Editing Channel Tunnel (section)

== Engineering ==
[[File:Channel Tunnel NRM.jpg|thumb|upright|The Channel Tunnel exhibit at the [[National Railway Museum]] in [[York]], England, showing the circular cross section of the tunnel with the [[overhead line]] powering a [[British Rail Class 373|Eurostar train]]. Also visible is the segmented tunnel lining.]]

Site investigation undertaken in the 20 years before construction confirmed earlier speculations that a tunnel could be bored through a chalk [[marl]] stratum. The chalk marl is conducive to tunnelling, with impermeability, ease of excavation and strength. The chalk marl runs along the entire length of the English side of the tunnel, but on the French side a length of {{convert|5|km|mi|abbr=in}} has variable and difficult geology. The tunnel consists of three bores: two {{cvt|7.6|m|ftin||adj=on}} diameter rail tunnels, {{cvt|30|m|ft|0}} apart, {{convert|50|km|mi|0|abbr=in}} in length with a {{cvt|4.8|m|ftin||adj=on}} diameter service tunnel in between. The three bores are connected by cross-passages and piston relief ducts. 

The service tunnel was used as a pilot tunnel, boring ahead of the main tunnels to determine the conditions. English access was provided at Shakespeare Cliff and French access from a shaft at Sangatte. The French side used five [[tunnel boring machine]]s (TBMs), and the English side six. The service tunnel uses Service Tunnel Transport System (STTS) and Light Service Tunnel Vehicles (LADOGS). Fire safety was a critical design issue.

Between the portals at Beussingue and [[Castle Hill, Folkestone|Castle Hill]] the tunnel is {{convert|50.5|km|mi|0|abbr=in}} long, with {{convert|3.3|km|mi|0|abbr=in}} under land on the French side and {{convert|9.3|km|mi|0|abbr=in}} on the UK side, and {{convert|37.9|km|mi|0|abbr=in}} under sea.<ref name="ICE p. 95">Institute of Civil Engineers p. 95{{inconsistent|reason=The book has only 65 pages: please verify the source.}}</ref> It is the third-longest rail tunnel in the world, behind the [[Gotthard Base Tunnel]] in Switzerland and the [[Seikan Tunnel]] in Japan, but with the longest under-sea section.<ref name="Daily Post 2006">{{cite news | first = Jane | last = Gilbert | title = 'Chunnel' workers link France and Britain |work=The Daily Post (New Zealand) | publisher=APN NZ Ltd | date = 1 December 2006 }}</ref> The average depth is {{cvt|45|m|ft|0}} below the seabed.<ref name="Kirkland pp.13">Kirkland p. 13</ref> On the UK side, of the expected {{cvt|5|e6m3|e6cuyd|lk=on}} of spoil approximately {{cvt|1|e6m3|e6cuyd}} was used for fill at the terminal site, and the remainder was deposited at Lower Shakespeare Cliff behind a seawall, [[land reclamation|reclaiming]] {{convert|74|acre|ha|0|abbr=off}}<ref name="Anderson Story p xvi-xvii"/> of land.<ref name="ICE p. 208">Institute of Civil Engineers p. 208</ref> This land was then made into the [[Samphire Hoe]] Country Park. Environmental assessment did not identify any major risks for the project, and further studies into safety, noise, and air pollution were overall positive. However, environmental objections were raised concerning a high-speed link to London.<ref name="Flyvbjerg p. 51">Flyvbjerg et al. p. 51</ref>

=== Geology ===
[[File:Channel Tunnel geological profile 1.svg|thumb|Geological profile along the tunnel as constructed. For most of its length the tunnel bores through a chalk [[marl]] stratum (layer).]]

