Consolidation and development of the CERN Leaky Feeder infrastructure

Size: px

Start display at page:

Download "Consolidation and development of the CERN Leaky Feeder infrastructure"

Blanche Chapman
5 years ago
Views:

1 Consolidation and development of the CERN Leaky Feeder infrastructure F. Chapron IT/CS/CS

2 Agenda What is the use of the leaky feeder (LF) cable at CERN? What is a LF cable? What is the issue? Risks assessment per machine Objective and challenges of the project Call for tender Recommendations

3 The leaky feeder infrastructure

4 What is the use of the LF cable at CERN? More than 50 km of cable at CERN Almost all tunnels and experiments Used to carry multiple signals CERN safety services (not everywhere): TETRA signals used mainly by the CERN fire brigade. Indoor localisation system (indirectly) External safety services (not everywhere): the TETRAPOL signals from the Swiss and French safety organisations General public mobile services: GSM (2G), UMTS (3G) and LTE (4G) signals Special users like the TIM ( Train d Inspection sur Monorail ) or remote maintenance robots access CERN intranet via a secured and controlled access (CERN APN). Leaky feeder cable TI8 injection tunnel

emitter. It is made of two coax cables isolated by foam and protected by a jacket.

equally distributed all along which allow to broadcast the signal in a clean way Affix to the wall/cable trays with

5 What is a leaky feeder cable? A leaky feeder cable: Is a cable used to carry radio signals in buildings It s like a long antenna connected to an emitter. It is made of two coax cables isolated by foam and protected by a jacket. Made for certain frequencies with better longitudinal loss (-3dB/m) and coupling loss The external coax cable has holes equally distributed all along which allow to broadcast the signal in a clean way Affix to the wall/cable trays with clamps Traditional antennas in tunnels are inefficient: Need of line of sight between transmitter and receiver Obstacles can easily cut the radio signals Over long distance, the multi-path effect degrades radio signals Holes Longitudinal loss Foam Jacket Coax cables

CQA50, CQB50, CQD50: installed since 1996 mainly in injector chain.

C50-11-1: installed since 1996 mainly in injector chain.

in LHC CQA50, CQB50, CQD50 and CMC50, C50-6-1, C50-11-1 Locate mainly in safe areas but few

Locate mainly in safe areas but few closed to beams 5 Termination types Installed probably

6 What is a leaky feeder cable? 4 types of LF cables and their clamps CQG50 : installed in 2002/2003 mainly in LHC and SPS CQA50, CQB50, CQD50: installed since 1996 mainly in injector chain. 4 types of patch cables CDC50, CMC50: installed in 2002/2003 mainly in LHC and SPS C50-6-1, C : installed since 1996 mainly in injector chain. 4 types of female/male connectors Connectors for CQG50/CDC50: installed in 2002/2003 mainly in LHC CQA50, CQB50, CQD50 and CMC50, C50-6-1, C Locate mainly in safe areas but few closed to beams 5 types of splitters, 1 combiner: Installed probably everywhere since? Locate mainly in safe areas but few closed to beams 5 Termination types Installed probably everywhere since? 2 types of load, 1 Isolator,2 type of Duplexer Mainly exposed to radiations 2 antenna types Installed in few locations Potentially exposed to radiations Splitter Omnidirectional antenna CQG50 CDC50 Duplexer

7 Risks evaluation

What is the issue? Mechanical tests have been performed on samples (jacket + foam) EDMS doc: 1077149 v.

