background image

PVA-3000 Reference Manual 

December 2, 2019

 

                                   

Sifos Technologies

  

Page 1 

 

 
 
 
 
 
 
 

Sifos Technologies 

  

PhyView Analyzer 

 

PVA-3000, PVA-3002 

 

 

 

 

Reference Manual 

 

Version 5.1 

 

December 2, 2019 

 
 

Copyright © 2013-2019 Sifos Technologies 

 

®

 

Summary of Contents for PhyView PVA-3000

Page 1: ...A 3000 Reference Manual December 2 2019 Sifos Technologies Page 1 Sifos Technologies PhyView Analyzer PVA 3000 PVA 3002 Reference Manual Version 5 1 December 2 2019 Copyright 2013 2019 Sifos Technologies ...

Page 2: ... user documentation Except as permitted by such License Agreement no part of this publication may be reproduced stored in a retrieval system or transmitted in any form or by any means without the prior written permission of Sifos Technologies Inc Information contained in the user documentation is subject to change without notice and does not represent a commitment on the part of Sifos Technologies...

Page 3: ... 12 Receiver Testing Metrology 19 1 3 13 Receiver Testing with Impairments 20 1 3 14 Reference Manual Overview 21 2 PhyView Analyzer Overview 23 2 1 Test Port Interfaces and Configurations 23 2 2 Test Port Resources 24 2 2 1 The PVA 3000 TEST PHY 24 2 2 2 PVA 3000 Meters Overview 25 2 2 3 PVA 3000 Meters and Link Rates 25 2 2 4 PVA 3000 Meters Configuration Maintenance 26 2 2 5 Triggered Meters 26...

Page 4: ...ation Menu 46 3 8 1 Link Config Menu Wide Controls 47 3 8 2 Link and Gigabit Configuration Sub Menus 47 3 8 3 Test Signal Sub Menu 48 3 8 4 Test PHY Sub Menu 48 3 8 5 Link Config Actions Sub Menu 49 3 9 Link Monitor Menu 49 3 9 1 Link Monitor Measurements 49 3 9 2 Link Partner Assessment 51 3 10 Meters Menu 52 3 10 1 Link Speed Configuration 52 3 10 2 Pair s Averages Configuration 52 3 10 3 PSD Fr...

Page 5: ...librations 80 4 9 DUT Transmitter Measurement Configurations and Queries 81 4 10 Link Status and DUT Receiver Measurements 85 4 11 PowerShell PSA Test Support Utilities 87 4 12 PowerShell Calibration Utilities 89 4 13 PowerShell Multi Pair Measurement Utilities 90 4 14 PowerShell Multi Port Measurement Utilities 91 4 15 PowerShell System Chassis Commands 92 4 16 PHY Performance Test Suite Commands...

Page 6: ...PVA 3000 125 6 1 Multiple Port Configuration and Operation 125 6 1 1 Broadcast Test Port Configuration 125 6 1 2 Event Triggered Multi Port Measurements 126 6 1 3 Multi Port Packet Transmission and Analysis 127 127 6 2 Merging the PowerShell PSA API into TCL and Wish Shells 128 6 2 1 Initial Connection Dialog Control PowerShell Tcl 128 6 2 2 Initial Connection Dialog Control PowerShell Wish and PS...

Page 7: ...odes not available to commercially packaged 10 100 1000BaseT ports At the other extreme link ups and packet flows can be achieved rather easily and inexpensively However this type testing seldom quantifies any characteristic of performance while relying on unknowns testing unknowns under uncharacterized conditions The PhyView Analyzer fills this wide coverage and productivity gap Measurements such...

Page 8: ...3 strict rules regarding backward compatibility and technology coexistence have assured smooth and seamless transitions to newer and faster networking technologies The introduction of Auto Negotiation with 100BaseTx enabled link partners to determine technology compatibility so that links start up with the highest possible performance level Auto Negotiation has since evolved to resolve numerous li...

Page 9: ...dB 60MHz 21 5 dB 40MHz 9 5 dB 100MHz Loss Tolerance 11 5 dB 10MHz 24 dB 100 MHz 24 dB 100 MHz Mismatch Tolerance 15 15 dB R L 15 dB R L to 20MHz 8 dB R L 100MHz 15 dB R L to 20MHz 8 dB R L 100MHz NEXT Crosstalk Tolerance 26 dB 10MHz 27 1 dB 100MHz 27 1 dB 100MHz Noise Tolerance 264 mVpk below 15MHz 40mVpp below 100MHz 40mVpp below 100MHz Jitter Tolerance 7 nsec Pk Pk 1 4 nsec Pk Pk 1 4 nsec Pk Pk ...

Page 10: ... topics such as receiver error detection and indication criteria 1 2 4 Time Domain Measurements in PHY Conformance Testing Originally the baseband nature of 10BaseT signals meant that time domain analysis using oscilloscopes was the natural approach to qualifying transmitted signal integrity to the standard Most of the above listed transmitter parameters see Section 1 2 3 were a good fit with time...

Page 11: ...ctional in nature and does not directly measure any physical layer parameters Packet flow testing is attractive because it is easy and often low cost Since Ethernet links universally discard defective packets the core requirement is a device that can transmit receive and count Ethernet packets Commercial packet analyzers from manufacturers such as Spirent and Ixia have been widely deployed for yea...

Page 12: ...annel impairments it can expect to encounter while in service Summarizing packet flow testing has thrived because it is convenient low cost and because the traditional PHY compliance tests are both expensive and laborious The PhyView Analyzer was developed to provide much of the direct physical layer test coverage associated with traditional PHY compliance tests while providing all of the convenie...

Page 13: ...ed to the lower frequencies This is critical to how these technologies overcome the insertion loss of 100M cable channels Secondly it is clear that these spectra are largely not impacted by packet flow This is critical to assuring that IDLE links remain linked with no loss of receiver synchronization The 100BaseTx spectrum shows a distinct peak in the 16 MHz range then rolls off to 20 dB by 90 MHz...

Page 14: ... longer period of time given an all zero s encoded data pattern For example 100KHz would correspond to 10 sec 1 250 symbol periods or a data pattern of 125 bytes Power over Ethernet can worsen droop if the delivery of DC current is not evenly split across both conductors of a twisted pair This impairment referred to as DC Unbalance causes saturation in the transformer which in turns clips off low ...

