIA Limits: Attenuation, NEXT, ACR 70.0 60.0 50.0 Limit in db 40.0 30.0 20.0 Att (limit) NEXT (limit) ACR (limit) 10.0 0.0 0 20 40 60 80 100 120 Frequency in MHz
Measured ACR Values 90.0 80.0 70.0 ACR in db 60.0 50.0 40.0 30.0 ACR (limit) ACR (DSP) 20.0 10.0 0.0 0 20 40 60 80 100 Frequency in MHz
DSP Series ACR Test Results
Power Sum NEXT Workstation Outlet Horizontal Cabling Patch Panel Hub Power Sum NEXT is the combined NEXT on one wire pair from all other wire pairs
Power Sum NEXT The amount of crosstalk induced on a pair when signals are applied simultaneously to all other cable pairs A harder test to pass than NEXT uses TIA NEXT pair-to-pair limits Will be included in new standard It s a calculation, not measurement.
Power Sum NEXT PS NEXT Margin PSNEXT Plot
NEXT and FEXT Workstation Telecommunications Outlet Cabling Patch panel Hub NEXT FEXT FEXT is additive all crosstalk along the link adds to the FEXT disturbance observed at the receiver
Multiple pair -parallel transmission Far-End Crosstalk (FEXT) adds disturbance Workstation Transmit (Output) Transmit (Output) Signal 1 Signal 2 FEXT LAN Equipment Receive (Input) Receive Input Example: 1000BASE-T: 4 pairs are used.)
ELFEXT (Equal Level FEXT) Workstation Outlet Horizontal Cabling Patch panel Hub FEXT attenuation ELFEXT (signal difference in db) ELFEXT is the FEXT level relative to the attenuated signal (FEXT-Attenuation)
Power Sum ELFEXT Workstation Transmit (Output) Signal 1 FEXT LAN Equipment Receive (Input) Example: 1000BASE-T: 4 pairs are used.)
ELFEXT: A factor in S/N Ratio Measure Far End Crosstalk Loss ( analogous to measuring NEXT loss) Measure attenuation ELFEXT: Subtract attenuation from FEXT loss ELFEXT: another indication of S/N for LAN systems where two or more signals travel in the same direction (1000BASE-T).
PSNEXT and PSFEXT Multiple Disturber NEXT (MDNEXT): Power Sum addition of NEXT Losses Multiple Disturber FEXT (MDFEXT): Power Sum addition of FEXT Losses Of particular importance in 2 situations: Use of 25-pair cable Network applications that use parallel transmission over 2 or more wire pairs
Return Loss A measure of reflected signal power over the frequency range of interest Result of variations in Characteristic Impedance Structural variations due to the cable manufacturing process Connectors Installation
Effect of Return Loss Full Duplex LAN systems System A Transmit Output Receive Input Signal A to B Reflection Signal B to A Directional Coupler System B Receive Input Desired signal = attenuated signal from other end. Noise = reflected signal on same wire pair.
Noise contributors in S/N Ratio ACR is the traditional S/N indicator for 2-wire pair LAN systems ACR no longer will be the only measure ELFEXT loss is another S/N indicator when multiple signals are transmitted in parallel Return loss causes another S/N degradation when signals on a wire pair are transmitted in two directions at the same time (Full Duplex)
Combined Noise Disturbance Transmit Output Receive Input Receiver must detect attenuated signal Disturbance at receiver input: NEXT (3 other pairs) ELFEXT (3 other pairs) Return Loss