In et 9/06 we introduced our '97 E36 M3 Sedan project car and looked at the impact on horsepower and acceleration caused by fitting lightweight wheels. This month, we start to seek more power from its 3.2 liter S52 motor with a popular intake manifold swap.
To give you some background on what we're doing, you should know that in 1996, BMW sought to increase the responsiveness of the E36 M3 by increasing displacement from 3.0 to 3.2 liters. They also changed the final drive ratio from 3.15 to 3.23:1 and added a smaller, choked-down intake manifold.
EHP's manifold conversion kit comprises a CNC-machined adapter that allows the OBD2 vacuum
The first two mods were nice upgrades with little compromise. However, the intake manifold changes increased the air velocity into the motor but only up to 4500rpm, and did so at the cost of reduced airflow. The result was a weaker top end on the later 3.2 engines. That's why, even though the OBD2 3.2 liter S52 motor in the '96-99 M3 made more torque than the '95 OBD1 3.0 S50 it replaced, peak power remained unchanged.
The same applies to the 328i that replaced the 325i in '96. Those engines are dubbed M52 and M50, respectively, and both use the same intake manifold as the M3 of the same year. Confusingly, the names of the manifolds on the S50/S52 M3 engines retained their respective M50/M52 codenames inside BMW.
The M50 manifold from the 325i or the '95 M3 is often used to upgrade the 328i or M3 3.2 engine. For a while, owners of the M3 3.2 retrofittted an OBD1 ECU along with the M50 manifold in order to run the larger 3.5" HFM (Hot Film Mass, or air mass sensor), as found in the Euro-spec M3. However, this was a costly and illegal procedure. As a result, the OBD1 '95 M3 3.0 became a more popular car for tuning.
EHP's stage 2.5 kit includes the Shark Injector, M50 manifold kit, 3.5" HFM and intake kit
Fortunately, things are now different for owners of the OBD2 M3 3.2, thanks to companies like Eurosport High Performance (EHP). Like many BMW specialists, it has various normally-aspirated upgrades. Its stage 1 kit, for example, includes an injection-molded carbon-fiber Evo2 intake kit plus software that removes the speed governor and sets the redline to 7000rpm. Its retails for about $738 and a peak gain of 15whp is possible, with gains of around 10whp in the upper reaches of the rev range.
With the software's new rev limit, you can hold the car in lower gears for longer. However, you get a significant drop of about 25whp between 6500-7000rpm when using the stock M52 manifold. This proves BMW was wise to set the rev limiter at 6500rpm, as shown in our baseline run (see graph).
In order to fit the fuel rail, the rear fuel lines have to be bent outward and around runn
So the EHP stage 2 kit adds an M50 intake manifold conversion kit to stage 1. It includes an adapter that allows the idle air control valve and all vacuum lines that go into the bottom of the M52 manifold to be fitted to the M50 manifold. It retails for $349 (plus the price of a used manifold - about $100 on ebay - and an optional $30 modified fuel rail cover).
We previously tested a stage 2 kit on a different car and it lost a significant amount of torque (about 8-15 lb/ft over stock) from 3200-4200rpm. However, there were gains over 5000rpm, showing a peak of 30whp at 7000rpm over the stage 1 kit. So now it would definitely pay to hold the car in-gear for longer.
The EHP stage 3 further ramps up the power (and cost) with everything in stages 1 and 2, in addition to a set of Schrick cams, a re-calibrated 3.5" HFM with an adapter to fit EHP's intake kit, as well as 24 lb/hr Bosch injectors, a fan delete nut, the relevant software and an optional traction control delete elbow.
The entire kit retails at $3396, including the first two stages. A peak gain of 35whp at 6400rpm over stock can be realized. And if you compare stage 3 to stage 1 at 7000rpm, the gains are over 45whp because stage 1 retains the stock M52 manifold that loses power over 6500rpm.
For our project car, we wanted a significant increase throughout the rev range without the cost of the full stage 3 kit. We also wanted more power than stage 1 could provide. So stage 2 appeared to be the best option, but only if we could somehow overcome the loss of low-end torque.