Page 15: ...itude and lower rate jitter is easily tracked out while high rate or high frequency jitter may easily flow through the receiver and turn up as residual noise This is why the 1000BaseT standard places a 5KHz bandwidth on the specification of peak peak jitter In 1000BaseT receivers have to deal with a special form of quasi random noise in the form of echoed transmitted signals and local crosstalk fr...

Page 16: ...sec This equates to just over 6 symbol periods This requirement exists to assure receivers can properly reassemble data streams split amongst the four pairs 1 3 7 Interface Verification Overview The PhyView Analyzer will verify two types of interface characteristics while linked at 1000BaseT Bulk Return Loss or Echo Bulk Pair Pair Crosstalk These characteristics are separated from Transmitter and ...

Page 17: ...Port Under Test Impedance Expected Return Loss Applicable Frequencies 100 Frequency Selective 83 5 20 9 dB Frequency Range 0 40 MHz 117 22 1 dB 49 5 9 4 dB 100MHz 197 9 7 dB 100 Wideband 82 1 20 1 dB Bulk Return Loss Limits that assure MDI Return Loss specs across 85 to 115 119 21 2 dB Using this table the wideband criteria for a compliant interface would be in the 20 to 21 dB Return Loss region T...

Page 18: ...etical performance limits need to be discerned from other specifications In a 10 100BaseT channel the only crosstalk that matters is Pair 2 3 since pairs 1 and 4 are not utilized for data communication In 1000BaseT all four pairs are utilized in a full duplex fashion and any crosstalk that can be measured on any pair will create an uncorrectable distortion at the receiving end of the link If the l...

Page 19: ...eceiver test methodology is largely the responsibility of Ethernet port designers and implementers The task of modeling all of the possible signal and channel impairments while assessing bit error rate performance can be daunting and the exact criteria for success pass or fail can be ambiguous 1 3 12 Receiver Testing Metrology The PhyView Analyzer offers two distinct measurement techniques for rec...

Page 20: ...nalyzer test port are restricted to MAC frames with programmable address size packet gap and repeating 4 byte payload This technique therefore requires that the port under test must forward any successfully received packets to another Ethernet port so that they may be returned to another PhyView Analyzer test port Hence the device being tested must be a multi port device with layer 2 bridging capa...

Page 21: ... 14 Reference Manual Overview The remainder of this manual will familiarize users with the various resources available within each test port and the various means of accessing and using those resources for link configurations impairment configurations measurements calibrations and fully automated test suites Section 2 will introduce basic test port architecture describe each testing resource in de...

Page 22: ...PVA 3000 Reference Manual December 2 2019 Sifos Technologies Page 22 ...

Page 23: ...aling that passes through the TEST interface will always experience a 2 6dB power loss as it traverses a passive RF 4 Pair Coupler see Figure 2 1 This loss is frequency flat through 100MHz and is implemented with 100 matching impedance so as not to disturb any 100 terminations It is uniformly applied to each of the four LAN pairs The PVA Switch in Figure 2 1 then determines whether the TEST interf...

Page 24: ...hat is essential to each of the following PhyView Analyzer metrologies Power Spectral Distortion PSD Tx Level Wideband Power SNR Echo Bulk Return Loss Bulk Crosstalk Timing Skew Link Monitor Link Partner Capabilities Through a cooperative arrangement between the TEST PHY device supplier and Sifos Technologies the PVA is able to access and manipulate numerous proprietary features within these devic...

Page 25: ... YES SNR Signal Noise Ratio Residual Distortion to 36 dB See Section 1 3 4 100 1000 dB per pair Link Rate 100 or 1000 Gigabit Mode Master or Slave or AUTO Pair Number 1 2 3 4 Averaging 1 to 64 Triggering Immediate or Event w timeout YES YES YES Echo Bulk Return Loss Wideband Echo Response to 26 dB See Section 1 3 8 1000 dB per pair Pair Number 1 2 3 4 Averaging 1 to 64 Triggering Immediate or Even...

Page 26: ...ggering Link Monitor PSD SNR Echo Response Bulk Crosstalk The default option is generally Immediate Triggering This means that when a measurement is started it will run until completion Completion will be marked by successful measurement or by notification of a measurement problem such as a DOWN link From PVA Interactive see Section 3 immediate triggering is the only supported trigger mode for all...

Page 27: ...est cables with each PVA test port and strongly recommends the exclusive use of these test cables PVA 3000 calibrations are simple to execute Each test port is considered to have a calibration partner test port which by default is the other test port on the same PVA 3102 test blade or on the PVA 3002 instrument So test port 2 is the calibration partner for test port 1 and vice versa see Figure 2 4...

Page 28: ...p to 100MHz with worst case return loss 14 dB The impairment is also designed to approximate the linear phase vs frequency response of a cable straight line response up to 100MHz The Line Loss impairment may be applied uniformly to pairs 1 4 or may be applied selectively to pairs 1 2 or pairs 3 4 This allows an outgoing 10 100BaseT signal to experience the impairment without a commensurate impairm...

Page 29: ... to within 100 ppm of nominal frequency that is 10MHz for 10BaseT and 125 Msymbol sec for 1000BaseT 100BaseTx is more discerning with a specification of 50 ppm for a 125 Msymbol sec baud rate The TEST PHY transmit frequency with zero offset will be within 10 ppm of absolute nominal frequency TEST PHY transmit timing may be deliberately offset by any of the following values 115ppm 100ppm 50ppm 50pp...

Page 30: ...mplitudes Within the PVA test port Tx Slew may be varied so that a 100BaseTx signal can transition anywhere between 2 8 nsec and 4 8 nsec based on 0 8V transitions The Tx Slew rate is by default managed automatically by the TEST PHY and may be configured to automatic control at any time When Tx Slew is reduced i e longer rise fall time frequency components in the transmitted signal above 70 to 75 ...

Page 31: ...merous extensions specific to the PhyView Analyzer and the PowerSync Analyzer PowerShell PSA is an interpreted command and query environment supporting interactive control of and automated program development for the PhyView instrument and the PowerSync Analyzer chassis PowerShell PSA is the subject of Section 4 of this manual Some test engineers will want to integrate the PowerShell API into pre ...

Page 32: ...e Chassis Selection menus in PVA Interactive and PowerShell Wish Default_Conf_Test_List_AF Not applicable to PVA 3000 Default_Conf_Test_List_AT Not applicable to PVA 3000 Default_Test_Results_Path Path to where test results from the PHY Performance Test Suite sequencer will be stored This has the default value of c Program Files Sifos PSA1200 Results chassis IP Address when PSA software is initial...