Successful tunnelling required a sound understanding of topography and geology and the selection of the best rock strata through which to dig. The geology of this site generally consists of northeasterly dipping [[Cretaceous]] strata, part of the northern limb of the Wealden-Boulonnais dome. It has:
* Continuous chalk in the cliffs on either side of the Channel, with no major faulting, as observed by [[Verstegan]] in 1605.
* Four geological [[stratum|strata]], marine sediments laid down 90–100&nbsp;million years ago; [[pervious]] Upper and Middle Chalk above slightly pervious Lower Chalk and finally impermeable [[Gault Clay]]. There is a sandy stratum of [[Glauconitic marl]] (tortia), between the chalk marl and the gault clay.
* A {{cvt|25–30|m|ft||adj=on}} layer of chalk marl (French: ''craie bleue'') in the lower third of the lower chalk appeared to present the best tunnelling medium. The chalk has a clay content of 30–40% providing impermeability to groundwater yet relatively easy excavation with strength allowing minimal support. Ideally, the tunnel would be bored in the bottom {{cvt|15|m|ft|0}} of the chalk marl, allowing water inflow from fractures and joints to be minimised, but above the gault clay that would increase stress on the tunnel lining and swell and soften when wet.<ref name="Eng Geol">{{cite book | title = Engineering Geology of the Channel Tunnel | editor = Harris, C.S. | year = 1996 | publisher=Thomas Telford | location = London | isbn = 0-7277-2045-7 | page = 57 |display-editors=etal}}</ref>

On the English side, the stratum [[Strike and dip|dip]] is less than 5°; on the French side, this increases to 20°. Jointing and faulting are present on both sides. On the English side, only minor faults of displacement less than {{cvt|2|m|ftin|}} exist; on the French side, displacements of up to {{cvt|15|m|ft|}} are present owing to the Quenocs [[anticline|anticlinal]] [[fold (geology)|fold]]. The faults are of limited width, filled with calcite, pyrite and remolded clay. The increased dip and faulting restricted the selection of routes on the French side. To avoid confusion, microfossil assemblages were used to classify the chalk marl. On the French side, particularly near the coast, the chalk was harder, more brittle and more fractured than on the English side. This led to the adoption of different tunnelling techniques on the two sides.<ref name="Kirkland geol pp.21–50"/>

The Quaternary undersea valley Fosse Dangeard, and [[Castle Hill, Folkestone|Castle Hill]] landslip at the English portal, caused concerns. Identified by the 1964–1965 geophysical survey, the Fosse Dangeard is an infilled valley system extending {{cvt|80|m|ft|0}} below the seabed, {{cvt|500|m|ft|0}} south of the tunnel route in mid-channel. A 1986 survey showed that a tributary crossed the path of the tunnel, and so the tunnel route was made as far north and deep as possible. The English terminal had to be located in the Castle Hill landslip, which consists of displaced and tipping blocks of lower chalk, glauconitic marl and gault debris. Thus the area was stabilised by buttressing and inserting drainage [[adit]]s.<ref name="Kirkland geol pp.21–50"/> The service tunnel acted as a pilot preceding the main ones, so that the geology, areas of crushed rock, and zones of high water inflow could be predicted. Exploratory probing was done in the service tunnel, in the form of extensive forward probing, vertical downward probes and sideways probing.<ref name="Kirkland geol pp.21–50">Kirkland pp. 21–50</ref>

=== Site investigation ===
Marine soundings and samplings were made by Thomé de Gamond in 1833–67, establishing the seabed depth at a maximum of {{cvt|55|m}} and the continuity of geological strata (layers). Surveying continued for many years, with 166&nbsp;marine and 70&nbsp;land-deep boreholes being drilled and more than 4,000{{nbsp}}line{{nbsp}}kilometres of the marine geophysical survey completed.<ref name="Kirkland geol pp.22–26"/> Surveys were undertaken in 1958–1959, 1964–1965, 1972–1974 and 1986–1988.

