This all-American racer was the final winning car built for the US team in Dorset. It wasn’t the most successful Penske, but its explosive acceleration still exhilarates
By Andrew Frankel
It takes time for certain cars to assume their place in history. Cars that achieve firsts are easy to recognise and remember, for their importance is clear to see from the moment they achieve their fame. But what about those that provide the other bookend, marking the last time its manufacturer achieved a certain kind of success? Those are much more difficult to spot, yet, in terms of their significance, perhaps no less important.
Take this gorgeous Mercedes-Benz-powered Penske PC26 as an example. When it was new in 1997 it was a well-engineered but, by Penske’s stratospheric standards, only reasonably successful Indycar. It started the season strongly, with Paul Tracy coming second at Homestead before winning at Nazareth, Rio de Janeiro and St Louis. But thereafter it faded, leaving Tracy a mere fifth in the championship and his team-mate, Al Unser Jr, way down in 13th spot. Hard to spot the significance, isn’t it? In fact this is not only the car that notched up Penske’s 99th victory when Tracy won at Gateway, St Louis, but by the time victory number 100 was scored three seasons later, it was with a chassis built by Reynard. This, then, is the last Indycar manufactured by Penske to win a race. And by ‘this car’ I mean the very machine you see here, Penske PC26, chassis 005.
It shouldn’t seem strange to see it on an English airfield, for not only was it designed and built in Poole, its 2.65-litre V8 Mercedes engine is also English, being of Ilmor design and built in Brixworth – a matter of minutes down the road from where we are today. But it is strange: despite its origins you’d need something from NASCAR to find a car more quintessentially American than this. Today, it resides with Patrick Morgan, son of the late Paul Morgan, who with Mario Illien founded Ilmor back in 1984.
Morgan runs Dawn Treader Performance and if the name is not familiar it’s because the cars rebuilt, restored and prepared there are not designed to compete in the modern world of historic racing, but to be returned to as close to original condition as possible. They are pieces of history that, through the extraordinary skills and dedication of those who work at Dawn Treader, are allowed to exist in the present.
Morgan remembers the Penske arriving: “It was the most ambitious rebuild we’ve ever taken on. When it got here it was just the tub and bodywork with some scrap wishbones. There was no engine, no gearbox and no electronics.”
The PC26 is also a fiendishly complex racing car, particularly on the electronics side, with one particularly irksome complication thrown in that would not have troubled its engineers at the time. “Not only do you need 1990s computer software to run it,” says Morgan, “but hardware too.” Put another way, unless you have a mid-90s laptop running Windows 95, it won’t even deign to talk to you. Simply extracting the data recorded and sending it to this magazine so we could publish it took Patrick several hours’ hard work and head-scratching.
The car was designed by Nigel Bennett for the 1997 season, but was, in fact, the ultimate iteration of an earlier concept. “It really dates back to the PC21 of 1992,” says Nigel Beresford, technical director of Penske Cars at that time. “Nigel [Bennett] is an evolutionary designer, developing one concept over a number of seasons. So just as the PC17-PC20 can be seen as one series of cars, so can the PC21-PC26.”
In fact, while Bennett was the chief designer responsible for the chassis and aerodynamics of the car, as he was going into semi-retirement he had been joined at Penske that year by John Travis, who’d been employed to come up with a clean-sheet design for 1998 but also heavily influenced the design of the PC26’s rear suspension.
At first, it seemed Penske had performed a miracle with the car. Beresford: “We always liked to finish a new car early and have a pre-Christmas test at Phoenix to try and get our heads around its issues. But when we took the PC26 there, Paul Tracy went out in it, went super fast and came back saying, ‘there’s nothing you can do to improve this car – let’s start tyre testing’.”
In fact what was less understood at the time was that while the PC26 was indeed brilliant around short ovals like Phoenix, its talents did not extend to the other types of course on which Champ Cars competed.
