A Lack of Horsepower for the CART

It’s generally accepted today that Ferrari did not create its F637 Indycar and F034 engine with a burning desire to actually race it, (though there are a few differing theories about this).  Rather, it was a tangible expression of the marque’s resistance to the proposal for new Formula 1 technical regulations being ventilated in the mid-80s by the FIA. Planned for the post-turbo era, these would exclude V12 engines.  The existence of an Indycar, ready to go, with the implied threat that this would become the focus at Maranello rather than Formula 1, was seen as effort and expense well worth expending.  In collaboration with the Indycar team, Truesports, run by Jim Trueman, a familiarisation/feasibility study began in ’85.  That autumn, Trueman took his driver, Bobby Rahal, to Italy and at Fiorano they showed what their car – a March 85C – could do.  Impressed with this, Ferrari delegated a small team, led by Harvey Postlethwaite, to develop the project.  Gustav Brunner, recently into his first stint at Ferrari, was responsible for the chassis design.  A go-ahead for engine design/development was given early in ’86.  This incorporated some features informed by experience with Ferrari’s Formula 1 turbocharged V6 Tipo 021 – 033 units, and the Tipo 268/282C/308C, Ferrari twin turbocharged V8s, utilised in the Lancia LC2.  In comparison with then-typical Indycar practice, the F034 was unusual in locating the exhaust headers within the cylinder head banks.  With a cubic capacity of 2648 cc, the 90° V8 gave some 700 bhp at 12,000 rpm.  Around mid-’86, chassis and engine were completed and the car made its initial runs.  By early autumn, progress was good enough to allow a press presentation of the F034 engine.  Meanwhile, commitment to the project was undermined by two key outcomes which flowed from the consolidation of John Barnard’s role as Ferrari’s Technical Director following his recruitment in August:  Barnard was not impressed by Brunner’s capabilities and the quality of his F637 Indycar design; Barnard became adamant that full corporate focus on the Formula 1 programme would be required if Ferrari was to regain Grand Prix-winning potential.  At the same time, there was a recognition in the FIA that Ferrari appeared serious about its Indycar plans and thus sought to placate Enzo, indicating that the new formula would accommodate V12 engines after all, and this was indeed formally confirmed in March ’87.  Thus, the F637’s raison d’etre dissolved and the car/engine was quietly consigned to an out of the way corner in Maranello.

Ferrari F034 engine.  Copyright Ferrari S.p.A.

While the Ferrari Indycar project was in its death throes in early winter ’86, it was finally announced that IRI/Finmeccanica would be entrusting the future of Alfa Romeo to Fiat rather than to rival bidder, Ford.  With both Ferrari and Alfa Romeo under the Fiat umbrella the following year, it is not surprising that the existence of and the canning of the F637 Indycar project piqued some interest in the offices on Viale Alfa Romeo, Arese.  Some, of course, didn’t even want to think about single seat racing cars, the demise of Alfa’s Formula 1 programme only too recent and dismal in terms of the woeful final two seasons with the 184T and 185T, to say nothing of the ongoing paucity of results being achieved with the 890T engine in the Osella FA1 chassis.  Nevertheless, there were those – perhaps mainly the younger, less long-served – who couldn’t see why their ‘brand,’ so historically associated with competition, should not still be participating in a high level of motorsport.  This found some expression in the beginnings of a new project to create a 4 cylinder Formula 1 turbo engine for Ligier, but this stumbled and was cancelled in early ’87.  There was an element of corporate politics involved with that, and Fiat’s group perspective was perfectly rational in compartmentalising effort and budgeting in terms of Formula 1 – Ferrari, Rallying – Lancia, and Touring Cars – Alfa Romeo.  However, for those keen in Arese, this guideline left a loophole – Indycars involvement would not conflict competitively with Ferrari and offered marketing opportunities, (in the large U.S. market), which could be of high sales and profitability importance to Alfa Romeo’s ongoing financial viability. 

