Optical and DWDM Testing

Transcription

1 application note and DWDM Testing with the T-BERD 310 1

2 and DWDM Testing with the T-BERD 310 T he rapid development of fiber optic technology has made possible the everincreasing flow of data and voice traffic over a single fiber. Increased usage places exceptional demands on service quality and makes errored or interrupted service more critical than ever before. Service providers are continuing to increase bandwidth using higher speeds or Wavelength Division Multiplexing (WDM) technology. Because many existing optical fiber cables were installed to transport lower rates and single wavelengths, intermittent errors commonly surface when systems are upgraded to increase bandwidth. As a result, technological advances create the possibility for problems that could delay service turn-up and increase callbacks. Testing the physical layer for various wavelengths and physical impairments during provisioning ensures future acceptability as networks expand, saving the time and expense of cable replacement. Today, service providers not only need comprehensive optical tests, but they also need an overall strategy for network installation and maintenance before live traffic is commissioned. This application note is one in a series of comprehensive guides for service providers using the TTC T-BERD 310 to install and maintain telecommunications networks. Each of the application notes in this series contains specific tests and troubleshooting appendices to isolate common problems associated with the transport network. If you have difficulty with any of the tests, call TTC s Technical Assistance Center (TAC) at The TAC staff also would appreciate hearing from you about additional testing tips to enhance the troubleshooting appendices. A Layered Testing Strategy The testing strategy outlined in this series is an efficient, bottom-up testing approach designed to systematically eliminate problems found at various transport testing layers that affect higher layer services. Unlike the layers in the Open Systems Interconnection (OSI) Basic Reference Model, which describe protocol relationships, the testing layers referred to in this application note represent categories of common problems and solutions gathered from numerous field installation and maintenance calls. Figure 1 shows the layers referred to in these testing applications. 2

3 Testing Strategy Figure 1: Transport testing layers Physical layer testing identifies a variety of problems caused by improper line build-out, connector or cabling faults, repeater failure, optical reflections, optical loss, and improper optical amplifier tuning in Dense Wavelength Division Multiplexing (DWDM) systems. Tests include pulse shape, jitter, signal measurements, Bit Error Rate Test (BERT), Return Loss (ORL), optical insertion loss, as well as channel peak power, center wavelength and Signal-to-Noise Ratio (SNR) in DWDM systems. Because these tests are performed primarily at several network locations (DS3 demarcation points and in the optical backbone), and DWDM Testing with the T-BERD 310 and DS3 Testing with the T-BERD 310 application notes were developed to thoroughly test the physical layer. Since physical layer problems commonly cause intermittent and hard-tofind trouble at higher layers, it is critical to verify DS3 and optical backbone operation during installation to prevent callbacks. SONET configuration layer testing eliminates common problems associated with SONET circuit setup. Tests include SONET timing, SONET path configuration, error and alarm reporting, and concatenated signal configuration. These tests, detailed in the SONET Testing with the T-BERD 310 application note, ensure the circuit is properly installed for the desired service. This testing results in reduced turn-up time and fewer maintenance calls. ATM configuration layer testing verifies ATM Virtual Path/Virtual Channel (VP/VC) configuration. Specific tests include ATM switch configuration, end-to-end channel setup, alarm and error reporting, errored cell handling, and bandwidth and priority configuration. These tests are ATM Configuration Layer SONET Configuration Layer Physical Layer performed during installation of the backbone ATM transport network or during configuration of ATM service. It is important to include in-service testing since data transport problems, such as delay variation and congestion, can only be roughly simulated during out-of-service testing. The unpredictable nature of data networks can cause problems to surface even after out-of-service simulations have functioned properly. In-service tests include monitoring ATM congestion, alarms, errors, and delay variation. The application note pertaining to this layer is ATM Testing with the T-BERD 310. Thorough testing verifies ATM transport configuration to ensure trouble-free service turn-up. Why Test Fibers? Testing the optical network during service turn-up is one way to avoid expensive upgrades and callbacks due to intermittent errors that surface after the optical transmission equipment is placed on the span. ORL and Insertion Loss (IL) are the two tests that, when performed together, ensure the fiber span has been thoroughly verified. ORL indicates the optical network s ability to transport high-speed signals because it measures the total accumulated power from reflections in the fiber. The ORL measurement is taken at the transmitting end of the fiber using an Continuous Wave Reflectometer (OCWR), also referred to as the stable source. Too much total reflected power can interfere with the operation of the transmitting laser, causing degraded service. These problems are especially evident at today s high speeds, since many existing optical fiber cables were installed to transport lower rates. In some cases, optical measurements were never performed on these cables. A high ORL measurement is desirable because it indicates less light is being reflected back to the laser source. A measurement of 30 db is sufficient for most applications, while a measurement of about 12 db indicates a large reflection caused by an open fiber. The equation for measuring ORL is shown below. Assuming the power transmitted stays constant, the lower the power received by the transmitter, the higher the total ORL measurement. ORL (dbm) = 10 log x TX Power (by Trans) RX Power (by Trans) Since ORL only measures what is happening at the transmitting end of the fiber, it cannot verify the total integrity of the span or that sufficient power actually arrives at the optical receiver at the far end. To illustrate, Figure 2 on the next page shows a sharp bend in the fiber, which allows light to escape the cladding. Because the bend reduces total reflection back to the transmitter, the ORL measurement actually improves. What the ORL test does not show is that the bend also reduces the total amount of power crossing the span to the receiver. A second test is needed to verify the receiving end of the fiber. An IL test measures total power loss across an optical span and indicates overall receiver health. A high IL means excessive power loss with inadequate levels reaching the receiver. A low IL indicates minimal power loss, which can cause receiver saturation, especially when high power lasers are used. Long reach lasers require a minimum of 10 db insertion loss. To prevent intermittent errors or system failure, total IL should be within a defined range. This application note defines 5 db to 15 db power loss as an acceptable range for single mode fibers. This is calculated using a known output of the T-BERD 310 stable source as dbm, while optical power is verified between dbm and dbm. 3

