Ugrás a tartalomhoz

Electric Vehicles

Gyuláné Vincze, Gergely György Balázs

Budapest University of Technology and Economics Department of Electric Power Engineering

Energy supply of overhead line powered railway vehicles

Energy supply of overhead line powered railway vehicles

Public network connected high power substations are established along the railway track in defined distances for supplying the overhead line powered railway vehicles. All the switching, converter and protecting devices are in the substations that are required for the overhead line supply.  

The following table summarizes the railway overhead line systems, the vehicle drive system solutions and the required energy converters. According to the table there are several solutions, each electric traction mode need several energy conversion processes, therefore for the traffic designers it is difficult to find the optimal solution.

Table 3‑1.: Contact line voltages and electric energy conversion solutions.

Main converters of the substations

Overhead line voltage

Internal energy consumption: electric drive and the required internal energy conversion



phase change among segments

single-phase, standard frequency voltage

(each of the listed vehicle drives is equipped with built-in transformer, that has suited-voltage and galvanically isolated)

  1. DC motor with rectifier and split transformer

  2. DC motor with controlled rectifier

  3. AC drive with DC-link frequency converter The DC-link can be:



frequency converter

single-phase, low-frequency voltage

The application is similar to the former

(nowadays the single phase brushed DC motor without rectifier is not used)




DC voltage

(the vehicle drive is not galvanically isolated from the contact line)

  1. DC motor with switchgears, resistor gears

  2. DC motor drive with DC chopper

  3. AC drive with DC-link frequency converter


three-phase voltage

induction motor drive with gear-switches (Italian Kandó-system, it is not used nowadays)

Railway electrification systems

Features of the overhead line powered systems: supply voltage amplitude, number of the phases and the frequency; at railway applications these three together is called “current type”. Several railway electrification systems exist all over the world. If a vehicle can operate in two different systems it is called dual-voltage vehicle. In Europe six different railway electrification systems have been evolved because in different countries the railway electrification had begun separately under various conditions:    

  1. DC, 850 V: England

  2. DC, 1500 V: France, Netherlands

  3. DC 3000 V: Spain, Belgium, Italy, Poland, Slovenia, Czech Republic, Slovakia

  4. three-phase, AC: Italy (nowadays it is not used because of the problems of the current collection)

  5. single-phase, AC, 15 kV, 16 2/3 Hz: Austria, Switzerland, Germany, Sweden, Norway

  6. single-phase, AC, 25 kV, 50Hz: Hungary, France, Denmark, Great-Britain, Czech Republic, Slovakia, Croatia, Yugoslavia, Romania, Bulgaria

The content of the former list continuously changes because of the reconstructions and the installation of new systems. According to the list, in practice the voltage of the overhead line system is DC voltage or single-phase AC voltage (three-phase overhead line is already not used). Two different types of the single-phase system exist: the standard frequency and the low-frequency. It is possible when two different electrification systems can be found in one country. In each European country at the installation of the modern high-speed railways the 25kV, 50Hz supply system is used.

The following diagram presents the distribution of the European railway electrification systems. According to this diagram the ratio of the different systems is approximately the same, except the 850V DC system. This causes a significant problem at the transcontinental traffic because locomotive must be changed at the border of different electrification system that can take quarter an hour which increases the travel time.  


Distribution of the European railway electrification systems

Installing a unified European railway electrification system is not feasible because of several reasons. If either system is selected at least 67% of the European electrified lines should be adopted. It requires significant cost; moreover numerous devices (converters, supplying systems, substations, etc.) would become unnecessary. Perhaps these devices operate at the beginning of their life cycle or could be expensive. More than 50% of the electric locomotives would become inappropriate to operate causing lack of locomotives; moreover it would result in disturbances of the railway traffic. Adopting needs extremely great cost; moreover it should be performed in a short time, just in a few days.  

The cheaper solution is to buy tractive vehicles that can operate under different electrification systems. These are called dual-, triple- or quad-voltage locomotives. These shall stop for a short period at the border of the different electrification systems and switch to the proper system by keeping appropriate rules. This can be executed in one minute; therefore it does not cause significant loss of time.    

Comparing DC and single-phase railway systems

The DC railway electrification system has been evolved for DC traction motors. The 1500V voltage value was defined by the maximum permissible nominal voltage of the DC motor commutator bar voltage. The 3000V DC overhead line voltage is applicable if at least two motors are always connected in series. The low voltage level is a great disadvantage of the DC system, therefore a few thousand ampere current supply shall be provided for the high power traction. While at direct current the voltage drop of the current conductors is resistive, such a large current causes large voltage drop even in increased diameter contact line (500-600 mm2). Therefore the energy supplying substations shall be installed relatively densely, at 1500V systems the distance between two substations is 10-15km.    