Page 33: ...ry Files C Program Files Sifos PSA1200 Note On 64 Bit Windows C Program Files will become C Program Files x86 Note International versions of Windows may choose a different name for Program Files PowerShell PSA Script Libraries PowerShell PSA Wish and PowerShell PSA Tcl Executables PSA Initialization Script PowerShell PSA Wish PowerShell PSA Tcl resource files documentation Various PSA reference do...

Page 34: ...T local Configuration Files including Config env environment file sub directory Cal Sub directories by chassis IP address that store all calibration files for PSD Bulk Return Loss and Bulk Crosstalk meters Results PVA PSA Test Report Files PVA specific report templates include pva_report xlsm pvaDCunbalance_report xlsm and psd_trace xlsm PSA specific report templates include psa_report xlsm mp_rep...

Page 35: ...nsiderations Certain aspects of the PhyView Analyzer and its host based software PowerShell PSA PVA Interactive rely on a low latency network connection and uninterrupted run time execution in a host computer Applications such as the PHY Performance Test Suite and pva_veriphy can be impacted in subtle but detrimental ways given very long and typically very intermittent network connection latencies...

Page 36: ...test port or a device under test With small patch cables any fixed wideband 100 padding between 1 dB and 15 dB in 1 dB steps can be applied The PVA LI4 will help verify Bulk Return Loss and Bulk Crosstalk measurements Return losses are synthesized from both low and high impedance mismatches The IEEE 802 3 MDI return loss limit line impairment synthesizes two return loss profiles that reflect how a...

Page 37: ...Link Stability Jittered or Offset Xmit Signal Packet Loss Jittered or Offset Xmit Signal Test PHY Jitter Offset Synthesis Line Emulator AND OR Alien Crosstalk OR Passive Mismatch Link Stability Multi Impairment Packet Loss Multi Impairment Transmit Receive Tests with PoE Impairments PVA 3102 Test Port to PSA 3102 OUT Port Connection s All Above Transmitter Tests 1000 100BaseT see above All Above R...

Page 38: ...e 4 75 MHz 1000Base T Pairs 1 to 4 Measurement Floor 26 dB Accuracy Repeatability normalized to calibration 1 0 dB 20 dB 1 5 dB 20 dB Ideal Gigabit Config DUT MDI MASTER Bulk Crosstalk Response 4 75 MHz 1000Base T Pair Combo s 1 2 1 3 1 4 2 3 2 4 3 4 Measurement Floor 39 dB Accuracy Repeatability normalized to calibration 1 0 dB 32 dB 1 5 dB 32 dB Ideal Gigabit Config DUT MDI MASTER Pair Timing Sk...

Page 39: ...Amplitude Accuracy 0 9dB typical Frequency Shaping 100BaseTx Spectrum Impairment Application Pairs 1 3 2 4 or 1 2 3 4 Transmit Frequency Offset Test Port with Test PHY Connected Nominal Transmit Frequency 125 MHz 10 ppm Nominal Transmit Duty Cycle 50 12 5 Fixed Frequency Offsets 115 100 50 0 50 100 115 ppm Transmit Frequency Jitter Test Port with Test PHY Connected Jitter Magnitude Range relative ...

Page 40: ...85 C Operating Humidity 5 to 95 RH Non Condensing 2 7 10 Certifications Description North America Europe International Emissions FCC Part 15 Class A Meets EN55011 VCCI AS NZS 3548 ICES 001 Safety CSA Listed CSA22 2 No 61010 Meets EN61010 1 General Certification Low Voltage Directive 2014 35 EU Electromagnetic Compatibility Directive 2014 30 EU CE Marking Directive 93 68 EEC FCC Statement This equi...

Page 41: ... time 3 2 Initial PVA Connection When PVA Interactive is opened the user will be prompted to select an instrument to which PVA Interactive will initially connect see Figure 3 2 The default selection will be the most recently connected PhyView Analyzer This brief dialog assures that various users on a common network sharing multiple instruments can connect to an instrument without interfering with ...

Page 42: ...bdirectories that are specific to a particular device under test DUT Without using DUT specific configurations all test reporting will appear in a default subdirectory named for the PVA chassis address DUT specific configurations files also contain other information that may be used by future versions of PVA software A DUT specific configuration file can be created saved and loaded using the File ...

Page 43: ...ned It will also be updated if a selected address is found not to exist on the network when a connection is attempted Users may add a new instrument to the list by entering the associated IP address in the Connect To entry field then pressing Add Chassis If the entered address is discovered on the network it will be added to the list and become available for future selection It must then be select...

Page 44: ...ee Figure 3 11 At this point pressing the Apply Selected Port Settings button will effect the selected configuration setting in the current test port if the corresponding checkbox for Port Connections and Port Impairments is also selected The three checkboxes Port Connections Port Impairments and Active Impairments serve as filters for which settings can be applied or replicated to other test port...

Page 45: ... Apply operations are not in any way governed by the Active Impairments checkbox in the Actions frame discussed earlier in Section 3 6 2 The Remove button in the Random Noise frame shuts off the noise insertion at the TEST interface As mentioned earlier the 12dB Mismatch setting can be used to inhibit noise impairment on two pairs by substituting the Mismatch impairment on two pairs The actual sta...

Page 46: ...e menus This dialog indicates the current Copy from port that is the port from which settings are to be copied The menu also includes enabled checkboxes for every other available PVA test port in the currently connected instrument To copy settings to particular PVA test ports the slot port check boxes for those ports are selected and the DONE button in pressed Configuration settings can be copied ...

Page 47: ...l the Reset Test PHY operation can be broadcast to all PhyView Analyzer test ports in the currently connected instrument by selecting the Reset All Ports checkbox prior to pressing the Reset Test PHY button 3 8 2 Link and Gigabit Configuration Sub Menus Each PhyView Analyzer test port offers a wide range of link configurations that can be used primarily to force a device under test DUT into a desi...

Page 48: ...signal 3 for jitter all four pairs Gigabit Random 4 802 3 test signal 4 for distortion all four pairs 100BaseTx Idle MLT 3 signal pair 2 These signals can be useful for evaluating TEST PHY transmit characteristics at the TEST interface For example they will impacted by Transmit Level and Transmit Slew configuration settings Note that they are also impacted by the 4 pair RF Coupler loss described e...