4 SONET Network Element Bend in fiber A single optical fiber that transports all the channels throughout the network. EDFAs which boost total optical power in the fiber without extracting information carried by the optical signals. A splitter, or passive demultiplexer. (Splits one fiber back into its original multiple of n fibers.) A selector (filter) to separate specific wavelengths. Figure 2: Sharp bend in optical fiber SONET Network Element A splitter and a selector are often combined into a single optical structure. DWDM Primer In the recent past, the telecommunications industry has seen unprecedented growth and investment in the search for increased bandwidth. WDM offers an effective solution that increases the capacity of existing fiber optic networks and provides increased flexibility in new ones. Broadband-WDM is a process by which several optical signals are carried on a single fiber at varying wavelengths. This method involves the use of repeaters, which convert the signal into electrical pulses for extraction, amplifies them, and then reinserts the regenerated signal back into the subsequent fiber section. However, because regeneration requires each signal be separated with its own regenerator, there has been little advancement in Broadband-WDM technology. In contrast, Narrowband-WDM systems, referred to as DWDM, link multiple channels in the 1550 nm spectral range. Today, DWDM allows up to 240 channels to simultaneously transmit on a single fiber, each having a different wavelength. Currently, most DWDM systems function within C-band, between 1530 nm and 1565 nm. Systems are also available using the L-band, between 1570 nm and 1605 nm. Ultra-Dense Wave Division Multiplexing (UDWDM or UWDM) with hundreds of channels will be available in the near future. DWDM systems are made possible by the evolution of a purely optical amplifier, the Erbium-Doped Fiber Amplifier (EDFA). This amplifier directly boosts optical signals and provides low-noise amplification of wavelengths ranging from 1530 nm to 1565 nm. Despite being highly effective as an optical amplifier, EDFAs suffer from gain variations across the passband. They add optical noise (Amplified Spontaneous Emission [ASE]) to the amplified signal, and worse, apply a constant gain to the total input energy. This cascading noise gain, amplified by every EDFA with which the optical signal comes in contact, raises the noise floor progressively through the DWDM span. This noise gain reduces the SNR at each EDFA and, consequently, only a limited number of EDFAs can be put in a span. A DWDM system can be broken down into the following elements: A transponder (for transponder-based systems) for each channel, which converts the incoming signal at 1310 nm or 1550 nm into a wavelength within the EDFA amplification range. Currently, transponder laser outputs can be spaced as tight as 0.2 nm apart, and they will be compressed to an even greater degree in the future. A coupler, or optical multiplexer, which receives modulated carrier signals from all sources and couples them into one optical fiber. (Converts multiples of n fibers into one fiber.) Receivers that process some SONET/SDH overhead such as B1 byte (section BIPs) and section trace (found in secondgeneration DWDM systems). These receivers also function to recover lowpowered signals and help to prevent errors from being introduced into a DWDM system when additional channels are added to systems carrying live traffic. See Figures 3.1 and 3.2. How a DWDM System Performs At the transmit end, multiple n signals are fed into transponders where the signals are converted to defined wavelengths within the 1530 nm to 1565 nm spectral range. An optical WDM coupler joins the signals, inserts them into one optical fiber, and forwards that one multiplexed, optical signal to an EDFA. Depending on the path length, the optical signal can encounter none (metro DWDM systems) or several fiber amplifiers, which boost the attenuated signal. However, since noise gain can diminish the effective goal of signal gain, the recommended maximum number of cascaded EDFAs is five or six, with a minimum spacing of 60 km to 120 km between each amplifier. EDFA improvements may increase this number in the future. At the receive end, the multiplexed signals are separated by a splitter and wavelengths are filtered out by a selector. Then they are fed into the receiver of the transmission element (SONET ADM). Second-generation DWDM systems have an intermediate receiver for increased performance, as shown in Figure

5 Transmitter from Transmission Element Transponder Selector (Filter) Second-generation DWDM systems have a receiver for increased performance WDM 1 MUX EDFA EDFA Demux Transponder-Based n Figure 3.1: Transponder-based system Transmitter Coupler Splitter Selector (Filter) WDM WDM EDFA EDFA Non-Transponder-Based Figure 3.2: Non-transponder-based system Types of Fiber An important consideration in DWDM development is the fiber type. There are three basic types of single-mode optical fiber in use today for commercial telecom applications, which have potential ramifications in DWDM deployment testing. They are: Non-Dispersion-Shifted Fiber (NDSF) Dispersion-Shifted Fiber (DSF) Non-Zero Dispersion-Shifted Fiber (NZDSF) While all three fiber types incur similar optical attenuation in the 1310/1550 nm single-mode transmission windows, they differ in chromatic-dispersion characteristics. NDSF, also known as standard fiber, is the most prevalent type found in telecom networks. NDSF exhibits low chromatic dispersion at 1310 nm and high dispersion in the 1550 nm region. However, NDSF limits the effectiveness of an EDFA, which offers its greatest gain in the 1550 nm range. Since dispersion is high, this limits the distance that an optical signal can travel before regeneration. For DSF, the wavelength of minimum chromatic dispersion is over the 1550 nm range, but the zero-dispersion point is within the passband of the EDFA. The significance of this factor is that near this zero-dispersion point, transmitted DWDM channels have an increased probability of non-linear effects such as four-wave mixing (described below). This disadvantage led to the development of NZDSF, also known as lambda-shifted fiber. It reduces the susceptibility of non-linear effects by providing a zero-dispersion point outside the passband of the EDFA. It also introduces just enough dispersion to suppress four-wave mixing while maintaining signal integrity. Problems Encountered in Systems As with any transmission system, physical impairments can impede data flow. At high power levels, optical channels may mix non-linearly and form new optical frequency components, as shown in Figure 4 on the next page. Commonly referred to as four-wave mixing, but having potentially more than four waves, this phenomenon causes both a loss of energy from desired signals and a generation of unwanted signals, which may coincide with existing channels. To reduce the effects of four-wave mixing, unequal or increased channel spacing may be considered. Another problem encountered in DWDM systems is dispersion. In an 5

