Click on the icon above for

the Brighton Motive Power Depots

Click on the icon above for

the Sussex Motive Power Depots & ASLEF Branches






extracted and adapted from the report by


Lieutenant Colonel



On Saturday, 16th December, 1972, at Copyhold Junction near Haywards Heath in the Southern Region of British 

Railways, a collision between two passenger trains that occurred at about 22.00, in patchy fog the 21.28 electric multiple-

unit (EMU) stopping passenger train from Brighton to Victoria, formed of 8 coaches, having passed at Danger the signal 

on the Up Local line protecting the converging junction with the Up Through line, was on the junction points when it was 

struck at the sixth vehicle by the 12-coach 21.45 Brighton to Victoria EMU express passenger train running under clear 

signals on the Up Through line.

The leading 4 coaches of the stopping train remained on the track but the rear 4 coaches became uncoupled, and were 

derailed and badly damaged. The first 10 coaches of the express train were derailed and 3 were badly damaged.

The derailment blocked all lines and the traction current was discharged automatically as a result of the collision. 

Fortunately both trains were very lightly loaded and there were no fatal casualties, but 11 passengers were treated for 

shock and minor injuries and two were detained in hospital for one and two nights respectively.

During the period of blockage there was considerable disruption of traffic and London-Brighton services were diverted 

via the Mid-Sussex line. Owing to the difficulty of access to the site, which lies in a steep sided cutting, clearance of the 

damaged vehicles and restoration of the track was not easy and the line was not reopened until 19th December, with the 

connections between both Up and Down Local and Through lines temporarily replaced with plain line. New high-speed 

turnouts had to be specially fabricated and the Down side connections were not restored until 14th January 1973 and the 

Up side connections one week later.


Site and Signalling

Copyhold Junction, where the former branch line to Horsted Keynes, now a siding serving a quarry at Ardingly, joins the 

Victoria-Brighton Main line on the Down side, lies I t miles north of Haywards Heath. Between Haywards Heath station, 

which has 2 island platforms, and Copyhold Junction the line is 4-tracked with Up and Down Local lines flanking the 

Through lines. With the exception of the branch, all lines are electrified on the conductor rail system at 750v DC. At the 

junction the four tracks merge into two for the double track section across the Ouse Viaduct and through Balcombe 

Tunnel. The line speed on the Up and Down Through lines is 90 miles/h and on the Local lines and through the 

connections between the latter and the Through lines at Copyhold Junction, 60 miles/h.

The gradient is rising in the Up direction generally at 1 in 264, from south of Haywards Heath to a summit at Balcombe 

Tunnel. At Copyhold Junction the line lies in a deep steep-sided cutting and, just at the point where the double track 

section begins, it is spanned by a substantial brick 2-arch bridge, the centre pier of which stands between the Down Main 

line and the Ardingly siding, which at this point is just starting to diverge eastwards from the Main line.

The signalling in the area is on the Track Circuit Block system with colour-light running signals, installed in 1932, since 

when the points and signals at Copyhold Junction have been controlled electrically from Haywards Heath. Train 

description is by means of magazine type train describers. The signal box at Haywards Heath, located on the Up side of 

the line, contains a 60-lever frame with full mechanical interlocking between the levers, some of which operate points 

and signals mechanically and others, as at Copyhold Junction, by electrical means. Above each of the levers controlling 

the colour-light running signals multiple aspect indicators are provided, displaying the actual aspect of the signal 

regardless of the position of the lever. Above the lever frame is a point and signal diagram of the layout with track circuit 

occupation shown by the illumination of two bulbs in an oval display for each separate track circuit, located and 

numbered appropriately.

Those points and signals relevant to the accident are shown on the drawing attached to this Report. On the approach to 

Haywards Heath from the South, the first controlled signal, CH 112, is a combined junction home signal with R/Y 

aspects, controlled by lever No. 1, reading to the Up Local line and R/Y/G aspects, controlled by lever No. 2, reading to 

the Up Through line. It is located on the approach to Haywards Heath tunnel, 648 yards in rear of No. 5 facing points 

where the Up Local line diverges from the Up Through line and 1,070 yards in rear of Signals C118 and CH19, the 3-

aspect platform starting signals on the Up Local and Up Through lines respectively. The 3-aspect signals protecting No. 

