TRACK CIRCUITS

CONTENTS

1. Introduction

2. Principles of Operation

3. Practical Considerations

4. Insulated Block Joints

5. Track Circuit Types

6. D.C. Track Circuits

7. AC. Track Circuits

8. Jointless Track Circuits

9. Impulsing Track Circuits

10. Rail Circuits and Overlay Track Circuits

11. Shunt Assisters

12. Alternatives to Track Circuits

1. INTRODUCTION

The track circuit is an important component of  modern  signalling systems. Its main purpose is to continuously detect the presence and position of traffic within the signalled area.

Track circuits are electrical circuits formed in part by the rails of the track. The detection of vehicles is possible because their wheels and axles reduce the rail to rail resistance significantly below the value relating to a clear track condition.

Eventhough modern communication based train control system make use of track installed transponders otherwise know as balises or beacon (French)  as a primary  means to identify the position,over laid secondary train systems are implemented (Can be axle counter as well) to recover the train ,if communication is lost and primary train detection is impossible.

The first track circuits were used only as indication devices, to remind the signalman of the presence of a train, for example standing at the home signal or on a section of track which might be obscured from his view.

Later they were used to lock facing points, as a replacement for fouling bars. In this role, the track circuit directly controlled the operation of the point lever through an electric lock.

Track circuits are now used to control the signalling directly in legacy fixed block signalling system,by preventing a signal being cleared if the track ahead of the signal is occupied by a train, by holding the route after a train has passed the signal, and to lock a set of points if a train is standing on them.

It is important to remember that a track circuit is designed primarily  to detect the absence of a train and not its presence. There are however many signalling controls which require track circuits to be occupied (e.g. release of overlap after a train has come to a stand) and the track circuit must therefore be reliable enough to be used for this purpose also.

A rail circuit is the reverse of a track-circuit, and is used to prove the presence (not absence) of a train at a particular place. Rail-circuits are not as commonly used as track circuits.

To understand the basic principles of track circuit operation, the easiest type of track circuit to study is the simple, battery-fed D.C. track circuit. Many other types have been developed to cover particular situations such as electrified areas, jointless track etc. These will be discussed later. Most of the principles of the simple D.C. track apply equally to other types, although the configuration of the equipment may differ.

2. PRINCIPLES OF OPERATION

A track circuit  employs  the rails as a transmission  line. The electrical  signal, in this case a d.c. voltage, is applied to one end. At the other end a relay is connected  across the rails as a receiver of the voltage applied at the opposite end.

The physical limits of the track circuit must be defined by insulated rail joints to prevent interference between adjacent track circuits or isolate the track circuit from sections of line where track circuits are not required.

When there is no train on the track circuit, the relay will be energised via the full length of each rail. When a train stands on the track circuit the wheels and axles create a short circuit between the rails. The current takes the lower resistance path via the train and the relay becomes de-energised.

All track circuits depend for their operation on the wheels of a train forming a low resistance path between the rails.

Current from a battery (or rectified power source) at one end of the section, energises a relay at the other end via an adjustable resistor and the rails. Each rail forms one leg of the circuit. Although the rail joints are metallically joined by fishplates, they are electrically unreliable and it is therefore necessary for the rails in each leg to be bonded together to form a good path for conduction of the electrical current. The end rails of the track circuit are insulated from the adjoining rails by means of insulated rail joints. The track circuit is an independent circuit with its own power supply at one end, and its own relay at the other end. Normally, when the track is clear of trains, current flows through the rails and through the coil of the tractive armature electromagnetic relay, thereby energising the relay which closes its front contacts.

When a train occupies the track circuit, the two rails of the track circuit are short-circuited by the axles of the train. The relay is shunted by an electrical path of extremely low resistance, provided by the wheels and axles of the train, the result being that virtually no current flows through the relay coil because all but an insignificant amount is diverted through the axles. The relay becomes de-energised and the armature drops to close the back contacts.

Thus the condition of the track relay determines whether the track circuit is clear or occupied. The most important feature of the track circuit is the fact that it fails to safety, i. e. is "fail-safe". The moment a rail or a wire breaks, or a metallic obstruction straddles the track, or the power supply fails, the relay will become de-energised indicating the track occupied condition

 

Figure 1 Track Circuit Clear (relay Energized) -Simple DC Track Circuit

Figure 2 Track Circuit Occupied (relay De-Energized) -Simple DC Track Circuit

 

The relay is therefore energised to prove that there is no train on the track,  and de-energised when there is a train on the track.

The relay will also de-energise if:

1. a) A rail connection becomes detached
2. b) The power supply fails
3. c) A rail to rail bond is broken or becomes detached
4. d) A rail breaks

The circuit is therefore designed to be "fail-safe", as under most fault conditions it indicates that the track is occupied. This may not be totally true in practice, as will be seen later.

3. PRACTICAL CONSIDERATION

Figure 3 Practical Track Circuits 

A practical track circuit differs significantly from the theoretical example shownin Figure 1 & Figure 2  The two major difficulties are:-

1. a) The rails are not perfectly  insulated  from each other. Current "leaks" from one rail to the other, through the rail fixings, the sleepers and the ballast. This rail-to-rail resistance is known as "ballast resistance". It can  vary  widely 300  metres  of standard gauge (1435mm) track may  have a ballast  resistance  as  high  as 50  ohms for well ballasted, dry track, or as low as 0.5 ohm with waterlogged or dirty ballast.
2. b) The presence of the wheels on the rails may not provide a perfect short circuit. The axles may have a small resistance, but the main problem is the contact between wheel and rail. This may be due to rust or other causes such as fallen leaves ,or ice on track during winter .

Although the rails also have a resistance, this is  normally  insignificant  compared  to  the above two problems. Rails will also have an inductance. This  will  be  significant  when  dealing with some types of a.c. track circuit.

