Presentation title (IMPORT THROUGH MENU View or Insert -> Header

1
Presentation title (IMPORT THROUGH MENU View or
Insert -gt Header Footer)
2

• " To be or not to be - shielded or unshielded?
  That is the question!"

• Matthias Gerber, Reichle De-Massari AG

• September, 2010

3
Motivation
STP vs. UTP
- New application? new demands - Susceptibility
to disturbances ? - Marketing smog screen.
4
Modulation e.g. PAM 16

Reduces signal bandwidth, increases
susceptibility to noise
12 Bit 12 Pulses
t
1 0
U2 V
1 1 0 0
0 0 1 0
1 0 0 1
NRZ Coding
5
Development signal strength senders
10Base-T
1V
10 MHz
Signal Level
0V
Spectral Bandwidth
-1V
Manchester Coding 20 MSymbols/s
6
Development signal strength receivers
10Base-T 100BASE-TX 1000BASE-T
10GBASE-T
1V
10 MHz IL 6.5 dB
31 MHz IL 11.4 dB
62 MHz IL 16.3 dB
417 MHz IL 44.7 dB
0.81V
Signal Level
0V
Spectral Bandwidth
0.08V
-1V
Manchester Coding MLT-3 PAM 5
PAM 16
20 MSymbols/s 25 MSymbols/s 125
MSymbols/s 800
MSymbols/s
7
Noise common mode disturbances
8
Coupling raceway to UTP cabling
Test setup
9
New Parameter Coupling attenuation(CA)
PCom
PDif
Coupling attenuation acoupl10Log(PCom/PDif)
10
CA _at_ ISO/IEC 11801 Amd.1 (2008-04)
ISO/IEC 24702 Industrial Premises MICE
Categorization E1 (Office) ? acoupl 40 dBE2
(Light industry) ? acoupl 50 dB E3 (Factory
floor) ? acoupl 60 dB
11
New parameter Alien Crosstalk (AXT)

• X-Talk of neighboring channels
• Increases in parallel runs

• Primarily between same colored pairs (same lay
  length)
• Difficulties to test in the field

12
Active Compensation with 10GBase-T

• Active DSP Echo- and NEXT compensation
• NEXT gt40 dB
• FEXT gt25 dB
• RL gt55 dB

10GBase-T Transmission
13
Independent LABs Study goals

• What EMC properties do shielded and unshielded
  cabling components possess?
• What special measures are necessary to take for
  screened or unscreened systems?
• Which advantages and disadvantages are linked to
  screened or unscreened systems?
• Give guidance to planners or end users to prevent
  confusion and uncertainty when select systems.
• Providing a fair and independent comparison of
  the EMC behavior of screened and unscreened
  cabling

14
The study was commissioned by
The study was carried out by
GHMT AG In der Kolling 13 66450
Bexbach www.ghmt.de info_at_ghmt.de
15
Cabling sample selection
16
Active components used
4x Switch 7x Server
Manufacturer Extreme Networks IBM
Typ Summit X450a-24t X3550
Slot Summit XGM2-2bt INTEL 10Gigabit AT Server Adapter
17
System set-up schematic
Installation cable
Patch panel
Patch panel
Patch cord
Server
Server
Switch
Equipment cord
Equipment cord
18
System set-up anechoic chamber
19
(No Transcript)
20
Emission Radiated field strength
Tests the ability of a device not to disturb
other equipment in their normal operation
21
Emission Radiated field strength
10GBase-T System 02 (U/UTP) vs. System 05 (S/FTP)
22
Emission Radiated field strength
1000Base-T System 02 (U/UTP) vs. System 05
(S/FTP)
23
Emission Radiated field strength
10GBase-T vs. 1000Base-T System 02 (U/UTP)
Spectrum 1000Base-T
Spectrum 10GBase-T
24
Emission Radiated field strength
10GBase-T System 05 (S/FTP) different earthing
single-sided earthing
double-sided earthing
No significant differences in terms of EM
emission !
25
Emission Radiated field strength
10GBase-T Overview
10 Gbps _at_ 800 MHz
The better the EMC quality of the cabling, the
lower the radiated radio-frequency disturbance
field strength.
26
Emission Radiated field strength
10GBase-T Summary

• Screened cabling will be able to transmit
  10GBase-T and fulfil Class B emission limits (for
  residential areas)
• Unscreened cabling with 10GBase-T fulfils Class A
  emission requirements (for business areas, e.g.
  dedicated Office spaces, data centre)
• The better the CA (and therefore the screening
  efficiency) the smaller the radiated field
  strength.
• No recognizable improvement of the emission from
  double sided earthing to single sided earthing.

27
EMC immunity tests performed

• EN 61000-4-3 Radiated, radio-frequency,
  electromagnetic field immunity test
• EN 61000-4-4 Electrical fast transient/burst
  immunity test
• EN 61000-4-6 Immunity to conducted
  disturbances, induced by radio-frequency fields
• EN 61000-4-8 Power frequency magnetic
  field immunity test
• EN 61000-4-2 Electrostatic discharge

28
Test criterion example plot
Stress level 1
Stress level 4
Stress level 3
Stress level 2
29
MICE-Classification EN 50173-12008-04
Excerpt for criterion E (Electromagnetic
compatibility)
30
Immunity Radiated, radio-frequency,
electromagnetic field
Tests the ability of a device to operate normally
in the presence of electromagnetic fields
31
Immunity Radiated, radio-frequency,
electromagnetic field, Overview
32
Immunity Radiated, radio-frequency,
electromagnetic field, Summary

• All screened systems allow for 10Gpbs operation
  in E1 / E2 environments.
• For harsher environments, an S-FTP screen
  construction is necessary.
• Double-sided earthing does not lead to any
  improvement over single sided earthing.
• Unscreened systems meet the requirements of class
  E1 / E2 for 1 Gbps. For 10Gbps additional
  protection measures are necessary.