16 points, where the Up Local and Up Through lines merge at Copyhold Junction, are CH13 and CH20 respectively 

located 1,453 yards in advance of the platform Starting signals and 659 yards in rear of the points at Copyhold Junction. 

Both signals CH13 and CH20 are provided with a combined berth and overlap track circuit, the overlap lengths being 

386 and 388 yards respectively. Beyond these overlaps, a common track circuit extends to beyond No. 16 points.

The gantry that supports Signals CH13 and CH20 stands on the left of the Up Local line with CH13 located on the 

supporting mast, 6ft 4ins from the running edge of the nearest rail, with the red aspect 11ft 74 ins above the rail level of 

the Up Local line. Signal CH20 is located on a cantilever arm 5ft 4ins to the left of the left hand running rail of the Up 

Through line and with the red aspect 16ft 6ins above rail level. From the Up Local line, on account of a gentle left hand 

curve, the Up Through line Signal CH20 comes into view first, at a distance of 572 yards, followed by CH13 at 528 

yards. A continuous full view of both signals is obtained from the Up Local line over a distance of 506 yards. 



The Trains

The 21.28 stopping train from Brighton to Victoria was formed of two 4-VEP through corridor semi-fast electric multiple 

units. Its overall length was 531ft and its tare weight 294 tons.

The 21.45 Fast train from Brighton to Victoria consisted of three 4-car express units, the leading and trailing units being 

of the 4-CIG type with a 4-BIG (Buffet) unit marshalled between them. Its overall length was 796 ft and its tare weight 

456 tons.

All the units of both trains were of a similar layout, each comprising two driving trailers with a non-driving motor brake 

second and a trailer car marshalled between them. All were of modern all-steel construction with buckeye couplings and 

Pullman-type gangways throughout and were fitted with both electro-pneumatic and Westinghouse automatic air brakes. 

Their maximum service speed was 90 miles/h. They were all built at York works at various dates between 1966 and 


The Course of the Accident and Damage caused

The actual point of collision, the fouling point where the Up Local line merged with the Up Through line at Copyhold 

Junction, lay 561 yards beyond Signals CH13 and CH20 and the initial contact was between the grab rail outside the 

driving cab of the Fast train and the side panelling in the area of the guard's compartment on the sixth coach of the Slow 

train. The marks of this contact indicated that the speed of the Fast train was only slightly higher than that of the Slow 

train at the moment of collision and the immediate result was that the leading driving trailer of the Fast train rolled over 

onto its right side, glanced off the centre pier of Copyhold Lane bridge and came to rest on the Down Main line about 80 

yards beyond the bridge and separated by a coach length from the rest of the train, which was derailed all but the last 2 

vehicles. The Slow train became divided between the 2 units, the leading 4 coaches being completely undamaged and 

coming to rest 250 yards beyond the bridge. The rear 4 coaches were all derailed towards the cutting side leaning over at 

an angle of about 45o with the fifth coach only a few feet from the brick abutment of the bridge.

The most extensive damage was to the underframes of the middle unit of the Fast train and to the rear unit of the Slow 

train, the bodies of which were pushed off their bogies. The degree of damage to the bodies was comparatively slight, 

most being caused by the detached bogies being carried along in the space between the two trains.

The damage to the permanent way and signalling equipment was severe. Both the connections between the Through and 

Local lines were demolished together with over 300 yards of plain track. The point machines were destroyed and the 

main line side signalling and telecommunications cables severed, thus cutting off power supplies to Haywards Heath 

signal box. The traction current was also discharged automatically as a result of the collision.

Operating Instructions

When both the Slow and Fast Brighton-Victoria trains are running on time, the former is booked to leave Haywards 

Heath at 52a minutes past the hour and precede the latter through the double track section from Copyhold Junction to 

Balcombe Tunnel Junction, the actual booked passing times at Copyhold Junction being 0.55 and 0.00 respectively. 

However, since the maintenance of the service depends on the Fast train being given as far as .possible a clear run, the 

following special instruction was issued after the introduction of the present service pattern in the new timetable that 

came into force in July 1967.

