Why did the Mercedes CLR Flip at Le Mans?

Text copyright Michael J. Fuller
Images copyright Frederic Le Floc'h, reproduced from the 1999 ACO Le Mans Annual
1999 Mercedes Benz CLRI get asked this question a lot.  It certainly was a what-the-fuck moment.  It's counter intuitive to see something so large, that you're so used to being contained to only two-dimensions, suddenly decide to take on a third.  Especially so when it's entirely unexpected.   I was at the 1998 Petit Le Mans, working for Downing/Atlanta, and I recall the moment when the replay showed on the TVs in pit lane and there was an audible gasp amongst those watching the Porsche GT-98's flip.

A couple of years ago I wrote a response on 10-10ths (most likely), the inquiry there was if the tires contributed to the CLR's accident.  It was a spurious inquiry then, such as it is now.  But in the process of saying no, I laid down the foundation of what I felt contributed to the incidents.  I cut and pasted my response into a document and over the years I've expanded on it.  But it was just sitting on my hard drive completely forgotten until recently and I figured I'd post it up on line:

In my opinion the the root of the CLR's issues can be traced to its dimensional architecture, lack of total downforce, and ultimately, some very bad luck.

For the record, the CLR was designed to the maximum overall length, 4890 mm.  Its wheelbase was 2670 mm, with a 1080 mm long front overhang and a 1140 mm rear.

Beginning in 1997 the ACO allowed the flat bottom LMP and LMGTPs to run the addition of a rear diffuser to the flat bottom.  The regulations defined the start point of the diffuser, rear wheel centerline, but allowed the trailing point to be defined by the car designer only as long as it ended where ever the bodywork ended; the trailing edge of the rear diffuser was matched to the trailing edge of the bodywork.  And on the CLR, it stuck out further than on any other car, a lot further; in comparison the rear overhang on the GT-One was 990 mm, Audi R8C: 940 mm, Nissan R391: 880 mm vs. the CLR’s 1140 mm.  

At the front, the CLR's 1080 mm front overhang was certainly comparable to its contemporaries, if hedging towards the long side.  Underneath, was a very small and comparatively muted front diffuser (certainly compared to the Toyota GT-One or even the BMW LMR).

In the middle of all this was a 2670 mm wheelbase.  Le Mans cars are typically of the long wheelbase variety in that it makes for the most stable aero platform.  But the CLR's wheelbase was the shortest of the entire LMP category.  Compare that to the 2850 mm of the Toyota GT-One, the Audi R8C's 2700, or the BMW LMR's 2790.  And none of the CLR’s competitors had overhangs, front and rear, as long.

You can imagine this architecture led to a pretty sensitive aero platform.  A small change in platform attitude due to that narrow wheelbase say in braking, accelerating, leads to a large change in ride height at the front or rear across that very long lever (overhang).  It was reported the CLRs were seen porpoising at various sections of the track throughout the weekend, this certainly confirms an underlying aero platform issue.

Add to that a coupe bodywork.  The cockpit shape on a closed top car actually contributes to the generation of lift and is an element the car’s downforce works to counteract.  Certainly most racecars in this genre generate significant downforce and the negative effects of the cockpit bubble are over come if regretted.  It is a fact of life and the designers simply juggle the compromise with the positive benefits in drag reduction that the closed top car allows.

It has also been reported that the CLR was running soft rear springs.  While this may be anecdotal and ultimately prove inaccurate, if true, it simply is one more thing stacking into this deck that worked against the CLR those moments at Le Mans in ’99.  Soft rear springs are sometimes used at high-speed tracks to improve straight-line speed.  At speed, downforce generated at the rear helps squash the rear end down reducing the car’s overall drag allowing for more top speed.

Furthermore, between the practice and morning warm up accidents and the race, it was reported that the team consulted with F1 aerodynamicist Adrian Newey about finding a solution.  If the CLR were to get into a crash the damage would be massive.  If anyone were injured they would be calling a car accident lawyer in San Diego or anywhere else around the world.  One of the ideas to come out of the discussions was attaching front nose dive planes to add a touch of front downforce to the cars.  Both CLRs would start with race with them in the afternoon.  In general, the cars of this era were pretty downforce-light to start, especially in Le Mans configuration.  Looking at data for the open top Nissan R391 LMP900, this is a car with between 2000-2500 lbs of downforce at 200 mph.  

Intriguingly, according to Mercedes-Benz, in a post-warm up crash press release, attempting to reassure that their cars were viable for the race, indicated that the dive planes they would be using in the race added as much as 25% more front downforce.  Assuming a 45/55 split, 2000 lbs total, 900 lbs front, 25% gains you 225 lbs of downforce and is certainly within reason. Rebalanced to 45/55, and you've added as much as 500 lbs to total downforce.  But ultimately it goes back to one thing, how absolutely little aerodynamic downforce the cars that were racing during this era had.     
Finally we get to the “moment”.  Bringing all these issues, and a couple more, together to see why this occurred.  Without the benefit of having witnessed all of the events, I can’t analyze the exacting specifics and will make some generalizations.  In most cases it would appear that the CLR had leading traffic ahead of it.  This certainly took downforce off the car’s nose.  In most cases the CLR was experiencing an “attitude” change, it was either cresting a undulation in the track, running over a curb, something appeared to be changing the car’s disposition, if ever so slightly, leading to a sudden (if in itself insignificant) change in forces.

So working forwards, downforce is reduced off the front of the car due to the turbulence coming off the leading car, the CLR’s pitch is changing due to terrain variations leading to additional downforce lost, the CLR is more pitch sensitive than most (due to those large overhangs and short wheelbase) and these issues lead to a larger than expected downforce change and the nose lifts as the low pressure being produced underneath the CLR approaches zero at the front and suddenly the lift created by the cockpit and top side bodywork begins to take effect lifting the nose even further.  The rear wing is still working pretty well, firmly planting the rear wheels and providing a nice pivot point, the rear wheel centerline.  As the nose lifts at the front, at the back of the car the rear diffuser, hanging way out past the rear wheel centerline, gets closer to the track and begins to generate even more downforce further accentuating the lift.  By now the underside is exposed and the lift being generated by the cockpit, coupled with the face of the exposed underfloor, completely takes over and the car gets airborne in a rather dramatic fashion.

The irony, such was the price of failure that Mercedes Benz decided not to withdraw the cars after the second incident in morning warm up.  I've often wondered if indeed the three incidents were all there were.  At the time I recall hearing rumors of one occurring in testing, but there's never been any traction behind that.  And with the intervening years, Mercedes has decided to forget the incident entirely, and indeed even forget their Le Mans history.  And chances are you'll never see them return to Le Mans.
ęCopyright 2016, Michael J. Fuller