The International Light Rail Magazine
+44 1733 367610

Considerations for depot design

The wind turbines at Angers’ Raymond Perron Centre Technique des Transports are just one example of using innovative sustainable energy technology to create an efficient facility; it is soon to be expanded as the French city seeks to double the length of its system. Neil Pulling

Given the number of new tramway systems that are being built – and the number of existing systems that are being extended and/or radically renewed – there is an ongoing problem that the amount of practical experience and design talent available to the industry is being spread rather thinly.

One can see that many new systems are simply copying that which went before and similar mistakes or inefficiencies are repeated. What is needed is a ‘Tram Depot Design for Dummies’ handbook that demonstrates the good, the bad and the ugly amongst existing designs, supporting its recommendations with a wide range of detailed data.

Such textbooks have existed and the US Electric Railway Handbook, last revised in 1924 and reprinted in facsimile in the 1990s, still has many useful pointers for designers, notwithstanding the huge changes in technology that have come about since it was first written. The US Transportation Research Board has examined the issue of depot location and design on a number of occasions over the last three decades and such documents may be purchased from its website.[1]

In Europe, the VdV alliance of transport companies has published a number of advisory papers, which can also be downloaded. These cover various aspects of depot planning, including considerations of internal transport and storage, traffic infrastructure and parking space.[2]

Many depots and maintenance centres – especially those on larger systems or those that are combined with bus or trolleybus operations – operate 24 hours per day. As potentially hundreds of employees will be working on site at all hours, the provision of a safe and comfortable environment is key to a facility’s success without unnecessarily increasing a site’s footprint.

With a modern depot costing anywhere from EUR20m upwards, these are major investments and must therefore be well-planned with close consideration of function as well as form, not only for current operations but also with future development in mind.


The most recent international study of depot issues has been undertaken by TramStore21, a co-operative project between the Netherlands, the United Kingdom, Belgium, France and Germany, co-funded by the partners of the project and the European Union Development Fund under the INTERREG IVB co-operation programme.

Between 2008 and 2013, the European partners worked to pool existing best practices aiming to improve transport efficiency and the urban integration of tram depots. New tram depots in Brussels, Rotterdam, Blackpool and Dijon feature, including integration with a bus depot or a park-and-ride, specific measures to deter corrosion, sustainable designs, and peer reviews at the planning stage. The project was intended to encourage innovative practices such as local authorities working in an open manner in an international setting, accepting peer reviews in the planning process (which can be more cost-effective than hiring in expert consultants), overcoming language barriers and confidentiality clauses in contracts.

Unfortunately, much of this work has consisted of rapportage, simply recounting what has been done on different sites rather than a gathering of data and undertaking a rigorous analysis of the effectiveness or value for money of a project. If research funding could be found and various transport undertakings were willing to give accurate and honest data on the costs and benefits of their depot strategy then a rigorous analysis and commentary could be provided that would be of inestimable value to tramway developers.

The consensual nature of much working within the European Union also tends to reduce the willingness of participants to review the work of colleagues with a critical eye and to challenge assumptions and established practice. It must also be recognised that much work has been carried out in wealthier countries that are willing to invest money in high quality public services. The current political/economic conditions may mean that all EU partners adopt the penny-pinching attitudes evident in the UK outside London. We need to maximise the ‘bang for each buck’ if tramway projects are to have a future.

Given the scope of the subject, this article can only touch on a few issues, but I hope that this will result in reflection and debate about how best to locate and design a depot.


Ideally a depot should be located as close as possible to the centre of a system to minimise the amount of dead running at the beginning and end of the traffic day. Older readers may remember the Mersey Square Depot in Stockport, similarly Rigby Road was fairly central to the Blackpool system – particularly when the inland routes were running – and today the Therapia Road depot on London Tramlink is only approximately 2km (1.2 miles) from the town centre loop. Unfortunately land values in city centres can be so high that sites for depots are unaffordable. Where sites can be found there is often objection from local residents who fear noise and light pollution from depot operations.

If a central depot is not possible then some of the effects of out-of-centre depots can be mitigated by careful timetable planning to ensure the best revenue mileage/dead mileage ratio is achieved. For example, the more robust nature of modern car designs, with longer periods between routine maintenance means that it is not necessary to have all the trams return to the central maintenance depot every night. An efficient maintenance works, coupled with simple, secure tram stabling on each route can be acceptable if each car’s working diagram is adjusted to ensure that it passes through the maintenance works at regular intervals.