The surveying in 1958–1959 catered for [[immersed tube]] and bridge designs, as well as a bored tunnel, and thus a wide area was investigated. At that time, marine geophysics surveying for engineering projects was in its infancy, with poor positioning and resolution from seismic profiling. The 1964–1965 surveys concentrated on a northerly route that left the English coast at Dover harbour; using 70&nbsp;boreholes, an area of deeply weathered rock with high [[permeability (earth sciences)|permeability]] was located just south of Dover harbour.<ref name="Kirkland geol pp.22–26"/>

Given the previous survey results and access constraints, a more southerly route was investigated in the 1972–1973 survey, and the route was confirmed to be feasible. Information for the tunnelling project also came from work before the 1975 cancellation. On the French side at Sangatte, a deep shaft with [[adit]]s was made. On the English side at Shakespeare Cliff, the government allowed {{cvt|250|m|ft|0}} of {{cvt|4.5|m|ft|0|adj=on}} diameter tunnel to be driven. The actual tunnel alignment, method of excavation and support were essentially the same as the 1975 attempt. In the 1986–1987 survey, previous findings were reinforced, and the characteristics of the gault clay and the tunnelling medium (chalk marl that made up 85% of the route) were investigated. Geophysical techniques from the oil industry were employed.<ref name="Kirkland geol pp.22–26">Kirkland pp. 22–26</ref>

=== Tunnelling ===
[[File:Eurotunnel schema (empty service).svg|thumb|Typical cross section, with the service tunnel between the two rail tunnels; shown linking the rail tunnels is a piston relief duct, necessary to manage changes in air pressure caused by the very fast movement of trains.]]

Tunnelling was a major engineering challenge; the only precedent was the undersea [[Seikan Tunnel]] in [[Japan]], which opened in 1988. A serious health and safety risk with building tunnels under water is major water inflow due to the high [[hydrostatics|hydrostatic pressure]] from the sea above, under weak ground conditions. The tunnel also had the challenge of timescale: being privately funded, an early financial return was paramount.

The objective was to construct two {{cvt|7.6|m|ft|0|adj=mid|-diameter}} rail tunnels, {{cvt|30|m|ft|0}} apart, {{convert|50|km|mi|0|abbr=in}} in length; a {{convert|4.8|m|ft|0|adj=mid|-diameter}} service tunnel between the two main ones; pairs of {{cvt|3.3|m|ftin||adj=mid|}}-diameter cross-passages linking the rail tunnels to the service tunnel at {{cvt|375|m|ft|0|adj=on}} spacing; piston relief ducts {{cvt|2|m|ftin|}} in diameter connecting the rail tunnels {{cvt|250|m|ft|0}} apart; two undersea crossover caverns to connect the rail tunnels,<ref name="Kirkland pp.63–128">Kirkland pp. 63–128</ref> with the service tunnel always preceding the main ones by at least {{cvt|1|km|mi|1|abbr=in}} to ascertain the ground conditions. There was plenty of experience with excavating through chalk in the mining industry, while the undersea crossover caverns were a complex engineering problem. The French one was based on the [[Mount Baker Tunnel|Mount Baker Ridge]] freeway tunnel in [[Seattle]]; the UK cavern was dug from the service tunnel ahead of the main ones, to avoid delay.
[[File:Channel Tunnel service road midpoint.jpg|thumb|Midpoint of the tunnel as seen from the service road]]
Precast segmental linings in the main [[tunnel boring machine]] (TBM) drives were used, but two different solutions were used. On the French side, neoprene and grout sealed bolted linings made of cast iron or high-strength reinforced concrete were used; on the English side, the main requirement was for speed, so bolting of cast-iron lining segments was only done in areas of poor geology. In the UK rail tunnels, eight lining segments plus a key segment were used; in the French side, five segments plus a key.<ref name="Wilson p.38">Wilson p. 38</ref> On the French side, a {{cvt|55|m|ft||adj=on}} diameter {{cvt|75|m|ft|0|adj=on}} deep grout-curtained shaft at Sangatte was used for access. On the English side, a marshalling area was {{cvt|140|m|ft|0}} below the top of Shakespeare Cliff, the [[New Austrian Tunnelling method]] (NATM) was first applied in the chalk marl here. On the English side, the land tunnels were driven from Shakespeare Cliff—the same place as the marine tunnels—not from Folkestone. The platform at the base of the cliff was not large enough for all of the drives and, despite environmental objections, tunnel spoil was placed behind a reinforced concrete seawall, on condition of placing the chalk in an enclosed lagoon, to avoid wide dispersal of chalk fines.{{clarify|date=November 2024}} Owing to limited space, the precast lining factory was on the [[Isle of Grain]] in the Thames estuary,<ref name="Kirkland pp.63–128"/> which used Scottish granite aggregate delivered by ship from the [[Foster Yeoman]] coastal super quarry at [[Glensanda]] in [[Loch Linnhe]] on the west coast of Scotland.

