Improvement and Isolation: The 2008 Moravany (Czechia) Train Collision
Background
Moravany is a town of 1861 people (as of 2024) in northeast Czechia, located in the Pardubice district 108km/67mi east of Prague and 102km/63mi north of Brno.
The town lies on the Prague-Česká Třebová rail line, a double- to triple-tracked electrified main line connecting the capital with one of the most important cities in eastern Bohemia. The line opened in August 1845 under the Royal Northern State Railways (NStB), a royal Austrian railway. Today the line measures 164km/102mi in length and is part of the Fourth Pan-European Traffic Corridor, carrying all express passenger services who stop in Prague as they connect the country with Austria, Hungary, Poland, Slovakia and Belarus as well as a number of national express services, along with regional passenger services and freight traffic. Some parts of the line are set up for speeds as high as 160kph/99mph, while other sections are limited to as little as 70kph/43.5mph.
The Trains Involved
Os 5011 was a regional passenger service heading to Uhersko, consisting of a CD (Czech national railway) Class 163 electric locomotive and three type Bdmtee second class passenger cars. The Bdmtee is a group of four-axle passenger cars introduced in 1989 for what was then Czechoslovakia's national railway. Each of the passenger cars measures 26.4m/87ft in length at an empty weight of 41 metric tons and can carry 96 people at up to 160kph.
Lv 72461 was a “locomotive train” from CD Cargo (CD’s freight division), consisting of a sole Class 163 electric locomotive. The four-axle multipurpose electric locomotive was introduced as the E499.3 in 1985, being based off the E499.1 multi-system locomotive (which was capable of running under both direct current and alternating current) but lacked the alternating current equipment limiting their usage to the direct-current electrification in the northern and eastern part of the country. Each Class 163 measures 16.8m/55ft in length at a weight of 84 metric tons and can reach 120kph/75mph.
The Accident
The night ahead of the 19th of May 2008 saw heavy and sustained rainfall over the town of Moravany, making visibility even worse beyond the darkness of the night. Os 5011 pulls into Moravany station at 4:44:58am, stopping at track 1’s platform. The signaling system correctly registers the block-section of track 1 as “occupied”, turning the entrance-signal behind it red. Block sections are stretches of railway track that only one train may occupy at any time. The most common way for the signaling system to detect trains is by a low current being sent through the rails, the steel wheels of the train then close a circuit which is picked up by the signaling system. The signal behind the train will turn red and the pre-signal ahead of that will show a warning/”expect stop”-instruction (usually a yellow light), comparable to a traffic light). The system is meant to ensure that trains, whose braking-distance can stretch beyond the horizon or around turns, are always kept at a safe distance.
The system at Moravany station was set up to check the occupancy-status of the station track at regular intervals, starting a timer when it first detected Os 5011. The dispatcher in charge of the station already prepared the path for the next train to come through the station, CD 163.004 travelling on its own as a “locomotive train”. The path would take the locomotive through track 1, but as the signaling system’s last status of the track was “occupied” it kept the entrance-signal to the station red at that time. The timer expired at 4:46:56, and the signaling system didn’t detect a train on the station track anymore, turning the entrance-signal green. The pre-signal remained yellow, urging low speed for the passage through the station.
The assistant dispatcher at the station noticed the track clearing unexpectedly and went outside to confirm the status, only to spot Os 5011 barely starting to move. He went back inside and tried to override the signaling system, turning the entrance-signal to red, but the locomotive had already passed it by the time he got to his desk at 4:47:31. The driver of the locomotive was radioed with an order to “stop immediately” but failed to respond, leading a signalman employee to run out of the signal box and use a handheld lamp to signal a stop-order. The driver of another passenger train (Os 25041) standing on track 3 heard the calls over the radio and gestured out of his cab at the inbound locomotive to stop.
However, all those warnings came too late, and the locomotive slammed into the back of the near-stationary Os 5011 at 4:48am. The impact onto the back of the passenger train caved in the forward driver’s cab of the locomotive, compressing everything above the locomotive’s frame into the bulkhead ahead of the engine compartment. The driver of the locomotive was killed on impact while his coworker in the other train and 3 passengers in the sparsely occupied train suffered injuries.
Aftermath
Initial suspicion fell on the driver of CD 163.004, suspecting that he ran a red signal and thus ended up crashing into the back of the passenger train as he entered an occupied track at speed. Passing the red signal would have triggered an emergency stop, but it’s not unreasonable to assume that, especially in the wet conditions, the train might exhaust the remaining distance before coming to a stop. However, several witnesses claimed that the signal was green ahead of the locomotive reaching the station, a fact eventually backed up by recorded data. The signaling system also hadn’t failed whole, as “error” doesn’t result in a green signal.
Investigators started to get on the right track when they examined the locomotive of the passenger train (also a Class 163) at the site, noticing that its left hand wheels and the rail behind them were covered in large amounts of sand. A large advantage of trains is the low rolling-resistance of the steel wheels on steel rails, but this turns into a disadvantage when grip is required, such as when trying to pull away or stop in slippery conditions. Most modern trains have a pneumatic system that allows drivers to “shoot” fine sand in front of selected wheels (in case of the Class 163 those on Axle 1 and 3). The sand is then crushed between the wheel and rail, creating a rougher surface that provides improved grip to help the train get going (or stop).