0.28 Brighton to Victoria

"The 0.28 Brighton to Victoria should be run to the Up Local Platform at Haywards Heath as a regular practice. If it 

leaves Haywards Heath 3 or more minutes late and the 0.45 Brighton is running to time, the 0.28 must follow the 0.45 

from Copyhold Junction. These arrangements to apply off peak weekdays, all day Saturdays and Sundays as applicable."





Relief Signalman J. E. R. Budd assumed duty at Haywards Heath at 21.25. He told me that the 21.28 Brighton to Victoria 

was described to him from Keymer Crossing at 21.48 and he routed it into the Up Local platform where it arrived at 

21.53, 3 minutes late. It was still standing in the station at 21.55 when the 21.45 Brighton to Victoria was described from 

Keymer Crossing and Signalman Budd decided to give the latter train preference. The points at Copyhold Junction were 

already set for the Up Through line, and the Up Local Starting signal (CH8) was showing Yellow with the repeater for 

Signal CH13 showing Red. As he cleared the Up Through signals for the Fast train the Slow train started away from the 

station and he restored CH8 as soon as the train passed the signal. Perhaps a minute later the Fast train emerged from the 

tunnel and whistled as it approached the station. Signalman Budd described it forward to Balcombe Tunnel Junction and 

replaced Signal CH2 and CH19 behind it. He then remained standing at the frame ready to replace CH20 lever after the 

Fast train had cleared Copyhold Junction and saw Track Circuit 18, covering the junction, show occupied about a minute 

after the train had passed the station. This he assumed to be the head of the Fast train passing the block joint at the 

overlap of Signal CH20. However, an instant before all his indications were lost he saw the berth track circuit for Signal 

CH13 show clear, indicating that the Slow train had passed beyond the overlap of that signal. Realising at once that it 

was almost certain a collision had occurred he sent the 'Obstruction Danger' signal to Balcombe Tunnel Junction and 

operated the emergency replacement switch for Signal CA16 on the Down line. He then called out the Emergency 

Services. His first actual confirmation that an accident had occurred was not until one of the permanent way men 

travelling on the Slow train went back to Signal CH13 and telephoned him from there.

In charge of the Fast train was Guard J. P. Farrell stationed at Brighton. He had taken over the train at Brighton at 21.30 

and it had left on time at 21.45. He told me that he was on his way to the rear of the train to carry out his duties as train 

ticket collector as the train passed through Haywards Heath, which he noted as being passed on time, and soon afterwards 

felt the brakes being applied in emergency. He started at once to return to his brake compartment but within about 5 

seconds he felt the train make 3 lurches and the coach started to lean over, coming to a stand with all the lights out.

Guard Farrell went immediately to his brake compartment, collected his detonators and was about to set out to protect in 

rear when he met a permanent way man who undertook to protect the train while he got out an extinguisher to deal with a 

small fire that had broken out underneath one of the rear coaches of the other train. He then handed the extinguisher over 

to the guard of the other train and then got out some more detonators and went forward to protect the Down line. On the 

way he met another permanent way man who told him that the driver was unhurt and was carrying out the protection, so 

he went back to Signal CH13 where he spoke on the telephone to the signalman and asked for the Fire Brigade and 

Ambulances to be sent to the scene. He then noticed that the visibility seemed to have worsened so he put down 

additional detonators, one on each Up line 100 yards behind the two trains and then went back the full mile to put 3 more 

detonators on each line.

The driver of the Fast train was Driver E. Rooke, who took the train over at Brighton at about 21.35, after his mid-shift 

break. He told me that he had had clear signals until after passing Wivelsfield where he first saw a Yellow signal. He 

made a slight application of the brake but the signal cleared to Green before the train had lost any significant amount of 

speed. He recalled that he then ran into an area of patchy fog and the next signal, which was the junction signal for the 

Up Local line at Haywards Heath, cleared from Yellow to Green for the Up Through line almost as it came into view. 

