S Agilent J-BERT N4903B High-Performance Serial BERT User Guide s Agilent Technologies
Notices © Agilent Technologies, Inc. 2014 Manual Part Number No part of this manual may be reproduced in any form or by any means (including electronic storage and retrieval or translation into a foreign language) without prior agreement and written consent from Agilent Technologies, Inc. as governed by United States and international copyright laws. N4903-91021 Edition Release Edition, June 2014 Printed in Germany Agilent Technologies, Deutschland GmbH Herrenberger Str.
Contents 1 Planning the Test Planning the Test - Concepts 11 Introduction to the Serial BERT 11 Introduction to the Serial BERT - Concepts 11 Navigating the Serial BERT GUI 18 Navigating the Serial BERT GUI - Concepts 18 Which Test is Appropriate? 25 Which Test is Appropriate? - Concepts Connecting the DUT 29 Connecting the DUT - Concepts 2 3 25 29 Setting up External Instrument(s) Setting up External Instrument(s) - Concepts 37 Setting up External Instrument(s) - Procedure 42 Setting up Exte
User-Defined Sequences 96 User-Defined Sequences - Concepts 96 User-Defined Sequences - Procedures 99 User-Defined Sequences - Reference 101 Sequencer-Pattern Editor 108 Sequencer-Pattern Editor - Concepts 108 Sequencer-Pattern Editor - Procedure 108 Sequencer-Pattern Editor - Reference 110 4 Setting up the Pattern Generator Setting up the Pattern Generator - Concepts 121 Input and Output Ports 122 Input and Output Ports - Concepts 122 Input and Output Ports - Procedures 131 Input and Output Ports - Refe
Setting up the Error Detector Setting up the Error Detector - Concepts 167 Inputs and Outputs 168 Inputs and Outputs - Concepts 168 Inputs and Outputs - Procedures 171 Data Input Setup - Reference 172 Clock Setup 173 Clock Setup - Concepts 173 Clock Setup - Procedures 178 Clock Setup - Reference 179 Error Ratio 182 Error Ratio - Concepts 182 Error Ratio - Procedures 185 Error Ratio - Reference 190 Sampling Point Setup 193 Sampling Point Setup - Concepts 193 Sampling Point Setup - Procedures 196 Samplin
Advanced Analysis Advanced Analysis - Concepts 227 Advanced Analysis - Procedures Advanced Analysis - Reference 229 234 DUT Output Timing/Jitter 241 DUT Output Timing/Jitter - Concepts 241 DUT Output Timing/Jitter - Procedures 250 DUT Output Timing/Jitter - Reference 252 Output Levels 263 Output Levels - Concepts 263 Output Levels - Procedures 265 Output Levels - Reference 270 Eye Opening 287 Eye Opening - Concepts 287 Eye Opening - Procedures 289 Eye Opening - Reference 294 Error Location Capture 30
Accumulated Measurements 376 Accumulated Measurements - Concepts 376 Accumulated Measurements - Procedures 377 Accumulated Measurements - Reference 378 Eye Measurements 392 Eye Measurements - Concepts 392 Eye Measurements - Procedures 392 Eye Measurements - Reference 393 8 Jitter Tolerance Tests Jitter Tolerance Tests - Concepts 395 Jitter Setup 404 Jitter Setup - Concepts 404 Jitter Setup - Procedures 407 Jitter Setup - Reference 413 Interference Channel Setup 431 Interference Channel Setup - Concepts 4
N4876A System Setup Adjustment 490 N4876A System Setup Adjustment 490 M8061A System Setup Adjustment 492 M8061A System Setup Adjustment 492 10 Problems with the N4916A 493 Problems with the N4916A - Concepts 493 Problems with the N4916B 500 Problems with the N4916B - Concepts 500 Problems with the N4876A 511 Problems with the N4876A - Concepts 511 Problems with the M8061A 518 Problems with the M8061A - Concepts 518 Customizing the Instrument Customizing the Instrument - Concepts Restoring the Syst
File Management 558 File Management - Procedures 558 Preset Instrument State 562 Preset Instrument State - Procedures 562 Self Test 563 Self Test - Concepts 563 Self Test - Procedures 563 Self Test - Reference 564 Index 567 Agilent J-BERT N4903B High-Performance Serial BERT 9
Agilent J-BERT N4903B High-Performance Serial BERT
S Agilent J-BERT N4903B High-Performance Serial BERT User Guide 1 Planning the Test Planning the Test - Concepts The following topics provide some information that can help you in planning tests with the Serial BERT: • “Introduction to the Serial BERT - Concepts ” on page 11 explains the basics of bit error rates (BER) and BER testing with the Serial BERT.
1 Planning the Test • Clock Out • Aux Data Out • Aux Data Out • Trigger/Ref Clock Out • Trigger/Ref Clock Out Termination of output ports improves the test performance. Important Information Regarding Security The Serial BERT is a PC-based instrument with a standard Windows operating system. As such, it is subject to the same security protection measures as any other PC. See the Microsoft web site for more information regarding data security: http://www.microsoft.com/security/default.
Planning the Test 1 Table 1 Fraction Exponent Instrument Display 1/10,000 1 x 10-4 1E-4 1/100,000 1 x 10-5 1E-5 1/1,000,000 1 x 10-6 1E-6 1/10,000,000 1 x 10-7 1E-7 Understanding SER Symbol error ratio (SER) is the symbol ratio of the number of errored symbols to the total number of symbols received. For example, a single error will result in a one symbol error in 1000 symbols where each symbol consists of 10 bits that corresponds to a SER of 1/1000 or 1x10-3.
1 Planning the Test Understanding the Serial BERT The Serial BERT, also referred to as a Bit Error Ratio Tester (BERT), is a powerful instrument that enables you to analyze systems and components in the telecommunication and enterprise communication industries. It is an essential tool for designing and troubleshooting communications systems, high-speed integrated circuits (ICs), and photonic components. Its intuitive operation and leading edge performance will help you quickly verify error performance.
Planning the Test 1 Option/Upgrade for jitter injection. Total jitter can be composed from random, periodic, sinusoidal, and bounded uncorrelated sources. • J11, U11 Spread Spectrum Clock (SSC) and residual Spread Spectrum Clock (rSSC) generation. • J12, U12 Option/Upgrade for automated jitter tolerance compliance tests. Use this test to ensure that the jitter tolerance of the device under test complies with a certain standard.
1 Planning the Test Express®, SATA 3 Gb/s, USB3, fully buffered DIMM, Hypertransport, CEI, or 10 GbE. The following De-Emphasis Signal Converter can be used: N4916A De-Emphasis Signal Converter The N4916A is a 2-tap de-emphasis signal converter. Following are the features and benifits of N4916A: • Generates 2-tap de-emphasis • Support data rates from 1 to 13.
1 Planning the Test • Programmable via J-BERT N4903B or stand-alone • USB and LAN connectivity Once the N4916A/B is connected, it can be conveniently controlled via the Serial BERT. N4876A 28 Gb/s Multiplexer 2:1 The Agilent's N4876A is a 2:1 multiplexer with an output data rate of up to 28.4 Gb/s. Following are the features and benefits of N4876A: • 2:1 multiplexer driven by Serial BERT data and aux data or by ParBERT Data rate up to 28.
1 Planning the Test • 2-slot AXIe module that can be controlled from J-BERT N4903B user interface via USB • Expands data rate of J-BERT N4903B pattern generator up to 28.
1 Planning the Test For understanding, the Serial BERT GUI can be divided into three panes; namely the upper, middle and the lower pane. Upper Pane The following picture illustrates the upper pane of the Serial BERT. The upper pane provides access to the Navigation Menu which allows you to launch the different controls and dialogs of the Serial BERT. For more information on the Navigation Menu, refer to the section “How to use the Navigation Menu” on page 21.
1 Planning the Test • Middle Pane Control’s Title: Displays the title of the control window. The following picture illustrates the middle pane of the Serial BERT. The middle pane allows you to launch the different controls and dialogs of the Serial BERT. These will be described in detail in the subsequent sections. Lower Pane The following picture illustrates the lower pane of the Serial BERT.
Planning the Test 1 How to use the Navigation Menu The navigation Menu comes with a pull down menu which provides a central launching point for the various controls of the Serial BERT. NOTE The availability of the menu items depends on the available hardware and licenses. The navigation Menu includes single menu items like File, Utility and Help and group menu items like External Instument(s), Pattern, PG Setup, ED Setup, Analysis, Jitter and Results.
1 Planning the Test Each pull down menu and its option are described in the following section: File The File menu allows you to perform the following tasks: • Generate a new pattern file • Open the existing Instrument State/Pattern/Sequence/Measurement • Save the Instrument State/Pattern in Editor/Screen Capture in Editor Sequence/Measurement • Preset Instrument State • Print the screen • Explore the File System • Open the Text Editor • Exit the application External Instrument(s) The Exter
1 Planning the Test • Sampling Point Setup • Trigger and Aux Setup • Pattern Sync • Accumulation Setup • Start Accumulation • Stop Accumulation • BER Location • Audio For more information on ED Setup, refer to the section “Setting up the Error Detector - Concepts ” on page 167.
1 Planning the Test Results The Results menu provides the following: • BER Results • Accumulated Results • Eye Results For more information on Results, refer to the section “Evaluating Results Concepts ” on page 361. Utility The Utility menu allows you to adjust the instrument as per your personal preferences. Using this menu, you can: • Use Edit submenu for the cut, copy and paste functionality.
1 Planning the Test In addition, the About button on the Help displays the license information of the Serial BERT. Which Test is Appropriate? Which Test is Appropriate? - Concepts These topics help you to decide on the appropriate test setup for your device. Determining How to Test Your Device The Serial BERT can help you test the performance of components and systems for high-speed digital transmission equipment.
1 Planning the Test The following references may also be helpful in determining how to test your device. • For a list of recommended patterns for your device and application, see “When to Use Which Pattern?” on page 64. • For a list of recommended connection diagrams for your device and application, see “Connecting the DUT - Concepts ” on page 29.
Planning the Test 1 The Serial BERT allows you to run tests from 150 Mb/s up to 12.5Gb/s (depending on the options with which you bought the instrument). See “Error Detector Bit Rate” on page 179 for more information. • Does your device require differential inputs? As complementary outputs, Data/Clock Out and Data/Clock Out may satisfy this requirement. See “Diagram 2. Connections for Differential Inputs” on page 30.
1 Planning the Test Table 3 BER STM-64/ OC-192 (9.95328 Gb/ s) STM-16c/ OC-48c (2.48832 Gb/ s) STM-4c/ STM-1/OC-3 OC-12c (155.52 Mb/s) (622.08 Mb/s) 1E-12 ~ 5 minutes ~ 20 minutes ~ 80 minutes ~ 5.3 hours 1E-11 ~ 30 seconds ~ 2 minutes ~ 8 minutes ~ 32 minutes 1E-10 ~ 3 seconds ~ 12 seconds ~ 48 seconds ~ 3.2 minutes The formula for confidence level is as follows: C = 1 - e^-nb Where: C = degree of confidence (0.
1 Planning the Test Connecting the DUT Connecting the DUT - Concepts This section provides information on how to connect your DUT for several common test scenarios. NOTE The pattern generator's Data Out, Aux Data Out, Clock Out and Trigger/Ref Clock Out ports must be terminated with 50 Ω if they are not connected. NOTE The error detector's Clock In connector must be terminated with 50 Ω if they are not in use.
1 Planning the Test Diagram 1. Connections for Different In/Out Data Rates Often used for multiplexers, or when the data rate received by the device is not equal to the data rate sent from the device. Diagram 2. Connections for Differential Inputs Application is the same as in Diagram 1. Illustrates how to connect to a device requiring differential signals.
Planning the Test 1 Diagram 3. Connections for an Amplifier Often used for amplifiers. Diagram 4. Connections for a Flip-Flop Often used for flip-flops.
1 Planning the Test Diagram 5. Connections for a MUX/DEMUX Pair Used to test multiplexer/demultiplexer (MUX/DEMUX) pairs. For accurate measurement results, the timing of data signals between the MUX/DEMUX pair must be set properly. Diagram 6. Connections for a DCA or Oscilloscope (PGonly) Used for measuring the output waveform of your device. For the least amount of jitter, the clock output signal may be used as a trigger for the oscilloscope.
Planning the Test 1 Diagram 7. Connections for a SONET/SDH Receiver (PGonly) Often used with network equipment, such as SONET/SDH receivers. The device shown may or may not require a clock signal. Diagram 8. Connections for an External Data Source (EDonly) Used when the device is stimulated by another data source. Illustrates the use of an oscilloscope in addition to the error detector.
1 Planning the Test Diagram 9. Connections of the De-Emphasis Signal Converter The Data Output of the Pattern Generator is connected to the Input of the N4916A/ B. The unused output of the Pattern Generator has to be terminated with 50 Ohm. The N4916A/B provides a differential signal to the DUT. The N4916A/B is controlled by the Serial BERT via a USB cable connected to ports at the rear of the instruments. Diagram 10.
Planning the Test 1 Diagram 11. Connections of the M8061A with J-BERT N4903B The Data In 1, Data In 2 and Aux Clk In of M8061A have to be connected to Aux Data Out, Data Out and Aux Clk Out ports of the J-BERT N4903B, respectively. The connections should be made using the matched cable kit with the part number M8061-61601. The Data Output and Data Output have to be connected to the device under test. NOTE Any unused output of the Pattern Generator has to be terminated with 50 Ohm.
1 Planning the Test 36 Agilent J-BERT N4903B High-Performance Serial BERT
S Agilent J-BERT N4903B High-Performance Serial BERT User Guide 2 Setting up External Instrument(s) Setting up External Instrument(s) - Concepts The External Instrument(s) menu allows you to enable external instruments connected to Serial BERT through a user interface. Currently, the following external instruments can be connected to Serial BERT: 1 N4916A De-Emphasis Signal Converter For more details, refer to the section “N4916A De-Emphasis Signal Converter” on page 16.
2 Setting up External Instrument(s) NOTE N4916B De-emphasis box is supported for Aux Data Out channel of the pattern generator as well in second channel mode. The figure below shows the N4916A De-Emphasis Signal Converter connected between the Data Out port of the pattern generator and the DUT. The figure below shows the N4916B De-Emphasis Signal Converter connected between the Data Out port of the pattern generator and the DUT.
Setting up External Instrument(s) 2 De-emphasis is a method that reduces the voltage of a digital signal if the generated level is high or low for more than one clock period. The principle is illustrated in the figure below. 0 1 1 1 0 0 1 0 1 1 0 Input Signal Output Amplitude (Vpp) De-emphasis Amplitude The de-emphasis amplitude is specified as a fraction of the output amplitude (in percent or dB). Post-cursor de-emphasis The figure above refers to a so-called post-cursor de-emphasis.
2 Setting up External Instrument(s) the roles of the two branches. The delayed signal has now a larger amplitude than the direct signal. A waveform example is illustrated in the following figure. 0 1 1 1 0 0 1 0 1 1 0 Input Signal Output Amplitude (Vpp) De-emphasis Amplitude When pre-cursor de-emphasis is generated this way, the complementary Output of the De-Emphasis Signal Converter becomes the normal output and vice versa.
Setting up External Instrument(s) 2 Understanding the N4916B Clock Multiplier The clock multiplier option enables error counting and error analysis of devices using half-rate clocking. The full-rate clock is needed to use the error, eye or jitter analysis capabilities of Serial BERT. NOTE The clock multiplier functionality is only supported by N4916B. Understanding the N4876A 28 Gb/s Multiplexer 2:1 The Agilent Technologies N4876A is a 2:1 multiplexer with an output data rate of up to 28.4 Gb/s.
2 Setting up External Instrument(s) The Data In 1, Data In 2 and Aux Clk In of M8061A have to be connected to Aux Data Out, Data Out and Aux Clk Out ports of the J-BERT N4903B, respectively. The connections should be made using the matched cable kit with the part number M8061-61601. The Data Output has to be connected to the device under test. De-emphasis on pre-cursor and 2 postcursors with M8061A and J-BERT N4903B The following example illustrates how to emulate transmitter de-emphasis up to 28.
Setting up External Instrument(s) 2 Now since the physical connections are OK, you can enable the external instruments through the Serial BERT's Config window. 1 Click Config menu item from the External Instrument(s) submenu. The Config window shows the list of externally connected instruments (Ensure that the external instruments are properly connected and installed). For more information on Config window, refer to the section “Config Window” on page 51.
2 Setting up External Instrument(s) Example - 3 The following image shows the Config window when the M8061A 28Gb/s Multiplexer with De-emphasis is connected to the J-BERT N4903B. 2 Choose the external instrument from the instrument list shown on the Config window.
2 Setting up External Instrument(s) NOTE Use Refresh button to reload the external instruments list, in case, if you are not able to see them in the instrument list. However, if the Config window still do not display any externally connected instruments, restart the J-BERT N4903B software and firmware. 3 Click Update button to update the software revision of external instrument (If required).
2 Setting up External Instrument(s) How to Enable/Disable N4916B De-Emphasis Function To enable/disable the de-emphasis function: 1 From the Config window, select the De-Emphasis function check box. It opens the Deemphasis Signal Converter Connection dialog. The following image shows the an example of N4916B. 2 Click Enable button. It enables the de-emphasis signal converter that is connected between Data Out of Serial BERT and the DUT.
2 Setting up External Instrument(s) How to Enable/Disable N4916B Clock Multiplier Function To enable/disable the Clock Multiplier function: 1 Select the Clock Multiplier function from the given list. It opens the Clock Multiplier Connection dialog. The following image shows the an example of N4916B. 2 Click Enable button. It enables Clock Multiplier function that is connected between Clock Out of PG of Serial BERT and Clock In of ED of Serial BERT.
2 Setting up External Instrument(s) 1 Select the Multiplexer function from the given list. It opens the Multiplexer Connection dialog. The following image shows the an example of N4876A. 2 Click Enable button. It enables multiplexer function that is connected between Data and Aux Data Output of Serial BERT's Pattern Generator and the input of the N4876A. Once the multiplexer function is enabled, a Multiplexer (4876A) menu entry is added to the External Instrument(s) sub-menu.
2 Setting up External Instrument(s) NOTE Ensure that the timing adjustment has to be run at least once when either the N4903B, N4876A or the cables connecting both are being exchanged in the test setup. Also, ensure that the timing adjustment has to be repeated whenever the operating temperature differs by more than 5o Celsius from the temperature at the previous timing adjustment.
2 Setting up External Instrument(s) Once the multiplexer function is enabled, a M8061A menu entry is added to the External Instrument(s) sub-menu. 3 To disable the multiplexer function, clear the Mux with DeEmphasis function check box present on the Config window. It will disable the multiplexer function and the M8061A menu item disappears. NOTE Ensure that the timing adjustment has to be run at least once when either the N4903B, M8061A or the cables connecting both are being exchanged in the test setup.
2 Setting up External Instrument(s) Controlling N4876A Multiplexer After the multiplxer function has been enabled through the Config window, the following parameter are visible in Multiplexer (N4876A) window. 1 Provide f/2 Jitter between -10 ps to 10 ps.
2 Setting up External Instrument(s) The Config window contains the following elements: Refresh Instrument List Identify NOTE 52 The Refresh button allows you to reload the external instruments list, in case, if you are not able to see them in the instrument list. However, if the Config window still do not display any externally connected instruments, restart the J-BERT N4903B software and firmware.
Setting up External Instrument(s) 2 provides the De-Emphasis function while the N4916B provides De-Emphasis and optional Clock Multiplier functions. Software Revision NOTE Visa Resouce Name System Setup Area Load Calibration Settings NOTE The Update button allows you to update the software revision of external instrument(s). This button will be either disabled or not available (in case of M8061A) if the software updates are not required.
2 Setting up External Instrument(s) For more information on input timing adjustment of N4876A, refer to the section “N4876A System Setup Adjustment” on page 490. Clock Multiplier Window The Clock Multiplier enables BER measurements using the forwarded clock as sampling clock for Serial BERT.
2 Setting up External Instrument(s) Termination • If the Half Rate Clock check box is selected, the multiplier of the clock multiplier will be set automatically to 2 and the frequency sent to the clock multiplier is Data Rate/2. • If the Half Rate Clock check box is not selected, the multiplier of the clock multiplier will be set automatically to 1 and Data Rate itself is sent to the box.
2 Setting up External Instrument(s) M8061A Window The M8061A is a 2:1 multiplexer to characterize serial interfaces of up to 28.4 Gb/ s with optional de-emphasis to extend the rate of J-BERT N4903B pattern generator. The M8061A window is shown in the following figure. The M8061A window contains the following tabs: 1 DataOut Tab - provides parameters to set the amplifier, deemphasis and interference. For more information, refer to the section “DataOut Tab” on page 56.
Setting up External Instrument(s) 2 • Amplitude: Used to set the amplitude of the signal. This parameter is enabled if the amplifier coupling is set to AC. • High: Used to set the upper voltage level of the signal. • Low: Used to set the lower voltage level of the signal. • Offset: Used to set the offset of the average voltage level from 0 V. • Coupling: Used to set the amplifier coupling to AC or DC. Selecting AC coupling will remove the DC component from the signal.
2 Setting up External Instrument(s) ClkGen Tab The ClkGen tab allows you to select clock source. It provides the following parameters: • Clock Source: Used to select the clock source to either CLK IN (External clock) or AUX CLK IN (Internal clock). • Input Timing Adjustment: Click Adjustment... to start the input timming adjustment. For more information on input timing adjustment of M8061A, refer to the section “M8061A System Setup Adjustment” on page 492.
Setting up External Instrument(s) Table 4 M8061A Parameters and Range Parameters Range PostCursor3 -12.04 dB to 12.04 dB PostCursor4 -6.02 dB to 6.02 dB PostCursor5 -6.02 dB to 6.
2 Setting up External Instrument(s) 60 Agilent J-BERT N4903B High-Performance Serial BERT
S Agilent J-BERT N4903B High-Performance Serial BERT User Guide 3 Setting up Patterns Setting up Patterns - Concepts The purpose of data patterns is to simulate the type of data that your device might receive in the real world. Different patterns present different data loads to your device, which can cause variations in the bit error ratio. A bit pattern is sent from the pattern generator to your device.
3 Setting up Patterns User Patterns User patterns are file-based, editable patterns. These may be patterns that you set up using the Serial BERT's Pattern Editor, or any of the patterns delivered with the Serial BERT. You can easily define your own patterns for any special requirements. Software-Generated Patterns The Serial BERT can also generate various PRBS 2^n patterns with a range of polynomes from 7 to 23.
Setting up Patterns 3 They are referred to by their pattern lengths, which are powers of 2. Longer pattern lengths give a better approximation of random data and provide a more rigorous test for the device. However, the pattern repetition time is significantly longer. PRBS patterns provide a means to simulate the type of traffic that a system is likely to see: random traffic.
3 Setting up Patterns designed to lock onto a single frequency may have a tendency to lock onto harmonics. A long PRBS may uncover such undesirable behavior. When to Use Which Pattern? Selecting the appropriate pattern and test setup for your application is important. Below is a list of recommendations that may be helpful. • To simulate random traffic, choose PRBS. • For functional and alarm testing, choose alternating memory patterns.
Setting up Patterns 3 User Patterns User Patterns - Concepts User patterns are user-editable patterns that are written to the Serial BERT's memory for pattern generation (see “How does the Serial BERT Generate Patterns?” on page 61 for details). User patterns can contain up to 32Mbit of any arbitrary data pattern. They may be generated by the Serial BERT software or defined in a pattern file and loaded into the memory. The Serial BERT reads the memory bit by bit, and generates the output accordingly.
3 Setting up Patterns If you save a modified text format pattern to the same file name, it is always saved in binary format. If you wish to preserve the original text format file, use the Save As file command and save the data to a different file. NOTE Pattern File Specifications For your instrument to download a pattern file, the pattern must be composed of at least 7 lines. The first 6 lines start with a keyword to identify the field, followed by an '=' sign.
Setting up Patterns 3 • In hex or text format, each character represents four bits-only the characters '0' to '9' and 'A' to 'F' are allowed (not case-sensitive). • In binary format, each character represents eight bits-all ASCII characters are allowed. • In symbol format, each symbol represents 10 bits. Example: 'X' in binary format or '88' in hex or text format produce the same bit sequence as 10001000 in dual format. In hex or text format, the data must be separated by one white space.
3 Setting up Patterns is an odd number of bits long, it is loaded into the RAM 512 times. Because of the restricted memory space, there are also restrictions to the pattern lengths. For example, a user pattern that is 256,001bit bit long would have to be loaded into the RAM 512 times to reach the 512-bit boundary. Such a pattern would occupy more than 128 Mbits in the RAM. The Serial BERT handles such patterns by rounding them off to the nearest length that would fit in the memory.
3 Setting up Patterns How the Serial BERT Generates Memory-Based Patterns The basic concept of how the Serial BERT generates memory-based patterns is relatively simple: The Serial BERT has 32Mbits of internal RAM memory used to store data patterns. The RAM is organized in 64k of 512-bit blocks. During test run, the pattern generator reads this memory bit by bit and generates the corresponding voltage at the output port (1=high, 0=low).