Page 49: ...tton opens the port replication dialog explained earlier in Section 3 7 3 9 Link Monitor Menu The Link Monitor tab menu see Figure 3 18 serves three purposes 1 Configure Link Monitor 5 Perform Link Monitor Measurements 6 Perform Link Partner Assessments The configurations performed in the Link Monitor menu only apply to the Link Monitor metering As with previously described menus there is a Read S...

Page 50: ...f 100 samples collected with 50 msec interval would total 100 samples over 100 50msec 5 seconds When any Link Monitor setting is selected the Apply Settings button will annunciate red see Figure 3 19 indicating that the new configuration will be applied when the Apply Settings button is pressed It should be noted that if a Link Monitor measurement is initiated using the Measure button while a conf...

Page 51: ...eg is COMPLETE or NOT_COMPLETE DET_FAULT Indicates if link rate is properly determined RATE_RESOLVED or not DET_FAULT MASTER_FAULT Indicate problem FAULT with gigabit auto neg given local port configured as MASTER SLAVE_FAULT Indicate problem FAULT with gigabit auto neg given local port configured as SLAVE AUTO_MS_FAULT Indicate problem FAULT with gigabit auto neg given local port configured to au...

Page 52: ... disabled Link Speed is an independent configuration setting to the SNR and PSD meters meaning that these meters will attempt to force the desired link speed as part of the measurement setup So if a test port is linked at 100BaseT to a port under test and a PSD measurement is configured for 1000 then the test port will force a new 1000BaseT link before performing the PSD measurement s Unlike SNR a...

Page 53: ...cally By default the PSD meter will automatically pop open windows with PSD traces of power versus frequency for each pair included in the measurement These traces are color coded by TIA wire color convention see Section 1 3 10 Figure 3 25 shows a PSD trace window for a 1 100 MHz PSD measurement on Pair 1 PSD trace windows can be inhibited by selecting the Inhibit PSD Traces checkbox Each PSD trac...

Page 54: ... support packet flow testing is to allow PSA software to automatically assign addresses such that each test port has a unique and valid unicast source address This is accomplished by selecting the Auto Address All Ports checkbox then pressing the AUTO Address button Furthermore the addresses scheme created can be stored in non volatile memory in each test port by selecting the Retain Address es No...

Page 55: ...ave the values of IDLE Packet transmission halted ACTIVE_CONT Transmitting packets continuously ACTIVE_BURST Transmitting packet burst UNLINKED Packet transmission prohibited by down link 3 11 3 Packet Reception and Counting The test port TEST PHY can also receive and count MAC frames using the Received Packets sub menu Within this menu there are three pushbutton controls and three displays The Re...

Page 56: ...diagrammed in Figure 3 29 would be 1 Configure AUTO Addressing on all PVA test ports 2 Configure impairments on PVA test port 1 3 Start packet count on PVA test port 2 4 Transmit fixed burst of packets from PVA test port 1 5 Wait for burst completion on PVA test port 1 6 Complete packet count on PVA test port 2 A successful result would then be a match of packets counted to the packet burst size t...

Page 57: ...seT receiver stress testing 5 pva_rx_100 Automated 100BaseTx receiver stress testing and analysis 6 pva_rx_1000 Automated 1000BaseT receiver stress testing and analysis These tests are described in greater detail in Section 5 3 12 1 Selecting and Executing Individual Tests The PHY Tests sub menu is utilized for two purposes Selecting individual tests to execute and report in the Results display se...

Page 58: ... a sequence and will automatically analyze all test data on a per test port basis relative to test limits for Excellent Marginal or Problem performance Results are generally color coded accordingly The Spreadsheet report also synthesizes color coded graphs for many parameters and is an extremely productive tool for rapidly assessing any performance problems Each test port is reported on a separate...

Page 59: ...PSE that uses 2 event classification and Type 2 LLDP is a 30 watt capable PSE that uses LLDP for powered device classification Note Type 2 LLDP specification will required that the PSA instrument be enabled for the LLDP emulation feature Max Unbalance mA This is the maximum amount of DC Unbalance current that will be created as the DC Unbalance Analyzer runs Measurements will be conducted at vario...

Page 60: ...on of sample acquisition and analysis that can be repeated and averaged in order to remove effects of noise and short term digital signal processing artifacts In general more averaging should produce better statistical accuracy with a small time penalty It is recommended that calibrations generally be done with high averaging that is either 64 or 32 In particular if PSD will be measured in frequen...

Page 61: ...opened 3 15 PVA Help Menus From the Help drop down there are three options available 1 PVA Config 8 PSA Features 9 About The PVA Config option opens a window see Figure 3 41 with the test blade configuration of the currently connected PVA instrument This configuration shows populated test slots test blade hardware version and test blade firmware version by test port Unused test slots will report E...

Page 62: ...t port needs to be set for MDI polarity Menu Task Screen Shot Port Select Select the test port to configure Click on Port 1 in Slot 1 on the Slot Port Select menu Port Config Configure the port to add the 100m Line Impairment on Pair 2 which will be used by the test port for Tx when the polarity is set to MDI Select the Port Config Tab Select Impair 1 2 under 100 Meter Cat5e Click Apply Selected P...

Page 63: ...port to add the 100m Line Impairment on all pairs and set up 115 ppm timing offset Select the Port Config Tab Select 115 ppm under Transmit Timing Offset Click Apply Offset Select Impair All under 100 Meter Cat5e Click Apply Selected Port Settings Link Config Configure the test port speed polarity and duplex settings Select the Link Config Tab Select 1000BaseT under Link Speed Select Automatic und...

Page 64: ...elect the port to configure Click on Port 1 in Slot 1 on the Slot Port Select menu Calibrations Perform Calibrations Select the Calibrations Tab Select PSD de select Echo and Xtalk under Meters Select Selected Port under Ports Select 100BaseTX and 1000BaseT under Link Speed Select 64 or desired value under Averages Click Run Calibrations Verify Test Port Connections Test cable should connect the t...

Page 65: ...ion Select the Link Config Tab Click Reset Test PHY Port Select Select the test port to use for measurements Click on Port 1 in Slot 1 on the Slot Port Select menu Link Config Configure the test port speed polarity and duplex settings Select the Link Config Tab Select Nominal Fixed under Test PHY Click Apply Rx Level Select 1000BaseT under Link Speed Select Automatic under Link Duplex Select Autom...

