Alton Towers "crash"

[Bellend Sebastian]

http://en.m.wikipedia.org/wiki/Incidents_at_European_amusement_parks

There have been one or two issues in the past, seemingly.

I've not been to one since 1989.

I may have a false memory about this but I'm sure something bad happened on The Black Hole (which was replaced by Smiler, apparently) around when it opened, but I cannot find anything on it.

Any oldies remember this?

[stripeyfox]

Yes but all rollercoasters should need to go through a series of safety checks, from a technological point of view the ride was safe, which means the rigorous testing standards for rollercoasters and rides is sufficient and there is not a fundamental flaw in the safety checks that could pose a risk to the safety of people on other rides.

Imagine it was a car, if you heard that a Ford Focus just careered off the road and crashed into another car killing people, if it was the fault of the driver then it is just as tragic, but there is no issue with the Ford Focus, if there was an error with the Ford Focus that caused it to crash then you would never get in another one and if you owned one you would sell it but nobody would buy it so you would scrap it and make a huge loss.

Roller coasters are safe and there was a human error, is more comforting than roller coasters aren't safe and there wasn't a human error by someone at Alton Towers.

I agree with you. I suppose it is the same that when an Airbus flies into the side of a mountain or a 777 vanishes without trace, it is better on balance, that it was caused by a human (error or deliberate) rather than by a fault with the system. Still scary though.

Be interesting to see what level of competence a ride operator is. What training is involved. Or is it just youngsters on a gap year being paid minimum wage and told "green button for go, red button for stop".

[Captain...]

I agree with you. I suppose it is the same that when an Airbus flies into the side of a mountain or a 777 vanishes without trace, it is better on balance, that it was caused by a human (error or deliberate) rather than by a fault with the system. Still scary though.

Be interesting to see what level of competence a ride operator is. What training is involved. Or is it just youngsters on a gap year being paid minimum wage and told "green button for go, red button for stop".

 

Hopefully they would not have the ability to override a safety warning, that would be worrying, you would expect an experienced engineer would be needed to override a safety warning.

[stripeyfox]

Looking over on coasterbuzz forums - like Foxestalk but for theme parks, and without the negativity - it seems that most rides have a "maintenance mode" activated by key turn. People posting there seem to think that this would be the mode which would disable the automatic braking system which prevents collisions. An empty train was sent out and "valleyed" - which I assume means wasn't able to crest the next hill and became stuck.

http://coasterbuzz.com/Forums/Topic/alton-towers-smiler-trains-collide-at-least-four-serious-injuries

[ADK]

I'm not a rollercoaster expert, but this sort of thing shouldn't happen with a modern safety system. It shouldn't be possible to simply override the safety mechanism.

 

 

 

All systems need a manual override, if the situation was that the safety check was in error and there were people stuck at the top of the big drop because of an erroneous safety warning and they couldn't be released, and the system couldn't be reset then they would still be stuck there.

 

All the points where the rollercoaster is designed to be held have some sort of emergency access in case the passengers need to be evacuated. It shouldn't be permissible to deactivate the safety mechanism with passengers on the ride, that would be negligence. If there is a fault then passengers are evacuated.

 

It doesn't really make sense, any sort of fault should set a flag on the control system to stop the ride. Then everyone should be evacuated and maintenance called and the fault investigated. The ride operators shouldn't have the capability to simply override the safety mechanism.

 

I can only think of a software/hardware fault or really major breaches of procedure being able to cause this and both seem unlikely.

[ramboacdc]

Looking over on coasterbuzz forums - like Foxestalk but for theme parks, and without the negativity - it seems that most rides have a "maintenance mode" activated by key turn. People posting there seem to think that this would be the mode which would disable the automatic braking system which prevents collisions. An empty train was sent out and "valleyed" - which I assume means wasn't able to crest the next hill and became stuck.

http://coasterbuzz.com/Forums/Topic/alton-towers-smiler-trains-collide-at-least-four-serious-injuries

you are correct in what valleying is there.  the next question is why wasn't it counted in? 

[Fox92]

http://en.m.wikipedia.org/wiki/Incidents_at_European_amusement_parks

There have been one or two issues in the past, seemingly.