6 Power λ 2 λ 3 λ 4 optical system, different wavelengths and polarizations do not travel at the same rate through fiber. These different speeds of light are commonly referred to as dispersion. Dispersion broadens the signal as each of these wavelengths arrives at different times. A transmitted pulse also spreads as it travels down the fiber, causing the signals to span more wavelengths. Compounding that effect, dispersion increases exponentially for high-speed systems. There are two types of dispersion which affect DWDM systems; namely, Chromatic Dispersion (CD) and Polarization Mode Dispersion (PMD). CD is caused by a variation of the index of refraction with wavelength. The signal broadening that results from dispersion causes crosstalk of adjacent bits, and becomes a more serious problem as the bit rate increases, as shown in Figure 5. Polarization is a direct result of asymmetries in optical fiber and occurs when the electric-field vector of the wave is perpendicular to the main direction of the optical beam. PMD increases the likelihood of bit errors and is a main concern in OC-192 deployment since much of the installed fiber base is non-dispersion shifted. The end result for both types of dispersion is a high Bit Error Rate (BER). Original Level New Level Figure 4: Four-wave mixing Why Test DWDM Systems? Testing DWDM systems is imperative to qualify, maintain, and troubleshoot high-speed transport networks and to determine any effects of four-wave mixing and dispersion. Since DWDM systems consist of complex modules, special measurement techniques are required to characterize components and network elements as a function of wavelength. Accurate measurement of channel wavelengths and channel spacing for a DWDM system provides an indication of possible wavelength shifts for individual laser sources in the system. Shifted wavelengths can cause crosstalk, resulting in bit errors. The center wavelength of each channel must be measured precisely to ensure design specifications are met and to detect unacceptable drifts in laser sources. Drifts in this parameter can be caused by factors such as temperature change, back-reflection, aging, and laser chirp phenomena. The laser source must be analyzed to ensure signals remain within their assigned wavelength limits, under all operating conditions. See Figure 6. Peak power is the energy that lies beneath the filter skirt of each laser line. Each channel at every available interconnection point should be measured to verify design goals have been met and system power flatness is within limits. Peak power is particularly important in systems using cascaded optical amplifiers, where gain tilt has an important effect on the power in each channel. In a properly functioning DWDM system, power levels across subsequent channels should remain constant without significant fluctuation. When gain tilt is present, channel power levels incrementally increase, creating a linear-sloped gain. In some DWDM designs, a preset tilt is applied to the power level across channels to counterbalance the gain tilt of typical optical amplifiers. See Figure 7. Some EDFAs require manual tuning for even power distribution. By checking individual signal levels, it is possible to detect whether each channel is amplified. A channel with low power can result in bit errors on that channel. See Figure 7. Overall power is a summation (total input) of the power of combined lasers tested. Information about the overall power in the system is very important since, in DWDM systems, it is critical to determine whether power exceeds the power budget, while keeping in mind gain control performance of the EDFA. A key measurement for testing the DWDM is SNR. SNR is defined as the ratio of overall power measured at the wavelength of the data signal and the optical noise power at the same wavelength (see Figure 8). A low SNR correlates to a high BER. However, a high Power Power Time Time Crosstalk 6 Figure 5: Chromatic dispersion

7 Power ASE Wavelength Figure 6: Wavelength shift analysis Channel Peak Power Channel Center Wavelength ASE = Amplified Spontaneous Emission Gain Tilt Power ASE Wavelength Figure 7: Gain tilt Power SNR ASE Wavelength Figure 8: SNR 7

8 SNR does not necessarily indicate a low BER because of four-wave mixing (see Figure 4 on page 6) or dispersion (see Figure 5 on page 6), both of which can cause a high SNR and a high BER. It also characterizes the optical noise level detected at the receiving end of the dense WDM optical link. Therefore, the ultimate test is end-to-end BER testing. The decisive parameter in determining the transmission quality of a system is the BER in an end-to-end test. Since DWDM systems typically carry SONET/SDH framed traffic, a transmission quality cover of BER or better is expected as per Bellcore GR2918 and ITU-T G.692. The following approaches are commonly used for BERT analysis. A lengthy soak test is the most common approach for qualifying DWDM systems. Equipment Required for and DWDM Testing In order to perform this test, the following equipment is required: Fiber Optic Cleaning Kit (Model No. FO-Cleaner) Two T-BERD 310 test sets with these options: SONET OC-12 TX/RX R SONET STS-1, OC-1/3, OC-3c RX T SONET STS-1, OC-1/3, OC-3c TX Media Test nm SONET OC-48, OC-48c TX/RX DWDM Channel Monitor Four tags to label fiber jumpers Equipment Verification To determine if the necessary options are installed in your T-BERD 310, complete the following steps: 1. Turn off the power to the test set. 2. Press and hold the MODE switch down while turning power on. 3. Release the MODE switch when the first hardware revision appears. 4. Press the MODE down arrow to scroll through a list of installed boards and options. Look for the appropriate option code: Testing OPTIC MEDIA TST OPT (310-16) DWDM Testing w/oc-48 DWDM OPTION (310-22) SONET OC-48 OPTION ( /1550) or OC-48 DUAL nm OPT ( DUAL) DWDM Testing w/oc-12 DWDM OPTION (310-22) SONET OC-12 OPTION (310-12) or OC-12 DUAL nm OPT ( DUAL) BER Testing w/oc-48 SONET OC-48 OPTION ( /1550) or OC-48 DUAL nm OPT ( DUAL) BER Testing w/oc-12 SONET OC-12 OPTION (310-12) or OC-12 DUAL nm OPT If the option is installed, the display shows Installed or Module Code = a revision number. If the option is not installed, the option code may not appear or shows Not installed. If the required option is not installed, please call your local TTC sales office or distributor listed on the back cover, or call TAC at Pre-Test Setup Each test setup in this Application Note assumes that the T-BERD 310 is restored to factory default settings before testing begins. Follow this procedure to restore all factory default settings: 1. Turn off power to the test set. 2. Press and hold RESTART. 3. Turn on power. Look for the RELOADING NOVRAM message. When the message appears, release the RESTART button. Factory default settings restore in about 60 seconds. 8