Page 66: ...ir All under 100 Meter Cat5e Click Apply Selected Port Settings Link Config Configure the Test Port speed polarity and duplex settings Select the Link Config Tab Select 100BaseTX under Link Speed Click Arm Configuration Click RE LINK Meters Measure Power with line impairment switched in Select the Meters Tab Select 100 under Link Speed Select Power under Meter Type Click Measure Port Config Config...

Page 67: ... Select menu Packets Start the Received Packets Counter Select the Packets Tab Click Clear Start Count under Received Packets Port Select Select the port to transmit packets Click on Port 1 in Slot 1 on the Slot Port Select menu Packets Configure and Transmit Packets Select 524288 under Frame Count Select Wait for Completion then Click Start Packets under Packet Transmission Port Select Select the...

Page 68: ...rt 1 1 speed and timing mode Select the Link Config Tab Select 1000BaseT under Link Speed Click Arm Configuration Select Slave under Timing Mode Click Arm Configuration Click RE LINK Verify Link to 1000BaseT Port Select Select the port to count received packets Click on Port 2 in Slot 1 on the Slot Port Select menu Link Config Configure the test port 1 2 speed and timing mode Select 1000BaseT unde...

Page 69: ... and the network connection between that host PC and the PVA instrument to run automated test applications that rely upon some degree of real time interaction and run time execution Depending on many factors including properties of the device being tested network connection latencies exceeding 20 msec or host preemption of executing test programs for periods of 50 msec or more may have detrimental...

Page 70: ...PVA 3000 Reference Manual December 2 2019 Sifos Technologies Page 70 ...

Page 71: ... It supports interactive user prompting via standard input from a script The Tcl Shell will not support Tk graphical interface extensions The Wish Shell see Figure 4 3 is a newer shell designed to enable GUI application development PowerShell Wish uses the top level window created by the Wish Shell to present a Close PowerShell button since the console window opened by Wish is subsidiary to the to...

Page 72: ...rs If no valid addresses are discovered PowerShell Tcl will open to Demo Mode see Section 4 2 1 The PSA Chassis Selection dialog used by PowerShell Wish will remain open indefinitely until a valid and available PVA or PSA instrument address is entered or selected If no such addresses exist the only way to open PowerShell Wish is to select Demo Mode and enter with demo mode activated see Sections 3...

Page 73: ...port is always displayed in the PowerShell command prompt The command line will reject values of port that are not available in the current selected PSA chassis See Section 6 1 1 for description of 99 99 99 1 and 99 2 broadcast configuration ports Indicates logical OR Indicates logical AND stat A special optional query available to many commands to report Operational State and in some cases to ret...

Page 74: ...version Outputs current version of PowerShell PVA software Module Versions pva_exists alias to psa_exists ipaddress 1 PSA found 0 No PSA found Returns 1 if PSA is found on network and 0 if PSA is not found on the network ipaddress The IP address of a PSA chassis pva_dut dut_name Loads the DUT specific configuration file for a specific DUT type The DUT specific configuration file specifies the foll...

Page 75: ...THROUGH Configures switching to either terminate to Test PHY or route Test Port through to Auxiliary Port phy Terminate the test port to the Test PHY wake up default through Connect the Test Port via low loss coupler to the Auxiliary Port pva_line port impair normal pair12 pair34 all IMPAIRED NORMAL ALL PAIR12 PAIR34 Inserts or removes a worst case IEEE 802 3 insertion loss 100M Cat5e passive line...

Page 76: ... to 40mVpp noise on the Test Port output high Configures the amplitude noise source to 4 dB relative to 40mVpp noise dB noisepwr Configures the amplitude noise at the Test Port from 6 dB to 21 5 dB 0 5 dB steps relative to 40mVpp wake up default is 0 dB pva_tx_clk port normal offset ppm_offset jitter low std high dB jitter_pwr NORMAL OFFSET ppm_offset JITTER jitter_pwr Configures Test PHY transmit...

Page 77: ... timing mode following the reset pva_relink port Causes an immediate re link to occur using any Auto Negotiation configurations previously specified and returns link status Accepts broadcast port 99 99 pva_speed port auto 1000 100 10 10 100 duplex auto full half arm LINKED UNLINKED ARMED Configures Test PHY to advertise selected speed and link duplex mode through either an immediate re link defaul...

Page 78: ...ls and returns ARMED if arm specified Accepts broadcast port 99 99 only in combination with arm argument auto Configure Master Slave relationship during link up master Force local Test PHY to be Master slave Force local Test PHY to be Slave arm Configure 1000BaseT timing mode but suppress the link up to effect this until a later re link operation Enables use with port 99 99 AUTO MASTER SLAVE pva_t...

Page 79: ... a specified level between 88 nominal levels and 112 nominal levels Nominal is 6 for 10 1000BaseT 7 for 100BaseT auto Configures the Test PHY to automatically set transmit level gain on each link up wake up default This control will also restore Rx Gain and Tx Slew to their respective AUTO modes as well low Configures Test PHY for a low transmit level 3 out of 10 med Configures Test PHY for a medi...

Page 80: ...air If link is 100BaseT only Pair 2 or 3 may be specified and depending on pair number link will be forced to MDI_X or MDI respectively This configuration is not maintained if PSA Software is closed PSA Software wake up default 3 link 100 Specifies a calibration using 100BaseT This configuration is not maintained if PSA Software is closed PSA Software wake up default link 1000 Specifies a calibrat...

Page 81: ...ng commands and queries are specific to transmitter and interface measurements described earlier in Sections 1 3 and 2 2 2 All measurement queries require link to the port under test Certain measurements such as PSD Echo and Xtalk will attempt to force re links to specific link configurations required by those measurements If a channel is unlinked upon starting any measurement or if the channel fa...

Page 82: ...eout 10 specifies that an Event Triggered measurement should time out if no trigger is received within 10 seconds following stat query Test Port wakeup default timeout 100 specifies that an Event Triggered measurement should time out if no trigger is received within 100 seconds following stat query Trigger_Mode off ext pairNo Link_Speed 100 1000 nAvgs Timeout 10 100 pva_echo port trig off ext pair...

Page 83: ... and depending on pair number link will be forced to MDI_X or MDI respectively This configuration is not maintained if PSA Software is closed PSA Software wake up default 3 link 100 Specifies a measurement using 100BaseT This configuration is not maintained if PSA Software is closed PSA Software wake up default link 1000 Specifies a measurement using 1000BaseT This configuration is not maintained ...

