Tech Analysis: BMW E36 Exhaust Testing

When the E36 M3 was released to the U.S. in 1995, tuners throughout the country went wild. Cams, chips, intakes, brakes, suspension systems and powerful forced induction kits made the M3 as popular with the go-faster crowd as another tuner favorite, the vaunted 911.

One of the more popular and easiest upgrades for the E36 M3 has been the cat-back exhaust. Why? They're street legal, sound cool, look better, weigh less, improve throttle response and, sometimes, even make power.

We rounded up 10 premium exhaust systems and over the course of two days tested build quality, fitment, sonic performance and horsepower against a stock E36 M3 and also against an uncorked M3. The tested M3 was a 1999 3.2-liter model, the perfect candidate: It already had an ECIS intake system and Conforti reprogrammed ECU, plus, some 75% of all E36 M3s are 3.2s (1996 to 2000), a better representation than the 1995 3.0-liter model.

We chose a car with an upgraded intake system, because most M3 owners looking for exhaust systems usually end up with an intake system as well--it's practically the best horsepower bang for the buck on an M3. And exhaust gains are usually realized with less restriction on the intake side.

Rather than using european car's floor-mounted dyno, we contacted evosport to use its lifted Dynojet dyno for faster exhaust installations. Thanks to Evosport technicians Josh Rickards and Steve Lee, everything went smoothly.

We used evosport's Snap-On Graphing Scanner tool to monitor the car's temperatures. In order to provide as much fairness as possible, the car's cooling system was allowed to cool down to between 190- to 200° F before each dyno run. The oil and intake charge temperatures were also monitored to make sure no exhaust had unfair advantages due to any possible heat soak.

Downs Oil Company, a major distributor for Sunoco Race Fuels, was contacted for a supply of 100-octane race gas. Sunoco's GT Unleaded race gas has a much higher octane rating than 91 pump piss, which makes it capable of withstanding more heat before pre-igniting. Some of you may be wondering why we would try and skew the numbers with race gas rather than pump gas. In reality, with a car like this, race gas does not increase power but allows the car to make its maximum power throughout the rev range without pinging. So what we do here is kill two birds with one stone: The car is less susceptible to pinging during multiple hard dyno runs--for our and the owner's peace of mind--and we also get a smooth dyno curve, eliminating power dips from knock-sensor actions which could unfairly alter the results. This gas basically levels the playing field in all aspects.

Important Things to Consider
Power gains and sounds from a brand new exhaust may vary slightly from an exhaust that's been used several months. All of the exhausts tested were brand new and had never been previously installed or broken in.

The first few exhausts were tested in the morning, giving them a slight advantage because of cooler weather--and they were tested in the order shown here. The SAE correction on the Dynojet 248C dynamometer works well, but it is not perfect. For this reason, another baseline run was done at the end of the day to provide a fairer comparison against the exhausts tested in the afternoon. This new baseline number yielded a loss in peak hp and torque measurements of 2.0 and 1.2, respectively, from the initial baseline run done in the morning. And by the end of the day the UUC exhaust dropped from its peak 219.0 hp to 216.7 hp due to the increase in ambient temperature, even with SAE correction incorporated. With this in mind, all of the exhausts performed within about 1 percent of each other.

Whether that's a small enough margin of error to say that all of these exhausts are pretty much the same in performance is hard to say. Some people may have been hoping for similar results to what was seen in our Jetta 1.8T test (ec, 11/01), but we knew that wasn't going to happen, because a turbocharged car will respond much differently, and often more positively to exhaust upgrades, than a normally aspirated car such as this M3.

Choosing a Favorite
We left the task of choosing a favorite exhaust to our test car's owner, and by no means was it a simple task for him. He bounced back and forth between four exhausts before he finally made his decision. After a long hour of going through the exhausts and all of the objective and subjective ratings, he finally chose the single-canister UUC System U. "I like the way the tips look and the way the car sounds with this exhaust," he said, "plus I don't mind the fact that it's light and put out more horsepower than almost all the rest."

UUC is a company well known for its drivetrain upgrades and exhausts for German cars. We used one of its short-shift kits and transmission braces on our Project S4 and loved them (ec, 9/02).

About a month after the install, we asked the owner how he was doing with the new exhaust on the car. He replied, "The sound is awesome under full-throttle acceleration, especially when it bounces off the concrete walls of the California Speedway!"

We concur--we were there to hear it from the pit lane, and it sure sounded sweet. "But for my taste," he added, "although I am getting more used to it, I still wish it would be just a little quieter for cruising around town. Also, the fit has become a bit of an issue, since one of the tips has seemed to have melted a small part of the rear valance." The exhaust is a prototype version and hopefully UUC will have this sorted out by the time you read this.

Every exhaust here had its pros and cons, but at the same time none of them made substantial peak horsepower. If there were any power gains, most of them were achieved in the midrange, save for the Rogue unit, which found peak gains up top.

The UUC exhausts were light, sounded great and, although the numbers were minimal, made some of the best horsepower gains of all. But the fit was an issue for both of them. The B&B exhaust was very light, made power, had nice staggered tips, and sounded wonderful at idle. But the sound became noisy and a little less desirable at full throttle. The Ansa exhaust didn't show power gains but sported the factory look with much nicer tips and a deeper rumble. It was a perfect fit, too, retaining the factory heat shield and hangers. The HIOP was light and a very easy install. It had the biggest gains around a very usable 5200 rpm and maintained similar gains to those of the UUC System U all the way through redline. But it also had the biggest torque losses up until 3500 rpm. The HKS exhaust was the lightest and the easiest to fit, but the sound was one of the least desirable of the bunch. Also, some people will love the massive, single oval tip--some people will hate it. The Rogue exhaust was gorgeous, fit well, and it probably could have taken the cake if it were a little bit lighter. The Racing Dynamics exhaust showed minimal horsepower improvements and was difficult to align, but it was light and the build quality was superb. Lastly, if there was one exhaust that had the highest and most consistent gains from 4000 to 6000 rpm, it was the Supersprint DTM exhaust. Unfortunately, this also turned out to be by far the heaviest exhaust of the bunch, giving up more than 12 lb to the lightweight HKS.

Now, go ahead and take your pick.

Special thanks to evosport for the use of its dyno and the installation of 10 exhaust systems.