3.1        Ballast Resistance

This is a factor over which the signal engineer usually has very little control. It will vary according to the type of ballast, its condition, the type of sleepers, the method of fixing of rails to sleepers, track drainage and, most importantly, the weather conditions.  These will all vary in time during the life of the track circuit. Some factors will change slowly, such  as the condition of the ballast. Others, like the amount of moisture in the ground, can vary rapidly.

The design of the track circuit equipment must be able to compensate for the long term changes by some means of manual or automatic adjustment. The short term changes due to weather must be contained within the normal operation range of the track circuit. If not, failures (either right side or wrong side) will result.

The inclusion of a series resistance at the feed end is an essential feature of most track circuits to avoid excessive power consumption when the track is occupied. It also permits components of a lower current rating to be used. This will also make the track circuit more difficult to set up for reliable operation. As ballast resistance decreases, the current drawn will increase, the voltage drop across the feed resistor will increase and the relay voltage will be correspondingly reduced. Too large a reduction in ballast resistance could cause a right side failure of the track circuit.

Under the most adverse conditions, the minimum ballast resistance should not be less than 4 ohms per 1000 ft. Short track circuits will of course have a higher ballast resistance.

The value of ballast resistance may be calculated from the measurements taken for rail resistance: -

Vf = Volts across rails at feed end.

Vr = Volts across rails relay end.

It = Current to track.

Ir = Current to relay.

Mean voltage across rails = (Vf + Vr) / 2

Current in ballast = It - Ir

Ballast Resistance = (Mean voltage across rails) / (Current in ballast)

For a typical DC track circuit calculation, the following figures for a track circuit will be used: -

EMF of battery                                                          1.2 volts.

Resistance of relay                                                    9 ohms

Pick-up value of relay                                               0.36 volts

Drop-away value of relay                                         0.27 volts

Minimum ballast resistance                                    2 ohms

Working voltage 25 per cent above pick-up          0.45 volts

Current to relay                                                        = Voltage across relay/Resistance of Relay

                                                                                  = 0.45/9  = 0.05 ampere.

Current to ballast                                                    = Voltage across rails/Ballast resistance

                                                                                  = 0.45/ 2 = 0.225 amp.

Current to track                                                      = 0.05 + 0.225= 0.275 ampere.

 

Voltage across feed resistance                               = EMF of battery - EMF across rails

                                                                                 = 1.2 - 0.45 = 0.75 volts

Value of feed resistance                                         = Voltage across feed resistance/Current to track

                                                                                 = 0.75 / 0.275 = 2.73 ohms

The value of the resistance which, when placed across the rails of the track circuit with a ballast resistance of 2 ohms, would reduce the voltage across the relay to the drop-away value, i. e. 0.27 volts, can be calculated as follows

Voltage across feed resistance                              = 1.2V - 0.27V = 0.93 volts.

Current to track                                                     = 0.93V/2.73= 0.341 ampere.

Current to relay                                                     = 0.27V/9 = 0.03 ampere.

Current to ballast                                                  = 0.27V/2 = 0.135 ampere.

Current to relay and ballast                                  = 0.03 + 0.13 5

                                                                               = 0.165 ampere.

Current to shunt resistance                                 = 0.341 - 0.165 = 0.176 amp.

Shunt resistance                                                   = Voltage across track/Current through shunt= 0.27/ 0.176

                                                                               = 1.53 ohms.

The above figures are given as an example to show how the various values can be obtained.

3.2       Rail Resistance

This is the total resistance of rails, fishplates and bonds. Its value remains appreciably constant. The rails of a track circuit form the conductors of a simple circuit in that current is supplied at the feed end and less current appears at the relay end. The loss in current is due to leakage and the loss in voltage to rail and cable resistance. The fall in current and voltage is not uniform along the length of the track circuit, but falls faster near the supply end than the relay end. In most cases, however, the rail resistance is small compared with the ballast resistance and it may be assumed that the voltage and current fall off uniformly along the length of the track circuit.

The value of the rail resistance in a track circuit may be obtained from the following measurements

Vf                                                                         = Rail voltage at the feed end of track circuit.

Vr                                                                         = Rail voltage at the relay end of track circuit.

It                                                                          = Current to circuit.

Ir                                                                          = Current to relay.

From Ohms law it is known that:

Rail resistance = (Voltage drop in rails) / (mean current in rails)

3.3       Train Shunt Resistance

This is the joint resistance of all the parallel paths from rail to rail of a train or vehicle. The train shunt of a vehicle is greater than that of a train and that of a pair of wheels greater than that of a vehicle.

It should be noted that the ordinary resistance of a pair of wheels and axle is not the same as the train shunt of a pair of wheels resting on the track, due to films of oil, scale and rust frequently found between the wheels and rail.

3.4       Train Shunt

The train shunt resistance varies with the voltage between the rails, in that as the voltage increases the train shunt resistance decreases. The lowest allowable drop shunt is 0.5 ohm irrespective of the track voltage.

The shunt and pick-up of a track relay can be taken with the aid of a variable resistance with 0.1 ohm tappings.

When taking the pick-up value the resistance should be set at a low value and gradually increased until the relay armature just picks up. This should be repeated two or three times until consistent results are obtained.

When taking the train shunt value the resistance should be set at a high value so that the relay remains in the energised position, and then the resistance is decreased until the armature of the relay just falls away and breaks its energised, i.e. front, contacts. The reading of the adjustable resistance is the drop train shunt of the track circuit. This should also be repeated two or three times to obtain the true value of the drop train shunt.

3.5       Relay Characteristics

There are two different values for train shunt. The drop shunt is the largest value of train resistance that will cause the track relay to release. The pick-up shunt is the smallest value of train shunt which will allow the relay to operate.

Due to the method of operation of all relays, the drop shunt will always  be the lower  of  the two values. As the energisation of the relay closes the air gap in the relay's  magnetic circuit, the reluctance of the magnetic circuit reduces. The relay therefore requires less current to

hold it in the energised position than to pick it up initially. Conversely, it will require a significant reduction in this current to release again. Whenever track circuits are tested the drop shunt should always be observed and recorded.