3 Setting up Patterns A0 A1 A2 B0 B1 B2 A0 A1 A2 AAAAAAAA BBBBBBBB AAAAAAAA Implications of Using Memory-Based Patterns There are several implications to the way Serial BERT handles memory-based patterns, a few of which are: • If you set up an alternating pattern where pattern B is a replica of pattern A except for the addition of one error, and the pattern is an odd number of bits long, the error will be repeated 512 times when B is sent (because pattern B is sent 512 times before switching bac
Setting up Patterns 3 User Patterns - Procedures You can do the following tasks in the Edit Pattern window: • “Creating New Patterns” on page 71 • “Opening Existing Patterns” on page 71 • “Editing Patterns” on page 72 • “Saving Patterns” on page 72 • “Loading Patterns to the Pattern Generator and Error Detector” on page 73 • “Loading Patterns Directly from the Pattern Editor” on page 75 Creating New Patterns To start a new pattern: 1 Click Edit Pattern on File menu item from the Pattern submen
3 Setting up Patterns 2 In the toolbar, click the Open icon. The Select User Pattern dialog box opens. 3 Use this dialog box to locate and open the desired pattern. You can select files of the following types: – 71612A BERT Pattern Files (*.dat) – 86130A Pattern Files (binary, hex, symbol, "01") (*.ptrn) – ASCII Text Pattern Files (*.txt) See “ Supported Pattern Types” on page 65 for descriptions of these files. Click New File to create new pattern.
3 Setting up Patterns If the pattern has already been saved earlier, the saved file is updated. Otherwise, the Windows standard Save As dialog box opens, where you can define the file's path and name. The available file formats to save the file in are: • Pattern File (*.ptrn) in binary (8 bits/byte), hex (4 bits/byte), 01 (1 bit/byte), or symbol format • ASCII Pattern Files (*.txt) See “Pattern File Specifications” on page 66 for descriptions of these file types.
3 Setting up Patterns Table 6 NOTE Pattern Type Next Required Action Sequence Enable this checkbox if you intend to download a user-defined sequence. Patterns generated within a sequence are defined in the Sequence Editor. Your instrument also contains user patterns that mimic real data packets and are designed to stress specific characteristics of your device. For more information, see “Example Patterns” on page 70.
3 Setting up Patterns Loading Patterns Directly from the Pattern Editor After you have finished editing a pattern in the pattern editor, you can load this pattern directly to the pattern generator and error detector. NOTE Keep in mind that your modifications in the Pattern Editor do not affect the pattern at the pattern generator or error detector until you download the pattern to the pattern generator/error detector.
3 Setting up Patterns NOTE Once the pattern of a specific length is captured from the error detector , it can be sent back to the error detector and the pattern generator. For later use of this captured pattern it should be saved using "SaveAs" option. The user will be prompted to save the captured pattern if a new pattern is loaded, and the same pattern is recaptured.
3 Setting up Patterns Table 9 Icon Name Description Export Click this icon to to open Export Trace dialog. See “Export Trace” on page 118 for details. To PG Click this icon to send the pattern from the editor to the pattern generator. To ED Click this icon to send the pattern from the editor to the error detector. Capture Click this icon for Pattern Capture. See “Pattern Capture Dialog Box” on page 83 for details.
3 Setting up Patterns Table 9 Icon Name Description INS/ Click this icon to toggle among the insert, dynamic insert and overwrite editing mode: OVR With Insert mode you insert bits at the cursor position, shifting the rest of the pattern to the right. The pattern length is not changed on insertion. For every inserted bit at the cursor position, one bit at the end of the pattern will be removed.
Setting up Patterns 3 Table 9 Icon Name Description Properties Click this icon to change the properties of the current pattern. See “Pattern Properties Dialog Box” on page 82 for details. Find Click this icon to open the Find Pattern Segment dialog box and perform the search and replace operation for a specified segment in the pattern. See “Find Pattern Segment Dialog Box” on page 84 for details. Block Edit Click this icon to insert a block of data in the pattern at the current cursor position.
3 Setting up Patterns Table 9 Icon Name Description Pat Click this icon to open the Pattern Select Form dialog box. For more information, see “Loading Patterns to the Pattern Generator and Error Detector” on page 73. Sel TIP If you are working on a remote PC, you can use the keyboard shortcuts for Cut, Copy, and Paste. Also, you can use the clipboard function to copy strings in either binary or hex format to and from other applications.
Setting up Patterns 3 Pattern Editor Canvas and Status Bar The pattern editor canvas displays the pattern and offers you to edit it. In case of symbols, the pattern editor canvas displays the symbols as following: The symbols marked with "+" & "-" denotes symbol disparity, while the symbols without signs denotes neutral disparity. The symbols with red color signifies the error in runnning disparity. The symbols with "N/A" signifies an invalid symbols (Not a valid 8B/10B symbol).
3 Setting up Patterns Disparity Error (D): This indicator turns red if there is any disparity error in the pattern. In case of alternate patterns, the indicator will indicate the disparity error of individual pattern trace. Loop Disparity Error (LD): This indicator turns red if the pattern causes any disparity errors when being looped. The available functions to change the view and to modify the pattern are described in “Edit Pattern Window” on page 76.
Setting up Patterns 3 To find out more about appropriate pattern lengths, see “Pattern Resolutions and Lengths” on page 67. Pattern Type Select Standard if you want to use this pattern as a standard pattern. Select Alternate if you want to define two alternating patterns A and B. Conversion Options On changing the pattern type, you have the conversion options given below.
3 Setting up Patterns Description Length in Bits Description for the captured Pattern. Pattern capture length in bits. Current ED pattern length Captures the entire pattern length of currently loaded pattern in the error detector. Maximum Memory Depth Captures the maximum (32Mbit) pattern length of currently loaded pattern in the error detector. Find Pattern Segment Dialog Box The Find Pattern Segment dialog box allows you to search for a specified bit sequence or symbols in the pattern.
Setting up Patterns 3 In addition, The Find Pattern Segment dialog box also allows you to replace the searched pattern segment with the desired pattern segment. Next Looks for the next occurrence of the pattern from the current position to the end. If there is no further occurrence, the current selection does not change. Make sure the search patterns entered in symbol format are seperated by one white space.
3 Setting up Patterns With the Block Edit Configuration dialog box, you define the Range that is to be modified. The available options for the Range are: • Select All Choose this option to select the entire pattern for editing. • Range Choose this option to select the range of bits specified by the bit positions entered in the From and To fields. • Fill From Cursor to End Choose this option to select the bits from the cursor location to pattern end.
3 Setting up Patterns • Fill with Ones • Invert. The Invert does the following functions: Symbol Mode off: Invert the bits. Symbol Mode OOB: 0 becomes 1, 1 becomes 0, Z remains unchanged. Symbol Mode PAM4: Invert bits on per channel basis. • Clock Pattern 0101…0101 • Clock Pattern divide by • Symbol NOTE All the above options are not available in all modes. The Symbol fill is available if the Symbol Mode is enabled.
3 Setting up Patterns NOTE The Recode tab is not accessible if Symbol Mode is set to OFF. 3 Traces This tab provides the following options: – Copy Choose this option to copy data from one trace to another. – Swap Traces Choose this option to swap data between the traces. NOTE The Copy and Swap features are enabled only when there are two traces (Pattern A and Pattern B when alternate pattern is used for single channel or Data and Aux Data pattern for second channel).
Setting up Patterns 3 – Rotate Right Treats the pattern data in the specified range as a circular buffer and rotates the bits to the right by the specified amount. No data will be lost and what is at the end of the buffer will be at the start of the buffer after the rotation. – Align to Sequence Aligns the pattern data in the specified range to a specified pattern sequence. 5 8B/10B This tab provide you an option to either Invert Running Disparity or Auto Correct Disparity in the patterns.
3 Setting up Patterns Error Detector Pattern Tracks the Pattern Generator Pattern With this checkbox you can determine whether both pattern generator and error detector use the same data pattern. If you clear this checkbox, the Error Detector Pattern tab appears, where you can select a different pattern for the error detector. The checkbox is automatically cleared when you load a user-defined sequence to the pattern generator. See “Sequence Mode Characteristics” on page 98 for details.
3 Setting up Patterns Bit Recovery Mode / BRM Only available on the Error Detector Pattern page. Enables the error detector's bit recovery mode. For details see “Understanding Bit Recovery Mode” on page 195. Software-Generated PRBS Software-Generated PRBS - Concepts Software-generated patterns are PRBS patterns that can be modified. They are calculated by the Serial BERT software and then loaded to the memory, from which the necessary output is generated.
3 Setting up Patterns Table 10 n Sequence Length Longest Run of 1's Longest Run of 0's 7 128 7 7 10 1,024 10 10 11 2,048 11 11 13 8,192 13 13 15 32,768 15 15 23 8,388,608 23 23 Zero Substitution Patterns A potential risk to bit errors are longer rows of zeros within a data stream. The longest run of zeros in a 2^n PRBS consists of n zeros. The Zero Substitution function can be used to stress the DUT additionally by inserting a longer row of zeros in the data stream.
3 Setting up Patterns The ones (mark) density can be varied to put the pattern out of balance in a randomly distributed way. This may be useful for systems that are AC-coupled. This helps to check for effects such as baseline wander. To test these cases, the Serial BERT provides PRBS patterns with the mark densities 1/8, 1/4, 1/2, 3/4, and 7/8.
3 Setting up Patterns How the Hardware Generates PRBS 2^n -1 PRBS is an inverted, hardware-generated pattern that is created by a series of shift registers with adjustable feedback. The example below shows the register configuration for a 2^7 -1 (127-bit) pattern. The following table describes the operation of XOR'ing two points for the different patterns.
Setting up Patterns 3 Range of Hardware-Generated PRBS Patterns PRBS Lengths The following table lists the hardware-generated 2^n -1 PRBS pattern lengths that correspond to different n values. This is before the pattern is finally inverted.
3 Setting up Patterns User-Defined Sequences User-Defined Sequences - Concepts This section describes the basics of user-defined sequences. A sequence is created and maintained by means of the Sequence Editor that can be enabled from the Pattern window. A sequence consists of up to 120 blocks than can be looped. Each block can generate a pause signal (constant 0 or 1), a divided clock signal, a 2^n -1 PRBS, or a user pattern. Single or multiple blocks can be looped.
3 Setting up Patterns Description= optional Start= optional Block #= repeated for each block, numbered Loop= repeated for every loop, not numbered Example of a SequenceExpression: (Version= 1.0 Start= IMM Block 1= PRBS11, 1024, TrigOn Block 2= C:\\Pattern\Upat10.ptrn Block 3= P0, 512, TrigOff Loop= B1, B1, 2) A sequence is in common for both channels, Data Out (Channel 1) and Aux Data Out (Channel 2).
3 Setting up Patterns NOTE Length is related to master channel (Data Out) while the other channel (Aux Data) will always follow this length. • Trigger: Refers to the pattern generator's Trigger/Ref Clock Out port. If the Trigger/Ref Clock Out is set up to generate a divided clock pulse, the block setting is ignored. But if the Trigger/Ref Clock Out is set to "Sequence Trigger" and the block to TrigOn, a trigger pulse is generated whenever the block starts or restarts.
Setting up Patterns • 3 Every sequence can be stored in a file and recalled from that file. However, the SequenceExpression does not include user patterns. Recalling a sequence from its file may fail if the referenced user patterns are not available. A recalled sequence is not automatically downloaded to the pattern generator. • In sequence mode, the function "Error Dector Pattern tracks the Pattern Generator Pattern" is automatically disabled.
3 Setting up Patterns – Decide on Trigger On or Off. This setting becomes effective if the Trigger/ Ref Clock Out port is set to Sequence Trigger (trigger at block begin). See “Sequence Trigger ” on page 147 5 Create loops (if desired): – Click the out-arrow of a block. A loop condition box labeled INF appears. – Click the in-arrow of the same or a previous block. This closes the loop. – Click the condition box and choose the appropriate break condition.
Setting up Patterns 3 User-Defined Sequences - Reference The elements of the Sequence Editor and the dialog boxes that can be opened from the Sequence Editor are described below: Sequence Block Display A new sequence consists of one block that is infinitely repeated (looped). By default, this block has a length of 512 bits and generates a Pause0 signal (a continuous stream of zeros). All this is shown on the display.
3 Setting up Patterns In this example, the number of blocks has been set to three. A "None" block generates no pattern and is completely ignored (indicated by a straight vertical blue line). Sequence Trigger at Block Begin After downloading the sequence to the pattern generator, you can also change the behavior of the Trigger Out port (for details see “Sequence Trigger ” on page 147).
Setting up Patterns 3 a browser and search for the file that contains the pattern. You can search for files with the extentions .ptrn, .txt, or .dat. If the pattern file contains two alternate patterns, you can also specify whether the A or B pattern shall be generated (see also “Pattern File Specifications” on page 66). If the pattern file contains just one pattern, your choice is ignored. You cannot change the block length. It shows the length of the pattern as stored in the file.
3 Setting up Patterns Every loop has a delete button [×] and a loop condition button. Loop Delete Button Loop End Condition Button Clicking the delete button [×] removes the loop from the sequence. The initial (default) setting of the loop end condition is INF(inite), which means, this loop will continue until the instrument is switched off. Clicking the loop end condition button allows you to specify the loop end condition. Choices are: • Counted: The loop ends after the specified number of repetitions.
Setting up Patterns 3 Table 14 Icon Agilent J-BERT N4903B High-Performance Serial BERT Name Description New Click this icon to create a new sequence. This opens the Properties dialog. See “Creating a New Sequence” on page 99 for details. Open Click this icon to open a sequence from a file. See “Loading a Sequence From a File” on page 100 for details. Save Click this icon to save the current sequence in its original file. Save As Click this icon to save the current sequence in a new file.
3 Setting up Patterns Table 14 Icon Name Description From PG Click this icon to upload the present sequence from the pattern generator to the Sequence Editor. This allows you to inspect and edit a sequence that has been loaded with a program. Properties Click this icon to change the properties of the current sequence. See “Setting Sequence Properties” on page 107 for details. Break Click this icon to terminate an infinite loop that is set to "manual" break condition.
3 Setting up Patterns Table 14 Icon Name Description CrossTalk Click this icon to open the CrossTalk dialog. Use this dialog to fill every sequencer block of the second channel (Aux Out) with a selected PRBS automatically. Note: This icon will only appear in the second channel mode.
3 Setting up Patterns NOTE If you edit the SequenceExpression manually, take care to adhere to the syntax and order of the keywords. Remember that you must download the updated sequence to the pattern generator if your changes shall take effect. Start Button The Start button of the Sequence Editor indicates that the sequence has been set up for manual start. Click this button to start the sequence execution.
3 Setting up Patterns By default, the Sequence Editor shows one block that is automatically repeated. It has a length of 512 bits and generates Pause 0 data (pure zeros). 2 Click Edit to launch Set Data Block Parameter dialog for second channel. 3 Click Create Pattern to launch Create New Pattern dialog. Use this dialog to create new patterns. See “Create New Pattern” on page 115 for details.
3 Setting up Patterns 7 In the toolbar, click Discard to reject all the new changes. Exporting Patterns in Sequencer-Pattern Editor To export the current pattern under a new name: 1 In the toolbar, click Export icon. The Export Trace dialog opens. 2 Use this dialog to locate and save a copy of the current pattern. See “Export Trace” on page 118 for details.
3 Setting up Patterns Table 15 Icon Name Description Apply Click this icon to apply the pattern to the hardware (in case the sequence is already active), or simply update the corresponding pattern file(s) if the sequence is not yet active. Applying closes the editor and brings the user back to the sequence editor. Discard Click this icon to close the editor without updating anything and bringing the user back to the sequence editor. Cut/ These functions follow Microsoft Windows conventions.
3 Setting up Patterns Table 15 Icon Name Description Find Click this icon to open the Find Pattern Segment dialog box and perform the search and replace operation for a specified segment in the pattern. See “Find Pattern Segment Dialog Box” on page 84 for details. Block Edit Click this icon to edit a block of data in the pattern (s). See “ Block Edit Configuration Dialog Box” on page 85 for details. Select Click this icon to select all the bits in the pattern.
Setting up Patterns 3 Table 15 Icon Name Description INS/ Click this icon to toggle among the insert, dynamic insert and overwrite editing mode: OVR With Insert mode you insert bits at the cursor position, shifting the rest of the pattern to the right. The pattern length is not changed on insertion. For every inserted bit at the cursor position, one bit at the end of the pattern will be removed.
3 Setting up Patterns Table 15 Icon Name Description Symbol Codings Click this icon to open Symbol Settings dialog. See “Symbol Settings” on page 116 for details.
3 Setting up Patterns It uses a different set of background color to distinguish between the two channels. The RED background denotes the pattern for Data Out chanel while the BLUE background denotes the pattern for Aux Data Out channel. NOTE If any of the channel has anything beside pattern file, the corresponding channel would not be visible in the canvas.
3 Setting up Patterns Using this dialog, you can: • Choose Select Trace as Data Out/Aux Data Out to create a Standard pattern or as Both to create an Alternate pattern. • Enter a Description for the pattern • Provide a File Name and click Browse to locate the new pattern • Specify the pattern's Length in Bits Symbol Settings The Symbol Settings dialog is used to select the symbol mode. It displays the calculated output levels based on the current Data Out and Aux Data Out levels.
Setting up Patterns 3 This dialog contains a drop-down list which provides the following options. • Off Mode: Displays the pattern editor without any symbol coding. • OOB Mode: Displays bit combinations of both channels as shown in the table below. Table 16 NOTE NOTE Symbol Data Out Aux Data Out 0 0 0 Z 0 1 1 1 1 The OOB Mode is only valid in 'Bin' and 'Symbol' representations. The 'Hex' representation is disabled in this mode.
3 Setting up Patterns Export Trace The Export Trace dialog saves a copy of the currently edited pattern. The Export Trace dialog is shown in the figure below: NOTE The Export Trace dialog will look different for the different cases in respect to the selection of what shall be exported (Data Out, Aux Data Out, Both Pattern A, Pattern B, Both Even Bits, Odd Bits, All Bits).
3 Setting up Patterns Pattern Properties The Pattern Properties dialog is used to view/change the properties of the current pattern(s). This dialog provides the following options: • Description: Description of the pattern. This text field allows you to enter a description of the pattern's characteristics or purpose. It can be up to 256 characters long. • Pattern Type: Pattern type of pattern. This field is non-editable. • Length in Bits: Pattern length in bits. This field is editable.
3 Setting up Patterns 120 Agilent J-BERT N4903B High-Performance Serial BERT
S Agilent J-BERT N4903B High-Performance Serial BERT User Guide 4 Setting up the Pattern Generator Setting up the Pattern Generator - Concepts The Serial BERT's pattern generator generates an output signal based on a data pattern. It has the following possibilities for generating an output signal: • Providing a wide range of clock frequencies You can use the pattern generator's internal clock or an external clock for defining the frequency of the outgoing stream.
4 Setting up the Pattern Generator Input and Output Ports Input and Output Ports - Concepts Input Ports The pattern generator's input ports are used to set the pattern generator's clock frequency and to manipulate the output signal with respect to jitter, error insertion and signal output.
4 Setting up the Pattern Generator • Delay Control In Allows you to connect a signal for generating jitter in the data outputs. • Aux In The signal at the Aux In port can be used to switch between patterns A and B or to "blank" (suppress) the output signal. When a user-defined sequence is executed, it can also be used to control the sequence execution.
4 Setting up the Pattern Generator Understanding the Output Protection Circuit The generator module offers a huge flexibility for external termination schemes and external termination voltages to address common technologies. For details, please refer to the Technical Data Sheet. An internal protection circuit continuously monitors the voltages of clock, data, aux data and trigger output. It becomes active if the termination voltage is wrongly adjusted.
4 Setting up the Pattern Generator The parameters can be changed to configure the output according to the DUT’s termination requirements. Be aware that the changes are not applied immediately but only when the output protection circuit button is clicked. Output Protection on M8061A Output(s) The output protection state on M8061A output(s) occurs when the amplifier detects an overload condition which is caused by the externally connected circuitry (DUT).
4 Setting up the Pattern Generator Incorrect Termination Detection on M8061A Output(s) The incorrect termination detection state on M8061A output(s) occurs when the amplifier detects an overload condition which is caused by the externally connected circuitry (DUT). In the incorrect termination detection state, the Auto Range, Amplitude Range, Amplitude, High, Low, Offset, Coupling, Termination Model, Termination Voltage and CMI state are re-programmed to safe values.
4 Setting up the Pattern Generator As shown in this figure, the signal output levels have the following components: • Vhi is the upper voltage level of the signal. • Vlo is the lower voltage level of the signal. • Vof is the offset of the average voltage level from 0 V. • Vampt is the amplitude of the signal. When adjusting the output levels, it is important to understand the concept of how the Serial BERT handles voltages.
4 Setting up the Pattern Generator NOTE Crossover Note that the delay is on Data and Aux Data. Crossover is the voltage level where the overlapped rising and falling edges of the logic levels intersect. This adjustment varies the widths of the logic highs and lows. The following figure shows examples of crossover at 50 %, 80 %, and 20 %: Understanding how the Serial BERT Uses Logic Families and Terminations The clock and data inputs of your device have load impedances (or terminations).
Setting up the Pattern Generator 4 The following table lists the characteristics generally associated with some logic families. All values are nominal. Table 18 Logic Family Vhigh[V] Vlow[V] Ampl. [mV] Term. [V] ECL - 0.95 - 1.7 750 -2 SCFL 0 - 0.9 900 0 LVPECL + 2.35 + 1.6 750 + 1.3 LVDS + 1.425 + 1.075 350 + 1.250 CML 0 - 0.
4 Setting up the Pattern Generator Note that an internal protection circuit becomes active if the termination voltage is wrongly adjusted. The protection circuit sets the output voltages to safe levels, typically: Vhi = Vlo = Vterm = externally measured termination voltage. AC Coupling and Bias Tees The pattern generator's outputs are normally DC-coupled; even when AC termination is selected. For this reason, extreme caution must be taken when connecting your instrument to a device or test setup.
4 Setting up the Pattern Generator Input and Output Ports - Procedures You have to do the following to set up the pattern generator's output ports: Setting Logic Levels and Terminations Before you can start sending signals to your device, you have to set the logic levels and terminations: 1 From the PG Setup submenu, click the Data Outputmenu item to modify Data and Aux Data and the Clock/Trigger Output menu item to modify Clock and Trigger.
4 Setting up the Pattern Generator Adjust Output Levels (optional) Data, Aux Data, Clock and Trigger/Ref Clock offset and voltage levels can be adjusted. This is typically done when you want to tune your BER measurement or stress the device. You can use the knobs on the Serial BERT's front panel to fine-tune the data and clock amplitudes and offsets. If you want to set a specific value, you can use the numeric keyboard.
4 Setting up the Pattern Generator To change the de-emphasis amplitude ratio value (for details see “Understanding the N4916A/B De-Emphasis Signal Converter” on page 37), click inside the text field and either: • Enter the desired value directly with the numeric keyboard. • Use the knob by the numeric keyboard to set the value. You can toggle between dB and percent. Toggling does not change the value. NOTE CAUTION It is possible to set the de-emphasis ratio to a negative dB value (amplification).
4 Setting up the Pattern Generator NOTE Once you have connected the pattern generator to the N4916A/B De-Emphasis Signal Converter, the Output Levels and Termination shown on the screen refer to the outputs of the N4916A/B. Check the indicated signal levels. The output voltage range of the N4916A/B is lower than the range of the Serial BERT. The N4916A/B is first of all meant for low voltage amplitudes. If the N4916A/B cannot generate the specified signal voltages, they are reduced to safe levels.
4 Setting up the Pattern Generator Input and Output Ports - Reference The pattern generator produces clock and data outputs that serve as frequency reference and device stimulus for the device under test. NOTE NOTE The Serial BERT will not allow you to adjust a voltage beyond its limits. The limit is determined by the Serial BERT's internal hardware. If a limit is encountered, the Serial BERT sends a message to the status bar.
4 Setting up the Pattern Generator Vlo This text field allows manual entry of the logic low voltage level and displays the current value. To modify the value, click inside the text field and either: • Enter the desired value directly with the numeric keyboard. • Use the knob by the numeric keyboard to fine-tune the value. See “Understanding the Output Level Parameters” on page 126 for information about how the Serial BERT modifies the output levels.
4 Setting up the Pattern Generator Half Rate Clocking and Clock Duty Cycle A checkbox allows to enable half rate clocking: the clock at the clock output runs at half the bit rate. In half rate clocking mode, the duty cycle of the clock can be adjusted in the range of 40 to 60%. Xover This text field allows manual entry of the data's crossover percentage, and displays the current value.
4 Setting up the Pattern Generator 0V (Disable) CAUTION Before clicking this button, ensure that your device's inputs tolerate ground potential. Use this button to clamp the pattern generator's output ports (Data, Clock, Aux Data and Tigger/Ref Clock) to ground. You can see the status of the outputs in the display area of the pattern generator. Note that you can also use the buttons present below the display, to enable or disable the output ports.
Setting up the Pattern Generator 4 De-Emphasis Pre-Cursor Use this text box to enter the pre-cursor value. The value is interpreted as dB or % depending of the specified mode. This is applicable ony for N4916B. NOTE Post-Cursor 1 Use this text box to enter the pre-cursor value. The value is interpreted as dB or % depending of the specified mode. Post-Cursor 2 Use this text box to enter the pre-cursor value. The value is interpreted as dB or % depending of the specified mode.
4 Setting up the Pattern Generator The Serial BERT can be connected to an external clock to allow it to run as part of a larger external system. • Use of a modulated clock A frequency- or delay-modulated clock can be used to provide a small amount of jitter to the clock signal. • Use of a precision clock A precision clock with very low phase noise can be used to enhance the instrument's performance. This is especially interesting for long-term measurements.
4 Setting up the Pattern Generator For the pattern generator the following rules apply: • Below 620 Mbit/s, the pattern generator can only be operated with an external clock source, because the internal clock source can only produce signals higher than 620 Mbit/s. • The trigger output cannot be set up to trigger on certain pattern positions or pattern sequences.
4 Setting up the Pattern Generator To set the bit rate: 1 If the external clock source has a frequency of 10 MHz, connect the clock source to the pattern generator's 10 MHz Ref In port. If the clock has another frequency, connect it to the Clock In port. 2 Click Bit Rate Setup menu item from the PG Setup submenu. 3 Select the appropriate clock source: – Internal Clock Source This setting uses the internal clock oscillator.