The next two signals, the Haywards Heath Up Through Starting signal and the signal protecting the convergence of the 

Up Through and Up Local lines at Copyhold Junction were both at Green, but just after passing the latter Driver Rooke 

suddenly saw the illuminated red screens of a train on the Up Local line close ahead. Realising that a collision was 

inevitable, he made an emergency brake application, let go of the controller and crossed to the off side of the cab where 

he clung to the handbrake column and waited for the impact. He thought the speed of his train was about 75 miles/h at 

that time. He told me of the terrible noise of the collision with sparks and debris flying everywhere and of how the coach 

rolled over onto its right side. He thought he must have fainted for a moment because when he came to all was still and 

he found himself trapped by the cab door into the right hand corner of the cab. He was very worried in case the Down 

line was obstructed and after a few moments he managed to force open the small sliding window and wriggle out 

between the cab side and the ballast. He had no lamp and no detonators, but there were several permanent way men who 

had been passengers on the Slow train nearby, and while one gave him a lamp another fetched detonators from a nearby 

platelayers cabin and accompanied him.

17. Driver Rooke told me that he then ran northwards, trying the signal post telephones as he went, but it was not until he 

got to Signal CA17, beyond the Ouse Viaduct and nearly 1½ miles from the scene of the accident, that he found a 

telephone in working order and was able to speak to the signalman at Balcombe Tunnel Junction who assured him that 

the Down line was protected and that there were no trains coming. He then made his way to Balcombe Station, a further l 

½  miles, where he was later seen by a doctor before making his way home by taxi.

Guard R. A. Lawal, stationed at Victoria, was in charge of the Slow train, which he had taken over at Brighton at 21.20 as 

his last trip of the day. He told me that the driver had not yet arrived and did not do so until about 21.25. He appeared 

perfectly normal and there was no indication that he might have been drinking. After making the necessary brake test the 

train left about 2 minutes late. It made the normal stops at Preston Park and at Hassocks but at Burgess Hill the train 

overshot the platform somewhat. It then made a normal stop at Wivelsfield and arrived in the Up Local platform at 

Haywards Heath 3 minutes late at 21.53. Guard Lawal recalled that the Starting signal was at Yellow when they left 

Haywards Heath at 21.55 and ran forward slowly along the Up Local line. He then felt the train accelerate as it 

approached Copyhold Junction, as if the driver had received a clear signal, but a few seconds later the collision occurred 

and the sixth coach, in which he was riding, shook violently and all the lights went out. As soon as the train stopped he 

jumped down and, after meeting the guard of the other train, set about extinguishing the small fire that had broken out 

under the rear portion of his own train.

The driver of the Slow train was Driver A. Gilbert, stationed at Horsham; he was 50 years of age and had been in his 

present grade for 17 years. He had booked on duty at 17.30 and after standing by had travelled as a passenger on the 

18.39 train from Horsham to Redhill where he had again waited until 20.00 when he took over the semi-fast from 

Victoria to Brighton where it was due to arrive at 20.37. Driver Gilbert's next task was to take the 21.28 Slow train from 

Brighton to Victoria and he told me that he spent the intervening 50 minutes in making a long telephone call to a lady 

friend, buying some fish and chips and in going to the lavatory. He said it was about 21.24 when he got back to take over 

his train and that they left about 2 minutes late.

Driver Gilbert was unable to give a convincing explanation of his overrun at Burgess Hill other than to say that he had 

misjudged his braking. He recalled that the station stop at Haywards Heath was fairly long, about 2 minutes, he thought, 

and that the Starting signal was at Yellow. He remembered that the fog was patchy after leaving Haywards Heath and that 

when he saw a Green signal ahead he took it as applying to him and then accelerated towards Copyhold Junction. In 

retrospect, Driver Gilbert told me he was now aware that he must have taken the Up Through line signal in mistake for 

his own and he frankly admitted his error. After the collision he had no idea an accident had occurred and he thought that 

his train had been brought to a stand by a brake application made by the guard.

I asked Driver Gilbert about his health and home life and he told me that he had recently been through a divorce case and 

that he was living on his own and cooking for himself. He had had an operation for a stomach ulcer and had to be careful 

about what he ate. He found he could not take a heavy meal before going on duty and that this was why he had bought 

fish and chips at Brighton to eat on the train. He assured me that he had taken no alcohol that evening but that, on account 

of his domestic and health worries, he found it difficult to concentrate fully on his job. He made the point that, if the line 

had been equipped with AWS, this accident, the responsibility for which he accepted, might well have been prevented.