Track relays are designed specially to minimise the difference between the pick-up and hold-up currents. Nevertheless, a hold-up current of only 0.5 to 0.7 times the pick-up current is typical.

Another problem is the time of release of the track relay. With both relay and rail being inductive, a loop will be formed by the rails, the train and the relay. The back-emf caused by the reduction in current as the train shunts the track will tend to cause a circulating current in this loop which will delay the release of the relay. Short, fast moving trains could momentarily be undetected as they pass from one track circuit to the next. This has serious implications for the release of route locking and the holding of points.

If the type of track relay used is not inherently slow to pick up, a repeater relay which is slow to pick must be used in all vital signalling controls.

3.6       Maximum Length

The combination of the above factors will generally limit the maximum length of any type of track circuit.

The longer the track-circuit, the lower will be its ballast resistance, and so the attainable voltage at the relay will be lower, as more current leaks between the rails. Eventually, a length is reached at which it is no longer possible to operate the track relay.

This maximum length will be lower for lower values of ballast resistance. The maximum length must therefore be based on worst case ballast conditions.

A track circuit adjusted for minimum ballast resistance (e.g. during wet weather) will have a relay voltage higher than the minimum pick-up voltage when ballast conditions improve. Adjustment must be such that the track relay does not remain energised when a train of maximum value of drop shunt occupies the track circuit. This requirement also effectively imposes a maximum length on the track circuit.

Maximum length will depend on the type of track circuit, the components used and the conditions under which it operates. Most track circuits have maximum lengths in the 600  to 2000 metres range.

3.7       Minimum Length

For short track circuits, adjustment for reliable operation does not usually present any problem. But if a track circuit is too short, a vehicle with a long wheel-base may stand completely over the track circuit without it being detected. The minimum length of a track circuit is set by the vehicle with the)opgest wheel-base.  For  bogie vehicles,  this distance is measured between the inner axles of each bogie. In  most cases, this minimum  distance is more likely to be determined by bogie vehicles than by two axle vehicles.

3.8       Staggered Block Joints

Ideally, insulated joints are installed in pairs opposite each other. In practice, it  is frequently difficult to ensure that they are exactly opposite, such as within pointwork. Joints may be provided to separate adjacent track circuits or to allow a polarity change within a track circuit.

Where joints are not exactly opposite, they are said to be staggered. There is a short length of track where both rails are of the same polarity. In this area, an axle will not be detected. If the stagger is too great, a bogie or even a complete vehicle may remain undetected. The maximum stagger is determined by the vehicle with the shortest wheelbase. Although it is still possible to "lose" one axle, the . other axle will be detected, Care must be taken to correctly position joints for clearance purposes to take account of this problem. Staggered joints must not be used to define critical clearance points.

If two sets of staggered joints are too close together, both axles of a short, four wheel vehicle could be undetected at the same time. A set of joints which is not staggered but defines the end of the track circuit could also fail to detect a vehicle if too close  to  staggered joints. Therefore, unless joints are exactly opposite, they must always be separated by the length of the longest vehicle.

Figure 4 Track Circuit Minimum Length             Figure 5 Staggered Joints -Maximum Stagger

Figure 6 Joints in Close Proximity -Loss of Train Detection

3.9       Equipment Details

To ensure effective operation equipment types and values must be carefully chosen.

Whether a.c. or d.c. power supplies are normally of low voltage. This is desirable as the presence of a train presents a short circuit to the feed set or transmitter. Similarly, there is usually a series impedance in the transmitter/feed set to limit output current, avoid damaging circuit components and prevent excessive power consumption.

Where power supplies are derived from an ac mains supply, each track feed will  normally have its own supply to prevent track circuit signals feeding through between adjacent track circuits via the power supply.

Where a feed resistor is provided it is often adjustable to deal with different track circuit characteristics. When a track circuit is set up, account must be taken of the weather & ground  conditions  at the time. An incorrectly  adjusted  track circuit  may either fail to detect all trains under conditions of high ballast resistance (dry weather) or fail to energise the relay with no train present if the ballast resistance is extremely low (wet weather).

Many audio frequency track circuits are designed to be adjusted by varying the gain of the receiver. This makes setting up easier and also permits track circuits to be centre fed where required.

Frequencies for a.c. track circuits are normally within the audio frequency range. Higher frequencies are increasingly attenuated by the inductive impedance of the rail. This would make the maximum length impractically low. Mains frequency may  be  used  in non-electrified or d.c. traction areas (with suitable precautions).

Due to the low power used, relays are normally of low coil resistance (typically  4  to 20 ohms). For the same reason, relays may only be provided with two front contacts.

Electrical continuity of both rails is essential. Any rails not welded or bonded for  traction return must be bonded  by  the signal  engineer.  Bolted  fishplates  do not in practice  provide  a reliable electrical connection.

For rails which carry track circuit currents only, an adequate  connection  can  be made  by using two galvanised steel bonds pinned at each  end.Welded  or  brazed  connections  may also be used. Where electric traction is used, the return traction current will be many times greater than the track circuit currents. The bonds must be much heavier and will often be provided by the electrification engineer.

Connections from the rail to lineside equipment should be firmly secured to the rails and employ flexible cable to counteract the effects of vibration.

4. INSULATED BLOCK JOINTS

The insulated rail joint is installed in the track at the extremities of a track circuit and also within a track circuit at points and crossings in order to achieve reversal of polarity. When 40-ft. rails were used, the tensile stresses were negligible and to provide an insulating medium the steel fishplate was planed down by 1/4 inch and a specially formed sheet of bone fibre was inserted between the fishplate and the fishing surfaces of the rail. Fibre ferrules and washers were used to insulate the bolts and provided the joint was correctly packed in the ballast bed to prevent excessive flexing of the joint and crushing of the fibre insulation, the result was satisfactory. To insulate the gap between the rails, a fibre sheet shaped to the contour of the rail was inserted between the rail ends. This latter insulation, which is known as an end-post or "T" piece, was amply strong to resist the compressive load when the 40-ft. rails expanded.