4 Setting up the Pattern Generator Bit Rate - Reference The Bit Rate Setup window contains the following elements: Clock Source Clock source can be one of the following: • Internal: The clock generator uses the internal oscillator. • 10 MHz Ref: The clock generator uses the 10 MHz Reference that must be connected to the 10 MHz Ref Input. • External: The pattern generator uses the external clock that must be connected to the Clock Input. You can choose between “Automatic and Manual Mode” on page 143.
4 Setting up the Pattern Generator External PLL Clock Divider and Multiplier The internal clock (i.e. the bitrate) is the external PLL divided by the value specified in the divider field and multiplied with the value specified in the multiplier field. The external PLL clock divider and multiplier field are available if you have chosen an external PLL clock source. Value and Units This text field allows manual entry of the clock rate value.
4 Setting up the Pattern Generator Clock Rate Indicators The PG (Pattern Generator) and ED (Error Detector) bit rate indicators shown in the lower pane display their current bit rate. The ED bit rate is measured from the incoming clock signal or derived from the data signal.
4 Setting up the Pattern Generator 3 Select the trigger pattern that you want to be generated. See “ Trigger/Ref Clock Output - Reference” on page 146“ Aux Data Output Reference” on page 154 for descriptions of the available trigger signals. NOTE If you have downloaded a user-defined sequence of patterns to the pattern generator, the pattern-related settings are ignored.
Setting up the Pattern Generator 4 Alternate Pattern Trigger Level NOTE This option applies for alternating patterns only. Select this option to set the trigger level high whenever alternate pattern B is sent. The trigger output for 1024-bit patterns looks as follows: Data Out A A B A B B A Trigger Out NOTE This pattern must be at least 1024 bits long. Alternate Pattern Trigger Pulse Select this option to send a trigger pulse whenever the pattern being sent changes (A to B or vice versa).
4 Setting up the Pattern Generator Pattern Trigger Position NOTE This option is not available for alternate patterns. Select this option to send a trigger signal that is synchronized to a certain position in the pattern.
4 Setting up the Pattern Generator Data Out 10 1 1 1000100 . . . . 10 1 1 1000100 . . . . Trigger Out • Shift Trigger Position Clicking the left and right buttons will move the bit trigger position back and forward by one bit. Aux Data Out Aux Data Output - Concepts The Aux Data Output port on the Pattern Generator's Output ports allows you to generate a pulse with a frequency that is a fraction of the present clock. Note that this output is not phasesynchronized with the clock.
4 Setting up the Pattern Generator • Clock Divider • Second Channel • Multiplexer mode • De-Emphasis NOTE The options on the Aux Data Output dialog are disabled when the N4876A Multiplexer is connected to the N4903B Serial BERT. NOTE The De-Emphasis option works only if there is no external De-Emphasis box connected to the N4903B Serial BERT.
Setting up the Pattern Generator • 4 Create the pattern then save it. 1 Click "Properties" icon to set up the properties of the current pattern, including parameters and length. 2 Use the pattern editor canvas to set up your pattern. 3 Click "Save" icon to save your pattern.
4 Setting up the Pattern Generator 152 • From the Pattern menu, click Sequence Editor. • You will be now able to select edit and drop in the data you want on both the Regular Channel and the Aux Channel.
Setting up the Pattern Generator • 4 Send the patterns to the outputs by pressing “To PG”. You should be now able to send the data.
4 Setting up the Pattern Generator Aux Data Output - Reference If you wish to use the pattern generator's Aux Data Out port, set the mode from the following options: NOTE • Set the Aux Data Out divider to a suitable value (between 2 and 128). • Set the Second Channel • Set the Multiplexer mode • Set the De-Emphasis If the Divider Factor n is uneven (e.g. 3), the clock's duty cycle will not be 50%, but the signal will stay high for (n+1)/2 and low for (n-1)/2.
4 Setting up the Pattern Generator having to care for the correct setup of the required data streams. The PG pattern is considered to be the serialized output of the connected multiplexer. De-Emphasis In De-emphasis mode, an external connected passive resistor network adds two bit streams. The bit streams provided at ‘Data Out’ and ‘Aux Out’ are identical except for the bit stream at Aux Out which is shifted by one bit. The bitshift value can only be -1 (Pre-Cursor) or +1 (Post-Cursor).
4 Setting up the Pattern Generator You can enable the input after reducing the voltage to a safe level. Delay Control Input - Procedures NOTE The following procedure does not apply to Serial BERT on which the calibrated and integrated jitter injection option J10 is installed. For such instruments refer to “Jitter Setup - Procedures” on page 407. To set up a jitter tolerance test on an instrument without option J10: 1 Press PG Data Setup in the PG Setup menu.
4 Setting up the Pattern Generator NOTE This checkbox is not available on instruments on which the calibrated and integrated jitter injection option J10 is installed. On such instruments, the Delay Control In port is enabled or disabled from the Jitter Setup window. Error Addition/Insertion Error Addition/Insertion - Concepts To test error correction algorithms, alarms and other functions that are imbedded in the data pattern, you can insert logic errors (flipped bits) into the pattern.
4 Setting up the Pattern Generator Using an External Signal for Inserting Errors To configure the Serial BERT to insert a single error into the output stream according to an external signal: 1 Connect an external instrument to the Error Add port. The signals received at this port must be TTL-compatible. 2 Click Error Add Setup menu item from the PG Setup submenu. 3 Select External (Error Add) and click OK to close the dialog box.
Setting up the Pattern Generator TIP NOTE 4 To find out how your DUT reacts on very small bit error rates, set up the pattern generator to enter errors once every 10-12 bits and run a longer accumulative test. You can then find the DUT's true error rate by calculating the difference between the bit error rate set up in the pattern generator and the accumulated bit error rate found by the error detector.
4 Setting up the Pattern Generator Error Add Setup Dialog Box The Error Add Setup dialog box provides the following configuration options: Channel Select With this option, you can apply the error insertion either on Data, Aux or Both channel(s). NOTE Aux and Both options will be disabled in single channel mode.
4 Setting up the Pattern Generator Internal Select this option to use internal error insertion functions: • Error Rate Allows to select from a preset bit error ratio from 1.000E-02 to 1.000E-09. Errors will be added to the output pattern to produce this error ratio. • Average Number of Bits Between Errors Allows you to enter the average number of valid bits between two errors. This is the inverse of the bit error ratio. Off Select this option turn off all internal add rates and external error inputs.
4 Setting up the Pattern Generator further configuration possibilities. This pattern is sent as long as the output port is enabled. See “Setting up Patterns - Concepts” on page 61 for more details. • Alternating patterns consist of two patterns, one of which is pattern A (the A half), the other is pattern B (the B half). Both patterns are of equal length, each up to 16Mbit. The Alternate Pattern Control dialog box lets you control when which pattern is sent.
Setting up the Pattern Generator 4 – Continuous A Only pattern A is output repeatedly. – Continuous B Only pattern B is output repeatedly. – Alternate AB The output stream switches between pattern A and pattern B. – Single Shot B With this option selected, the Insert B button is enabled. 3 Select OK when your selection is complete. Manually Inserting Pattern B into the Output Stream You can insert a single shot of pattern B into the continuous stream of pattern A: 1 Click Alt.
4 Setting up the Pattern Generator 3 In the Aux In section, define whether the Serial BERT should insert pattern B in Level Sensitive or Edge Sensitive mode. See “Aux In” on page 165 for more information on the different available modes. 4 Click OK to close the dialog box. Suppressing the Outputs via External Signal You can configure the Serial BERT to suppress the output according to an external signal: 1 Connect an external instrument to the Aux In port.
4 Setting up the Pattern Generator Continuous B Select this option to send out only pattern B repeatedly. B0 Data Out B1 B2 B B B B B B B B B B B B Alternate AB Select this option to alternately send out patterns A and B (A, B, A, B...). A0 Data Out B0 A0 B0 A A A A B B B B A A A A B B B B Single Shot B Select this option in the Alternate Pattern Control dialog box to enable the Insert B button on the main display.
4 Setting up the Pattern Generator Aux In Data Out • A A B A B B B Edge Sensitive Pattern A is sent out until a rising edge at Aux In is detected. Then, at the next complete repetition of pattern A, a single occurrence of pattern B is inserted into the output. Aux In Data Out • A A B A B A A Output Blanking Pattern A is sent when the signal at Aux In is low. Pattern B is not sent in this mode.
S Agilent J-BERT N4903B High-Performance Serial BERT User Guide 5 Setting up the Error Detector Setting up the Error Detector - Concepts The error detector analyzes an incoming bit stream, compares it to the expected pattern, and locates any inconsistencies. The error detector requires the following settings to work correctly: • The expected pattern The error detector needs to "know" which data to expect so that it can detect bit errors.
5 Setting up the Error Detector • BER location mode You can specify whether all errors are counted or only the errors that occurred on a particular bit position or range of positions. • Audio warnings You can set up the audio warnings so that the instrument beeps when a certain BER is exceeded. • Trigger output for external measurement instruments This allows you to connect other devices for further error analysis.
5 Setting up the Error Detector Understanding Error Detector Input Ports The error detector has the following input ports: • Clock In This port needs to be connected to a clock signal unless you use the error detector in Clock Data Recovery (CDR) mode, where it derives the clock frequency from the data port. See “Clock Setup - Concepts ” on page 173 for more details. • Data In and Data In This port is connected to the data signal and the inverted data signal.
5 Setting up the Error Detector single trigger pulse is sent. If continuous errors are detected, the error out signal would be a clock signal. n Errors within Error 128bit granulation detected (n<128) n Errors exceeding 128bit granulation (n>128) A Data IN A Error Out single error marking • Aux Out This port can be used to output the clock signal or the data signal to another connected device.
5 Setting up the Error Detector Why Can Wrong Terminations Damage Your Device? Choosing wrong terminations may cause your device to output voltage levels that are not as expected. It may also cause excessive current or current flow in the wrong direction, which can damage your device. NOTE The clock termination is set to 50 Ohm, AC-coupled, and cannot be changed.
5 Setting up the Error Detector Data Input Setup - Reference The ED Input Setup dialog is accessed from the Sampling Point Setup window. It is used to set up the error detector's data input port: Input The selection in this list defines how the signals arriving at the Data In and Data In connectors are interpreted. The following options are available: • Differential If differential mode is selected, both input ports need to receive a signal.
5 Setting up the Error Detector Termination In this field, enter the termination voltage that is appropriate for the incoming data signal. This selection should be made before the device is connected to the analyzer. If the input is set to differential mode, a termination voltage can only be set for a device that is DC-connected. CAUTION Selecting the wrong terminations may damage your device. The Data In port is connected to a 50 Ohm load impedance (or termination) within the error detector.
5 Setting up the Error Detector clock from the pattern generator), or extract the clock signal from the incoming data (CDR mode). CDR mode does not work for all kinds of data patterns. For example, if the device under test sends only blocks of ones and zeros, there are no transitions in the data stream and the Serial BERT cannot recover the clock. Also, if you are testing bursts, there are some special considerations for setting up CDR. See the following sections for details.
5 Setting up the Error Detector NOTE The “frequency” is the modulation frequency, and “amplitude” is the tracking of the PLL. Data rate and loop parameters The data stream contains multiple frequencies, and the CDR needs to know the expected data rate. The user entered information is necessary to lock the CDR, and to detect a loss in lock condition. The entered frequency should be accurate, and better than 100 ppm.
5 Setting up the Error Detector The threshold voltage can be derived from the input signal via a low-pass filter. This will work fine for most applications. But applications that do not provide a continuous data stream at the input (for example, any application using bursts) cannot use this low-pass filter, because the threshold voltage will drift from the correct level when there is no input. In such cases, the threshold can be specified manually.
5 Setting up the Error Detector The following figure clarifies the behavior in the range between 615Mbit/s and 620 Mbit/s: Limitations do not apply in this range 620 615 Bitrate [Mbit/s] Limitations apply in this range For the error detector the following rules apply: • For low frequencies, you cannot use the automatic data alignment functions (Auto Align and Data Center). Instead you need to align the error detector manually.
5 Setting up the Error Detector With this method you can still place the sampling point anywhere in the clock cycle to find the optimum sampling point, even at low frequencies. 3.226 ns Clock 310 Mbit/s Covered Range However, if the bit rate falls below appr. 310 Mbit/s, even with this method gaps occur in the range of possible sampling points. 6.
Setting up the Error Detector 5 8 Either enter the transition density manually, or run a measurement to update this value. Clock Setup - Reference The Clock Setup window contains the following elements: Error Detector Bit Rate The bit rate of the data reaching the error detector is displayed in the lower right corner of the user interface. The bit rate of the error detector is set by the clock signal received by the error detector.
5 Setting up the Error Detector Preset List The Preset list contains clock rate values that are commonly used by the error detector. These presets also modify the loop parameters of the tunable CDR according to the selected standard. The factory presets consider the parameters as given in the selected standard. Click a preset to select it for the error detector clock rate. Double-click a preset to modify it.
5 Setting up the Error Detector transition density. In such a case enter the value given in the specification, so that the CDR behaves according to the standard. If a standard from the preset list is selected, this field is preset. Loop Bandwidth This is the range of the CDR loop bandwidth. In this field the user should enter the loop bandwidth value; the range is within 100 KHz to 12MHz. If a standard from the preset list is selected, this field is preset.
5 Setting up the Error Detector • press the Measure button to measure the voltage level of the incoming signal while a burst signal is applied. Once the level is defined, it remains fixed for the following measurements. Note that the entered voltage level has to be within the input range of the error detector. If the value lies outside this range, the change is rejected and an error message is displayed. CDR Spread Spectrum Clocking This control is used to adapt the CDR to an input bit stream with SSC.
Setting up the Error Detector NOTE 5 All measurements (except Jitter Measurement) in "8B/10B Symbol Comparison", "Bit Comparison without PCIe3 SKPOS" and "Bit Comparison without USB3.1 SKPOS" modes, automatically switch to BER mode and work in Bit Recovery Mode. These changes are reflected in GUI too. Jitter measurements also works in "8B/ 10B Symbol Comparison" mode and uses calculated BER (cBER) to check against Target BER.
5 Setting up the Error Detector Understanding the Error Free Receiving in PCI® Express 3.0 128B/ 130B Encoded Data Comparison When RX detects incoming data correctly, the pattern looped back has the "same" content except the length of the Skip Ordered Set (SKPOS) primitives. The change in the SKPOS length by DUT is to compensate for the speed differences of the clock domains.
Setting up the Error Detector NOTE 5 The "Bit Comparison without USB3.1 SKPOS" error ratio mode supports the bit rate up to 10.35 Gbit/s. However, if you want to use bit rate greater than 10.35 Gbit/s then switch to "8B/10B Symbol Comparison" or "Bit Comparison without PCIe3 SKPOS" error ratio mode and then go to "Bit Comparison" mode. Error Ratio - Procedures This section explains how to set up an Error Ratio.
5 Setting up the Error Detector 4 Choose the error ratio mode from the following options: a Bit Comparison b 8B/10B Symbol Comparison c Bit Comparison without PCIe3 SKPOS d Bit Comparison without USB3.1 SKPOS NOTE NOTE The Bit Comparison option is selected as default. Switching among the error ratio options might take some time; please wait untill the switching process is completed.
5 Setting up the Error Detector However, if you select error ratio based on "8B/10B Symbol Comparison", follow the below steps. 5 Click cBER Setup. It opens a cBER Setup dialog. This dialog offers the choice to calculate cBER based on either SER, FER, FSR, ISR, or DER with either automatic conversion factor or manually specified conversion factor.
5 Setting up the Error Detector 9 You can view the details of the selected presets. To do so, either click View/ Edit Preset button or double-click the preset. Remember, the details of the system presets will be in the disable mode. User presets can be modified. To do so, either click View/Edit Preset button or double-click the preset. System presets can not be modified.
5 Setting up the Error Detector NOTE The "Automatic" mode is selected by default for symbol alignment. The "ReAlign" button is disabled in "Automatic" mode. During the Jitter Tolerance Characterization/Compliance measurement, the "Automatic" symbol alignment mode switches to "Manual" mode and restore back to "Automatic" mode when the measurement is finished/aborted. 12 From the Sampling Point Setup window, set the sampling point. For more information, see “Sampling Point Setup - Procedures” on page 196.
5 Setting up the Error Detector Error Ratio - Reference 8B/10B Presets Dialog The 8B/10B Presets dialog allows you to enter a short description, values for Filler Primitives definition and Alignment Symbol for the new preset. This dialog contains the following elements: Filler Primitives Definition Filler Primitives are inserted or deleted for clock tolerance compensation. These are not compared and therefore cannot be counted as errors. Filler Primitives contain filler symbols.
Setting up the Error Detector 5 Table 19 Standard Filler Symbols USB3.0 Skip: K28.1, K28.1 10B Symbol alignment: K28.5 SATA Align: K28.5, D10.2, D10.2, D27.3 10B Symbol alignment: K28.5 MIPI MPhy Filler: K28.1 10B Symbol alignment: K28.5 Display Port Filler: Empty String ("") 10B Symbol alignment: K28.5 SAS Align: K28.5, D10.2, D10.2, D27.3 K28.5, D7.0, D7.0, D7.0 K28.5, D1.3, D1.3, D1.3 K28.5, D27.3, D27.3, D27.3 10B Symbol alignment: K28.5 PCle 1&2 Skip:K28.5, K28.0; K28.
5 Setting up the Error Detector 10B Alignment Symbol The 10B Alignment Symbol contains the K28.1, K28.5 and K28.7 symbol for 10B alignment. You can use the drop-down list to specify symbols. All standards supported by Serial BERT use the K28.5 symbol for 10B alignment. However, the standards differ in the way if the K28.5 is only used during some training sequence or if it comes also later in the data stream. In SATA for instance, K28.
Setting up the Error Detector 5 Sampling Point Setup Sampling Point Setup - Concepts This section provides basic information on the sampling point setup and eye diagrams. How Does the Sampling Point Setup Work? The sampling point of a data signal is defined by two values: a point in time and a voltage level. Each bit of the data signal is sampled at this point in time and in reference to this voltage level.
5 Setting up the Error Detector observed. The eye is bounded by overlaid logic 1 and 0 voltages (top and bottom) and multiple 0 to 1 and 1 to 0 transitions (left and right). Different portions of the bit pattern Eye diagram Overlay of all portions of the bit pattern The sampling point of the error detector must be set within the data eye. The error detector uses the eye diagram to graphically display the location of the sampling point.
5 Setting up the Error Detector Understanding Bit Recovery Mode In bit recovery mode (BRM), the error detector does not expect any particular pattern. Nevertheless, it detects bit errors. In bit recovery mode, the error detector uses two sampling points. The second sampling point is not visible. Both are set to the expected optimum position when the Auto Align button is pressed (with respect to clock rate, signal voltages and offset, decision threshold, and so on).
5 Setting up the Error Detector • Error Location Capture cannot be used. • The function Error Detector Pattern tracks the Pattern Generator Pattern is disabled. This function is not re-enabled when the BRM is terminated. • BRM does not work above 11.5 Gb/s when the CDR is enabled. BRM indicators: • The mode is indicated in the status panel and in the BER Bar. • The mode is indicated in all screen dumps and log files. • The Measurement User Interface shows when this mode is used.
5 Setting up the Error Detector The Sample Point in the eye diagram moves horizontally as you change the value. TIP NOTE While adjusting the data input delay, you can monitor the BER bar on the analyzer display, or listen to the BER warning tones. Higher BERs may indicate that you are measuring at the edges of the data eye. Pattern synchronization can be lost during sampling point adjustment. If Auto Sync is enabled, the Serial BERT will resynchronize the patterns automatically.
5 Setting up the Error Detector 2 Click the Bit Recovery Mode button. The Sync Now button is no longer available. Note that you can also enable the bit recovery mode from the Pattern window. Click the Pattern Select icon and then the Error Detector Pattern tab. If this tab is not available, disable the checkbox Error Detector Pattern tracks the Pattern Generator Pattern. 3 Click the Auto Align button.
5 Setting up the Error Detector Data Delay The data input delay defines the point in time (in reference to the clock signal) at which the incoming data signal is measured. Specifically, it is the time delay from the active clock edge to the time at which the data is actually sampled. This field allows the manual entry of the data input delay, and displays the current value in picoseconds. This delay can be set as high as 1 bit period or 10 ns (10,000 ps), whichever is less.
5 Setting up the Error Detector Sync Now Click this button to manually start a pattern synchronization. See “What Type of Synchronization Should You Use?” on page 208 for information on when you should use this function. Note that this button is not available in bit recovery mode. 0/1 THReshold Center Starts an auto-search function that sets the 0/1 threshold to the optimum point of the incoming data eye on the vertical voltage axis without changing the data input delay.
5 Setting up the Error Detector Cancel Click this button to cancel the Auto Align, 0/1 Threshold Center, or Data Center functions while they are in progress. The following parameters will be returned to their previous value or status: • Auto Align Canceled – Data Delay and 0/1 Threshold values returned to previous. – Status of Data Inverted and Avg. 0/1 Threshold checkboxes returned to previous. • 0/1 Threshold Center Canceled – 0/1 Threshold value returned to previous. – Status of Avg.
5 Setting up the Error Detector This function is useful for providing a "starting point" for sampling somewhere within the data eye on the vertical voltage axis. It is recommended to use this prior to clock/data alignments.
5 Setting up the Error Detector Trigger and Aux Output Trigger and Aux Output - Concepts The error detector has a Trigger Out and an Aux Out port that you can use to send signals to other external devices such as an oscilloscope. The Trigger Out can be set up to send a trigger to an external device either according to the clock, or according to the data pattern being generated.
5 Setting up the Error Detector threshold is set below or above the data eye, the output at Aux Out will be constant high or low, respectively. 4 Click OK to finish the Trigger and Aux Out setup. Trigger and Aux Output - Reference The available options in the ED Trigger & Aux Output window are: Trigger You have the following options to configure the signal at the Trigger Out port: • Clock Divided by n The trigger is set every n bits.
5 Setting up the Error Detector The usage of the enhanced error detector trigger is based on the PCIe3 SKPOS (Skip Ordered Set). The 130 bit SKPOS is used as detect word and is not configurable. The SKPOS removal unit triggers upon a detected word of 66, 98, 130, 162 and 194 bit SKPOS which is used with the pattern generator AUX input to advance a pattern sequence and thus implements a simple handshake.
5 Setting up the Error Detector Introduction to Pattern Synchronization Pattern synchronization (sync) refers to aligning the incoming data pattern with the internal reference pattern. This is accomplished in one of two ways: Hardware-Generated Patterns Memory-Based Patterns For 2^n-1 PRBS patterns, bits from the incoming data pattern "seed" the error detector's pattern generator, causing it to generate a precisely aligned internal reference pattern.
Setting up the Error Detector Expected Pattern Detect Word 48 bits Detect Word 48 bits 5 Incoming Bit stream Detect Word 48 bits Reference Point Correct Sync Detect Word 48 bits Detect Word 48 bits • Multiple instances of the detect word with false synchronization Expected Pattern Detect Word 48 bits Incoming Bit stream Detect Word 48 bits Reference Point False Sync Detect Word 48 bits Detect Word 48 bits Detect Word 48 bits If the error detector attempts to synchronize on the incorrect de
5 Setting up the Error Detector The detect word on which the error detector attempts to resync is chosen strictly by chance. So if there are two instances of the detect word in the pattern, the error detector has a 50% chance of selecting the correct one. The more instances of the detect word exist in the pattern, the higher are the chances for incorrect synchronization. The software attempts in any case to identify a 48-bit pattern that occurs as seldom as possible in the pattern.
Setting up the Error Detector NOTE 5 Adjusting the data input delay may cause momentary clock loss. If you select Manual Sync mode, this may also result in sync loss. • Burst sync mode is a special operating mode for measuring data in bursts of bits, rather than one continuous stream of bits. For more information, refer to “Introduction to Burst Sync Mode” on page 211.
5 Setting up the Error Detector How Can You Tell if Your Synchronization is False? You may suspect false synchronization under the following conditions: • You are using a pattern other than PRBS and the error detector gains sync, but it measures a constant, fixed error ratio. • You are using a pattern other than PRBS and the error detector gains sync, but auto-search functions (Auto Align, Clock/Data Center, 0/1 Threshold Center) repeatedly fail.
Setting up the Error Detector 5 Introduction to Burst Sync Mode The burst sync mode is a special operating mode for measuring data in bursts of bits, rather than one continuous stream of bits. The burst sync mode measures bit error rates for each burst of data after the error detector synchronizes to the incoming pattern. The signal at the Gate In port controls the timing of synchronization and error counting for each burst. This function is useful, for example, to analyze recirculating loop data.
5 Setting up the Error Detector How Burst Sync Mode Works The following figure presents the basic order of events that make up a burst mode measurement. It also illustrates how the instrument operates while in burst mode. no 1 Gate In? yes 2 CDR? no yes CDR Settling Time Synchronization Time 3 success Sync? Register Incoming Bits 4 5 failed 7 Bad Burst +1 no Gate In? yes no BER OK? yes Total Burst + 1 NOTE Incr.
Setting up the Error Detector 5 3 The error detector can then synchronize to the incoming signal. For PRBS, the first received bits are used to seed the synchronization. If there is an errored bit in this phase, the synchronization fails (bad burst). For memory-based patterns, a unique 48-bit detect word is used for the synchronization. This pattern should be available one time only in the pattern. If the detect word is not found, synchronization fails (bad burst).
5 Setting up the Error Detector The Bit Count Time is the part of the burst from which the bits can actually be counted. The remainder of the burst is covered by the Begin Margin and End Margin. The CDR Settling Time is the time that the error detector requires in CDR mode to get the clock from the data stream. The Synchronization Time is the time the error detector requires to synchronize to the pattern. This time depends on the pattern type (PRBS or memory-based).