Like a conspiracy theory, it’s been said that the end of the Ferrari programme not only prompted interest in Arese, but also provided Alfa with an engine – the F034 – for a project of its own.  But this was not so.  The Alfa Romeo engine, designated M191 – was designed by four of its own people, led by Claudio Lombardi, in autumn ’88.  A prototype ran for the first time the following February.  A month on, it was installed in a CART-specification March 88C chassis and under test at Monza.  Its capacity, (and bore/stroke dimensions), was the same as the F034’s – 2648 cc – and this may have been something that sparked the rumour that it was a badge-engineered version of the Ferrari unit.  That it was not is easily appreciated when the induction/exhaust layout is viewed – on the M191, the exhaust headers are on the outer side of the heads, i.e. fundamentally differing from the F034’s unusual, ‘inside-the-vee’ arrangement.

Alfa Romeo M191 V8 engine – Courtesy Race Cars Direct

A key player in Alfa’s Indycar adventure became involved in the latter part of ’88.  This was Johnny Capels.  Capels grew up and started his motor sport participation in New Mexico.  In the late Sixties he relocated to Indianapolis and ran in USAC events, working firstly with the Unsers and subsequently Parnelli Jones, before joining forces with Alex Morales, for whom he operated as lead technician and team manager.  Following the death of Morales in’88, Capels took over the team’s ownership.  After an initially low key approach from Arese, Capels was asked to facilitate the launch of on-track activities and manage the ’89 race and development programme.  An early step was to engage the interest of driver Roberto Guerrero in the opportunity and have him try the M191-powered March 88C at Monza.  This apparently confirmed the potential to Capels and Guerrero sufficient for them to start planning for a race debut at the 73^rd running of the Indianapolis 500 on May 28^th ‘89.  Cesare Fiorio – recently appointed as the head of Squadra Alfa Corse within the Fiat Group – confirmed this intention on a visit to the Miami CART race in early November.  Given that Alfa Romeo’s Formula 1 programme had yielded so little over the preceding 10 years, the public commitment to its Indycar programme was already a PR-risky venture.  Yet the vulnerability to a negative outcome was compounded by the utilisation of the March chassis.  March had seen considerable success in CART in the mid-Eighties, but then lost focus, partly no doubt because of distractions as the company was converted to a stock market-quoted enterprise and as a related outcome, joint founder, Robin Herd, was forced to relinquish his chairmanship.  The Indycar models through the latter half of the decade lost competitiveness.  The 88C, especially, was disappointing and a trend set in whereby teams began to opt instead for Lola chassis.  It was considered that the Lola was more adaptable aerodynamically and easier to set up for differing circuits.  March responded by creating a new car for ’89 which was more extensively/deeply revised than was usually involved with a ‘next year’s’ car.  The resulting 89C was designed for improved aerodynamic efficiency, better ride/attitude control and ease of general set-up/adjustment.  A further ‘improvement’ was the recruitment of Maurice Philippe to lead and co-ordinate the design and development team – with this came the implied promise of innovation to come, an expertise Philippe had so clearly demonstrated with cars such as the Lotus 49 and 72.

March 88C – used to develop the M191 engine

Modified 88C chassis were provided by March to enable a concerted test and development programme to be started in early ’89 by Capels’s team.  Then, by Spring, two Alfa engine-specific 89Cs were to be completed.  Drivers would be Guerrero, Bruno Giacomelli and Giorgio Francia at the Monza, Nardo and Mugello circuits. Engine dyno testing and shop development was undertaken at Abarth’s Turin base.  At this time, when continuity was all the more desirable, March’s corporate status was once again in flux, with a reversion to private ownership mooted and a Japanese take-over negotiated.  In March ’89, 88Cs were under test with the M191 engines installed and generating around 685 bhp – some 7% down on the output of the Chevrolet ‘A’ V8, (and a similar deficit likely in relation to the Porsche 9M0 unit).  Finally, in late April, the new car, now designated 89CE, made its first on-track appearance at Mugello.  But, as the test programme intensified, it became apparent that there were driveability problems with the engines, significant enough to make the planned race debut at the 500 unfeasible.  However, at least one complete car had reached the U.S. by mid-May and this was afforded a shakedown at Grattan Raceway.  Guerrero‘s times were encouraging, especially after a revised rear wing was tried, and a decision to debut the car at Detroit, 16^th/18^th June in was confirmed.  With this target in mind, the team moved on to Mid-Ohio determined to prove it could optimise the performance of the car on this relatively convoluted road course.  Though track time became limited by rainfall, considerable learning was derived, and it was mostly in terms of causes for concern.  A combination of poor engine performance when picking up from low revs and lack of grip left Guerrero considerably slower than a Lola T89 testing at the circuit at the same time.  Team morale was not helped by the news from Italy that Giacomelli was unable to extract competitive performance from another 89C at the Fiorano circuit.  Back in the U.S. Guerrero next took to the track at Michigan, hoping no doubt that the car could do better on an oval.  But, again, disappointment was the order of the day – the 89CE was 4% down on the previous year’s race pole time.