9 Testing A s a rule, optical testing should be done before other types of testing because optical problems affect other International Standards Organization (ISO) layers. Additionally, since handling fibers and connectors may invalidate ORL and IL tests, optical tests should be repeated after all other testing is completed. The instructions in this application note are written for the T-BERD 310-S SONET/ATM User Interface as shown in Figure 9. Testing Return Loss The objective of this test is to verify ORL is within specification at the SONET transmitters. Test Setup 1. Place a T-BERD 310 at each end of the circuit and arbitrarily identify one as East and the other as West. Clean all fiber optic connectors and test set connectors. Tag optical fibers with arbitrary TRANSMIT and RECEIVE designations. 2. At East and West, connect the fiber designated TRANSMIT to the RETURN LOSS/SOURCE connector on the side panel of the T-BERD 310. Leave the fiber designated RECEIVE connected to the SONET network element. It is important that this end of the fiber is closed. If the network element is not available, connect the fiber directly into the POWER METER connector on the T-BERD 310. Figure 10 on the next page shows the proper connections for this test T-BERD 310 COMMUNICATIONS ANALYZER ATM CELL SYNC CELL SYNC LOSS AIS RDI SONET SIGNAL PRESENT FRAME SYNC PATH PTR PRES HISTORY VT PTR PRES CONCAT PAYLOAD DS3 SONET RX SONET TX AUX DS3-DS1 SONET-VT SONET-STS MODE RATE INSERT VOLUME PATTERN PAYLOAD DROP LASER ENABLE ERROR INSERT RESULTS I SECONDARY SONET TRANSMIT TIMING INT RECOVD DS1 BITS CLK PAYLOAD SUMMARY DS3 LOGIC BPV PARITY FRAME SIGNAL TIME SONET SECTION LINE PATH VT ATM RESULTS II RESTART DISPLAY HOLD SECONDARY DS3 TRANSMIT TIMING INT RECOVD EXIT ALARM INSERT PRIMARY DS3 SIGNAL PRESENT FRAME SYNC C-BIT FRAME IDLE PRESENT DS2 FRAME SYNC PATTERN SYNC SIGNAL LOSS FRAME LOSS DS2 FRAME LOSS PATTERN LOSS BLUE (AIS) YELLOW FAR-END ALARM POWER LOSS HISTORY SECONDARY SIGNAL LOSS FRAME LOSS SECTION BIP LINE AIS LINE RDI LINE BIP PATH AIS PATH LOP PATH RDI PATH BIP VT AIS VT LOP VT RDI VT BIP PATH PTR ADJUST VT PTR ADJUST HISTORY INSERT SONET FRAME SECTION BIP LINE BIP LINE FEBE PATH BIP PATH FEBE VT BIP VT FEBE DS3 LOGIC DS3 BPV DS3 FRAME ATM HEC INSERT LINE AIS LINE RDI PATH AIS PATH LOP PATH RDI VT AIS VT LOP VT RDI ATM AIS ATM RDI HISTORY RESET DS3 TRANSMIT DS3 RECEIVE DS3 SOURCE EXT SONET Figure 9: The T-BERD 310 with SONET/ATM user interface Legend: Setup MODE PATTERN LASER ENABLE button RESULTS I arrowed switch RESULTS I blank switch RESULTS II blank switch RESULTS II arrowed switch RESTART DISPLAY HOLD AUX 12. Channel Control 13. INSERT 14. DROP 15. ERROR INSERT button 16. SONET TRANSMIT TIMING 17. ERROR INSERT select switch 18. ALARM INSERT button 19. DS3 TRANSMIT TIMING 20. ALARM INSERT select switch 21. DS3 SOURCE 9

10 West East Power Meter Return Loss/Source Power Meter Return Loss/Source T-BERD 310 (Side Panel) T-BERD 310 (Side Panel) SONET Network Element Span SONET Network Element Figure 10: Connecting the T-BERD 310 to the network to test optical return loss 3. At East and West, configure as follows: Press AUX. Press MODE; select OPTICAL TEST. Press PATTERN; select STABL SOURCE. select 1310 nm. Press AUX to exit. Test Results 1. At East and West, access the following result: Press RESULTS I blank switch; select SIGNAL. select RETURN LOSS. Verify the following result: RETURN LOSS > 30 db 2. Repeat Step 3 in Test Setup, selecting 1550 nm. Verify the following result: RETURN LOSS > 30 db If any results fail, refer to Appendix B on page 19 for troubleshooting tips. When Testing Return Loss results are all okay, continue to Testing Insertion Loss. Testing Insertion Loss The objective of this test is to verify that total IL is acceptable for the laser type and the receivers. Test Setup 1. Place a T-BERD 310 at each end of the circuit and arbitrarily identify one as East and the other as West. Clean the fiber optic connectors and test set connectors. Tag the optical fibers with arbitrary TRANSMIT and RECEIVE designations. 2. At East and West, connect the fiber designated RECEIVE to the POWER METER connector on the side panel of the T-BERD 310. Verify that the fiber designated TRANSMIT is connected to the RETURN LOSS/SOURCE connector on the side panel of the T-BERD 310. Figure 11 shows the proper connections for this test. 3. At East and West, configure for optical testing: Press AUX. Press MODE; select OPTICAL TEST. Press PATTERN; select STABL SOURCE. select 1310 nm. Press PATTERN; select OPTICAL PWR. select 1310 nm. Press AUX to exit. Test Results 1. At East and West, access the following test result: Press RESULTS I blank switch; select SIGNAL. select OPTICAL PWR. Verify the following result: OPTICAL PWR is between dbm and dbm 2. Repeat Step 3 in Test Setup, selecting 1550 nm. Verify results. 3. Repeat ORL and IL tests in the opposite direction. Re-tag cables and swap the TRANSMIT and RECEIVE cables. 10