This form of insulated rail joint gave reasonably good service even when 120-ft. rails came into use. However, the crushing of the fibre fishplate resulting in electrical failure of the insulated joint could not be observed and therefore the failure could not be anticipated. This fact necessitated the use of a joint where signs of impending mechanical failure could be seen long before replacement became necessary. For this joint, the steel fishplate was discarded and replaced by one made of an insulating material, the most common one being resin impregnated, compressed laminated wood. Such joints will take a tensile load of 2000 psi but the bending strength is limited to a smaller loading. The fishplate will be very quickly smashed unless the joint is well supported. From a track point of view a rigidly supported joint is not favoured since track resilience is the ultimate aim.

Modern railway practice is to provide a running top without any joints whatsoever as distinct from the use of 40 ft. or 120 ft. rail lengths joined together by means of steel fishplates. Rails of 45 ft. are depot-welded to 360-ft. lengths, transported to site and then thermit-welded into 1440-ft. lengths in the track. Thus the civil engineer has a track which is smooth, resilient and silent, requiring comparatively little maintenance and one which increases the life of rolling stock. The signal engineer, however, is obliged to request that the long-welded rail be cut at selected points in order that insulated rail joints may be installed to suit the track circuiting arrangements. Insulated rail joints, for which a design to resist the stresses set up in long rails has yet to be perfected, are inserted in the track. In order to protect the insulated joint against the high tensile forces, it is normal practice to install a splice joint next to the insulated joint. The splice joint allows the long rail to expand and contract, but apart from its prohibitive cost, it breaks the pattern and, together with the insulated joint, tends to nullify the advantages of long-welded rails.

A possible solution to the problem lies in the application of an insulated joint which reverts to the use of steel fishplates which, after all, give maximum bending strength. The joint is known as a "frozen" joint or "glued" joint because over and above the use of high-tensile steel bolts, the entire assembly is rigidly held together by means of an adhesive glue, 1/16 inch thick, which acts both as a strengthening bond and as insulation. The entire assembly must be made to close tolerances under workshop supervision. Two short lengths of rail are used, the completed joint then being taken to site and thermit-welded into the track.

Except for jointless track circuits on plain line, insulated block joints are used to define the limits of a track-circuit. It is important that the failure of an insulated  joint does not cause  a wrong side failure.

Joints may be provided in one rail only or in both rails. This will usually be decided by requirements for electric traction return.

With joints in both rails, if a single block joint fails, both track circuits should continue to work satisfactorily. If both block joints of a pair fail, there is a danger of a relay being falsely energised from the adjacent track-feed. To avoid this problem, the polarities of the rails on either side of every insulated joint must opposite. If both block joints of a pair fail, the two feeds will short-circuit each other, and both relays will de-energise.

In the vicinity of points and crossings,it may be necessary for sections of one rail to be common to two track circuits. Failure of any common blockjoint should cause both track circuits to show occupied. In electrified territory, joints may be in one rail  only. Maintenance of opposite polarities at every blockjoint is essential as only one joint separates the track circuits.

Figure 7 Track Circuit polarities -Double Rail (Joints in Both rails)

Figure 8 Single Rail (Joints in One Rail Only -Other Rail Continuous)

5. TRACK-CIRCUIT TYPES

Many different types of track circuit are available, designed to be used in particular applications such as electrified lines, continuous welded  rail, etc. A list of the main types of track circuit is given below. As track circuit development continues, further types will no doubt become available. It is not necessarily an exhaustive list.

5.1        D.C. Track Circuits

D.C. track circuits have the advantage that they can be simply fed from a battery, either as the main or a standby supply. Installation, setting up and subsequent maintenance is straightforward. They have the disadvantage that they are unsuitable where d.c. traction systems are in use.

Battery - fed (primary cell or trickle charged)

Non AC - immune

AC - immune Coded "Swept"

5.2       Fixed Frequency A.C. Track Circuits

All track circuits in this category work from a mains power supply at the same frequency.

Mains frequency (e.g. 50Hz) vane relay (single-rail or double-rail)

Traction immune, vane relay (operates at a different frequency to the traction supply for immunity from a.c. and d.c. traction currents)

A.C./D.C (a.c. track circuit employing a d.c. relay) various types

5.3       Multiple Frequency A.C. Track Circuits

Track circuits in this category are normally designed and installed so that adjacent track circuits operate at different frequencies. In most cases they may be used without insulated blockjoints.

Reed

Aster

Hitachi UM71

ML TI-21

Siemens FS2500/FS3000

Alstom CVCM

5.4       Impulse Track Circuits

These track circuits employ short pulses at a relatively high voltage to overcome contact resistance at the rail while keeping power consumption at a low level.

Jeumont Lucas

Although in some cases proprietary or manufacturer's names have been shown above, similar equipment may be available from other manufacturers.

5.5       Functions Other Than Train Detection

As the rails are a continuous transmission line from one end of the track circuit to the other, it is possible to use the track circuit for more than simple train detection.

To economise on lineside circuits, additional controls can be added into the feed end of the track circuit, typically in automatic sections. For example, the track feed can include proof that the signal ahead of it bas returned to normal after the passage of the previous train and/or proof that the signal is alight. The disadvantage of this type of circuit is that it complicates fault finding and may give the signalman misleading track circuit indications.

A more recent application is the coded track circuit. The track circuit current is interrupted at different intervals or modulated at different frequencies to convey information to the train, usually for automatic train control (ATC). The current passing from one rail, through the wheels and returning to the other rail is detected by one or two receivers on the front of the train. Different signals may correspond to different commands to the on-board ATC equipment.