I've not been to one since 1989.

I may have a false memory about this but I'm sure something bad happened on The Black Hole (which was replaced by Smiler, apparently) around when it opened, but I cannot find anything on it.

Any oldies remember this?

 

I can remember the Runaway train incident.

 

As I said on the previous page, I'm not a massive fan but the last time I went to Alton Towers I was 13 which was ten years ago now. 

[stripeyfox]

you are correct in what valleying is there. the next question is why wasn't it counted in?

http://www.themeparkreview.com/forum/viewtopic.php?f=2&t=55756&start=4160

These are not my words - but from another forum (link above). Seems plausible though.

100% human error. This is the only conceivable scenario considering the amount of safety systems and redundancies that are involved in modern ride control systems:

(Assume train 1 = train dispatched first, train 2 = train dispatched next)

Assumptions:

•Ride control system allows train 2 to be dispatched (to lift, let's call it Block A) before train 1 has cleared the block immediately following lift (let's call this Block B). This dispatch is allowed because it is assumed that train 1 will clear Block B (reach the next brake or lift) in time for train 2 to be allowed to continue into the block.

•If train 1 has not cleared Block B by the time that train 2 reaches a certain point on the lift, the lift stops and is not allowed to restart until that block is clear. (The lift may automatically restart when the block is clear, or the operators may need to manually restart the lift; which of these is the case varies from ride to ride)

Alton Towers incident

In this incident, we can deduce the following:

There are reports that the loaded train (train 2) was sitting idle on the lift for a while (I think I read 5 minutes somewhere). We also know from the collision that there was a train (train 1) that became stuck in a valley in the next block (which we called Block B). Putting 2 and 2 together, we can assume that the ride's safety systems acted as designed and stopped the lift, holding train 2 in Block A until Block B cleared.

Now, these complex control systems don't just "forget" which blocks are occupied. A power failure or "system reset", as someone in this thread suggested, of a modern ride control system would result in all of the ride's blocks being set, not the opposite. Otherwise the unthinkable would occur if a ride were to lose and regain power in a short amount of time. Blocks are also not accidentally cleared by anomalies, such as a bird flying in front of a photoeye. Safety-critical control systems are designed to fail to a safe state, and they are very redundant. For example, a bird flying in front of a photoeye may result in an additional block being set, not in a block being cleared. There are all sorts of sensors and components that can have a say in whether or not a block is occupied: proximity sensors, photoeyes, brakes, and even motors. If even the smallest anomaly is detected with any of them, the block will be set, just to be safe. "We don't know what caused X, but it could be a train. We'd rather assume that it is a train than assume otherwise and cause a collision." This is what causes "ghost trains."

With that in mind, let's go back to the above scenario. A ride technician responds to a breakdown, and sees that train 2 has been sitting on the lift (in Block A) for 5 minutes or so. He looks at the logs and sees that the control system is holding it there because Block B is still occupied, meaning by extension that Block B has been occupied for at least 5 minutes. A valley certainly is far from a normal occurrence, so it was not likely anywhere near the front of his mind. Additionally, judging from the picture, it appears that this ride is low-capacity, which means that it likely runs at least 4 or 5 trains at all times. It seems very likely that one could easily mis-judge the number of units on the track, pair that with the fact that a free-running block has been occupied for at least 5 minutes, and assume that some anomaly has resulted in a ghost train (as mentioned above). (In reality, however, we now know that Block B actually still is occupied.) The technician in this situation then likely cleared Block B, allowing train 2 to continue, where it then collided with train 1.

The technician should have at least verified visually that there was not a train in the block that he was clearing, but he really should not have been clearing a block with a loaded unit on the course, anyway. (Doing so is actually in violation of an ANSI standard.) However, clearing a block without even verifying that it was indeed clear would certainly be the most horrendous portion of this violation.

I hope that didn't come off as rambling, but I'm trying to make the point that modern control systems--especially safety-critical ones--are redundant, fail to a safe state, and were not the cause of this accident. This is simply the only reasonable chain of events that could've led to this collision. "Glitches" don't cause this. "Bad design" doesn't cause this. Human error does.