11 West East T-BERD 310 (Side Panel) Power Meter Return Loss/Source T-BERD 310 (Side Panel) Power Meter Return Loss/Source Span Figure 11: Connecting the T-BERD 310 to the network to test for insertion loss DWDM Testing Testing the Spectrum The objective of this test is to verify overall and individual carrier measurements through non-intrusive testing. Test Setup: 310 Module Only 1. Clean all fiber optic connectors and test set connectors. 2. Connect the fiber from the system to the side panel of the T-BERD Configure receive signal as follows: Press AUX. Press MODE; select WDM RX. Define UNITS: Press PATTERN; select UNITS. select nm or THz. 4. Define PEAK THRESHOLD: Press PATTERN; select PEAK. select THRESHOLD value (which must be between 0 and 40 db). The peak threshold limit defines the minimum value of optical power required for peaks to be considered valid DWDM channels. Each laser line must exceed the peak threshold limit. It is determined by subtracting the PEAK THRESHOLD value from the power of the largest laser line. The default value for PEAK THRESHOLD is 10 db. 5. Define PEAK EXCURSION: Press RESULTS II arrowed switch; select EXCURSION value (which must be between 0 and 40 db). The PEAK EXCURSION defines the rise and fall in amplitude that must take place for a laser line to be recognized. The default value for PEAK EXCURSION is 15 db. 6. Configure for SCAN MODE: Press MODE; select WDM RX. Press PATTERN; select MODE. select SCAN MODE. 7. Press AUX to exit. Test Results 1. Overall results: Press RESULTS I switch; select SIGNAL. scroll through the following measurements: WDM Channels WDM Total Power (dbm) WDM AVG WAVE WDM AVG FREQ 2. Individual carrier results: Press AUX. Press MODE; select WDM RX. Press PATTERN; select CHANNEL. select the desired channel. Press AUX to exit. Press RESULTS I switch; select SIGNAL. 11

12 scroll through the following measurements: WDM Channel Wavelength (nm) WDM Channel Frequency (THz) WDM Channel Power (dbm) WDM Channel SNR (db) 3. Repeat steps for subsequent individual carrier wavelengths. Test Setup: 310 Module w/laptop 1. Set Baud Rate: Press AUX. Press MODE; select PRINT. Press PATTERN; select BAUD RATE. Select Clean all the fiber optic connectors and test set connectors. 3. Connect the fiber from the system to the side panel of the T-BERD Connect the PC to the T-BERD 310 via an RS-232 cable. 5. Double click on the DWDM icon to run the program. 6. Verify the connection. The LED on the PC screen will turn green. A red LED indicates no connection. 7. Configure the user interface as follows: Define UNITS: Select nm or THz. Define PEAK THRESHOLD: Enter value between 0 db and 40 db. The peak threshold limit defines the minimum value of optical power required for peaks to be considered valid DWDM channels. Each laser line must exceed the peak threshold limit. It is determined by subtracting the PEAK THRESHOLD value from the power of the largest laser line. The default value for PEAK THRESHOLD is 10 db. Define PEAK EXCURSION: Enter value between 1 db and 30 db. The PEAK EXCURSION defines the rise and fall in amplitude that must take place for a laser line to be recognized. The default value for PEAK EXCURSION is 15 db. 7. Scan Waveform Press RESTART SCAN button on the PC GUI. After the waveform updates (approx. 25 secs. due to Baud rate); select STOP SCAN. Test Results 1. Select the desired wavelength: Using table on PC software GUI, scroll to desired wavelength and click to select. (See Figure 12.) Figure 12: DWDM scan mode 12

13 End-to-End BER Testing The four methods described in the following paragraphs are applicable for non-transponder-based systems whose access is provided at the tributary level of the add-drop MUX. Daisy Chain. At first glance, the Daisy Chain method could be seen as the most cost effective and time-efficient method of testing because it allows complete testing of the system. In theory, the Daisy Chain method generates a SONET signal through the DWDM span using a single test set. The output is looped back to the receiver on the far-end of another channel, sending the SONET signal back to the near-end. Each output is cabled to each receiver in turn until the last channel is connected to the receiver of the test set. This method is useful for system verification. The real-case scenario of this method, however, is that sectionalization can be very time consuming if a problem is found. Each section must be tested separately, and the inability to load up each channel during sectionalization testing may not cause the impairments that led to the original failure. Moreover, not all DWDM systems have enough signal strength from each selector for the receiver to recover the signal, disallowing the daisy chain method. These factors limit the effectiveness of this method. (See Figure 13.) In addition, typically no more than seven channels can be daisy chained at the same time. Multi-channel BERT. This technique employs a single SONET transmitter with an optical splitter to load all channels of the DWDM MUX. Here, all channels are loaded (causing impairments), and impairments are detected as each channel loads. But while this method provides the best balance of cost and time effectiveness, impairments that occur at the same time on multiple channels are not detectable, also limiting the effectiveness of this approach. (See Figure 14.) Multiple Receiver. This method often provides the best solution for accuracy and efficiency. It uses one transmitter with an optical splitter and an optional EDFA (to boost output power if required) to load all channels of the DWDM MUX. All channels are loaded simultaneously, allowing impairments from other channels. On the receiver end, multiple receivers monitor each receiver concurrently during the soak test and allow BERT measurements to be made simultaneously. This approach significantly reduces test time and is the most effective in locating problems, but it also requires significant test equipment investment. (See Figure 15.) Transponder Selector Test Set MUX Demux EDFA EDFA 16 Selector MUX Demux Transponder Figure 13: Daisy chain testing Transponder Selector Test Set Test Set EDFA (optional depending on signal strength) Splitter MUX EDFA Demux Figure 14: Multi-channel BERT with single transmit and receive testing 13

