Originally designed for subways or metros, the electro-pneumatic brake has more recently been used on main line passenger railways and some specialised freight operations. Its main advantage over the air brake is its speed of control and quick on-vehicle reaction times, giving instantaneous control of the whole train to the driver. Its speed of operation makes it ideal for automatic train operation (ATO). E.P. braking is not the same as ECP braking. ECP brakes have been introduced recently in an attempt to overcome the drawbacks of the air brake system on long freight trains. An article on this site here ECP Brakes has been written by Randy Buchter.
Background - Principles of the E-P Brake - A Simple E-P Brake System - E-P Application - Brake Cylinder Pressure - E-P Brake Release - E-P Control - E-P Variations - Self-Lapping Brakes - Retardation Controller - Variable Load Control - P-Wire Control.
Even the most modern, purely air brake systems rely on the transmission of an air signal along the brake pipe. This is initiated from the front of the train and has to be sent to all vehicles along the train to the rear. There will always be a time lapse (called the propagation rate) between the reaction of the leading vehicle and the reaction of one at the rear. This time lapse is a considerable restraint on operation. It causes the braking of vehicles to happen at different times along the train so that while some cars are slowing down, others are still trying to push, unbraked, from the rear. When releasing, the front of the train is pulling the rear, still braking, and causes stress to the couplers. Another drawback is the lack of a graduated release, an elusive goal for many years.
The introduction of electric traction and multiple unit control was the spur which eventually produced electrically controlled air brakes. The rise of rapid transit operations in cities, with their high volume and frequent stops and starts, meant that quick responses to brake commands and accurate stopping at stations was an essential ingredient in getting more efficiency. E-P brakes first appeared in the US. They were tried on the New York Subway in 1909 and then on London Underground in 1916.
Principles of the E-P Brake
There are many types of e-p brake systems is use today and most of them were developed as an "add-on" to the original air brake system and, as a result, incorporated some common principles in their design as follows:
The e-p brake operates as the service brake while the air brake is retained for emergency use
The e-p brake does not compromise the fail-safe or "vital" features of the air brake
The air brake normally remains in the "Release" position, even while the e-p brake is in "Application" and the same brake cylinders are used.
E-P brakes are invariably used on multiple unit passenger trains.
E-P brakes use a number of train wires to control the electrically operated brake valves on each car.
The train wires are connected to a brake "valve" or controller in the driver's cab.
E-P brakes should not be confused with ECP (Electronically Controlled Pneumatic) brakes. E-P brakes are used on multiple unit passenger trains whereas ECP brakes have been developed recently for use on freight trains. ECP brakes do not always require a train wire and, if they do, it is usually a single wire. A detailed description of an ECP brake system is here.
A Simple E-P Brake System
The standard air brake equipment is provided as the safety system for back-up purposes. A main reservoir pipe is provided along the length of the train so that a constant supply of air is available on all cars. A connection pipe is provided between the main reservoir and the brake cylinders on each car. An "application valve" in this connection pipe will open when required to allow main reservoir air into the brake cylinders. Because the brake pipe is fully charged during an e-p application, the triple valve is in the release position so the brake cylinder is connected to the exhaust. For e-p operation, a "holding valve" is added to the triple valve exhaust. When an e-p application is called for, the holding valve closes and prevents brake cylinder air escaping through the exhaust.
The application valve is energised and open while the holding valve works the opposite way, being energised and closed. Main reservoir air feeds through the application valve into the brake cylinder to apply the brakes in the usual way.
Brake Cylinder Pressure
It is essential to ensure that, during braking, the train wheels do not skid. Skidding reduces the braking capability and it damages wheels and rails. Wheels involved in a skid will often develop "flats", a small flat patch on the tyre which can normally only be removed by reprofiling the wheel in a workshop. To reduce the risk of skidding, brake cylinder pressure must be restricted. In a pure air brake system, a natural restriction is imposed by the maximum allowed brake pipe pressure and in the proportion of volume between the auxiliary reservoir and the brake cylinder. In an e-p equipped train, the main reservoir supply is not restricted, so it would be possible to go on pumping air into the brake cylinder until it burst. Of course, this will not happen because the brake cylinder is fitted with a safety valve (not shown in the diagram) set at the maximum pressure normally obtained in full braking.
E-P Brake Release
In the "Release" position (diagram left), both electrically operated valves are de-energised, the application valve being closed and the holding valve being open. Once the holding valve is open, brake cylinder air can escape and release the brakes. It is possible to stop the release by energising the holding valve again. This prevents any more brake cylinder air escaping. By adjusting the applications and releases of the brake during the stop, the driver is able to get a very precise stopping position. In addition, the response of the equipment to his commands is instantaneous on every car. This sort of control is essential for a rapid transit service on a metro line with frequent stops, heavy patronage and short headways.
Electro-Pneumatic brakes are controlled by the driver's brake valve handle. It is usually the same handle used to control the air brake. Electrical contacts are provided so that selection of a position will energise the train wires required to operate the e-p valves on each car, as shown left.