To operate correctly, the track feed must always be at the end of the track at which the train leaves. Otherwise the train will shunt its own track circuit signal and the receiver on the train will see no signal· current in the rails. This is not a problem where trains run in one direction only. On bi-directional lines, this means that the track feed and relay circuits must be capable of being switched from one end to the other or two sets of track circuit equipment must be provided for each bi-directional track.

Coded track circuits also create complications in train detection at the relay or receiver end. The simplest approach is to build in some form of delay longer than the code pulses to maintain the relay energised. The relay/receiver could be set up to recognise only the available set of codes, any invalid code causing the track circuit to show occupied. The third option is to compare the signals at the feed and relay ends.

Figure 9 Coded Track Circuits Operating Principle 

6. D.C. TRACK CIRCUITS

In its simplest form the d.c. track circuit is as described in the earlier sections. Power was originally derived from one·or more primary cells which would be replaced· as necessary during routine maintenance. Where a lineside signalling power supply is available (normally a.c.) this will be used to operate the track circuit, either via a transformer & rectifier or by trickle charging a battery of secondary cells. In more recent years, solar power has become a viable power source in areas where the climate is suitable.

The choice of feed arrangements will depend upon the reliability of the main power supply.

If there are standby arrangements for the supply itself, a battery is not normally used.

In ac. electrified areas, d.c. track circuits may be used but with special ac.  immune feed sets and relays. Immunity to the a.c. traction currents must ensure that the feed circuit cannot be damaged by traction currents and  the relay will not pick or be damaged  by a.c. A combination of fusing, series chokes and a.c. immune relays will normally provide acceptable protection. Purpose built a.c. immune feed sets are available.

Obviously, d.c. track circuits cannot be employed where d.c. traction is in use as there is  no satisfactory means of preventing traction currents from passing through the relay.

7. A.C. TRACK CIRCUITS

A.C track circuits, normally at mains frequency, can be used in areas where the presence of high levels of d.c. would prevent the use of d.c. track circuits. This is most commonly due to the use of d.c. traction, but may also be due to the presence of certain chemicals in the ground, causing a voltage to appear across the rails due to electrolytic effects. It is also possible to use a.c. track circuits in non-electrified areas.

In place of a d.c. feed, an a.c. source is used. A two element a.c. vane relay is usually used in places of the d.c. relay. AC. vane relays cannot be energised by a d.c. voltage. The vane relay is operated by two windings, both of which must be energised simultaneously. One winding is fed locally, the other through the rails. The two relay feeds are then approximately 90" out of phase. If either feed is absent, or the phase angle between  them  is wrong, the relay will not energise.

To isolate the track circuit equipment from the effects of very large traction currents, it is normal to connect the track circuit equipment to the rails via either a transformer or a series capacitor. If a capacitor is used in the feed end, this can be variable to perform the function of the feed resistor in a d.c. track circuit. Fuse protection is also desirable.

One point to be borne in mind when designing locations for a.c. track circuits is that the supplies for the two windings must come from the same source. Signalling power supplies are single phase. This is normally derived from a 3-phase supply and adjacent power feeders may not be fed off the same phase. As the relay is phase sensitive, failure to  observe this rule could lead to incorrect operation and/or adjustment of the track circuit.

If d.c. traction is used, voltages are usually low and currents are therefore extremely high. To provide an acceptable low resistance return current path to the substation, both rails may be needed for traction return. Two basic configurations of A.C. track circuit are required, Double Rail and Single Rail. This refers to the traction current, not the track circuit current. Phase must be alternated between adjacent tracks for the same reason as polarity is staggered with D.C. track circuits (i.e. failure of blockjoints, connecting two supplies in antiphase will produce an effective voltage which is insufficient to energise either of the relays).

7.1        Double  Rail Track circuits

With low voltage d.c. traction systems, currents of 3,000A or more are commonplace. If current is to pass along both rails, special arrangements are made at track circuit joints to pass d.c. but not a.c. The arrangements include the use of an "impedance bond", which presents a low impedance to d.c., but a high impedance to a.c. by virtue of its two coils being wound in opposite directions.

Often the impedance bond is "resonated". The impedance bond is provided with a secondary winding of a larger number of turns than the primary windings. This is connected in circuit with a capacitor, giving a band-stop filter effect at the track circuit frequency to further increase the a.c. impedance.

The impedance bond is effectively a centre tapped inductor. The d.c. traction currents in each half of the winding set up opposing fluxes in the magnetic core which cancel each other out. To the a.c. track circuit current, the impedance bond presents a rail to rail impedance which is high compared with the ballast resistance.

Failure of a blockjoint or the disconnection of a traction return bond between rails may unbalance the d.c. currents in the impedance  bond  sufficiently  to saturate  the magnetic  core of the impedance bond and cause the track circuit to fail.

In addition to the extremities of a track circuit, impedance bonds will also be found:-

1. a) Where it is necessary to bond the traction return between adjacent tracks (to provide a lower resistance path). Depending on the power supply arrangements, this is usually necessary at approximately 0.9 km (BR third rail 750v) to 1.5 km (TfNSW 1500v overhead) intervals.
2. b) Immediately adjacent to substations for connection to the traction supply.

7.2       Single Rail Track Circuits

If traction currents can be kept low enough it may not be necessary to use double rail track circuits throughout an electrified line. Through points and crossings the use of double rail track circuits may not always be possible or desirable. Single Rail  track  circuits will then be used. The length of single rail track circuits through pointwork should be kept as low as possible where D.C. traction is in use as the impedance of the traction return path will be increased.