Given that many depot designers come from a heavy rail background it is not surprising that many modern tram depots have a through-flow layout, with a fan of turnouts at each end of the sidings. Indeed VdV Recommendation 823 is that ‘track crossings should be avoided through anti-clockwise vehicle driving (right driven tracks) and it is advised to connect the different depot units directly, thus achieving more efficiency’.[3]

This layout may be useful when one is dealing with long trains of vehicles, but it is worth questioning whether it offers efficient land use and value for money on tramway systems. It is claimed that with a dead-end siding a single failed car can trap others behind it; however in 30 years in the industry I have never seen any data on the frequency of this happening. Lack of data leads me to conclude that this is an ex post facto justification for double-ended sidings. It is notable that of the four depots examined by TramStore21 21 two (Blackpool and Dijon) are single-ended, as are the modern depots in Nice, Tenerife and Lille.

What can be much more frustrating is a derailment during morning run out – this can block a whole depot throat. If the layout uses ballasted track then the car can settle on the ballast and will require lifting, packing and slewing to get it back on the rails, a time and labour intensive process. However if the depot fans are laid in paved track often all that is needed is a reversal of the car to get it to drop back into the grooves. If this fails a little ‘persuasion’ with another car and some stout lads armed with levers can often sort the problem out. An examination of the paving around the decaying track fan at Rigby Road, Blackpool will show how often trams have suffered a minor derailment and then been run back onto the track.

Covered stabling?

Most modern depots take advantage of the weather resistance of 21st Century trams by stabling the cars in the open. This is obviously the cheapest solution, but it can result in an increased need to wash the cars and a more rapid deterioration of any applied vinyl graphics. In colder countries outdoor stabling means the cars absorb more energy when they are heated and condensation is driven out before cars enter service – and the conditions for cleaning staff or for drivers leaving or returning to the cars is poorer.

A roofed depot contains more of the operating noise and light pollution that can be a source of complaint from local residents. A fully-enclosed building also offers increased security in areas suffering from vandalism.

However, a roof can be very expensive to construct and maintain. The ideal solution is to find a source of revenue from leasing out the air space above. In Tuen Mun, Hong Kong, this space accommodates a number of 29-storey residential towers, but it is rare that land is scarce and expensive enough to make this worthwhile. The need for utility and services access to the buildings – and the great weight of the structures – means that a massive and expensive structure will be required to support the development.

A better solution is to provide a lightweight structure that can be used for a low-rise car park. Prefabricated buildings like this are often used to provide additional parking on space-constrained hospital and supermarket sites. The car park can be used for staff parking and to form the basis of a park-and-ride site, generating extra revenue for the tramway. The allocation of reserved parking to residents of the immediate area can often help reduce local objections to the structure and also reduce unsightly on-street parking. Such an approach can mean that everyone gains, turning opponents of the scheme into supporters.

Equipping the Maintenance Hall

The choice of rolling stock can have a significant effect on depot design and the amount of space required can be influenced by the type of bogie and the accessibility of equipment cases. Thus many depots have large fixed roof-level walkways and plenty of air space to allow work to be carried out on control and auxiliary equipment fixed to the roof. However other designs (most notably the Eurotram supplied to Strasbourg, Milan and Porto) have removable equipment cases that can be lifted from the car. This approach minimises the number of gantry workspaces, their size and headroom requirements.

Access to a wheel lathe is essential for the efficient running of a modern tramway. The complex layout of traction motors (particularly in earlier monomotor cars) means that all the wheels on a bogie have to be exactly the same size, and difficulty in getting the bogie out from under a car means that an underfloor lathe is usually specified. Such a lathe is an expensive item and needs to be put in a shed at least twice as long as the longest car specified for the system, so that all the wheels can be dealt with whilst the shed doors are closed. Where mistakes are made and doors have to be opened this can let bad weather in and noise out into the local community.

Thus if the operator expects that it will ever have to run cars in the 40-45m length range it will need to provide a building over 90m long, a large structure for a small-start system. It is to be hoped that innovative designs with individually controlled wheel-hub motors may allow easier tolerances and wheels to be machined singly off the car, resulting in savings in plant and buildings.