5 Setting up the Error Detector Table 21 Parameter Values Begin Margin CDR Settling Time + Synchronization Time End Margin Valid after Gate Bit Count Time Burst Length - Begin Margin - End Margin Gate Active Begin Margin + Bit Count Time Optimizing the Timing There are three things to watch when optimizing the timing for burst sync mode: • BER The bit error rate increases when the gate closes too late. The duration of the signal should typically be reduced.
5 Setting up the Error Detector Starting too late: If the Gate In signal is applied too late, the Bit Count Time will be shortened, and thus the burst sync ratio will be low (which forces the measurement of more bursts to obtain the necessary level of reliability).
Setting up the Error Detector 5 2 Select the synchronization mode (Normal or Burst sync mode). NOTE The burst sync mode is disabled in "8B/10B Comparison", "Bit Comparison without PCIe3 SKPOS" and "Bit Comparison without USB3.1 SKPOS" error ratio modes. 3 If you selected Normal synchronization mode, choose whether you want Automatic or Manual synchronization. 4 Specify the Sync Threshold at which the pattern will be recognized at synchronized.
5 Setting up the Error Detector 7 Press OK to confirm your settings. Pattern Synchronization - Reference The error detector can synchronize the data patterns in the following ways: • In Normal Sync Mode, you can choose between – Automatic Sync With this option selected, the error detector constantly tries to synchronize the patterns when the BER threshold is exceeded.
5 Setting up the Error Detector anymore. Also test results would not correlate with other error counters such as protocol testers anymore. The high error ratio also serves as a trigger to perform an automatic resynchronization. So on one hand automatic pattern re-synchronization is desired during receiver testing but it will be necessary to mask respectively do not account the high error count (which was caused by the not synchronized error detector) into the measurement.
5 Setting up the Error Detector Error Accumulation Error Accumulation - Concepts The Serial BERT instantly starts measuring the BER when the error detector receives data. However, to perform tests that can be repeated and compared, you can collect the measurement data in several ways and save it to a log file. The available options are to test: • for a specified time, • until a certain number of errors occurred, • until a specified number of bits have been sent, • until you stop it manually.
5 Setting up the Error Detector Error Accumulation - Reference The Accumulation Setup window contains the following elements: Activation Mode • Manual Activate this option to configure the error detector to start and stop accumulation when the front panel buttons are pressed. • Single Activate this option to configure the error detector to accumulate over one accumulation period and then stop.
5 Setting up the Error Detector Period The accumulation period cannot be specified for Manual tests. • Time Select this option to configure the error detector to accumulate bit errors for a specific period of time. You can enter the desired time period in the Days, Hr (Hours), Mn (Minutes), and Sec (Seconds) fields. You may use the numeric keypad or front panel knob. The minimum value is 1 second and the maximum value is 99 days, 23 hours, 59 minutes, and 59 seconds.
Setting up the Error Detector 5 BER Location BER Location - Concepts The Serial BERT can be run in BER location mode. In this mode, the errors are not counted on all bits of the data signal, but on a particular bit or block of bits in the pattern. You can specify which bit position(s) are considered during the measurement. This allows, for example, to calculate the BER for the header bits or the payload of a data stream exclusively.
5 Setting up the Error Detector All Bits With this option, all errors are counted. Block With this option, only errors occuring in the block between the specified Start Bit and Block Length are counted. Single Bit With this option, only errors occuring at the specified position are counted.
5 Setting up the Error Detector Audio Signals - Procedures To set up the warning sounds, do the following: 1 Click Audio menu item from the ED Setup submenu. 2 Check the Audio on checkbox to switch on the sounds. TIP To switch the audio signals on and off, you can also use the Audio On/Off front panel button. 3 Use the Main Volume slider to adjust the volume. After adjusting the slider, a test tone is played at the new volume. NOTE If no sound is heard, please check whether the system volume is muted.
5 Setting up the Error Detector Main Volume Click any point on the slider or drag it to change the main volume level. You will hear a test tone at the new volume level. Audio on BER Alarm Click this option to make the analyzer play warning tones when a specific BER is exceeded. The BER Alarm Threshold is the threshold at which tones are produced. Enter the BER threshold in the following format: (number)E(exponent). For example, 1.5E-3.
S Agilent J-BERT N4903B High-Performance Serial BERT User Guide 6 Advanced Analysis Advanced Analysis - Concepts The Serial BERT offers several different kinds of advanced measurements for various purposes: DUT Output Timing/Jitter This type of measurement is used to measure the timing and jitter behavior for a device under test (DUT). It uses a bit error rate (BER) measurement to evaluate the shape of the eye for the output signal of the DUT.
6 Advanced Analysis Fast Eye Mask The Fast Eye Mask measurement is first of all meant for production and screening tests. It allows to determine very quickly whether the eye opening seen at the output signal of a device is within specifications, that means, within certain timing and voltage limits. This is achieved by measuring the bit error rate at a limited number of test points. Eye Diagram The Eye Diagram allows a quick check for the DUT’s signal output, and determines the signal quality.
Advanced Analysis 6 changing the measurement settings after the measurement has been run, please note: • Parameters that affect the data capture. Changes on the Parameters page take only effect if you run the measurement again. • Parameters that change the display of the measured data. Changes on the Pass/Fail, View, Graph, and Color tabs only affect the display of the results. There is no need to repeat the measurement.
6 Advanced Analysis How to Export Measurements To export the measurement data into a .txt file for later use in external spreadsheet applications: 1 After your measurement has finished and the results are displayed, select Export Data from the Analysis menu. 2 In the Export dialog box, select the path and file name for the .txt file. Alternatively, you can activate the Clipboard checkbox to copy the data to the clipboard. In this case you can then paste it into any other application.
Advanced Analysis 6 4 Repeat the measurement with modified parameters and observe the new results. 5 To distinguish the two graphs, change the color of one graph by tipping the red Color field. You can now directly compare the graphical and numerical results of the two Output Levels measurements. How to Rename Copied Measurements To rename a copied measurement for easy distinction: 1 Highlight the copied measurement. 2 Press Enter (or choose Rename from the context menu).
6 Advanced Analysis How to Change Measurement Properties and the Graphical Display The various tabs in the Properties dialog box allow you to specify the parameters for the measurement. The available options are not the same for all measurements. See the Reference information of the respective measurement type for details. To modify the measurement properties: 1 Press the Properties button to open the Properties dialog box.
6 Advanced Analysis Depending on the current measurement type, you can change: – BackColor: The background color of the graphs (default is white) – BERMarkerColor: The color of the BER Threshold indicator (default is red) – ForeColor: The foreground color of the scales and frame of the graphs (default is black) – FreqRangesColor: Color of the selected frequency ranges (default is light yellow) – GridColor: The color of the dashed grid lines (default is gray) – PowerMarkerColor: The color of the Noise Thre
6 Advanced Analysis Advanced Analysis - Reference This section covers reference information such as definitions of important parameters and the descriptions of supplementary functions. Exporting Result Data If you want to use the measurement results with other applications, you can export the data to a file via Analysis - Export Data.... The contents of the resulting file may look as follows: Date:;02/11/05 03:33:17 Version:;1.0 Type:;TM Fast Eye Mask SB-Electrical UI:;9.
Advanced Analysis 6 Timing Unit Definitions The timing unit specifies the timebase for the measurements. It is possible to switch between the unit interval (UI) or seconds for the time scale. The timebase is set on the View page of the Properties dialog box.
6 Advanced Analysis Example At 100MHz, the pulse period is 1/100 × 10-6s = 10-8s = 10ns; this is the unit interval (UI). For example, a value - in fact, any parameter that is specified in time units of 37ns is equivalent to 37ns/10ns=3.7UI (time values are expressed as multiples of the unit interval).
6 Advanced Analysis The following values are displayed for the current marker positions: Table 23 Value Description Upper Right Time and BER of the upper right intersection of the displayed marker lines (X2 and BER1) Lower Left Time and BER of the lower left intersection of the displayed marker lines (X1 and BER2) Delta Distance between the two vertical marker lines on the time axis (X2-X1) and distance of the two horizontal marker lines on the BER axis (BER1-BER2) The markers can be moved to kee
6 Advanced Analysis This formula describes a bell-shaped Gauss curve. If μ is zero and σ varied, you would get the curves illustrated in the figure below: The height and position of a normal distribution can be specified in terms of two parameters: μ and σ. The parameter μ is the mean, the parameter σ is the standard deviation. The Gaussian marker shows such a curve. Position, height, and width of this curve can be changed by dragging the handles, and the actual parameter values are displayed.
Advanced Analysis 6 Table 24 Parameter Symbol Meaning Mu μ Mean. DUT Output Timing/Jitter measurement: The position of the marker center on the time scale. Output Levels measurement: The position of the marker center on the vertical threshold scale. DUT Output Timing/Jitter Measurement Sigma σ Standard deviation. The RMS value of the marked area. Kappa κ Linear scaling factor. A Gaussian marker is used when the jitter graph is displayed. It is most useful if deterministic jitter is present.
6 Advanced Analysis As the bathtub borders are not uniform (both have two edges), the linear derivative (the jitter) will show two peaks: If you switch to linear scale and enable the marker, you can see its bell shape. You can measure the random jitter distribution of each peak as well as the distance between the peaks, which means the deterministic jitter. You can also use the marker with logarithmic scale.
Advanced Analysis 6 But your dBER distribution may also look like this: The measurement will calculate the Level and Standard Deviation results from all data points. The marker allows you to measure the μ (Mu) and σ (Sigma) of the individual peaks. DUT Output Timing/Jitter DUT Output Timing/Jitter - Concepts This type of measurement is used to measure the timing and jitter behavior for a device under test (DUT).
6 Advanced Analysis Output Timing Characteristics The sampling point is swept automatically within a 1.5 clock period to generate a "bathtub" curve. The resulting graph is centered around the optimum sampling point of the port. In addition, the results are available in a tabular view. If a clock signal is defined, the software measures the data to clock alignment and displays the absolute delay.
Advanced Analysis 6 Example Results The following illustration shows the resulting graph of a typical DUT Output Timing measurement: The following figure shows a typical jitter histogram with two peaks indicating the presence of random and deterministic jitter: For a detailed explanation of the Fast Total Jitter measurement results see “Explanation of the Fast Total Jitter Measurement Results” on page 262.
6 Advanced Analysis 2 All measurement points that have BER between the BER Threshold and Minimum BER for RJ/DJ Separation are transformed into Q-space. The Q-factor describes the signal-to-noise ratio at the decision circuit. It is described in “Understanding the Q-Factor Results” on page 282. 3 Linear regression is performed for both the left and right edges. 4 The mean and sigma are calculated for both lines: – RJ is calculated as the mean of the two sigmas.
6 Advanced Analysis Estimated Total Jitter The Estimated Total Jitter (TJ) allows you to predict the jitter expected for very low bit error rates that would take a long time to measure. It is obtained by extrapolating the measured BER curves. The TJ is estimated by extending the BER curves (based on the points detected between the BER Threshold and the Minimum BER for RJ/DJ Separation) to the Residual BER for RJ/DJ Separation level. The estimated TJ is the period minus the width of the measured eye.
6 Advanced Analysis Explanation of the Fast Total Jitter Measurement The Fast Total Jitter measurement is an optimized method to determine the total jitter for devices that generate a very low error density (BER well below 10-10). To measure (not estimate) the total jitter for a device with a BER of 10-12 with conventional methods, one usually needs to compare more than 1012bits for each sample point.
6 Advanced Analysis operates with a BER better than let us say 10-12; whether the true BER is 1.1 × 10-13 or 2.7 × 10-15 is irrelevant. To abort the measurement for a single point and proceed to the next, we need two limits that tell us whether the BER is above or below the given threshold. These limits have been calculated from the error probability density functions applicable to BER measurements. The equations were solved for a level of confidence of 95%.
6 Advanced Analysis Note that there is a gap where the BER is so close to 10-12 that we cannot decide. For example, if we compared 3 × 1012 bits and got two errors (a measured BER of 0.667 × 10-12), we are in the "uncertain" white area on the graph. In such a case, we need to transmit more bits until the number of bits either reaches the upper limit (6.296 × 1012), or until we see more errors.
Advanced Analysis 6 We do not need to know the exact BER values at x+ and x-. It is sufficient to assure that BER(x-) is greater than 10-12 and BER(x+) is lower than 10-12 at a confidence level of 95 %. The algorithm then assumes that xL (for the left-hand slope of the bathtub curve) is in the middle of the bracketing interval. After repeating the procedure to determine xR (for the right-hand slope), it calculates the total jitter peak-to-peak like in the standard timing/jitter measurement.
6 Advanced Analysis DUT Output Timing/Jitter - Procedures This section shows how to set up and use the DUT Output Timing/Jitter measurement. As an example, we measure the output timing and jitter behavior of a shielded cable.
Advanced Analysis 6 9 Press Sync Now and then Auto Align to find the optimum sampling point. Check that the synchronization and the alignment were successful. None of the error indicators at the top of the user interface should show red and the resulting BER should be zero. How to Execute the DUT Output Timing/Jitter Measurement To run the DUT Output Timing/Jitter measurement: 1 Switch to the Analysis area. If the Output Timing screen is not yet displayed, press the Output Timing icon.
6 Advanced Analysis These settings are used for data collection. Changes here require the test to be run again. See “Parameters Tab” on page 252 for details. – Pass/Fail tab These settings determine whether the calculated results are recognized as passed or failed. However, a new test run is not required when doing changes here. See “Pass/Fail Tab” on page 254 for details. – View tab, Graph tab, and Color tab All settings on these tabs only affect the way the data is displayed.
6 Advanced Analysis • Number of Errors After this number of errors, the measurement stops for the current sample point and moves to the next one. This allows you to speed up the measurement. You can switch off this option if only the number of compared bits is important. NOTE The measurement moves to the next sample point when the first of the two criteria is reached. Both numbers are ignored if the Fast Total Jitter measurement is selected.
6 Advanced Analysis – Fast Total Jitter at BER This enables the Fast Total Jitter measurement. Before enabling this measurement, you need to know the BER floor of the device and to specify a BER threshold that is above that floor. For details see “Explanation of the Fast Total Jitter Measurement” on page 246. Pass/Fail Tab The Pass/Fail tab of the Properties dialog box allows you to specify the criteria to decide whether the DUT passes or fails the test.
6 Advanced Analysis NOTE If the results of a Fast Total Jitter measurement are displayed, only the appropriate parameters are compared and flagged: • Phase Margin • Optimal Sample Delay • Total Jitter Peak to Peak • Fast Total Jitter Uncertainty Other pass/fail limits may be enabled but are ignored. In the tabular view, each of the calculated values will be marked with an icon if it failed the test.
6 Advanced Analysis View Tab The graph shows either the bathtub curve or the jitter distribution vs. time. Bit Error Rate Graph The BER graph (the bathtub) shows the BER vs. sample delay. The BER graph can be viewed in either linear and logarithmic view. The logarithmic view is shown above. Jitter Histogram 256 The DUT Output Timing/Jitter measurement calculates the jitter histogram as the absolute of the derivative of the measured bit error rate (jitter=dBER/dt).
Advanced Analysis NOTE 6 Because the right-hand slope of the tub does not provide additional information on the jitter, the measurement's jitter display shows only the portion at the left-hand side of the optimum sampling point. The jitter histogram allows you to visually inspect the jitter components: • Random Jitter (RJ) • Deterministic Jitter (DJ) • Estimated Total Jitter (TJ) “Jitter Measurement Parameters” on page 260 describes how these components are calculated.
6 Advanced Analysis If this is selected, RJ/DJ separation is not available. • BER Threshold To calculate the parameters for the given BER threshold. This is the BER level for which output timing numerical values (phase margin, skew, etc.) are calculated. It is also the upper limit of the BER range for RJ/DJ separation. The BER threshold influences some of the parameters of the DUT Output Timing measurement.
6 Advanced Analysis Markers To analyze the graphs at a particular point, you can use the markers. Additionally, you can display all related values for the markers in the marker readout. Pressing the Reset Markers button will set the markers back to the default positions. Zoom Show Measured Points Several zoom factors are available. When you show the zoom graph, you can also allow the zoom graph to track the mouse (or your finger, if you are working directly on the Serial BERT).
6 Advanced Analysis Table 26 Parameter Description Definition Optimal Sample Point Delay The average of the left (A) (A+B)/2 and right (B) bathtub/BER threshold intersections. Phase Margin The period of time where the bit error rate is lower than the BER threshold. B-A The A and B values are the left and right intersections of the bathtub curves with the BER threshold. Obviously, all values change if the BER threshold is modified. The following illustration shows an example for a measurement.
6 Advanced Analysis • Jitter Mean Mean value for total jitter. Calculated as the weighted average of the left edge jitter histogram. • Random Jitter RMS The total jitter component with Gaussian distribution. After transforming a contiguous range of measured points into Q space and performing a linear regression, it is calculated as the mean of the sigmas of the two straight lines. The contiguous range is limited by the the BER Threshold and the Min. BER for RJ/DJ Separation threshold.
6 Advanced Analysis RJ/DJ values can be calculated in this case, they are not shown because confidence in the results is too low. Explanation of the Fast Total Jitter Measurement Results The Fast Total Jitter measurement provides both graphical and numerical results: Explanation of the Result Display The example below shows a copied result, and the display of measured points was enabled.
6 Advanced Analysis Peak-to-peak value of the total jitter. Calculated as the pulse period (unit interval) minus the Phase Margin at the Total Jitter BER Threshold. • Total Jitter Uncertainty The maximum of the uncertainties of both slopes. Measured as the time between a point with a BER greater than the specified Total Jitter BER Threshold and the next point with a BER less than the specified Total Jitter BER Threshold (left slope) and vice versa (right slope).
6 Advanced Analysis Three Available Views The Output Levels measurement provides three different graphical views to visualize the calculated results: • BER versus Threshold This graph shows the relationship between the analyzer decision threshold and the resulting BER. It presents the raw data. • dB Histogram versus Threshold This graph shows the relationship between the analyzer decision threshold and the derivative of the bit error rate (dBER/dTh).
6 Advanced Analysis Changes on the Pass/Fail, View, Graph, and Color tabs only affect the display of the results. There is no need to repeat the measurement. Variable Decision Threshold Method The method used by this measurement is commonly known as Variable Decision Threshold Method. It provides a "vertical" analysis of the eye opening seen by the receiver. This method allows you to determine more than just the actual levels.
6 Advanced Analysis 4 Use a shielded cable to connect the pattern generator's Data Out port and the error detector's Data In port. 5 Switch to the Pattern panel and press Pattern Select. Select an appropriate pattern for this test. We use a pure 2^23-1 PRBS segment. 6 For the pattern generator setup you need to specify the logic levels and the bit rate. Select ECL levels and a clock speed of 1250MHz in this example. This corresponds to a clock period of 0.8ns.
Advanced Analysis 6 In this example, we expect the signal voltages to be between -1.75V and -0.95V. The Sample Threshold values proposed above cover this range well. The Resolution is the distance between the measurement points when the threshold moves from the low to the high level. A resolution of 10mV results in 100 measured points per Volt. Note that we have disabled the Edge Resolution Optimization. 4 Press OK to close the Properties dialog box. 5 Press the Start button to execute the measurement.
6 Advanced Analysis How to Improve the Output Levels Display You can change the display of an existing measurement, for example, if you wish to see more details to investigate the graph. This can be done on the Graph tab of the Properties dialog box: 1 Press the Properties button. If you have a mouse connected to your Serial BERT, you can also click the right mouse button on the graph and select Properties from the context menu. 2 Switch to the Graph tab. 3 As an example, select Show Measured Points.
Advanced Analysis 6 How to Change the Output Levels Properties In the example measurement, we have set the focus on speed: 100 threshold levels and 1,000,000 compared bits per measuring point. You may wish to obtain more precise results. 1 Press Properties and switch to the Parameters tab. 2 Increase the Number of Compared Bits to 100,000,000. Remember: One failure per 1 million bits yields a BER resolution of 10-6. One failure per 100 million bits yields a BER resolution of 10-8.
6 Advanced Analysis This graph shows the absolute values of the derivative of the bit error rates over the thresholds (dBER/dTh). It visualizes the data that forms the basis for the calculations of the level and noise values. The graph provides a special marker that allows you to estimate the data distribution by approximating it by means of a Gaussian normal distribution. The Output Levels measurement provides a third graphical display: the Q from BER versus Threshold graph.
6 Advanced Analysis After this number of errors, the measurement stops for the current sample point and moves to the next one. This allows you to speed up the measurement. You can switch off this option if only the number of compared bits is important. NOTE The measurement moves to the next sample point when the first of the two criteria is reached. Set the criteria for the sample threshold: • Resolution Specifies the distance between the sampling points.
6 Advanced Analysis As you can see from the figures above, Edge Resolution Optimization does not sacrifice the precision of the measurement. But it can reduce the measurement duration considerably, especially if you compare a large number of bits at a low data rate. Pass/Fail Tab The Pass/Fail tab of the Properties dialog box allows you to specify the criteria to decide whether the DUT passes or fails the test. You can change pass/fail criteria without rerunning a test.
Advanced Analysis 6 In the following figure, you can see how errors are flagged.
6 Advanced Analysis If you compare the Low Level result with the limit we have set on the Pass/Fail tab, you will find that the measured result fails the upper pass/fail limit for this parameter. View Tab The graph shows either the BER vs. Threshold, the dBER vs. Threshold, or the QBER vs. Threshold. BER vs. Threshold Graph This graph shows the relationship between the analyzer decision threshold and the measured BER. The BER considers all errors. It is calculated as: dBER vs.
6 Advanced Analysis QBER vs. Threshold Graph This graph shows the extrapolation of the optimum Q-factor and the optimum threshold level from a limited number of measured points. The measured data points to be used for the calculation have to be within a contiguous BER range. This range is defined by specifying the Min BER for Q (lower threshold) and the BER Threshold (upper threshold). Both thresholds can be set in the lower section of the View tab.
6 Advanced Analysis This is the bit error rate threshold at which the Threshold Margin is determined. It is also the upper threshold for the Q-factor calculations. The BER Threshold is displayed in the BER vs. Threshold graph. There, it can be positioned with the mouse (or your finger, if you are working directly on the Serial BERT). • Min BER for Q This is the lower threshold for the Q-factor calculations. Table Number Format You can select the number of Decimal Places to be displayed in the table.
Advanced Analysis NOTE 6 If you have a mouse connected to your Serial BERT, you can access many parameters and display options conveniently from the context menu.
6 Advanced Analysis • Low Level The Low Level is the mean of the lower dBER/dTh distribution. It is calculated as: • Mean Level The Mean Level is the middle between the High and Low Levels, calculated as: • Amplitude The Amplitude is the difference between its High and Low Levels. • Threshold Margin The Threshold Margin is the distance between the upper and the lower BER curves at the position given by the BER Threshold setting.
Advanced Analysis 6 where Mean is the High Level of the terminal. • Low Level Std. Dev. The Low Level Standard Deviation is calculated as: where Mean is the Low Level of the terminal. • Peak Peak Noise The peak-to-peak Noise is calculated as: Note that the Threshold Margin depends on the position of the BER Threshold.
6 Advanced Analysis Table 29 Parameter Name Pass/Fail Q High Level Nr. Points Q High Level R^2 Q Low Level Q Low Level Std.Dev Q Low Level Nr. Points Q Low Level R^2 For some of these parameters, pass/fail limits can be set, as indicated in the table. The numerical Q-factor parameters are defined as follows: • Q Factor The Q-factor is calculated as: where μ1,0 is the mean level of the 1 and 0 rails, respectively, and σ1,0 is the standard deviation of the noise distribution on the 1 and 0 rails.
6 Advanced Analysis The Q High Level is the mean, calculated from the linear regression curve for the high level data: m= • -A B Q High Level Std.Dev The Q High Level Standard Deviation is the σ (Sigma), calculated from the linear regression curve for the high level data: s= • 1 B Q High Level Nr. Points This is the number of data points used for the calculation of the Q High Level value. It depends on the setting of the BER Threshold and also on the setting of the Min BER for Q parameter.
6 Advanced Analysis • Q Low Level Std.Dev The Q Low Level Standard Deviation is the σ (Sigma), calculated from the linear regression curve for the low level data: s= • 1 B Q Low Level Nr. Points This is the number of data points used for the calculation of the Q Low Level value. It depends on the setting of the BER Threshold and also on the setting of the Min BER for Q parameter. The minimum for calculating Q-factor values is two points. It is recommended to include more than 5 points.
Advanced Analysis 6 Mathematical Background Bit errors are caused by noise, and the Q-factor describes the signal-to-noise ratio at the decision circuit. It is possible to calculate the Q-factor from a limited number of measured BER vs. threshold data points. It is also possible to calculate expected bit error rates from the Q-factor. This is a method for predicting very low bit error rates (typically below 10-11) that would take a long time to measure.
6 Advanced Analysis This function, applied to the high level and low level data points, yields new threshold vs. value combinations. In the area of low BER (typically below 10-4), these new data pairs should fit to two straight lines, although a couple of assumptions and approximations have been made. To determine the gradient and offset of these lines, a linear regression is performed. This is illustrated in the figure below.
Advanced Analysis 6 This graph shows two straight lines. The intersection of these lines marks the Qfactor and the Q optimum Threshold. Mathematically, the standard deviation and mean values are calculated as: s= - m= 1 B -A B This calculation leads to the values of μ1,0 and σ1,0. Notes on the Q-Results You can specify the range of data points used for these calculations by means of the Min BER for Q and BER Threshold parameters (see also “View Tab” on page 274).