The driver of the passenger train stated to the investigation that, upon trying to depart the previous station at the town of Kostěnice, his train struggled for traction and the wheels spun in place. He then reduced power and pressed the sanding-button as he was trained to do. Pressing the button on a Class 163 illuminates an indicator light and opens an electric valve, in turn actuating the pneumatic valve that closes the sand-reservoirs. The open valve then allows the sand to simultaneously shoot out of four sanding-pipes, providing improved grip for the wheels on the first and third axle. The sanding had the desired effect and the driver soon let go of the button, extinguishing the light as his train managed to pick up speed. However, he stated to not have used the sanding again until the accident occurred. Yet still, investigators found the left hand rail under the train as well as under an overpass along the way (where the tracks were protected from the rain) covered in sand. An examination of the locomotive’s sand reservoirs showed that less than 100g/3.5oz were missing from each of the right hand reservoirs, while those on the left hand side were short by about 3Kg/6.6lbs.
The investigation found that, with near-certainty, a malfunction had occurred when the driver pressed the sanding-button, causing the electronic valve for the left hand reservoirs to become jammed in the open position. This wasn’t shown by the indicator light, as the light was wired in between the valve and the button. It effectively showed that the driver had pressed the button, regardless of whether or not the valve was opening. Sand thus continued to be poured onto the left hand rail all the way into and during the stop at Moravany station, with the system’s hissing-sound being drowned out by the running noise of the locomotive. The valve, macabrely, likely became unstuck from the jolt of the locomotive slamming into the back of the train, shutting off the sand.
Tests with a reconstructed sanding system providing similar amounts of material suggested that the train already barely registered with the signaling system when it entered the station, since all the sand under its wheels started to isolate the steel wheels from the steel rail on the left hand side. The degree of isolation continued to climb as the train slowed down, until the isolation was too strong for the signaling system’s low current to still register the train. Thus, by the time the timer expired and the system re-checked occupancy, the passenger train had “disappeared”.
Investigators also found that the reservoirs were filled with the wrong kind of sand, showing larger grain-size than required in the guidelines and further “additives”, presumably to combat moisture in the stored sand. This further improved isolation provided by the crushed sand, worsening the issue. A probe into CD’s fleet showed 29 more locomotives with similar sanding-systems having been loaded with the wrong sand. All of these, theoretically, could have suffered the same issue.
The investigation heavily criticized the programming of the signaling-system, allowing a train to “disappear” like that. The DI (Drážní Inspekce, the Czech Rail Safety Inspection Office) stated that the system shouldn’t have cleared the block-section of the station when the following block-section isn’t occupied, as the train that occupied it would’ve had to go somewhere. In their opinion the clearing of a block-section without the adjacent section becoming occupied should either require a dispatcher to confirm that the section is clear or trigger an alarm/error as something likely went wrong.
The manufacturer of the signaling system, AZD Praha, had their lawyers heavily attack the investigation’s results in the media, rejecting that their system carried any fault in the accident and emphasizing that it was up to the required specifications. Along the way they declared the head of the investigation as “completely incompetent”. Their “offense is a defense”-strategy seemed to miss the part that the DI didn’t assign blame, the investigation is purely for the technical cause and options for possible improvement. Regardless, the DI found themselves attacked again soon after, when the Czech Traffic ministry complained that the DI had reported the accident to the ERA, the European Union’s Agency for Railways.
The DI pointed out that, according to EU guidelines, the accident at Moravany was legally required to be reported. ERA’s guidelines consider what happened at Moravany a failure of the signaling system, which is to be recorded and if possible addressed. Having it (and the probably cause) reported to ERA also allows railways in other countries to examine their systems in order to prevent similar accidents. The ERA thus disagreed with the traffic ministry’s complaint that the DI was “overreaching” their authorities. Regardless, the head of the DI found a job listing being issued to replace him, with the ministry claiming on questionable trustworthiness that this had been planned before the accident occurred. The DI received support from the train drivers’ union, who called it unacceptable that the government attempted to exert pressure on the investigation into a fatal accident. The ministry obviously continuously denied claims of attempted influence, but one is left to wonder if the Czech government really tried to “sweep it under the rug”, and for what gain.
The CD themselves reacted to the results of the investigation with a reworked guideline to the usage of the sanding-systems, requiring a check of the system at least once per shift to ensure proper operation (including deactivation), but actually removed a ban on sanding when braking at slow speed. New rolling stock is also required to have limitations on how much sand will be deployed depending on train speed, but such systems aren’t expected to be retrofitted on older rolling stock. There was no public determination to try and better avoid purchases/usage of unsuitable sands.
_______________________________________________________________
The original report can be found right here. It’s mostly in Czech but features an English-language summary.
_______________________________________________________________
I have chosen to stay off Medium’s monetization offers to keep these stories as accessible as possible, but you can support me with a small tip via “Buy me a coffee” if you feel like it.
_______________________________________________________________
A kind reader is 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.