[VLC86]

http://www.themeparkreview.com/forum/viewtopic.php?f=2&t=55756&start=4160

These are not my words - but from another forum (link above). Seems plausible though.

100% human error. This is the only conceivable scenario considering the amount of safety systems and redundancies that are involved in modern ride control systems:

(Assume train 1 = train dispatched first, train 2 = train dispatched next)

Assumptions:

•Ride control system allows train 2 to be dispatched (to lift, let's call it Block A) before train 1 has cleared the block immediately following lift (let's call this Block B). This dispatch is allowed because it is assumed that train 1 will clear Block B (reach the next brake or lift) in time for train 2 to be allowed to continue into the block.

•If train 1 has not cleared Block B by the time that train 2 reaches a certain point on the lift, the lift stops and is not allowed to restart until that block is clear. (The lift may automatically restart when the block is clear, or the operators may need to manually restart the lift; which of these is the case varies from ride to ride)

Alton Towers incident

In this incident, we can deduce the following:

There are reports that the loaded train (train 2) was sitting idle on the lift for a while (I think I read 5 minutes somewhere). We also know from the collision that there was a train (train 1) that became stuck in a valley in the next block (which we called Block B). Putting 2 and 2 together, we can assume that the ride's safety systems acted as designed and stopped the lift, holding train 2 in Block A until Block B cleared.

Now, these complex control systems don't just "forget" which blocks are occupied. A power failure or "system reset", as someone in this thread suggested, of a modern ride control system would result in all of the ride's blocks being set, not the opposite. Otherwise the unthinkable would occur if a ride were to lose and regain power in a short amount of time. Blocks are also not accidentally cleared by anomalies, such as a bird flying in front of a photoeye. Safety-critical control systems are designed to fail to a safe state, and they are very redundant. For example, a bird flying in front of a photoeye may result in an additional block being set, not in a block being cleared. There are all sorts of sensors and components that can have a say in whether or not a block is occupied: proximity sensors, photoeyes, brakes, and even motors. If even the smallest anomaly is detected with any of them, the block will be set, just to be safe. "We don't know what caused X, but it could be a train. We'd rather assume that it is a train than assume otherwise and cause a collision." This is what causes "ghost trains."

With that in mind, let's go back to the above scenario. A ride technician responds to a breakdown, and sees that train 2 has been sitting on the lift (in Block A) for 5 minutes or so. He looks at the logs and sees that the control system is holding it there because Block B is still occupied, meaning by extension that Block B has been occupied for at least 5 minutes. A valley certainly is far from a normal occurrence, so it was not likely anywhere near the front of his mind. Additionally, judging from the picture, it appears that this ride is low-capacity, which means that it likely runs at least 4 or 5 trains at all times. It seems very likely that one could easily mis-judge the number of units on the track, pair that with the fact that a free-running block has been occupied for at least 5 minutes, and assume that some anomaly has resulted in a ghost train (as mentioned above). (In reality, however, we now know that Block B actually still is occupied.) The technician in this situation then likely cleared Block B, allowing train 2 to continue, where it then collided with train 1.

The technician should have at least verified visually that there was not a train in the block that he was clearing, but he really should not have been clearing a block with a loaded unit on the course, anyway. (Doing so is actually in violation of an ANSI standard.) However, clearing a block without even verifying that it was indeed clear would certainly be the most horrendous portion of this violation.

I hope that didn't come off as rambling, but I'm trying to make the point that modern control systems--especially safety-critical ones--are redundant, fail to a safe state, and were not the cause of this accident. This is simply the only reasonable chain of events that could've led to this collision. "Glitches" don't cause this. "Bad design" doesn't cause this. Human error does.

Sounds logical (what I understood of it).

Always struck me how young the people operating roller coasters are. Granted you aren't going to get many pensioners wanting to operate them but still.

One thing that really gets me is, how he hell didn't any of the staff working on the ride notice another car on the track?

[ramboacdc]

http://www.themeparkreview.com/forum/viewtopic.php?f=2&t=55756&start=4160

These are not my words - but from another forum (link above). Seems plausible though.