6 Advanced Analysis • The R2 values are excellent (greater than 0.95). • The mean levels and standard deviations returned by the Q-factor calculations differ from the measured values. The QBER vs. Threshold graph may look as shown below: If you increase the BER Threshold to include more points, you will find: • The mean levels and standard deviations returned by the Q-factor calculations approach the measured values. • The Q-factor decreases. • The R2 values deteriorate.
6 Advanced Analysis Eye Opening Eye Opening - Concepts The Eye Opening measurement generates a three-dimensional bit error rate (BER) diagram as a function of the sample delay and the sample threshold. With this measurement, the complete eye of the DUT output signal is measured. The results comprise the voltage and timing of the eye opening and the optimum sampling point. To get the result, the sampling delay and the input threshold of the signal are shifted within 1.
6 Advanced Analysis The contour plot shows discrete lines of equal bit error rate - just like the contour lines on a map. The color of a line indicates the respective BER value. This graph is useful to visualize in which areas the BER changes - a homogeneous BER field will give you no lines at all. • Pseudo Color Plot This plot visualizes the BER by a continuous color gradient. It uses different colors for the regions between the lines of equal BER.
Advanced Analysis 6 Example Results The following figure shows the graphical result of a typical Eye Opening measurement: Eye Opening - Procedures This section shows how to set up and perform an Eye Opening measurement. As an example we measure the eye diagram of a shielded cable.
6 Advanced Analysis • Optimizing the view of the results (see “How to Optimize the View of the Results” on page 291) • Using the color bar (see “How to Use the Color Bar” on page 292) • Adding or changing colors (see “How to Add or Change Colors” on page 292) • Changing the BER threshold (see “How to Change the BER Threshold” on page 293) • Changing the BER range of a color (see “How to Change the BER Range of a Color” on page 294) How to Prepare the Eye Opening Measurement To prepare an Eye Open
Advanced Analysis 6 How to Execute the Eye Opening Measurement To run the Eye Opening measurement: 1 Switch to the Analysis panel and then press the Eye Opening icon. 2 Press the Start button to execute the measurement. The measurement is run and the display is continually updated.
6 Advanced Analysis – View tab, Graph tab, and Color tab All settings on these tabs only affect the way the data is displayed. You do not need to run the measurement again. See “View Tab” on page 298, “Graph Tab” on page 300, and “How to Change the Colors of the Graph” on page 232 for details. 3 Press OK when you have made all required changes to close the Properties dialog box. How to Use the Color Bar The color bar at the right-hand side of the diagram shows the assignment of BER thresholds to colors.
6 Advanced Analysis Table 30 Option Description Add color... To add an additional color to the color gradient at the cursor position. The BER range for this new color will be assigned automatically. Update Plots Dynamically Rainbow To display a large variety of colors for the bit error ranges. Yellow-Blue To display a color gradient from color1 to color2 for the bit error ranges.
6 Advanced Analysis How to Change the BER Range of a Color The bit error ranges are set automatically, but you can change these areas. To do so: • Move the handles of a color with the mouse (or your finger, if you are working directly on the Serial BERT). This is particularly useful if you have set the display to show the Pseudo Color Plot. By default, you may see a graph like the one below: Now you may wish to know more precisely what happened between yellow and red.
6 Advanced Analysis Parameters Tab NOTE If you modify the parameters on this page, you have to rerun the measurement to update the results. Set the criteria for moving to the next sample point: • Number of Compared Bits After this number of compared bits, the measurement stops for the current sample point and moves to the next one. The default is 1 million bits. That means, you can measure a bit error rate down to 10-6 (one error per million).
6 Advanced Analysis • Low Level Specifies the lower limit of the measurement voltage range. Enter the value in mV. This value should be slightly lower than the lowest expected signal voltage. • High Level Specifies the upper limit of the measurement voltage range. Enter the value in mV. This value should be slightly higher than the highest expected signal voltage.
Advanced Analysis 6 In the tabular view, each of the calculated values will be marked with an icon if it failed the test. The following illustration shows an Eye Opening measurement that has failed the criteria for the eye opening voltage.
6 Advanced Analysis View Tab The graph shows either a Contour Plot, a Pseudo Color Plot, or only one curve for the selected bit error rate threshold. 298 Contour plot The contour plot shows discrete lines of equal bit error rate-just like the contour lines on a map. The color of a line indicates the respective BER value. This graph is useful to visualize in which areas the BER changes - a homogeneous BER field will give you no lines at all.
Advanced Analysis Equal BER at BER threshold 6 It shows the contour line at the BER threshold. Furthermore, the following parameters can be set on the View tab of the Properties dialog box: Analyze You can analyze for: • All Errors To display all errors. • Errors if 0s Expected To display the errors if "0" is expected, but "1" received. • Errors if 1s Expected To display the errors if "1" is expected, but "0" received.
6 Advanced Analysis and drop the horizontal BER threshold in the graphical display to change this value. Table Number Format You can select the number of Decimal Places to be displayed in the table. Graph Tab On the Graph tab, you can use the several options to optimize the graphical display according to your needs. Timing Units Choose between Unit Interval and Seconds to select the timebase for the display's x-axis. Markers To analyze the graphs at a particular point, you can use the markers.
6 Advanced Analysis Explanation of the Numerical Results Additionally to the graphical results, the measurement provides numerical results: Measurement Parameters The measurement parameters are defined in the following List: • Time Eye Opening This is the maximum extension of the BER threshold contour line in sample delay direction (eye width). NOTE This value is different from the horizontal extension of the BER threshold bounding box.
6 Advanced Analysis Understanding the Result Parameters All result parameters are calculated from the BER threshold contour line and its bounding box. So, all parameters change with the BER threshold. The result display of the Eye Opening measurement shows the maximum eye opening time, the maximum eye opening voltage, and the position of the optimum sampling point.
6 Advanced Analysis NOTE NOTE The Error Location Capture measurement is disabled in "8B/10B Comparison", "Bit Comparison without PCIe3 SKPOS" and "Bit Comparison without USB3.1 SKPOS" error ratio modes. The measurement run is aborted by various actions, like selecting a new pattern or starting synchronization or alignment.
6 Advanced Analysis • No other advanced measurement (Output Timing, Output Levels, etc.) can be running. • Error Location Capture can only run when the BER Location Mode is set to more than one bit (for example, all bits, or a block with a length > 1). Error Location Capture - Procedures This section shows how to set up and perform an Error Location Capture measurement. As an example first add a couple of errors to an alternating pattern and then capture their position.
Advanced Analysis 6 How to Demonstrate the Error Location Capture Measurement To demonstrate the Error Location Capture measurement: 1 Switch to the Pattern panel and press the Open button. Load a Fiber Channel random data pattern from the demo patterns (Demo > Fiber Channel > RPAT.ptrn). Press the Properties button and set the Pattern Type to Alternate. To display both halves of the alternate pattern, press the Alt pat view button. Ensure also that the error detector tracks the pattern generator.
6 Advanced Analysis Choose between a binary or hexadecimal display of the pattern. The current setting is shown in the status bar below the pattern window. Note that in hexadecimal view the captured error can only be located as being one of a group of four bits. To display which of the four bits is the errored bit, you have to switch to the binary display. • Exp. / Cap. (Expected/Captured) Toggle the data view between the display of the expected data, i.e.
6 Advanced Analysis • Length: This field indicates the length of the captured data. Note that the value here does not equal the length of the pattern. Compare Pattern File The captured data is saved as an alternating pattern: • Pattern A contains the expected data. • Pattern B contains the errored data: 0s if the expected bits were also received, 1s for errored bits. To calculate the captured pattern, XOR the bits from pattern A with the bits from pattern B.
6 Advanced Analysis This is achieved by measuring the bit error rate at a limited number of test points. Up to 32 measurement points can be specified, each defined by a sampling time relative to the actual sampling point (which can be the optimum sampling point) and a threshold voltage (which is adaptive). In practice, six measurement points will often suffice to approximate the shape of the eye. Six measurement points are preset by default.
6 Advanced Analysis Instead of UI (one Unit Interval is equal to one system clock period), the relative time can also be specified in seconds. • The Voltage is the decision threshold voltage at this measurement point. The voltages of the measurement points can be set as absolute voltages, as offset voltages, or as percentages. This is done on the Parameters page of the Properties dialog.
6 Advanced Analysis • Optimizing the View of the Results. (See “How to Optimize the View of the Results” on page 311 ). How to Prepare the Fast Eye Mask Measurement To prepare a Fast Eye Mask measurement to test a shielded cable: 1 Disable the pattern generator outputs by pressing the 0V (Disable) button in the PG Setup -> Data Output screen. 2 Use a shielded cable to connect the pattern generator's Data Out port and the error detector's Data In port.
6 Advanced Analysis The measurement is run and the result window shows the bit error rates measured at six measurement points. The Relative Time refers to the current sampling point. The Voltages are the decision threshold voltages for measuring the bit error rate at this measurement point. The voltages of the measurement points can be set as absolute voltages, as offset voltages, or as percentages.
6 Advanced Analysis Parameters Tab Note that if you modify the parameters on this page, you have to rerun the measurement to update the results. Set the criteria for moving to the next sample point: • Number of Compared Bits After this number of compared bits, the measurement stops for the current sample point and moves to the next one. The default is 1 million bits. That means, you can measure a bit error rate down to 10-6 (one error per million).
Advanced Analysis 6 Depending on the quality and characteristics of the eye opening, the resulting 50% threshold may deviate from the decision threshold defined in the loaded setting. • Number of Valid Points Change the Number of Valid Points, if you wish to use less or more measurement points for the measurement. • Relative Time and Voltage In the table at the bottom of the Parameters tab you can change measurement points. Enter the Relative Time and Voltage according to the above settings.
6 Advanced Analysis The BER Threshold will usually be 0, since you want the measurement to fail if a single bit was received in error. NOTE The pass/fail threshold applies to all measurement points. An icon indicates all measurements where the bit error rate is higher than this threshold, as shown in the figure below: View Tab The following parameters can be set on the View tab of the Properties dialog box: Analyze You can analyze for: • All Errors To display all errors.
6 Advanced Analysis Table Number Format You can select the number of Decimal Places to be displayed in the table. Eye Diagram Eye Diagram - Concepts The Serial BERT provides quick design analysis with the Eye Diagram capability. The Eye Diagram allows a quick check for the DUT’s signal output, and determines the signal quality. Due to the higher sampling depth of a BERT value, the eye contour lines display the measured eye at a deeper BER level, for accurate results.
6 Advanced Analysis Methods of Representations There are two methods of representing the eye diagram: • Waveform • Contour Waveform Waveform is the shape, and form of a signal. The waveform graph shows the periodical variation of voltage against time. The waveform in the Serial BERT is similar to the one in the oscilloscope. In this case, the waveform initially gives a coarse, but quick picture of the signal quality; while the ‘smooth waveform’ quickly generates a high resolution graph.
Advanced Analysis 6 The BER Threshold is configurable. The BER Threshold is the level up to which the signal is represented as waveform, and BER values below this threshold are represented as contours. BER Contour Contour is a curve connecting points where the BER has a same particular value. The contour graph is plotted within the Eye Diagram, and it helps to determine the Eye Opening at deep bit error rates, such as 1e-10, 1e-12, and so forth.
6 Advanced Analysis italics. A BER value for which a Contour does not exist is struck across. You can select multiple BER values. The lower BER values are in red and pink, while the deep BER values are in green and blue. The screen shot below displays the contour and the legend. The outer contour represents the measured BER contour, while the inner lines are extrapolated.
Advanced Analysis NOTE 6 The mask test is enabled when "center" is set to "middle of eye", and "persistence" is set to "infinite" in the Parameter Tab. Automated Eye Parameter Measurement Automated Eye Parameter Measurement characterizes an Eye Diagram by measuring the rise time, the fall time, the eye amplitude, and so forth. These measurements are called Automated Eye Parameter Measurement, they display the Eye behavior, and they depend on the parameters set in the View Tab.
6 Advanced Analysis NOTE The scroll bar at the bottom of the table shows the entire table and its contents. For more details refer to “Eye Diagram - Reference” on page 332 Eye Diagram - Procedures This section lists out some of the generic processes to set up measurements, to run them, and to create Eye Diagrams.
6 Advanced Analysis 7 Set up the error detector to match the input range and the termination with the pattern generator’s levels: – Select an Input Range from –2 V to 0 V – Set the Data Termination to –2 V – Set the Alignment BER Threshold to 1E–6 – Set the Clock Setup to Clock Data Recovery to get the error detector’s clock from the incoming data stream. 8 Enable the pattern generator outputs by pressing the 0V (Disable) button.
6 Advanced Analysis How to Change the Default Settings of an Eye Diagram? To achieve desired results, you can change the measurement parameters: 1 Click the Properties button to open the Properties dialog box. 2 Use the different tabs in this dialog box to make the required settings: – Parameters tab The Parameter settings used for the measurement can be changed only when the measurement is not running, and if the settings are changed, then you will have to re-run the measurement.
Advanced Analysis 6 The view tab options set the parameters to measure, and choose units for the displayed results in the numerical table. (Refer to “View” on page 333 for the detailed options) – Graph tab The graph settings give the display parameters like timing units, waveform, contour, grid and markers.
6 Advanced Analysis – Eye Measurements tab The Eye Measurements tab lists out all the measurements. The user can select specific results.
Advanced Analysis 6 How to Set Markers? Markers help in tracking the BER at specific points, so to set the markers: 1 Click on the Graph tab 2 Select Show Maker Lines 3 Apply Setting This enables the markers, and a marker readout button appears on the top left corner. You also have an option to Reset Markers to their original position.
6 Advanced Analysis The Marker reading is shown on the left side of the graph. 4 Click the marker readout button on the top left corner to see the detailed marker readout. How to Run the Mask Measurement? To run the Mask Measurement, the Eye Diagram measurement must be running. 1 Click on the Mask tab 2 Click Load button 3 Select a Mask File. 4 Select a Mask Alignment.
6 Advanced Analysis 5 Start This starts the Mask Measurement.You also have options for setting Mask Margins, and Mask Scaling. 6 Click the mask readout button on the bottom left of the screen to either see, or close the detailed mask readout. How to Customize Mask Files A Mask file can be modified as per requirements. Following are the sections in a mask file: Mask File Identifier The mask file identifier is the first line in a mask file.
6 Advanced Analysis /* Region Number */ 2 Region Type You can define three types of regions for each region number you have specified: • STD defines the actual mask region. • MARGIN_MAX defines the maximum margin area when test margins are set to 100%. • MARGIN_MIN defines the minimum margin area when test margins are set to -100%.
Advanced Analysis 6 :MTES:INP:MODE DIFF In the above example, the Y1 and Y2 values define the mask with an absolute voltage value. In some mask files, Y1 and Y2 values are not specified, and so the voltage value is relative to either Vtop and Vbase, or to the eye boundaries specified by the user in the GUI. A mask can be set in two modes, either differential, or single. The above example, ":MTES:INP:MODE DIFF", sets the mode of the mask file to differential.
6 Advanced Analysis • Property Page • Legend States • Left Panel Values • Automated Eye Parameter Measurement 1 Click on the Analysis Menu 2 Click on ‘Save Measurement’ To Recall the measurements: 1 Click on the Analysis Menu 2 Click on ‘Load Measurement’ How to Export Fetched Data? You can export various counters to an exported file location.
Advanced Analysis 6 When you click on ‘Export Data’ the following screen appears: The Clipboard option saves the data on a clipboard. The ‘Format Options’ takes you to the following screen. You can select the ‘Format’, ‘Save to File’ and ‘Error Options’ as per requirements. The ‘Save to File’ option specifies, either saving only the BER values, or saving the BER, the Compared Bits, and the Errors.
6 Advanced Analysis The ‘Error’ option gives the user the choice of saving, the BER, the Compared Bits, and the Errors, for different combinations of expected 1s, 0s, and all. Eye Diagram - Reference The Eye Diagram measurement returns the results in graphical and numerical forms. The following section explains the measured parameters, and the displayed options that are specific to these measurements. Additionally, some information is provided to explain the theoretical background to these measurements.
6 Advanced Analysis With ‘middle of eye’, the center of the eye coincides with the center of the screen. • Transition With ‘transition’, the transition of the eye coincides with the center of the screen. Number of eyes Set the criteria for the number of eyes. • 1.5ui This option displays ‘1.5’ eyes on the graph. The transitions of the complete eye are placed such, that, they show ‘0.25’ eyes on both sides. So, the total number of displayed eyes is 1.5. • 2ui This option displays ‘2.
6 Advanced Analysis Marks the lower edge of the eye window boundary • Right Boundary Marks the higher edge of the eye window boundary Eye Width/Height Set the criteria for eye width and eye height. • Eye width at crossing point This parameter gives the eye width at the crossover. Total is 100% of the display screen. • Eye width custom This parameter gives the eye width at a specific percent of the eye amplitude. (Starting from low level to high level).
Advanced Analysis 6 Ratio of two optical power levels, low level and high level • dB Unit for measuring the extinction ratio. ratio[dB]=10*log(ratio) • % Extinction Ratio in percentage Graph Timing Units Set the criteria for the Timing Units: • Unit Interval Unit used to measure delay relative to the eye width. • Seconds Unit used to measure eye width in absolute terms of time. Waveform Set the criteria for the waveform: • Show Waveform This option shows the waveform graphic.
6 Advanced Analysis Markers Set criteria for the markers: • Show Marker Lines This option enables the markers on the screen. • Reset marker This brings the marker back to its default position. Mask Mask Selection • Load Mask File With Load, a mask file is loaded. Once the file is successfully loaded the Start button gets activated. Mask Alignment Set Mask Alignment: • Display The Display mode aligns the mask using the Vtop and Vbase values of the displayed waveform.
6 Advanced Analysis The Mask Run Control aligns the mask, switches it on, starts displaying the number of waveforms, and the number of mask violations per mask region, and the worst BER per Mask region. Exit stops the computation, hides the mask, and the violations. Automated Eye Parameter Measurement Automated Eye Parameter Measurement Table 31 S.No Parameter Description 1 Rise time Measures the mean transition-time taken by the data on the rising edge of the eye diagram.
6 Advanced Analysis Table 31 338 S.No Parameter Description 4 0-Level Zero level is the measurement of the mean value of the logical 0 of an eye diagram. Note: This measurement is made in a section of the eye referred to, as the eye window boundaries. The default value for the NRZ eye window is the central 20% of the bit period. You can change the eye window boundary settings in the Configure Measurement dialog box. These settings determine what portion of the eye will be measured.
Advanced Analysis 6 Table 31 S.No Parameter Description 7 Eye Width The horizontal measurement of the eye opening at a specified BER Threshold. Note: The Eye Width is calculated according to the BER Threshold set in the View Tab. Eye Width is calculated as per the selection of either Width at Crossing Point or Custom defined Width. 8 Eye Height Measures the vertical opening of an eye diagram with respect to BER threshold. This determines “eye closure” due to noise.
6 Advanced Analysis Table 31 340 S.No Parameter Description 10 Jitter RMS A standard deviation of the crossing point histogram with respect to BER threshold. Note: The Jitter RMS is calculated according to the BER Threshold set in the BERT from Calculate Measurement Parameters of the View Tab. 11 Cross Voltage Crossing percentage is a measure of the amplitude of the crossing points relative to the 1 level and 0 level.
Advanced Analysis 6 Table 31 S.No Parameter Description 14 OMA The Optical Modulation Amplitude (OMA) is the difference between the 1level power and 0-level power. Note: The OMA is calculated using the Dark Level and Conversion Gain of the View Tab. 15 Duty Cycle Distortion (DCD) This value is the difference between the period of a 1 bit and a 0 bit. 16 Average Power The mean of level 0 power and the level 1 power.
6 Advanced Analysis Eye Diagram Screen This graph gives the picture of the Eye Diagram Screen. Spectral Jitter Spectral Jitter - Concepts The Spectral Jitter measurement allows you to analyze the jitter inherent in the output signals of your device under test (DUT) as a function of the frequency. This measurement can be used for investigation of the behavior of the DUT, for example to identify crosstalk effects.
6 Advanced Analysis The Spectral Jitter measurement detects even small periodic components that may be hidden in a high level of random noise. It informs you about the frequencies of such components and measures their contribution to the total jitter. This helps to identify jitter sources and to reduce or eliminate their influence. Prerequisites for Using the Spectral Jitter Measurement It is important that the initial sampling point is in optimum position (i.e.
6 Advanced Analysis These records are automatically processed. The error information is subject to a fast Fourier transform (FFT). FFT reveals the spectral components and their power. Several window algorithms are provided to reduce the influence of leakage. Jitter Distribution Over Time Jitter has a more or less characteristic distribution over time.
6 Advanced Analysis Jitter caused by a triangle signal shows an even distribution: Last, but not least, the histogram of jitter modulated by a sinusoid exhibits two significant peaks: The jitter histogram of a real world signal shows most often a mixture of these characteristic distributions. Measurement Results The Spectral Jitter measurement provides: • A graphical view of power vs. frequency. This makes it easy to identify prominent spectral jitter components.
6 Advanced Analysis • You can switch between linear and logarithmic scales. • Graphical markers and the zoom function assist you when you are analyzing the graph visually. • The numerical results include bit error rate, total power, and noise power. They provide also frequency and power information about the dominant peaks in the spectrum. • Absolute and relative power values are available. Relative values can be normalized to the total jitter power or the power of a selected tone (frequency bin).
Advanced Analysis Signal without jitter 6 1 0 Expected and correctly sampled data 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 Periodical, sinusoid jitter increasing right shift Signal with jitter decreasing right shift increasing left shift decreasing left shift 1 0 Captured data 1 0 1 0 1 0 1 0 1 1 0 1 0 1 0 1 0 1 Expected data 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 Error data 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 The jitter source m
6 Advanced Analysis NOTE The Spectral Jitter measurement should only be used in conjunction with data that has an equal distribution of ones and zeros over time. Otherwise, the results are hard to predict and may be not reproducible. Signal Processing If the error signal is obtained as explained above, an analysis in the frequency domain reveals the absense or presence of deterministic jitter. Dominant frequency components become visible and their contribution to the total jitter can be measured.
Advanced Analysis 6 The fundamental frequency and its harmonics appear. Such spectra have been measured with the Spectral Jitter measurement. A logarithmic power scale shows the details: When the repetition period of the characteristic pattern in the error record increases, you will also find the typical sine-x-over-x decay of the spectral power. Leakage and Windowing FFT assumes that the time record contains a representative section of an endless periodic signal.
6 Advanced Analysis Leakage makes it impossible to detect minor adjacent spectral components. The following two figures refer to a slightly disturbed sine wave. When you perform a Spectral Jitter measurement, it is likely that some degree of leakage occurs. The measurement therefore provides a choice of FFT windows that allow you to detect leakage and to reduce its impact. An FFT window is a filter that sets the beginning and end of the time record smoothly to zero.
6 Advanced Analysis • Every window reduces the spectral power. • Results obtained from different devices can only be compared if the same window is used. Spectral Jitter - Procedures This section shows how to set up and use the Spectral Jitter measurement. As an example, we measure the spectral jitter behavior of a shielded cable.
6 Advanced Analysis 9 Press Sync Now and then Auto Align to find the optimum sampling point. Check that the synchronization and the alignment were successful. None of the error indicators should show red, and the resulting BER should be zero. How to Execute the Spectral Jitter Measurement To run the Spectral Jitter measurement: 1 Switch to the Analysis area. If the Spectral Jitter screen is not yet displayed, press the Spectral Jitter icon. 2 Press the Start button to execute the measurement.
6 Advanced Analysis These settings are used for data collection. Changes require to run the test again. See “Parameters Tab” on page 252 for details. – Pass/Fail tab These settings determine whether the calculated results are recognized as passed or failed. However, a new test run is not required when doing changes here. See “Pass/Fail Tab” on page 254 for details. – View tab, Graph tab, and Color tab All settings on these tabs only affect the way the data is displayed.
6 Advanced Analysis data rate / Acquisition Depth For example: If you have a data rate of 2.5 GHz and an Acquisition Depth of 128 Kbit, the frequency resolution is 19.0735 kHz. The relations are illustrated in the following figure: NOTE A high Acquisition Depth requires a high degree of computational effort and hence time. • Sample Point Offset By default, the sampling point for the measurement is positioned 0.5 clock periods or UI ahead of the present analyzer sampling point.
6 Advanced Analysis The other windows have the shapes illustrated in the figure below: You can see from the figure that the Blackman window is the strongest filter. The windows are based on the following formulas: NOTE In case of leakage, FFT windows improve the spectral resolution. FFT windows generally reduce the measured spectral power. For an introduction to FFT windows see “Leakage and Windowing” on page 349.
6 Advanced Analysis NOTE The pass/fail criteria do not control measurement execution. The measurement run will be completed even if the measurement fails for one or more of the criteria. The following Spectral Jitter Pass/Fail criteria can be selected for the pass or fail decision: • Bit Error Rate • Total Power (in dB) • Noise Power (in dB) You can also define the pass/fail criteria for each of the defined frequency ranges.
Advanced Analysis 6 View Tab The graph shows the noise versus frequency. The View tab provides you with various options for analyzing the data and setting frequency ranges for evaluation. Power scale The linear power scale is well suited for identifying large peaks in the spectrum. The View tab allows you to change the power scale from linear to dB. The dB scale is a logarithmic scale. The dB scale allows you to examine the whole power range.
6 Advanced Analysis Absolute vs. relative values The figure above shows the absolute power scale and values as calculated by the FFT. You can also calculate relative values. Relative values can be normalized to either the total power (True Relative) or to the power of a selectable frequency bin (Relative). The following figure shows an example where the scale and the power values have been normalized to the power measured at 1 MHz.
6 Advanced Analysis Frequency Scale Power Unit Frequency Axis Range Markers Choose between logarithmic and linear scale for displaying the frequencies. Display the power either in a logarithmic (dB) or linear scale. Show the entire frequency range or zoom in one part of it. To analyze the graphs at a particular point, you can use the markers. Additionally, you can display all related values for the markers in the marker readout.