“It actually had a very narrow performance envelope and all three of its victories came on short ovals. On superspeedways and road courses, it was nothing like so competitive,” says Beresford. It is possible that these deficiencies could have been addressed, but with Bennett ramping down his involvement and Travis concentrating on the clean-sheet PC27, the PC26 did not perhaps receive the refinement and development attention its raw talent deserved.
Looking at it today, none of this matters. Not only is it an extraordinarily beautiful car, prettier by far than Formula 1 machines of the era, but it is also built to a simply astonishing standard for a disposable race car with a life expectancy of substantially less than a year. Such is the finish of every component, from tiny joints to the carbon tub, that it becomes a thing of pure engineering beauty as well. Beresford parks the credit for this firmly at the door of Nick Goozee, then MD of Penske Cars, as well as Penske’s philosophy that everything had to be the best.
But it is one of the non-Penske parts – that Ilmor engine – that fascinates me most. Having never so much as sat in an Indycar before I find its specification as fascinating as the soon-to-be-realised prospect of actually driving the car to which it is attached.
As mentioned earlier, and as regulations mandated, the 82-degree V8 displaces 2.65 litres and is boosted by a single, massive turbocharger. Maximum boost pressure is regulated by a pop-off valve, but this ‘phase two’ version of the engine still produces 820bhp. In its day it would race with a limit of 14,400rpm and qualify at 14,800rpm. And Patrick Morgan has just apologised in all sincerity for the fact that, in the interest of preserving its longevity, today the limiter has been set at a mere 14,200rpm. Even so Patrick tells me it’s “an excellent engine, very strong and with a nice, wide power band. Really it’s as good as gold, terribly easy to drive”. Doesn’t look it from where I’m standing.
And I’m wondering if that nice, wide power band he refers to is race engineer speak. Before Dawn Treader, Patrick was an Ilmor engineer who followed this and many other engines through their careers in both Indycars and F1. Certainly it’s hard to see how the word ‘progressive’ is going to figure much in the vocabulary of such a small engine giving so much power thanks to such a big turbo.
There’s something else I need to factor in too: like all such Indycars, it runs on methanol. Much harder to ignite and therefore less dangerous than petrol, methanol is a fairly horrid substance, noted for its extreme corrosiveness (the PC26 engine has to be flushed through with petrol every time after it is fired up) and the fact that it burns with a colourless flame, making fire nigh-on impossible to see. But despite also being volumetrically less efficient than petrol it’s rather good at producing power, which is why, despite producing more than 300bhp per litre, this engine actually runs an extremely modest 1.35-bar boost pressure. There are road cars with more than that. This also accounts for its extraordinary 14.6:1 compression ratio, about double what you might expect from a turbo race engine. Fuel consumption? About 1.8mpg.
But if this engine guarantees the PC26 will be different to any other race car in my experience, it is the gearbox that’s likely to render useless all other racing reference points. It’s a Penske design but with British Xtrac internals and, insofar as it’s a sequential six-speeder, all is perhaps as you might expect. It’s the ratios it contains that so utterly divorce it from anything else. This PC26 has superspeedway gearing and these are the maximum speeds given for each of its six ratios: first goes to a little under 110mph, second to around 138mph. Third goes to 193mph, fourth to 224mph, fifth to 226mph. Sixth will take the car past 230mph. Clearly the first three gears simply get you up to speed but fourth, fifth and sixth are all flat-out gears: one for when running into a headwind, one for a tailwind and the last for slip-streaming. On a short airfield I’m clearly not even going to get out of third gear, but what’s worrying me more is how the hell to even get it moving. It has a tiny 5in sintered clutch, and if I burn that there’s not another.
At least it’s easy to climb in and out of. Indycars always tended to be built for the fuller frame compared to F1 cars, which seem to exclude all bar the snake-hipped from driving. You just step on the floor, support yourself with your hands, thread your legs through a carbon-fibre hoop and gently lower yourself down. As your backside hits terra firma, so your feet find the pedals. There are three as and where you’d expect. Ahead lies an LCD readout, most of which I can’t see, and numerous switches for various pitlane and fuel strategies, most of which I won’t need.
