A good, quantifiable value is a good selling point. It’s an even better selling point if few people know what it actually means, so they can’t really challenge it or compare it.
If you follow Formula 1 these days, you will hear a lot, an awful lot, about ‘aero’. Assuming the drivers don’t all have a fixation on the bubbly, chocolate snack, we can assume this means that the aerodynamics of the overpriced racing cars are very important. They are important for road cars too but, oddly, nowadays manufacturers don’t make a big deal of it in their marketing, leaving you to guess from the often excessively racetrack mimicked shapes of splitters, spoilers and diffusers we see on so many cars.
The fact is that today’s cars are generally far better behaved aerodynamically than their predecessors, but you can’t really quantify general behaviour.
The drag coefficient of a car is an indication of how little air resistance it encounters whilst travelling in a straight line – the lower the figure the better. In the early 1980s Audi were positioning themselves as the cooly, technical company and they weren’t shy in their marketing. The smooth looking 1982 C3 model 100, with its novel flush side glazing, was released with its then class-leading Cd of 0.30 emblazoned on its rear side window.
It’s not exactly the first time the Cd was used as a marketing idea. The 1974 Citroen CX model name paid service to an alternative expression of the value. Its Cx, or Cd, was actually 0.36, good but not fantastic, being the same as the deceptively less sleek looking Mercedes W126 S Class of 1979.
But drag just tells you how smoothly air flows over the surface of the car. It does nothing for stability and there’s much more to aerodynamics than a slippery shape, as all the tortuous little features on today’s Formula 1 cars testify. The Ford Sierra, introduced the same year as the Audi, also featured a low drag shape with a Cd of 0.34, though not as slippery as the Probe III concept that preceded it. But the car was hastily modified with a couple of small side spoilers on the C pillar, after reports of instability at speed.
The Audi just had a good Cd for a saloon. The 1970 Citroen SM’s Cd was 0.26 (note : see comments below) which, apart from the car’s obvious shape, was helped by paying equal attention to the car’s underside and having a full undertray to the engine bay. Since then designers have paid a lot more attention to detail, helped in a large part by improved manufacturing techniques.
So the bulbous looking 2005 W221 Mercedes S Class also had a Cd of 0.26 and we now find a good number of current models in the 0.24 to 0.28 range, with EV’s making more effort than most, helped of course by not needing to gulp a load of air into a radiator.
Over 30 years ago, the W126 Mercedes, the Sierra and the C3 Audi 100 made drag figures fashionable for a while, with journalists trying to tease out of manufacturers the Cd of any vaguely fast looking new model likely to break the Cd 0.29 barrier but, fickle as we are, interest soon moved on to other things.
Driven To Write 
My neighbour bought a Renault Fuego 1981 and he was very proud of the low cd of his car – 0,34 was indeed pretty good in the early eighties. I was more impressed by the (spectacular) seats and the nice interior, but the Fuego was to my eyes exactly 10 times more beautiful than his previous car – the Renault 14. But the Renault Fuego was the first streamlined car that was part of my car live.
I always prefer streamlined curvy cars to edged cars – more Pininfarina and less Bertone.
The Audi 100 of 1982 was always a soap box car to my eyes.
In contrast to cars like the Panhard CD – what a curvy female erotic shape !
p.s. That Panhard is a beauty; I’ve never seen one, I’ll be reading up on it now.
I am not an expert in physics, but as far as I know, a number for the drag coefficient tells only half of the story of the drag of a particular object. But then, there are probably good reasons why car manufacturers never made a big deal out of face/surface area values. And in the face of the crossover/SUV paradigm this is unlikely to change anytime soon.
Exactly what I was thinking. I suspect they have stopped publishing drag coefficients and frontal areas because it would show up the fiction of their mpg and CO2 claims. I have noticed that some of the recent figures are totally fanciful; for example, without being specific, you might find a small hatchback with a 1.6 TD engine managed 55 mpg 5 years ago while the same engine in some hulking MPV or SUV now will magically manage 70 mpg, despite no obvious progress in engine design and the greater weight, frontal area and probable higher cd value.
A drag coefficient is just that – a figure which, when it multiples the frontal area, gives the actual aerodynamic drag of an object. Most of the good work in the wind tunnel is these days is overbalanced by the greater width and height of modern vehicles, although the impressive top speeds achievable by quite ordinary cars suggests that actual drag is reducing.
Most of the clever stuff is below and out of sight. My newest car’s first MOT came with a rather haughty observation that “Undertrays are fitted obscuring some underside components”. It’s true – just about everything’s hidden under a massive plastic bellypan, and this seems to be normal industry practice.
And Sean, EVs do have cooling systems. Even our humble Mitsubishi i-MiEV campus runabouts have a cooling system for the traction motor very similar to that of an IC engine, and also air cooling for the battery pack. Probably nothing like the cooling load for a V8 or an Old Testament diesel, but the radiator looks about the same size as I’d expect to find on the equivalent petrol car.