6 Advanced Analysis 360 Agilent J-BERT N4903B High-Performance Serial BERT
S Agilent J-BERT N4903B High-Performance Serial BERT User Guide 7 Evaluating Results Evaluating Results - Concepts The Serial BERT offers several different kinds of tests that can be run: • Instantaneous Measurements This type of measurements is used to monitor the instantaneous "BER" / "8B/ 10B Comparison" / "Bit Comparison without PCIe3 SKPOS" / "Bit Comparison without USB3.1 SKPOS" during measurement setup, experiments, and adjustments.
7 Evaluating Results • Frame Error Ratio (FER) • Calculated Bit Error Ratio (cBER) • Filler Symbol Ratio (FSR) • Disparity Error Ratio (DER) • Illegal Symbol Ratio (ISR) In order to stress your device under test, you can then, for example, add errors to the data stream or switch between different patterns and view the resulting "BER" / "8B/10B Comparison" / "Bit Comparison without PCIe3 SKPOS" / "Bit Comparison without USB3.1 SKPOS".
Evaluating Results 7 This can be used to see how changes to your device affect the "BER" / "8B/ 10B Comparison" / "Bit Comparison without PCIe3 SKPOS" / "Bit Comparison without USB3.1 SKPOS". You can determine what adjustments improve or degrade the "BER" / "8B/10B Comparison" / "Bit Comparison without PCIe3 SKPOS" / "Bit Comparison without USB3.1 SKPOS". Instantaneous Measurements - Procedures This section describes how to run, monitor, and analyze instantaneous measurements.
7 Evaluating Results How to Monitor BER Results There are different ways, how you can monitor instantaneous BER values: 1 Listen to the BER warning tones. See “Audio Signals - Procedures” on page 225 for details. 2 View the BER Results window, which is described in “BER Results Window” on page 366. 3 View the BER bar and errors indicator. See “Status Indicators” on page 372 for details.
7 Evaluating Results 3 View the Bit Comparison without USB3.1 SKPOS bar and errors indicator. See “Status Indicators” on page 372 for details. How to Analyze Instantaneous Measurements Results To analyze the behavior of your device, you can do the following modifications to the test setup and monitor the effect on the resulting "BER" / "8B/10B Comparison" / "Bit Comparison without PCIe3 SKPOS" / "Bit Comparison without USB3.1 SKPOS": 1 Add errors to the data stream.
7 Evaluating Results Instantaneous Measurements - Reference This section describes all elements of the BER Results window and 8B/10B Comparison Results windows and their various status indicators: BER Results Window This window displays the actual (BER) or cumulative (AccumBER) BER results: • The BER is the current BER, calculated upon a period of 200ms.
7 Evaluating Results 8B/10B Comparison Results Window This window displays the actual or cumulative results of the following measurements: • Symbol Error Ratio (SER). See “SER Results Window” on page 367. • Frame Error Ratio (FER). See “FER Results Window” on page 368. • Calculated Bit Error Ratio (cBER). See “cBER Results Window” on page 368. • Filler Symbol Ratio (FSR). See “FSR Results Window” on page 369. • Disparity Error Ratio (DER). See “DER Results Window” on page 370.
7 Evaluating Results Accumulated/Actual Button Error Count/Accum Error Count Click this button to toggle between SER or AccumSER results. This area displays the actual symbol frame error count or cumulative error count. FER Results Window This window displays the actual (FER) or cumulative (AccumFER) FER results: • The FER is the current FER, calculated upon a period of 200ms.
7 Evaluating Results You can click Accumulated/Actual button to toggle between cBER or AccumcBER results. The Error Count/Accum Error Count area displays the actual error count or cumulative error count.
7 Evaluating Results • The FSR is the current FSR, calculated upon a period of 200ms. • The AccumFSR shows either the accumulated FSR of the current accumulation, or if no accumulation is running, the results of the most recent accumulation. This enables you to monitor real-time FSR behavior as you do things such as to manually adjust the sampling point, to add errors, or to make adjustments to your device.
7 Evaluating Results • The DER is the current DER, calculated upon a period of 200ms. • The AccumDER shows either the accumulated DER of the current accumulation, or if no accumulation is running, the results of the most recent accumulation. This enables you to monitor real-time DER behavior as you do things such as to manually adjust the sampling point, to add errors, or to make adjustments to your device.
7 Evaluating Results • The ISR is the current ISR, calculated upon a period of 200ms. • The AccumISR shows either the accumulated ISR of the current accumulation, or if no accumulation is running, the results of the most recent accumulation. This enables you to monitor real-time ISR behavior as you do things such as to manually adjust the sampling point, to add errors, or to make adjustments to your device.
Evaluating Results 7 You can drag the yellow alarm threshold mark to change the SER Alarm Threshold. FER bar The FER bar displays the FER calculated upon a period of 200 ms. You can drag the yellow alarm threshold mark to change the FER Alarm Threshold. cBER bar The cBER bar displays the cBER calculated upon a period of 200 ms. You can drag the yellow alarm threshold mark to change the cBER Alarm Threshold. FSR bar The FSR bar displays the FSR calculated upon a period of 200 ms.
7 Evaluating Results NOTE The two additional error indicators "Symb Lock" and "8b10b Error"will only appear if you select the error ratio as 8B/10B Symbol Comparison from the Error Detector - Error Ratio's window. For more details, see “Symbol Lock Indicator” on page 192 and “8b10b Error Indicator” on page 192. For troubleshooting information, see “Setup Problems - Concepts” on page 469.
7 Evaluating Results • No new errors: The triangle icon on the button will remain yellow. If an error occurs, the detailed error message is displayed for a few seconds on the top of the status indicator area. You can click the message to make it disappear immediately. Click the Show Error Messages button to open an Error Message dialog box with all error messages. On this dialog, click Clear & Close button to clear the error messages list and quit the dialog box.
7 Evaluating Results Clock Rates The PG and ED measured bit rates are displayed here for your reference. Accumulated Measurements Accumulated Measurements - Concepts Accumulation refers to collecting measurement data over a specific period. This can be used to create test scenarios that are reproducible and comparable. Also, you can let tests run over long times and then evaluate the results afterwards.
7 Evaluating Results NOTE During measurement logging, the Serial BERT logs data in ten-second intervals. Your log file may be missing up to the last ten seconds of data. To avoid this condition, accumulate for 10 seconds longer than desired.
7 Evaluating Results How to View the Results You can view data from accumulated measurements in several ways: 1 Click the Accumulated Results menu item from the Results submenu to view the test results. For easy comparison, the results of the current and previous accumulation are listed here. See “Accumulated Measurements - Reference” on page 378 for details on the reported values. 2 Examine the measurement log files on an external PC.
Evaluating Results 7 Ratios Graph This graph displays the delta errored 1's ratio, delta errored 0's ratio, and total delta error ratio at data points over the entire accumulation period. The error ratios on the y-axis are set to a range of 1E+0 (100 % errors) to 1E-12. The accumulation period is on the x-axis. Display Change During accumulation, data will appear to move from left to right on the ratios graph. When the graph is completely filling the display, the x-axis time scale will double.
7 Evaluating Results NOTE NOTE The zooming function cannot be used before the accumulation period has ended. To view precise BER data for every point in time, view the measurement log file. While zoomed in, you can move data left or right by dragging the graph or by using the front panel knob. Sizing Handles You can also zoom vertically by dragging the top and bottom sizing handles. The current view can be moved vertically if you click between the handles and drag them up or down.
7 Evaluating Results • Errored 0's Ratio The ratio of the errored 0's count to the bit count is displayed here to 4 significant digits. • Errored 0's Count The number of logic 0's measured as logic 1's during the accumulation period is displayed here. • Auto Resync Counter The total number of Auto Re-Sync Count received in a time interval is displayed here. G.821 Measurements • Availability The ratio of the available seconds to the total gating period expressed as a percentage.
7 Evaluating Results Interval Results • Error Free Seconds The number of seconds in which no errors were measured is displayed here. • Error Free Deciseconds The number of deciseconds in which no errors were measured is displayed here. • Error Free Centiseconds The number of centiseconds in which no errors were measured is displayed here. • Error Free Milliseconds The number of milliseconds in which no errors were measured is displayed here.
Evaluating Results • 7 Accumulation Period For single and repeat accumulation modes, the accumulation period is displayed here. The period will be displayed in elapsed seconds, error count, or bit count, depending on the selected method. Burst Results The recommended application for burst mode is to measure multiple data bursts in quick succession. In this way, you can accumulate data for all bursts together.
7 Evaluating Results The status is unknown. This can occur if accumulation has not been started, or if Burst Sync mode has not been activated. • Burst Sync Ratio Percentage of time in burst (while the Gate In signal is low) that patterns are in sync and error counting is active. A higher percentage is desirable, because it indicates that more data is measured during each burst. • Total Burst Count Total number of bursts received while in burst mode during the accumulation period.
7 Evaluating Results Ratio Graph This graph displays the delta symbol errored 1's ratio, delta symbol errored 0's ratio, and total delta symbol error ratio at data points over the entire accumulation period. For more information, see “Ratios Graph” on page 379. Accumulated Results In the 8B/10B Comparison Accumulated Results table, the following values are listed: NOTE If the Block Length is equal to zero, then use Frame Count instead of Pattern Count for calculating Accumulated Results.
7 Evaluating Results • Compared Symbol Count (Symbol counter) The number of compared symbols considered for the accumulation period is displayed here. This may not be equal to the total number of symbols sent from the Pattern Generator. Compared Symbol count excludes any filler symbols (i.e. K28.1,K28.1 for USB 3.0) and considers block length in Block Mode or Pattern length in All Bits mode during comparison with Expected Data.
7 Evaluating Results • Illegal Symbol Count (Illegal symbol counter) The total number of illegal symbol count received in a time interval is displayed here. • Disparity Error Ratio (DER) The ratio of the number of illegal disparity change count to the number of symbols received in the current (or last completed) accumulation period, specified by the gate period is displayed here.
7 Evaluating Results The raw number of received symbols by the error detector equals the sum of filler primitives (which are being dropped before comparison) and the actually expected symbols. The actually expected symbols may be further distinguished into compared symbols and not compared symbols when using block mode. The number of filler primitives is a key indicator for DUT behavior. This number must also visible if there is NO pattern lock.
7 Evaluating Results one single frame error. In contrast to this, the Serial BERT loses pattern synchronization and must re-synch before the test can be continued. Currently, the detection mechanism for loss of synchronization is a high error ratio. This causes the Serial BERT to show many errors in the "Accumulated Results" window though there was only one single error. The serial BERT shows with such DUTs high error number, the number is not accurate.
7 Evaluating Results Ratio Graph This graph displays the delta symbol errored 1's ratio, delta symbol errored 0's ratio, and total delta symbol error ratio at data points over the entire accumulation period. For more information, see “Ratios Graph” on page 379.
7 Evaluating Results Accumulated Measurements for Bit Comparison without USB 3.1 SKPOS The Accumulated Results window for Bit Comparison without USB 3.1 SKPOS contains a graph and two tables: NOTE The Accumulated Results window for "Bit Comparison without USB 3.1 SKPOS" does not contain tables for G.821Measurement, Internal Results and Burst Results.
7 Evaluating Results Accumulation Parameters for Bit Comparison without USB 3.1 SKPOS For detailed information on "Bit Comparison without USB3.1 SKPOS" accumulation parameters, see “Accumulation Parameters” on page 382. Eye Measurements Eye Measurements - Concepts The purpose of eye measurements is to measure the eye height and width of the incoming data signal at specific alignment BER thresholds. This information is displayed on a representation of an eye diagram.
Evaluating Results 7 Eye Measurements - Reference Eye results are displayed for the Current Align and Previous Align process. They are also graphically displayed on a representation of an eye diagram. Eye Width and Height Eye width is the inside horizontal opening of the data eye in an eye diagram. This measurement is made by the error detector after an Auto Align or Data Center. The eye width at the current 0/1 decision threshold is displayed on a simulated eye diagram.
7 Evaluating Results NOTE Most oscilloscopes measure eyes differently, using statistical data. Therefore, the eye results of a Serial BERT and an oscilloscope may be different. The table in the top half of the Eye Results window additionally lists the following values: • Eye Voltage Center One of two components of the sampling point, the 0/1 decision threshold (eye voltage center) is displayed here.
S Agilent J-BERT N4903B High-Performance Serial BERT User Guide 8 Jitter Tolerance Tests Jitter Tolerance Tests - Concepts NOTE This chapter refers only to Serial BERT instruments on which the calibrated and integrated jitter injection option J10 is installed. A jitter tolerance measurement is used to determine the ability of a device or system to maintain communication quality in the presence of jitter.
8 Jitter Tolerance Tests • Intersymbol interference • Level noise This refers particularly to the performance of phase-locked loops (PLLs) or clock data recovery circuits (CDR). Jitter tolerance can be measured by applying a distorted data signal to the DUT and measuring the resulting bit error ratio. To make jitter tolerance tests reproducible, the signal distortion must also be reproducible. This requires some definitions.
8 Jitter Tolerance Tests The Serial BERT provides the Interference Channel option J20 to simulate level variations by adding sinusoidal interference. For more information see • “Crosstalk” on page 400 PLL Performance Test Testing a PLL (or CDR) is not complete until the entire frequency range of the PLL has been checked under worst-case conditions. For this purpose, the Serial BERT provides the Jitter Tolerance Characterization measurement.
8 Jitter Tolerance Tests Bounded Uncorrelated Jitter Similar to random jitter, bounded uncorrelated jitter (BUJ) has also a Gaussian distribution, but this distribution is cut (bounded) at both sides. This kind of jitter can be caused, for example, by crosstalk on a parallel bus or by intersymbol interference of random or long pseudo random binary bit sequences. To simulate bounded uncorrelated jitter for jitter tolerance tests, it can be generated from a filtered (frequency-limited) PRBS.
8 Jitter Tolerance Tests The jitter amplitude is usually specified in UI. One UI (Unit Interval) is always the reciprocal of the present data rate. For example, if the data rate is set to 2 Gbit/s, one UI corresponds to 500ps. Sinusoidal Jitter This kind of jitter can be picked up from any adjacent signal or clock. The jitter histogram has a U-shape. Triangular Jitter Triangular jitter is always generated by a spread spectrum clock but can also be caused by other periodic sources.
8 Jitter Tolerance Tests Intersymbol interference can be simulated by inserting a defined transmission path between the pattern generator and the DUT. Crosstalk Crosstalk between adjacent signal paths modulates the vertical eye amplitude. This narrows the voltage range of the receiver needed for capturing data correctly. For jitter tolerance tests, amplitude modulation is most often simulated by modulating the data output with a sinewave signal. This is called Sinusoidal Interference (SI).
8 Jitter Tolerance Tests Edge, level, and shape variations impact width, height, and symmetry of the eye opening. An example is shown below: How the N4903 Generates Jitter An Serial BERT on which the calibrated and integrated jitter injection option J10 and the interference channel option J20 are installed, combines all the necessary jitter sources in one instrument.
8 Jitter Tolerance Tests and the RJ sources, i.e. BUJ and RJ are mutually exclusive with sRJ. In addition, an external jitter signal can be applied at the Delay Ctrl Input. Note that all these sources can be individually switched on or off. Clock modulator This is a special modulator that modulates the generated clock with a lowfrequency signal. The resulting jitter superimposes all other jitter components.
8 Jitter Tolerance Tests The Interference Channel provides a switchable signal path with Nelco 4000-6 FR4 characteristics. This path can be connected between the pattern generator's Data Out port and the DUT. For instructions refer to “Connecting the J20 Option” on page 537. The Interference Channel provides also a sinewave generator and a resistive combiner. This makes it possible to add sinusoidal interference (amplitude modulation) to the data signal.
8 Jitter Tolerance Tests If P1 is used as input, sinusoidal interference is (or can be) added after the signal has passed the chosen trace. Seen from the DUT, this is called "near end" injection. If P2 is used as input, sinusoidal interference is (or can be) added at the beginning of the trace. This is called "far end" injection. Jitter Setup Jitter Setup - Concepts The Jitter Setup function is used for composing the total jitter in a defined and calibrated way.
8 Jitter Tolerance Tests The delay lines show markers which indicate the used and the remaining jitter amplitude in UI. Colored bars allow you to judge the impact of each jitter component on the total jitter. UI is the acronym for Unit Interval. One UI is the present clock period. Sinusoidal Jitter is not generated by a delay line but by modulating the clock. Sinusoidal Jitter and Spread Spectrum Clock cannot be used together with residual Spread Spectrum Clock because they share the same modulator.
8 Jitter Tolerance Tests • Random Jitter • Spectrally Distributed Random Jitter • External Jitter Source (Connected to the Delay Ctrl input) It depends on the jitter configuration which seven out of these nine jitter sources are displayed. Each jitter type can be separately turned on or off by clicking the LED button on the left side. The corresponding jitter source is enabled if the LED is on. The Jitter Setup area allows you to directly change the most commonly used parameters.
8 Jitter Tolerance Tests • NOTE Specify which delay line (220ps or 610ps) is used for the generation of PJ and BUJ. If the 610ps delay line is selected and has any jitter source connected to it, then, it is not possible to disable SJ/SSC/rSSC on the data output. • Specify the delay between 220ps jitter on Clock and Data. Jitter Setup - Procedures Do the following to compose the desired jitter: 1 Click Jitter Setup from the Jitter submenu. This opens the Jitter Setup window.
8 Jitter Tolerance Tests Set Jitter Configuration Use Jitter Configuration area to specify whether you want to use: • SSC or rSSC • RJ and BUJ or sRJ Set Jitter Distribution 1 Specify whether the corresponding jitter source is applied to the data and/or the clock output by clicking the corresponding LED button. 2 Specify the delay line (220ps or 610ps) used for the generation of PJ and BUJ. 3 Define the delay between 220ps delay line jitter on Clock and Data.
8 Jitter Tolerance Tests Select the Delay Line To select the delay line for (sinusoidal) periodic jitter 1 and 2 and BUJ, press 610ps button. The LED will turn green if the 610ps delay line is selected. Note that sRJ can not be enabled if the 610ps delay line is selected. NOTE The 610ps delay line cannot be used for bit rates above 3.37 GHz. Specify the Jitter Components In this section, an example of periodic jitter 1 is shown to specify the jitter components.
8 Jitter Tolerance Tests A parameter that must be set for any jitter type is Amplitude. This defines the timely width of the jitter component. 4 Observe the consumption of delay line capacity indicated by the delay line markers. Ensure that the total jitter remains within the margins of the chosen delay line.
8 Jitter Tolerance Tests To specify the jitter components: 1 Activate a jitter source (press the corresponding button). 2 Press Edit button to access all parameters of the corresponding jitter type. If you want to specify the jitter components for Constant Amplitude Sweep, skip steps 5 to 6. However, if you want to specify the jitter components for Variable Amplitude Sweep, skip steps 3 to 4 and go to step 5. 3 Activate Constant Amplitude Sweep (press Constant Amplitude Sweep button).
8 Jitter Tolerance Tests This opens a window which allows you to set all parameters available for the given jitter type. 6 Specify the following parameters: – Standard – Sweep Time – Nr. of Steps – Step Distance For more information on parameters, see “Periodic Jitter 2 – Variable Amplitude Sweep Parameters” on page 424. 7 Observe the consumption of delay line capacity indicated by the delay line markers. Ensure that the total jitter remains within the margins of the chosen delay line.
8 Jitter Tolerance Tests Jitter Setup - Reference The Jitter Setup function is used for composing the total jitter in a defined and calibrated way. Jitter Enable The Jitter On/Off button is used enable or disable jitter generation. When this button is disabled, no jitter is generated. This refers to the whole composition shown on the screen. When this button is enabled, the present status becomes active. NOTE The Jitter On/Off button has no impact on SSC.
8 Jitter Tolerance Tests Defining Jitter Delay on Clock and Data Use this entry to define the delay between 220ps jitter on clock and data. Delay Line Switch This switch can be pressed to switch the source of periodic jitter (1 and 2) and BUJ to either the 610ps or the 220ps delay line. NOTE 414 The 610ps delay line cannot be used for data rates above 3.37 GHz.
8 Jitter Tolerance Tests Delay Line Display The delay line display shows red markers that indicate the consumption of delay line capacity by the present jitter setup. Colored bars indicate the consumption by the various jitter components. The consumption is shown in UI (Unit Intervals) and is calculated from the present data rate. Because the delay lines have a fixed time capacity, the scales change when the bit rate is changed.
8 Jitter Tolerance Tests The Jitter Setup area allows you to enable a jitter source, to change its most commonly used jitter parameters directly or to go into advanced edit mode by pressing the corresponding Edit button. Spread Spectrum Clock The Spread Spectrum Clocking setting controls the pattern generator's spread spectrum (SSC) clocking feature. When the SSC is enabled, it impacts the Data Out, Clock Out, Aux Data Out, and Trigger/Ref Clock Out ports.
Jitter Tolerance Tests 8 • Change the ‘centerspread’value: the bitrate remains unchanged, while the upper and lower frequency changes according to the selected deviation. The deviation value specifies ½ p-p value. • Change the ‘downspread’ value: the upper frequency remains unchanged, while the bitrate is adjusted. The deviation value specifies the p-p value.
8 Jitter Tolerance Tests The selected deviation type is also indicated in the jitter composition shown in the Jiter Setup area. The following are the deviation type indications: • Downsperad: • Centerspread: • Upspread: • Arbitary: Profile You can click on the corrosponding button to choose between the Triangular and Arbitrary profile. The green LED on the button indicates the currently selected profile. For Arbitary profile, you need to specify the Arbitary Waveform file.
8 Jitter Tolerance Tests Residual Spread Spectrum Clock The Residual Spread Spectrum Clock (rSSC) is generated by modulating the clock that is used for data generation. It can therefore not be enabled if SJ or SSC is active. Residual Spread Spectrum Clock is characterized by: • Amplitude (peak-to-peak) • Frequency There is a dependency between frequency and acceptable amplitude. A graph indicates whether the chosen setting of frequency and amplitude is acceptable.
8 Jitter Tolerance Tests Amplitude(p-p) The maximum peak-to-peak Amplitude is limited by the chosen frequency. Frequency The maximum Frequency of the rSSC jitter source depends on the amplitude. Sinusoidal Jitter Parameters Sinusoidal jitter (SJ) is independent from periodic jitter. It is generated by modulating the clock that is used for data generation. It can therefore not be enabled, if SSC or rSSC is active.
Jitter Tolerance Tests 8 Periodic Jitter 1 Parameters The Periodic Jitter is characterized by: • Amplitude • Frequency • Waveform A graph indicates whether the chosen setting of frequency and amplitude is tolerable. Amplitude The maximum peak-to-peak Amplitude is limited by the free capacity of the chosen delay line. Frequency The maximum Frequency of the periodic jitter source depends on the chosen waveform. Refer to the technical specifications.
8 Jitter Tolerance Tests You can click on the corrosponding button to choose among the different modes of Periodic Jitter 2. The green LED on the button indicates the currently selected mode of Periodic Jitter 2. The ability of sweeping the PJ settings supports testing DUTs according to the PCIe 2.0 and PCIe 3.0 standard without the need of providing a jitter modulation signal externally.
Jitter Tolerance Tests 8 Amplitude The maximum peak-to-peak Amplitude is limited by the free capacity of the chosen delay line. Frequency The maximum Frequency of the periodic jitter source depends on the chosen waveform. Refer to the technical specifications. Waveform You can switch between sine, rectangular, or triangular jitter source waveforms.
8 Jitter Tolerance Tests Amplitude The maximum peak-to-peak Amplitude is limited by the free capacity of the chosen delay line. Frequency Range The stop frequency has to be higher than start frequency and the range should be in accordance with the selected waveform. Waveform You can switch between sine, rectangular, or triangular jitter source waveforms. Sweep Time You can specify the duration for sweeping the specified frequency range once. Nr.
8 Jitter Tolerance Tests Standard You can use this drop-down list to specify whether you want to select a pre-defined standard or a user-defined standard. All the available pre-defined standards will be shown in this list. However, if you select the user-defined standard, press Load button to locate the Jitter Tolerance standard. The user-defined standard uses the same file format like the Jitter Tolerance Compliance measurement. See “Jitter Tolerance Compliance - Concepts” on page 446.
8 Jitter Tolerance Tests • Amplitude • PRBS polynomial • Low pass filter A graph indicates whether the chosen setting of frequency and amplitude is tolerable. You can specify your own setup or choose one of the predefined settings. Setting You can choose one of four predefined settings: • CUSTom: The individual parameter values apply. • CEI6G: Meant for CEI 6 Gbit/s tests: PRBS data rate is 1.1 Gbit/s, the PRBS polynomial is 29-1, the low-pass filter is 100 MHz.
8 Jitter Tolerance Tests Filter The bounded uncorrelated jitter source is equipped with three low-pass filters with cut-off frequencies at 50, 100, and 200 MHz. One of these filters is always active. Random Jitter Parameters Random Jitter is characterized by: • Amplitude • Filter Settings The graph shows the possible and the current jitter spectrum, which is defined by filter settings. Amplitude The random jitter Amplitude must be entered as an rms (root mean square) value.
8 Jitter Tolerance Tests Filter The random jitter source is equipped with a 10 MHz high pass and a 100 MHz and 500 MHz low pass filter to limit the spectral range. All these filters can be disabled or enabled by pressing the respective buttons. Spectrally Distributed Random Jitter Spectrally distributed random jitter is composed of two jitter sources: low frequency jitter and high frequency jitter. It is characterized by the amplitudes of the low and the high frequency jitter.