Travelling on the 21.28 from Brighton as a passenger was Driver L. D. H. Churcher (West Worthing), on his way to take 

over a train that divided at Haywards Heath. He took his seat in the last vehicle of the train and, as departure time 

approached realised that no driver had yet appeared. He told me that in order to save delay he went to the rear cab, 

released the handbrake and started the compressors. He said it was a recognised thing for one driver to do this for another 

but, while he was in the cab, at about 21.28 or 21.30 Driver Gilbert arrived and asked him, in a rather surly manner, what 

he thought he was doing. Driver Churcher explained to me that Driver Gilbert, whom he had known for a number of 

years, seemed perfectly normal and that such a reaction was one that was to be expected of him.

Also travelling as a passenger on the Slow train was Leading Trackman W. L. Palmer who joined the train at Burgess 

Hill. When the train arrived he thought it was not going to stop and in fact it came to a stand with the third coach half off 

the end of the platform. He joined the front part of the train and when the lights went out at Copyhold Junction and the 

train came to a stand he was one of the permanent way men who got down to assist. He met Driver Rooke and so learnt 

that there was a second train involved and, as he appeared somewhat shaken, accompanied him as far as Signal CA17, 

from where Driver Rooke spoke to the signalman at Balcombe Tunnel Junction. Mr. Palmer then returned to assist at the 

scene of the collision.

Mr G. K. Summerfield Assistant (Special Duties) in the Divisional Manager's Office made his way to Haywards Heath 

signal box on learning of the collision. He ascertained that the frame had not been touched since the accident occurred 

and confirmed that the route was set up for the Fast train with levers 3, 4, 10, 20 and 45 reversed in the frame. The 

illuminated diagram and all signal and point indicators were out or showing 'wrong'. He then carried out a check on the 

mechanical interlocking and confirmed that lever 13 could not be moved from the normal position unless lever 16 was 

reversed and that lever 20 could only be moved from the normal position when lever 16 was normal. 

Subsequently, a full mechanical and electrical test of the installation was carried out, including the selection circuits for Signals CH13 and CH20, and the cabling between the signal box and signal location, and all was found in proper order.




The Speed of the Trains

Calculated from the distance run by the front unit of the Slow train, approximately 300 yards, after it became divided, its 

speed at the moment of collision was probably between 50 and 55 miles/h and this agrees with the speed the train would 

have attained at Copyhold Junction after accelerating normally 4 from a station stop at Haywards Heath. The speed of the 

Fast train, as recalled by Driver Rooke in his evidence, was about 75 miles/h but, from its calculated performance on the 

rising gradient of 1 in 264 its actual speed is not likely to have been in excess of 70 miles/h at the time it passed Signal 

CH20, and the brake application initiated by Driver Rooke on becoming aware of the Slow train ahead of him on the 

converging Up Local line would have reduced its speed by perhaps L5 miles/h by the time the first contact between the 

two trains took place, at a speed differential of probably less than 5 miles/h. On this basis, the tail of the Slow train would 

have been some 70 yards ahead of the front of the Fast train as the latter passed Signal CH20 and its leading end would 

have reached the overlap track circuit joint, 386 yards beyond Signal CH13, some 6 seconds later, at which moment 

Signal CH20 would have reverted to Red. The tail of the Slow train would have cleared the overlap joint, thus allowing 

the berth track circuit to show clear on the signal box diagram and alerting the signalman to the impending collision, only 

4 seconds before it occurred.

The cause of this accident was that the 21.28 Slow train from Brighton to Victoria, having started from Haywards Heath 

on the Up Local line under a Caution signal, passed at Danger the signal protecting the converging junction between the 

Up Local and Up Through lines at Copyhold Junction. I have no doubt that the signalling was in proper order and that the 

signal concerned was displaying a red aspect as the train approached and passed it.

