The F1 heritage is real.

The F1 heritage is real.

"In this new Special Series named “S85-002”, I earn a badge of honour: I blew up my V10.

I document the service history of my original S85 engine and diagnose its failure. I go through the process to source, refresh, cam and refit the replacement V10 into the E60.

In this entry, I go in-depth on the history of the V10 F1 era, the BMW Motorsport F1 program and what remained of its influence on the S85 V10 development."
- Matt

A caricature of an era. 

In most conversations around the S85, the F1 lineage starts and ends at an unnamed foundry where both the F1 & S85 engine blocks where made. As you're about to discover, this was merely the tip of the iceberg. 

For most, the F1 heritage a laughable proposition in light of fitting this V10 into a mid size, 4100 lbs saloon. At best, it's a a rationalization used by V10 owners when they ultimately face yet another 3,000$ repair bill. 

The perception was compounded with BMW M using and abusing the comparison in its superficial marketing campaigns around the launch of the M5 V10. The technical lineage was left to technical training documentations.

The journalists salivated: Fifth Gear famously reminded us our V10 was fitted in a saloon chassis.

The S85 was from a different breed. 

The S85 was engineered with a vastly different core design philosophy than the S50B32.  Along with 4 extra cylinders and 12.0:1 compression ratio, the S85 was made to rev beyond 8,000 RPM. It was designed as an over-square with a 92mm bore x 75.2mm stroke, equating to a Bore : Stroke ratio of 1.22. 

The major shift in design philosophy, engineering and manufacturing capabilities had to come from somewhere. 

The S85 had purely bespoke parts, differing from every other Sxx engines ever designed. 

Standing on the shoulders of giants. 

The S85 was designed and engineered following some of the greatest of BMW M street engine masterpieces: the S50B32 in the Euro E36 M3 and the S72 found in Gordon Murray's legendary creation.

These engines all have one thing in common: they were developed by Paul Rosche's team. 

M. Rosche held the post of Technical Director for BMW Motorsport. He was a fixture of BMW Motorsport, having headed development of production and racing engines for the BMW M1 amongst others, from 1979 to 1996.

He was responsible for the development of various technological features of S series engines such as dual VANOS first found in these legendary engines.

With a bigger stroke than bore, the S50B32 and S72 were under square powerhouses. They made significant torque for their displacement, but were rev limited below 8,000 RPM. 

The E36 M3's Euro engine was essentially half of the Mclaren F1 engine - or a quarter of those pictured here.

Yet, Paul's signature achievement was the Mclaren F1's V12 engine. 

M. Rosche retired in 1999 on top of his craft. 

I've compiled the comparison below between the S85 and S72. There were fundamental differences. If you find further information about the S72, reach out!

Specs  S72 S85
Displacement 6.1 liters 5.0 liters
Max Power  618 hp at 7,400 rpm 507 bhp at 7,750 rpm
Max Torque 479 ft/lbs at 5,600 rpm 384 ft lbs at 6,100 rpm
Max Rev 7,500  8,250
Power per liter 101.31 hp 101.40 hp
Weight (fully dressed) 584 lbs 529 lbs
Oil System Dry Sump Quasi-Dry Sump
Bore 86mm 92mm
Stroke 87mm 75.2mm
Compression Ratio 10.5.:1 12.0:1
Construction Aluminum Aluminum
VANOS Dual, high pressure Dual, high pressure
VANOS Pressure 1,500 PSI 1,160 PSI
Valves per cylinder 4 4
Intake Valve 35mm
Exhaust Valve 30.5mm
Camshaft Drive Chain Chain
Pistons Forged Aluminum Cast Aluminium
481.7 grams per
Rods

Forged Steel
623 grams per

 

The F1 bloodlines. 

To better understand the context in which the S85 was developed, we need take in the scope of what was a bonanza-era of F1 engine development.

The sound.

We need to get the fanboy argument out of the way first: the unique sounds produced from this era. The V10 sang at higher pitch than the V12s, and rev'ed way, way higher - by over 2,000 RPM. 

This video has been on repeat for the hours spent researching, compiling, uploading, writing, aligning narratives correcting, re-editing and staying true to its goal:

The S85 is the greatest BMW M engine ever made. 

Peak money.

The FIA imposed little regulations beyond restricting engines to 10 cylinders, its 3.0 displacement limitations and a yearly focus on increasing engine life.

The hype around the V10 era is in large part due to the implication of many large scale automotive manufacturers in what was the best manufacturer-represented era of Formula One. 

  • Mercedes Benz partened with McLaren
  • Renault had bought the Benetton Team. 
  • Jaguar took over Stewart.
  • Honda first supported BAR and eventually bought them out.
  • Toyota entered with a factory team. 
  • Peugeot was an engine supplier until 2000. 

With budgets adjusted to inflation, there is no other F1 era with such condensed development funds to a single engine architecture. From the nearly unlimited budgets of manufacturers, approximately half went to engine developments.

This resulted in some of the most impactful naturally aspirated V10 engine developments. The 11 teams used 10x V10 different engines in 2000, while Ferrari, Honda and Ford supplied their customer teams with older versions of their engines.

Below are the 2004 operational budgets per teams according to F1 Racing, and their WCC standings. It peaked to 500MM in 2005 led by Toyota, but data is unfortunately scarce on other teams. 

Team  Yearly Budget (US$ MM) WCC Standings
Ferrari 418.2 1
Toyota 368.5 8
Mclaren-Mercedes 359.2 5
Renault 359 3
BMW Williams 309.9 4
BAR Honda 255.2  2
Jaguar 154.6 7
Sauber Petronas  141.9  6
Jordan 79.92 9
Minardi 46.58 10

 

Budgets peaked in 2005 before the V8s were introduced and more cost cutting measures were introduced.

Ferrari still dominated

We could debate all day about GOAT eras of F1s - yet it is undeniable that Michael Schumacher and Ferrari triumphed in what was the most purely competitive era outside the race track.  

In it in these circumstances that Ferrari won 5 straight Driver and Constructor Championships from 2000 to 2004. The massive budgets didn't create the competitiveness we would have hoped - but that is what apex domination is. 

I've always remember being at the Senna S'es in Montreal when I first heard a damn Minardi start up for practice session.

Still, it paled in comparison to the race start with 20 + V10s diving into the first turn with Schumi leading the way. 

The French got one. 

Only Renault managed to snatch one away from the Italians, edging out Mclaren and Mercedes in the process. Some will say it had little to do with their RS25's 72 degrees V10 engine performance and more to do with regulation changes to aerodynamics.

Hats off to Renault's risky budget allocations to aerodynamics as they were taking Ls in the Summer of 2024. 

My man, Alonso - he had every right to rejoice, and date Taylor Swift 15 years later. 

Toyota would call BMW next time.

For as much vitriol was directed at Ferrari for driving costs of F1 to the moon in its obsessive, often times disgusting ways to win with no absolutely no regards for the fans [...] there was no bigger loser than Toyota.

They proved to us all you can't buy your way to a championship. In 2005, Toyota outspent Ferrari by 60 MM US$ for a total budget of half a billion dollars. They still took the L against la scuderia. 

Most tragically, they never managed to win a single race. Back to making Corollas and time to outsource engines to BMWs my guys!

The Tifosis had a Frenchman. 

Ferrari's engine family of V10 spanned a decade, from 1996 to 2006. Starting in 1999, the Tipo engines were developed under the guidance of Gilles Simon.

The Frenchman had been working at Peugeot, developing V10s for their LeMans and contributing to the McLaren V10 engine development. In 1993, he followed Jean Todt to Ferrari and worked under the mentorship of Paolo Martinelli on engine development. 

He's standing French next to Luca Di. The famed direttore had an eye for talent. 

His first: the 048. 

In 1997, Gilles was promoted to the role of Head of Engine Development and introduced his first Tipo engine for the 1999 season: the Tipo 048. 

This engine continued the trend of increasing bores and shortening strokes. It was massively oversquare with a ratio far exceeding the famed 2.0. The 94mm bore and stroke of 43.19mm had a B:S of 2.18. 

The 048 had a 80 degree angle: helping in reduced the engine's height, and its center of gravity. 

Ferrari were notably the first to make sure of the top exit exhaust. Engines were closely intertwined with aerodynamics: the shorter exhaust path benefited aero. It also matched the shorter stroke pulses more efficiently to generate more power.

All teams followed suit with this design in following years. 

Ferrari won their first World Constructor Championship in 20 years in 1999. 

Gilles had an angle.

In 2000, Gilles led the design and operations development of the latest Tipo engine: the 049. The Tipo 049 continued the trend of oversquared engines with further enlarging bores by 2mm, increasing to 96mm. They also reduced the stroke to 41.4mm, bringing the B:S ratio to to 2.318. 

Still, this wasn't the most impressive technological advance. Ferrari introduced a ground breaking, immensely counterintuitive 90 degree V angle design. It lowered the center of gravity of the F2001 beyond what the competition could achieve. 

It was one of many technological edges found in the F2001. 

The angle would create an uneven firing order and traditionally caused balancing issues. Still, Ferrari knew how to solve that problem thanks to its massive budget edge. They were the only team in 1999 to use complete engines for development, whereas other teams used 1 cylinder rigs. 

Ferrari had been optimizing their V10 design based off their years of experience. Back in the early V10 days, their V10s did not have a balance shaft and initially cracked the titanium casings of gearboxes in 1998. They knew the fixes by 2000. 

In 2000, they were the only team using the 90 degree angle. 

Team

 V Angle (degres)

Ferrari Tipo 049 90
BMW E41/42 72
Jaguar Cosworth CR2 72
BAR Honda RA 000E 80
Jordan Mugen MF-301 72
Mercedes IImor F110J 72
Prost Peugeot A20EVO4 72

Ferrari shocked the F1 grid and dominated the season: winning the WCC and WDC. Michael scored what was then a record 123 points while taking home 9 wins. 

His masterpiece: the Tipo 050. 

In 2001, they introduced their 2nd generation Tipo V10, the 050. The latest engine had further enlarged bores of 96mm with short strokes of 41.4mm bringing up the B:S ratio to 2.32. 

The engine block was made of die-cast aluminum with 7% silicone content, referred to as Nikasil. 

The Tipo engine could rev higher than anyone: up to 19,000 in qualifying trim, making up 900 horsepower out of 2,997cc. It was dialed down to 17,300 RPM in race trim, making an estimated 825hp.

It had a unique sound, contributing to the Ferrari mystic of the era - even though they never fitted a V10 to a road car. 

The 050 was the most powerful engine on the grid, only to be exceeded a year later by you know who. 

BMW Motorsport's V10 development.

In 2000, Mario Theissen's took over management as director of BMW Motorsport. BMW went all out: the team grew from 20 up to 200 and a new brand F1 factory was built. 

The development phase of the V10 engines can be broken down into three stages: the lead up from 1997 to 2000, its collaboration as engine partner for Williams, and the final four years with its own BMW Sauber F1 Team.

It all started here: a blank canvas. 

The stages. 

In the early years of its involvement in F1, BMW developed and built a new engine for each season. Still, BMW F1's engines were always designed, developed and manufactured in Munich - lessons were learned and improvements could be implemented. 

Little care will given to V8s in this entry.

The pre-historic (!) E41/4.

BMW started development on F1 V10 engines in 1997, starting from a small team of 20 that was initially headed by Paul Rosche who eventually retired in 1999. 

The timeframe was short, the team wasn't accustomed to one another, operational processes were lacking and what were QA standards anyways? Derived from various V12 concept engines from the 1990s, the E40/41 was the first engine produced. 

The even firing order was a straight take away from V12 engine development. 

The E41/4's cylinder angle of 72° degrees ensured reduced vibrations with the even firing intervals. The low angle V design allowed a slim, but tall design. This enable a narrower body which had a direct negative impact of aerodynamics.

The E41/4 was an improvisation based on old engine philosophies that called for even firing orders at all costs. 

 

It wasn't perceived as ideal in terms of vibrations and it made for a tall engine with a high centre of gravity. As the 2000 Australia GP rolled around, the E41/4 was not yet race-ready and the failure rate was high.

The intake ports and pneumatic valvetrain of the E41/4 wasn't well optimized for the V10 - still, pay attention to the head's casting. 

The 94mm bore and 42.3mm stroke had a 2.2 B:S ratio. They were behind Ferrari. 

P8x engines: the prime years. 

BMW Motorsport took notice of Ferrari's success with a narrow angle V10 and dropped Paul's 72 degrees heritage. 

P80. 

The best overall package solution was considered to be a cylinder bank angle of 90° for the P80. However, it wasn’t just its cylinder bank angle that set the P80 apart from the E41/4, this was a completely new design.

The P80 was further improved with revised ports, modified rocker arms and tuned valve lift curves allowed maximum engine speed to be increased to 18,000 rpm and power output to be increased up to 880 hp.

The P80 had an impressive bore of 95mm and stroke of 42.3mm for a B:S ratio of 2.245

Still, the it fell short of Ferrari's masterpiece Tipo 050's B:S ratio of 2.32.

 

 

The P8x series saw rapid pace of improvements leading to BMW engines reaching cult status.

The development on new engines would start before the previous powerplant ever saw a single race: 1,900 + CAD drawings and over 1,000 upgrades were made for, and to  each P8x series engine development.

Little information is publicly available. Below is the engine dyno room used for the first V8 development: the P86.

Records were broken.

This frenetic development pace brought about impressive engine performance. With the P80, BMW had moved to the head of the field on engine power. Famously, BMW first broke the 19,000 RPM barrier with the P80 in 2002. 

Eventually, BMW Motorsport generated over 900 hp beyond 19,200 RPM in 2003 with the P83. The engine was more powerful, yet lighter, coming in at less than 200 pounds. 

More power, less weight meant heat management was more crucial than ever. The exhaust temperatures of up to 1,742F  were reached along with maximum air temperature of 482F.

A legend was born.

BMW's engine prowesses were symbolized by Montoya's lap record at Monza with his P84 engine. The previous year, BMW P83 endured the highest full-throttle proportion on the Monza circuit at 73 per cent per lap.

This time around, Montoya averaged speeds of 262.242 km/h (162.950 mph) across the Temple of Speed. The engine' performance was only to be bested by his teammate Antonio Pizzonia hitting 369.9 km/h (229.8 mph) on the back stretch the following day.

Montoya hit 372 the year after [...] in a Mclaren-Mercedes.

What could have been: the P85. 

The P85 was the last F1 V10 engine ever created by BMW Motorsport. The recent advancements in manufacturing processes enabled designers to lighten its alloy components. 

The cylinder spacing was reduced to 102 mm and bore was increased to 98 mm, giving a wall thickness between cylinders of just 4 mm.With a stroke of 39.75, the B:S ratio was a staggering 2.465

BMW Motorsport would have been sitting at the top alongside of Ferrari, the apex of the V10 food chain with its Tipo 053 engine. 

Unfortunately, it never raced. Regulation changes on engine life requirements made it too risky to run and BMW opted to revised the previous year's engine.

The P85 produced over 950hp beyond 19,500 RPM at 82 kilograms / 180.7 lbs. 

 

Auf Wiedersehen!

The V8s came around in 2006 and 18,000 rev limits were mandated. It was the end of the race to the moon, budgets were cut and we all wondered what if. 

To make matters worse, the Great Recession showed up, the new F1 regulations didn't correlate to road car development and the V10 CSL was shelved. 

BMW wasn't Ferrari, Formula One wasn't a core component of their brand. They sold the BMW-Sauber team back to Peter Sauber and that was it. 

At least we got to see our "homegrown" Jacques Villeneuve drive for BMW in 2006 - only to get canned by BMW that very same year. 

The heritage. 

The Landshut foundry.

BMW's Landshut foundry was built for the F1 program and manufactured the blocks for the P-series F1 powerplants.

The introduction of an improved aluminium sand casting process at the BMW foundry in Landshut, using sintered sand cores, provided greater latitude for the engine designers. F1 engine components could now be casted with wall thicknesses down to 2 mm. 

Similar processes were used to manufacture the S85's block at this very same location.  

The S85's cylinder crankcases were cast using low-pressure gravity casting methods and are made purely of aluminium-silicon alloy. This special alloy contains at least 17 percent silicon referred to as Alusil. The S65/S85s were the first M engines to be casted with this alloy. As a comparison, the S85's cylinder wall thickness is 6mm. 

These days, they make +/- 25,000 S engines per year in this facility. 

The 90 degree V, and its uneven firing order. 

The S85 was designed in similar fashion: the V10 arranged the two cylinder banks at an angle of 90° with a 17 mm offset to create a compact design. The 90° angle in a V10 is a direct descendant of the P8x engine series. 

They used learnings from the F1 Program to achieving the right balancing in an inherent unbalanced V10 design. The 90 degrees and offset were deemed as the equilibrium between limiting vibrations and maximizing stiffness of the block.

The uneven firing order is a fundamental component to the legendary exhaust sound. 

The equal length headers. 

The F1 V10 era pushed the designed of exhaust components to their limits: using thin walled inconel, engineers had to design systems that optimize for aero, heat, and exhaust pulses of ever more powerful, high revving engines. 

The S85's header design is a direct successor to the F1 program, it featured the first equal length header design to maximize performance.

On the V10, BMW M used stainless steel materials using hydroforming - eliminating needs for welding.The alloy's wall thickness was +/- 0.8 mm, optimizing for weight in the process. 

Unfortunately, BMW M fitted a primary catalytic converter right at the merge collector, limiting performance and incorporating design compromises in the primary tubes.

The V8 got the perfect set of F1 derived headers later on with the cats moved further downstream. 

The block's bedplate design. 

Th S85 was the first Sxx V engine to feature a bedplate design with integrated iron inserts as found on the E40/1E. While the F1 program ditched this design to favour aero, the E60 does not have such immediate concerns. 

The bedplate designs from Paul Rosche had a significant durability benefit: they were extremely stiff. It enabled precise fitment of the crankshaft to maintain main bearings clearance minimal over the entire operating temperature range.  

Clearances were tight - some say, too tight. 

The diamond like coatings.

During the F1 program, the mechanical production facility designed to deliver maximum precision was expanded to cover various surface treatment processes. 

It is this team that was responsible for the development of in-house DLC carbon coating technologies, allowing BMW to break up a supply monopoly. This expertise is now also used in the BMW production network.

The S85 made use of hydraulic tappers, reducing valvetrain weight, maintenance and improving performance. 

The oiling system. 

The F1 program required extreme oiling requires in 4G settings.  During extreme cornering centrifugal forces force the engine oil to the cylinder bank facing the outside of the bend, thereby preventing the natural return of oil from the cylinder head, which might lead to inadequate oil supply in the oil sump. 

While the dual VANOS had been developed by Paul Rosche, the S85 engine designers connected the VANOS high pressure oil pump driven off the crankshaft with the main oil pump to efficiently lubricate the engine. 

While the E60 barely hits a quarter of the F1 G forces, those two pumps weren't enough. BMW M designed two electrically-operated, continuously variable scavenger pumps bolted to each side of the main oil pan.

Contrary to popular belief, they don't push oil upwards, they do the opposite. They suck oil down the cylinder heads to the main oil sump if lateral acceleration rates exceed 0.6 g.

 

The engine management electronics. 

The F1 program required internal development of engine management hardware and software manufacturing capabilities. Notably, the KERS system was created by BMW in this very facility. This department has left its legacy, today, it produces new hybrid concepts for BMW production cars. 

The MSS65 DME used with the S85 was an in-house development and produced by Siemens. It marked a leapfrog in engine management with the ability to manage calculations more than 200 million calculations per second up to 9,000 RPMs. 

The Ionic knock-control technology used in the S85 was ground breaking for a street engine, enabling higher compression ratios due to precise combustion management. The by-product of higher combustion efficiencies were improved gas consumption (!?).  
See the board on the front section valve covers? This is one of two Ionic modules on the S85. 

It also marked the first widespread use of MAP engine protocols in BMW M engines, first seen in the CSL S54. I documented the evolution of protocols in the "Alpha V10" entry here.

Unfortunately, durability was a minimum threshold. 

In 2000, a new engine was used every day, on Friday, Saturday and Sunday of a race weekend. Back then, Formula 1 racing engines only had to travel 800 kilometres in one racing weekend. 

The V10s eventually evolved into a planned engine life up to 1,600km. When they V8s were introduced, 1,600km was the regulatory benchmark and eventually evolved to 2,000km by 2009. 

With the goal of achieving longer service life, the road going V10 had a reduced piston speed of approximately 20 metres per second while the BMW Williams F1 reached as much as 40 metres per second.

In 87 starts, BMW Motorsport suffered 38 retirements. 

 

The S85 had to last the lifetime of the vehicle, regardless of the climate, traffic situation and style of motoring - and we know how that went. 

Up next: I document every little changes and production updates made to the S85 by BMW M to the S85 in an attempt to improve reliability. 


Leave a comment

Please note, comments must be approved before they are published

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.


The latest entries to the Stripper build journal

CONNEXPLUS
I help you save money on everything you need for your V10.

ConnexPlus is a premium membership for the most demanding BMW M owners.

A select group of members will save on everything they will ever need for their BMW M cars at the world's biggest BMW parts retailers.