Dyno Test: BMW 335i and 335is

A few years ago, we were among the first to get a brand-new BMW 335i on the dynamometer. We don’t often dyno cars, but we couldn’t resist. Before the first 335i was delivered to a paying customer, the rumor mills were bubbling about the 335i’s twin-turbo straight-six being underrated.

It turns out those rumors may have been right. The 2007 BMW 335i’s N54 twin-turbo six is rated by its manufacturer at 300 hp and 300 lb-ft of torque. And when we put the first 335i coupe on the rollers, it laid down a very impressive 275 hp and 300 lb-ft of torque. In awful conditions, too: 92 degrees of humid, Fahrenheit heat.

Remember — the power measured at the 335i’s rear wheels should be lower than the engine’s rated power. Some of the engine’s output is lost to friction in the transmission, driveshaft, differential, axles, wheel bearings, and tires. The dynamometer we used to measure the 335i’s output is a DynoJet model, and DynoJets tend to read higher than other dynos, but the output was a good bit higher than we expected.

So a few weeks later, when a second 335i coupe came to our office, I dragged it right to the dyno shop to find out whether that first car’s output had been an anomaly. It wasn’t. The second car put down an equally impressive 282 hp and 285 lb-ft.

One thing to remember-always, when looking at dyno numbers-is that the peak numbers tell only part of the story. Even though these two motors had slightly different peak outputs, their curves looked largely the same. So the difference between the two is easily merely just normal variation from engine to engine.

That the N54 is a consistent engine in testing was highlighted three years later when I tested a 135i coupe on the dyno. That engine’s output was right in that same brawny ballpark: 276 hp and 285 lb-ft of torque.

The 135i used a later version of the N54 that includes a slightly different wastegate design. The only real difference between it and the earlier cars was the turbo’s response at low revs (look at the torque curve at the beginning of the test to see the late N54 produce more torque, earlier.)

Another important thing to note is that we chose to publish each engine’s LOWEST output to ensure we didn’t publish a nonrepeatable number. The first N54’s actual power peaks were 276, 276, 280, and 283 hp. The second’s were 285, 282, and 287 hp. The third’s were 276 and 277 hp. Such minimal variation from run to run shows that the N54 is a very consistent engine in addition to being a very powerful one.

So it’s no wonder we jumped at the opportunity to be the first to publish dyno results for a manual-transmission 2011 335i coupe. (We’ve seen dyno results for a 335i with a twin-clutch transmission, but for a true apples-to-apples comparison, we wanted a car with the same transmission. The DCT could experience greater loss due to its hydraulic pump and different gear ratios.)

The new engine, the N55, is different from the N54 in that it uses only one larger turbocharger in place of the N54’s two smaller turbos. To keep lag to a minimum, BMW used a twin-scroll turbo and its Valvetronic throttle-less, variable-valve-lift system. On the street, after driving N54 and N55 cars back to back, it’s definitely safe to say that the N55 suffers from even less turbo lag than the N54 did. And that engine set the benchmark.

BMW’s quoted output for the N55 remains 300 hp and 300 lb-ft, which raises a question: what if the N55 actually does only make 300 hp? And, given the simultaneous appearance of a 335is (which uses an even more potent, 320-hp version of the N54) we couldn’t help but wonder: what if the 335is didn’t actually make any more power than the old, underrated 335i?

There’s only one way to find out, right? Call the dyno shop.

To answer the questions quickly: BMW isn’t hiding anything. The 335is put down 293 hp and 343 lb-ft, beating every previous N54 we’ve tested by a big margin. And if you look at the torque curve, the midrange overboost function BMW promised (rated at 369 lb-ft) is definitely alive and well. The 335is engine is, like the original N54, probably still underrated.

And now to the N55: well, it doesn’t, in fact, make as much power as the others N54s we’ve tested, but it does make slightly more torque. Peak numbers were 266 hp and 302 lb-ft. Before you start typing your “N55 sucks!” forum posts, let’s remember that while the N55 put down 10 less horses than two of the N54s, a third N54 produced 10 more horsepower than the others. Meaning: 10 hp is within the noise when you’re at this power level — it’s a variation of less than 4%. And you can easily see 4% variance between runs on the same engine.

(Case in point, while the N54 is a supremely consistent engine on the dyno — especially for a turbocharged engine — the variation between the best run and the weakest run on the three N54s we tested ranged from 1.0 to 2.6 percent. That’s between runs done on the same dyno, on the same day, with the same exact car, in the same conditions. These engines were tested on four different cars in two different states in temperatures ranging by 30’F over four years and three of them were within 4%. Okay?)

[Obviously, the N54 “is” engine is vastly more powerful in the midrange and slightly more so up top.]

In fact, the only real difference between the regular N54s and their successor, the N55, is that the N55’s output drops off more as it approaches its maximum speed. This isn’t a surprise-the N55 has only one turbo with which to force-feed the engine with air. To keep turbo lag at a minimum, BMW likely used the smallest turbocharger possible that could achieve the (modest) boost levels that the N55 runs on. A slightly larger turbo might have increased high-rpm boost (and thus, power) slightly, but it’d come at the expense of greater lag. And besides, we’re talking about an engine that over-delivered to begin with.

And remember, only a fool looks at peak numbers. It’s the shape of the curve — and the area under it — that really determines how fast a car is and how fast it feels in everyday driving. Looking at the graphs, it’s immediately clear that there’s not much of a difference between the regular N54s and the N55.

The end result? We can finally and directly compare an early N54, late N54, high-output N54, and an N55. The N55 exhibits less turbo lag in normal street driving, and even though its peak output was near the bottom of the group of engines we’ve tested, it’s still well within its rated peak horsepower output — and well above its rated peak torque output. Let the N54 vs N55 war continue!

Previous N54 Dyno Features:
2007 BMW 335i Dynamometer Run

2007 BMW 335i Dynamometer Revisited: We Test Another One!