Instalaciones de fibra optica en entornos de centro de datos y campus para suportar 40, 100G y mas

Transcription

1 Instalaciones de fibra optica en entornos de centro de datos y campus para suportar 40, 100G y mas Jim Davis Regional Marketing Engineer Fluke Networks

2 Agenda Inspection and Cleaning Loss Budgets TIA (Cabling) vs IEEE (Application) Loss Testing Best practice to reduce uncertainty 3 jumper reference for MPO cable testing with OLTS OTDR Testing - Troubleshooting MPO Trunks Results management Cloud based services

3 INSPECTION, AND, IF NECESSARY, CLEANING (REPEAT AS NEEDED)

4 #1 Problem: Dirt! Contaminated connector end-faces: Leading cause of fiber link failures Particles of dust and debris trapped between fiber end faces cause signal loss, back reflection, and damaged equipment Many Sources of contamination: Equipment rooms & Telecommunication rooms in filthy environments Improper or insufficient cleaning tools, materials, procedures Debris and corrosion from poor quality adapter sleeves Hands of technicians Airborne 4

5 Video Microscopes for Inspection Protect your one good eye A variety of tips available Patch Cord and Bulkhead Auto Centering and Auto Focus options Ability to store images Can be used to compare against standard

6 Automated Analysis IEC IEC ED.2 MM Zone Name Scratches Defects A: Core (0-65µm) B: Cladding (65-115µm) No Limit 3 µm None > 3 µm No Limit 5 µm None > 5 µm 4 5 µm None > 5 µm No Limit < 5 µm 5 From 5-10 µm None > 10 µm C: Adhesive No Limit No Limit D: Contact ( µm) No Limit No Limit < 20 µm 5 30 µm None > 30 µm

7 Inspect Clean - Inspect

8 Inspect Clean - Inspect

9 Inspect Clean - Inspect

10 If Dry Clean fails, try Wet Clean

11 Pass!

12 MPO Inspection & cleaning Requires special adapter for microscope Two axis camera is easier even for 12 fiber connections Second knob on tip allows centering of fibers

13 MPO Inspection & cleaning Cleaning is more important than ever With large contact surface, any contaminants can cause and air gap and this leads to reflectance problems Example of a dirty MPO end face

14 The cover lifts to insert MPO connectors You may put a drop of solvent on the cleaning tape

15 MPO Inspection & cleaning That fiber from the earlier slide is clean now Example of a cleaned MPO end face

16 Tier I testing is loss, length, and polarity Is there enough light coming out of the fiber for the receiver to detect? Is the transmission on the right (left) fiber?

17 A Test Result! Yes! 17

18 A Test Result! What Test Limit was Passed? TIA Length Based Limit for Multimode Fiber 18

19 How Fiber Loss is Measured Set a reference Run a test Find the difference

20 First set a reference Then find the Difference Optical Loss Measured Direct connection (No bulkhead adapter!) All connections are included in the loss measurement

21 Why not 2 jumper reference? Sadly, most folks are setting a reference this way? db Issues You have no idea what the loss is in the adapter Whatever it is, it s subtracted from your measurement The uncertainty is horrendous negative loss

22 What is done today Let s take an example 0.75 db Issues You have no idea what the loss is in the adapter Whatever it is, it s subtracted from your measurement The uncertainty is horrendous negative loss

23 What is done today Let s take an example 0.3 db 0.1 db 0.3 db Measurement = = db Issues You have no idea what the loss is in the adapter Whatever it is, it s subtracted from your measurement The uncertainty is horrendous negative loss

24 Example of Negative Loss Bad test results do not accept!

25 Accurate test process and values will reduce uncertainty For less uncertainty in our fiber testing, especially of multimode fiber, there are 4 key ingredients to loss testing A one jumper reference An LED source Reference Grade Connectors Encircled Flux compliance 25

26 Accurate test process and values will reduce uncertainty For less uncertainty in our fiber testing, especially of multimode fiber, there are 4 key ingredients to loss testing A one jumper reference An LED source Reference Grade Connectors Encircled Flux compliance LED Source VCSEL Source 26

27 Accurate test process and values will reduce uncertainty For less uncertainty in our fiber testing, especially of multimode fiber, there are 4 key ingredients to loss testing A one jumper reference An LED source Reference Grade Connectors Encircled Flux compliance < 0.10 db of loss per mated pair <0.20 db for Single-Mode 27

28 Encircled Flux compliance The light source s launch condition determines how and where the light is distributed within the fiber Cladding Core Source 1 Over filled Source 2 Under filled Source 3 EF Compliant

29 Encircled Flux Improves Precision EF compliant = just enough light Measure the right loss Passes good links Fails bad links Overfilled = too much light near the edge Measures high loss Fails good links Underfilled = not enough light near the edge Measures low loss Passes everything

30 CALCULATING A LOSS BUDGET Difference between a TIA limit and an application limit The Loss Budget determines what Passes and Fails

31 There is no Cat 6A for fiber Which Limits to use? Does this link really pass? There is conflict between what the standard will support and what the application requires Installers should use Custom Limits to certify links Cabling Standards: TIA 568 ISO Application Standards: 1000BASE-T 10GBASE-T Application Standards: 1000BASE-T 10GBASE-T