The Leaves which are taking their place have water cooling to the traction motor and inverter, and some sort of battery cooling arrangement involving the air-con. Impressive machines, I call them the Poor Man’s Teslas. My bovine colleagues aren’t convinced. They avoid them like the plague, as they have neither a clutch, gearlever, nor a Vauxhall or Ford badge.
You’re not going to get movement without things getting hot but EV’s have the advantage of greater efficiency and, also, that all the cooling isn’t generally confined to just one place. Hence, no need for a gaping maw at the front of Teslas, even if they choose, disappointingly, to mimic a rad grille.
The good thing is, the grille has already disappeared on the model X, and it will do so on the others, too.
Am I the only one who thinks that the grilles on fossile burners could be smaller as well, if the designers would like so? Or at least better hidden below the bumpers?
Simon, VW tried that with a Passat more than 20 years ago, and the customers didn’t like it. Hence Tesla’s initial reluctance to do without a Trompe-l’œil (I copy-and-pasted that for you, Laurent!) grille.
The gargantuan size of today’s upper class grilles is mainly due to Chinese aesthetic preferences, hence the BMW Seven’s rabbit teeth and the S-class’ chromed maw. I fear the giant-sized grill will stay with us for a while, maybe even until EVs take over all of automobilekind.
I liked the unadorned Passat. So did Richard I believe. But apart from us two, you’re right, it didn’t make owners feel important enough. Also, I liked the 1975 Vauxhall Cavalier as a grille-less version of the otherwise identical grilled Opel Ascona.
@Kris
I appreciate the effort.
(that said the capital T and hyphen are unnecessary
http://dictionary.cambridge.org/dictionary/english/trompe-l-oeil?q=trompe+l%27oeil
)
Thanks for an excellent article Sean. Another car i believe deserves to be discussed in this context is the eminently likable Mercedes W124, in spite of its rather angular design, the base model had a Cx of 0.29, the careful shaping being particularly noticeable when viewed from above.
Also, was the SM really that good? I raise the question as the only source i could find quotes a Cx of 0.339 for the SM. ( http://www.citroenet.org.uk/miscellaneous/aero/aero01.html )
Roberto. You know, I’ve been wondering the same ever since I wrote the piece. I chose the SM, not for reasons of bias, but because I wanted to find something old with a streamlined shape that has the same Cd as something newer and ostensibly less slippery. My source was a (too?) convenient list on Wikipedia.
https://en.wikipedia.org/wiki/Automobile_drag_coefficient
As it says ‘citations needed’.
The SM’s smooth underside and the rear horizontal taper (another way of achieving what the W124 did) would certainly have helped, but it does seem extraordinarily low for the period. Though not as low as the sub 0.22 claimed for the Tatra 77 in the same list.
I also doubt that low figure for the SM.
Another example to be cited is the Citroën GS. I don’t know what the Cd was in its initioal form in 1970, but with the 1979 facelift, becoming the GSA, it’s claimed to have something around 0.30 or 0.31. I don’t know if there are any reliable sources for that, though.
The Citroen AX has a cd of only 0,31, which was really good for a small car in the eighties. With 60 horsepowers (i had this engine) it runs 168 km/h, thanks to his small and low shape.
For example a Citroen C2 was not faster than 158 km/h with the same engine and the Citroen C1 (2005) was even slower with 68 horspowers (157 km/h)….
Streamlined cars are often pretty unagressive because they can not afford a dominant grille and fat tyres. Maybe that is why they are not so hot at the moment. Look at the Audi A2 – with a cd of 0,25 it is still better than for example an Opel Ampera (0,27) or a BMW I8 (0,26 – which is really disappointing to me).
But no one wants to build a new A2 again – first of all Audi…
That 0.25 A2 figure is ONLY for the 1,2 “three litre” version that had very narrow tyres and a longer spoiler out back. The normal ones were 0.29. Now even something as mundane (looking) as a C-class Mercedes sedan manages 0.24.
Lowest of course is the VW XL1 at 0.189.
It’s likely that many contemporary Cds are theoretical, being calculated by computer simulations, rather than in actual wind tunnels, though that doesn’t necessarily mean they are less accurate. There is obviously a fixed formula for calculating Cd, but how you measure the components in a real (or virtual) wind tunnel may vary. Also, do you measure the actual vehicle, with door handles and all those other drag inducing bits, or just the basic shape? My point really was that there are so many variables that it was a perfect figure to use for marketing, since no-one was likely to go to the bother of disproving it. The downside, of course, is that your competitors can come up with their own figures that you might find it hard to disprove.
Form a further trawl of the internet, I think I can say with complete authority, that the SM’s Cd lies exactly somewhere in the vicinity of between approximately, give-or-take 0.26 to 0.46. I think I’m going to have to take this up with Jimmy Wales – I didn’t pay my last contribution for this level of indecisiveness.
Hold on. Correction. Martin Buckley in an EVO article mentioned 0.25.
Yesterday, I altered the Wikipedia entry for the SM to the 0.339 Roberto mentions, which I consider a far more credible figure.