8 Jitter Tolerance Tests Amplitude LF (rms) This is the low frequency jitter amplitude as rms value Amplitude HF (rms) This is the high frequency jitter amplitude as rms value Filter The Spectrally Distributed Random Jitter is equipped with a 100MHz Low Pass filter to limit the spectral range, which can be enabled by pressing the corresponding button.
8 Jitter Tolerance Tests NOTE Even if the External jitter source is enabled, you can still add or change internal jitter components. The range of the external jitter source is automatically updated when you add, change, or turn off one of the internal sources. Amplitude This is the maximum jitter Amplitude that can be generated. Voltage This is the maximum voltage that can be accepted. It would generate the maximum jitter amplitude.
Jitter Tolerance Tests 8 Interference Channel Setup Interference Channel Setup - Concepts The optional Interference Channel module can be plugged into the instrument. For an introduction please refer to “Interference Channel” on page 402. This module is to be connected between the pattern generator's Data Out port and the DUT. For installation instructions refer to “Installing Hardware Options Procedures” on page 536. The Interference Channel simulates intersymbol interference (ISI).
8 Jitter Tolerance Tests If you have enabled sinusoidal interference, enter appropriate values for Frequency, Amplitude, and Mode. Interference Channel Setup - Reference The Interference Channel module must be enabled before it can be used. This is done by clicking the corresponding button. When the Interference Channel is disabled, it establishes an almost direct connection between P1 and P2. For details about the minimum trace length refer to the technical specifications.
8 Jitter Tolerance Tests Frequency The frequency range of sinusoidal interference is limited (see the technical specifications). Mode One of four modes can be chosen for sinusoidal interference: • Common: Requires that a differential data signal is generated and connected to P1/P1 or P2/P2 of the Interference Channel. In common mode, both components of the differential data signal are amplitude-modulated with the same sinewave. This adds common mode noise.
8 Jitter Tolerance Tests Green points (BER below limit) and red crosses (BER above limit) are used to display the results. After a measurement has exceeded the target BER or when moving to the next frequency point, a measurement settling or relax time can be applied. This allows the DUT to settle after changes and before the next measurement; this is important because some devices may need time to recover from error conditions caused by jitter exceeding the margin during the search for the threshold.
8 Jitter Tolerance Tests When the measurement has finished, the jitter composition shown in the Jitter Setup window is restored. Target BER The target bit error ratio must be specified. The condition for moving from one jitter amplitude step to the next can be set to • numbers of received bits and errors • a confidence level Additionally, a relax time can be specified.
8 Jitter Tolerance Tests Vertical Search Methods You can specify that the jitter amplitude changes in linear steps from maximum to minimum or vice versa. The step size is adjustable. You can also specify that the jitter amplitude changes in logarithmic steps from maximum to minimum or vice versa. Minimum and ratio are adjustable. Alternatively, you can also specify one of two dynamic (binary) search methods. Searching Down/Up Both search directions, upwards and downwards, are supported.
8 Jitter Tolerance Tests Searching upwards When you search in upward direction (from minimum to maximum), the test for one frequency stops as soon as the measured BER exceeds the target bit error ratio. Dynamic Search If you perform the measurement over a wider frequency range, you measure many useless points with the downwards/upwards methods. To construct the tolerance curve, we actually need only the two points between which the measured result turns from green to red.
8 Jitter Tolerance Tests Jitter Tolerance Characterization - Procedures To access the Jitter Tolerance Characterization measurement, press the corresponding menu item from the Jitter submenu. To set up a Jitter Tolerance Characterization test, press the Properties button. This opens the Jitter Tolerance Properties dialog. Setting the Frequency Range Choose Auto or Manual mode. In Auto mode, the jitter frequencies are automatically calculated: 1 Enter the Start Frequency. 2 Enter the Stop Frequency.
8 Jitter Tolerance Tests 3 Set the Number of Steps between start and stop. The Number of Steps yields a set of logarithmically equidistant frequencies. In Manual mode, the automatically calculated jitter frequencies can be edited. You can also enter your own comma-separated list of frequencies. The Sequence Advance checkbox advances the PG sequence during jitter tolerance characterization. It will be enabled only if the sequence is send to the PG.
8 Jitter Tolerance Tests 3 If desired, set a Relax Time. For details see “Relax time” on page 435. Setting the Search Mode 1 Choose one of six search modes: – Downwards linear – Downwards logarithmic – Upwards linear – Upwards logarithmic – Binary search – Extended binary For details refer to “Vertical Search Methods” on page 436. 2 If desired, change the related parameters. 3 Check the Minimum Jitter Curve and Maximum Jitter Curve checkboxs.
Jitter Tolerance Tests 8 This opens a dialog box which allows you to load a file. By default, the browser searches for files with the suffix .jcs, but you can load any file. NOTE The required file format is described in detail in “User-Defined Standards” on page 452. 5 Check the Show Jitter Curve checkbox. It shows the jitter curve on the Jitter Tolerance Characterization graph.
8 Jitter Tolerance Tests You can also abort the test at any time by pressing the Abort button. Once the measurement has finished, the jitter composition shown in the Jitter Setup window is restored. When the measurement is either completed or aborted, you can obtain a list of measured points by clicking the following tab which appears once the measurement is stopped. Clicking on the above tab you get the list of measured points.
8 Jitter Tolerance Tests 2 Select 'Generate HTML report'. 3 Select the folder where you want to save the report, name the file, and save it. Jitter Tolerance Characterization - Reference Jitter tolerance characterization determines the actual jitter levels where the device under test can no longer maintain a desired BER. Frequency Specification You can choose Auto or Manual mode. Auto mode In Auto mode, the jitter frequency is automatically swept from Start Frequency to Stop Frequency.
8 Jitter Tolerance Tests Manual mode In Manual mode, you can edit the automatically calculated frequencies. You can also enter your own list of frequencies, with a minimum of two frequencies specified. For example, you can jump between low and high jitter frequencies. Enter the frequencies in Hz, separated by commas, for example: 2e4, 1.75e6, 6.334e5, ... Show Compliance Curve Enabling this compliance curve helps in ensuring that the device under test complies with a certain standard.
8 Jitter Tolerance Tests next. The chosen method impacts the measurement duration. The specified parameters influence both measurement duration and precision. You can choose one of six search methods: • Downwards linear • Downwards logarithmic • Upwards linear • Upwards logarithmic • Binary search • Extended binary For a description of these methods refer to “Vertical Search Methods” on page 436. Linear search For linear search, you can set the step size.
8 Jitter Tolerance Tests This defines the minimum jitter amplitude step size. The test for one frequency stops when this step size is reached. For a description of this method refer to “Binary search” on page 437. Extended binary search For extended binary search, you can set the coarse and fine accuracy. The test for one frequency stops when the BER limit is crossed with fine accuracy amplitude steps. For a description of this method refer to “Extended binary search” on page 437.
Jitter Tolerance Tests NOTE 8 The Jitter Tolerance Compliance measurement is a software option that requires a license. For details on how to obtain and install such licenses refer to “Obtaining a License” on page 535. Measurement Principle The Optical Internetworking Forum (OIF) and other institutions have proposed and published standards for testing the performance of data receivers and receiver circuits in the presence of jitter.
8 Jitter Tolerance Tests During a jitter tolerance compliance test, jitter with the specified jitter amplitude is sequentially applied at a number of frequency points in the frequency range of interest. The receiver checks for transmission errors and measures the bit error rate. Measurement Results The graphical display shows the results. A green circle indicates that the BER at this point did not exceed the target BER. A red cross indicates that the BER at this point exceeded the target BER.
8 Jitter Tolerance Tests Jitter Type The standards demand and the measurement generates jitter with sinusoidal distribution. In terms of the Serial BERT, sinusoidal jitter (SJ) and periodic jitter (PJ) with sinusoidal characteristic are used for the test (see also “How the N4903 Generates Jitter” on page 401). Whether SJ or PJ is used depends on the jitter frequency. SJ supports wide amplitude variations up to some MHz. PJ supports narrow amplitude variations over a wide frequency range.
8 Jitter Tolerance Tests • numbers of received bits and errors • a confidence level Additionally, a relax time can be specified. Number of Received Bits or Errors When numbers of received bits or errors are enabled, the measurement proceeds with the next frequency as soon as one of the two numbers is reached. But we are often dealing with very small tolerable bit error ratios, 10–12 and below. Measuring such a bit error ratio precisely (and just for one single point) takes time.
Jitter Tolerance Tests 8 Predefined Standards The Jitter Tolerance Compliance option provides a number of predefined standards. Standard Specifications Predefined standards specify: • Jitter frequency range The jitter frequency range may depend on the bandwidth of the receiver being tested. • Bit rate range Some standards define the jitter tolerance curve as a function of the bit rate.
8 Jitter Tolerance Tests User-Defined Standards You can create your own standard in a text file. User-Defined Standard Specifications A user-defined standard consists of a paired list of numbers. Each pair specifies a jitter frequency (Hz) and the associated jitter amplitude (UI), for example: 2e3, 110 3.24e4, 50 71.234e4, 10 10.654e6, 0.2 18.333e6, 0.2 stands for or just .
Jitter Tolerance Tests 8 Jitter Frequency Test Points You can specify the jitter frequency range to be used for the test, from 1 kHz up to some hundred MHz. You can hence measure only a portion of the standard or points outside the standard. Points outside the standard will always pass the test. If a standard prescribes a certain range, for example from 2 kHz to 20 MHz, you can select a checkbox to adjust the jitter frequency range to the loaded standard.
8 Jitter Tolerance Tests Jitter Tolerance Compliance - Procedures To access the Jitter Tolerance Compliance measurement, click Jitter Compliance from the Jitter submenu. To set up a Jitter Tolerance Compliance test, press the Properties button. This opens the Jitter Tolerance Properties dialog. Setting the Standard Select either a predefined standard or load a standard from a text file. 1 Switch to the Standard tab. A preview and a short description of the selected standard are displayed.
8 Jitter Tolerance Tests The list contains all predefined standards, including user-defined standards. User-defined standards are preceded by an asterisk (*). 3 Select one of the predefined standards. or Press Browse... to load your own standard, stored in a text file. This opens a dialog box which allows you to load a file. By default, the browser searches for files with the suffix .jcs, but you can access any file.
8 Jitter Tolerance Tests TIP With the Auto Frequencies checkbox, you can adjust the test frequencies automatically to the selected standard. This is described belowin “Setting the Frequency Range” on page 456. Setting the Frequency Range You can freely specify the jitter frequency range and the number of points to be measured. By default, the test points are suitably distributed within the frequency range – according to the loaded standard. But you can change the distribution according to your needs.
8 Jitter Tolerance Tests 2 If desired, change the default values. For example, if only a certain frequency range has to be tested: Enter the Start Frequency and the Stop Frequency. 3 Set the Number of Steps between start and stop. TIP When Auto Frequencies is activated, the Number of Steps can also be set on the Standard tab.
8 Jitter Tolerance Tests 2 Choose the Verification Method. Setting a Confidence Level is generally recommended. See also “Target BER” on page 449. If you enable Absolute Bits/Errors, the measurement proceeds with the next amplitude as soon as one of the two numbers is reached. 3 If desired, set a Relax Time. For details see “Relax Time” on page 450. Before Running the Test Before running the test, set up the instrument to ensure that the requirements of the selected standard are met.
8 Jitter Tolerance Tests This opens the Jitter Setup window. 2 Enable jitter generation. Press the respective checkbox. 3 Activate Sinusoidal jitter. 4 Activate Periodic jitter 5 Set the jitter amplitude to 0 UI for both Sinusoidal jitter and Periodic jitter. NOTE When rSSC is enabled the measurement does not use SJ. The maximum available jitter amplitude is reduced for the jitter tolerance measurement.
8 Jitter Tolerance Tests Selecting a point, and clicking on 'Set', updates the Sinusoidal and Periodic jitter of that point on the Jitter Setup page. Generate HTML Report The measurement results can be saved as HTML file. To generate the HTML report: 1 Click on the >> icon on the Jitter menu. The following submenu will appear. 2 Select 'Generate HTML report'. 3 Select the folder where you want to save the report, name the file, and save it.
8 Jitter Tolerance Tests Managing User-Defined Standards User-defined standards are saved in the folder C:\\JTolStandards. The extension is .jcs. Adding a User-Defined Standard To add a user-defined standard to the list: 1 Press the Properties button and switch to the Standard tab. 2 Select the last entry in the list: Browse…. This opens a dialog box which allows you to load a file. NOTE The file must contain a comma-separated list of frequencies and jitter values.
8 Jitter Tolerance Tests Exporting Measurement Results To export the measurement results into a .txt file for later use in external spreadsheet applications: 1 After your measurement has finished and the results are displayed, select Export Data from the Analysis menu. 2 In the Save as dialog box, select the path and file name for the .txt file. 3 Press Save to export the data to the specified destination.
8 Jitter Tolerance Tests frequency range and the jitter amplitude range of the standard is given in the preview. A description is displayed at the bottom of the Standard tab. The last entry in the list is Browse…. Select this entry to load a file that contains a user-defined standard. See “Setting the Standard” on page 454 for details. Margin To aggravate test conditions, the standard's test jitter amplitude can be increased by Margin percent.
8 Jitter Tolerance Tests Consider these recommendations when setting up the pattern generator (see “Setting up the Pattern Generator - Concepts ” on page 121). • NOTE Additional Jitter components you have to add for the compliance test if the standard requires non-sinusoidal jitter components, such as random jitter (RJ) or intersymbol interference (ISI).
8 Jitter Tolerance Tests Manual mode In Manual mode, you can edit the automatically calculated frequencies. You can also enter your own list of frequencies, with a minimum of two frequencies specified. For example, you can jump between low and high jitter frequencies. Enter the frequencies in Hz, separated by commas, for example: 2e4, 1.75e6, 6.334e5, ...
8 Jitter Tolerance Tests Select Point HTML Report This displays the list of measured points. The list classifies the points with their frequency, amplitude, and status. The status indicates whether the test passed or failed at that point. This contains the results as follows: 1 Test Configuration Details: gives the User Comments, Device Type, Last Test Date, Model Number, Serial Number and Software Revision number.
Jitter Tolerance Tests 8 Jitter Tolerance Measurement 4/10/2006 12:12:37 PM Standard: XAUI Margin: 0 % Condition: 2^23-1 PRBS@2.488 Gb/s; Comment: Passed Points: (frequency; amplitude; number of bits; bit error rate; number of errors; result) 1000; 8.5; 3013355404; 0; 0; passed; 1790.61312892; 8.5; 3015843721; 0; 0; passed; 3206.295377462; 8.5; 3015843721; 0; 0; passed; 5741.234598079; 8.5; 3015843721; 0; 0; passed; 10280.33004753; 8.5; 3015843725; 0; 0; passed; 18408.09395274; 8.
8 Jitter Tolerance Tests 468 Agilent J-BERT N4903B High-Performance Serial BERT
S Agilent J-BERT N4903B High-Performance Serial BERT User Guide 9 Solving Problems Solving Problems - Concepts This section provides information that can help you to troubleshoot the Serial BERT in case any problems occur. Setup Problems Setup Problems - Concepts You may run into the following problems when setting up the Serial BERT.
9 Solving Problems If the input signal is still out of range for any of these parameters, the error detector will be immediately disconnected again. Error Detector Does not Count Bits If the analyzer does not count any bits, press Sync Now. This might happen after the Error Detector had glitches at the clock input. DLL Alignment Failed If an error message containing the text "... DLL alignment failed ..." is displayed, increase or decrease the bit rate by 1bit/s.
9 Solving Problems BERT connected to your device: • There is no clock signal from your device. • Your device is faulty. DATA LOSS This indicator turns red when no data signal is detected at the ED Data In port. BERT connected looped back or to your device: • The PG data output is off or not connected. • There is no connection to the ED data input. • The 0/1 threshold is not in the eye limits of the incoming data signal. Use Auto Align or select Avg 0/1 Threshold. • Your cables are faulty.
9 Solving Problems Error This indicator turns red when errors are detected. View the BER bar or BER Results to see the nature of the errors. BERT connected looped back or to your device: • Stable errors caused by the error add function: Turn error add off. • Stable errors caused by false sync: Select a lower sync threshold BER. • Variable and high errors may be caused by faulty connectors/cables. BERT connected to your device: • Stable or variable errors can also be caused by your device.
Solving Problems 9 Some phenomenon you might observe with the possible causes are listed below: Constant Errors (More Errored 0's than 1's) Possible Cause A hardware failure may have occurred in your device. The errors became constant at the point of failure. Digital circuitry often has parallel architecture in which data lines are multiplexed in stages into one serial data line. This example assumed that a PRBS pattern was going through digital circuitry that had a total multiplexing ratio of 256:1.
9 Solving Problems Example Log of Constant Errors The error ratio during periods of constant errors was 0.00195313. This corresponds to 1.953E-3. Random Errors 474 Possible Cause Random errors may have been caused by a noisy waveform. Noise can be caused by bad cables or connectors. Additional Information The output of your device may contain errors due to noise, although a noisy waveform is not present at the output.
9 Solving Problems Sync Loss Seconds Possible Cause The sync loss seconds in this example may have been caused by high bursts of errors or momentary clock loss. The clock signal from a clock recovery module may have momentarily ceased. To confirm periods of momentary clock loss, check the measurement log file. Additional Information In this example, the analyzer was in auto sync mode. If the analyzer was in manual sync mode, the results would not have been measured and displayed in the same way.
9 Solving Problems Using the Measurement Log to Identify Problems Measurement Logs are saved as CSV (comma separated variable) files. Follow the steps below to view results in a measurement log file: 1 Copy the measurement log file from the analyzer to your PC. By default, log files are saved in the folder C:\\log. 2 Open a spreadsheet application on your PC. 3 Import the measurement log file.
Solving Problems 9 Table 32 Instantaneous Cumulative BIT COUNT BIT COUNT Cumulative ERROR COUNT ERROR COUNT Cumulative ERROR RATIO ERROR RATIO Cumulative ERRORED 0 COUNT ERRORED 0 COUNT Cumulative ERRORED 0 RATIO ERRORED 0 RATIO Cumulative ERRORED 1 COUNT ERRORED 1 COUNT Cumulative ERRORED 1 RATIO ERRORED 1 RATIO Cumulative SYNC LOSS ERROR SECONDS Cumulative PG CLOCK LOSS ERROR FREE SECONDS Cumulative ED CLOCK LOSS ERROR DECISECONDS Cumulative DATA LOSS ERROR FREE DECISECONDS Cumula
9 Solving Problems Other Messages Other Messages - Concepts This section covers additional error messages that may occure when working with the Serial BERT. Overheat Protection The Serial BERT is equipped with an overheat protection function to prevent it from overheat damage.
9 Solving Problems The Serial BERT will be shut down, your instrument settings will not be saved. Troubleshooting Overheating CAUTION If the Serial BERT has indicated overheating, do the following: • Shut down the instrument and let it cool (45minutes to 1hour). • Reduce the environmental temperature. • Make sure that the fans are running and the ventilation holes are not blocked. • If the problem continuously recurs, contact Agilent support.
9 Solving Problems N4916B System Setup Calibration N4916B System Setup Calibration The input timing of the N4916B has to be calibrated to ensure error free operation for all data rates. This calibration is required once for a dedicated setup of N4903B, N4916B and the cable kit being used, and has to be repeated whenever one of these are changed in the setup. NOTE The Input Timing setting that is determined during this calibration is only valid for all data rates when using the N4915A-010 cable kit.
9 Solving Problems WARNING The instrument settings will be lost once you start the calibration process. It is recommended to save the instrument settings before you start the calibration process. 5 Click Yes to continue if you have already saved the settings. Calibration Procedure The following steps explain the calibration prodecure: 1 Store current instrument setting if the current setting is required after the calibration is done 2 Load the calibration setting.
9 Solving Problems – Press the Auto Align key on the front panel – Switch to the Accumulated Results Screen – Press the Start Accum key on the front panel to start the BER measurement – For each measurement write down the number of errors at the end of the measurement – The correct Input Timing is selected by the following rule: Look for the BER measurement that produced no error, and that has the biggest distance to a measurement with errors and select the Input Timing setting that was used in this measur
Solving Problems 9 Table 33 Input Timing Errors Forward distance to ‘bad’ setting Backwar d distance to ‘bad’ setting 1 0 3 1 2 0 2 2 3 0 1 3 4 50 0 0 5 1000 0 0 6 100 0 0 Case-2 (Calibration using an Oscilloscope) – Connect Data Out of N4916B to the oscilloscope. – Terminate Data Out of N4916B. – Connect Trigger-Out of PG with Trigger-In of oscilloscope. – Terminate Trigger Out of PG with 50 Ohm termination resistor.
9 Solving Problems How to Create the Calibration Settings Use the following calibration settings: • Preset Instrument State • Enable De-Emphasis • Set all cursors to 0dB • Set Output Offset at N4916B => 0V • Set Output Amplitude at N4916B => 400mV • Set Data rate to the maximum possible data rate of your setup. This depends on the maximum data rate of the PG, ED and N4916B. Note that a N4916B that is specified for 10.5Gb/s should be calibrated at 12.
9 Solving Problems Calibration Preparation • Connect the N4903B’s Trigger Output with the oscilloscope’s Front panel Trigger input • Connect the N4916B’s Data Output with channel 1 of the oscilloscope • Terminate the N4916B’s complement output using a 50Ohm termination resistor 1 N4903B a Open the configuration page for external instruments b Enable the De-Emphasis function c Click the Load Calibration Setting… button d Set the PG’s data rate to 14.
9 Solving Problems 486 • Input Timing Setup = 2 • Input Timing Setup = 3 Agilent J-BERT N4903B High-Performance Serial BERT
Solving Problems • Input Timing Setup = 4 • Input Timing Setup = 5 Agilent J-BERT N4903B High-Performance Serial BERT 9 487
9 Solving Problems • Input Timing Setup = 6 Input Timing Setup #1 shows timing violations at the input of the N4916B. All other measurements do not show any violations during the measurement interval (6 minutes).
Solving Problems Notes on the measurement duration 9 The example above was done for a target BER measurement of 10-3 and a confidence level of 95%. As a rule of thumb, this requires to measure approximately 3 times the number of bits given by the target BER.
9 Solving Problems When using lower target BER values, the required measurement time per Input Timing Setting is getting longer very quickly. The required measurement time for a target BER of 10-9 is approximately 338e6 seconds per Input Timing Setting, which is not practicable in a real measurement setup. N4876A System Setup Adjustment N4876A System Setup Adjustment The N4876A has to be adjusted to ensure error free operation for all data rates.
Solving Problems 9 4 Click Yes to continue if you have already saved the settings. Once the auto adjustment starts, you will see the adjustment progress in the following screen: NOTE The adjustment process may take several minutes for completion. The following message will pop up after the successful completion of adjustment process.
9 Solving Problems M8061A System Setup Adjustment M8061A System Setup Adjustment The input timing of M8061A has to be adjusted to ensure error free operation for all data rates. This adjustment is required once for a dedicated setup of N4903B, M8061A and the cable kit being used, and has to be repeated whenever one of these are changed in the setup or the operating temperature is changed by more than 5o Celsius, compared to the previous adjustment.
Solving Problems NOTE 9 The adjustment process may take several minutes for completion. The following message will pop up after the successful completion of adjustment process. Problems with the N4916A Problems with the N4916A - Concepts When you have opened the External Instrument(s) - Config window and have selected N4916A from the external instruments, select De-emphasis under Enable Function column in the list, the N4916A Connection window appears.
9 Solving Problems If this happens, ensure that the N4916A has been correctly installed (see “Installing the N4916A/B - Procedure” on page 539). Check the Power Supply Ensure that the power cord is connected. The green LED close to the power switch must be illuminated when the unit is turned on. Check the USB Communication Ensure that the USB cable is connected between the USB port at the rear of the N4916A and one of the USB ports at the rear of the Serial BERT.
Solving Problems 9 correctly, the VISA Assistant should indicate an instrument named "D4916". Click UsbDevice1 or D4916. 5 Click the Formatted I/O tab. 6 In the Instr. Lang. box (Instrument Language), enable SCPI. 7 Click the *IDN? button. This generates the identification query in SCPI format. The instrument should return "Agilent Technologies, N4916A,," and the software revision. If you could proceed until here, the N4916A power supply and the processor are fully functional.
9 Solving Problems 4 From the Navigation Menu, select External Instrument(s) and then click Config sub-menu. The Config window shows a list of all instruments connected to the Serial BERT. Confirm the presence of N4916A enrty in that list. 5 Corrosponding to N4916A entry, select the De-Emphasis function check box from the given options under Enable Function column. It opens the Deemphasis Signal Converter Connection dialog. 6 Click Enable button.
9 Solving Problems Then the oscilloscope should show an output eye like the following: If you see a display like above everything is ok. If you see merely one fixed level, either the N4916A or the output of the Serial BERT is defective. Check the Output of the N4916A Using the Error Detector If you have connected the N4916A to a Serial BERT and have no oscilloscope at hand, you can use the error detector for checking the output of the N4916A.
9 Solving Problems 4 Connect the Output of the N4916A to the Data Input of the error detector. 5 Ensure that the error detector follows the generated pattern. – Press the Auto Align button. – From the Analysis panel select the Eye Diagram page. – Press the Start button.
Solving Problems 9 Note that a de-emphasis value of 6 dB causes a 50 percent amplitude reduction. Check the Output of the Pattern Generator with a Scope If the N4916A does not produce a signal, check the Data Out of the Serial BERT. • NOTE Connect the Data Output of the pattern generator to an oscilloscope. As long as the N4916A is in "connected" state (after pressing the Connect button), the pattern generator's output is programmed to a fixed level. The amplitude is set to 1.2 V, the offset to 0.
9 Solving Problems 3 Connect the Data Output of the Serial BERT to the Data Input of the error detector. 4 Ensure that the error detector follows the generated pattern. – Press the Auto Align button. – From the Analysis panel select the Eye Diagram page. – Press the Start button. After some seconds, you should see an eye diagram like the following: This is the eye of a non-distorted pattern.
