Technical comments only: car to car crash safety

[roon]

With respect to the recent sad news, please, in this thread:
--don't mention your sporting preferences
--no blaming
--no fault finding or inquiries
--no personal comments
--debate only upon physics

-----

With racing cars as in road cars, there are some types of impacts which are truly hard to safen. High speed impacts involving two vehicles at odd angles are truly a huge engineering challenge, and no concessions are given by physics in the situation.

Can the cars be made to survive, let alone safely dissipate, such energies?

My initial, vague suggestions are:
-how the driver is braced by safety hardware
-methods/technologies for car-based absorbers/deflectors
-yet-to-be-invented trackside devices which can intervein
-gps/transponder/optics based locating & control software

To the lattermost point: if software can immediately identify problem situations based on said sensors and data, an automated caution mode could be enabled upon all cars instantly, decelerating and rerouting all cars in concert. Robust automated/driverless car tech could be drawn upon. Put simply: criteria would engage an immediate field-wide driverless safety mode.

In lieu of a software-based approach, physical devices on the car might be considered. The first thing on my mind are some sort of large, durable airbag, like what one of the Mars rovers used. There would be an obvious mass and packaging issue accompanying that approach. If both cars use them, airbag-to-airbag impacts may avoid puncturing while providing a crucial extra ~meter of deceleration.

Trackside devices could take a variety of forms, in concept. I think of destroyable foam or cellular blocks or air cushions of sufficent mass to slow a car that drives through them. For example, if a bunch of them were held aloft above run off areas, they could be dropped upon a stationary car, or in front of an uncontrolled car. The car then crashes into this rain of absorbers.

Or, synthetic netting stretched upon run off areas which the cars can snag onto to slow them down. Stretching or tearing at a controlled rate. Something like a tow hook could deploy to engage this ground-net. If the net is not sacrificial, the tow hook could be attached to a spool of tether, with braking achieved by a braking force applied to the spool. Still a violent decel, but less violent should be the goal.

Further to that point: a pyrotechnic anchor could fire from the car down into a controlled-compound runoff surface. A spool of tether with a brake would be attached to this anchor and stop the car a short distance from the anchor. Decelerate strongly yet safely. Perhaps 10-20g. Whatever can be deemed a safe limit.

[NL_Fer]

Electronics are to slow, can give false alarms and sudden loss of power can be dangerous in a fast corner.

But barriers which can prevent a crashed car to bounce back to the track, that would be something.

[Jolle]

Good to split the items indeed.

Looking at the spa crash, on first view, the second impact, with another driver, was the fatal one, happening in the run off area. This brings us to a problem, for a one sided crash you need indeed good materials to slow the car down at the right rate for that corner but that would mean that in high speed corners the speed differential with other cars is massive. At 270km/h any runoff is short and if one car hits that barrier there is a risk that that same incident will cause another car to hit the first after the barrier did its work on car number one. We see this in MotoGP as well, the circuits are pretty safe but now all real scary and fatal crashes are when drivers are hit by other bikes.

This might be one of the vulnerabilities that are very difficult to address. A (relative) stationery car effectively transforms in a piece of barrier for oncoming traffic.

[izzy]

It's worth differentiating between crash prevention and crashworthiness, as they are quite different aspects of it. Anyway one thing is that the crash tests are pretty modest: the front is only 33 mph, side 22 and rear 25.

Also they don't line up, someone thought Billy Monger's legs were broken by the rear crash structure of the car he hit.

So good thread, I'm sure there's more they can do. Halo is made to a much higher standard, like 116kN.

[Jolle]

There have been a couple of crashes where one car his another at high speed, while one was (relative) stationary. Zanardi, Monger and now Huber. With Monger, it seems that the jack lifter, that was never used in crash tests, worked as a can opener, ripping trough the carefully designed front end. With Zanardi there is a theory that him loosing his legs saved him from death. Because the front of the car shattered at impact, the g-forces on the rest of his body were within limits.

To prevent big injuries you could put it into an equation. Take the worst case scenario, a car t-boned at max velocity, without the engine (extra weight) attached, how many cm of deceleration/time do you need to stay within safe limits? I think the amount of foam needed would make a car underivable.

[Just_a_fan]

In the case of side impacts, the issue is that it's difficult to build sufficient energy absorption in to the side of the car. The nose and rear structures give relatively gentle impacts, being hit from the side means the driver suffers massive lateral acceleration. One could, I suppose, mandate that the side pod extends along the side of the cockpit and design them to be impact structures. That would certainly help. Having said that, the front of the other car at Spa was gone following the impact. The driver's feet were visible in the end of the safety cell meaning the entire energy absorption ability of the nose cone had been used up and the front of the safety cell - just about the strongest part of the car - hit the side of Huber's car.

At Spa, the car was split in two which might suggest that the impact was somewhere around the joint between the tub and engine. The tub then went through a series of motions, all of which would have been bad for the driver's brain stem.

Sometimes, the energy is too high for any system to deal with it in such a small vehicle as a single seat racing car. Spa was a "worst case scenario" in that regard as the impact was a full speed car hitting, effectively, a stationary one in the worst possible way.

[Edax]

Unfortunately I am afraid that the thing that can be learned from this tragedy is that freak events happen. It was a collision outside of the track by two cars which lost control in a very short succession, very difficult to prevent.

In terms of mitigation, when a 750 kg car hits a stationary object at 250 km/hr, you have to dissipate ~2 Mj of kinetic energy. The only way to get around that is by deflecting the cars with an as shallow angle as possible, to minimize energy transfer. I think the only practical way is by sending one of the cars flying. Which construction wise is certainly possible but presents a host of other problems. It is Ironic that we have seen several accidents that perhaps would have been less severe with a high nose. But the risk of a car flying into the stands is simply too big.

