Microsoft PowerPoint - TP2_TP3_ParamMeasTestReadiness_v8 110th Oct 2005 TP2 and TP3 Parameter Measurement Test Readiness Jonathan King, Sudeep Bhoja, Jeff Rahn, Brian Taylor 10th Oct 2005 210th Oct 2005 Contents • Tx and Rx Specifications • TP2 Testing – Tx: Eye Mask OMA, ER, Average Power Encircled Flux (and CPR) RIN (and measured SNR (Tx to Rx)) TWDP • TP3 Testing – Stressed Eye Generator – Rx: B-B Sensitivity Performance TP3 Test Waveforms Overload Aim is to show experimental verification of test procedures & compliance to specifications 310th Oct 2005 Summary of Major LRM Parameters at TP2 & TP3 >30% inside 5um radius >86% inside 11um radius Encircled Flux 4.7dBTWDP 5x10-5 hits in margin0%Eye Mask Margin -128dBmOMA/HzRIN +0.5dBmMean Power +1.5-4.5dBmTx OMA CommentsMaxMinTP2 Parameter TP3 informative waveform-6.5dBm OMAInformative Sensitivity Test TP3 normative waveforms+1.5 +0.5 dBm OMA dBm mean Overload TP3 normative waveforms-6.5dBm OMAStressed Sensitivity Test +0.5-8.5dBmMean Power +1.5-6.5dBm OMARx OMA Range CommentsMaxMinTP3 Parameter 410th Oct 2005 TP2 Testing Parameters measured at module level for TP2 Readiness evaluation: • Average launch power, OMA, ER, Jitter, Eye-mask Margin – Test kit required: • 231 PRBS data source (eg XAUI, self generated), bit trigger • 8x1 8x0 (or similar) data source • Digital Oscilloscope with O-E • Optical power meter • TWDP – Test kit required: • 29 Data source, pattern trigger • Digital Oscilloscope with O-E • Computer • Encircled Flux – Test kit required: • IR camera or equivalent method TP2 tests use conventional equipment and established methods 510th Oct 2005 TP2 Measurement Results • Typically >4 σ margin to spec • Confirms TP2 specifications are consistent with high volume manufacture, as required for PAR 11.9-4.5 (min)0.27-1.27Min/coldSMF OMA 12.30.25-1.48Nom/nom 7.90.37-1.6Max/hot 3.805.2820.0Min/coldEye Margin % 4.14.9720.6Nom/nom NB: 0 hit limit used 4.04.6518.5Max/hot 5.0331.5225.4Nom/nom Jitter p-p, ps 8.50.164.41Max/hot 13.30.094.31Nom/nom Min/cold Max/hot Nom/nom Min/cold Supply, Temp 5.60.38-1.64 6.9 6.5 6.2 Margin to Spec. σ 3.5 0.5 (max) Spec ER, dB SMF power, dBm Parameter 0.164.62 0.29-1.35 0.26-1.12 σAverage 610th Oct 2005 TP2 Encircled Flux and CPR Testing & Results Encirc led flux 0 10 20 30 40 50 60 70 80 90 100 1 2 3 4 5 6 7 8 9 % 11um 5um >30%, 5um >86%,11um 4.4 σ 4.9 σ 2.3 σ Margin to spec 2dB 86% 30% Spec CPR ** 11 um radius 5um radius Encircled Flux 0.25dB0.9dB 2.32%97.3% 13.3%60.4% Sigma Average Encircled Flux: • Non-IEC method used here (scanned SM fibre technique) • Comparison underway with IR camera approach 710th Oct 2005 TP2 Measurements: RIN • Test conditions – Nominal: 30oC, nominal module bias, & nominal back-reflection – Stressed: 70oC, low supply (-6%), and 12dB back-reflection • Nominal Tx OMA -2dBm into SMF – Worst case RIN: -140dB/Hz nominal, -137dB/Hz stressed -155 -150 -145 -140 -135 -130 -4 -3 -2 -1 0 1 2 3 Nominal RIN1 (dB/Hz) Nominal RIN0 (dB/Hz) Stressed RIN1 (dB/Hz) Stressed RIN0 (dB/Hz) R IN (d B /H z) OMA (dBm) RIN dB/Hz OMA into SMF 810th Oct 2005 TP2 Measurements: Example of Tx-Rx SNR • Test condition – TOSA & ROSA: 70oC, low supply (-6%) • At -2dBm OMA Tx output power – SNR > 38dBe at -6.5dBm OMA Rx input 25 30 35 40 45 -14 -12 -10 -8 -6 -4 -2 0 2 -3.9dBm_SMF -3dBm_SMF -2dBm_SMF -1dBm_SMF 0dBm_SMF SN R (d B ) OMA (dBm)Received OMA, dBm LRM spec 30dBe 8dBe margin to spec 910th Oct 2005 TP2: TWDP • TWDP measurements here used Draft 2.2 stressors, 100/50 TFF/DFE EDC emulation – Normalization of captured waveform to OMA added within code • 4.7dB test limit, as in Draft 2.2 • Needs repeating with latest finite EDC emulation and latest OMA extraction code • 10 samples typically >4 σ margin to spec Max/hot Nom/nom Min/cold Supply, Temp 3.9σ0.074.41 6.1σ 4.4σ Margin to Spec. 4.7 Spec TWDP, dB Parameter 0.064.31 0.084.32 σAverage 1010th Oct 2005 TP3 Testing Parameters measured at module level for TP3 Readiness Evaluation: • Normative sensitivity tests – Test kit required: • Stressed eye generator (SEG) with E-O (3.5dB ER, 27dB SNR) • BER detection • Optical power meter • Overload – Test kit required: • 