100% human error. This is the only conceivable scenario considering the amount of safety systems and redundancies that are involved in modern ride control systems:

(Assume train 1 = train dispatched first, train 2 = train dispatched next)

Assumptions:

•Ride control system allows train 2 to be dispatched (to lift, let's call it Block A) before train 1 has cleared the block immediately following lift (let's call this Block B). This dispatch is allowed because it is assumed that train 1 will clear Block B (reach the next brake or lift) in time for train 2 to be allowed to continue into the block.

•If train 1 has not cleared Block B by the time that train 2 reaches a certain point on the lift, the lift stops and is not allowed to restart until that block is clear. (The lift may automatically restart when the block is clear, or the operators may need to manually restart the lift; which of these is the case varies from ride to ride)

Alton Towers incident

In this incident, we can deduce the following:

There are reports that the loaded train (train 2) was sitting idle on the lift for a while (I think I read 5 minutes somewhere). We also know from the collision that there was a train (train 1) that became stuck in a valley in the next block (which we called Block B). Putting 2 and 2 together, we can assume that the ride's safety systems acted as designed and stopped the lift, holding train 2 in Block A until Block B cleared.

Now, these complex control systems don't just "forget" which blocks are occupied. A power failure or "system reset", as someone in this thread suggested, of a modern ride control system would result in all of the ride's blocks being set, not the opposite. Otherwise the unthinkable would occur if a ride were to lose and regain power in a short amount of time. Blocks are also not accidentally cleared by anomalies, such as a bird flying in front of a photoeye. Safety-critical control systems are designed to fail to a safe state, and they are very redundant. For example, a bird flying in front of a photoeye may result in an additional block being set, not in a block being cleared. There are all sorts of sensors and components that can have a say in whether or not a block is occupied: proximity sensors, photoeyes, brakes, and even motors. If even the smallest anomaly is detected with any of them, the block will be set, just to be safe. "We don't know what caused X, but it could be a train. We'd rather assume that it is a train than assume otherwise and cause a collision." This is what causes "ghost trains."

With that in mind, let's go back to the above scenario. A ride technician responds to a breakdown, and sees that train 2 has been sitting on the lift (in Block A) for 5 minutes or so. He looks at the logs and sees that the control system is holding it there because Block B is still occupied, meaning by extension that Block B has been occupied for at least 5 minutes. A valley certainly is far from a normal occurrence, so it was not likely anywhere near the front of his mind. Additionally, judging from the picture, it appears that this ride is low-capacity, which means that it likely runs at least 4 or 5 trains at all times. It seems very likely that one could easily mis-judge the number of units on the track, pair that with the fact that a free-running block has been occupied for at least 5 minutes, and assume that some anomaly has resulted in a ghost train (as mentioned above). (In reality, however, we now know that Block B actually still is occupied.) The technician in this situation then likely cleared Block B, allowing train 2 to continue, where it then collided with train 1.

The technician should have at least verified visually that there was not a train in the block that he was clearing, but he really should not have been clearing a block with a loaded unit on the course, anyway. (Doing so is actually in violation of an ANSI standard.) However, clearing a block without even verifying that it was indeed clear would certainly be the most horrendous portion of this violation.

I hope that didn't come off as rambling, but I'm trying to make the point that modern control systems--especially safety-critical ones--are redundant, fail to a safe state, and were not the cause of this accident. This is simply the only reasonable chain of events that could've led to this collision. "Glitches" don't cause this. "Bad design" doesn't cause this. Human error does.

this makes sense until i look at pics that first started emerging showing 2 trains (in this scenario A and B) right at the top of the lift and stopped. a start wouldn't have released A and then B. they were that tight together (already touching) it would have released both.  My Physics isn't the best but this empty one at the front would have sped off with about 4-5 seconds head start and then when got the loop started on its way back. its then been shunted and they have both gone through the loop and down to its resting spot.

i could be wrong there but thats my theory. i think if one was coming back at speed and one forward at full pelt, it would have been a lot worse (aka derailment)

[BoneDog]

http://en.m.wikipedia.org/wiki/Incidents_at_European_amusement_parks

There have been one or two issues in the past, seemingly.

I've not been to one since 1989.

I may have a false memory about this but I'm sure something bad happened on The Black Hole (which was replaced by Smiler, apparently) around when it opened, but I cannot find anything on it.