As for trackside improvements. I like the netting idea, and I think it is doable, and could prevent cars from bouncing back to the track like in Brasil (alonso accident). But in this case the accident happened on the runoff area, so one could argue that the barrier did its job.

A crash avoidance system, similar to ACARS in airplanes could be a solution. But implementation would IMO be very difficult in a car that is supposed to almost crash during the whole race. But auto drying is taking huge strides. And fully automated racing (roborace) could perhaps in a few years deliver the logic that is needed to take over the car in an out of control situation.

I think we will see that one day. Leave the track and the computer takes over.

[Zarathustra]

Did the Halo block the view going up eau rouge?

[izzy]

Unfortunately I am afraid that the thing that can be learned from this tragedy is that freak events happen. It was a collision outside of the track by two cars which lost control in a very short succession, very difficult to prevent.

in safety there's no such thing as a 'freak accident', and anyway there's nothing surprising about two racing cars colliding, with one of them stationary side-on after hitting the barrier, especially just there. It's really predictable.

There are always things you can do, improvements, and just watch now: there will be a big effort, just like secretly Halo was made strong enough to have saved Jules Bianchi, or after Senna.

It's a tragedy but there will be energy from it, is the one thing to cling on to.

[MtthsMlw]

Did the Halo block the view going up eau rouge?
http://formule1nieuws.nl/uploads/img6ab1dfccfc4e.jpg

Not really. With two eyes you get two perspectives (3D view to simplify it). You're kinda looking through that middle post of the halo.

[izzy]

In the case of side impacts, the issue is that it's difficult to build sufficient energy absorption in to the side of the car. The nose and rear structures give relatively gentle impacts, being hit from the side means the driver suffers massive lateral acceleration. One could, I suppose, mandate that the side pod extends along the side of the cockpit and design them to be impact structures. That would certainly help. Having said that, the front of the other car at Spa was gone following the impact. The driver's feet were visible in the end of the safety cell meaning the entire energy absorption ability of the nose cone had been used up and the front of the safety cell - just about the strongest part of the car - hit the side of Huber's car.

At Spa, the car was split in two which might suggest that the impact was somewhere around the joint between the tub and engine. The tub then went through a series of motions, all of which would have been bad for the driver's brain stem.

Sometimes, the energy is too high for any system to deal with it in such a small vehicle as a single seat racing car. Spa was a "worst case scenario" in that regard as the impact was a full speed car hitting, effectively, a stationary one in the worst possible way.

I think there are things they can do tho. Your point about the sidepods is a good one, there's space there and that as you say is what you need to spread the acceleration out over time. They can raise the nose high enough to match the other safety structures and make the nose absorb more energy. Probably in this crash the nose crushed early on and then it was tub on tub.

Car-to-car has been more or less ignored up to now, look at the situation when a car ends up pointing the wrong way so they collide head on - Schumi and Liuzzi (?) at Abu Dhabi! Halo is there now but even so, the nose isn't designed at all for that and not for hitting the rear or side either. So i think there is lots of scope, when they start looking.

[Jolle]

Did the Halo block the view going up eau rouge?
http://formule1nieuws.nl/uploads/img6ab1dfccfc4e.jpg

The top of the halo is out of the field of view of the driver. The pic you shared isn't what the divers see.

[Big Tea]

My only thoughts on it, and triggering correctly would be all but impossible, is airbags. Not as in a passenger car, that inflate to hold the driver, but somewhat more rubbed and would stay inflated for 'seconds' and surround the driver cell. It would decelerate objects and spread out the impact, using much of it to 'vent' the gas.

As I said, triggering would be very difficult and could cause as much problem as not having it.

(Not truly representative, but a short vid of what I was thinking here with a landing https://www.youtube.com/watch?v=y0FSeF2jNuI at about 2 min)

[Jolle]

My only thoughts on it, and triggering correctly would be all but impossible, is airbags. Not as in a passenger car, that inflate to hold the driver, but somewhat more rubbed and would stay inflated for 'seconds' and surround the driver cell. It would decelerate objects and spread out the impact, using much of it to 'vent' the gas.

As I said, triggering would be very difficult and could cause as much problem as not having it.

(Not truly representative, but a short vid of what I was thinking here with a landing https://www.youtube.com/watch?v=y0FSeF2jNuI at about 2 min)

The speeds make it quite impossible. Now we have a crash as "just" 270, while the aim is to prevent a worst case scenario, at 300+, bounding off the last corner in Brazil. The rate of deployment and size of the bubble needed and the forces that come with such an explosion, might be almost as bad as the crash itself.

[Big Tea]

My only thoughts on it, and triggering correctly would be all but impossible, is airbags. Not as in a passenger car, that inflate to hold the driver, but somewhat more rubbed and would stay inflated for 'seconds' and surround the driver cell. It would decelerate objects and spread out the impact, using much of it to 'vent' the gas.

As I said, triggering would be very difficult and could cause as much problem as not having it.

(Not truly representative, but a short vid of what I was thinking here with a landing https://www.youtube.com/watch?v=y0FSeF2jNuI at about 2 min)

The speeds make it quite impossible. Now we have a crash as "just" 270, while the aim is to prevent a worst case scenario, at 300+, bounding off the last corner in Brazil. The rate of deployment and size of the bubble needed and the forces that come with such an explosion, might be almost as bad as the crash itself.

Agreed. But I was not thinking of any part to be in contact with the driver, more sort of as a 'bumper' around the outside. I also realise that inproper triggering could launch the car etc.