Jan 15
Rolled Up: The 2002 Wampersdorf (Austria) Train Collision
Background
Wampersdorf is a town of 857 people (as of 2022) in the far east of Austria, located in the Baden District 22km/14mi northwest of the Hungarian border at Klingenbach and 27.5km/17mi south of Vienna (both measurements in linear distance).
The town lies on the Pottendorf Line (Pottendorfer Linie), a 51km/32mi long electrified and mostly double-tracked main line connecting Vienna with Vienna-Neustadt (a different, geographically city, somewhat confusingly). Opening in sections between 1871 and 1875 on a slightly different routing the line’s original purpose was to help improve the connection of Vienna with the “Transleithania”-area of the then-existing Austro-Hungarian empire. Nowadays the line sees mostly freight services and regional passenger trains at speeds of up to 160kph/99mph.
The trains involved
Gz 46682 was an international freight service from Sopron (Hungary) to Vienna (Austria), consisting of 28 various freight cars pulled by ÖBB (Austrian National Railway) 1142 685. The series 1142 is a four-axle multipurpose electric locomotive introduced in 1995. The type was created by adapting series 1042 locomotives for push-pull trains, allowing them to be used for modern regional passenger trains where the locomotive is sometimes controlled remotely from the cab car at the far end of the train. Each series 1142 measures 16.22m in length at a weight of 83.5 metric tons and can reach a top speed of 150kph/93mph. The type was going through a refit-process around the time of the accident, recognizable by a different livery and smaller headlights. However, the unit involved in the accident had not been refitted yet.
At the time of the accident a southbound RoLa-Serivce (“Rollende Landstraße”, “Rolling Road” in English) was parked at Wampersdorf station, waiting to continue its journey to Hungary. RoLa-trains consist of a row of specially developed eight-axle low-floor flatbed cars, each one large enough to carry a semi-truck, along with a sleeper car at the front of the train where the truck drivers can rest during the journey. The service was introduced in Europe in the mid-90s to take trucks off the road and thus reduce emissions and congestion, plus allowing drivers to still make progress during their mandatory rest-periods. By 1999 the ÖBB counted 254 thousand trucks having used their RoLa-services alone. On the day of the accident the RoLa sitting at Wampersdorf station was carrying 21 trucks, with their drivers travelling in a Bcmz-type sleeper car right behind the locomotive. The Bcmz-type sleeper cars used for the RoLa offer nine compartments for sleeping, each with six beds, at an overall length of 26m/85ft.
Pulling the RoLa-Service on the day of the accident was ÖBB 1044 213. Introduced in 1976 the series 1044 is a four-axle multipurpose electric locomotive measuring 16.1m in length at a weight of 84 metric tons. At the time of its introduction the type held the title of the world’s strongest four-axle electric locomotive with a power output of 5280kW/7080hp, enough for a top speed of 160kph even with a heavy train coupled to it.
The accident
On the 26th of February workers were busy assembling a freight train at Sopron (Hungary) to be sent to Austria as Gz 46682. The train was a mixed freight service, consisting of various kinds of freight cars carrying different kinds of cargo. During the assembly they noticed a faulty valve in the brake system of a type Tadgs 959 four-axle swivel-top car, which wouldn’t allow air to pass through it. This meant that the brakes on train cars past the faulty car could not be properly operated. It’s unknown if the workers had the option of having the car replaced, either way they chose not to, instead calculating that the train had enough braking-power even with the faulty car in it. All they had to do was place it at the back of the train, as the 26th and last car. That way all the brakes except for the last car’s ones were working, providing enough braking-power. With a brake-check conducted and the necessary paperwork signed the train was sent on its way.
The train operated without incident for several hours, crossing into Austria and eventually reaching Ebenfurth (pronounced like “Ebenfurt”), 8km/5mi down the line from Wampersdorf station. The train entered Ebenfurth station from the southeast and was meant to depart to the northeast, meaning it had to be turned around. During the same process the locomotive was also replaced, separating the Hungarian locomotive which had delivered the train here from the train and adding the waiting ÖBB 1142 685 along with two more freight cars. With the locomotives being swapped the change of direction was easily achieved by workers coupling the Austrian locomotive and its two freight cars to what used to be the rear end of the train, making it point in the correct direction and increasing its length to 28 cars.
Having completed the shunting-operation the train was cleared for departure towards Vienna at approximately 2:50pm. With the Austrian locomotive and two freight cars having been attached to the former rear end the defective freight car was now in third place, meaning 26 of the 28 freight cars had no working brakes.
The freight train was meant to hold outside Wampersdorf station to let the oncoming RoLa-service pass, so the driver activated the brakes a few hundred meters outside the station. According to the calculations and paperwork he should have needed approximately 400m/1312ft to come to a stop. Instead he was met with wildly insufficient deceleration, making the train almost completely unwilling to slow down even with an emergency stop triggered. Knowing he was headed for an occupied track the driver laid on the horn and flashed the headlights, trying to warn crew and passengers of the train at the station.
