Spannerhead Dot Com

How far would you go to win one race?

Would you design an engine completely from scratch with the knowledge that it would almost certainly be banned by the next race, expending hundreds of thousands of dollars in development and production costs just to score that lone victory?

Mercedes and Ilmor would—and did. Their 500I 3.4l turbocharged, methanol-fueled V8 was a clean-sheet design constructed with a sole purpose: To exploit a loophole in the 1994 IndyCar rulebook and thus win that year’s Indianapolis 500. It could be argued that the nature of the race in question—the prestige of winning the Indy 500 being up there with the Monaco GP or Le Mans—made the resources committed to the project slightly more worth the investment. Still, the scale and audacity of the endeavor shocked the racing world.

Some background: The IndyCar technical regulations for the 1994 season allowed pushrod engines to run higher turbocharger boost pressures, as well as additional displacement, over their multi-valve counterparts as part of a kind of equivalency formula. IndyCar reasoned that some constructors would simply use non-optimized off-the-shelf production-based pushrod engine designs as a less-expensive starting point, and gave teams who chose to do so a leg up on their more affluent competition by allowing them the benefit of a few extra cubic inches and pounds of boost.

Prior to the 1991 season, engine manufacturers had indeed been limited, by the rulebook, to production-based engine blocks. After 1991, however, that requirement had been dropped, a development that went largely unnoticed—except by engine maker Ilmor. They figured that a clean-sheet, carefully-designed pushrod engine could take advantage of the regulatory loophole to overcome its inherent design disadvantages compared to a multi-valve engine, and thus trounce the competition for at least one race before the powers-that-be cracked down on the rulebook’s oversight. Ilmor placed their bets on the 1994 Indy 500, and before long Mercedes came on board to help finance and lend technical assistance to the top-secret project.

It worked. Mated to Penske’s all-conquering PC-23 chassis, the 500I’s extra 650cc and 2.5 psi of boost (as allowed by the regs) netted it an extra 150-200 horsepower compared to its multi-valve rivals, and Al Unser Jr won the race going away. Some sources quote a remarkable figure of over 1,000 hp over the duration of the 500-mile race. And naturally, caught with their pants down, as it were, the regulators banned the engine almost as soon as the checkered flag waved at Indy. Still—it’s to Ilmor’s and Mercedes’ credit that they had the creativity to consider an “inferior” engine design in light of the racing series’ rules, and the commitment to follow through with it, with spectacular results.

For a more in-depth look at the engine and the racing politics that surrounded it, check out this 8W article.

Editor’s note: This post is part of an ongoing series examining unique and significant powerplants. Read the other installments here:

• Saab Variable Compression Engine
• Rolls-Royce Goshawk
• Napier Nomad
• Mazda R26B
• Pratt & Whitney R-4360

Prompted by a recent online forum discussion wherein I explained the difference between BMW’s and Audi’s AWD powertrain layouts, I thought I’d put together a short rundown of front-engined AWD solutions.

I’m sure I’ve overlooked a configuration or two. If you think of one I haven’t covered, post a comment and I’ll add it to this primer!

Standard layout – The illustration above depicts what I’ll call the standard layout—basically a RWD setup with a transfer case attached to the rear of the transmission, sending power to the front wheels via a secondary driveshaft, front differential and axles. All four wheels must be able to turn at different speeds for dry pavement use, so the standard layout incorporates a center differential within the transfer case.

Advantages of this layout include the ability to position the engine further rearward for better weight distribution within the chassis, and the fact that it’s a relatively simple affair to convert a RWD car to AWD using a few extra parts. Among the downsides are additional complexity compared to the other methods described below, as well as a higher center of gravity since the front differential and axles must fit under the engine, and the transfer case and secondary driveshaft run beneath the car as well.

Automakers that use this layout: BMW, Mercedes, Nissan, Infiniti, Cadillac, etc.

Standard transverse layout – Similar to the first configuration, the standard transverse layout is a conversion of a “regular” 2WD (in this case FWD) powertrain setup to AWD. A transfer case, including a center diff, is attached to the rear of the transmission and sends power to the rear wheels through a (long) driveshaft and conventional rear diff and axles.