In other words, after a certain level of irradiations: The foam can easily get damaged if the cable is touched The external jacket gets burned 100 KGy Foam radiation tests How to understand these

8 What is the issue? Mechanical tests have been performed on samples (jacket + foam) EDMS doc: v.1 The foam Compression tests: Samples exposed to more than 100 kgy, the maximum force is around 3 mm compression and the samples start crumbling away. In other words, after a certain level of irradiations: The foam can easily get damaged if the cable is touched The external jacket gets burned 100 KGy Foam radiation tests How to understand these results? If the LF cable exposed to more than 100 kgy is manipulated, there is a high risks of short-cut with impossibility to repair preventing normal LF operation Any repair will also introduce additional power loss The cable will continue to work as long as the coaxial structure remains intact The foam losing its dielectric feature, first the longitudinal loss value of the cable will degraded (problematic for LHC), and the coupling loss. Picture in SPS1

9 Other issues to take into account Radiation effects on other components The connectors*, patch cables*, clamps**, antennas were not tested. This needs to be organised for the last type of equipment installed in 2002/2003. Install a LF segments in high radiation areas to test its proper functioning in live. Environmental effects Humidity + Ozone = Nitric acid This is particularly risky in confined environment Unless exception, we are not concerned. Normal aging of the cable More than thirty (30) years under normal environmental in-tunnel conditions RFS has plenty of cables with 30 years of service behind them. Damages Severe event like a fire destroying the cable can prevent communications in large tunnel sections: Could be mitigated with a second LF fire resistant cable Local damage (cable hit by transport) Local repair can be performed but will affect the max propagation distance * Teflon often presence in connectors and patch cables get degraded after few kgy ** multiple clamp break will let the LF fall

10 Risk evaluation in LHC Radiations Cumulative dose since LHC start In most cases: few kgy Worst cases (*): 10 kgy Estimated dose from 2016 to LS2 and LS3 In most cases : few kgy/y Cable aging Triplet and similar: 1-10kGy/y (peaks) Collimation: 10-50kGy/y (peaks) In 2023 (LS3), CQG50 cable will be 20 years old (Not an issue) In 2019 (LS2), other cables will be 24 years old Conclusion No issue for CQG50 LF in most cases up to LS3 (and probably above), Worst cases (Collimation), short section where a local replacement can be done but any repair work will induce significant power loss for TETRA stopping the overlapping of signals between REs segment Repeat TETRA signals from REs (?) Action at collimator locations during E-YETS (LHC7 = 3.6 km ; LHC3 = 3 km) Other LF types need further studies (probably a replacement at LS2) (*) Collimators at LHC7, LHC3 ; Inner triplets at LHC1, LHC5 Cable type CQG50 LF (installed after 2002) CQB50 LF (installed before 2002 probably in 1996) Unknown LF (installed before 2002 probably in 1996) Length (m)

11 Risk evaluation in SPS Radiations Cumulative dose since 1996 In most cases: ~60 kgy (3.1 kgy/y at cable tray level) Worst cases : TT10, BA1 and BA2 Cumulative dose since 2002 In most cases: ~40 kgy (3.1 kgy/y at cable tray level) Worst cases : TT10, BA1 and BA2 Estimated dose from 2016 to LS2 and LS3 In most cases : 3.1 kgy/y Cable aging CQB50 will be about 24 years old at LS2 In 2023 (LS3), CQG50 cable will be 20 years old (not an issue) Conclusion CQB50 LF type need probably to be replaced at LS2 CQG50 LF type to be replaced everywhere, or at least in TT10, BA1 and BA2 at E-YETS/LS2 SPS coordinator recommends LS2 instead of EYETS (too hot!) SPS coordinator recommends to have similar services than in LHC Two LFs in SPS if two in LHC The integration team is trying to do it in SPS6 Figures are uncertain In particular for BA1 BA2 Cable type CQG50 LF (installed after 2002) CQB50 LF (installed before 2002 probably in 1996) Length (m)