Current to operate the brake control is supplied from a battery through a control switch, which is closed in the operative cab. In the release position, all contacts are open and the e-p valves on each car are de-energised. In the "Application" position, the holding and application contacts are energised and the holding and application valves will be energised on each car to cause the brakes to apply. Note that the contact for the holding wire is arranged to close first so that no air will escape when the application valve is opened.
In the "Holding" position, only the holding wire is energised. If this position is selected after an application, the brake cylinder pressure remains at the value reached at that time. If after a partial release, the brake cylinder pressure will remain at the lower value achieved at that time. In effect, the driver can add or subtract air at will and can obtain an infinite variety of braking rates according to the requirements of each stop.
In all other positions, only the holding wire is energised. In reality, it is not needed to allow the operation of the air brake but it is closed anyway to act as a back up.
There have been a number of developments of the e-p braking system over the years, including a common addition - the "Self Lapping" brake. There have also been "retardation controllers" and, more recently, variable load control and single wire or P-wire control.
Self Lapping Brakes
A "self lapping" brake is really a brake controller (brake stand or brake valve, call it what you will) in the driver's cab, where the position of the brake handle between "Release" and Application" corresponds to the brake rate achieved by the equipment - in theory at least. This is similar in principle to the self lapping controllers fitted to some air braked locomotives. A number of different systems have been adopted, including one which uses a pressure sensitive valve detecting brake cylinder pressure and comparing it with the position of the brake handle. When the pressure corresponds to the position of the brake handle, the application electrical connection is opened to keep the brake cylinder pressure at that level.
Another version was developed, using a mercury filled tube inside the brake controller. The mercury was used to conduct the control current to the application and holding wires. The shape of the tube was oval and it was aligned "forward and aft" so it allowed the mercury to flow forward if the train started braking. When "Application" was called for, the movement of the brake handle towards full application tilted the mercury tube backwards and caused the holding and application valves to be energised. As the train brakes applied, the mercury detected the slowing of the train and it ran forward in the tube. This had the effect of cutting off the application so that the rate of braking conformed to the angle of the tube set by the driver's movement of his brake handle.
The mercury brake controller was an adaption of a device introduced to London Underground in the mid-1930s called the "mercury retarder" or "retardation controller".
The mercury retarder is a dynamic switch set into the e.p. brake application circuit, comprising a glass tube filled with mercury. It is mounted parallel to the motion of the train so that the mercury fluid reacts to the train's braking. The tube is curved so that the electrical contact at the base is always covered with mercury but a second contact, set higher up the rear of the tube, becomes exposed when the mercury runs forward during braking. It has the effect of measuring the deceleration rate. It cuts off application at a pre set level, no matter how much more the driver tries to put into the brake cylinders. Its main purpose was to reduce flatted wheels. It also acted as a crude form of load compensation.
In the London Underground version, two retarders were provided and they were stationary, being fixed in the driving car. They were used to regulate the rate of braking at the full application end of the range, primarily to reduce skidding and the dreaded "flats" on wheels. One retarder limited the application while the second was used to reduce the brake cylinder pressure by releasing some air through a special "blow down" valve.
Retardation controllers were later used to control braking rates on the world's first ATO railway, the Victoria Line. Four were used in all, each being set at a different angle and selected as necessary to give the required braking rate. They were also used by British Rail as self-lapping brake controllers provided on the EMU stocks built in the 1960s and 70s.
Variable Load Control
Although the retardation controller is a form of load control - because the braking rate is monitored, a heavier train will require more brake cylinder pressure, so the retarder will not reach its setting until the right rate is reached - it is rather crude. It only monitors the whole train, not individual cars. This means that lightly loaded cars in a generally heavy train are still at risk from a skid or wheelslide, as it is called. The solution is in variable load control. The car weight is monitored, usually by a lever fitted between the car and the bogie, which detects the bogie spring depression as weight increases. The lever is connected to a regulating valve in the brake cylinder feed pipe, so that the brake cylinder pressure is varied in relation to the weight of the car. With the introduction of air suspension, load control is achieved by monitoring the level of air in the suspension system and regulating brake cylinder pressure accordingly. Nowadays, the same load signals are used to vary acceleration and dynamic braking according to car weight.
As train control systems grew more complicated, more train wires were required and the traditional 10-wire jumper used by so many railways grew to the 40-wire jumper often seen today. In an attempt to reduce wiring, a novel form of e-p brake control appeared in the 1970s called the P-wire system. The brake rate was controlled by a single wire carrying pulses of different lengths to correspond to different brake rates. The pulse width was modulated to correspond to the brake demand required and it became know as the PWM (Pulse Width Modulation) system or P-wire, for short. The system was "fail-safe" in that no pulse activated the full brake while a continuous pulse kept the brake released.
No survey of the electro-pneumatic brake would be complete without a reference to the European system known as PBL90. This is not a pure e-p brake system as used on metros and suburban systems but more of an electrically assisted air brake control system. It is designed to allow vehicles with no electro-pneumatic brake controls to operate in a train with e-p control available on the locomotive or power car. For a description of the control system including diagrams, go to the PBL 90 EP Brake Control Page.