 

With these testing outcomes proving so confidence-sapping, the team members needed to pull together, yet at this very point continuity took a very hard knock – Maurice Philippe died by suicide on 5^th June.  Frank Cuppock was appointed to take technical lead of the project.  Though it seems apparent that his abilities were not on the same level as his cousin, Gordon’s, (successful designer at McLaren), his scope for improving the car in quick order was clearly limited by such an abrupt and major change in role.  For his first test session in charge, Cuppock was at Indianapolis Raceway Park, (at Clermont, to the west of Indianapolis Motor Speedway), to see the car able to set slightly more competitive lap times.  Here, and at the other venues visited since the car’s arrival in the U.S., a redeeming feature was engine durability, with no big failures/blowups experienced.

 

Finally race-ready, the Detroit weekend did not get off to a very good start for the Alex Morales Autosports team.  Qualifying was not problem-strewn but the 89CE just was not fast.  Guerrero was unable to get anywhere near Michael Andretti’s pole time and the three other Marchs present were all quicker, whether that was Teo Fabi’s Porsche-powered 89P or the Cosworth-engined examples driven by Kevin Cogan or Steve Saleen.  So Guerrero was down in a disappointing 23^rd in the starting order.  But the race sprung a surprise – though not finishing on the lead lap of winner, Emerson Fittipaldi, Guerrero brought his 89CE home in 8^th place.  Though Fabi’s 89P was the best-placed March, in 4^th, it was generally agreed that Guerrero’s 5 points-earning result was a creditable one for a car featuring an engine on debut.

 

Within 5 weeks, Guerrero and the 89CE had run in four more events and inadvertently proved that the ’89 season was not going to be a glorious one for Alfa Romeo in CART.  In those races, at Portland, Cleveland, Meadowlands and Toronto, the car finished only once, (in 13^th, at Cleveland), qualified no better than 19^th and retired as early as lap 4, (Toronto).  Things were no better over the remaining six races – only 2 finishes and only 1 championship point won, (12^th at Mid-Ohio).

 

In early September, in consultation with Alfa Corse, March gave notice to Capels that a contract was being agreed with the Patrick Racing team, (recently bought-into by Chip Ganassi), for the running of the Alfa-engined car in 1990.  Patrick was in especially high regard at that point, having won the ’89 Indianapolis 500 with driver Emerson Fittipaldi in a Penske PC-18, and would end the season as CART/PPG Indycar World Series champions.  Whilst the Penske chassis had proven title-winning, March’s Porsche-powered 89P had seen some success, more, certainly, than the 89CE.  Thus, learning from the 89P was brought into play as work began on the ’90 Alfa-engined March.  At the same time, the Ilmor Chevrolet engine was studied and a number of features ‘copied’ in order to improve the Alfa V8.  In another attempt to strengthen the Alfa effort, Al Unser Snr was signed to drive a second chassis at the Indy and Michigan 500 oval races.

March 89CA   Copyright 2015 Stellantis

The ’90 car was denominated March-Alfa 90CA, the design credited to John Baldwin.  March supplied three chassis to Patrick Racing, (some say four – this being the number that had been agreed contractually).  The team’s engineering expertise was strengthened by the involvement of Mo Nunn, best known for his founding of the Ensign Formula One operation in the ‘70s.  With prominent Miller Genuine Draft/High Life sponsorship, Guerrero was again at the wheel.  At launch, the engine was to ’89 specification, though development work was continuing at pace back in Arese.  Suspension was inboard, front and rear, actuated by pushrods.  Some comment from observers of the static car questioned the body’s bulk and profile rear of the cockpit, suggesting that this would compromise the drag and downforce characteristics.  Whether or not such inferences were valid began to be evaluated in March ’90 at Memphis, where an initial test session with the first-completed chassis was conducted.  Shortly afterwards, at the beginning of April, a second test took place at Phoenix, a week in advance of the championship’s opening round at that track.  By this point, engine performance was improved and Guerrero expressed satisfaction that the 90CA was a definite step up from the 89CA.  This seemed to be confirmed by the stopwatch, since it recorded slightly faster times than a March-Porsche 90P also testing at Phoenix.

Guerrero qualified the 90CA 14^th for the Autoworks 200 at Phoenix, an encouraging performance given that his average qualifying result with the 89CE/88C had been 21^st.  But in the race he was in the wall after 145 of the 200 scheduled laps, leaving him classified down in 16^th place.  Things were not much better at the next race, at Long Beach, where Guerrero qualified 19^th and finished in 14^th, completing 91 laps of the 95 run.  Whatever the reality of the criticisms of car’s aerodynamics as evaluated at launch, the main issue at Phoenix and Long Beach appeared to be engine power, especially as delivered at lower rpm, rendering notably poor pick up characteristics.

 

All too soon, the Indianapolis 500 was the next challenge.  For this race Guerrero was at least not carrying all the Alfa Romeo expectation alone, Al Unser being entered in a second 90CA.  The first days of practice were not inspirational – both cars suffering engine cover fires for example.  In that first week, the fastest runner was Unser’s son, Al Jr – dad and teammate Guerrero’s speed in comparison was over 6% down on that mark.  The following week, Guerrero had a substantial accident – into the wall as a result of a rear wing support strut failure.  Though Roberto was not significantly hurt, the car was damaged beyond repair at the Speedway.  Subsequently after back-at-base evaluation it was decided that it would not be economic to rebuild it.  A replacement chassis was eventually entrusted to Guerrero and he managed to lap a little faster in this.  Emerson Fittipaldi qualified his Penske PC-19 on pole at a speed of 225.6 mph, whilst both 90CAs managed only a disappointing 212 mph.  Of the race’s 33 starters, Guerrero was in 28^th place, Unser 30^th.  That Unser finished 13^th and Guerrero 23^rd therefore doesn’t appear too bad, but Unser completed only 186 of the 200 laps and Guerrero retired with a rear hub failure with consequent loss of drive after 118 laps.  At least Unser had the distinction of being the best placed March driver – both Porsche-powered 90Ps, (Teo Fabi and John Andretti), failing to finish.

 

There was no time to mull over the Indianapolis disappointment as the next championship round – at Milwaukee - followed within 7 days.  Guerrero failed to record an official qualifying time, so was put to the back of the starting grid.  From there he was able to achieve very little – retiring after 88 laps with a suspension failure and classified in 18^th place.  Two weeks later it was back to Detroit, but the relative success of Alfa Romeo’s CART debut there twelve months previously was not to be repeated – the engine failed at less than half distance.  In addition to the circumstances of this in its own right, there was a wider significance.  Alfa Romeo engineers had been talking up the quality, reliability and competitiveness of the engine, maintaining that the continuing run of lack-lustre results was attributable to design and build shortcomings of the March chassis.  It is notable however that Al Unser had been much more forthcoming than Guerrero about performance limitations imposed by the engine’s characteristics from a driver’s viewpoint.  It was also clear that an upgrade rumoured to be under development was long overdue.

 

So, a much better showing at the Portland round in late June was especially welcome.  Having qualified the 90CA 16^th, Guerrero took the car to an 8^th place finish.  There was still a slight disappointment though – as well as being on nowhere near the pace of the 1-2-3 finishing Lolas, Roberto was even 2 laps down on Fabi’s March 90P.  Round 7 of the championship was next up, at Cleveland.  It was not to be a continuation of better performance, Guerrero qualifying only 19^th and retiring at just over one third distance with a damaged transmission.  One week later, and at the championship’s mid-point at Meadowbanks, New Jersey, Roberto had a very similar experience, though the 90CA’s gearbox did at least get past mid race distance before failing.  Despite the upturn in fortune at Portland, the team had become despondent as a result of the car’s lack of competitiveness coupled with a feeling that March was unable to do anything substantial to help improve the situation.  Whilst it was clear that there was a basic shortcoming stemming from the 90CA’s design, it was also increasingly evident that March’s operational resources were degrading, largely as a consequence of financial issues.  Just a few years previously, sales had been good, prospective customers’ confidence being buoyed by good results for example in ’86 when March cars were victorious in every round of the American Racing League.  Subsequently, however, sales slipped as Lola and Reynard products came to the fore, and in ’89 it had been necessary to submit to a take-over by the Leyton House concern.  In its new guise the company retained designer Adrian Newey for a while, but he moved on to Williams Grand Prix not long after March co-founder Robin Herd had also left.  Thus, come mid-’90, Patrick Racing, with Alfa Romeo’s agreement, had concluded that the March chassis should be abandoned and the Lola T90/00 adopted in replacement.  Two cars were acquired, the first arriving in mid-July.

Lola T9000.  © Finarte Auctions S.r.l

The change of car naturally caused considerable operational disruption – enough to dictate Patrick Racing’s absence from the late July championship round at Toronto.  It’s instructive to note that of the twenty seven runners no less than twenty were Lolas.  Only two March chassis appeared – the pair of 90Ps for John Andretti and Teo Fabi.  If these statistics alone validated the decision to switch to the Lola chassis, the next race outcome was even more of a confirmation.  At the Marlboro 500, Michigan, though Guerrero qualified only for the back row, he brought the Lola-Alfa home in 5^th.  But the possibility of enhanced glory was spoilt by the Al Unser’s inability to take the start in the second car in consequence of a big accident on the first day of practice.  Attributed to a front suspension failure, this occurred on only his 7^th lap and left the veteran driver with several broken bones. 

Following up the much improved result Guerrero had managed in debuting the Alfa-powered Lola at Michigan, he achieved the team’s best qualifying performance so far, when he took 11^th place on the grid for the Denver street race.  His race however was over with 20% left to run when he suffered a third transmission failure in four successive races.  As September came round, the next race venue was Vancouver.  The better characteristics of the Lola chassis were further demonstrated by Guerrero’s ability to lap quickly enough for a 10^th position start.  On race day he managed only 29 laps before retiring with collision damage.  In the middle of the month, at Mid-Ohio, Guerrero again qualified well – in 15^th – but was an early retirement with a broken clutch.  Then, at Road America, speed and reliability were improved - qualification in 12^th and finishing in 8^th.  With the advent of October, Nazareth saw another more encouraging result for the Lola-Alfa – Guerrero both qualified and finished in 9^th.  But there was disappointment at the closing round, at Laguna Seca, Guerrero running to near the race’s conclusion, but being forced to retire the car because of a fire.

The latter part of the ’90 season had certainly seen an improvement in Patrick Racing/Alfa’s fortunes and the year-end tally was 24 points and 16^th place in the PPG Indy Car World Series.  An intriguing ponderable concerns the contribution to the better results possibly made by an engine upgrade.  Whilst it is not difficult to find critical commentary on the design of the March chassis – especially rear end aero characteristics – confirmation of the specification and actual delivery of a long-promised ‘new version’ M191 is elusive.  The main items rumoured to feature were 40 valve heads, shorter stroke reciprocals and a revised two unit wastegate set-up.  In regard to that latter configuration incidentally, there is crossover to the question of March’s culpability on the rear aero deficiencies – the single wastegate seen on the M191 gave rise to an asymmetrical body outlet duct which did not help with the profiling of the engine cover’s lefthand side.  An aspect of the Lola design which made it significantly more successful than March’s was the attention to detail in the packaging of the exhaust/turbo system, allowing for a generally lower location of components, with centre-of-gravity, gas blow directional, and airflow-to-wing advantages.  

In recent months, through contact with various people, including former team members, I have renewed my efforts to confirm whether or not upgraded engines were made available to Patrick Racing during the second part of ’90 and ‘91.  Although apparent awareness of work having been carried out at the factory is commonly recollected, I have found no evidence that relevantly modified engines reached the Patrick Racing workshop.  As a power output disadvantage against the competition had been openly acknowledged since the M191 began track testing in ’89, it’s difficult to reconcile a situation whereby improvements had been developed but were not delivered to the race team.  My explanation is that internal pressures at Alfa Romeo forced the Alfa Corse management to become overly-fearful of implementing anything that might attract criticism of the brand’s ability to build a reliable product, be it for the road or the track.  In regard to the latter, a major factor in the failure of the 70s/80s Formula 1 programme was the durability of both engines and chassis.  Most notably, in the season in which the 183T was at its most competitive – 1983 - of 30 starts there were 14 technical failure DNFs.  There was perhaps near-paranoid sensitivity around the robustness of engines, historically a strong point for Alfa Romeo.  That concern was indeed seen to be justified when in May ’91 the higher stress running at Indianapolis resulted in many broken M191s!  Along with the limitations imposed by the conservative approach adopted by the Alfa Corse engine team, it was also hamstrung by a lack of familiarity with methanol fuel, this having very significant effects on performance/response at part throttle.  Looking for insight on this topic and the ‘secrets’ of the Chevy, Patrick had sent one of his ’89 season 265s for detail study in Arese, but it is not obvious what was learnt from this and as it might well be thought a subversive exercise, we don’t have any record of the conclusions.  Yes, of course, there was an upturn in performance/reliability in the latter part of ’90, but this is surely attributable in the main to the work put in by Mo Nunn.  So, another item to be added to Patrick Racing/Alfa Romeo’s ‘90 misfortune’s list was that Nunn would be elsewhere for ’91.

Patrick Racing took to the tracks in ’91 not only Nunnless, but also with a new driver.  True, Roberto Guerrero had not fulfilled the promise perceived by many when he began driving the March-Alfa in ’89 – he had, after all, been in with a real chance of winning the Indy 500, and indeed the championship, just two years previously – but the best he could cite for his ‘89/’90 seasons was the 5^th place finish at Michigan.  In contrast, Danny Sullivan, driving for Penske Racing, had enjoyed a competitive ’90, with two wins and three additional podium placings.  There was considerable surprise when it was announced that he was moving to the less fancied Patrick Racing outfit and the Lola-Alfa for ’91, particularly as he had been the winner of the final two races of the ’90 season.  Perhaps Sullivan had been presented with a very positive picture regarding the car and engine with which he would be working, but the reality – as far as the engine was concerned at least – was rather different.  Behind closed doors at Arese, the management team had just about given up on the Indycar project in terms of anything more than the impending new season.  The Alfa 75, the model with which Alfa Romeo had hoped to achieve a volume sales breakthrough in the U.S. had already been discontinued after just two years of marketing as the ‘Milano.’  For ’91, the new 164 would be marketed in the U.S., but volume expectations were reduced and the justification for major expenditure to directly influence the American market could not be argued on a positive cost : benefit ratio basis.  Instead, Alfa Romeo was once more returning to the concept of raising its profile by supporting motorsport participation on a global basis.  Thus, where engineering resources had been focused on the M191 engine, they were now redirected towards projects such as the 164-based Procar proposition and the ’92-‘96 155-based Touring Car programmes primarily for the DTM and BTCC series.  No further development work on the M191 was undertaken and related Alfa Corse operations effectively went into run-down mode, though Sullivan was impressed on a visit to Arese by what he perceived as activity apparently not constrained by a lack of budget.  That serious money was available for this ‘one last gasp’ season was further indicated by Sullivan’s pay being reported as around $1.75M, thanks to a significant degree to sponsorship income from Millers (beer) and Q8 (oil).

 

In announcing the T91/00, Lola had characterised it as generally a refinement of the T90/00, no major redesign work being required given that the car had taken Al Unser Jr to the Indycar title and was the winning machine at 12 of the 16 rounds.  Key modifications included widening of the monocoque to enable a lower-mounted fuel tank, new front wing for reduced pitch sensitivity, new 5 element rear wing, improved suspension component access and adjustability and both drivetrain and brake cooling systems revised for greater efficiency.  By the ’91 season’s end, it was clear that Lola had achieved its aim to maintain its product as the #1 Indycar choice, the T91 taking victory at 13 of the 16 championship rounds.  The first of these, in mid-March, was at Surfers Paradise in Australia.  Sullivan qualified in 10^th, fastest car not powered by a Chevy.  Thanks to a good race start, picking up places and then a number of retirements among the front runners, the Lola-Alfa defied general expectations and crossed the line in 4^th.  So, at this first time of asking, Patrick Racing in its ’91 guise, achieved the best result for the M191 since its ’89 debut.  Thus, the team arrived at Long Beach for round 2 in optimistic mood.  But this was immediately undermined by lack-lustre performance in practice and qualifying.  The car was handling well enough, but Sullivan was frustrated in being unable to extract a competitive level of engine performance.  The outcome was a grid position back in 16^th.  There was some race progress – mainly as several cars succumbed to mechanical failures and accidents – but the finish was as low as 11^th and quite flattering at that, given that Sullivan was three laps down on Al Unser Jr’s winning Lola-Chevy.  This was disappointing enough to prompt some intense activity as soon as the car was back at the Westside base, resulting in diagnosis of the power loss problem as a fuel injection fault, inducing over-lean running.  With that corrected, the next qualifying effort, at the Phoenix 1 mile oval, surpassed the encouraging Surfers Paradise mark – 6^th fastest and well in touch with the front row pace of the Chevy-engined Penske PC20 of Rick Mears and the Lola T91/00 of Michael Andretti.  A further morale boost was provided by two race laps being led by Sullivan.  In addition, Sullivan was credited with the race’s fastest lap and he held 4^th position in the latter stages.  But, with just two laps to go, the engine let go and the car was classified in 7^th.  The three pre-Indy 500 rounds had been productive for the team, with 20 championship points accumulated, (whereas the entire ’90 season’s total had been only 4 more than that), and the project-best 4^th place result in Australia.

Danny Sullivan in the Lola T91/00. Copyright Lola Cars

The largely upbeat team mood engendered by the March/April interlude was always going to challenged by May’s focus shift to the ultra high speed Indianapolis Motor Speedway.  Though those first three races had seen Sullivan running near the front, the Alfa engine’s power disadvantage in comparison with that of the Chevy had remained clearly evident.  But, more disappointingly, once any tweaks to tease a little more grunt from the M191 had been enacted, and the units had begun to run at the sustained higher speeds attainable at Indianapolis, reliability took a very big hit.  That development had indeed stopped at Alfa Corse is shown by Sullivan’s experience during practice and preparation for the race – Danny told me, ‘I blew up 13 during the month of May, and no power, which was worse.’  He does not recall being told by anyone at Patrick Racing that any upgrades were in preparation.  This left Sullivan struggling to achieve lap speeds above 220 mph, putting him at a deficit of around 2% against the cars vying for the pole.  Thus, his best-likely race expectation was that he’d be four laps down on the winner and probably good enough only for a finish in the 6^th – 10^th bracket.  Sullivan was of the opinion that the main issue was lack of usable torque rather than outright brake horsepower.  In attempting to compensate for the power issue and increase straightline velocity by running a near horizontal rear wing angle, a compounding of issues was inadvertently garnered, since the resulting reduced downforce hurt speed capability in the oval’s four turns.  Given all this, it’s testament to Sullivan’s skill and persistence that he was able to qualify as high a 9^th, with only Jim Crawford’s Buick-engined Lola faster of the non Chevy-powered cars.  For this event Patrick entered a second car, for Guerrero, but Roberto was nowhere near Sullivan’s pace, qualifying in a lowly 28^th.  Ironically though, Roberto’s race get away was significantly better since Danny’s car was stationary, under investigation for a lack of fuel pressure.  Though quite quickly diagnosed, he was unable to get going until the rest of the field had already completed the opening lap.  His persistence and expertise in the traffic nevertheless enabled him to make considerable progress and he was well into the points by the last quarter of the race.  Retirement as a consequence of a failed turbo was therefore all the more hard to take on lap 174.  As partial compensation, Sullivan was classified in 10^th place, whereas Guererro’s race was cut short with 24 laps completed, after tangling with Kevin Cogan’s Lola-Buik.  Cogan had pushed Guerrero outwards mid-turn, resulting in both cars hitting the wall.  A suspension arm detached from Guererro’s Lola was struck by A J Foyt’s car.  A J was uninjured, but unable to continue.  Guererro suffered a bruised shoulder, but worst hurt, with several broken bones was the accident’s instigator, Cogan, the injuries preventing him from competing at the Brickyard in ’92.

Guerrero emerges from his wrecked

 Lola-Alfa at the ‘91 Indianapolis 500

However downbeat the Patrick personnel may have felt as they packed up at the Speedway, just seven days later their spirit and optimism had been restored by Sullivan’s performance in Wisconsin.  At the Milwaukee Mile oval, Danny qualified 4^th and finished the race in 5^th.  And this time he was more in touch with the four Chevy-engined cars ahead, completing 197 of the 200 laps.  A fortnight later, the championship’s sixth round was hosted at Detroit.  Qualifying here was a step backwards for Sullivan, midfield mediocre pace enabling him to take no better than 13^th.  There was a little progress in the race, with the second 10^th place of the season recorded.  But, more positively, with one third of the season done, all six races had returned points, 36 in total.

At Portland, the customary monopolisation of the leading grid slots by the usual Lola-Chevy suspects was considerably disrupted – Fittipaldi and Mears took 1^st and 2^nd in their Penske PC-20s, Scott Pruett was 3^rd in a Truesport 91-Judd, Michael Andretti and Al Unser Jr were 4^th and 5^th, Sullivan 6^th.  Danny’s race was, however, quite soon over, engine failure occurring after just 30 of the 104 laps.  The following round at Cleveland saw Sullivan and the Lola-Alfa less competitive in qualifying, but yielded a points-paying 9^th place finish, with every car ahead being Chevy-engined.  At Meadowlands, Sullivan was even less successful in qualifying, (14^th), but was once again the highest finishing non-Chevy car, in 6^th place. 

The next three races – at Toronto, Michigan and Denver - were especially disappointing as in each, electrical problems led to retirement.  At Vancouver, however, a good qualifying performance, (taking 9^th place), was followed by a solid race, finishing in the same position.  But in the remainder of September and at the beginning of October, non-finishes occurred at Mid-Ohio, Road America and Nazareth.  At least at these events (and also at Denver) Sullivan had been able to qualify within the top twelve, but at the season’s final round at Laguna Seca, he could do no better than 15^th.  The race was somewhat better, a 9^th place finish, and 4 more championship points.  Indeed, Danny had scored points in 10 of the championship’s 17 rounds, for a total of 56 and 11^th place in the drivers’ standings table.  So, a much more productive season, but, once again a relatively uninspiring 4^th place was the highlight result.  With that in mind, it is perhaps not difficult to understand why there was no pause to reconsider the decision, communicated to Sullivan in August, that Alfa Romeo would not supply engines for a ’92 championship programme.

 

Given that Alfa Romeo embarked on the CART Indycar project with no previous experience of the category, it might well have been thought realistic to plan and budget for at least a five year programme.  In Autumn ’91, at the conclusion of the second full year of the programme, enough progress should have been perceptible and supportive of ongoing involvement along with a reinvigoration of engine development activity at the Factory.  But the budget situation was dire and unsalvageable.  Whatever had happened in ’91 on-track, it could not affect the diminished corporate ambitions regarding road car sales in the American market.  Meanwhile, the global sales/marketing results for Alfa Romeo remained highly problematic, leaving the marque dependant on further funding from Fiat, (possibly with support from IRI/Finmeccanica), for any motorsport initiatives.  For Fiat, there was a rationale to seek enhancement of Alfa Romeo’s sporting image, but it was seen that this should be biased towards its presence in the European markets and with a direct relevance to its higher volume road car products – hence the focus on the 155 (and, subsequently, the 156) in Touring Car guise. 

Not for the first time, Alfa Romeo had sought sporting success on a speculative basis, but fallen short, mainly because of an inability to plan -and provide funding – for the long term.