Page 84: ...diately upon stat query Test Port wake up default trig ext Configures calibration to start upon receipt of an Event Trigger following a stat query pairs pairGrp Specifies 1000BaseT LAN PAIR GROUP to measure Range is 12 13 14 23 24 and 34 where Pair 1 is TIA blue pair Pair 2 is TIA orange pair Pair 3 is TIA green pair and Pair 4 is the TIA brown pair This configuration is not maintained if PSA Soft...

Page 85: ...e used as an alternative to packet flow to assess link health Measurement may be either immediate or event triggered Configurations are retained in test port trig off Configures measurement to start immediately upon stat query Test Port wake up default trig ext Configures measurement to start upon receipt of the first Event Trigger following a stat query sample link Count instantaneous link status...

Page 86: ...ied See Note above store Retain MAC addresses in non volatile memory pva_tx_pkt port size frameSize rate line med slow count frameCount payload hexChars start stop stat Trigger_Mode off ext frameSize line med slow frameCount hexChars Configures transmission parameters for simple internally generated MAC frames from one or all test ports Initiates and terminates transmission of MAC frames given spe...

Page 87: ...LL 10BaseT NO HALF FULL HALF FULL 100BaseT4 NO YES PAUSE NO RESPOND XMIT RESPOND XMIT LINK_OK YES FAULT GIG_ERR N A Rx_OK NO YES N A MDI_MDI X AUTO MDI_ONLY MDI X_ONLY MASTER_SLAVE AUTO MASTER_ONLY SLAVE_ONLY N A NLP_LINK LINKED UNLINKED diag Adds additional diagnostic information including auto negotiation status Fault indications for link rate resolution gigabit timing resolution faults for forc...

Page 88: ...ase T training time NORMAL is nominal performance PRE_DISTORT indicates 1000Base T link up is unreliable without connecting Line Loss impairment an indictor of receiver AGC issues NO_RELINK indicates a 10 100 port fails to reliably respond to link partner induced Restart Auto Neg operation an indicator of Auto Neg issues NORMAL PRE_DISTORT NO_RELINK NOT_CONNECTED pva_linkup_time port res 20 50 10 ...

Page 89: ...ation partner on same test slot Default is current slot port May also specify 99 1 to convey all Port 1 s on all test blades and 99 2 to convey all Port 2 s on all test blades measurements Measurement types to calibrate Options are psd power spectral distortion echo bulk return loss and or xtalk bulk crosstalk Echo and Xtalk are calibrated at 1000BaseT only If no measurements are selected pva_cals...

Page 90: ... averages to use in measurement Range is 1 to 64 Default is 16 nAvgs must be between 48 and 64 if fstart 1 MHz plot Produce graphical frequency vs power plots of each pair s PSD measurement Must use PowerShell WISH for this feature Will erase any previously displayed PSD plots color assign a non default graph color scheme of classic sky rain sand desert or mellow Specify pair to configure default ...

Page 91: ...z to 100 MHz Default is 100 MHz avg nAvgs Specifies non default number of averages to use in measurement Range is 1 to 64 Default is 16 nAvgs must be between 48 and 64 if fstart 1 MHz plot Produce graphical frequency vs power plots of each port s PSD measurement Must use PowerShell WISH for this feature Traces will accumulate up to 10 ports then start overwriting after that color assign a non defa...

Page 92: ...PSA1200 chassis controller to output or input external trigger signals Default wake up mode is output in Trig port on PSA1200 Controller becomes an input to the external trigger bus out Trig port on the PSA1200 Controller becomes and output of the external trigger signal psa_setipaddress ipaddress mask mask gateway gateway Current IP Address Mask and Gateway Address Installs a new fixed IP address...

Page 93: ...download immediately to selected port Useful for automated updating to more than one port psa_enable feature_code enable disable Serial Number Feature Info Command is used for querying Serial Number and Features from a PSA Controller Also used to enable and disable features given a proper code NOTE This command will run from either PowerShell Wish or PowerShell Tcl and requires PSA Software Versio...

Page 94: ...f 20 seconds per impairment condition then report UP if link is 99 available or DOWN if less available Requires just a single DUT connection Packet Flow will report as percent of line rate packets received and forwarded Packet flow requires two test port connections to a switch or hub type of DUT unless loopback is specified link_mon Assess stressed receiver performance using Link Monitor This is ...

Page 95: ...f Link Monitor operates as Link Viability indicator there is no assurance that an UP indication means the port will forward some or all traffic A DOWN indication means it will not forward all traffic If Link Monitor operates as a Link Stability indication link stability should roughly correspond to hypothetical packet flow packets Assess stressed receiver performance using Packet Flow This require...

Page 96: ... Packet Flow will report as percent of line rate packets received and forwarded Packet Flow always requires two test port connections to a switch or hub type of DUT link_mon Assess stressed receiver performance using Link Monitor This is the default meter If Link Monitor operates as Link Viability indicator there is no assurance that an UP indication means the port will forward some or all traffic...

Page 97: ... receive and count packets This parameter must be furnished to the command pkt_count Count of packets to transmit in a burst Range is 32K 128K 512K and 1024K Default is 1024K packets pkt_size Size of packets to transmit Range is 60 to 1512 bytes in 4 byte increments Default is 60 bytes line Transmit packets at full line rate 96 sec gap This is the default rate med Transmit packets with 576 nsec pa...

Page 98: ...le name with txt extension The report will reside in a directory defined by the PowerShell global psaResultsPath n file_name Specify a non default file name for text file test results This name should be the full absolute path and file name EXCLUDING the txt file extension dir ext Specify a 1 6 character alphanumeric extension to the present reporting directory path that will be applied to just th...

Page 99: ...of physical layer performance The pva_verify application is designed to test DUT ports quickly in contrast to the PHY Performance Test Suite which offers a much more thorough test that is sequenced one port at a time Section 6 4 describes the pva_verify multi port test application in greater detail Command Command Parameters Returned Parameters pva_veriphy port1 port2 portN 10 100 1000 dce dte abo...

Page 100: ...ons Link success will be verified using pva_rx_pwr meter and pva_link_mon meter over a 5 second period Hint To force the DUT into MDI X polarity the test port needs to be set for MDI polarity Expected Result LINKED Speed 100 MDI Rx_Power_3 Example 1 script Using PVA 3102 in slot port 1 1 set port 1 1 Add the 100m Line Impairment on Pair 2 which will be used by the Test Port for Tx when the polarit...