14 Test Set EDFA (optional depending on signal strength) Splitter Transponder MUX EDFA Demux Selector Test Set Test Set Test Set Test Set Figure 15: Multiple receiver testing Single Channel. This process loads a single channel with a BERT pattern one at a time. End-to-end, or loopback, is useful for testing a single wavelength that is being added to an existing system with live traffic. However, testing a single channel at a time is not recommended for turning up a DWDM system since wavelengths are not loaded simultaneously and may not produce nor allow detection of impairments from other channels. (See Figure 16.) Test Set Transponder λ 1 MUX EDFA Demux Selector Test Set Test Setup 1. Determine which of the above mentioned methods is applicable. 2. Clean all the fiber optic connectors and test set connectors. Tag the optical fibers with TRANSMIT and RECEIVE designations as shown in Figure If only one test set is available, the fiber marked RECEIVE should be connected from the T-BERD 310 receiver to the network element transmitter, and the fiber marked TRANSMIT should be connected from the network element receiver to the T-BERD 310 transmitter, as seen in Figures 13 and 14 on page 13. If more than one test set is available, place a T-BERD 310 at each end of the circuit to emulate SONET network equipment. Figures 15 and 16 show the location of the test sets in relation to the network. One side will be arbitrarily designated West and the other East to facilitate test instructions. 4. At East and West, connect the fibers marked RECEIVE to one of the following T-BERD 310 side-panel connectors: OC-1/3 RECEIVE, OC-12 RECEIVE, or OC-48 RECEIVE, depending on the test access rate. Figure 16: Single channel testing T-BERD 310 (Side Panel) Power Meter Figure 17: Connections to the T-BERD 310 for testing optical power From the Network Element 14

15 5. At East and West, connect the fibers marked TRANSMIT to one of the following T-BERD 310 side-panel connectors: OC-1/3 TRANSMIT, OC-12 TRANSMIT, or OC-48 TRANSMIT, depending on the test access rate. 6. Ensure the test connections are as shown in Figure Verify the LED above the side panel OC-1/3 RECEIVE, OC-12 RECEIVE, or OC-48 RECEIVE connector is illuminated if the signal is optical. 8. At East and West, configure transmit signals as follows: Press Setup switch; select SONET TX. Press MODE; select signal rate (STS-1 or OC-1/3/12/48). Verify the LASER ENABLE button is illuminated if the signal is optical. Press PATTERN; select DS1 ASYN INT, DS3 ASYN INT, CONCAT(3c), CONCAT(12c), or CONCAT(48c). Press SONET TRANSMIT TIMING; select DS1 BITS CLK. This requires an external clock source, synchronized to the SONET network, to be connected to the bantam DS1 BITS CLOCK connector on the side panel of the T-BERD 310. If a clock is not available, then RECOVD timing can be used, but only if it can be verified that the device under test is timed to the BITS clock. 9. If testing DS1, perform this step. Otherwise, skip this step and proceed to Step 10. At East and West, configure the transmit signal using the lid as follows: Press MODE; select T1 D4. Press PATTERN; select 3 IN 24. Press TRANSMIT TIMING; select INTERNAL. 10. At East, configure transmitted path trace message as follows: Press AUX. Press MODE; select SONET TX. Press PATTERN; select PATH TRACE. select USER 1. Press AUX to exit. From the Network Element From the Network Element STS-1 Receive OC-1/3 Receive STS-1 Transmit to the Network Element OC-48 Transmit OC-1/3 Transmit to the Network Element OC-12 Transmit T-BERD 310 (Side Panel) OC-12 Receive OC-48 Receive Figure 18: Connections for testing SONET path continuity 11. At West, configure the transmitted path trace message as follows: Press AUX. Press MODE; select SONET TX. Press PATTERN; select PATH TRACE. select USER 3. Press AUX to exit. 12. If the transmit signal is OC-3/12/48 with a DS1 or DS3 payload at East and West, select the STS-1 channel as follows: Press Channel Control switch; select SONET-STS. Press DROP switch; select the STS-1 channel which maps the DS1 or DS3 payload. Press INSERT switch; select the same STS-1 channel as above. 13. For all signal rates, at East and West, configure receive signals as follows: Press DS3 SOURCE; select SONET. Press Setup switch; select SONET RX. Press MODE; select signal rate (STS-1 or OC-1/3/12/48). Press PATTERN; select AUTO. 14. Verify ds1 asyn, ds3 asyn, or concat(3c) is displayed (as selected in Step 9). Test Results 1. At East and West, verify the LED above the side panel OC-1/3/12/48 RECEIVE connector is illuminated if the signal is optical. To the Network Element To the Network Element From the Network Element From the Network Element 2. At East and West, examine the SONET LEDs. Verify no red LEDs are illuminated. Verify the following SONET LEDs are illuminated (green): SIGNAL PRESENT, FRAME SYNC, and PATH PTR PRES and also VT PTR PRES will illuminate if a DS1 payload is being tested, CONCAT PAYLOAD will illuminate if OC-3c/12c/48c is being tested. 3. At East and West, Press RESULTS I blank switch; select SUMMARY. Verify RESULTS I window indicates ALL RESULTS OK. 4. At East and West, verify timing of the signal as follows: Press DISPLAY HOLD. Press RESULTS II blank switch; select SECTION. Press RESULTS II arrowed switch; select SON RX FREQ. Verify the result is within the range according to the following table, depending on the line rate. Line Rate Min. Freq (Hz) STS-1/OC-1 51,838,963 51,841,037 OC-3 155,516, ,523,110 OC ,067, ,092,442 OC-48 2,488,279,234 2,488,369,766 Press DISPLAY HOLD. Max. Freq (Hz) 15