In the past, at the substations the DC voltage was generated by synchronous motor driven DC generators. Nowadays it is generated by a diode rectifier connected to the public supply network through a matching transformer. Generally the contact line is at positive polarity, the rail is at negative polarity. If DC voltage supplying substation is installed by a rectifier, the braking energy cannot recuperated into the public supply network. The energy transfer is limited but it can be achieved by two vehicles connected to the same contact line. Therefore the recuperated energy (current) of a braking vehicle can be consumed by other vehicles connected to the same line and operating in motor mode.

The contact line is distributed to segments that can be separately released, the energy supply of the segments can be unidirectional or bidirectional.  

Sing le-p hase AC supply system can be operated with standard frequency or low-frequency. The nominal value of the overhead line voltage can be high (in Hungary: 25kV) and it is a great advantage of the AC systems. The built-in main transformer of a vehicle produces the most adequate voltage level for the electric drive. On the other hand the AC supply has a disadvantage, in addition to the resistive voltage drop on the overhead line there is a significant inductive voltage drop (at 50Hz the X/R~3, where X=2πfL). At high voltage transmission less current belongs to the transmitted power that causes less voltage drop in spite of the increased impedance. The substations can be installed in 30…50km.

The low-frequency (16 2/3 Hz in Europe) AC system was evolved based on the tradition of the single-phase series commutated motor traction and nowadays it still remains in several countries. It has a disadvantage, because of the low-frequency the iron core size and the weight of the vehicle’s main transformer shall be designed for a much larger value then it would be necessary at standard frequency. It has another disadvantage, the substations shall be built with frequency converters capable of the low-frequency energy supply.

Kálmán Kandó was a pioneer in applying and wide-spreading the standard voltage railway traction system. The single-phase voltage of the overhead line is produced by transformers installed in the substations that connect to one of the three-phase public supply network’s line voltage. The two phase load of the network causes asymmetry that is reduced by cyclically connecting the transformers – that supply consecutive segments – to different line voltage of two phases of the public network (Fig.3.2.).   


Figure 3-2.: Standard voltage traction system.

Simple segment isolator cannot be applied between rail segments supplied by different line voltage, because if a vehicle powered by two pantographs is running through a simple segment isolator, it can cause line-to-line fault. If the segment boundary is a phase boundary as well then no-voltage isolated overhead segments must be installed, the vehicles run through with their momentum. If the vehicle accidentally stops under a no-voltage, isolated segment then it can be connected to the next supplied segment temporarily. The standard frequency system has more advantages: the connections to the public supply network and the energy recuperation at braking can be easily achieved. The network-friendly operation and the requirement of the regenerative braking are important aspects at the design of the standard frequency railway network based modern vehicles.

Elements of the energy flow at overhead line powered vehicles

At the electric railway traction the supplied current of the substation flows through the contact wire and the pantograph and it closes through the wheels and the grounded rail (Fig.3.2.).

The rail is the part of the supplied current circuit, therefore the metallic contact, the protection against increase of the potential and in constant distances the grounding of the rails must be ensured. At DC supplying systems significant current can flow out of the rail, if the resistance of the parallel current paths is less or comparable with the resistance of the rail. Particularly dangerous if the leak current flows through conductive pipes or metal-sheathed cables because if the ground is wet the current can perform electrolysis at its entrance and exit places on the metallic parts that causes corrosion. This phenomenon does not exist at AC voltage systems, because of the high impedance of the parallel current paths.

Wheel-rail circuit. Without any measure the motor current would flow to the rail through the bearings and the wheels. The bearings can be damaged, therefore generally the current is conducted to the wheels through a slip ring – brush installation by bypassing the bearing box, the number of the slip rings depends on the amount of the current.   

The pa ntograph has a sliding shoe connector that is pressed to the contact line by a sprung, armed and hinged mechanism. It can flexibly adapt to the instantaneous height of the contact line (the sag is 15-25cm). The lever apparatus is flat if it is folded and always possesses with an element that can break if the pantograph get stuck by accident. Two kinds of current collector is widespread the pole and the pantograph. The pantograph is used at railways. The pantographs have two different types: Z-shaped (asymmetrical) and diamond-shaped (symmetrical). The icing, the frosting and the pollution influence the life cycle and operational reliability of the pantographs.

