Jump to content


Photo
* - - - - 1 votes

Fatal Anthoine Hubert crash Spa 2019


  • Please log in to reply
1169 replies to this topic

#1151 Kalmake

Kalmake
  • Member

  • 3,611 posts
  • Joined: November 07

Posted 11 February 2020 - 09:09

Those are panels on top of the inner structure.



Advertisement

#1152 A3

A3
  • Member

  • 29,601 posts
  • Joined: February 00

Posted 11 February 2020 - 09:23

Then why did the monocoque break at exact those points?


Edited by A3, 11 February 2020 - 09:24.


#1153 Kalmake

Kalmake
  • Member

  • 3,611 posts
  • Joined: November 07

Posted 11 February 2020 - 10:15

Maybe because of the access holes on top. Maybe because there is a seam in the inner structure as well. Can't know based on those pictures.

 

From what I can make out in video, the brake is more of a crack that clean separation.


Edited by Kalmake, 11 February 2020 - 10:17.


#1154 Nemo1965

Nemo1965
  • Member

  • 5,626 posts
  • Joined: October 12

Posted 11 February 2020 - 10:19

Putting on brakes no, but would cutting engine power cause problems? Serious question, I'm not a physics PhD!

 

I am not a physics engineer either, but cutting the engine on any car on high speed in any corner is NOT a good idea, especially if you over a hump. 



#1155 ANF

ANF
  • Member

  • 10,871 posts
  • Joined: April 12

Posted 11 February 2020 - 11:02

Maybe because of the access holes on top. Maybe because there is a seam in the inner structure as well. Can't know based on those pictures.

I think the visible seams seem to be on the outer shell in this picture.

0879536.jpg
https://www.motorspo...en-250000-euro/



#1156 cpbell

cpbell
  • Member

  • 2,275 posts
  • Joined: December 07

Posted 11 February 2020 - 11:38

Honest question, why would you lose control if the engine power cut? I can clearly understand why you would if the brakes were put on but why does cutting engine power cause a car to slide? The downforce at that moment isn't being affected by cutting engine power as that's largely (and almost entirely) due to air speed over the car, the tyres aren't going to suddenly lose their adhesive properties, the only change is the engine not turning over as much. Are you assuming that if the throttle is suddenly 'pulled' it'd be like the brakes being applied due to engine braking? If so surely that can be taken into account by reducing the power more slowly. However, it's all overly complicated even if it could be done. This was an absolute freak accident and even with the best will and technology in the world, I'd struggle to see how it could have been avoided while maintaining anything that looks or behaves like motorsport.

Combination of engine braking effect and weight transfer forwards.



#1157 cpbell

cpbell
  • Member

  • 2,275 posts
  • Joined: December 07

Posted 11 February 2020 - 11:44

so you'll see a yellow flag for every snap oversteer they correct?

Maybe even a safety car if I understand his proposal correctly.



#1158 cpbell

cpbell
  • Member

  • 2,275 posts
  • Joined: December 07

Posted 11 February 2020 - 11:49

FFS, I've tried to avoid seeing photo or footage from the crash, and then, as I'm scrolling through the thread, there's a large photo of it.  If you must include such images, please use a spoiler so that only those who wish to see them need to.  I only saw it for half a second before I scrolled off it, but that was still half a second too long.



#1159 A3

A3
  • Member

  • 29,601 posts
  • Joined: February 00

Posted 11 February 2020 - 11:55

FFS, I've tried to avoid seeing photo or footage from the crash, and then, as I'm scrolling through the thread, there's a large photo of it.  If you must include such images, please use a spoiler so that only those who wish to see them need to.  I only saw it for half a second before I scrolled off it, but that was still half a second too long.

Fixed.



Advertisement

#1160 absinthedude

absinthedude
  • Member

  • 1,016 posts
  • Joined: June 18

Posted 11 February 2020 - 20:14

The "survival cell" monocoque will be the part from behind the driver's head (the roll over hoop) to about where his feet go (the part with the circles). Not the whole car. You're also seeing parts such as aero devices, anti-intrusion panels etc. attached to it. 

 

The first picture above contains almost none of the monocoque. The noes and front section isn't part of the monocoque or survival cell. The bottom photo is mostly survival cell...with lots of other parts attached to it. In terms of the overall car...the monocoque "survival cell" is quite small.