Figure 10  A.C Double Rail Track Circuits (Typical TfNSW , Former SRA) 

Figure 11  BR(British)  Capacitor Fed 

Figure 12 Typical Single rail Track Circuits 

Figure 13 Double Rail Track circuits with Additional Impedance Bond 

7.3       Dual Electrified Lines 

In a few cases, more than one electrification system is in use on the same line (e.g. 750v d.c. third rail and  25kv  a.c. overhead)  or lines  on  different  systems  run  in  close  proximity. If mains frequency a.c. track circuits are used, they will not be immune to the a.c. traction current. Similarly, d.c track circuits  cannot  be  used  either.  It  is however  possible  to operate at a supply frequency different to the  traction  supply  (and  any  harmonics),  Immunity  will then be provided against both traction systems. For a 50Hz traction supply, frequencies  of 83.3Hz and 125 Hz have been used.

A separate signalling supply at the required frequency can be provided  and  distributed  along  the lineside. Alternatively, the supply for  each  track  feed  can  be produced  at  each  location by an inverter. Filters must be incorporated in both feed and relay ends of the track circuit. Impedance bonds on double rail track circuits must  be  resonated  at  the  track  circuit  frequency (not 50Hz). This will require a different (usually lower) value for the resonating c3pacitor. The impedance bond will then appear as a much lower impedance at 50Hz.

This arrangement is particularly useful where the traction system is to be converted from

d.c. to a.c. where much of the existing track circuit equipment can be retained and track circuit conversion can be undertaken in advance of the traction changeover.

The use of such track circuits is now falling in popularity  as  a  number  of  dual  trnction immune audio frequency track circuits are available.

7.4       A.C/D.C. Track Circuits

The simple d.c. track circuit has an inherent slow-to-release characteristic, because  the track relay has a high inductance, and a low resistance. The d.c. track relay is however much simpler and cheaper to manufacture than the a.c. vane relay. The high inductance delays the release of the relay. This is an unwanted characteristic for a track  circuit, because it delays detection of fast moving vehicles and could leave a short train or vehicle undetected as it passes from one track circuit to the next. This obviously has serious implications on the safe operation of route locking.

If a relay with higher resistance, and lower inductance is used, the relay releases quicker. However, a relay with higher resistance requires a higher operating voltage.

An equivalent effect can be obtained by feeding a d.c. track relay through a step up transformer rectifier arrangement. One example of this is the B.R. "Quick Release" track circuit. The voltage on the rails is ac. The resistance in series with the track relay increases the value of resistance in the shunted circuit, and reduces  the release time.  The fact that the relay end has a high resistance also makes it possible to locate the relay some distance from the end of the track circuit (useful in restricted access situations such as tunnels).

This type of track circuit is not traction immune. It is also not frequency selective, and will operate at frequencies other than 50Hz, which causes problems when used adjacent to audio-frequency jointless track circuits.

8. JOINTLESS TRACK CIRCUITS

Insulated block joints are an additional cost in modem welded rail. Although the strength and reliability of insulated joints has generally reached a high level, most permanent way engineers prefer to avoid insulated joints wherever possible. Jointless track circuits have been developed to operate without insulated block joints. Various arrangements of electrical components are used to terminate the track circuits, and define their limits, electrically.

All jointless track circuits employ an a.c. signal. A transmitter at a specified frequency is connected to one end of the track circuit. At the opposite end, a receiver is connected  which will only respond to the signal produced by  its own  transmitter.  The receiver  will  normally be used to operate a relay which interfaces with the signalling system in the normal manner.

The insulated joint normally takes the form of a filter which will prevent signals of the track circuit's operating frequency from penetrating more than a short distance beyond  the  transmitter or receiver and ensure that the track circuit is not  operated  by  signals  from adjacent track circuits.

Adjacent track circuits must therefore operate at different frequencies. The frequencies used must also be immune from mains interference from the signalling power supply and, where necessary, the traction power supply.

Early jointless track circuits were generally not traction immune. Several types of jointless track circuit are now available, many of which are immune to both a.c. and d.c. traction.

The GEC "Reed" track circuit was also developed to be used as both jointless and conventionally jointed track circuits, immune to both a.c. and d.c. traction. However, problems were encountered with the operation of the track circuit in its jointless form in certain situations and it is at present only used on BR as a conventional jointed track, often in areas with both ac. and d.c. traction together. It will however be described  in this  section.

Due to the absence of insulated joints, the limits of jointless track circuits cannot  be  as precisely defined as for conventional track circuits. This generally presents  no problem  on plain line but, as the extremities of track circuits through pointwork need to  be  precisely defined to ensure safe clearances, block joints must be used. Track circuits  which  define precise clearances will thereftill need insulated joints  although  the  same  jointless  track circuits transmitters and receivers may often be used. Joints are also  required  at  the  extremities of jointless sections where they adjoin conventional track circuits.

To permit the filters to operate correctly, each type of track  circuit  will  often  have  a minimum length. This may be of the order of 50 metres. Shorter  track circuits  may  therefore be unsuitable for use with jointless track circuit equipment.

To permit adequate flexibility in the positioning of track circuits to meet all requirements, at least four distinct frequencies must be used, two on each line of a double line railway. This prevents interference between track circuits on adjacent lines. An early type of jointless track circuit, the Aster 1 watt, employed six frequencies, three for each line. On railways with more than two parallel lines, each set of frequencies could be used on alternate lines. Aster track circuits were not traction immune.

A further development was the Aster U type. This employed only four track frequencies (two on each line) and was also not traction immune. To permit alterations (addition or removal of a track circuit) with only two frequencies per line, the track circuit may be centre fed if required. It is still a simple example of a jointless track circuit to study and  will be described first.

8.1       Aster Track Circuits

The connection of the transmitter and the receiver to the rails is  by secondary  coils of  the  track transformer. This defines the approximate position of the boundary between  the  two track circuits.

The tuning units operate in conjunction with the inductance of the rails to form filters. In the diagram below, L1-4 are the inductances of the rails between the tuning units and the track transformer.