The overhead wire generally consists of a contact wire and a catenary wire. The catenary wire is produced from a high mechanical strength material and it suspends the contact wire. The catenary wire and the contact wire is not isolated, this make a parallel current carrying branch. The contact wire contacts with the pantograph, it is approx. 120 mm2 solid copper wire in Hungary at 4.8…6.5m above the vehicle. If the contact and the catenary wires are viewed from above, their tracing is zigzag shaped and has a symmetrical ±0,5m deviation from the center line of the track to even the wear on the pantograph’s shoe.

The contact line is distributed to segments that can be separately released. The rails are grounded and cannot be distributed to segments. The current can leave the rail and can flow in the ground as leak current until the return wire.

The contact wire shall be selected according to the mechanical and electrical features. If the mechanical aspects are considered, the contact wire shall be bearing, weather proof, well-mountable, and shall possess with adequate strength to endure the stress caused by the moving pantograph. If the electrical aspects are considered the contact wire shall have the better conductivity. The voltage level defines the insulation and the breakdown strength that shall be kept. The designed current stress of the rail line defines the diameter of the overhead line.

Multi-voltage locomotives

The realization of the long-distance transnational rail traffic is difficult because of the different railway electrification systems. If the locomotive can only operate in one kind of system, then it shall be changed on the border of the system. Multi-voltage locomotives can operate in different railway electrification systems, the change between the systems can be performed by an electric switch.

In the locomotives operating in two DC voltage systems (e.g. from 1500V or 3000V) two exactly the same drive systems are built that are designed for 1500V supplying voltage. The two drive systems are connected in parallel if the locomotive is powered from the 1500V overhead line, and these are connected is series if the locomotive is powered from the 3000V system.  

In th e locomotives operating in two A C voltage systems ( e.g. from 15k V, 16 2/3Hz or 25kV, 50Hz) an on-board special transformer or a group of transformers is connected to the overhead line. The voltage ratio can be changed by the transformer; the change remains undetected for the electric drives. Two different solutions of the switching between two voltage systems are presented in Fig.3.3.


Figure 3-3.: Switching methods for dual-voltage locomotives operating in standard and low-frequency AC voltage system a./ Switching the primary number of turn, b./ Switching the secundary number of turn.

Fig.3.3.a. presents a solution operating with swi t ching the primary number of turn. The iron core and the primary number of turns of the transformer shall be designed for 16 2/3Hz, 15kV supplying system. The nominal primary current of the transformer shall be also defined for the 16 2/3Hz, 15kV supplying system. The transformer secondary voltage shall not change at 50Hz, 25kV supply, therefore an increased primary number of turn coil (ratio: N 2 /N 1=25kV/15kV) shall be connected to the overhead line. Therefore the secondary voltage does not change, but the magnetic stress of the iron core, its flux density is only 1/3 at 50Hz supply. If same vehicle power and phase constant is assumed, at 25kV, 50Hz supply, the transformer primary current will be reduced with the rate: 15kV/25kV. The (N 2 -N 1) additional turns can be designed for this reduced current. Compared to the 50Hz supply, the transformer shall be significantly overdesigned, considering the flux density and the current also.

Fig.3.3.b. presents a solution operating with switching the secondary number of turn. In this case, similarly to the switching the primary number of turn, the transformer shall be designed for 16 2/3Hz, 15kV supplying system. The three times higher frequency 25kV voltage can be switched to the primary coil without difficulties and the magnetic stress of the iron core will be even approximately 50% at 50Hz (25kV). The secondary voltage would increase at 25kV supply; therefore on the secondary coil a 15kV/25kV ratio tap change is required, at 25kV from K1 to K2 shall be changed. The switches are on the secondary part; therefore a separate switch is required to each secondary coil. The switches shall be designed for much higher secondary current than then primary current. Other solutions exist, but the presented two switching methods are the most common.  

In the locomotives operating in two AC and one DC voltage systems the switch between the 15kV, 16 2/3Hz and 25kV 50Hz supplying system can be performed by transformer switch that was detailed above. The third voltage system - that the vehicle can be made suitable for– is e.g. the 1500V DC supply. It can be realized if the vehicle drive system is operating with a DC-link, and its voltage is designed for the DC overhead line voltage level. In this case the DC-link can be connected directly to DC overhead line.