 

The monocoque,  by definition, is one piece. It's not "glued together". It will be attached to other parts of the car for, aft and on the sides. 

 

I've already said my opinion.No survival cell is perfect. The fact that Correa's intruded into Hubert's and Correa's was damaged itself does not mean they're in any way deficient. 



#1161 Sterzo

Sterzo
  • Member

  • 1,703 posts
  • Joined: September 11

Posted 11 February 2020 - 21:32

I've already said my opinion.No survival cell is perfect. The fact that Correa's intruded into Hubert's and Correa's was damaged itself does not mean they're in any way deficient. 

Agreed, and the survival cell is designed to accept some damage. A rigid one would simply transmit all forces to the driver.

 

The most obvious way to incease the driver's safety would be to put more space around him, with deformable structures further away. That's the current approach to road cars, and the reason they're getting bigger and bigger.
 



#1162 cpbell

cpbell
  • Member

  • 2,275 posts
  • Joined: December 07

Posted 14 February 2020 - 12:53

Fixed.

Thanks.



#1163 ElectricBoogie

ElectricBoogie
  • Member

  • 678 posts
  • Joined: March 19

Posted 14 February 2020 - 13:26

Why would a snap of oversteer that can be corrected have to trigger anything? Sensors and PEOPLE are way too advanced to have to be triggered by that.
But a car actuallly (half-)spinning out of an unsighted corner taken at high speed, would you like to know before you get around the corne or just gamble that most involved will live?

How many seconds were there between the moment Alesi knew he'd lost control and Hubert arriving at the scene, let alone Correa? I bet Correa could act on an audio signal and never catch up to Hubert's car. Hubert himself might have slowed just enough to prevent serious harm, let alone the lethal impact.

If Eau Rouge by its nature prevents drivers to take any sort of avoiding or retarding, apart from the insighted aspect, it's going to be hard defending still racing there. I think that corner, any, can be made signficifantly safer with tech largely on board anyway.


Edited by ElectricBoogie, 14 February 2020 - 13:41.


#1164 Dmitriy_Guller

Dmitriy_Guller
  • Member

  • 4,857 posts
  • Joined: July 01

Posted 14 February 2020 - 15:10

For a real world demonstration of power-off oversteer, you can look at 1992 Indy 500 month.  Ironically, two of the most experienced drivers with five F1 championships between them demonstrated it.  Nelson Piquet was driving at full speed in practice during a yellow period, and decided to lift off at the last moment to get into the pits.  That was in the middle of turn 4, and the results for him were much of what Correa is going through right now.  Later on, during the race, Emerson Fittipaldi had seen an accident unfold in front of him, between Jim Crawford and Rick Mears.  He lifted off in the middle of the turn to slow down, and swapped ends and crashed into the wall before even getting to the scene of the accident.


Edited by Dmitriy_Guller, 14 February 2020 - 15:10.


#1165 absinthedude

absinthedude
  • Member

  • 1,016 posts
  • Joined: June 18

Posted 14 February 2020 - 21:34

Agreed, and the survival cell is designed to accept some damage. A rigid one would simply transmit all forces to the driver.

 

The most obvious way to incease the driver's safety would be to put more space around him, with deformable structures further away. That's the current approach to road cars, and the reason they're getting bigger and bigger.
 

 

Indeed, I was explaining to an American who moved to the UK in 2017 why our cars got visibly wider in the mid 2000s, due to that approach to safety. The FIA could perhaps mandate more deformable structure to the side of the driver, potentially mandate large sidepods? But the actual survival cell itself is almost certainly performing as intended and as well as any in the world. 

 

As for this idea of cutting the engine....ludicrous and extremely dangerous. See the examples re the 1992 Indy 500 for a start. But other incidents where an Indycar or F1/F2 car or sports prototype has experienced an unexpected loss of power in the middle of a fast turn. It's a recipe to maim and kill if actually done deliberately. 

 

Spotters might help, if someone had got onto Correa's radio perhaps he could have slowed more. But in reality there's likely no blame here. It is entirely possible for a tragedy to happen and for nobody to be at fault. They're called accidents. 



