Fury 325 Support Failure

[super7]

The fact that this ride operated unsafely for nearly a week with this structural failure shows how much tide safety is being neglected so Cedar Fair can cut costs.  
 

Carowinds has chased off qualified constantly cutting their hours at the last minute with early closures. now this cost cutting has translated to less safe rides.    This is exactly how the big thunder mountain tragedy happened at Disneyland.  Cost cutting    This all falls in the VP

[silver2005]

To me, I think its just B&M maybe miscalculated the forces at that particular spot.  That's one of those lower high speed turns, which is a pretty forceful part of the ride.  Add in those trains which are pretty dense with the weight and there you go.  Carowinds should have fault for not finding out as its not like Fury is a secluded ride and its in a spot pretty obvious to spot.   The support doesn't look all that different than other supports in similar areas on Fury, Leviathan, or Orion.   Fury is the more forceful of the 3.   

Good news to me on this- it wasn't catastrophic, no one was injured or worse, and its a relatively easy fix, both in hardware and making adjustments to maintenance.  I do suspect that no one will ride it again until 2024 though. 

[Taylor.B03]

Depending on what information comes out, we may divert our Carowinds plans to go to Kings Dominion instead. Likely will be a long while before it reopens but your never know!

[SonofBaconator]

29 minutes ago, silver2005 said:

Carowinds should have fault for not finding out as its not like Fury is a secluded ride and its in a spot pretty obvious to spot.

Whether someone from the park noticed it or not, Carowinds screwed the pooch by allowing the ride to operate with a cracked support. I wonder if their media team will address this or be quiet about it.

[SonofBaconator]

4 minutes ago, jsus said:

What more do you expect them to say at this point than:

https://www.wcnc.com/article/news/local/carowinds-fury-325-ride-close-local/275-9caaab31-5e8a-421e-a5a6-ecd877304ec5

I retract my statement regarding their media team, I guess I didn’t dig deep enough.

[silver2005]

I feel like I dodged a bullet with this, too, regarding trip plans.  For my next trip, I was 50/50 on making the part after my Virginia portion either Carowinds or the current Pennsylvania plans.  I was also contemplating a SCarowinds trip in October, which is a big nope now. 

[SonofBaconator]

[NegativeGs]

As a structural engineer, I'd say this is a significant concern. The top of the support appears to have completely fractured – likely due to fatigue at the weld joint between the adjoining circular sections. The concern is this connection detail is likely used by B&M in many of their modern coasters. The question is whether Fury 325, Orion, or other modern B&M coasters have similar connections that should be inspected and monitored.

[Kenban]

I keep seeing people assuming B&M or Clermont had to have made a mistake.  If a footer moved, especially if it’s the one supporting the diagonal, the forces imparted to the support would be increased and I believe the failure would look exactly like what we have seen.

The top of the support appears to be pushed outward away from the support, which is the direction it broke in.  My theory is they do a survey and find the diagonals footer sunk into the ground more than expected.

[SonofBaconator]

19 hours ago, NegativeGs said:

The question is whether Fury 325, Orion, or other modern B&M coasters have similar connections that should be inspected and monitored.

Maybe I’m wrong but, I’d like to assume that KI staff have seen the video and pictures and conducted their own inspections of their B&Ms.

[DeltaFlyer]

1 hour ago, Kenban said:

I keep seeing people assuming B&M or Clermont had to have made a mistake.  If a footer moved, especially if it’s the one supporting the diagonal, the forces imparted to the support would be increased and I believe the failure would look exactly like what we have seen.

The top of the support appears to be pushed outward away from the support, which is the direction it broke in.  My theory is they do a survey and find the diagonals footer sunk into the ground more than expected.

Not sure if I have seen anyone else here mention it, but Orion had some large additional footings added last off season. Take a look next time you ride, you can see the different shades of concrete where they were added. 

[Orion742]

32 minutes ago, DeltaFlyer said:

Not sure if I have seen anyone else here mention it, but Orion had some large additional footings added last off season. Take a look next time you ride, you can see the different shades of concrete where they were added. 

Do you know specifically where they added them?

[disco2000]

Diamondback got that treatment in recent years as well.

[IndyGuy4KI]

Looks like inspecting of track is underway. Photo from Carowinds fans only.

https://twitter.com/CarowindsFansOn/status/1675603541866545159?t=bzXhKwPTn1jglFepjul9dA&s=19

[disco2000]

^they are earning their pay with that job.  Can you imagine the difficulty of that having to constantly hook and unhook and the angles.