Calculations would later show that the freight train would have needed approximately 2000m/1.24mi to stop in the condition it was in, it had about two thirds of that. Thus, at 3pm, it slammed head-on into the locomotive of the parked RoLa-Service, still travelling at 30kph/19mph on impact. The collision was a case of a (just about) unstoppable force meeting an (just about) immovable object. 26 unbraked freight cars pushed against two freight cars and a locomotive that had brakes, and the RoLa’s flatbed-cars with the heavy trucks on them weren’t going anywhere either. Some of the impact-energy was absorbed by the faulty swivel-top car, buckling it and folding it upwards by several feet. However, fatally, most of the energy was absorbed by the sleeper car, which got caught as the weak link between the heavy truck-carrying cars, the freight train and its own locomotive. The back of the car was shoved upwards with such violence that the car was torn into two pieces, the rear 3/4 falling into an adjacent field while the leading section was compressed between the locomotive and the first flatbed car’s truck. The train’s driver had managed to jump out of the locomotive ahead of the collision, but six passengers in the sleeper car died in the collision as much of the sleeper car suffered a near-total loss of survival space. Another 15 passengers and the freight train’s driver survived with mostly severe injuries.
Aftermath
The first responders reached the scene at approximately 3:10pm, being met with a scene of mayhem. Most of the sleeper car was sitting in a field, save for a nearly unrecognizable chunk jammed between the RoLa’s locomotive and the truck on its leading car. Pieces of the sleeper car’s interior were strewn around the site and witnesses were trying to recover survivors from the wreckage. Both locomotives had suffered fairly localized damage to their cabs, and apart from the destroyed swivel-top car the freight cars were undamaged. One of the first emergency doctors on site climbed onto and then rappelled into the sleeper car, which was sitting on its side, rendering first aid to the nearmost survivors until they could be rescued from the wreckage.
Forces of the impact had torn the interior fittings in the sleeper car off their mounts, creating a pile of victims, injured survivors and furniture. Firefighters had to cut the mangled interior apart bit by bit and use hydraulic spreaders to gain access to trapped survivors, making for a painstakingly slow rescue before the survivors could be placed in ambulances or taken aboard rescue helicopters that had landed at the site. 4 hours after the accident a lightly injured truck driver became the last survivor to be pulled from the wreckage, having been trapped upside down in a tiny cavity in the forward section. With no sign of further living survivors a pair of large cranes was used to upright the rear section of the sleeper car and recover the crushed forward section, allowing recovery of the last victims as the wreckage was cut apart at an accelerated pace. Five victims were recovered from the forward section, which had no survival space left, one victim was found in the rear section of the sleeper car.
Most responders left the site by 11pm, with only a few firefighters remaining to illuminate the site for the investigators. How the accident had happened in a direct context was fairly clear rather fast, with both the statement from the train driver and the freight train’s data-logger painting the same picture. Somehow, the train’s braking-performance was more than 4x worse than it should have been according to the paperwork. But how had it come to that? How did a train with seemingly next to no working brakes clear two brake-checks and make it quite a distance before things went fatally wrong?
The distribution of the damage across the trains was an early sign of what had gone wrong. Going south to north investigators came across 25 undamaged freight cars, followed by a severely damaged freight car and two more undamaged freight cars. The freight train’s locomotive had suffered mild collision damage on both ends, while the RoLa’s locomotive had suffered severe damage to only its rear end where the remains of the sleeper car had caved in the driver’s cab. Following the utterly destroyed sleeper car responders passed a severely damaged semi truck on a nearly undamaged RoLa-car. The rest of the RoLa consisted of undamaged train cars loaded with mildly damaged or undamaged trucks.
The investigation concluded that the rear cars of the freight train had all had the same deceleration ahead of the accident (that being, next to none), while the locomotive and leading two freight cars had decelerated stronger. The third freight car had rear-ended the decelerating car in front of it, being buckled upwards as the rest of the train kept pushing from behind by momentum. On impact with the RoLa enough of the energy had been absorbed for the freight train to push back the RoLa’s locomotive rather than mounting its frame or derailing, but with over 20 eight-axle RoLa-cars sitting behind the locomotive with brakes applied, resisting motion, the sleeper car ended up having to absorb the remaining forces, being forced into a “headstand” before tearing in two as the rear of the car tipped over while the forward end was stuck and couldn’t follow the motion. This resulted in the somewhat odd outcome of the most severely damaged cars not being anywhere near one another during the collision.
Since the RoLa had been parked investigators focused on the freight train, especially the suspiciously damaged third car. They found that it brakes hadn’t been applied as they were supposed to be, and soon found the faulty (different sources say it was physically damaged or just severely rusted) valve which had kept the pneumatic lines separated between cars 2 and 3. Looking into the train car’s history they found that the fault had actually been known ahead of the train’s departure in Hungary, but with the train car placed at the back of the train enough braking-power was achieved and the train could be sent on its way while being in accordance with all guidelines regardless of the faulty car.