Car washing

The exterior cleaning of trams and LRVs is an important issue. The outer appearance is the first impression given to passengers as these are the moving advertisements for any system, so there is good economic sense in having them look their best.

There are two main types of wash plant: drive-through and gantry. The drive-through type is most common as it takes less time (for example, Dijon claims 2.5 minutes for a 30m tram, in contrast to six minutes required for a gantry-type machine). Drive-through plants are good at cleaning car sides and can handle simple, smooth end-shapes, but are generally less-effective at cleaning roofs, especially if they feature a jumble of equipment cases.

The washing facility location is generally on the way to a train stabling area and should be directly linked to the depot entrance. Washing the car on the way into the depot means that trams are cleared of potentially damaging dirt at the earliest possible moment and any delay at the wash plant does not delay revenue service. However, this can result in a tram standing wet on the stabling sidings overnight; as it cools it will tend to draw cold, damp air into the car and equipment cabinets, at worst potentially damaging electrical equipment.

Washing the car on the way into service means that cars ‘blow dry’ as they travel to their start of duty point and damp air is drawn away from the saloons. However, this requires the tramwash to have a quick throughput and the provision of an emergency bypass track
in the event of a delay.



With strong political implications, it is of little use to proclaim the green credentials of light rail schemes if auxiliary functions are consuming scarce resources. Of course the construction phases can include the use of recycled and renewable materials in areas such as insulation, noise mitigation, surfacing and paving, but other key areas include the reduction of water and energy usage.

Rainwater can be harvested from the roof of depot buildings and used for most purposes, other than drinking, in the depot. This is one advantage of having covered stabling, albeit the Starr Gate depot at Blackpool does not achieve this as well as it might as the proximity of the sea means that the rainwater is contaminated with salt. Of course the run-off from a car park built over the depot can also be used, once treated to remove fuel and oil contamination, for most purposes.

One significant saving can be effected by recycling the water from the tramwash; modern plants can recycle over 80% of the water used, reducing water demand and cost. In France, the recycled water cost is estimated at about EUR0.2/m3. By comparison, the current cost of drinking water in Dijon is approximately EUR3.40/m3. The water has to be filtered and stripped of contaminants between each use and this can lower the throughput of the wash unless a large buffer tank is provided.

Another useful environmental activity is the harnessing of sun and wind to generate electricity. The Kirnitzschtalbahn in Germany has been using solar cells on the roof of its depot for over a decade; these contribute approximately 20% of the necessary electricity for the system’s operation. The tram depot in Tenerife has had solar panels installed since it was built in 2005. In accordance with Metropolitano de Tenerif policy, the rail system – train and LRT – aims to be 100% self-supporting in terms of energy supply by 2020.

In the US, the installation of a solar-voltaic plant capable of generating 780kW per annum has recently been completed at the Valley Metro LRT operations and maintenance facility in Phoenix, estimated to save approximately 16% (USD100 000 average per year) in energy consumption.

Another example is the depot in Angers, opened in 2011, that was specified by Angers Loire Métropole to be a ‘positive energy’ structure – therefore producing more electricity than it consumes. The installation of 400m2 of photovoltaic solar panels help achieve this aim; the recently tendered 9.9km (6.2-mile) line B also includes provision for an expansion of this programme to meet an expanded site’s increased energy requirements.

Additionally, a pair of efficient wind turbines generate electricity, while thermal solar energy panels provide the facility’s hot water. In a further innovation, a prototype autonomous photovoltaic system provides power for lighting of the site’s car park.

Tram systems tend to have small areas of unused land around the edges of depots and permanent way yards etc, and these could be ideal for the installation of wind turbines or solar plants to provide additional green energy.

The topics above only refer to a number of the more obvious design aspects, but they do show the importance of an integrated approach to the specification, design and procurement as poorly thought-out decisions can have knock-on effects all over the system.

Careful consideration must also be given to the delivery and storage of spares, social and training facilities and the location of Control Centres. It has become standard practice to
co-locate these within modern depots, but land availability may make it necessary to disperse them across the system. It is hoped these issues can be addressed in a future article.


 Feature originally appeared in October Tramways & Urban Transit (934).



2 cf. VdV Recommendation 823: Recommendation on the Design of Depots for LRVs and Tramcars, Köln 10/01 2