I am guessing they felt like it was unsafe to use the inspection crawl car.

I suspect this incident will probably increase the frequency use of said car.

[Tyler2012]

On 7/1/2023 at 12:02 AM, jsus said:

Some supports are braced behind the track where the train would be pushing it.  Some are in front of the track in the opposite direction of the imparted forces.  Would really like to hear from a structural engineer as to what they were thinking designing this.  As someone who isn't a structural engineer, surely the only way to fix this is to redesign the entire support structure of the ride to actually brace against the imparted forces and not in the complete opposite (wrong) direction...

Structural (though non-roller coaster) engineer here. There are many factors that dictate support placement. For instance, existing structures and/or other interferences above ground (like TR:TR’s entrance cave as shown in the picture of Diamondback that was posted earlier), soil conditions underground that determine where footings can go, structural efficiency, aesthetics, etc. Assuming soil conditions are generally uniform in this area of Fury 325 (which may or may not be the case), it appears the only other condition that would determine where footings could be constructed is the location of the parking lot. Supports in this area may have been designed to avoid being built in the parking lot, but structural efficiency may have also played a role. Let me explain: steel members in tension (i.e., pulling) are generally more efficient (i.e., less material is needed to resist the same force) than steel members in compression (i.e., pushing). Think of a rope: you can pull a truck with it, but if you “push“ on a rope, it just crumples; it can’t resist compressive forces at all. The structural engineering equivalent of “crumpling” (more or less) is buckling, or the sudden lateral (i.e. sideways) deflection (i.e., movement or displacement) of a supporting member. Think of a piece of uncooked spaghetti: if you push on the ends, the spaghetti bends, or buckles. The diagonal portion of the Fury 325 support that failed would have been loaded primarily in tension by the passing train, therefore it could be smaller than if it were loaded in compression (in which case the diagonal support would have been placed on opposite side of the vertical support). In this particular situation, using a tension brace or a compression brace isn’t right or wrong. It just depends on several other factors, and I imagine one of them was efficiency. 

Also, as I posted on another site, unless you have access to B&M’s inspection requirements for Fury 325, and proof those requirements were not met, blaming maintenance for this incident is ignorant and irresponsible. That a structural failure like this occurred is not an automatic indication that maintenance was not being done properly. Design, construction, and/or material issues could have contributed to this just as much as anything else. Until (if) a report detailing the cause of the failure is released, no one without detailed knowledge of the ride and its operations should be blaming anyone. Statements regarding what the park or maintenance “should” have been doing are likely uninformed. Engineers can roughly determine how many cycles (i.e., when a structure is loaded and unloaded, or stressed and unstressed) a steel structure can withstand, which can be used to determine how often a component should be inspected. Keep in mind, most of the bridges in the US have arm’s length inspections only every 12 to 24 months. I don’t see why the structure of a steel roller coaster would be much different.   

[disco2000]

1 hour ago, Tyler2012 said:

Structural (though non-roller coaster) engineer here. There are many factors that dictate support placement. For instance, existing structures and/or other interferences above ground (like TR:TR’s entrance cave as shown in the picture of Diamondback that was posted earlier), soil conditions underground that determine where footings can go, structural efficiency, aesthetics, etc. Assuming soil conditions are generally uniform in this area of Fury 325 (which may or may not be the case), it appears the only other condition that would determine where footings could be constructed is the location of the parking lot. Supports in this area may have been designed to avoid being built in the parking lot, but structural efficiency may have also played a role. Let me explain: steel members in tension (i.e., pulling) are generally more efficient (i.e., less material is needed to resist the same force) than steel members in compression (i.e., pushing). Think of a rope: you can pull a truck with it, but if you “push“ on a rope, it just crumples; it can’t resist compressive forces at all. The structural engineering equivalent of “crumpling” (more or less) is buckling, or the sudden lateral (i.e. sideways) deflection (i.e., movement or displacement) of a supporting member. Think of a piece of uncooked spaghetti: if you push on the ends, the spaghetti bends, or buckles. The diagonal portion of the Fury 325 support that failed would have been loaded primarily in tension by the passing train, therefore it could be smaller than if it were loaded in compression (in which case the diagonal support would have been placed on opposite side of the vertical support). In this particular situation, using a tension brace or a compression brace isn’t right or wrong. It just depends on several other factors, and I imagine one of them was efficiency. 

