Diapositiva 1

1
My experience at Commissioning
matteo
Matteo Volpi
2
Fiber tests
Matteo Volpi
3
Why to test fibers
Slide 1

• Fiber Attenuation measurements -
• Failure to deliver the necessary light level
  by either TTC or Glink fibers will prevent the
  data taking from the associated Tile module.
• Fiber Length measurements
• Uncertainty on the fiber length will result
  in
• for TTC uncertainty on arrival time of the
  clock and control commands gt BC-id and TOF
• The Tile studies from the TBs proves that while
  the fiber length is well controlled the
  resolution of the time measurement from the Tile
  can be as good as (0.7-1.0)nsec.

4
TESTs
Slide 2

• Fiber Attenuation measurements, were done for
  850nm (ROD wavelength) and 1300nm (TTC
  wavelength) with the LED Source/Meter
• Fiber Length measurements done with the OTDR
  (Optical Time Domain Reflectometer)
• All data went into the QCsheets linked to the
  wikipage of commissioning team 3

5
Test SETUP
Slide 3
Length tests done with OTDR (Optical Time Domain
Reflectometer)
Patch Panel USA15
Finger patch panel (SD)
OTDR
Tile Fibers 90 m ( LC/PC and SC/PC connectors)
Fiber 5 m (SC/PC and LC/PC connectors)
Fiber 2 m SM fiber (FC/APC and FC/PC
connectors)
6
Possible future setup
Gerolf Schlager
Slide 4
Reminder minimum level into TileCal Front-End
Electronics at the SD patch panel is -15 dBm
Patch panel
SINGLE MODE
MULTI MODE
(laser source)
116
TTCex
Optical attenuator
TTCoc
TileCal Front-End
(Optical Splitter 16 output at -13
dBm )
8 outputs
USA15
UX 15
Max. possible loss from TTCoc to TileCal
Front-End 2dB (safety margin 0,5 dB) Acceptance
level 1,5 dB in Multi Mode fibers (at the moment
we use MM fibers) Acceptance level 3 dB in Single
Mode (TTCoc and TTCex tested with SM

fiber that has 1.5 dB more of
tolerance). This must account for -Losses in
patch panel
-Losses in connectors
-Losses in the fibers
7
Nominal Patch Panel Loss (USA15)
Slide 4
Patch Panel
Nominal Value for each channel are known
8
Connector in USA15 Patch Panel
One big cable for each sector is spliced in 6
cables of about 20 meters before the end of the
fibers in the module.
We have some times Problems with the splice
box ( see the next slides ) We cleared and
checked the connectors a many times. All the
fibers components were found to be very sensible
to the dust, etc
Slide 5
9
Super Drawer Patch Panel
Slide 6
TTC ROD fibers
10
Measurement I The Attenuation
Slide 7
Sector 9 Side A
Expected value of attenuation Nominal loss in
the Patch panel Losses in the connectors
Loss inside the fibers 0,06dB at 1300 nm
for 100 m 0,2dB at 850 nm for 100
m Accuracy of measurement is 0,3dB from
stability of the measurement by LED
Source/Meter (that includes variations due to
connector)

• High loss for connector 5 (spare) at A side
• Out of specification using SM fiber to connect
  TTCex and TTCoc attenuation more than 3dB
• High loss for connector 19 (RO) at A side
• Out of specification attenuation more than
  1.5dB using MM fiber to connect TTCex and TTCoc
• Both have to be exchanged

11
Measurement II Length of the fibers
Slide 8
With the OTDR we measure the length of the
fibers. Speed is 4,9 ns/m ( 2.1x108 m/sec)
Length resolution (OTDR) lt 1cm
12
Measurement II Length of the fibers
Slide 9
Length for 36 fibers of sector 5 A Side
We see the long and the short fibers within
the pack, as expected
When the finger are connected we can measure the
length of the fiber inside the Super Drawer, the
average is 1,27 m (Patch panel to Interface
board)
All the data are stored in the data base. These
Measurements are important for the timing.
In advance we confront the limitation of these
machines, the OTDR and LED Source/Meter
13
More Information
Slide 10
What can be detected with OTDR only -length of
the fibers -the loss in extreme damages e.g.
bending, breaking.( I try the test with broken
fibers) -some effect in the patch panel
connectors. -Also when the fingers are already
inserted in the SD
we can check the connectors
we can measure the
length of the fibers inside the SD The good
resolution of the OTDR (lt1cm) is fundamentally
for the time reconstruction, necessary for the
system as the TOF. What can be detected with
LED Source/Meter only -the attenuation
in the fibers and in the patch panel connectors
-of course the loss in extreme damages e.g.
bending, breaking What can be seen by both
systems -some effect in the patch panel
connectors - the loss in extreme damages
14
Example of OTDR Test for the Bad Fibers
Slide 11
With the OTDR we checked the bad fibers analyzed
with the LED Source/meter
We can see a big reflection at about 16 meter
before the end of fibers in both cases. The
fibers are in the Sector 5 A Side and
are -Module 17 TTC fiber, color white, channel
20 in the Patch Panel. -Module 20 Spare fiber,
color orange, channel 35 in the Patch Panel. The
problem is in the same position for both fibers
located in two different cables.
15
New test with the OTDR after to check the bad
fibers in the splice box
Channel 20
The reflection 16 meter before the and
disappear. With LED Source Meter we measure
again the attenuation is Channel 35 0,60 dB
at 1300 nm 1,08 dB at 850 nm Channel 20 0,49 dB
at 1300 nm 0,89 dB at 850 nm Before we dont
have communication in these channels
Red lineold measure Blue linenew measure
Channel 35
Red lineold measure Blue linenew measure
Conclusion the fibers are reaper.


Slide 12
16
Some effect seen with the OTDR
Slide 13
Normally with the OTDR we see the back reflection
signal from the connectors. During the
measurement of the length we see this effect
-Loss in the connector in the patch panel
Normally this loss is much smaller than the OTDR
sensitivity. After to open and clean the
connector the loss disappear.
17
Conclusions
Slide 14
At the moment we have checked the Sector 5,7,9,11
and 13 with LED Source/Meter and OTDR. I was the
responsibile of the measurements and the OTDR
tests. My conclusion is With the OTDR we can
see some effects that belong to the finger
-problems in
the connector in the SD
-back reflection in the connector
-the
loss in extreme damages e.g. bending,
breaking We need both instruments the LED
Source/Meter to measure the attenuation and the
OTDR to measure the length of the fibers, because
the OTDR can not measure the attenuation of the
fibers. Before leaving I taught Valencia team
(now assigned to the tests) to use the
machines. Future plan Check the next sector and
if there are the problems, change the damage
fibers and check the bad connectors.