Any oldies remember this?

 

There was definitely a bad crash on the Black Hole in the late 80's. Can't remember what happened but I think that somebody died.

[Guest MattP]

I'm I the only one who thinks that adults who go to theme parks are a bit weird unless they have kids?

[BoneDog]

One thing that really gets me is, how he hell didn't any of the staff working on the ride notice another car on the track?

 

That's what I've been wondering. Were they just looking at a computer screen in the booth they sit in or did anyone actually step out and take a look at the tracks to check? I've not seen this new ride so don't know if there is any track hidden from view.

[stripeyfox]

That's what I've been wondering. Were they just looking at a computer screen in the booth they sit in or did anyone actually step out and take a look at the tracks to check? I've not seen this new ride so don't know if there is any track hidden from view.

I'm not familiar with the ride, but it seems that it would be impossible to see the entire track from any particular vantage point. It would seem that in the case of the safety system reporting the error, that the correct course of action would have been to verify the entire track. This obviously didn't happen, or if it did, it wasn't carried out correctly.

[AndWhat?]

Pretty sure the control tower or what ever it's called will have CCTV of the whole track. 

[Hollism]

Hello guys I know 2 of the people injured. They're a young couple from Barnsley a guy of 18 and a girl of 19. The guy has lost his little finger (That is the reason for the reports of blood) and both of his knees have popped out of place from what I was told last night he will walk again fortunately. His girlfriend however she has been told she could potentially be paralysed. she had to be airlifted to the nearest spinal injury specialist hospital. Praying to god they're ok.

From reddit.

[MrSpaM]

From reddit.

 

That's terrible news

[stripeyfox]

Pretty sure the control tower or what ever it's called will have CCTV of the whole track.

I dunno. The layout looks as if much of the track is below ground level and there are dozens of twists and turns. I'm sure there are CCTV cameras but there could well be blind spots.

The process must be to establish without doubt that the track is clear before overriding the error message. It would seem that the operator(s) did not achieve this.

But we just don't know. Could have been throwing up error messages all morning. It's a simple error I suppose. Terrible consequences though.

[Kitchandro]

I'm I the only one who thinks that adults who go to theme parks are a bit weird unless they have kids?

 

Yes.

Well, you and Bovril.

 

You're dead inside.

[Fox92]

From reddit.

 

That's horrific. 

[AndWhat?]

From reddit.

 Do you have a link to the thread on Reddit mate?

[benpicko]

 Do you have a link to the thread on Reddit mate?

 

http://www.reddit.com/r/unitedkingdom/comments/38casb/alton_towers_shut_amid_investigation_into/

[Nod.E]

Unfortunately for Alton Towers they'll take a massive hit in terms of footfall over the coming years. Even if only 5% of people that would have gone decide against going as a result of this, that translates as a LOT of money lost. I read that average daily attendance is 11,500 people. So that's 575 people less per day. If we assume an average price of about £45, that's £25,875 per day. With roughly 150 days left in the season, that's £3,881,250 for the rest of the season.

It's a shame because realistically it seems the blame falls at the feet of one guy who made a stupid mistake. 

The chances of something like this happening are probably less than the chances of suffering a similar fate on the drive home from the resort, and now that this has happened I would have thought they'll be taking extra measures to make double sure it will not happen again.

It's the best ride at the theme park by a stretch so I hope it doesn't get shut down. Absolutely tragic for the victims and they deserve all the compensation they get and more. 

As eluded to by somebody already, human error shouldn't be a possibility. I imagine they will review the safety system to ensure over riding the system is something only someone at the top of the pile can do after a full review. An extra half an hour or so of a ride being shut down is no price to pay if it stops somebody else losing a leg, or worse, both.

[ramboacdc]

From reddit.

i don't think this is correct*. if he had lost his little finger, he would have bled to death in the time between crash and rescue surely? or at least passed out, yet they said they were all conscious. ]

 

EDIT: It appears i forgot to finish my sentence *

[m4DD0gg]

Apparently legs have been lost

 

http://www.telegraph.co.uk/news/uknews/11647707/Alton-Towers-shut-amid-investigation-into-rollercoaster-crash-as-victim-loses-leg.html