9 Solving Problems Cannot-Connect Problems An error message may appear when you now press Enable. If this happens, ensure that the N4916B has been correctly installed (see “Installing the N4916A/B - Procedure” on page 539). Check the Power Supply Ensure that the power cord is connected. The green LED close to the power switch must be illuminated when the unit is turned on.
9 Solving Problems 3 From the Agilent IO Control menu, open the VISA Assistant. 4 If no VISA alias name has been assigned so far, the VISA Assistant should indicate an instrument named "UsbInstrument1". If the N4916B has been installed correctly, the VISA Assistant should indicate an instrument named "DATA_OUT". Click UsbInstrument1 or DATA_OUT. 5 Click the Formatted I/O tab. 6 In the Instr. Lang. box (Instrument Language), enable SCPI.
Solving Problems 9 7 Click the *IDN? button. This generates the identification query in SCPI format. The instrument should return "Agilent Technologies, N4916B,," and the software revision. If you could proceed until here, the N4916B power supply and the processor are fully functional. If not, the N4916B is probably defective. Ensure that the User Software can Access the N4916B 1 From the Windows task bar, restore the user interface – click GUI Agilent 4900 Series.
9 Solving Problems Once the de-emphasis function is enabled you can see the de-emphasis parameters in the PG Output screen. 7 Again, from the Config window, corrosponding to N4916B, clear the DeEmphasis function check box. It will disable the de-emphasis function. If you could proceed until here, the connection of the user software to the N4916B is ok.
9 Solving Problems Then the oscilloscope should show an output eye like the following: If you see a display like above everything is ok. If you see merely one fixed level, either the N4916B or the output of the Serial BERT is defective. Check the Output of the N4916B (De-Emphasis Part) Using the Error Detector If you have connected the N4916B to a Serial BERT and have no oscilloscope at hand, you can use the error detector for checking the output of the N4916B.
9 Solving Problems The eye diagram should look like this after some seconds: Check the Output of the Pattern Generator with a Scope If the N4916B does not produce a signal, check the Data Out of the Serial BERT. • NOTE 506 Connect the Data Output of the pattern generator to an oscilloscope. As long as the N4916B is in "connected" state (after pressing the Enable button), the pattern generator's output is programmed to a fixed level. The amplitude is set to 1.2 V, the offset to 0.
9 Solving Problems The oscilloscope should show the following picture: If you see only a straight line, ensure that the pattern generator is not disabled and that it is set up to generate a pattern (e.g. PRBS).
9 Solving Problems 1 Connect PG Clock Out to Clock Multiplier Clock In 2 Terminate PG Clock Out with 50 Ohm 3 Termiate PG Trigger Out with 50 Ohm 4 Terminate Clock Multiplier Clock In with 50 Ohm 5 Set Clock Multipiler to 1 6 Measure the signal on scope.
9 Solving Problems Check the Output of the N4916B (Clock Multiplier Part) Using a Scope ( Clock Multiplier 2) Following are the steps to test clock multiplier 2 using a scope: 1 Connect PG Clock Out to Clock Multiplier Clock In 2 Terminate PG Clock Out with 50 Ohm 3 Termiate PG Trigger Out with 50 Ohm 4 Terminate Clock Multiplier Clock In with 50 Ohm 5 Set Clock Multipiler to 2 6 Measure the signal on scope.
9 Solving Problems N4916B Status LEDs This following section explains about the different LEDs that are available on the front panel of the N4916B De-Emphasis Signal Converter.
9 Solving Problems Problems with the N4876A Problems with the N4876A - Concepts When you have opened the External Instrument(s) - Config window and have selected N4876A from the external instruments, select Multiplexer under Enable Function column in the list, the Multiplexer Connection dialog appears. Cannot-Connect Problems An error message may appear when you now press Enable. If this happens, ensure that the N4876A has been correctly installed (see “Installing the N4876A - Procedure” on page 544).
9 Solving Problems 3 From the Agilent IO Control menu, open the VISA Assistant. 4 If no VISA alias name has been assigned so far, the VISA Assistant should indicate an instrument named "UsbInstrument1". If the N4876A has been installed correctly, the VISA Assistant should indicate an instrument named "N4876A". Click UsbInstrument1 or N4876A. 5 Click the Formatted I/O tab. 6 In the Instr. Lang. box (Instrument Language), enable SCPI.
Solving Problems 9 7 Click the *IDN? button. This generates the identification query in SCPI format. The instrument should return "Agilent Technologies, N4876A," and the software revision. If you could proceed until here, the N4876A power supply and the processor are fully functional. If not, the N4876A is probably defective. Ensure that the User Software can Access the N4876A 1 From the Windows task bar, restore the user interface – click GUI Agilent 4900 Series.
9 Solving Problems 6 Click Enable button. It enables N4876A 28 Gb/s Multiplexer 2:1 that is connected between data out of Serial BERT and the DUT. Once the output levels of Data Out that you set on the PG Output screen will control the output levels of the N4876A's Data Output. The level setting of Aux Data will be disabled on the PG Output screen when the multiplexer function is enabled. 7 Again, from the Config window, corrosponding to N4876A, clear the Multiplexer function check box.
Solving Problems 9 Then the oscilloscope should show an output eye like the following: If you see a display like above everything is OK. If you cannot see a clear eye diagram, then check if the N4876A is connected properly to the Serial BERT. The N4876A must be connected to Data Out, Aux Data Out and Aux Clk Out to work properly.
9 Solving Problems 3 Terminate PG Data /Out with 50Ohm 4 Terminate PG Clock /Out with 50Ohm 5 Ensure that the error detector follows the generated pattern. – Press the Auto Align button. – The error detector should finish the auto alignment successfully and report a BER of 0.000. – Manually adjust the ED sample delay by 166 ps. – Press Sync Now. – The error detector should report a BER of 0.000.
9 Solving Problems If you see only a straight line, ensure that the pattern generator is not disabled and that it is set up to generate a pattern (e.g. PRBS). N4876A Status LEDs This following section explains about the different LEDs that are available on the front panel of the N4876A 28 Gb/s Multiplexer 2:1. Table 35 LED Name Description Output Signals the state of the output amplifier Constantly off when output is disabled (0.
9 Solving Problems Table 35 LED Name Description LAN Constantly green when LAN is operable Constantly red when LAN is not operable Flashing green when device identification is enabled Constantly orange during boot phase of the instrument Power Constantly orange when instrument is OFF (in power standby mode) Constantly green when instrument is ON (powered) Problems with the M8061A Problems with the M8061A - Concepts When you have opened the External Instrument(s) - Config window and have selected M80
9 Solving Problems on the front panel of M8061A. For more information on the functionality of these LEDs, refer to the section “M8061A Status LEDs” on page 522. Check the USB Connections Ensure that the USB cable is properly connected between the USB ports of the M8061A and one of the USB ports of the J-BERT N4903B. Use the matched USB cable to establish the USB connection. The mini USB port is available at the front side of the AXI frame which is mounting the M8061A module. Use the USB port (USB 2.
9 Solving Problems 5 Corresponding to the M8061A entry, select the Mux with Deemphasis function check box from the given options under Enable Function column. It opens the M8061A Connection dialog. 6 Click Enable button. It enables M8061A Multiplexer with Deemphasis that is connected between data out of J-BERT N4903B and the DUT. Once the Mux with Deemphasis function is enabled, you will see the Mux with Deemphasis parameters in the M8061A window.
9 Solving Problems 7 Again, from the Config window, corresponding to M8061A, clear the Mux with Deemphasis function check box. It will disable the multiplexer function, and the PG Output screen will allow to control the output levels of Data and Aux Data at the connectors of the PG again. If you could proceed until here, you have successfully connected the application software to the M8061A.
9 Solving Problems If you cannot see a clear eye diagram, then check if the M8061A is connected properly to the J-BERT N4903B. The M8061A must be connected to Aux Data Out, Data Out and Aux Clk Out to work properly. If the connection to the J-BERT N4903B is setup correctly, then check the output signal of the J-BERT N4903B at the Data Out, Aux Data Out (both should show a eye diagram with an offset of -70 mV and a amplitude of 800mV).
Solving Problems 9 • After the M8061A is ready, the red "FAIL" LED goes off and the green "ACCESS" LED stays on. • When the green "ACCESS" LED is on, you can start the firmware. • When the firmware is communicating successfully with the module, the green "ACCESS" LED starts flashing.
9 Solving Problems 524 Agilent J-BERT N4903B High-Performance Serial BERT
S Agilent J-BERT N4903B High-Performance Serial BERT User Guide 10 Customizing the Instrument Customizing the Instrument - Concepts The Serial BERT provides various utilities for adjusting the instrument to your personal preferences. Restoring the System Agilent Recovery System - Procedures The Agilent Recovery System can be used to repair the system in case of serious malfunction. CAUTION When the operating system is restored, all personal settings, programs and user data will be lost.
10 Customizing the Instrument How to Create the Agilent Recovery Partition When booting the instrument for the first time, the Agilent Recovery System will automatically set up the recovery partition. This process is indicated by a small window that pops up on first boot and will take less than one minute to complete. How to Boot from the Agilent Recovery System The following steps will guide you to boot from the Agilent Recovery System. 1 Plug in a keyboard and reboot the instrument.
10 Customizing the Instrument Table 36 Options Description Run CHKDSK to resolve Disk Issues If you encounter any malfunctions you should always try this option first. CHKDSK won’t delete any of your user generated data but try to recover the disk issues. The CHKDSK process may take more than 2 hours. Restore Factory Settings Use this option only if CHKDSK could not repair the system. This option will restore the originally installed operating system and software.
10 Customizing the Instrument Configuring the Instrument Configuring the Instrument - Concepts This section covers information on the configuration of the Serial BERT. Updating the Software New software features are offered for the Serial BERT from time to time; you may be able to enhance the capabilities of your instrument by updating the software. See the Agilent Serial BERT web site to see if updates are available for your instrument: http://www.agilent.
Customizing the Instrument 10 Configuring the Instrument - Procedures Follow the instructions below to configure the instrument according to your personal needs: Setting Date and Time Your instrument uses the date and time when: • Saving files • Saving self test information • Saving calibration information To set the date and time of your instrument, do the following: 1 On the Utilities menu, click Set Time and Date... This opens the Date/Time Properties dialog box. 2 Set the date and time.
10 Customizing the Instrument For more details refer to “Using an External Monitor ” on page 533 and “Using the On-Screen Keyboard” on page 532. NOTE NOTE If the touchscreen is disabled, the external keyboard and mouse are also disabled. See “Turning the Touchscreen Off/On” on page 531 on how to enable/disable the touchscreen. If caps lock is pressed on the keyboard, this will also affect the frontpanel keys, i.e.
Customizing the Instrument 10 Configuring the Instrument - Reference The Serial BERT has the following utilities for configuring the instrument: Set Date/Time Dialog Box This function opens the Date/Time Properties dialog box of Windows®. Use this dialog box to set the date and time as well as the time zone of your location. GPIB Address Change Dialog Box Your instrument is set to a default GPIB (General Purpose Interface Bus) address.
10 Customizing the Instrument 1 On the Utilities menu, click Touchscreen Off. The Enable Touchscreen icon is placed in the upper right hand corner of the instrument display. 2 Click the Enable Touchscreen icon to enable the touchscreen, keyboard, and mouse again. Calibrating the Touchscreen In normal usage, you should not need to calibrate the touchscreen display.
10 Customizing the Instrument 2 Press the On-Screen Keyboard button to open the on-screen keyboard. 3 Click letters or numbers as desired. For numbers, you must add the unit by pressing the G/n, k/m, M/µ, or x1 front panel button. 4 Press the On-Screen Keyboard button again to close the on-screen keyboard. 5 Continue with your work. NOTE You can also access the on-screen keyboard from the Utilities menu. A check mark appears next to the menu entry, indicating that it has been selected.
10 Customizing the Instrument TIP You can always move the On-Screen Keyboard window by clicking its title bar and dragging it. Additionally, you can switch between different views of the on-screen keyboard.
Customizing the Instrument 10 Installing Software Licenses Installing Software Licenses - Procedures The Serial BERT has optional features that require a license. The following procedures explain how to obtain and install a license: Obtaining a License To obtain a license: 1 In the Utilities menu, click Licenses. The Licenses dialog informs you about the options that require a license. 2 Record the HOST ID / MAC Addr shown in the window. This is the MAC Address of the Serial BERT.
10 Customizing the Instrument Licenses Dialog The Licenses dialog informs you about the options that require a license. The HOST ID / MAC Addr shown in the window is the MAC Address of your Serial BERT. You will be asked for this address when you request a license file for an option from Agilent. NOTE NOTE The MAC Address is used for generating the license key code. That means, the key stored in the license file refers to just one instrument.
Customizing the Instrument 10 At the present time, the J20 Interference Channel is available. This module can be used for adding intersymbol interference and sinusoidal interference to the signal generated by the Pattern Generator. Installing the J20 Option To install the J20 Interference Channel: 1 Perform the regular shutdown procedure provided by the operating system. 2 Disconnect the instrument from mains. CAUTION Never plug in or remove a module while the instrument is connected to mains power.
10 Customizing the Instrument If you connect P2 to the Data Out of the Pattern Generator, P1 provides the output. CAUTION The module has golden 2.4 mm connectors. Be sure to use the specific cables or adapters for the connections. Improper connection can damage the module. Installing External Instrument(s) Installing External Instrument(s) - Concepts This section describes the procedure to install the external instruments with the Serial BERT.
10 Customizing the Instrument Installing the N4916A/B - Procedure The N4916A/B De-Emphasis Signal Converter is an optional instrument that can be connected to the Serial BERT. For general information see “Understanding the N4916A/B De-Emphasis Signal Converter” on page 37. When you install the N4916A/B for the first time, you need to configure the USB interface of the Serial BERT.
10 Customizing the Instrument 3 Click the About item to check the details. If you wish to update the Agilent IO Libraries Suite, you can download the software from http://www.agilent.com/find/iolib. Installation instructions are given in the associated Readme file. Connecting the N4916A/B via USB The N4916A/B is controlled by the Serial BERT via USB. 1 Connect mains power to the N4916A/B and switch the instrument on.
Customizing the Instrument 10 5 Click Next and let the Wizard locate the appropriate driver. The USB driver is part of the Agilent IO Libraries Suite. 6 Step through the Wizard by clicking Next. The Wizard will finally set up a USB device named "USB Test and Measurement Device".
10 Customizing the Instrument Assigning a VISA Alias Name Assigning a unique VISA alias name to the N4916A/B simplifies remote programming and helps to make programs portable. The Serial BERT software requires a specific VISA alias name to access the N4916A/B. Using the Agilent Connection Expert The Agilent Connection Expert is available on from revision 14.0 of the Agilent IO Libraries Suite.
Customizing the Instrument 10 4 Click OK. The Agilent Connection Expert for N4916A shows a window like the following: The Agilent Connection Expert for N4916B shows a window like the following: For detailed instructions refer to the Connectivity Guide which is part of the Agilent IO Libraries Suite Documentation. Connecting to Pattern Generator and DUT (N4916A) After the USB port has been configured, you can remove keyboard and mouse and make the signal connections.
10 Customizing the Instrument 1 Use the N4915A-004 cable or a 2.4 mm cable (m-m) longer than 350 mm and connect the pattern generator's Data Out port to the Input port of the N4916A. 2 Mount the SMA 50 Ohm termination on the unused Data Out port of the pattern generator using the SMA to 2.4 mm adapter. 3 Enable the De-Emphasis Signal Converter as described in “How to Enable/ Disable N4916B De-Emphasis Function” on page 46 and set the levels such that the DUT will not be damaged. 4 Use a matched pair of 2.
10 Customizing the Instrument entering the required information. You can remove keyboard and mouse when you are done. Checking the Serial BERT Software Revision The Serial BERT software supports the N4876A on from revision 6.7 or later. 1 Switch on the Serial BERT. 2 In the Help menu, click About. 3 If the software revision is below 6.7, you need to update the software (see “Updating the Software” on page 528).
10 Customizing the Instrument If you wish to update the Agilent IO Libraries Suite, you can download the software from http://www.agilent.com/find/iolib. Installation instructions are given in the associated Readme file. Connecting the N4876A via USB The N4876A is controlled by the Serial BERT via USB. 1 Connect mains power to the N4876A and switch the instrument on. NOTE When you connect the N4876A via USB, it is important that the N4876A is switched on.
Customizing the Instrument 10 Assigning a VISA Alias Name Assigning a unique VISA alias name to the N4876A simplifies remote programming and helps to make programs portable. The Serial BERT software requires a specific VISA alias name to access the N4876A. Using the Agilent Connection Expert The Agilent Connection Expert is available on from revision 15.5 of the Agilent IO Libraries Suite.
10 Customizing the Instrument "N4876A" is the alias used by the software. 4 Click OK. The Agilent Connection Expert for N4876A shows a window like the following: For detailed instructions refer to the Connectivity Guide which is part of the Agilent IO Libraries Suite Documentation. Connecting to Pattern Generator and DUT After the USB port has been configured, you can remove keyboard and mouse and make the signal connections. 1 Disable the outputs of N4903B and use the matched cable kit (2x 2.
Customizing the Instrument 10 connect the pattern generator to the N4876A. Make sure to connect Data Out to Data In, Aux Data Out to Aux Data In and Aux Clk Out to Aux Clk In. 2 Mount the SMA 50 Ohm terminations on the unused Data Out and Aux Data Out ports of the pattern generator using the SMA to 2.4 mm adapters. 3 Enable the N4876A as described in “How to Enable/Disable N4876A Multiplexer Function” on page 47 and set the levels such that the DUT will not be damaged.
10 Customizing the Instrument 1 Connect the USB cable between the USB ports available on the front side of the ESM of chassis with the USB port available on the rear side of J-BERT N4903B.
10 Customizing the Instrument 4 Connect the differential outputs of the M8061A to the DUT. If only one output of the M8061A is needed, connect a 2.4 mm 50 Ohm termination to the unused output. Installing Webserver Customizing the Web Server- Concept What is a Web Server? A web server is a program that operates by accepting HTTP requests from the network, and providing an HTTP response to the requester. Enabling a web server on an instrument allows it to accept and respond to HTTP requests from clients.
10 Customizing the Instrument 2 The instrument web server responds with the instrument’s “Welcome Page”, as shown in the screen below. 3 Click on the left-side link to control/monitor the instrument. 4 The browser returns the instrument display and control applet in the main display frame which contains the instrument display and the instrument-centric controls provided in the applet. How to Install the Web Server The following steps explain how to download, and install the Web Server.
Customizing the Instrument 10 The WebControl dialog provides the user with the current IP address of the instrument. The IP address can be used on the internet explorer to address the instrument. Additionally the WebServer running on the instrument can be enabled/disabled, and a password can be specified. Without a valid password you can not access the instrument's GUI. The following screen displays the WebControl dialog.
10 Customizing the Instrument 4 On the web control dialogue shown above you can request for help by pressing the 'Help' button on the lower right hand corner. 5 The instrument IP will be automatically generated in the space labeled 'Instrument IP Address'. While adding the instrument on the network the IP address will be defined. The DHCP protocol automates the assignment of the IP addresses.
10 Customizing the Instrument Printing Printing - Procedures This section provides information on setting up and using a printer: Connecting a Printer to the Instrument The instrument supports parallel and serial printers as well as USB and LAN connected printers. To connect your printer to the instrument: 1 Turn off power to the instrument and printer before connecting. 2 Connect the printer to the instrument with the cable that came with your printer.
10 Customizing the Instrument Removing a Printer When you disconnect a printer from the instrument and no longer intend to use it, you can remove the printer driver and increase the free disk space available on the instrument's internal hard disk drive. To remove a printer driver: 1 On the File menu, click Print Setup, then Configure Printer.... This opens the Printers dialog box. 2 Click the printer driver you want to remove. 3 In the File menu in the Printers dialog box, select Delete.
Customizing the Instrument 10 Configuring Printer Properties You can change the properties of a selected printer: 1 On the File menu, click Print Setup 2 Click Configure Printer.... This opens the Printers dialog box 3 In the File menu in the Printers dialog box, select from the following options: – Server Properties Click this option to change such items as printer margins, spooler warnings and alarms, and port configurations.
10 Customizing the Instrument File Management File Management - Procedures This section provides information on saving and recalling instrument states, log files, patterns, and screenshots. You can save the files on the local disk. Note that the local disk must not be repartitioned in a Windows 2000 environment. You can also plug an USB stick into the rear of the instrument for saving and recalling files. No driver is required because Windows has a built-in driver that supports USB memory drives.
Customizing the Instrument NOTE NOTE 10 Ensure your instrument state files have the extension, *.btz. If you do not specify an extension or use a different extension, the instrument may not recognize the file as an instrument state file. The default pathname for user defined instrument states is C:\ \Settings The number of instrument states you can save is limited only by the available space on the internal hard disk or floppy disk.
10 Customizing the Instrument 2 Open a spreadsheet application on your PC. 3 Import the measurement log file. If your application has an import wizard, you may need to indicate that data is delimited with commas. 4 Once the file has been imported, you may need to resize columns. The imported log file should appear similar to the following example. NOTE During measurement logging, the Serial BERT logs data in ten-second intervals. Your log file may be missing up to the last ten seconds of data.
10 Customizing the Instrument Table 38 Instantaneous Cumulative ED CLOCK LOSS ERROR DECISECONDS Cumulative DATA LOSS ERROR FREE DECISECONDS Cumulative SYNC LOSS SECONDS Cumulative BURST STATUS Cumulative BURST SYNC RATIO Cumulative TOTAL BURST COUNT Cumulative BAD BURST COUNT Cumulative Saving and Recalling Patterns You can save your pattern in either of the following ways: • Click the Save icon on the Pattern Editor toolbar.
10 Customizing the Instrument Saving To save an image of the display screen, do the following: 1 In the File menu, select Save. 2 Select Save Screen Capture.... This opens a dialog box, which allows you to save the image as a *.bmp (256 color bitmap file format) file. Recalling Follow the steps below to view screen images on an external PC: 1 Copy the screen image file from the Serial BERT to your PC. By default, image files are saved in the folder C:\\User.
Customizing the Instrument 10 Self Test Self Test - Concepts The self test checks specific hardware components for basic functionality. At initialization, the following self tests are automatically run: NOTE • Pattern Generator Power Up Test • Error Detector Power Up Test The current system state is saved and the instrument is re-initialized before performing the selected tests. The saved system state is restored upon completion of the self test.
10 Customizing the Instrument This opens the Self Test Options dialog box. 2 Select the desired tests and click the Run Test button. 3 After the self test has finished, the results are displayed next to each selected test. 4 If a test failed, you can view detailed error messages by clicking the respective Message button. Self Test - Reference This section contains information about the Self Test Options dialog box used for setting up and running a self test and for viewing the results.
Customizing the Instrument 10 Select All/Unselect All Use the Select All button to select all tests in the Self Test Options dialog box. After clicking the button, it becomes the Unselect All button. Clicking the button again will then unselect all tests and display the Select All button again. Run Test After selecting at least one test, use this button to start the test. If you selected tests, that need a longer time to complete, a message box opens. Confirm your selection by clicking OK.
10 Customizing the Instrument 566 Agilent J-BERT N4903B High-Performance Serial BERT
Index 0/1 Decision Threshold, 197 0/1 Threshold, 198 0/1 THReshold Center button, 200 0V (Disable) button, 138 10 MHz Ref Input, 122 8B/10B Comparison Results Analyzing, 365 Monitoring, 364 8B/10B Comparison Results Window, 367 A Absolute threshold, 312 AC Coupling, 130 Accumulated Measurements, 376 Results, 378 Running, 377 Accumulated Results, 380 Accumulation Parameters, 382 Accumulation period, 222 Adjust Output Levels, 132 Advanced Analysis, 227 Agilent Recovery System, 525 Alternating Patterns, 161 A
Index Constant Errors, 473 Contour, 320 Contour plot, 288 Creating New Pattern, 71 Criteria for Moving to the Next Measurement Point, 312 Crossover, 127 CrossTalk, 107 Crosstalk, 400 D Data Center button, 200 Data In port, 169 Data Input Delay, 196 Data input delay, 199 Data Input Setup, 172 DATA LOSS, 471 Data Out port, 123 Data Termination, 137 Date and Time, 529 dBER vs.
Index Grid , 320 H Hardware-generated patterns, 61 High Level, 277 High Level Std. Dev.
Index Number of Valid Points, 313 Numerical Results Eye Opening, 301 Output Levels, 277 Output Timing, 259 Spectral Jitter, 359 O Offset threshold, 312 On-Screen Keyboard, 532 Optimal Sample Point Delay, 260, 263, 301 Optimal Sample Point Threshold, 301 Optimum Sampling Point, 196 Oscilloscope Connections, 32 Output blanking, 138 Output Level Parameters, 126 Output Levels measurement, 263 Output Protection, 124 Output Timing measurement, 241 Output Timing Measurement Optimization, 253 P Parameters Bounde
Index Q High Level Std.Dev, 281 Q Low Level, 281 Q Low Level Nr. Points, 282 Q Low Level R^2, 282 Q Low Level Std.Dev, 282 Q Optimum Threshold, 280 Q Residual BER, 280 Q-Factor Calculations mathematics, 283 Notes, 285 QBER vs.
Index View tab Fast Eye Mask, 314 Vlo, 136 Vof, 135 Voltage of measurement point, 313 W Web Server, 551 Z X Zero SubstitutionPattern zero substitution, 92 Zoom Function Eye Opening, 300 Output Level, 276 Output Timing, 259 Spectral Jitter, 359 Xover, 137 Waveform, 320 572 Agilent J-BERT N4903B High-Performance Serial BERT