Also, as I posted on another site, unless you have access to B&M’s inspection requirements for Fury 325, and proof those requirements were not met, blaming maintenance for this incident is ignorant and irresponsible. That a structural failure like this occurred is not an automatic indication that maintenance was not being done properly. Design, construction, and/or material issues could have contributed to this just as much as anything else. Until (if) a report detailing the cause of the failure is released, no one without detailed knowledge of the ride and its operations should be blaming anyone. Statements regarding what the park or maintenance “should” have been doing are likely uninformed. Engineers can roughly determine how many cycles (i.e., when a structure is loaded and unloaded, or stressed and unstressed) a steel structure can withstand, which can be used to determine how often a component should be inspected. Keep in mind, most of the bridges in the US have arm’s length inspections only every 12 to 24 months. I don’t see why the structure of a steel roller coaster would be much different.   

Excellent 1st post here!

Reading what you said in layman's terms - steel generally works best in tension and that design, while it may look awkward to the armchair engineer, is probably (and properly) designed with that in mind.  That diagonal member would be the tension component.

Based on your experience and what can be seen so far, would the weld at that location be a likely candidate to be looking at to see if it were poor or the material wasn't up to strength, or could running the coaster for extended periods near the minimum operating temperature had contributed to brittleness in the weld resulting in failure?

[Tyler2012]

1 hour ago, disco2000 said:

Excellent 1st post here!

Reading what you said in layman's terms - steel generally works best in tension and that design, while it may look awkward to the armchair engineer, is probably (and properly) designed with that in mind.  That diagonal member would be the tension component.

Based on your experience and what can be seen so far, would the weld at that location be a likely candidate to be looking at to see if it were poor or the material wasn't up to strength, or could running the coaster for extended periods near the minimum operating temperature had contributed to brittleness in the weld resulting in failure?

Thanks!  

I can't say the support was properly designed (or that it wasn't for that matter, even with the failure), but your summary is basically correct. The diagonal member would be loaded primarily in tension/pulling. I wouldn't say steel works "best" in tension, though, just that if a designer has the option of resisting a load with a tension or compression member, the tension member will likely be more efficient (but that doesn't mean better).

I could speculate what may have happened, but it would only be a random guess that could be right, or very, very wrong. Design defects, construction defects (e.g., weld installation issues, foundation settlement), material defects (i.e., weld material and/or base metal (i.e., the steel being welded together)), failure due to residual stresses (i.e., stresses left over from manufacturing the steel and/or welding it together), or all these in some combination could have contributed to the failure. But yes, a metallurgist will almost certainly be hired to look at the weld(s?) there.

Steel absolutely behaves in a more brittle manner at lower temperatures (meaning it tends to crack instead of stretch the colder it gets), but I doubt that contributed much to the failure. It certainly could have, but even if Carowinds operates the ride year-round (and I don't know if they do or not), I don't think the temperatures are so low that special considerations would be needed in this situation. I didn't deal with dynamic structures though (i.e., structures where loads are applied quickly), so I could be way off base. 

Edited July 3, 2023 by Tyler2012
Spelling

[Tyler2012]

57 minutes ago, jsus said:

Quote

True, it's entirely possible (and likely) that the "wrong-way" bracing is under tension, it just doesn't make itself immediately apparent.  Post-tensioning of steel cables inside of a concrete structure makes sense and is incredibly common.  But when watching roller coaster construction, there's nothing that makes it apparent to the public watching that any tension is indeed placed on any of the support columns.  It generally appears that a roller coaster structure is simply constructed at rest, essentially, that it's neither particularly under tension or compression.  It's just neutral.

I'm not sure I understand the reference to post-tensioned concrete, but if I'm reading the rest of this correctly, I think you're saying there isn't much load in the diagonal member when there isn't a train running on that portion of the track? If so, I agree. The tension load is caused by the dynamic forces of the train moving along the track, and the diagonal member probably doesn't have much to do when the coaster isn't operating. That being said, from an engineering perspective, that member is very clearly a tension member. That isn't to say it is not also loaded in compression at some point, but I'd be willing to guess the largest force it sees by far is a tensile one. This may not be immediately apparent to the public (there's no reason it would be), but I think most people could understand why with a quick 5-minute lesson.