32 There is no Cat 6A for fiber Which Limits to use? Does this link really pass? There is conflict between what the standard will support and what the application requires Installers should use Custom Limits to certify links Manufacturers may offer a custom link loss calculator for their components Cabling Standards: TIA 568 ISO Application Standards: 10GBaseSX 40GBase-SR4

33 IEEE 802.3ba (100 Gig) Fiber Channel Performance Requirements Per Table Maximum Channel Insertion Loss = 6.7 db at 1310 nm Maximum Length = 10 Km Per db at 1310 nm Maximum Discrete Reflectance shall be less than -26 db

34 TIA Limits Vs. IEEE limits 1 km Cabling Standards: TIA 568 ISO Per TIA: # Connectors * 0.75 db KM of fiber * 1 db/km (Tight Buffered) # Splices * 0.30 db In this example = 2.5 db Total Loss Budget Measured Loss = 2 db Pass for both TIA and IEEE Application Standards: 10GBaseLX 40GBase-LR4

35 TIA Limits Vs. IEEE limits 1 km Cabling Standards: TIA 568 ISO Per IEEE: Less than 6.7 db of loss, less than 10 KM of length In this example 6.7 db Total Loss Budget Measured Loss = 5 db Pass but only for IEEE Application Standards: 10GBaseLX 40GBase-LR4

36 Using a TIA limit without understanding the application Customer wants to run 10GBASE-SR on this multimode link 0.75 db 0.75 db 0.75 db nm (300 m) 0.75 db TIA (tester) Limit = 0.75 db db db db db = nm 10GBASE-SR Limit = nm This design will not support 10GBASE-SR

37 Which Limits to use? Does this link really pass? That 3.90 loss budget falls within the acceptable values for the cabling standard, but outside of the acceptable values for the application Cabling Standards: TIA 568 ISO Application Standards: 10GBaseSX 40GBase-SR4 X

38 Using a TIA limit without understanding the application Customer wants to run 10GBASE-SR on this multimode link 0.90 db (300 m) 0.75 db 0.75 db TIA (tester) Limit = 0.75 db db db = nm 10GBASE-SR Limit = nm This design will support 10GBASE-SR

39 Using a TIA limit without understanding the application Customer wants to run 10GBASE-SR on this multimode link 0.90 db (300 m) 0.75 db 0.75 db TIA (tester) Limit Confirm = 0.75 the db + performance 0.90 db of your db MPO cassettes Many manufacturers = 2.40 have 850 nm regular and High Performance or 10GBASE-SR Low Limit Loss = cassettes Regular 850 nmmay This have design > 1 will db of support loss for 10GBASE-SR the two connections

40 MPO/MTP Testing with OLTS

41 Multi-Fiber Connectors (MPO/MTP) 12 Fiber For plug and play cassettes in datacom environment 40 Gig 16 Fiber For SAN market, where switch & director blades come in eight fiber increments 24 Fiber High density for the data center server side 100 Gig

42 Two options for testing MPO to MPO cables Traditional duplex field tester Dedicated MPO field tester

43 MPO/MTP Testing with OLTS 3 Jumper Reference

44 How to tell 3 jumper reference is set properly Very important for end user Look for two TRC Verifications in test results

45 How to tell 3 jumper reference is set properly Very important for end user Second step check 2 nd TRC and reset reference

46 Testing MPO to MPO cables LC LC MPO MPO Loss reported Referenced out

47 OTDR testing of MPO/MTP connectors

48 OTDR Testing of MPO Connectors Ideally you will have at least 30 Meters of Launch and Receive fibers Provided that the link under test is short Use a fan out cord or cassette to convert from Single fiber Port on OTDR 30 Meters 30 Meters Launch Fiber Receive or Tail Fiber Don t forget your Pinned/Unpinned connections!

49 Port Protector/Sacrificial Cord in Action 30 Meter Launch Cord 30 meter Receive cord

50 MPO Launch Cord and Trace 30 Meters Fan-out to MPO Launch Fiber

51 MPO Launch Cord and Trace MPO

52 All 12 fibers stacked trace

53 Looks easy now let s try Single Mode MPO

54 First check launch/receive cords 50 Meter Launch Cord 50 meter Receive cord

55 Detail of launch fiber verification notice Gender Changing connector and tool

56 Example of Fiber Switch instead of Fan Out Switch 1 x 12

57 The Switch helps, but makes a large event

58 Launch and Receive Fibers Look Good

59 Here are Launch Link Receive Fibers

60 What is happening with the First Fiber?

61 Is the connector dirty?

62 Please be sure to Inspect ALL Connectors before installing, clean them if necessary, inspect again!! Video Microscope Brand new out of bag After Cleaning 62

63 Documenting Results Request your test results in Native Format, not.pdf Your tester only delivers results in Paper format? Consider using a cloud based results management service Check that the reference value is correct and recent Did they verify the known good leg? Deliver the results today, not in a month While your team still has access to the site

64 In Conclusion Inspect and clean if necessary Follow the right process to reduce uncertainty in your measurement MPO/MTP connectors are an accepted multi-lane solution Data Centers use them extensively Tier I testing can be done with OLTS and 3 jumper reference Tier II testing can be used for troubleshooting A 12/24 fiber launch fiber and tail fibers are required Leverage the benefits of cloud based results management

65 Thank you, Gracias, Obrigado Jim Davis Fluke Networks 6920 Seaway Blvd Everett, WA 98271