It’s a relatively compact setup, and commonly added to FWD cars. That said, the engine placement is still limited and the fact that so many components are concentrated near the front of the car means weight distribution is very often less than ideal. Also, the front-rear torque split is almost invariably biased toward the front wheels, only sending power to the rears when the front start to slip, making the car for all intents and purposes FWD, except in certain select circumstances.

Automakers that use this layout: Volkswagen, Mitsubishi, Mazda, etc.

All-in-trans layout – Subaru and Audi pioneered this layout in the ’70s and ’80s, respectively. Everything, including the front diff, is contained inside the transmission casing, and the front axles simply sprout from the sides of the bellhousing.

Both the upsides and disadvantages of the all-in-trans layout are significant. Its configuration is simpler and more compact than any other, and comparatively very robust. Also, having all the mechanicals “on the same plane” allows the car’s CG to be relatively low. All that said, a quick glance at the schematic above reveals the layout’s major downside: The engine must sit fully in front of the transmission and front axle line, utterly destroying any hope of reasonable weight distribution. To Audi’s credit, in the past 5 years or so, they’ve managed to rearrange components inside the transmission such that the front axle line can now move in front of the clutch, improving weight distribution slightly. Better—but they’re still a long way from a 50-50 front-rear balance.

Automakers that use this layout: Audi, Subaru, some Volkswagens (those built on platforms shared with higher-end Audis).

Both-ends layout – This one’s rather creative. It represents Ferrari’s solution to the “AWD problem” and was first introduced with their 2011 FF top-of-the-line shooting brake. Instead of siphoning power from the back of the engine to send forward, the FF’s V12 has a second transmission mounted to the front of the engine, without a differential, sending power to the front wheels via a pair of clutches. For its part, the actual transmission is mounted at the rear of the car in a classic performance car transaxle layout.

The boon to weight distribution from the both-ends layout is clear: The entire engine can, and indeed must, be pushed behind the front axle line. It’s relatively simple, mechanically. But the car’s wheelbase and styling proportions have to follow the layout, and can look rather stretched, as any consideration of the FF’s profile will attest to. Also, the lack of a front differential limits the power sent forward to only 20% of the engine’s output, not a great help in many situations.

Automakers that use this layout: Ferrari.

Editor’s note: This post is part of an ongoing series spotlighting obscure automotive engineering solutions. Read the other installments here:

• Opel’s Cam-In-Head Engine
• Nivomat Dampers
• The Desmodromic Valve
• Mercedes’ Monoblade Wiper
• Twincharging
• The Turbine Car
• The Sleeve Valve
• The Variable-Geometry Turbo
• The Hydraulic Cooling Fan
• The Laycock de Normanville Overdrive
• Audi’s UFO Brakes

I’ve been captivated by this thing ever since I was a kid. Is there any more fun car activity on a dreary, drippy Tuesday afternoon ride home from school than trying to find a mid-’80s Mercedes E-Class or 190E just to catch a glimpse of the wiper in action? There is? Well… I enjoyed it.

Attempting to achieve a cross between the coverage of two blades and the economy and aerodynamic advantages of a single blade, Mercedes’ Monoblade (also known as “eccentric clean sweep”) certainly wasn’t the first single wiper system on the market. Cars as varied as entry-level Fiat econoboxes, sporty VW Sciroccos and high-end Jaguar luxury sedans preceded it with their solitary blades. What made the German automaker’s wiper system unique was the hub mechanism used to increase the wiper’s coverage of the windshield beyond a simple arc.

As illustrated above in the “single arm (controlled)” drawing, the Monoblade’s substantial coverage was achieved by designing it to extend outward toward the corners of the windshield in the course of its travel across the glass. A cam-type device in the hub moved the arm away from the pivot twice, retracting it in between so the wiper would not overextend the top of the windshield. It all sounds ungainly, but to watch it in action is to witness a quasi-mesmerizing symphony of mechanical fluidity. Call it odd, but I’m transfixed whenever I see a Monoblade going through its paces on a rainy day. I nearly have to wrench my attention away from the spectacle in order to focus on the task of driving my car.

Introduced on the pioneering W201 190E model series and also fitted to the W124 and W210 E-Class generations as well as the W202 C-Class, the Monoblade’s primary advantage was aerodynamic, as it noticeably cut wind resistance at high (read: Autobahn) speeds. Even though it seemed more simple, with one linkage instead of two, as in a traditional wiper system, the Monoblade was actually more expensive to produce and repair, owing to the number of specialized parts. Additional disadvantages included the difficulty for owners in finding the required long replacement blades and a slight, but decidedly non-luxurious rocking motion introduced to the car when the big wiper was operating in high-speed mode.

Watch the clip below to see the wiper in action, and be on the lookout for an older C- or E-Class next time it rains. You’ll get a treat.

Editor’s note: This post is part of an ongoing series spotlighting obscure automotive engineering solutions. Read the other installments here:

• Opel’s Cam-In-Head Engine
• Nivomat Dampers
• The Desmodromic Valve
• Front-Engined AWD Systems
• Twincharging
• The Turbine Car
• The Sleeve Valve
• The Variable-Geometry Turbo
• The Hydraulic Cooling Fan
• The Laycock de Normanville Overdrive
• Audi’s UFO Brakes

Twincharging: Combining turbocharging and supercharging on the same engine. Blending of the low-rpm immediacy of a supercharger’s response with the high-end power and efficiency of a turbocharger.

Only a few automakers have attempted such a complex hybrid system in a production automobile. Continuing with the Lancia theme started with yesterday’s post, the Italian automaker was the first to offer a vehicle featuring twincharging for sale with the homologation version of their all-conquering Delta S4 Group B rally monster. The S4 Stradale, as it was called, was fitted with a detuned but otherwise intact version of the full-blown rally car’s mid-engined turbo- and supercharged 1.8l DOHC 4-cylinder. Downrated to “only” 250 hp for the Stradale, the powerplant was capable of cranking out upwards of 500 hp in race trim, and what’s more, able to deliver that power in a right-now, linear fashion ideal for low-grip rally competition.

As far as the mode of operation, there are a couple of way of doing it: Turbo first, or blower first. Lancia opted for the former, positioning the supercharger closest to the engine for immediate response, and incorporating a valve to allow the turbo to bypass the blower when up to speed and pressurize the intake charge with less restriction. The cost was obviously the complexity of the additional plumbing, drive belt, second intercooler and the like, but the system avoided the wasteful, manifold-destroying characteristics of contemporary anti-turbo-lag setups, which essentially dumped raw fuel immediately upstream of the turbine inlet in order to keep the turbo spooled between shifts.

More recently, Volkswagen has introduced a supercharger-first twincharging system on their 1.4TSI engine. Slightly less complex and more production-oriented than Lancia’s racing-focused effort, VW’s engine uses the blower’s output not only to boost the engine, but also to keep the turbo spooled at low rpm. For further reading on the 1.4TSI, check out this excellent Autozine article.

Editor’s note: This post is part of an ongoing series spotlighting obscure automotive engineering solutions. Read the other installments here:

• Opel’s Cam-In-Head Engine
• Nivomat Dampers
• The Desmodromic Valve
• Front-Engined AWD Systems
• Mercedes’ Monoblade Wiper
• The Turbine Car
• The Sleeve Valve
• The Variable-Geometry Turbo
• The Hydraulic Cooling Fan
• The Laycock de Normanville Overdrive
• Audi’s UFO Brakes

The Napier Nomad is unquestionably an incredibly complex piece of engineering. Whether it qualifies as brilliant or completely overwrought is more arguable.

Fresh off the success of the Sabre, most notably the powerplant for the V1 buzz bomb killer Hawker Tempest, Napier embarked on another ambitious engine development program, one that eventually led to the Nomad. First tested in 1949, the Nomad was Napier’s attempt to leverage their piston engine mastery and introduce a fuel-efficient alternative to the then-new jet engine technology.

At least with respect to the company’s goals of creating an economical engine, the Nomad succeeded. To this day, it remains one of the most efficient piston engines ever made when examined from the standpoint of specific fuel consumption, or fuel consumed per unit of power produced. That said, such economy came at a price: An almost Rube Goldberg-like level of complexity. At its core, the Nomad was a 41-liter, valveless, 2-stroke, horizontally-opposed 12-cylinder diesel engine. Around this, Napier added an array of combustion chambers, valves, couplings, clutches, turbines and compressors designed to capture every last bit of exhaust energy in service of turning the twin contra-rotating propellers bolted to the nose of the engine.

When it worked, it worked: Mounted in the nose of an Avro Shackleton testbed aircraft, the Nomad produced 3,000 hp with exceptionally low fuel consumption for its day. It’s a tribute to Napier’s expertise and persistence that they managed to get all its parts working harmoniously for 1,000 hours of testing.

Unfortunately, the Nomad, like other incredibly sophisticated piston engines of the day, was simply overtaken by progress—new jet engine technology promised much higher speeds with an exponentially lower level of engine complexity, and as for fuel consumption, it was farther down on most customers’ lists of priorities; the lean days of the early ’70s were still over 20 years away. The Nomad found no takers, and fell into obscurity.

Editor’s note: This post is part of an ongoing series examining unique and significant powerplants. Read the other installments here:

• Saab Variable Compression Engine
• Rolls-Royce Goshawk
• Mercedes-Ilmor 500I
• Mazda R26B
• Pratt & Whitney R-4360

I’ll say it: The American pushrod V8, and specifically the small-block Chevy (including its later incarnations, the Chevy LS- and LT-series) is the best all-around engine ever made. Period. The performance potential, the popular tuning expertise, the power-to-weight ratio of the latest iteration… Put it all together and you have the best engine ever. This position can be backed up by any number of car enthusiasts; venture to contest it and I’ll wager at least a half-dozen SBC buffs will produce numbers and stats reasserting its superiority.

This declaration may come as a mild surprise to some of you acquainted with my enthusiasm for imports and weird engine tech in general. But a clarification is in order: There’s more than one way to appreciate a powerplant, and the distinction to be made here rests on the difference between two of those ways: Respect and interest.

That said, do I respect the small-block Chevy V8 and its virtues? Absolutely. But does the engine actually interest me? Not in the least.

Put two vehicles side-by-side, say a Corvette Z06 with the factory performance catalog added and a Miata with a homebrew turbo system scavenged from various cars and run with a owner-programmed standalone EMS, and I’ll invariably be drawn to the latter. The Corvette may annihilate the Miata in every way, but how many times have the standard parts (cam, headers, etc) been added to an SBC in the pursuit of power? Yes, it may make a noise that amounts to catnip for most car buffs and spin the tires through third gear, and I have nothing but respect for that. It’s power and that’s always a draw. But let’s do something different. That’s where the interest lies.

I extend that lack of interest in the American pushrod V8 to swap projects involving the engine as well. There seems to be this especially virulent idea in many corners of the automotive world that dropping a pushrod V8 into any car will invariably make it better. Credit the late Carroll Shelby, perhaps, for the genesis of the notion; after all, he was the first to popularize it with his fusion of American muscle and British bodywork into the original Shelby Cobra. And to reiterate my earlier point: In most cases, I can’t argue the numbers—the V8 swap very often makes the resultant car more powerful, more reliable, cheaper to maintain and frequently even lighter and more economical. But is there anything more mind-numbingly boring than, say, a Jaguar XJ6 with an American pushrod V8 swap? Yes, it’s orders of magnitude more reliable, but the car’s distinctive character is gone, its engine note is completely out of sync with the car’s ethos and the whole idea is just so obvious that it loses all appeal.

Respect touches on those qualities I admire in a car: Power, handling and the like. Interest, on the other hand, has to do with qualities I’m drawn to in a scope wider than just the automotive realm: Innovation, creativity, new solutions to old problems that may at first be inferior to their elders but could potentially, with enough development, eclipse the traditional “best way” to do something.

I have a hunch more enthusiasts than just me view their automotive passion this way. If nothing else, maybe it will explain why a lot of car buffs persist in focusing on engine types and swap projects that seem inferior in so many ways to the tried-and-true-and-effective. It’s a matter of interest.

Fitted to the first- and second-generation (C1 and C2) Corvette V8s, Rochester fuel injection was GM’s first foray beyond carburetion.

Hard on the heels of Mercedes’ fuel injection experiments across the Atlantic in the mid-’50s, the Rochester mechanical fuel injection system was less a counterpart to the German automaker’s racing-focused program and more an effort to enhance low-speed driveability in their high-performance Corvette sports car.

One of the primary issues was hood height. Perched atop the intake manifold surmounting a tall small-block Chevy V8, a carburetor and its air cleaner assembly would fit under the low, sloping hood of a sports car only if the intake runners were relatively short. And while shorter, wider runners are ideal for airflow at high rpm, they’re not best for engine efficiency and cylinder filling at low engine speeds, compromising the low-rpm, off-the-line “punch” of the Corvette’s V8. Longer, narrower runners (and a consequently taller intake manifold) are desirable during the engine’s warm-up period too, when the manifold heats unevenly and has the potential to develop “hot spots” that affect individual cylinder operation.

GM’s solution to these conflicting requirements was to task one of their prime subcontactors—Rochester—with the development of a mechanical fuel injection system. First fitted to a ‘Vette in 1957, the resultant system wasn’t nearly as dependent a carb on intake runner length and profile. It solved all the issues, giving the engine a smooth warm-up sequence, muscular low-rpm behavior, a decent top end and allowed the use of a shorter intake manifold for low hood height.

Reminiscent of Bosch’s later K-Jetronic system, the Rochester’s injectors sprayed continuously into the intake ports upstream of the valves. The higher fuel pressure required over a carb was delivered by a cable-driven mechanical pump, and metering was controlled by a movable plate situated in the intake manifold whose position varied with airflow, and thus engine load. The plate’s movement sent a vacuum signal to the fuel pressure regulator, which controlled the amount of fuel flowing to the injectors.

It was a remarkably simple system, with only a few downsides, among them cost, as the lower production volumes of the Rochester fuel injection system couldn’t benefit from economies of scale. Also, fuel economy compared to contemporary carbs was poor (not that it really mattered much in the ’50s and ’60s) and all the components had to be adjusted with precision for the system to function properly, so it required some specialized knowledge and tools. Aftermarket hot-rodders found its ultimate performance potential poor as well, seeing as how there was an accepted body of knowledge about how to tune carbs for performance, and aftermarket injectors and high-flow intakes were unheard of during the Rochester’s heyday.

Overall, though, it’s a neat system, and its adoption way back in the late ’50s just goes to show that GM is frequently much more forward-thinking than I give them credit for.

Editor’s note: This post is part of an ongoing series highlighting various obsolescent methods of fuel delivery. Read the other installments here:

• Continuous Injection (K-Jetronic)
• Single-Point Injection
• The Dell’Orto carb
• The SU carb

Mazda’s 4-rotor, 2.6l, 700-hp R26B is the only engine by a Japanese manufacturer to win the 24 Hours of Le Mans race outright. In doing so, Mazda scored an achievement that has always eluded such pillars of the Japanese racing scene as Nissan, Toyota and Honda.

The year was 1991, and Mazda had something to prove. Perennially stung by criticism of their signature Wankel engine as an unreliable gas guzzler, the automaker had long sanctioned factory entries into endurance racing series across the globe. And though Mazda had achieved a remarkable amount of success through the years in that form of racing, Le Mans stood as the unconquered peak, the title that would perhaps finally demonstrate, to the racing world at least, that the rotary engine deserved to taken seriously from a competition standpoint.

The ultimate incarnation of a long series of endurance-focused rotaries, the R26B built on the foundation laid by its 4-rotor predecessor the 13J-M, itself a variation of the 3-rotor 20B, and added a number of refinements. At its core, the R26B was a basic non-turbocharged rotary engine, but with racing-derived features like intake ports on the periphery of the rotor housings instead of on the side plates (as in all production Mazda rotaries), an arrangement that produced a great deal of overlap between the intake and exhaust “stroke” of the rotor but which allowed for much greater airflow potential at high rpm, where racing engines live.

Also, the R26B was fitted with steplessly variable intake runners, able to optimize intake length and thus airflow seamlessly for any engine state, as well as 3 spark plugs per rotor instead of the usual 2, promoting more uniform burn of the fuel-air mixture. The engine was capable of cranking out 900 hp at upwards of 10,000 rpm, but was detuned to “only” 700 at 9,000 rpm to in the interests of durability.

And it worked. Fitted to a durable, proven 787B prototype chassis and driven by the trio of Johnny Herbert, Volker Weidler and Bertrand Gachot, the R26B vindicated Mazda’s efforts once and for all at Le Mans in 1991. Perhaps sweetest of all were the primary reasons for the win: Not power, where it was outclassed by the Jaguars and Mercedes running that year, but fuel economy and durability, two attributes which allowed the R26B-powered 787B to keeping lapping the circuit longer and shrug off failures that sidelined other teams. It’s an amazing engine, and Mazda is rightly proud of their success.

To see 1991 Le Mans winner Johnny Herbert driving his race-winning 787B just last year, click here. It’s a spine-tingling clip.

Editor’s note: This post is part of an ongoing series examining unique and significant powerplants. Read the other installments here:

• Saab Variable Compression Engine
• Rolls-Royce Goshawk
• Mercedes-Ilmor 500I
• Napier Nomad
• Pratt & Whitney R-4360

So yeah, it’s sideways. So what? At least there’s a brutally turbocharged 5-cylinder back under the hood of an Audi (the TT RS), where it belongs.

I give Audi a lot of credit for recognizing and acknowledging their engineering legacy, and I’d like to think that it was an intentional engineering decision; in other words, that it was more than just a happy coincidence that 5 cylinders ended up working better than 4 or 6, and Audi decided to capitalize on that fact from a marketing standpoint. Whatever the case, it’s wonderful to hear the distinctive rasp of the 5-cylinder firing order again. With the configuration’s reintroduction, the German automaker establishes a real, tangible connection with their legacy—it’s more than lip service paid to some abstract tuning philosophy; the engine layout that powered their all-conquering Group B rally cars is under the hood of a current offering, in honest-to-goodness cast iron and aluminum. It counts for a lot to be able to touch, hear and experience an automaker’s history in real time.

How incredible would it be if other automakers followed Audi’s lead in fitting a car or two with an engine configuration inspired by their manufacturers’ greatest hits? BMW could roll out a “reimagined” update of their classic “big block” (M30, M88, S38) six. Nissan and Toyota could reintroduce fresh takes on their legendary twin-turbo straight sixes (JZ, L, RB, etc). Volvo could develop a modern version of their vaunted Redblock turbo 4 from the 200- and 700/900-series. Porsche could fit at least a few of their myriad flavors of 911 with an actual air-cooled flat-6, an engine that left with the much-missed 993 generation. And finally, Mazda, whose signature engine, the rotary, has most recently been given the axe, could breathe new life into their stalled 16X program.

The sad fact is, though, that in our current regulatory climate, it’s doubtful the reintroduction of any of those configurations would be possible. Ballooning front crumple zones (as well as the marketing allure of the almighty V8) killed the big straight 6, and emissions and efficiency considerations spelled the end for the air-cooled mill and the rotary. Furthermore, part of the character of those classic powerplants is wrapped up in their lo-fi technology, like single overhead cams and distributor ignition systems. With the understanding that those certainly won’t be returning, how far would a prospective “throwback” engine have to go to maintain a connection with its inspiration? In Audi’s case, the recipe was simple—5 cylinders, between 2 and 2.5 liters of displacement, and a turbocharger—but it’s not as cut-and-dried with other automakers. It is fun to think about, say, a lighter, simpler post-F10 BMW M5 with a big, free-revving naturally-aspirated straight six under the hood, but with modern materials and styling. One can dream.