Our stock 3.2 liter S52 engine
EHP's instructions are so detailed that the throttle body removal occurs at Step 21
By Step 35, the intake manifold comes off. It's a good idea to label the connectors so you
After a chat with Josh MacMurray at EHP, we decided to try stage 3 without the expensive cams. We'd still use the same software (which would give us a bit more fuel than needed up top) but our main concern was idle quality. However, MacMurray was confident it would work. And at $2297, we'd save over a grand plus the six-hour cam install. And in the future, we could upgrade to stage 3 cams if we wished.
The first order of business was to get a new baseline dyno run. Our S52 motor was stock, with the exception of a K&N drop-in filter and it spun Imagine Auto's Mustang MD500 dyno to 221whp and 208 lb/ft in third gear.
We used an OBD2 scanner tool to monitor intake air temps for accuracy. It can show if a ca
Using Imagine's OBD2 scanner, we kept track of the ambient and intake air temps (IAT), which were 91 and 110 F, respectively.
That afternoon we got to work on our EHP stage 2.5 upgrade, which included detailed instructions on a CD covering every single bolt. It was our first time swapping intake manifolds, so it took a while. Fortunately, it's not complicated and EHP's instructions are fantastic.
Once the manifold was fitted, I inserted the software injector into the ECU port and fired her up.
After a fairly long install (it took me a total of eight hours), there's nothing sweeter than a car that starts on the first turn of the key, and the idle was perfect.
Driving it, I immediately noticed a drastic improvement in throttle response, as if it were a drive-by-wire BMW in Sport mode. So I was anxious to let it rip, and when I did the difference was intoxicating.
M50 manifold (left), M52 manifold (right). Although bore sizes are the same 68mm diameter,
Not surprisingly, the M50 manifold made the car noticeably quicker at the top end. Where the car would previously go flat after 6000rpm, the revs now zing to the new 7000rpm redline.
The intake roar at high rpm is one of the best sounds you will ever hear. Much to my surprise and satisfaction, the car felt noticeably quicker at all engine speeds.
Back at Imagine Auto the next morning, we ensured the ambient temp and IAT were the same as the previous day for accurate dyno results. What we found confirmed my observation; there was no loss in power anywhere. The car surpassed 240whp at 6250rpm and maintained it through its new 7000rpm redline. There was also a peak of 219 lb/ft in the mid-range - not bad for software that's tuned for different cams on a car with 128k on the odometer and the original exhaust!
Once in place, it's a factory install; we even used the original fuel rail cover. EHP's Eu
The gains peaked with 21hp at 6430rpm, plus an extra 16 lb/ft from 5900-6430rpm. And while the stock car shuts down after 6430rpm, we predict the gains with stage 2.5 would have been well into the 30whp range if we'd been able to run the stock car up to 7000rpm.
Author's note: As previous testing has shown, the advantage of the M50 manifold against the M52 is greater between 6500-7000rpm. However, it's not possible to compare this with the stock engine because the software doesn't rev to 7000rpm. Yet it's in this range that the M52 manifold will lose up to 25whp, while the M50 set up will sustain peak horsepower to redline, as most '95 M3s do. The real world gains are the ability to hold a lower gear for longer - the torque multiplier.
Down low there was a gain of 9-11 lb/ft of torque between 2400-3100rpm as well. I repeat, we unexpectedly saw a significant gain in the low-end with the stage 2.5 kit over the stock setup with the M52 manifold!
Overall, we've made a noticeable difference to our M3. On the dyno, the car was 0.6sec quicker to accelerate from 25-80mph in a third gear sweep (8.6sec vs 9.2sec). From 50mph on (where third gear is most useful), the car is now 0.4sec quicker. And in that short sprint, where the stock car will reach 80mph, the new setup is already 2mph ahead and just barely getting into the real meat of its powerband.
In the future, we hope to discover what the stage 3 Schrick cams are worth. But for now we're off to a great start. Stay tuned.
We enjoy a peak gain of 21whp and 17 lb/ft at 6430rpm (the stock redline). At low rpm, we
The M50 intake runners are 55x25mm, where the M52's are 45x24mm
The Shark Injector plugs into the DME port and uploads the new software. We suggest hookin
Project M3 on Imagine Auto's Mustang MD500 dyno
Eurosport High Performance