[[File:TML construction locos.jpg|thumb|2 Hunslet 900 mm gauge battery locomotives for Trans Manche Link construction trains]]
On the French side, owing to the greater permeability to water, earth pressure balance TBMs with open and closed modes were used. The TBMs were used in the closed mode for the first {{convert|5|km|mi|0|abbr=in}}, but then operated as open, boring through the chalk marl stratum.<ref name="Kirkland pp.63–128"/> This minimised the impact to the ground, allowed high water pressures to be withstood and also alleviated the need to grout ahead of the tunnel. The French effort required five TBMs: two main marine machines, one mainland machine (the short land drives of {{convert|3|km|0|abbr=in}} allowed one TBM to complete the first drive then reverse direction and complete the other), and two service tunnel machines. 

On the English side, the simpler geology allowed faster open-faced TBMs.<ref name="Kirkland geol pp.29">Kirkland p. 29</ref> Six machines were used; all commenced digging from Shakespeare Cliff, three marine-bound and three for the land tunnels.<ref name="Kirkland pp.63–128"/> Towards the completion of the undersea drives, the UK TBMs were driven steeply downwards and buried clear of the tunnel. These buried TBMs were then used to provide an [[earth (electricity)|electrical earth]]. The French TBMs then completed the tunnel and were dismantled.<ref name="Wilson p. 44">Wilson p. 44</ref> A {{convert|900|mm|0|abbr=on}} gauge railway was used on the English side during construction.<ref name="Kirkland pp.117–128">Kirkland pp. 117–128</ref>

In contrast to the English machines, which were given technical names, the French tunnelling machines were all named after women: Brigitte, Europa, Catherine, Virginie, Pascaline, Séverine.<ref>{{Cite web|last=Tempest|first=Rone|date=1 May 1990|title=Documentary : From France to England--Underground : Come with us down inside the $12-billion Channel Tunnel, where diggers have died and progress is measured in inches. When it's finished, Britain will no longer be an island nation.|url=https://www.latimes.com/archives/la-xpm-1990-05-01-wr-260-story.html|url-status=live|access-date=22 November 2021|website=Los Angeles Times|archive-url=https://web.archive.org/web/20211122222042/https://www.latimes.com/archives/la-xpm-1990-05-01-wr-260-story.html |archive-date=22 November 2021 }}</ref>

After the tunnelling, one machine was on display at the side of the M20 motorway in [[Folkestone]] until [[Getlink|Eurotunnel]] sold it on eBay for £39,999 to a scrap metal merchant.<ref>{{cite web|url=https://www.eurotunnel.com/build/ |title=How the Channel Tunnel was Built |publisher=Eurotunnel Le Shuttle |access-date=23 October 2017}}</ref> Another machine (T4 "Virginie") still survives on the French side, adjacent to Junction 41 on the [[A16 autoroute|A16]], in the middle of the D243E3/D243E4 roundabout. On it are the words "hommage aux bâtisseurs du tunnel", meaning "tribute to the builders of the tunnel".