Page 101: ...port needs to be set for Master Ideal Result LINKED Speed 1000 MASTER Count 100 Remote_Gig_OK Actual Result Count 78 Example 2 script Using PVA 3102 in slot port 1 1 set port 1 1 Add the 100m Line Impairment on all Pairs pva_line port impair all Add a 115ppm offset Impairment to the clock pva_tx_clk port jitter off offset 115 Configure the Test Port for 1000Base T duplex polarity auto Master pva_s...

Page 102: ... cable PSD will be evaluated over 1 to 100MHz with 16 averages Example 3 script Using PVA 3102 in slot port 1 1 Perform the calibration set rslt pva_cals 1 1 psd 100 1000 Figure 4 7 Exercise 3 Script Execution Example 4 script Using PVA 3102 in slot port 1 1 set port 1 1 Make sure that the cal partner is in a default state pva_reset calport port Configure the Test Port for 1000Base T duplex polari...

Page 103: ...ber 2 2019 Sifos Technologies Page 103 Ideal Result LINKED 36 dB on each pair Ideal Result 2 6dB Flat Frequency Response 1 100 MHz Figure 4 8 Exercise 4 Script Execution SNR Results Figure 4 9 Exercise 4 Script Execution PSD Results ...

Page 104: ... dB Example 5 script Using PVA 3102 in slot port 1 1 set port 1 1 Add the 100m Line Impairment on all Pairs pva_line port impair all Configure the Test Port for 100Base TX full duplex and re link pva_speed port 100 duplex full Measure the broadband power with the impairment in place set impr_dB lindex pva_rx_pwr port stat 5 Remove the 100m Line Impairment on all Pairs pva_line port normal all Meas...

Page 105: ...tore Configure both Test Ports for auto speed duplex polarity pva_speed TxPort auto duplex auto arm pva_polarity TxPort auto arm pva_speed RxPort auto duplex auto arm pva_polarity RxPort auto arm Force a re link to establish the desired line conditions pva_relink TxPort pva_relink RxPort Verify that the links are up pva_link_wait TxPort pva_link_wait RxPort Transmit Traffic on both ports to prime ...

Page 106: ... Enabled Test ports 1 1 and 1 2 need to be connected to a DUT that supports packet forwarding and supports 1000BaseT operation with Master timing mode Note Test port 1 1 and 1 2 MAC addressing were configured in previous example to send traffic to each other Switch bridging was also established and it is assumed to be in tact for this example Figure 4 11 Example 6 Packet Flow Test ...

Page 107: ...ons pva_relink TxPort pva_relink RxPort Verify that the links are up pva_link_wait TxPort pva_link_wait RxPort Configure impairments on the TxPort pva_line TxPort impair all pva_noise TxPort on dB 4 0 pva_tx_clk TxPort jitter dB 2 0 offset 0 Start the Received Packet Counter on the RxPort set rslt pva_rx_pkt RxPort start Transmit 1024k packets with the TxPort set rslt pva_tx_pkt TxPort size 60 rat...

Page 108: ...PVA 3000 Reference Manual December 2 2019 Sifos Technologies Page 108 ...

Page 109: ...nce Test Suite consists of the following sequence able automated tests DUT Link Capabilities Test 100BaseTx Transmitter Test 1000BaseT Transmitter and Interface Test 10BaseT Receiver Test 100BaseTx Receiver Test 1000BaseT Receiver Test Separately the PHY Performance Test Suite adds the following specialized automated testing and reporting for Power Sourcing Equipment PSE s PSE DC Magnetic Unbalanc...

Page 110: ...ase T4 YES DUT does support 100Base T4 Pause NO DUT does not support Pause operation RESPOND DUT capable of full duplex pause operation XMIT DUT desires asymmetric pause RESPOND XMIT DUT desires asymmetric pause capable of full duplex pause operation Link_OK FAULT DUT has detected remote fault YES Link is OK N A Could not be assessed GIG_ERR MASTER SLAVE configuration fault detected MDI MDI X AUTO...

Page 111: ...her tests 5 3 Transmission and Interface Tests The Transmission and Interface tests assess physical layer transmitter performance characteristics see Section 1 3 2 and interface characteristics see Section 1 3 7 Tests are available for 100BaseTx and or 1000BaseT capable DUT s Note There is no specific test for 10BaseT transmitters Generally these tests are readily sequenced to one or more DUT port...

Page 112: ... receivers with impairment was described earlier in Section 1 3 13 The PHY Performance Test Suite fully takes advantage of that concept 10 100 1000BaseT receivers universally must be designed to tolerate the types of insertion loss created by 100M of Cat5 cable while also tolerating other forms of line impairment All 10 100 1000BaseT receiver testing performed in the PHY Performance Test Suite is ...

Page 113: ...mode where the result for each impairment condition will be either UP or DOWN As discussed earlier in Section 1 3 12 Link Stability measurements are not likely to be indicative of potential packet loss given a 10BaseT link where there is no option for DUT to drop back to a lower rate and no real motivation to stop sending link pulses because of an impaired receiver Link Monitor measurements are co...

Page 114: ...ly impairments simultaneously to all four pairs thus reducing some of the asymmetrical impairment abilities used with 10BaseTand 100BaseTx receiver testing The impairments utilized with the DUT in the MASTER timing mode are Line Loss High Slew Rate Line Loss Low Transmit Level Low Slew Rate Line Loss Random Noise e g 1 1 5 and 4 dB where 0 dB Noise 40mVpk pk Line Loss Random Jitter e g 1 1 and 3 d...

Page 115: ...e Transmitter and Interface results are in the middle and the Receiver test results are at the bottom The report is structured so that each port tested in a sequence produces a specific worksheet tab or page dedicated to that port Up to 24 ports of data can be displayed in a single report The standard report includes colorized graphics and colorized annotations of test limit excursions Most of the...

Page 116: ...s The pva_tx_100 test reports results in each of the following reporting categories for pairs 2 and 3 Transmit Power Level Signal Quality SNR Low Frequency PSD Wide Band PSD The pva_tx_1000 test reports results in each of the following reporting categories for all four pairs Transmit Power Level Signal Quality SNR Low Frequency PSD Wide Band PSD Skew Echo Xtalk Transmit Power Level at both 100Base...