12 Risk evaluation in Injectors and experiments Facility Cable Type Length (m) State ISOLDE CQG50 (new) < 200 New infrastructure LINAC 400 CQG50 (new) < 500 New infrastructure LINAC III CQD50 (old) < 200 Probably no change CTF3 CQD50 (old) < 250 Probably to replace Radiations PS (at cable tray level) In most cases: few kgy/y Worst cases (*): 23 kgy/y PSB From few kgy/y to 70 kgy/y (peaks at 2500 kgy/y) Missing figures for exp. AD CQD50 (old) < 300 Probably to replace N-TOF CQD50 (old) < 500 Probably to replace DIRAC CQD50 (old) < 150 Probably to replace ISR CQD50 (old) < 700 Probably to replace BUT... Cable Aging CQD50 will 24 years old at LS2 In 2023 (LS3), CQG50 cable will be 20 years old (not an issue) PS PSB LINACII TT2 CQD50 (old) <1500 To replace Conclusion Detailed studies to performed with experiment s coordinators (on-going) DIRAC, N-TOF, CTF3, ISR, LINACIII are not equipped with TETRA services About 3400 m of LF to replace

13 Consolidation and development

14 Objective Replace the LF cable in Injectors (except SPS) during LS2 (or E-YETS) Add a second LF cable during LS2 (or even during YETS) For SPS and LHC Allow a smooth transition(s): Deployment without cutting the mobile and radio services Replacement of the existing LF cable during LS3 Replacement of the other cable for LSx Additional benefits: Fire resistant cable (cost) Radiation resistant cable (cost and 8% less propagation) For injectors and high exposed locations in LHC Coloured cable for a clear safety warning Allow higher data transfer capacity for future (4G+ and 5G?)

15 Challenges Technical constraints New cable at least 30 cm between the two LF cables The binding radius of the LF (min 500 mm) Long continuous sections Integration Finding space for a new LF cable is not easy... Drum size Can we lift the drum down to the pit? The tunnel path will be blocked deployment? Time to deploy Typically one week for an LHC sector (only tunnel) Manpower Project follow-up, integration, work supervision, cable pulling IT/CS, EN/EL, EN/MEF, Experiment coordination Budget Project not yet funded, but it seems there are consolidation budgets...

16 Work for E-YETS Double the LF in one LHC site and/or SPS site to confirm 4G/5G performances Imagine/deploy an active environment test for the LF Replacement campaign (provided RP will accept) LHC 3/7 = 6.6 km SPS (TT10, BA1, BA2) = 3 km Injectors = 5 km Material budget can be below 200 KCHF (without high radiation resistance cables) Manpower (EN-EL to make an evaluation) Single source of purchase (RFS)

17 Preparing call for tender

18 Call for tender Current provider: RFS Other European manufacturers: Nexans (FR), GeneralCable (Spain), Eupen (Belgium), etc... Selection criteria Company: Turnover, references, etc. Cable hardware Normal life-time (30 years) Resistance to radiations (have some of cable segment with better resistance) Size (1 ¼) Impedance: ohm 50 Minimum Distance to Wall (max 8 cm) Operation Temperature Minimum bending radius (50 cm) Easy of deployment Teflon free Halogen free Fire retardant Coloured cable* Cable communication features Stop bands Made for tunnels and indoor Compatible with TETRA, TETRAPOL, GSM900, U900, LTE800 Compatible with LTE-A (2 cables) Longitudinal loss and coupling loss FQ Coupling loss 450 MHz db 66/ MHz db 64/ MHz db 62/ Longitudinal loss

19 Recommendations Complete review/update of documentation and cross-check on the field Convert the documentation in new tools (AutoCad, GIS) Produce guidelines for LF deployment Develop a link-budget evaluation tool to anticipate impact of changes Perform radiation resistance tests of all components In Hi-Radmad Place samples at strategic locations in accelerators Preparing work for LS2 Write ECRs or Workunit descriptions Many: LHC, Experiments, SPS, PS, PSB, etc... Agree with FB and RP on a deployment process Integration if 2 LF Plan work with coordination

20 Thanks

High-Speed Mobile Communications in Hostile Environments

High-Speed Mobile Communications in Hostile Environments S Agosta, R Sierra and F Chapron CERN IT department, CH-1211 Geneva 23, Switzerland E-mail: stefano.agosta@cern.ch, rodrigo.sierra@cern.ch, frederic.chapron@cern.ch