In fact you just flick on the ignition switches and wait for the car’s external starter to fire you up. Patrick has put a large insulation blanket over the rear wing to stop it and the man wielding the starter from being singed. Even through an Arai, balaclava and ear plugs, the commotion when the engine fires makes you want to run and hide. Indeed if I weren’t already firmly strapped into position, the flight reflex might have got the better of me. Even when your brain has become sufficiently accustomed to extract some sounds from the noise, it’s hard to find any you might call pretty. Deeper, uglier and more raw than a modern F1 engine, it is pure aggression on wheels.
My instruction from Patrick was under no circumstances to slip the clutch: he’d much rather I lit up the slicks, so his recommended strategy was to rev the brute up to 12,000rpm and simply side-step the clutch. As someone brought up on vintage cars, in which your chances of making it to your destination is directly proportional to your level of mechanical sympathy, the very thought horrified me. But it was also almost impossible to do: not only could I not see the rev-counter, but the engine was the most trigger-happy I’ve ever encountered: one tremble of the toe seemed to add about 2000rpm to the score. I couldn’t even park it on the rev-limiter and drop the clutch from there as Patrick seemed only marginally less keen on that idea than slipping it from lower revs. So in the end I just guessed.
And guessed right, if you can call the fact I didn’t stall a good guess. It was probably the least dignified getaway an Indycar has had to endure, but at least we kept moving as the engine stayed above 8000rpm, the speed below which it had no interest in working at all.
There was no time to wonder what might happen next. When I opened its eight butterfly throttles, there was no turbo-lag I could discern at all, just pulverising, monumental thrust.
I changed gear almost immediately, not because another was required but because I’d figured that second could not possibly be as violent as first. Turned out I was right, but only just. In fact what second seemed to provide was just a bit more time for the assault to take place. Only when I hit third did the attack abate sufficiently for me to take in what was going on.
The big surprise was how much strength you needed just to operate the Penske. The steering has no assistance and even just turning around at the end of the runway required discernible effort: the thought of having to turn slightly left four times a lap at 230mph gave me new respect for Indycar drivers. Similarly the gearshift requires Popeye-grade forearms; I’m not sure why but I’d expected it to flick through its ratios with barely a nudge, but in fact it felt like you were picking up each cog and physically marrying it to its neighbour.
After a few more runs, during which I acclimatised just a little to what this car could do, I stopped for a chat with Patrick. “Is everything OK?” he asked, clearly concerned. “Er fine,” I replied, “why do you ask?” “Only because from what I can hear you don’t seem to be using very many revs.”
True, I’d not been able to see the rev-counter but I still thought I’d been revving the valves off it. So by adjusting the padding, wriggling down further into the cockpit and bending my knees some more, I afforded myself a glimpse of at least part of the rev-counter and set off again. Once more I short-shifted into second but then I let it go, realising I’d hitherto been shifting at least 2000rpm too early. And only then did the Penske let rip. The forces at work were so mighty it seemed fanciful that a mere mortal could exercise control over them. I felt like a passenger, as if it were coincidence that the car reacted to my stammering inputs, or at least that in doing so it was merely humouring me.
On the next run I finally summoned the courage to plant my foot in first, and despite the three-figure potential of the gear and those massive rear slicks, at 60mph they spun like a Corvette’s leaving the lights. The data explains just how fast it is: the Noble M600 has a better power-to-weight ratio than a Bugatti Veyron and accelerates from 60-160mph in 12.9sec. Yet despite the wheelspin, the ultra-long gears, more than 1.5 tonnes of downforce at 160mph and all the attendant drag, the Penske added the same 100mph in 7.5sec.
Never again will I look at one of these cars as some kind of poor relation to an F1 machine. Yes, it has steel brakes and wishbones and a manual transmission because that is what the rules mandated, but whether you look at it as a thing of beauty, an engineering masterclass or just an explosively powerful and ferocious racing car, this Penske PC26 is simply exceptional. The fact that it is also a car of such great significance means it’s an important part of motor racing history too.
Our thanks to Patrick Morgan and all at Dawn Treader for making this feature possible