#1166 ElectricBoogie

ElectricBoogie
  • Member

  • 678 posts
  • Joined: March 19

Posted 15 February 2020 - 21:02

Since the tech to signal to drivers when there's an accident happening up the track is just a matter of deciding to have it...there is some built here. It was deemed unnecessary. But one marshall on track where no cars are driving, at any speed, is reason to stay on FSD some more.
Now that I've posed it, the next time such an accident happens, everyone opposing merely exploring ways to alert drivers better, will have something to think about to themselves.

It's so simple. A driver loses control, the sensor know because the wheels are off the track for longer than that piece of the track would ever allow for. The car is pointed the wrong way. It happened too soon. It happened while the driver tried to prevent it. Beyond a random snap of oversteer.

Very clear signal to upcoming cars, together with yellow flag on dash. Zero time for interpretation or observation. You're slowing the car as hard as you can, safety. Not just avoiding a fastest sector of showing a blip of sub-100% throttle. If you're not in an emergency stop within half a second or even one, you're automatically earmarked for a possible black flag. In any case you get to explain to stewards why you took you sweet time.  When in the simulator, you'd clearly shown to be capable to respond quick enough in the most testing of situations, battling more cars (actual peers on sims), in order to obtain the license to race this series.
Alesi starts spinning, drivers are alerted. Correa doesn't even make it up the hill and finds himself having to keep the car on the brake until a marshall signals he can try to pull away, or how to steer off the track in reverse for him. Hubert was too close to Alesi and unable to take sufficient avoiding action. His car is destroyed, but he survives.

This tech is EASY for the right developers. Road race are predicting accidents to happen by using radar bounced off the road surface under the car in front to notice retardation of the car in front of that one. When the car in the middle cannot possibly avoid a tail ender anymore, the third car already activates anti lock braking and doesn't take part in the carnage that had not yet taken place at the time of the audible signal. 

This is more than 3 years ago now: https://www.business...ional=true&r=US

Heck, any driver behind Alesi hitting the brakes at the crest should already trigger a signal to other driver to STOP. It's not that complicated, people!
Cautions have been thrown for less.


Edited by ElectricBoogie, 15 February 2020 - 21:02.


#1167 ExFlagMan

ExFlagMan
  • Member

  • 4,035 posts
  • Joined: January 10

Posted 16 February 2020 - 15:53

 

 

A few comments on a few of your suggestions.

 

There are mountains of data on sensors in relation to spins and worse.

 

That might be true in F1 but I doubt there is much data from lower formulae - 99+% of race cars have no such sensors.

Applying F1 data to lower formulae seems a sure fire way of creating chaos - the usual software developers GIGO principal applies.

 

 

At one point, certain or all sensors post values outside what's normal for a given section.
For instance, steering input. Let's say in quali, no car even went beyond X degrees each way. Engineers and drivers agree that 1.5X would be a clear sign of trying to get out of a crash.

Now in the race, it reaches 1.5, the signal to upcoming cars is triggered and RC automatically throws waved yellow flags or even VSC or FSC. If no car leaves the track or incurs damage: oof, that was close. Continue.

 

 

Or maybe the driver just went wide onto/over the kerb or hit a patch of oil/gravel and made a quick recovery.

Seems like a sure fire way of causing a server meltdown on this forum - all those interruptions for no apparent reason.

 

 

If no driver ever brakes through a given section and yet it happens, that would be a clear reason to worry. Might be a kangaroo on the track. But even a good hedgehog could be a real danger to drivers and spectators when hit. The first driver who brakes causes all cars behind to brake. The animal gets to leave the track, race continues.

 

Unless the hedgehog rolls into a ball...

 

Or maybe one driver gets a bit too close to the car in front or gets chopped by a defensive manoeuvre and has to brake.

 

 

F1 level engineers and stewards will have zero difficulty identifying sensors to use and which signal thresholds to use. [b]Various scenarios could be programmed.[b] Say, FSC does not get triggered all the time on the steering angle being out of typical window. But if a spin is sensed, that's it, VSC. A big hit of the barreir as a result? No need to wait for RD to decide, that's a FSC right there.

 

What happens when some unprogrammed scenario happens?

I wonder how many F1 level engineeers have spent enough (if any) time trackside observing cars actually having incidents on track to enable them to identify the parameters required to define such a system.