=== Tunnel boring machines ===
The eleven tunnel boring machines were designed and manufactured through a joint venture between the Robbins Company of [[Kent, Washington]], United States; [[Markham & Co.]] of [[Chesterfield, Derbyshire|Chesterfield]], England; and [[Kawasaki Heavy Industries]] of Japan.<ref>{{cite news |last=Horvitz |first=Paul |date=6 October 1987 |title=Powerful Machines Readied for Channel Tunnel |page=C1 |url=https://www.nytimes.com/1987/10/06/science/powerful-machines-readied-for-channel-tunnel.html |work=The New York Times |access-date=10 November 2020}}</ref> The TBMs for the service tunnels and main tunnels on the UK side were designed and manufactured by [[James Howden]] & Company Ltd, Scotland.<ref name="hollingum-1993">{{cite journal|title=Howden Tunnel-boring Machine|first=Jack|last=Hollingum|journal=Industrial Robot|issn=0143-991X|oclc=473369390|date=1 June 1993|publisher=[[MCB UP Ltd]]|doi=10.1108/eb004595|volume=20|number=6|pages=33‒36|quote=This tunnel-boring machine (TBM), built by James Howden at [[Renfrew]], uses the main bearing retrieved from one of the service tunnel-boring machines from the Channel Tunnel. Howden built two of the machines for the main tunnels as well as two machines for the service tunnel in the Channel Tunnel project.}}</ref>

=== Railway design ===
[[File:ETunnelhoch.jpg|thumb|upright|Interior of the [[Eurotunnel Shuttle]], used to carry motor vehicles through the Channel Tunnel. These are the largest railway wagons in the world.<ref name="Anderson Story p xvi-xvii"/> ]]

====Loading gauge====

The [[loading gauge]] height is {{convert|5.75|m|ftin|abbr=on}}.<ref>{{cite web |url= https://intra.kth.se/polopoly_fs/1.273361.1397150299!/Menu/general/column-content/attachment/Fran-Scan%20(G2%20PC%20450)%20-%20A%20Hi-Cube%20Intermodal%20Corridor%20to%20Link%20Britain,%20France%20and%20Scandinavia%20-%20KTH%20Boysen%202011-11-30%20rev%201930.pdf |title= Fran-Scan (G2, P/C 450) – A Hi-Cube Intermodal Corridor to Link the UK, France and Scandinavia |last= Boysen |first= Hans |year= 2011 |department= Department of Transport Science Royal Institute of Technology |publisher= KTH Railway Group, Center for research and education in railway technology |page= 16 |access-date= 8 February 2019 |archive-url= https://web.archive.org/web/20190209124034/https://intra.kth.se/polopoly_fs/1.273361.1397150299!/Menu/general/column-content/attachment/Fran-Scan%20(G2%20PC%20450)%20-%20A%20Hi-Cube%20Intermodal%20Corridor%20to%20Link%20Britain,%20France%20and%20Scandinavia%20-%20KTH%20Boysen%202011-11-30%20rev%201930.pdf |archive-date= 9 February 2019 |url-status= dead }}</ref>

==== Communications ====
There are three communication systems:<ref name="Kirkland pp.129–132">Kirkland pp. 129–132</ref>
* Concession radio – for the tunnel operator's personnel and vehicles within the concession area (terminals, tunnels, coastal shafts)
* Track-to-train radio – secure speech and data between trains and the railway control centre
* Shuttle internal radio – communication among shuttle crew, and to passengers over car radios

==== Power supply ====
Power is delivered to the locomotives via an [[overhead line]] at [[25 kV AC railway electrification|{{nowrap|25 kV 50 Hz}}]]<ref name="Kirkland pp.134–148">Kirkland pp. 134–148</ref><ref name="wiki_railelectric">Article: Railway electric traction 9 August 2009</ref> with a normal overhead clearance of {{cvt|6.03|m|ftin|frac=2}}.<ref>{{cite web |url= https://www.getlinkgroup.com/content/uploads/2019/09/Eurotunnel-network-statement-2020.pdf#page=12 |title= Eurotunnel Fixed Link Usage Annual Statement – 2020 Working Timetable – |year= 2018 |page= 12 |access-date= 3 May 2021 |archive-url= https://web.archive.org/web/20210504060210/https://www.getlinkgroup.com/content/uploads/2019/09/Eurotunnel-network-statement-2020.pdf#page=12 |archive-date= 4 May 2021 |url-status= dead }}</ref> All tunnel services run on electricity, shared equally from English and French sources. There are two substations fed at 400 kV at each terminal, but in an emergency, the tunnel's lighting (about 20,000 light fittings) and the plant can be powered solely from either England or France.{{sfn|Foreign & Commonwealth Office|1994|p=9}}