Page 117: ...k peak signal amplitude and pulse droop The Estimated Peak Peak Voltage and Droop tables are produced using relationships between PSD peak to peak voltage and pulse droop that were established and verified during formal correlation exercises The Vpp and Droop values in this table directly relate to 802 3 published specifications for 100BaseTx and 1000BaseT Note Droop is expressed as of peak voltag...

Page 118: ... 5 10 and graphical form see Figure 5 11 within the Skew Echo and Xtalk section of the standard report These results are based upon measurements of the following parameters Gigabit Pair Timing Skew see Sections 1 3 6 and 2 2 2 Bulk Wideband Return Loss see Sections 1 3 8 and 2 2 2 Bulk Wideband Crosstalk see Sections 1 3 9 and 2 2 2 Gigabit pair timing Skew presents timing offset of three gigabit ...

Page 119: ...see Figure 5 13 and the test results will be in percentage of transmitted packets that were received at the calibration partner port The 100BaseT results see Figure 5 14 can appear with a Link Mon column header or a Link Chk column header similar to the 10BaseT results when the Link Monitor method of testing is applied The choice is based on the outcome of the 100BaseTx link stability pre assessme...

Page 120: ...SNR3 and Offset Jitter SNR4 In all cases the value reported for SNR1 SNR4 is the lowest SNR reading from all pairs These four values will warn if there is any danger that a low Link Stability reading or low Packet Flow reading could in any way be the result of the test port receiver getting into trouble with the applied impairment condition SNR readings below 23 dB will report as potential danger ...

Page 121: ...rns to the PSE on a second twisted pair As figure 5 18 indicates DC unbalance current Iunb is the magnitude difference in current flowing on each conductor of a twisted pair The IEEE PoE specification 802 3at stipulates that a link supporting PoE have a maximum DC Resistance Unbalance of 3 where DC Resistance Unbalance Conductor Resistances Conductor Resistances This means that PSE s and PD s must...

Page 122: ...rt under test in order to perform measurement The link is done over an electrically transparent DC decoupled channel that will be calibrated and compensated as part of the DC Unbalance Analyzer software The two PhyView Analyzer measurements of interest are 1 Low Frequency PSD e g 20KHz and 200KHz 16 Signal Noise Ratio SNR As the PSE port magnetics begin to saturate with DC bias both the Low Freque...

Page 123: ...r are reported graphically using a pop up Excel spreadsheet see Figure 5 21 The effects of DC Unbalance on SNR Low Frequency 50KHz PSD and estimated Pulse Droop are shown in graphical form Note Droop is waveform percentage of peak value at 500nsec The upper graph shows SNR degradation by pair designator to DC Unbalance levels ranging up to that maximum level requested Dotted lines show 100BaseTx S...

Page 124: ...PVA 3000 Reference Manual December 2 2019 Sifos Technologies Page 124 ...

Page 125: ... 99 99 will broadcast a configuration to every test port in a PSA chassis Similarly the slot port 99 1 will configure all Port 1 s excluding Port 2 s and 99 2 will configure all Port 2 s excluding Port 1 s in current chassis to the specified configuration Additionally the slot port N 99 will configure both test ports of slot N within a PSA chassis Broadcast port designations should only be used wi...

Page 126: ...s using stat query and retrieve results upon READY status The time savings comes about because the time to initiate measurements step 1 and the time to query meters step 3 is small relative to the time to execute the measurements The following script demonstrates this concept using a PSD measurement performed on every PhyView Analyzer test port in a chassis Build Test Port List using pvaSlotRange ...

Page 127: ...h Establish Transmit and Receive Test Ports set txPortList list 1 1 2 1 3 1 4 1 set rxPortList list 1 2 2 2 3 2 4 2 Configure burst packet transmission to all Test Ports pva_tx_pkt 99 99 size 128 rate line count 1024K Start all the Rx counters in parallel we can ignore the ones that we don t care about when we retrieve the results pva_rx_pkt 99 99 start Start the Tx packet flow on the specific por...

Page 128: ...l shell without modifying the standard prompt wishrc_psapi tcl Sourcing this file will layer or add the PowerShell API into a Wish shell without modifying the standard prompt To merge the PowerShell API into a Tcl shell execute the following command from Tcl source c Program Files Sifos PSA1200 tclshrc_psapi tcl Windows PC source env HOME Sifos PSA1200 tclshrc_psapi tcl Linux Unix WS To merge the ...

Page 129: ...d actions with the instrument and then transition those commands to the automation technique appropriate for the external application context the user needs to use PowerShell PSA can be accessed from external applications using the following methods 1 Application Programming Interface API Library accessible to any programming environment that can call functions in either a Windows dll or Linux so ...

Page 130: ...tility results are passed back through the socket I O as single lines with user specified delimiters between elements Commands will respond with either COMMAND_OK or with PowerShell_ERROR Error Message Queries will respond with either RESPONSE data or with PowerShell error message Special client command quit will terminate PowerShell psa_server_off will prevent new server connections show port wil...

Page 131: ...sole is opened or installed in Windows using the PowerShell TCL exe or PowerShell Wish exe executable and in Linux with the PowerShell_TCL sh or PowerShell_WISH sh shell scripts Batch mode method 1 takes advantage of the fact that any tcl file located in the PSA Software Contrib directory will automatically get sourced into PowerShell Batch mode 2 involves editing the PowerShell Initialization Fil...

Page 132: ...ssion parameters excluding wideband return loss where with DCE Pair 2 will be analyzed and with DTE Pair 3 will be analyzed DCE data communications equipment Fault Mode This control instructs the program to either abort in the event of one or more unreliable un linkable DUT ports or to continue testing excluding exclude any ports that are shown to be unreliable ABORT on unreliable port Report Mode...

Page 133: ...ual pva_veriphy command and associated arguments in presented in Section 4 17 6 4 2 pva_veriphy Reporting The pva_veriphy automated test program returns a standard ASCII text formatted table that can be further processed by a calling Tcl Tk script and it can also be recorded to a user specified data file Figure 6 1 depicts a nominally performing device under test where 6 ports are tested at 10 100...

Page 134: ...dB and 31 dB Finally if Link Stability measures below 97 or SNR measures below 27 1 dB the Summary will report FAIL Similarly the decision logic for the Transmission Test TEST2 requires that as many as 13 parameters meet or exceed certain criteria in order to be deemed GOOD and will produce a FAIL if any one of those parameters fall short of the MARGINAL criteria Users wishing to alter summary cri...

Reviews: