The Zeltweg Solution

Young, single minded, supremely confident and abrasive – such a character often makes waves but is usually defeated by the authority and inertia possessed of an elder generation of management.  But Ferdinand Piëch was a bit different.  Joining the ‘family firm’ – Porsche – in 1963, Piëch was already in charge of testing in 1966 at the age of 24.  As the son of Ferry Porsche’s sister, Louise, some may have thought there was a touch of nepotism about this – but Piëch would rapidly show that it was actually all about his ability and character. 

Piëch (right) at the 1965 Targa Florio with Fritz Huschke

Piëch soon became deeply involved in the creation and evolution of the Group 4/6 Porsche 906.  This was a ground-breaking project for Porsche as the car embodied the adoption of key, pure-race features, primarily, 200+ bhp, lightweight 6 cylinder engine, glass fibre body, super-light tubular space frame chassis and high focus on aerodynamics, (the 906 was the first Porsche wind tunnel-tested), with emphasis on lowest possible coefficient of drag.  The car was an immediate, class-winning success.  This was instrumental in reinforcing Piëch’s firm belief in the primary importance of vehicle top speed, based on lightness of construction and an exceptionally low Cd metric.  Several of his colleagues have said that this went beyond ‘firm belief’ into the realms of unshakable dogma.

Porsche team with Piëch, centre,  on return from 906’s winning debut at Daytona

In its first season, the 906 scored class victories in the major sports car races at Daytona, Sebring, Monza, Targa Florio, Spa, Nurburgring, Le Mans, Watkins Glen, Mugello, Zeltweg and Montlhéry.  It was successful too in the following two seasons.  And while all this convinced Piëch that his technical ‘recipe’ was unchallengeable, Ferry Porsche became certain that a racing project of the utmost importance could be entrusted to his dynamic young nephew.  Meanwhile, certainty was a state of mind difficult to find amongst the senior officials at the Fédération Internationale de l'Automobile.  It had felt some alarm that the topflight sports car racing it administered had seen an incursion of entries involving cars equipped with large capacity production stock-based American V8 engines.  Ford had led this trend with the introduction of the GT40, initially utilising 4.2 and 4.7, and, subsequently, 7.0 litre engines.  Once an American victory at Le Mans had been achieved in 1966, the FIA became concerned that the European makers such as Ferrari, Porsche, Maserati, etc, would be deterred from future competition.  So, in reviewing its technical rules, it sought to redefine Groups 4 and 6 in the favour of European-taste products, primarily by imposing engine capacity limits of a 3000 cc and 5000 cc respectively.

For Porsche too, the second half of the decade looked likely to be hugely significant.  Like Ford, it was eager to have its status as a premium marque confirmed by racing success at World Championship level.  And, as had been the view at Dearborn, an outright victory at Le Mans was seen at Zuffenhausen as a prize of inestimable value.  Ferry, his careful, conservative nature to the fore, blanched at the prospect of needing to build a 5.0 litre car to achieve this.  In every which way he contemplated the costs that would be involved – made all the worse by the homologation requirement to build 50 examples – he struggled to find the courage to make the commitment, fearing that even if successful in sporting terms, the project could bankrupt the company of which he was so proud.  But the pro-build lobbying of Piëch was suddenly given added persuasiveness when the FIA announced in April 1968 that the number of cars required for homologation would be reduced to 25.  Within two months the Porsche board had agreed a go-ahead for the 917 project and in the July Mezger’s team began work on designing the type 912 flat 12 engine.

Original 917 Longtail with front tabs and rear flaps/spoiler

Meanwhile, though the 907 was enabling several drivers to become used to being class victors, it was also bringing them regular moments of extreme concern.  For, whilst they enjoyed the availability of top speeds in the order of 190 mph, at such pace the car was notorious for its instability stemming from its ultra lightweight chassis and low drag/downforce body, the very features that enabled the speed.  For Le Mans and Daytona, the 907 ran with a longtail body, but this gave no better drivability characteristics than the short-tail version.  As the 1968 season dawned, the 907 ‘became’ the 908, with initial concentration on a longtail variant.  This proved just as liable to wander at high speed at the Le Mans test event.  Trim tabs and spoilers were found to have little beneficial effect, but the addition of two turbulence-constraining fins on the tail did.  The following year it was the new 917 which was in La Sarthe in June, and, as might have been expected, they tended to need the full width of the straight while carrying their even greater maximum speed of around 215 mph.  This remained a characteristic of the 917 through its maiden season despite the rear body movable flaps which had been a design feature from the outset.  This issue and how it was addressed is the main subject of this article, we’ll come back to it later.

Piëch (right) with Gerhard Mitter at the public debut of the 917 at Geneva,  March 1969




Piëch at the FIA homologation inspection, April 1969

The closing two decades of the twentieth century saw an increasing tendency for cars to be ‘designed by committee.’  For consumers of an enthusiast bent there grew a feeling that the products they were being offered lacked distinctive character, failed to be individually innovative and were more and more generically formulaic.  This contrasted with what had made the sixties and seventies so much more rewarding – an industry which allowed outstanding individuals to create vehicles brimming with their own personal creative DNA – stylists such as Giorgetto Giugiaro, Marcello Gandini and Tom Tjaarda, engineers like Sir Alec Issigonis, Giotto Bizzarrini and Rudolf Hruska.  To that second list we could add Ferdinand Piëch.  Of particular interest here, his creation of the Audi A2 is worthy of some consideration.  Although the model’s design is credited to Luc Donckerwolke, he specified and visualised the vehicle in compliance with the very detailed brief dictated by Piëch.  Key features, with ultra-high fuel use efficiency as an overriding objective in mind, were: aluminium body/chassis construction; optimum interior space utilisation/minimum external footprint and frontal area; small capacity/high efficiency petrol and diesel engines.  It is likely that progress of the concept through to volume production would have been halted at any factory other than Audi’s at that time since the use of aluminium significantly complicated and added cost to the build process.  What got the A2 through to dealer showrooms for its 1999 launch was Piëch’s force of personality and belief that he knew better than anyone what customers really wanted in a contemporary small car.  That Piëch was an exceptional ‘force of nature’ is recognised for instance by the existence of academic studies analysing his psychological make-up, most notably the Frontiers in Psychology paper, Ferdinand Karl Piëch: A Psychobiography of a Ruthless Manager and Ingenious Engineer, by Claude-Helene Mayer, Roelf van Niekerk and Nicola Wannenburg, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8490634/

However, five years later I was making presentations at Audi UK regional dealer meetings and witnessed a personal ‘first.’  I had previously attended many such meetings for the BMW, Alfa Romeo, Chrysler and Lotus marques and had frequently enjoyed seeing the attenders expressing enthusiasm for a new model/version announced at the event.  But in 2004, I was just a little taken aback by similar glee being forthcoming, but, and this for the first time in my experience, in response to confirmation of rumours that a model was to be canned.  Yes, that was the A2, a car that despite many good attributes was not liked at all by most Audi dealers.  For them, it didn’t ‘fit’ – the brand was building a ‘family’ of models which shared DNA associated with aggressive style, performance and ‘street presence.’  The A2 didn’t even hint at any of that, but, worst of all, it was seen as too expensive for its natural market sector.  All that aluminium and the specific build processes involved made that inevitable.  Piëch was credited with great vision in identifying a car of the future but berated for persisting with the concept and being responsible for consequent and considerable financial losses.  Nevertheless, it is likely that Piëch himself never accepted that his concept was incorrect for the particular context – it would be the customers and the dealers who hadn’t a clue.  Recognising that attitude of mind now helps us understand just what happened at Zeltweg in mid-October 1969.

The commonly quoted anecdote is that senior engineer, John Horsman, of JW Automotive, (the Gulf-sponsored team that would spearhead the Porsche World Championship campaign in 1970), noticed something missed by the factory engineers, Peter Falk and Helmut Flegl.  Brian Redman and Kurt Ahrens Jnr. flogged round the Osterreichring circuit for the best part of two days, failing to achieve the target lap time by a margin of 1.5 seconds.  Then, according to the ‘legend,’ it dawned on Horsman that whilst most of the bodywork of the two test cars was peppered with insect corpses, the rear bodywork surfaces remained clean and perfectly inoffensive to any entomological specimens minding their own business whilst making their way down the pitlane.

Two aspects of this missing midges mystery may have been misunderstood by certain parties and at certain times in the past.  Firstly, several writers, contemporarily and in more recent times, have concluded that those clean top surfaces indicated that no airflow was attaching there, robbing that part of the car of a downforce pressure influence.  However with today’s ever greater understanding of aerodynamics, a more likely reading is of a rear body shape that attracted and conducted an airflow with potential to exert downforce but which also tended to allow turbulence from the sides to spill over and undermine that downforce.  What is more important, of course, is that Horsman set JWA’s Ermanno Cuoghi and Peter Davies to the task of fabricating there and then a ramp-like cover which provided a new, raking rear section which anticipated the 1970 Kurz version of the car.  Though they were unable to get the work finished that day, the modified car was ready for Brian Redman to try again the following morning.  And how good do we think did Redman – and the engineers – feel when his immediate reaction was that the car was transformed and actually enjoyable to drive!

Misunderstanding 2: John Horsman was undoubtably an excellent engineer with a lifetime of achievement and a career marked by association with some of the most prestigious marques – Aston Martin in the 50s/60s; Ford Advanced Vehicles for the 60s Le Mans victories; JWE/Porsche; Gulf Research Racing with the Mirage model in the mid-70s.  However, it is most unlikely that it was only John who noticed the absent arthropods.  Falk and Flegl surely observed the same phenomenon but were quite possibly reticent for good reason.  That body shape, evidencing its aerodynamic flaw such as it was, was also demonstrating its supremacy in the low drag stakes.  To disturb that with revised rear panel sections, spoilers and flaps would surely be an act of sacrilege in the view of the man who had specified this automotive holy grail – Ferdinand Piëch.  And with their responsibilities within the 917 project, the last thing that Falk and Flegl really ought to be doing was challenging a basic tenet of the Piëch philosophy.  Thus they were perhaps being understandably self-protective when they 'did a Nelson' and didn’t mention the lack of gnats – safer to leave that to John!

Happily for Porsche, engineering niceties stood back, there was no overt challenge to Ferdinand’s principles – to his face at least - and the Zeltweg solution snuck through.  On the second day of the tests, Ahrens’s best lap had been 1.48.2, whereas he eventually got down to 1.43.2 before everything was packed up for return to Stuttgart.  In Karl Ludvigsen’s account of the tests, (in Porsche: Excellence was Expected, Book 2), he records that a management contingent including Piëch arrived at the circuit the next day.  Horsman apparently recalled, “. . . the group ignored our presence as though we were hired day labourers and departed without speaking to us!  Oh my, I thought, what have we gotten ourselves into.”  Ludvigsen’s narrative continues:

All too soon he (Horsman) would learn of the rival Piëch-backed Salzburg team.  When John Wyer and Ferdinand Piëch met later, the Austrian smiled wryly and said, “It seems that your tail is three seconds faster than ours.  So that is what we must do.”  It was a solution that came from all of us,” summed up Peter Falk.  “Later Mr. Wyer and Mr. Horsman said it was their work.  And we naturally told Mr. Piëch that it was our work.

 ‘Field’ rear body modifications made at Zeltweg, October 1969

Back at base Eugen Kolb designed a ‘proper’ version of the Zeltweg tail and also drew some changes to the car’s front wings and nose.  This new ‘Kurz’ 917 was also characterised by changes to the area around the engine, exposing the horizontal fan/intake trumpets, and facilitating better visibility to the rear.  The changes were homologated in mid-November by the FIA.  They paved the way to a season of success, including, at last, the longed-for victory at Le Mans – ironically, in view of the availability of the circuit-specific Longtail variant – by the 917 Kurz of Porsche Konstruktionen Salzburg, driven by Dickie Attwood and Hans Herrmann.

Revised profile incorporating required changes identified at Zeltweg


Modified 917 as at year-end, 1969

 

Comments

Popular posts from this blog

The Nearly Alfa - Part 2

Jo's Last Season

Abarth T140 Update