The traditional railway south of London uses a 750&nbsp;V&nbsp;DC [[third rail]] to deliver electricity, but since the opening of [[High Speed 1]] there is no longer any need for tunnel trains to use it. High Speed 1, the tunnel and the [[LGV Nord]] all have power provided via overhead catenary at 25&nbsp;kV&nbsp;50&nbsp;Hz AC. The railways on "classic" lines in Belgium are also electrified by overhead wires, but at 3,000&nbsp;V&nbsp;DC.<ref name="wiki_railelectric"/>

==== Signalling ====
A cab signalling system gives information directly to train drivers on a display. There is a [[train protection system]] that stops the train if the speed exceeds that indicated on the in-cab display. [[Transmission Voie-Machine|TVM430]], as used on [[LGV Nord]] and [[High Speed 1]], is used in the tunnel.<ref name="Kirkland pp.149–155">Kirkland pp. 149–155</ref> The TVM signalling is interconnected with the signalling on the high-speed lines on either side, allowing trains to enter and exit the tunnel system without stopping. The maximum speed is {{Cvt|160|km/h}}.<ref name="wiki_de-eurotunnel">Article-de: Eurotunnel#Betrieb 9 August 2009</ref>

Signalling in the tunnel is coordinated from a control centre at the Folkestone terminal. A backup facility at the Calais terminal is staffed at all times and can take over all operations in the event of a breakdown or emergency.

==== Track system ====
Conventional ballasted tunnel track was ruled out owing to the difficulty of maintenance and lack of stability and precision. The Sonneville International Corporation's track system was chosen because it was reliable and also cost-effective. The type of track used is known as Low Vibration Track (LVT), which is held in place by gravity and friction. Reinforced concrete blocks of {{Convert|100|kg|abbr=on}} support the rails every {{convert|60|cm||1|abbr=on}} and are held by {{convert|12|mm||2|abbr=on}} thick closed-cell polymer foam pads placed at the bottom of rubber boots. The latter separates the blocks' mass movements from the concrete. The track provides extra overhead clearance for larger trains.<ref name="bonnett782">Bonnett 2005, p. 78</ref> UIC60 (60&nbsp;kg/m) rails of 900A grade rest on {{convert|6|mm|4=2|abbr=on}} rail pads, which fit the RN/Sonneville bolted dual leaf-springs. The rails, LVT-blocks and their boots with pads were assembled outside the tunnel, in a fully automated process developed by the LVT inventor, Roger Sonneville. About 334,000 Sonneville blocks were made on the Sangatte site.

Maintenance activities are less than projected. The rails had initially been ground on a yearly basis or after approximately 100MGT of traffic. Maintenance is facilitated by the existence of two tunnel junctions or crossover facilities, allowing for two-way operation in each of the six tunnel segments, and providing safe access for maintenance of one isolated tunnel segment at a time. The two crossovers are the largest artificial undersea caverns ever built, at {{cvt|150|m}} long, {{cvt|10|m}} high and {{cvt|18|m}} wide. The English crossover is {{convert|8|km||abbr=in}} from Shakespeare Cliff, and the French crossover is {{convert|12|km||abbr=in}} from Sangatte.{{sfn|Foreign & Commonwealth Office|1994|p=14}}

==== Ventilation, cooling and drainage ====
The ventilation system maintains greater air pressure in the service tunnel than in the rail tunnels, so that in the event of a fire, smoke does not enter the service tunnel from the rail tunnels. There is a normal ventilating system and a supplementary system. Twin fans are mounted in vertical shafts where digging for the tunnel began, on both sides of the channel: two in [[Sangatte]], France, and two more at [[Shakespeare Cliff Halt railway station|Shakespeare Cliff]], UK. The normal ventilating system is connected direct to the service tunnel and provides fresh air through the cross- passages into the running tunnels, where it is dispersed by the piston effect of the train and shuttle movements. Only one fan on each side is ever running, the second being available as a backup. The supplementary ventilating system is a separate emergency system and can be used to control smoke or supply emergency air within the tunnels. On both systems, the fans are normally run on supply mode, pulling in air from the outside, but they can also be used in extraction mode to remove smoke or fumes from the tunnels.<ref>{{Cite web |last=Dodge |first=Terence M. |title=Ventilating the English Channel Tunnel |url=https://www.aivc.org/sites/default/files/airbase_7421.pdf }}</ref>

