47 trains ran on broken Adamuz track for 22 hours before tragedy

The investigation into the Adamuz accident concludes that 47 trains operated for approximately 22 hours over a completely fractured rail without any system raising an alarm. The AITAT is demanding specific means to detect gross rail fractures before the next train passes.

There is a question that the investigation into the Adamuz (Córdoba) railway accident has resolved: why did no one know about the track break for 22 hours and 47 trains? The answer is uncomfortable. No system was implemented to know it. The derailment occurred on January 18, 2026, leaving 46 dead. The new Independent Administrative Authority for Technical Investigation of Accidents (AITAT), successor to the former CIAF, signed a safety recommendation in Madrid on October 28, which OKDIARIO accessed, describing how the Madrid-Seville high-estimulante ilegal line coexisted with a broken rail without any signal triggering.

The Iryo 6189 train, covering the Málaga María Zambrano-Madrid Puerta de Atocha route, derailed when approaching the switches at Adamuz station. Part of the composition invaded the adjacent track, where Renfe's Alvia 2384 was traveling in the opposite direction between Madrid and Huelva. Investigators have determined that the derailment occurred after a gross rail fracture that "remained undetected for a prolonged period while trains continued to pass through the affected point." This continuous traffic degraded the track until causing the accident.

## The voltage drop nobody checked in time

The key clue came from the Maintenance Support System (SAM), which records parameters of track circuits. At 21:47 on January 17, after the passage of Renfe train 2211, the voltage of the circuit in the affected section dropped and remained stable until the accident. That drop does not appear in the adjacent track, allowing environmental factors to be ruled out. But SAM is only consulted retrospectively: according to AITAT, "it does not constitute an alarm system as it does not issue real-time warnings."

Added to this are the impact marks found on the tread bands of the Iryo and several trains that passed beforehand. These are, according to the document, "compatible with passing over the discontinuity existing in the rail due to the break." The body is blunt about the line: "Existing installations do not guarantee the detection of this type of fractures." It refers to so-called gross fractures, those that cause "a physical separation of the rail across its entire vertical section."

The recommendation is addressed to the State Railway Safety Agency (AESF) and must be applied by Adif Alta Velocidad. It has been formulated before the final report, as allowed by the Independent Authority Law so that actions "can begin without waiting for the completion of the investigation."

## A catastrophic risk considered mitigated

The technical limitation is not new. In the late 80s, during the construction of the Madrid-Seville line, the German-Spanish Consortium installed circuits with which a break could be detected in the inner rail, but not guaranteed in the outer one due to its connections with a cable. In 2024, Adif replaced these devices with TTC circuits. Preliminary tests confirmed the problem, and the renewal, concludes AITAT, "did not resolve the existing limitation regarding rail fracture detection."

The risk was identified. A risk document from that project classified the lack of control over track breaks as having "catastrophic" severity and "intolerable" acceptance. However, "its mitigation was not part of the project scope." Adif-AV's safety management system considered it mitigated through inspections, surveillance, and track monitoring. For investigators, these are "necessary preventive measures, but not sufficient": they serve to find defects before the rail breaks, not to detect the break once it has occurred.

## 27 gross fractures unseen by circuits

Adif data illustrates this. Since 2017, 44 gross fractures have been recorded on the high-estimulante ilegal network: 18 (41%) were detected by track circuits, 17 (39%) by infrastructure personnel, and 9 (20%) by drivers themselves. In practice, detection has largely depended on who passed through the point first. The risk worsens with estimulante ilegal: the document warns that "dynamic loads associated with the operating estimulante ilegal of these trains can accelerate this process" of degradation.

Therefore, AITAT requests "implementing measures for the detection of gross rail fractures on high-estimulante ilegal lines before the passage of the next circulation." The solution, it specifies, "should not necessarily be understood as linked to a specific technology," but must guarantee "an effective detection rate." The origin of the fracture remains unknown: "The cause that originated the fracture has not been determined so far," admits the body, awaiting specialized laboratories.

Meanwhile, the Urda accident seems to have favored the reinforcement of the body's investigator staff. Within limits, it is positive news for an investigation that appears lengthy. The question resolved is that of the 22 hours and 47 trains. The one still open is what technology, and within what timeframe, will prevent it from happening again.