Quote

So, given that the only logical explanation is that B&M is using tension vs. compression on these supports, there is apparently more to the construction process than many of us have historically been made aware.  Wouldn't there have to be some process of putting tension on the diagonal column in order to make this work?  That would've had to have happened during construction back in 2014-15.  For that matter, when I hear of steel under tension, it's usually steel cables.  How does it work when it's a hollow cylinder (or one sometimes filled with sand for sound deadening)?

The diagonal member will be loaded in tension when the train passes just by virtue of the support and track configuration. As shown in the video where the train rides over the failed support, the top-most portion of the support moves to the right since it is no longer attached to the rest of the support. If not for the separation, the diagonal member would resist that rightward movement by pulling the vertical member back into place like a rope or chain, therefore the diagonal member is in tension in that situation. The vertical member, however, is likely modeled as a compression (i.e., pushing) and/or flexural (i.e., bending) member, though. The diagonal member goes into tension when loaded by the train, but tension doesn't have to be put into the support during construction. Referring to your first paragraph, post-tensioning strands (and prestressing strands, for that matter), are installed in concrete for the sake of the concrete, not the steel. Steel does not have to be prestressed or post-tensioned to effectively carry load (most of the time...nor does traditionally reinforced concrete that uses only rebar). The sand just adds extra dead load the diagonal member has to deal with when there is no train- it wouldn't have much of an effect on the member's tensile capacity. But steel, whether a box, tube, angle, wide flange (i.e., an I-shaped beam or column), strand, wire, or cable, can be put into tension by any pulling force applied to the member. The shape doesn't matter as long as the member can be pulled. 

Quote

 

Very good point regarding bridge inspections.  Of course, structures, such as bridges, that are known to be deficient are typically required to have more frequent inspection intervals.  Bridges see take quite a beating between the weather and the weight of vehicles driving over them.  But we don't have any reason to believe that scheduled structural inspections revealed anything in recent years.

How thorough of a structural inspection should we expect Carowinds' maintenance department to do on a daily, weekly basis?  Even walking the track, which shared pictures indicate is taking place, won't make it easy to inspect the supports behind the track.

For that matter, I would've expected to see them inspect the supports before climbing the track.  Definitely wouldn't want to take the track inspection ball out and have it get stuck or anything.

 

I have no idea how often B&M expects the entire structure to be inspected, but I would be very surprised if it's the "daily" inspection some folks are saying "should" happen. 

We'll see if I figured out how to use the quote feature with interspersed comments. My apologies if not. 

[DoomPlague]

49 minutes ago, Tyler2012 said:

Thanks!  

I can't say the support was properly designed (or that it wasn't for that matter, even with the failure), but your summary is basically correct. The diagonal member would be loaded primarily in tension/pulling. I wouldn't say steel works "best" in tension, though, just that if a designer has the option of resisting a load with a tension or compression member, the tension member will likely be more efficient (but that doesn't mean better).

I could speculate what may have happened, but it would only be a random guess that could be right, or very, very wrong. Design defects, construction defects (e.g., weld installation issues, foundation settlement), material defects (i.e., weld material and/or base metal (i.e., the steel being welded together)), failure due to residual stresses (i.e., stresses left over from manufacturing the steel and/or welding it together), or all these in some combination could have contributed to the failure. But yes, a metallurgist will almost certainly be hired to look at the weld(s?) there.

Steel absolutely behaves in a more brittle manner at lower temperatures (meaning it tends to crack instead or stretch the colder it gets), but I doubt that contributed much to the failure. It certainly could have, but even if Carowinds operates the ride year-round (and I don't know if they do or not), I don't think the temperatures are so low that special considerations would be needed in this situation. I didn't deal with dynamic structures though (i.e., structures where loads are applied quickly), so I could be way off base. 

Fury was ran during Winterfest in 2022, I know that much. Though manufacturers do give parks a minimum operating temperature for their rides.

 

One thing I wonder (for anyone that might know) is how redundant are the supports on these coasters? There's a clip of a train successfully navigating that curve with the support broken, as the ride obviously had enough supports to keep it in place. Do manufacturers plan for the possibility of a certain number of pieces breaking?

[SonofBaconator]

5 hours ago, Tyler2012 said:

Until (if) a report detailing the cause of the failure is released, no one without detailed knowledge of the ride and its operations should be blaming anyone.

I understand where you’re coming from, but if the crack in question was noticed by park goers a week ago, if I’m reading posts correctly, then the park either didn’t notice it or cleared it to operate with a crack.