At frequency F1, C2 and L6 are series  resonant  and  are therefore  an  effective  short  circuit to F1. The remaining components present  a parallel  resonant  circuit  of  high  impedance  to the F1 signal. F1 will thus reduce in amplitude progressing from  Tuning  Unit 1 to Tuning Unit 2.

A similar situation arises for F2 with L5 and C1 acting as a series resonant short circuit.

Four frequencies are used. 1.7KHz and 2.3KHz would alternate on one line and 2.0KHz and 2.6KHz would alternate on the other line. Track circuit length may be from 50m to 1000m long. By using a centre feeding Transmitter the track circuit may be extended up to 2000m either as two individual track circuits or a single long track circuit (the TPR circuit would of course include both receivers in this case).

Figure 14 Aster U Type Track Circuit -Tuned Area Between Adjoining Tracks 

8.2       ABB /ML Style TI-21  Track  Circuits

The TI21 track circuit is similar in operation to the Aster track circuit already described, but has been designed to provide traction immunity.

This track circuit requires a tuned length of 20m with the tuning unit F1. acting as a short circuit to F2 and Tuning Unit F2 short circuiting F1.

Transmitter and Receiver connections are made via the respective tuning units. The security of the system is enhanced and traction  immunity  provided  by operating  each track circuit at two frequencies 34 Hz apart and modulating between them at a rate of 4.8 Hz. The receiver must detect both frequencies and the correct rate of modulation to energise the follower relay. It is considered that any signal produced by the electric traction supply  and its harmonics would be unable to simulate the track circuit signal for long enough  to energise the relay.

For traction return purposes, parallel tracks are normally bonded together at regular  intervals. It is therefore not possible to re-use the same frequencies on third and fourth parallel tracks. Eight frequencies are available, four of  which  are preferred  and thus used on two track lines. The remaining four are used on third and fourth lines when required.

A minimum length of 200m is required (300m if centre fed) and a maximum of 1100m is permitted (1000m per end if centre fed). Length can be reduced to 50m in a special low power mode of operation but the adjoining track circuits either side must then be connected with the transmitter nearest to the low power track to prevent the signal feeding through and falsely energising the receiver of the next track of the same frequency.

On d.c. electrified lines, double rail traction return is usually required and impedance bonds will be required at feeder points and wherever adjacent tracks are bonded together. They  will not be required at the extremities of  the  tracks. Special  resonating  units are provided to resonate the bonds to the track circuit frequencies.

Figure 15 TI21 Transmitter & Receiver Connections

8.3       GEC Type RT Reed  Track Circuits

The GEC Type RT Jointless Track Circuit operates by detecting the current in the track circuit loop, as opposed to the voltage at the end of the track circuit like the jointless track circuits already described. It is always centre fed in the jointless mode, although the last track before block joints may be end fed. The Type RT Track circuit  may also be used  as a jointed track circuit, in which case it may be centre or end fed.

Each track circuit operates at a frequency defined by a mechanical reed filter in the transmitter. The filter is very stable in frequency and has a bandwidth of less than 1 Hz. This allows track frequencies to be only 3 Hz apart, using 366, 369, 372, 375, 378, 381 and 408 Hz. On each line a four frequency cycle should be maintained.

The transmitter is connected directly to the rails. The receiver only responds to its own frequency due also to a reed filter at its input. When used in its jointless configuration, each track circuit is centre fed. The receiver is energised by a_n aerial positioned between the rails. The aerial comprises 36 cores of a 37 core cable wired to give a 36 turn inductive loop. The 37th core is connected to a series resonant shunt which terminates the track circuit. At any frequency other than the track circuit frequency, insufficient current will flow in this loop to energise the receiver. As the limits of the track circuit may not extend as far as the rail connection to the loop, the ends of adjacent tracks must overlap as shown.

The inefficiency of the shunt has led to problems in practice on B.R. where  track  circuit lengths are often short and/or more than two parallel tracks occur on  electrified  lines. The result is that signals from nearby track circuits of the  same  frequency  feed  through sufficiently  to be detected  by other receivers. This type of  track circuit  is only  used on B.R.  in its conventional jointed form, where it provides dual (a.c. & d.c.) traction  immunity. Receiver connection is then directly to the rails. Tracks may be either centre or end fed.

Figure 16  GEC Reed Track Circuit Arrangement for Jointless Operation

8.4       Siemens FS2500 Track Circuit

This is another recent addition to the range of jointless track circuits available. The location equipment can be mounted on a standard BR930 relay rack. The track circuit is traction immune and its circuitry includes the use of microprocessors to process the track circuit signal digitally. It employs a similar signal to the 1121 track circuit, a modulated ac. signal where the level of modulation and the modulation frequency must be correct to indicate a clear track circuit.

An interface has been designed to connect direct to a Solid State Interlocking (SSI) trackside module without the use of a track relay. The receiver has a built in "slow to operate" feature.

It employs four sets of frequencies, two on each line. It may be centre or end fed. In addition, the receiver may be positioned remote from the end of the track circuit, useful in tunnel situations. Different termination arrangements are used in each situation.

For a series of end fed track circuits, each transmitter and receiver is connected to the rails via a tuning unit. The tuning unit performs three functions:-

1. a) A high impedance to its own frequency
2. b) A low impedance to the frequency of the adjacent track circuit. This prevents transmission of the signal through to the next track circuit of the same frequency and shunting of the track by a train outside its nominal limits.
3. c) A transformer for transmission of the signal from transmitter to rails and from rails to receiver.

Adjustment of the signal level can be performed at both transmitter and receiver. It is usual for the transmitter to be set up on installation according to the length of the track circuit. Finer adjustment according to individual track characteristics is at the receiver.

At the end of a jointless section, an end tuning unit, which is slightly different to the normal tuning units for track circuit separation, is connected across the rails. The same tuning unit is used at any centre feed to provide a high impedance connection between transmitter and rails. It can also be used in conjunction with an impedance bond where connection between tracks, to traction feeders or to conventional jointed track circuits.