So called brake-weight, the mass a train’s brakes can stop within a certain distance, is a major topic in the configuration and operation of trains. If a train’s brake-weight comes in below the train’s actual weight the train’s speed has to be reduced or further cars have to be added to achieve a positive relation. Austrian guidelines offer a third option, with trains being given different routes with the signals further apart, thus creating longer braking-distances for trains. The exact weight and brake-weight of the freight train aren’t known, but it’s safe to assume that brake-weight was high enough to have one unbraked car run at the back of the train. An empty Tadgs 959 weighs just 25 metric tons, to compare, the sleeper car had an empty weight of approximately 41 metric tons.
With that in mind there was nothing to be held against the workers at Sopron (Hungary), they had cleared a train for departure that wasn’t in perfect working order but still safe to operate under the circumstances. The issue arose when the train changed locomotive and direction at Ebenfurth station, just a few kilometers away from Wampersdorf. A breakdown of communications resulted in the faulty car not being moved from one end of the train to the other, turning it from the rear car into the third car when the Austrian locomotive and two cars were added to the former rear end of the train, turning it into the leading end.
Shunting workers coupled the train cars together but missed or ignored the faulty valve. Regulations from the ÖBB dictated that following the addition of new cars and/or a new locomotive a new brake-check is to be conducted by sending a shunting-worker to the far end of the train to ensure the brakes properly apply and release on input from the locomotive. However, as it turned out, the worker on duty a Ebenfurth chose to stay at the leading car of the train rather than using his bicycle to head 660m/2165ft down the train to the rearmost car. He ensured that the brakes on the leading car worked properly, signed off on paperwork saying all the train’s brakes worked properly, and thus the dispatcher got word back that the train was ready to depart. Neither the train driver nor the dispatcher knew that 26 out of 28 train cars had no working brakes from that point onwards. The driver, not knowing that he had been given wrong information, had no chance to avoid the accident once he got even only near the speed he was instructed to reach by the dispatcher, who was similarly clueless about the fatal negligence by the shunting worker.
The accident was still dubbed the “Wampersdorf Wonder” by some, due to the number of people who did survive, especially the truck driver who had managed to survive in the forward section of the sleeper car, being trapped upside down 5m/16.5ft off the ground for hours before being freed and still surviving. His precarious position had kept the firefighters from moving the section earlier, but also made it difficult to access him, especially with the heavy tools needed to slowly peel away the wreckage around him.
The shunting-worker responsible for the train’s preparation at Ebenfurth station was eventually put on trial on charges surrounding dangerous interference with rail traffic as well as negligent cause of bodily harm and negligent manslaughter. The employee, referred to as “Mister T” by the media, had been working for the ÖBB since 1971 and appears to have had an unproblematic record until the accident. The trial, during which 14 experts and witnesses were heard, concluded that T was the main person at fault, backed up by the defendant openly admitting that he had skipped the trip to the far end of the brake for no particular reason (“I had enough time, but I wanted to move things along”). He even admitted that this was far from the first time he had done a brake-check that way, a fact made worse when both his superior and the dispatcher confirmed that it was well known that shunting-workers would occasionally skip the lengthy trips up and down the length of the train.
The defendant’s lawyer argued that his client wasn’t alone in his fault, explaining that he felt that the list of defendants was missing several people. In his opinion the shunting worker who had signed off on the assembled train in Hungary should have been put on trial as well, along with the head of the maintenance-department for the Raab-Ödenburg Railway where the train car had been in service for months with the defect in place. The judge, pointing out that putting the faulty car on the back of the train was perfectly fine, still saw sufficient guilt with the defendant, sentencing him to two years in jail set out for probation. After the accident the ÖBB promised to keep a closer eye on the shunting workers, working to ensure a higher adherence to the rules and avoid operational negligence from routine (referring to a mindset of “I don’t have to check that because it will be fine anyway”).
The sleeper car’s remains and the swivel-top freight car were scrapped after the accident, but the rest of the two trains actually returned to service. ÖBB 1044 213, the locomotive that had pulled the RoLa-train, was fitted with remote control for push-pull trains during the repairs, making it 1144 213. As of 2023 a few locomotives of the type are still in service in Austria, mainly with push-pull trains as regional passenger services or in freight service. Their days are very much numbered, with more and more new locomotives and multiple units replacing the aging locomotives, making operations cheaper along the way. RoLa-Services are still in operation and still popular, despite being the least efficient form of so-called combined transport. More and more companies chose to have only the trailers or containers carried by train and have them received by a new truck and driver at their destination. This also allows more cargo to be carried on the same length of train.
ÖBB 1142 685, the locomotive that had pulled the freight train, is also still in service with the ÖBB as one of the type’s last surviving units. The final retirement of the type is scheduled for 2028. 1142 685 received a new cab during the repairs as its rear cab had been severely damaged, giving the rear end the refit-appearance with the new headlights. The other end, however, was left alone, remaining in the pre-refit configuration. This “two-face”-condition, which was never addressed since the repairs, makes the locomotive unique and recognizable among units of the type.
_______________________________________________________________
A kind reader has started posting the installments on reddit for me, I cannot interact with you there but I will read the feedback and corrections. You can find the post right here.