In looking for a reason for this signal being passed at Danger, I am satisfied that neither the siting of the signal, which 

was unmistakably positioned with respect to the line to which it applied, nor the prevailing weather conditions of patchy 

fog which was not thick enough to prevent the clear reading of colour light signals, nor the pattern of train working 

which was in accordance with an instruction that had been in force over 5 years, can be held in any way responsible. I 

can only assume that Driver Gilbert's failure to see the signal or to comprehend its message was the result of his allowing 

his concentration to wander from his job at the critical time as he approached and passed the signal. The earlier events of 

the evening, including his late arrival to take over his train at Brighton after having spent much of his layover time in 

making a long telephone call to a lady friend and his misjudged stop at Burgess Hill, tend to point towards his having 

allowed his personal affairs to occupy his thoughts to the extent that he just did not react to a well-sited colour-light 

signal which was displaying a red aspect as he approached and passed it. There was no way in which any other person 

could have averted or mitigated the effects of the collision which followed, for which Driver Gilbert must bear the sole 

responsibility. In view of the circumstances, it was indeed fortunate that this accident did not result in serious or fatal 


The Automatic Warning System (AWS) of train control, by which a driver is warned on his approach to a signal at 

Caution or Danger by a horn sounding in his cab, which warning must be acknowledged, in default of which the brakes 

are automatically applied, might well have prevented this accident, and the fact that AWS is not provided on this 

important line cannot be allowed to pass without comment.

The situation with regard to the provision of AWS in the Southern Region was explained, as far as it obtained at that 

time, by Colonel J. R. H. Robertson in his Report on the collision that occurred on 4th January 1969 between Paddock 

Wood and Marden on the main line from Charing Cross to Dover which, like the Victoria to Brighton main line, was not 

provided with AWS.

Since that time, the main signalling effort in the Southern Region, within the limitations imposed by restrictions on 

capital investment, has continued to he directed to the extension of multiple-aspect colour light signalling, concentrating 

in the main on the densely trafficked suburban areas for which, for the reasons set out by Colonel Robertson in his 

Marden Report, there were doubts as to the suitability of the BR standard system of AWS. However, this system is 

suitable for the less intensively worked areas of the Southern Region and its installation as part of a continuing 

programme is now proceeding on the Portsmouth line and in the area to be controlled from the new signal box at 

Feltham. In the case of the Brighton main line, I am informed that the installation of AWS will have to wait the 

replacement of the existing signalling, which was installed in 1932, since the braking distances then provided for the 3-

aspect signalling were calculated to include sighting distances and not measured on a post-to-post basis which is a basic 

requirement on lines fitted with AWS. I accept that it would not be economic to adapt the existing signalling to permit the 

installation of AWS at this stage in its life.

It was suggested at my Inquiry that this accident would have been averted had Signals CH13 and CH20 been provided 

with separate berth and overlap track circuits, because the occupation of the overlap track circuit of Signal CH13 by the 

Slow train would have resulted in Signal CH20 changing from Green to Red when the Fast train was some 300 yards on 

the approach side of it, at which juncture an emergency brake application by Driver Rooke would have prevented a 

collision. In the particular circumstances of this accident this is in fact fortuitously true, but in my view an isolated 

incident of this kind cannot be advanced as an argument for the general read option of separate berth and overlap track 

circuits with the additional costs involved.

It has also been suggested that the provision of trap points and sand drags could help to prevent this sort of accident. The 

real object of traps, however, is to prevent slow speed movements obstructing the adjacent line and, had they been 

provided at Copyhold Junction, the outcome would have been that the Slow train would have hit the solid brick abutment 

of Copyhold Lane Bridge at 50 miles/h at the very moment the Fast train was passing on the adjacent line at 70 miles/h. 

The results would have been truly catastrophic.

It would not be right for me to conclude this report without remarking on the devotion to duty shown by Driver Rooke, 

the driver of the Fast train, who, whilst suffering from shock, ran l ½ miles in the dark to reach a telephone. His action 

was in the highest tradition of the railway service.




For the full report into this accident, click 75A icon


 Scene at Copyhold Junction a few months earlier




Click on the icon above for

the Brighton Motive Power Depots

Click on the icon above for

the Sussex Motive Power Depots & ASLEF Branches


Make a free website with Yola