33
Practical test 10GBase-T
Walkie-talkies, mobile phone, GSM card
34
Practical test 10GBase-T
Walkie-talkies, mobile phone, GSM card
? Interference (abort data transmission) ? No
interference observed
Usage of a wireless communication device in
proximity of an unscreened cabling (e.g. in a
data center) can impair the data transmission.
35
Immunity Fast transients (burst) test
Tests the ability of a device to operate normally
in proximity of disturbing power lines and
equipment
36
Immunity Fast transients (burst) test
10GBase-T, Overview
37
Immunity Fast transients (burst) test
10GBase-T, Earthing influence
38
Immunity Fast transients (burst) test
10GBase-T, Summary

• All screened systems allow for 10Gpbs operation
  in all environments (E1 E3). Quality of the
  screen improves immunity.
• Double-sided earthing does improve immunity
  against fast transients over single sided
  earthing.
• The screen connection has to be continuous from
  Switch to Server (no U-UTP patch cords).
• Unscreened systems meet the requirements of class
  E1 for 1 Gbps. For 10Gbps operation carefully
  routing necessary to separate from power cables

39
Practical test - fluorescent lamp -
Practical test 10GBase-T
Fluorescent lamp
40
Practical test 10GBase-T
Fluorescent lamp
? Interference (abort data transmission) ? No
interference observed
Fluorescent lamps can impair the 10GBase-T in
U-UTP systems, when distance between data cable
and power cable or fluorescent lamp is not
sufficient.
41
Experiment Mesh cable tray
10GBase-T, Fast transients (burst) test
42
Experiment Mesh cable tray
10GBase-T, Fast transients (burst) test, Overview
43
Experiment Mesh cable tray
10GBase-T, Fast transients (burst) test, Summary

• Screened systems meet the environmental class E3
  at all distances, even without separation.
• For unscreened systems separation to power
  cabling is essential. For class E1 distances
  must be 30 cm.
• The standardized capacitive coupling clamp
  simulates a cable installation with power and
  date cable very near together (lt 1cm).

44
Immunity test Conducted disturbances, induced by
radio- frequency fields
Tests the ability of a device to operate normally
in proximity disturbing power lines
45
Immunity test Conducted disturbances, induced by
radio- frequency fields, overview
46
Immunity test Conducted disturbances, induced by
radio- frequency fields, Summary

• The screened systems meet the requirements of
  MICE class E3 for 1000Base-T and 10GBase-T.
• Higher quality screening provides higher immunity
  against induced radio frequency fields.
• Double-sided earthing does not improve the
  immunity significantly over single-sided
  earthing.
• Unscreened systems meet the requirements for E1 /
  E2 environments in the 1Gbps operation mode.For
  10Gbps additional protection against conducted
  induced radio-frequency fields is needed.

47
Immunity Power-frequency magnetic field
Tests the ability of a device to operate normally
in the presence of large magnetic fields
48
Immunity Power-frequency magnetic field
10GBase-T, Overview
49
Immunity Power-frequency magnetic field
10GBase-T, Overview

• All systems tested meet the requirements of MICE
  class E3 in all three operation modes.
• The equalizing currents ascertained with screened
  systems do not have any impact on data
  transmission.
• Equalizing currents can be reduced by
  double-sided earthing.

50
Immunity Electrostatic discharge
51
Immunity Electrostatic discharge
10GBase-T, Overview
Conducted discharge was not tested for U/UTP
systems

52
Immunity Electrostatic discharge
10GBase-T, Summary

• The screened systems meet the requirements for
  all environments in all three operation modes
  both for air and for contact discharge.
• The unscreened systems react very sensitive to
  electrostatic discharges. Additional protection
  measures are necessary for a safe operation.
• The screen seams to offer a primary discharge
  path to earth, preventing the discharge to reach
  the vital intestines of a data cable.

53
Conclusions of UTP/STP study
54
EMC results, Summary
55
Conclusions of UTP/STP study (1)

• Secure 10GBase-T transmission, requires careful
  consideration of the environment to choose
  cabling components accordingly.
• Coupling attenuation as a parameter (especially
  when measured up to 1 GHz) represents the EMC
  behavior for emission and immunity with good
  correlation.
• The more elaborate the screen construction, the
  lower the emission and the better the immunity
  against disturbances from the outside.

56
Conclusions of UTP/STP study (2)

• Screened cabling systems with S-FTP construction
  are well suited to carry 10GBase-T under all
  environmental classesScreened solutions with
  F-UTP construction are good for office
  environments.
• Unscreened cabling systems can only be used for
  10GBase-T outside residential areas (e.g.
  dedicated offices, data centers).

57
Conclusions of UTP/STP study (3)

• For unscreened cabling to transport 10GBase-T
  special measures must be taken to reduce the
  exposure of the cabling to external disturbances
• Careful separation of data cabling to power
  cabling and disturbing equipment
• Usage of dedicated metallic raceways for data
  cabling
• Prevent usage of wireless devices in the
  proximity of the cabling
• Prevent the possibility of ESD

58
Conclusions of UTP/STP study (4)

• Only the very best unscreened solution will be
  able to deliver 10GBase-T
• For the decision of screened or unscreened
  cabling for 10GBase-T, the whole of the
  environment and the installation specifics have
  to be taken into account.

59
(No Transcript)
60
Thank you for your attention!
Presentation title (IMPORT THROUGH MENU View or
Insert -gt Header Footer)