Either way it doesn’t seem like a good look on the park’s end regardless of anything else.

[disco2000]

7 minutes ago, SonofBaconator said:

I understand where you’re coming from, but if the crack in question was noticed by park goers a week ago, if I’m reading posts correctly, then the park either didn’t notice it or cleared it to operate with a crack.

Either way it doesn’t seem like a good look on the park’s end regardless of anything else.

From what I read, the person that took the photo that is circulating around mentioned that she did not zoom in to the photo she took a week ago until AFTER the break happened.  She has been posting and making it clear, but nobody is listening because sensationalism sells LOL, that she did not know about the crack prior to the video of the full break.

And when you see the original photo compared to the zoom, that was a lot of digital zoom so there is no guarantee that it was a crack. Maybe it was or maybe it was the lighting and shadows at that time giving the appearance of a crack.

[robintodd]

It surpises me that they don't already use drones to inspect tracks on roller coasters.  Seems like they should be able to get a scan of the entire structure, weld points, joints, etc. and after getting one visual confirmation of a "good scan" it can then be compaired frame by frame to new scans and possible anomilies be identified.  I would think 2-4 drones per park could be used to scan coasters on a weekly rotation.  Finally it would also seem that putting a camera on the front or rear car would also provide a video feed that could be accessed and assessed during the day every hour or two.  Hopefully these technologies will be implimented sooner rather than later to increase the ability to finid/inspect anamolies on coasters.  

[Stoan]

Someone else mentioned using drones for inspections, which I agree with. I am surprised more parks haven't moved towards having a pre-programed drone flight path with a camera literally just fly the entirety of the track, giving a 10 minute video each day to the maintenance team to inspect.

 

If you wanna get SUPER fancy with it, add in some kind of IR/QR trackers on the track itself, use that as "checkpoints", then run through a AI algorithm to give an early alert system where it goes through every single daily video at those points and compares to see if anything at all has changed. Color differences, size differences, angles or shapes could all be assessed on the daily automatically while the engineering team does their walkthroughs and video inspections. I don't fault the maintenance team for possibly missing this IF it happened suddenly, who is going to spot that crack from all the way on the ground. But I think this crack probably started forming a while ago and it should have been spotted.

 

That said, I do think a programed drone would be the smart way to go. Add a LZ just outside the entrance walkway, incorporate it into the theming even, just to give a nice reminder to "hey, we are taking this safety thing serious". Heck add a little TV and show that day's video footage on a loop so people can see the whole track from a new perspective.

[SonofBaconator]

 

[Taylor.B03]

4 hours ago, Stoan said:

Someone else mentioned using drones for inspections, which I agree with. I am surprised more parks haven't moved towards having a pre-programed drone flight path with a camera literally just fly the entirety of the track, giving a 10 minute video each day to the maintenance team to inspect.

 

If you wanna get SUPER fancy with it, add in some kind of IR/QR trackers on the track itself, use that as "checkpoints", then run through a AI algorithm to give an early alert system where it goes through every single daily video at those points and compares to see if anything at all has changed. Color differences, size differences, angles or shapes could all be assessed on the daily automatically while the engineering team does their walkthroughs and video inspections. I don't fault the maintenance team for possibly missing this IF it happened suddenly, who is going to spot that crack from all the way on the ground. But I think this crack probably started forming a while ago and it should have been spotted.

 

That said, I do think a programed drone would be the smart way to go. Add a LZ just outside the entrance walkway, incorporate it into the theming even, just to give a nice reminder to "hey, we are taking this safety thing serious". Heck add a little TV and show that day's video footage on a loop so people can see the whole track from a new perspective.

The only downside to drones would be rain. If it is raining, drones are harder to operate

[Captain Nemo]

34 minutes ago, Taylor.B03 said:

The only downside to drones would be rain. If it is raining, drones are harder to operate

Thankfully rain is something you can prepare for.

[Gordon Bombay]

Looks like Leviathan is closed too pending Fury’s investigation. Orion didn’t get closed down cause it’s not a giga. 
 

* rimshot *

[disco2000]

22 minutes ago, Gordon Bombay said:

Looks like Leviathan is closed too pending Fury’s investigation. Orion didn’t get closed down cause it’s not a giga. 
 

* rimshot *

You had me going for a second about Levi lol

[SonofBaconator]

According to another video, the guy who recorded the video called 911: