DTMiller

Mostly a testament to the power of extremely muddy off track surfaces + lack of talent + speed + sideways + luck both good and bad.

OP, don't go halfway, safety systems are a system and once you change one component it requires changing the whole system.

docwyte

As an instructor I won't get in a car with a harness bar/harnesses. All the way or none of the way...

mglobe

If you search the literature, I think you'll find that there is little to no correlation between the amount of roof crush, and the severity of injury in belted passengers. It looks like the majority of spinal injuries come from the driver and/or passenger hitting the roof due to the amount of mobility afforded from three point belts. A cage is optimal, but I don't think it's been proven that harnesses with a harness bar are inherently unsafe. Personally, I will take harnesses over three-points all day long, regardless of the presents of a roll bar.

hf1

With 3pt belts and standard seats, there is probably little to no correlation between roof crush and the severity of injury, as the body would either be shifted around or pushed down with the (collapsed) seats.

It's a completely different story when the roof crushes onto a head/neck on top of a body strapped with a 6pt harness into a rigid, bolstered race seat. When the roof comes down, the body and seat have nowhere to go. It's simple physics to conclude the consequences.

Pls link to literature that shows this not to be a major concern. I don't think it exists but even if it did, I'd still go with common sense and either use (1) 3pt belts with standard seats and no roll bar or (2) full 6pt harness, bucket seats, roll bar and HANS. Everything in between would be too risky of the worst (catastrophic) kind.

Astroman

+1
I didn't realize there were so many crash test engineers on this forum.
Plus, I would add, many "roll bars" I see in DE cars are useless crap. Almost all of them are bolted in to less than ideal locations.

CosmosMpower

Scary stuff, for street cars most guys just run a back half cage. It doesn't look like a back half cage would have helped much in this scenario as the A pillar collapsed.

Friend of mine was instructing in a S2000 and the car rolled and the A-pillar partially collapsed, he was wearing a 3 point seat belt and walked away. The driver went to the hospital but lived.

mglobe

Too tired to ask the google?

http://citeseerx.ist.psu.edu/viewdoc...=rep1&type=pdf

There's no doubt scenarios that could happen where being harnessed in with roof collapse could cause serious harm. You have to ask yourself though if you are increasing or decreasing the overall risk of being injured when you install safety equipment. In the case of a rollover, as I mentioned previously, you are much more likely to be hurt by bouncing (for lack of a better term) around in the car in your three points than if you are in harnesses even without a roll bar.

In my opinion, when you risk weight the potentials for harm, harness + cage> harness + harness bar >> three point belts. I'd much rather have a cage or roll bar. But to say harnesses with a harness bar is more dangerous than three point belts just does not make sense to me.

hf1

It's a paper from 1976:
"Abstract--Rollover crashes involving 997 automobile occupants of the front outboard seating positions
were reviewed. Injury levels, using the Abbreviated Injury Scale, were studied for their association with
ejection, restraint use, roof crush, car weight, and the absence or. presence of the B-pillar.
The results indicate that the more serious injuries and fatalities are sustained by those ejected from the
car, with ejection occurring most often through the side glass area. Restraints are related to less severe
injuries to the non-ejected occupant as well as to a reduced rate of ejection. Roof crush is found to be a
factor that interacts with restraint use and with the vehicle factors."

As expected, talks about injuries from ejections presumably from occupants not wearing seat belts and says nothing about risk comparison of 3pt + standard seats vs 6pt + race seats during a roof crash.

That's not what the linked paper was talking about. The bouncing around they studied happened to occupants that wore NO seat belts, in cars with 70's safety tech, to boot. So where did you get this likelihood estimation?

There are two types of risks: risk of injury and risk of death (or very serious injury). Harness + rollbar/cage lowers BOTH risks vs the other two alternatives, we all agree. Harness + harness bar may have a lower risk of injury (evidence?) vs 3pt + standard seats + airbags (OEM design satisfying increasingly rigorous govt crash standards) but my gut tells me that it definitely carries a higher risk of death for the reasons already explained. Many racers have told me that their gut feeling is exactly the same. I probably first heard it from them, to start.

Hopefully these discussions are helpful towards everyone making their own decisions.

Smokey the Talon

Newer info on harnesses and rollovers from a company most of us trust with our primary safety equipment.

https://www.schrothracing.com/docs/TechTactics.pdf

Starting on page 32 - "Aftermarket harnesses and racing seats ARE safer in the majority of incidents in vehicles with OR WITHOUT a rollbar – INCLUDING ROLLOVERS"

I'd be much more concerned with making sure you're using a proper seat, have proper belt routing/angles and mounting points. Those will play a larger role in your safety for the majority of incidents.

Dr911

Agreed! I wished I'd done it 18 months before I actually did it.
One idea to consider that I wish somebody had suggested to me: spend some time every weekend "breaking-in" the new harness. It does seem to make a differemce.

Terry L

Take a look at the location of the roll bar - it's well behind the B pillar and can't prevent the collapse of the A pillar. I'm no engineer but the B pillar seems far more robust, at least in Porsches, and that is the area where the passengers' heads are located. I have gone to a fully caged (and therefore not streetable) car, but I'd rather see a harness bar with harnesses and a proper seats than stock seats with three point belts and air bags. I hope that I never have to test this hypothesis.

white924s

Perhaps its splitting hairs, but one more thing to consider is the impact of harnesses on the probability of a serious crash. I think most would agree that properly installed/positioned harnesses will help the driver better control the car and be less fatigued - potentially reducing the risk of an incident in the first place!

One more piece of food for thought - in rollover scenarios, an occupant could be injured either a) if they flail into something solid or b) if some part of the roof structure is impacted inwards into them. I believe that most would agree that a proper harness system reduces the risk of a). I think perhaps an unintuitive way that a good harness bar can help in b) is by transferring some of the force to the pillar to the rest of the chassis (reducing deflection of the roof - obviously not as much as a proper roll bar, but as was already mentioned here - many roll bars aren't designed in a way to bear significant loads). So net/net, it seems to me that in the vast majority of scenarios, a harness bar/harness system is a net positive, though like all safety systems (except for maybe a full cage/containment seat/nets), it is a compromise.

All this makes me want to go out and get harnesses for my current car - if only there was a harness system that worked with stock Porsche seats. My old car had harnesses, Recaro Profi seats, and a harness bar, but that's not in the cards for now - I am thinking about getting a Simpson Hybrid Head & Neck system though.

Hella-Buggin'

Bought it used. Just out of spec from a Spec Boxster Club car.
They have to replace every two years I think but for DE they are fine for
a few more years.

mglobe

+ a billion

There's a lot of conventional wisdom that goes on within our hobby that is often based on what others have said but is not founded in real data. I'm guilty of it too. Here's some light reading.

http://acurazine.com/forums/car-talk...t-cars-378607/

http://www.transport-safety.com/docs...ionSystems.pdf

http://www.upload.hayesassoc.com/ROY...0al%201999.pdf

PJorgen

Thanks for posting this link, very informative. A couple of other interesting points from that Schroth document:

1. The greatest risk of serious injury in a rollover is head contact with the door frame or other things outside the vehicle. A proper 4+ point system will greatly reduce the potential for such contact.

2. Serious injuries in rollovers are very seldom due to the collapse of the roof on the occupants.

And finally:

Rollovers result in loads of 10g to 35g deceleration to the body eliminating the possibility of controlled “escape” by an occupant to “duck” from the collapsing roof.

I've always been skeptical of the "never use a harness without a roll bar/cage" argument. This information does not appear to support that argument.