• Our booking engine at tickets.railforums.co.uk (powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!

Catastrophic derailment in Taiwan 21/10/18

Status
Not open for further replies.

Intermodal

Established Member
Joined
3 Nov 2010
Messages
1,255
Location
I wonder how long I can make my location on this f
There has been a catastrophic derailment in Taiwan today, BBC is reporting at least 17 dead and 137 injured at the present time. The derailment occured at 9am GMT.

Taiwanese news reporting that potentially dozens are still trapped inside.

Track appears to be completely destroyed, and the train is in a very bad way. The derailment appears to have occured on a curve.

https://www.bbc.co.uk/news/world-asia-45932991
At least 18 people have been killed and 187 injured after a passenger train derailed in north-east Taiwan.

Railway authorities say they are investigating the accident, which happened in Yilan County at about 16:50 local time (08:50 GMT) on Sunday.

A total of 366 passengers were on the train when all eight of its carriages derailed.

The incident is Taiwan's worst rail accident in 27 years. At least three children were among the dead.

"The train was going very fast. I hit a wall when the car started to flip. Around five to six people were thrown out of the carriage door," passenger Henry Tseng told Reuters news agency.

"There [was] no time to think what happened. Everyone was in a rush to get out."

It is not clear if any passengers remain unaccounted for - with reports providing conflicting details.

_103949638_hi050097851.jpg
 
Last edited by a moderator:
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R

RailUK Forums

Laryk

Member
Joined
11 Jun 2011
Messages
188
Location
Taiwan
Looks to be a terrible tragedy.
I have travelled on this route a few times. It has to deal with some challenging terrain in a particularly active seismic zone.
The train "Puyuma" or "TEMU200" tilts to allow an increased speed over the older push-pull stock.
Taiwan's railways are very busy on weekends, however Puyuma services operate on a reservation only system, so all passengers would have had a booked seat with no standing passengers allowed.
From the north travelling south this station sits on the first significant curve since Yilan (approx 20km).
 
Last edited:

Laryk

Member
Joined
11 Jun 2011
Messages
188
Location
Taiwan
Linespeed drops from 120 km/h down to 85km/h for tilting stock and 65 km/h for non tilt.

Sources:
Southbound approaching curve: (at 1:11:11)

Northbound leaving curve: (at 42:42)
 

Attachments

  • upload_2018-10-21_17-39-15.png
    upload_2018-10-21_17-39-15.png
    1.4 MB · Views: 10
Last edited:

najaB

Veteran Member
Joined
28 Aug 2011
Messages
30,785
Location
Scotland
I do not wish to speculate but clearly there is one possible cause at least.
The photgraphic evidence certainly points towards a speed-related cause for the derailment, if so the investigation will need to focus on if the root cause lies in design or operation.
 

MarkyT

Established Member
Joined
20 May 2012
Messages
6,245
Location
Torbay
It may be another overspeed in curve with protection not provided/active incident. These seem to be dominating recent high casualty rail accidents the world over. Lack of effective technical speed supervision and protection on approach to significant speed reductions for curves seems to be extraordinarily dangerous, especially where the curves coincide with 'holes', temporary or otherwise, in protection that applies widely elsewhere. I was surprised the train was able to run normally at full speeds if, as suggested, the ATP was turned off. It looks like they may be able to fallback to an older protection and warning system, in the way that Thameslink trains can step down from ETCS/ATO to manual with signals and TPWS/AWS if neccessary, but if the backup/legacy system can't do full speed supervision or even overspeed trapping then its no protection at all against this kind of incident.

http://focustaiwan.tw/news/asoc/201810220014.aspx
The driver had reported unusually low pressure from the air compressor, which controls the power system, said Lai Sui-chin (賴隨金), deputy head of the TRA's Rolling Stock Department.
...
"The driver reported the problem many times as the train traveled between Toucheng and Jiaoxi stations," Lai said, but added that the glitch should have been fixed because the train was serviced by technicians when it stopped at Yilan Station.
...
While the malfunction itself is unlikely to trigger an acceleration, the TRA is also considering the possibility that the automatic train protection (ATP) system was turned off manually, said director-general Lu Chieh-shen (鹿潔身).
...
It has been suggested that in order to make the power system work, the driver turned off the ATP.
According to TRA video surveillance footage, the train was likely traveling above the speed limit of 75-80 kilometers per hour when the accident occurred.

https://en.wikipedia.org/wiki/Automatic_train_stop#Taiwan
Taiwan
Many Taiwan Railways Administration trains are equipped with the Japanese ATS-SN and ATS-P systems, which serve as fallback for an ATP system introduced in 2006, of which the latter system replaced the older AWS system originally introduced in 1978 on some express trains.
 

Groningen

Established Member
Joined
14 Jan 2015
Messages
2,866
I believe this curve has not tilted tracks (where the track is higher (or lower) than the other one). To be honest i In Harderwijk (Netherlands) such a curve is taken with 110 kilometers an hour. Question to Laryk: i have not the time to watch 2 hours of Taiwan railways. At what point is the fatal curve?
 

Laryk

Member
Joined
11 Jun 2011
Messages
188
Location
Taiwan
Question to Laryk: i have not the time to watch 2 hours of Taiwan railways. At what point is the fatal curve?

The videos were linked to specific times. The first should be 1:11:11 and the second 42:42. I will edit my post for clarity.
 

Groningen

Established Member
Joined
14 Jan 2015
Messages
2,866
Word is that the speed was 140 kilometers an hour, where it should have been 75. The speedcontrol was switched off by the engineer. Even this curve is not suitable for 140 in the Netherlands.
 
Status
Not open for further replies.

Top