16 5. At East and West, Press RESULTS II blank switch; select LINE. Press RESULTS II arrowed switch. Verify the following results: POINTER JUST = 0 POINTER DATA = stable, any value from 0 to At West, verify path configuration as follows: Press RESULTS II blank switch; select PATH. Press RESULTS II arrowed switch; select PATH TRACE. Verify the result indicates: The quick brown fox jumps over the lazy dog !@#$%* 7. At East, verify path configuration as follows: Press RESULTS II blank switch; select PATH. Press RESULTS II arrowed switch; select PATH TRACE. Verify the result indicates: T-BERD 310: Communications Analyzer for SONET, DS3, DS1, & DS0 8. At East and West, Press RESTART to initiate test. Allow to run undisturbed for at least 30 minutes. 9. At East and West, verify results as follows: Press RESULTS II blank switch; select PATH. Press RESULTS II arrowed switch; verify results: PATH EQU BER is less than 1.0 E -10 PATH BIP ES = 0 PATH BIP ESB = 0 PATH BIP SES = 0 PATH UAS = 0 Press RESULTS II blank switch; select LINE. Press RESULTS II arrowed switch; verify results: POINTER JUST = 0 POINTER NDF = 0 LINE EQU BER is less than 1.0 E -10 LINE BIP ES = 0 LINE BIP ESB = 0 LINE BIP SES = 0 LINE UAS = At East and West, Verify no HISTORY LEDs or red LEDs are illuminated. Verify RESULTS I window indicates ALL RESULTS OK. Mid-Span BER Testing The objective of this test is to provide the capability to filter out a single channel and locate the source of the impairment in a DWDM system. An optional EDFA may be required if the individual signal strength is -19 dbm or less. (See Figure 19). If the termination point and the monitoring point at the mid-span are not co-located, then two T-BERD 310s will be needed for this test (one at the West side and the other at the East side). Test Setup 1. Clean all the fiber optic connectors and test set connectors. Tag optical fibers with TRANSMIT and RECEIVE designations as shown in Figure 17 on page At East, the fiber marked RECEIVE should be connected from the T-BERD 310 DWDM input port to the EDFA monitor jack. 3. At West, the fiber marked TRANSMIT should be connected from the OC-1/3/12/48 system receiver to the T-BERD 310 transmitter. 4. At West, connect a fiber cable from the channel output connector to the T-BERD 310 side-panel connectors: OC-1/3 RECEIVE, OC-12 RECEIVE, or OC-48 RECEIVE, depending on the test access rate. West East Transponder Selector Test Set Test Set MUX Demux Test Set = Amplifiers may be necessary to boost power Figure 19: Mid-span BER testing 16

17 5. At East, configure receive signal as follows: Press AUX. Press MODE; select WDM RX. Define UNITS: Press PATTERN; select UNITS. select NM or THz. Define PEAK THRESHOLD and PEAK EXCURSION: Press PATTERN; select PEAK. select THRESHOLD value (which must be between 0 and 40 db). The peak threshold limit defines the minimum value of optical power required for peaks to be considered valid DWDM channels. Each laser line must exceed the peak threshold limit. It is determined by subtracting the PEAK THRESHOLD value from the power of the largest laser line. The default value for PEAK THRESHOLD is 10 db. Press RESULTS II arrowed switch; select EXCURSION value (which must be between 0 and 40 db). The PEAK EXCURSION defines the rise and fall in amplitude that must take place for a laser line to be recognized. The default value for PEAK EXCURSION is 15 db. 6. Configure for DROP MODE: Press MODE; select WDM RX. Press PATTERN; select CHANNEL, and press RESULTS I arrowed switch until desired wavelength displays. Press PATTERN; select MODE. select DROP MODE. Press AUX to exit. 7. Ensure the test connections are as shown in Figure 18 on page 15. Test Results 1. At West, insert Line BIP errors as follows: Press ERROR INSERT select switch; select LINE BIP. Press ERROR INSERT button five times to insert Line BIP Errors. Verify no red LEDs are illuminated. Examine the SUMMARY category. Verify the following error is shown: LINE FEBE = 5 2. At West, configure to insert AIS alarm as follows: Press RESTART. Press ALARM INSERT select switch; select LINE AIS. Press ALARM INSERT button. Verify the SONET NE goes into alarm and generates required local alarms. Verify remote management center alarms, if applicable. Verify only the LINE RDI LED (red) is illuminated. 3. At East, verify AIS results as follows: PATH PTR PRES is not illuminated. CONCAT PAYLOAD is not illuminated (only applicable for concatenated OC-3c/12c). PATH AIS is illuminated. 4. At West, cancel LINE AIS alarm as follows: Press ALARM INSERT button so it is not illuminated. 5. At East and West, verify network cleanup as follows: Press RESTART. After five minutes, verify the SUMMARY category window indicates ALL RESULTS OK. Verify the SONET NE clears alarms. (You may need to wait several minutes for the network to cancel alarms). 6. At West, remove the fiber from the OC-1/3/12/48 TRANSMIT connector, clean it and connect to the OUT jack of the SONET NE. Remove the fiber from the OC-1/3/12/48 RECEIVE connector, clean it and connect to the IN jack of the SONET NE. 7. At West, remove the fiber designator tags used during tests. Verify any new equipment that is installed (such as customer premises equipment) is configured with the proper timing. 17