29 Data source, pattern trigger • Digital Oscilloscope with O-E • Computer 1110th Oct 2005 TP3 Testing: SEG (Stressed eye generator) for LRM • Creates a stable, stressed channel that can be used to test and compare receiver performance • ISI generator is based on TFF plus linear E-O – Avoids issues with long period reflections associated with lumped element implementations • On board PRBS generation and error detection • Large number of taps permits better fits – Short period reflections can be compensated for with extra taps – DUT sees accurate stressor 1210th Oct 2005 SEG Block Diagram USB interface uC XAUI bias T TOSA Linear DriverEDC Ext Data source ROSA LA choice of external data source electrical noise source 231-1 PRBS • Like a transponder, but with EDC in Tx chain, linear driver & E-O • USB interface for easy PC based GUI control • Implements 802.3aq TP3 Test Waveforms; fiber pulse response emulations are also possible. 1310th Oct 2005 SEG Board Photo Power Supplies USB TOSA Output Bias Tee Linear Driver EDC 1410th Oct 2005 SEG: Symmetric Channel • Blue = SEG fit, Red = Target channel • PIE-D value 3.85 dB – PIE-D error = +0.03dB • SNR = 30.46dBe – SNR error = -0.54dBe • ER = 3.6dB – ER error = +0.1dB • OMA = -3.0 dBm temporal frequency 1510th Oct 2005 SEG: Pre-cursor Channel • Blue = SEG fit, Red = Target channel • PIE-D value 4.1 dB – PIE-D error = +0.06dB • SNR = 30.65dBe – SNR error = -0.35dBe • ER = 3.63dB – ER error = +0.13dB • OMA = -2.96dBm temporal frequency 1610th Oct 2005 SEG: Post-cursor Channel • Blue = SEG fit, Red = Target channel • PIE-D value 4.2 dB – PIE-D error = 0.0dB • SNR = 31.66dBe – SNR error = +0.66dBe • ER = 3.6dB – ER error = +0.1dB • OMA = -2.99dBm temporal frequency 1710th Oct 2005 SEG Characteristics* dH2 dH3 dB/Hz-128-135RINOMA dBc [4]-<-40 dBc [4]-<-38 Transmitter Linearity 0.08 2e-4 - 16 3.7 -2.6 -1.5 Typ MSE4e-4-Pulse Response Fit Error dB0.175-0.175PIE-D Error dBm0.5-4Average Power dB-20-Optical Return Loss Hz100Dynamic Repetition Rate dB (optical) [3]1715Signal-to-Noise Ratio dB [2]TBD3.5Extinction Ratio dBm [2]TBD-4.5OMA UnitMaxMinItem [*] Measurements with 62.5µm MCP [2] Max upper bound is determined by linearity requirements. [3] Consistent with IEEE 802.3aq D2.2 [4] 500MHz tone 1810th Oct 2005 SEG: Calibration • Measure the step response of the channel using a 128 bit square wave • Calculate the pulse response • Calculate PIE-D using the folded frequency method • Linearity is guaranteed by design bounds • Combination of SNR, ER, goodness of fit, and PIE-D value used to determine suitability of channel for testing • Procedure iterates until the target SNR and ER are achieved • BER tested on reference RX 1910th Oct 2005 Test results with SEG 2010th Oct 2005 Transponder Measurements: 220m (Draft 2.2) Normative Tests 15 units tested over 3 corners: typically >4σ margin to spec 4.5 σ0.72-9.7Min/coldSymmetric 4.5 σ0.79-10.1Nom/nom 4.6 σ0.83-10.3Max/hot 4.7 σ0.85-10.5Min/coldPre-cursor 6.4 σ0.69-10.9Nom/nom 7.7 σ0.57-10.9Max/hot Post-cursor Stressor 8.3 σ0.55-11.1Max/hot 7.9 σ0.58-11.2Nom/nom 6.3σ0.64-10.6Min/cold Margin to spec (-6.5dBm) σ dB Average sensitivity dBm OMA Supply/temp. 2110th Oct 2005 Rx Overload Measurements • SEG overload test capability recently achieved, enabled with high power cooled DFB used as O-E – Max output +3dBm OMA max (2dBm mean, 6.4dB ER) >2.6 >2.6 >2.6 Overload dBm OMA (Module) Min/cold220m Symmetric 3.0Nom/nom Max/hot Min/cold220m Pre-cursor 3.6Nom/nom Max/hot 220m Post-cursor Stressor Max/hot 3.5Nom/nom Min/cold Overload dBm OMA (Engineering test bed) Corner 2210th Oct 2005 Summary • Overview presented of the measurement capabilities & the technical feasibility for many of the 802.3aq 10GBASE-LRM TP2 and TP3 Parameters – Test capabilities for TP2 parameters established – Test capabilities for TP3 parameters established • Excellent TP2 and TP3 performance and margins demonstrated – Essentially all measurements were completed at the TP2 & TP3 points connected to implementations complying to different industry MSA modules • Supports the practicality and manufacturability of 10GBASE-LRM PMD for the TP2 and TP3 specifications