Two cooling water pipes in each rail tunnel circulate chilled water to remove heat generated by the rail traffic. Pumping stations remove water in the tunnels from rain, seepage, and so on.{{sfn|Foreign & Commonwealth Office|1994|p=8}}

During the design stage of the tunnel, engineers found that its aerodynamic properties and the heat generated by high-speed trains as they passed through it would raise the temperature inside the tunnel to {{convert|50|C|F}}.<ref name="CoolingPost">{{Cite news|title=HFO chillers to cool the Channel Tunnel|date=14 September 2016|work=Cooling Post|url=https://www.coolingpost.com/world-news/hfo-chillers-to-cool-the-channel-tunnel/|accessdate=12 June 2016}}</ref> As well as making the trains "unbearably warm" for passengers, this also presented a risk of equipment failure and track distortion.<ref name=CoolingPost/> To cool the tunnel to below {{convert|35|C|F}}, engineers installed {{convert|480|km|mi|abbr=in}} of {{convert|61|cm|in|1|abbr=on}} diameter cooling pipes carrying {{convert|84|e6l|e6impgal|abbr=off}} of water. The network—Europe's largest cooling system—was supplied by eight [[York International|York Titan]] chillers running on [[Chlorodifluoromethane|R22]], a [[hydrochlorofluorocarbon]] (HCFC) refrigerant gas.<ref name=CoolingPost/><ref name="CoolingPost2">{{Cite news|title=Tunnel vision proves R1233zd efficiency|date=1 June 2018|work=Cooling Post|accessdate=12 June 2018|url=https://www.coolingpost.com/features/tunnel-vision-proves-r1233zd-efficiency/}}</ref>

Due to R22's [[ozone depletion potential]] and high [[global warming potential]], its use is being phased out in developed countries. Since 1 January 2015, it has been illegal in Europe to use HCFCs to service air-conditioning equipment; broken equipment that used HCFCs must be replaced with equipment that does not use it. In 2016, [[Trane]] was selected to provide replacement chillers for the tunnel's cooling network.<ref name=CoolingPost/> The York chillers were decommissioned and four "next generation" Trane Series E CenTraVac large-capacity (2,600&nbsp;kW to 14,000&nbsp;kW) chillers were installed—two in Sangatte, France, and two at Shakespeare Cliff, UK. The energy-efficient chillers, using [[Honeywell]]'s non-flammable, ultra-low GWP [[1-Chloro-3,3,3-trifluoropropene|R1233zd(E)]] refrigerant, maintain temperatures at {{convert|25|C|F}}, and in their first year of operation generated savings of 4.8{{nbsp}}[[GWh]]—approximately 33%, equating to €500,000 ($585,000)—for tunnel operator [[Getlink]].<ref name=CoolingPost2/>