18 Appendix A: Troubleshooting Tips for Testing T his appendix provides a list of items to check when your test setup or test results do not match what is described in this application note. Trouble Found Items to Check OPTICAL PWR result is NO SIGNAL RETURN LOSS is too low (less than 30 db) Indicates received power is too low to be detected. Verify fiber marked RECEIVE is connected to the POWER METER connector on side panel of T-BERD 310. Replace possibly defective optical patch cord. Verify optical connector keys are lined up at both ends of the jumper cable. Verify opposite end of fiber is connected to the RETURN LOSS/SOURCE connector on the T-BERD 310. The LED above that connector should illuminate. Try another fiber. Indicates too much light is reflected in the span. The ratio of the power of the outgoing signal measured against the power of the signal reflected back shows impedance discontinuity. Other end of fiber may be open. This can cause a large air-to-glass reflection. Ensure the end of the fiber is closed. The reflection needs to be attenuated. One way to cause attenuation is to use a mandrel wrap on the fiber. Wind the fiber five times around an object (like a pencil) close to the RETURN LOSS/SOURCE connector. If the return loss result improves (gets higher), the bulk of reflections are downstream from the mandrel wrap. If result does not improve, then the largest reflection occurs upstream of the mandrel wrap. Clean the test set connector, ensure the connection is secure, and polish or replace the fiber. Verify fibers are clean. Polish if necessary. Verify test set connectors are properly cleaned. Verify connector keys are properly aligned. If problem persists, scan the fiber with an Time Domain Reflectometer (OTDR) and repair faults. OPTICAL PWR is too low Indicates too much light is lost across the span. This optical power loss can cause intermittent (between db and errors because insufficient power is reaching the receiver db or less) Verify fibers are clean. Polish if necessary. Verify test set connectors are properly cleaned. Verify connector keys are properly aligned. If problem persists, scan the fiber with an OTDR and repair faults. Replace the fiber. OPTICAL PWR is too high Indicates not enough light is lost across the span. This reading could indicate intermittent errors (between db and because excess power is crossing the span, saturating the receiver db) Install an attenuator and adjust until power measurements fall within acceptable limits. Table 1: testing troubleshooting tips 18

19 Appendix B: Troubleshooting Tips for DWDM Testing T his appendix provides a list of items to check when your test setup or test results do not match what is described in this application note. Trouble Found Number of channels found does not correspond to number of system wavelengths PEAK POWER of all channels is too low (based on manufacturer s specification) PEAK POWER of a specific channel is too low (based on manufacturer s specification) Inaccurate CHANNEL CENTER WAVELENGTH SNR <20 db (or below manufacturer s specification) for all wavelengths SNR <20 db (or below manufacturer s specification) for specific wavelength Items to Check Indicates channels are not within PEAK EXCURSION and/or PEAK THRESHOLD range. Verify channels are within 1528 nm to 1565 nm range (based on manufacturer s design). If number of channels is too low, increase PEAK THRESHOLD and/or decrease PEAK EXCURSION levels. If number of channels is too high, decrease PEAK THRESHOLD and/or increase PEAK EXCURSION levels. Indicates a system-wide power loss. Verify fibers are clean. Polish if necessary. Verify test set connectors are properly cleaned. Verify connector keys are properly aligned. Verify span amplification. Verify amplifier tuning. Verify power level of initial SONET signal. Indicates a channel-specific power loss. Verify fibers are clean. Polish if necessary. Verify test set connectors are properly cleaned. Verify connector keys are properly aligned. Verify power level of a specific channel. Verify span amplification. Verify power level of initial SONET signal. Indicates improper laser tuning. Verify CHANNEL CENTER WAVELENGTH (is correct according to manufacturer s specifications). Consult system manufacturer for possible explanation. Indicates excessive or improper amplification, or optical amplifier malfunction. Verify span amplification. Verify amplifier tuning. Verify power level of the initial SONET signal. Indicates a decreased power level for a specific channel. Verify power level of a specific channel. Verify span amplification. Verify power level of the initial SONET signal. Table 2: DWDM testing troubleshooting tips 19

20 Trouble Found Daisy Chain Bit errors are detected at SONET layer End-to-end BERT Bit errors are detected at SONET layer Mid-span BERT db level cannot be recovered at SONET layer Items to Check Indicates too many channels may be daisy chained (TTC recommends <7 channels). Verify fibers are clean. Polish if necessary. Verify test set connectors are properly cleaned. Verify connector keys are properly aligned. Reduce number of channels being daisy chained. Sectionalize errors via end-to-end BERT using the T-BERD 310 SONET capabilities. Verify fibers are clean. Polish if necessary. Verify test set connectors are properly cleaned. Verify connector keys are properly aligned. Sectionalize errors via mid-span BERT with DWDM drop capabilities. Indicates there is not enough signal power on the dropped channel. Verify fibers are clean. Polish if necessary. Verify test set connectors are properly cleaned. Verify connector keys are properly aligned. Verify signal power level using the optical power meter on the Media Test Option (310-16). Input Power Ranges OC-1/3-7 dbm to -35 dbm OC-12-8 dbm to -26 dbm OC-48-9 dbm to -28 dbm Amplify the signal where necessary to ensure sufficient signal power. Table 2: DWDM testing troubleshooting tips (continued) 20