Network-friendly operation for AC voltage powered rail vehicles

The network-friendly operation is a general requirement of modern standard- or low-frequency AC voltage powered vehicles. Those devices are called network-friendly consumers which cause minimal harmonic distortion on the network, and consume the P 1 =U 1 I 1 cosφ1 active power - required for their operation – with minimal I 1 current (index 1 refers for the fundamental harmonic quantities), i.e. the cosφ1 power factor well approximates the cos φ 1=±1 value. This involves that the motor mode is optimal if the phase angle is φ 1=0°, i.e. the consumed current is in phase with the network voltage, and the regenerative braking mode is optimal if the phase angle is φ 1=180°, i.e. the current is in antiphase with the network voltage.

In the past the control for the optimal power factor could be solved by rotating machines, e.g. in single-phase synchronous motor driven, DC generator powered Ward-Leonard type locomotives, because the built-in synchronous motor – that was connected to the network - is able to correct the phase angle with the control of the excitation. The Ward-Leonard type locomotives were withdrawn from the traction, but they are still applied for phase compensation (as a synchronous compensator) in busy railway junctions.

Nowadays the network-friendly operation is solved with electronics, with a 4qS (four-quadrant-supply) converter, that is connected to the network.  It is a single-phase voltage source inverter type  an AC/DC converter which is controlled by pulse width  modulation (PWM). Generally the 4qS is applied with voltage source inverter fed induction motor driven vehicle drive system, as it is detailed afterwards in the examples. Fig.3.4.a. represents the simplest circuit of the IGBT transistor based 4qS converters.


Figure 3-4.: 4qS network-friendly supply a.) electric circuit, b.) time function of main electric signals

The input of the 4qS is connected to the secondary voltage of the vehicle’s main transformer, and the output is connected to the DC-link voltage of the vehicle, that feeds the inverter powered vehicle drive. The 4qS converter is responsible for the network-friendly operation in motor and brake mode, and for controlling the DC-link voltage. The reference value of inverter powered drive DC-link voltage is U ea that is approximately 1,2…1,5 times higher than the secondary voltage peak value (u sz1). If the u e voltage of the C capacitor is smaller than the reference U ea value, then the FSZ voltage controller gives charging command, if the u e is higher than U ea, then the FSZ gives discharging command. It is realized indirectly by a cascaded current control circuit. The controller output does not specify directly the i d DC current required for the change of the (i d -i e) charging current, but it specifies the i sz1 current through the AJK reference signal generator. Based on the amplitude and sign of the voltage controller output signal, the AJK specifies the network current reference signal for the ÁSZ current controller, that has a sinusoidal shape and its phase angle is φ =0° if C has a charging demand (i.e. in motor mode), and its phase angle is φ =180° if C has a discharging demand (i.e. in brake mode). So the phase angle specification depends on the required power flow direction for the DC voltage control.


Figure 3-5.: Simulated results of 4qS voltage and current time functions (brown: usz1[V], blue: isz1[A], green: Ue[V], red: ie[A])

The first part of Fig.3.5 shows motor mode, the C shall be charged to keep the DC-link voltage level, the consumed i sz1 secondary current is in phase with u sz1, P>0. The second part of Fig.3.5 shows regenerative braking mode, where i sz1 is in antiphase with u sz1, P<0.

In the 4qS circuit (Fig.3.4.a.) the i sz1 secondary current control is fulfilled by PWM modulation, where the feasible switching configurations of the IGBT transistors are as follows:

  1. if T1-T4 are switched on → uv=ue and id=isz1, independently of the current direction,

  2. if T2-T3 are switched on → uv=-ue and id=-isz1, independently of the current direction,

  3. if T1-T3, or T2-T4 are switched on → uv=0 and id=0.

The current control is performed by these switching cases. Fig.3.4.b presents typical time functions of the control purpose fulfillment in motor mode, where i d>0 (charging current direction), the consumed power from the network is P>0. The harmonic content of the controlled isz1 secondary current depends on the switching frequency and the value of the L filter choke. If the vehicle drive consists of a twin drive connected to two individual secondary coils (u sz1 and u sz2) as it is shown on Fig.3.4.a, shifting the PWM pattern of the two individual 4qS converters can be applied for the reduction of the network harmonics.

According to the figure, the i d charging current of the C capacitor is pulsating with 2f H frequency, therefore the u e voltage is pulsating with the same frequency. For reducing the pulsation, in some vehicles a L sz -C sz filter is built in, tuned for 2f H frequency (at 50Hz frequency for 100Hz).

Nowadays there are still several vehicles powered from the single-phase AC electrification system that do not possess network-friendly, controlled supply, but in these vehicles it is tried to reach more favorable features for the network.