Air cored inductors (often referred to as SI units) are used where it is necessary to make earth connections to the track without direct connection to the rails. These are installed in the centre of the tuned area between tuning units.

Intermediate receivers can be installed within a track circuit to indicate the occupation of part of a track circuit, for example approaching level crossings.

Maximum track circuit length is 1.3km end fed (although TfNSW adopts a maximum of 900m). Minimum track circuit length is 50 metres.

8.4.1        Siemens FS3000 Track Circuit

FS3000 is the latest addition to the family .This Audio frequency track circuit has tranceiver unit and different types of tuning units for various configurations .This type dosent have impedenace bond for the application .Singapore Thomson East Cost Line use this type 

8.5       CSEE UM71 (Hitachi) Track Circuit

The operation and configuration is similar to other types of traction inunune jointless track circuits. It is not intended to repeat the details contained in earlier sections. Only the differences will be identified.

Instead of the tuning unit, separate tuning and  matching  units are  provided  at the trackside for connection and track circuit separation.

Resonating units for each  frequency  are  available  for  use  with  impedance  bonds. Maximum length is 600m and minimum length is 50m. Length may be extended up to 2km

by the use of compensating capacitors connected at intervals between the rails.

Two types of equipment are in use, the "T1" for most normal applications and the "T2" for extended connections in tunnel situations.

9. IMPULSING TRACK CIRCUITS

Rusty rails, as may occur on lightly used lines, will tend to cause wrong side failures whereby vehicles will remain undetected. To overcome this problem two basic methods are used. One is to weld a stainless steel strip onto the surface  of such  rails in critical areas; the strip will not go rusty. The other is to use impulsing track circuits.

The principle of an impulse track circuit is to  use  high  voltage  "spikes"  to  be fed  to the rails. If a train or single vehicle occupies the track the voltage of these spikes is enough  to break down the resistance caused by the rust on the rails. The voltage spikes applied to the  track have insufficient energy  to electrocute  personnel  on the line, although  their voltage is  of the region of 150V. Frequency is of the order of 3 spikes per second.

Although a higher ac. voltage on the rails would produce a similar effect, the power c:onsumption would be excessive. An impulse type track circuit generally consumes no more power than a conventional track circuit.

Two types, the Jeumont and the Lucas track  circuit,  are  in  use on  B.R.  although  they  are not very common. TfNSW uses the Jeumont track circuit extensively.

On the Jeumont track circuit, / a typical voltage waveform on the rails is shown on the following page. The receiver detects the correct  polarity  of  the pulse (important  in the event of a failed insulated joint) and the correct ratio of voltage and duration  between  the positive and negative parts of the waveform. It must also provide an output to the track relay in the interval between pulses.

The arrangement of connections to the track is also shown.

Figure 17  Jeumont Track Circuit -Typical Rail Voltage Waveform 

Figure 18  Jeumont Track Circuit -Typical Rail Voltage Waveform 

It is not intended to describe the construction and operation of Jeumont tracks in detail. They are generally considered to be traction immune due to the low frequency of operation (3Hz). Due to the high voltage on the rails, they can be used for long track circuits, up to 1500 metres on electrified lines and 3km on non-electrified lines.

Double rail track circuits on d.c. electrified lines will be connected at each end via an impedance bond.

The relay is peculiar to the Jeumont track circuit and is designed to operate with the receiver.

10. RAIL CIRCUITS AND OVERLAY TRACK CIRCUITS

Unlike a track circuit, which detects the absence of a  train,  a  rail  circuit  detects  the presence of a train. It may be used at automatic level crossings, for approach control of  signals, or in place of treadles on high speed lines.

It has also been used in situations such as bump marshalling yards to control the operation of points, due to its higher speed of operation when compared with the track circuit. To detect the absence of trains, a rail circuit is not fail safe.

The simplest form of rail circuit consists of a standard track relay connected in series with its feed across the rails. A train or vehicle on the track will then complete the circuit, energising the relay.

An alternative to the rail circuit is the overlay track circuit. It uses a much higher

:frequency signal than conventional track circuits. The inherent impedance of the rail limits its extent of operation, so insulated joints are not required and it may be overlaid on most conventional track circuits.

11. SHUNT ASSISTERS

As mentioned earlier, increasing problems are being experienced with track circuits due to the introduction of lightweight vehicles with improved riding characteristics. Obviously, unreliability of track circuits puts the integrity of the whole signalling system at risk. Large scale replacement of existing track circuits with those that will operate satisfactorily with the new vehicles may not be practical or economic.

If it can be shown that the problem is confined to certain types of train, it may be possible to deal with it by modifying the vehicles instead of the track circuits. On B.R. certain classes of one and two car diesel multiple unit vehicles are being fitted with shunt assisters. These operate by applying a high voltage (compared to most track circuits) signal to the rails. This is outside the frequency range of normal track circuit operation (50khz) and does not affect the operation of relays and/or receivers. It does however break down the resistance between rail and wheel and allow the track circuits to operate more reliably.

Such a solution is not without problems. The on-board equipment must be  monitored  for correct operation and its failure could prevent further operation of the  vehicle  until  the problem has been rectified. It is not therefore a substitute for ensuring reliable track circuit operation for normal traffic.

12. ALTERNATIVES TO TRACK CIRCUITS

In locations where track circuits cannot be installed or where a long track circuit is required, axle counters may be a possible solution. A set of axle counter equipment is generally more expensive than a single track circuit. If one set of axle counters can substitute for several track circuits, the overall cost may be cheaper.

If required, axle counter sections can be designed to overlap, reducing the overall quantity of equipment. The track equipment may also be located remotely from the electronic logic equipment.

Axle counters will be covered in more detail in a later section.

Although not yet used in any working system, the use of Automatic Vehicle Identification transponders on vehicles has the potential for use as a train detection system. An esential requirement is, of course that all vehicles are fitted with transponders and that they can be proved to be functional.