=== Rolling stock ===
{| class="wikitable"
|- style="background:#f9f9f9;"
! rowspan="2" |Class
! rowspan="2" |Image
! rowspan="2" |Type
! rowspan="2" |Cars per set
! colspan="2" |Top speed
! rowspan="2" |Number
! rowspan="2" |Routes
! rowspan="2" |Built
|- style="background:#f9f9f9;"
!mph
!km/h
|-
!colspan=9| Eurotunnel
|-
| [[Eurotunnel Class 9|Class 9]]
| [[File:Folkestone Channel Tunnel car shuttle 3496.JPG|170px]]
| [[Electric locomotive]]
| rowspan="3" | Car Shuttle: 2 × 28<br> HGV Shuttle: 2 × 30 or 32
| 99
| 160
| 57
| rowspan="4" | [[Eurotunnel Folkestone Terminal|Folkestone]] to [[Eurotunnel Calais Terminal|Calais]]
| rowspan="3" | 1992–2003
|-
| [[Car shuttle train|Car Shuttle]]
| [[File:Channel Tunnel car shuttle 2010 3481.JPG|170px]]
| rowspan="3" | [[Passenger carriage]]
| 99
| 160
| 252
|-
| [[Car shuttle train|HGV Shuttle]]
| [[File:Eurotunnel Class 9705 - Sortie Tunnel sous la Manche à Coquelles.jpg|170px]]
| 99
| 160
| 430
|-
| Club car
| [[File:Navette Camions Eurotunnel.jpg|170px]]
|
|
|
|-
!colspan=9| Eurostar
|-
|[[British Rail Class 373|Class 373 <br> ''Eurostar e300'']]
|[[Image:3015 at Calais Frethun.jpg|170px]]
| rowspan="2" |[[Electric Multiple Unit|EMU]]
|2 × 18
|186
|300
|28
|[[St Pancras railway station|London]]–[[Gare du Nord|Paris]]<br />London–[[Brussels-South railway station|Brussels]]<br />London–[[Marne la Vallée-Chessy railway station|Marne-la-Vallée&nbsp;– Chessy]]<br />London–[[Bourg Saint Maurice]]<br>London–[[Gare de Marseille-Saint-Charles|Marseille Saint-Charles]]
|1992–1996
|-
|[[British Rail Class 374|Class 374 <br> ''Eurostar e320'']]
|[[File:Eurostar Class 374 on HS1.jpg|170px]]
|16
|200
|320
|17
|[[St Pancras railway station|London]]–[[Gare du Nord|Paris]]<br />London–[[Marne la Vallée-Chessy railway station|Marne-la-Vallée&nbsp;– Chessy]]<br />[[St Pancras railway station|London–]]{{stn|Amsterdam Centraal}}
|2011–2018
|-
! colspan="9" | Freight: [[DB Cargo]]
|-
|[[British Rail Class 92|Class 92]]
|[[File:Class-92-db-red-92009-dollands-moor-1.jpg|170px]]
|[[Electric locomotive]]
|1
|87
|140
|46
|Freight routes between the UK and France
|1993–1996
|-
! colspan="9" | Eurotunnel Service Locomotives
|-
|[[Eurotunnel Class 0001|Class 0001]]
|[[File:Railion 6456.jpg|170px]]
| rowspan="2" |[[Diesel locomotive]]
|1
|62
|100
|10
| rowspan="2" |Shunting
|1991–1992
|-
|[[Eurotunnel Class 0031|Class 0031]]
|
|1
|31
|50
|11
|1988 (as 900&nbsp;mm gauge locomotive);<br>1993-1994 (rebuilt as shunter)
|}

=== Rolling stock used previously ===
{| class="wikitable"
! Class
! Image
! Nickname/Nameplate
! Production
! Builder
! Note
|-
| [[SNCF Class BB 22200|SNCF Class BB 22200/British Rail Class 22]]
| [[File:BB22394-Antibes.jpg|120px]]
| Yellow Submarine
| 1976–1986
| [[Alstom]]
| Electric locomotives used in 1994/95 pending delivery of [[Eurotunnel Class 9|Class 9s]]<ref>First Revenue Earning Freight Through Tunnel ''[[The Railway Magazine]]'' issue 1120 August 1994 page 10</ref><ref>SNCF Class 222xxs bow out on RfD services ''The Railway Magazine'' issue 1136 December 1995 page 12</ref>
|-
| [[British Rail Class 319]]
| [[File:319058 - Bedford (8959164642).jpg|120px]]
|
* 319008: ''[[Cheriton, Kent|Cheriton]]''
* 319009: ''[[Coquelles]]''
| 1987
| [[York Carriage Works]]
| Electric Multiple Unit used on demonstration runs in 1993/94<ref>Naming Notes ''[[Rail (magazine)|Rail]]'' issue 224 13 April 1994 page 59</ref>
|}