As for stewards knowledge, from my experience of stewards, while many of them will have spent time trackside, that experience may be somewhat out of date.

 

 

It's so simple. A driver loses control, the sensor know because the wheels are off the track for longer than that piece of the track would ever allow for. The car is pointed the wrong way. It happened too soon. It happened while the driver tried to prevent it. Beyond a random snap of oversteer.

 

That list sounds like a recipe for many false positives - the consequences of which would lead to many races being subject to unnecessary interuptions and drivers having no faith in the system.

The time off track would need to be pretty high, given some F1 drivers tendancy to treat run-off areas as being a legitimate part of the track.

 

 

 Very clear signal to upcoming cars, together with yellow flag on dash. Zero time for interpretation or observation. You're slowing the car as hard as you can, safety. Not just avoiding a fastest sector of showing a blip of sub-100% throttle. If you're not in an emergency stop within half a second or even one, you're automatically earmarked for a possible black flag. In any case you get to explain to stewards why you took you sweet time.  When in the simulator, you'd clearly shown to be capable to respond quick enough in the most testing of situations, battling more cars (actual peers on sims), in order to obtain the license to race this series.

 Alesi starts spinning, drivers are alerted. Correa doesn't even make it up the hill and finds himself having to keep the car on the brake until a marshall signals he can try to pull away, or how to steer off the track in reverse for him. Hubert was too close to Alesi and unable to take sufficient avoiding action. His car is destroyed, but he survives.

 

And on any other day - Alesi starts spinning, system alerts drivers but Hubert has already managed to avoid Alesi, Correa has already been slowed down, this triggers the following car to be slowed down even more - repeat until the whole field except Hubert is stationary.

 

What happens next?  I guess all you can do is red flag it, return the cars to the start line wait the prescribed length of time and try again.  I wonder how many GPs would be able to run the required no of laps to count for full points in the 2 hr time limit.

 

OR maybe

 

Alesi starts spinning, system alerts drivers - Hubert is already taking avoiding action but system applies his brakes and his car gets t-boned/launched by the spinning car instead of avoiding it. 

 

 This tech is EASY for the right developers. 

 

Is it that easy - from almost 40 years experience of developing software systems I would suggest that the main cause of many software malfuctions can be put down to 'boundary conditions' - events or combinations of events that the developers had not foreseen.

 

I would suggest that unless those 'right developers' defining the system have spent a considerable time trackside they would probably have little idea of the variety of circumstances that can lead to a driver having an incident and the even greater variety of subsequent events - 99% of which do not result in any significant contact between cars.

 

 

One additional thought.

 

I assume that the developers of the software for the Roborace cars will have factored in the need for crash avoidance - It will be interesting to see how it performs if they ever get round to actually having a race....



#1168 ElectricBoogie

ElectricBoogie
  • Member

  • 678 posts
  • Joined: March 19

Posted 16 February 2020 - 19:06

@ExFlagMan
Good work. Now it's my turn.

You come up with a solution, to anything, and I'll find some reasons why it'll never work. Or might need the slight bit of tweaking.

You don't seem to realize that in tech, there is more than an on/off switch nowadays.
And I'm pretty sure even in lower formula they do data logging of steering'/throttle/brake input. 



#1169 ExFlagMan

ExFlagMan
  • Member

  • 4,035 posts
  • Joined: January 10

Posted 16 February 2020 - 22:27

I would suggest that no matter how much 'fuzzy logic' the tech allows, the ultimate result has to come down to an on/off switch decision as to whether or not to trigger the intervention system.

 

With any proposed new system someone has to ask the awkward question - no matter how much it 'p*****s on the proposers chips'.

 

Some lower formula might have the appropriate data logging systems, but that is a country mile away from them having the ability to analyse the data and to reliably send a signal to any other car and for them to respond.


Edited by ExFlagMan, 16 February 2020 - 22:28.


#1170 noikeee

noikeee
  • Member

  • 16,321 posts
  • Joined: February 06

Posted 16 February 2020 - 22:47

I think sudden automated power cuts to cars would be bad, but surely it's technically possible to a) do it in a controlled, managed way, over a couple of seconds ; and b) automatically warn drivers in their cockpits through a flashing light or an alarm sound or something?

I'm not saying it'll work for sure, just that these solutions could be investigated.