There is a dangerous assumption people sometimes make when rebuilding an older performance engine:
Take it apart, clean it up, order some new pistons and put it back together.
That may sound like an engine rebuild.
It isn’t how a performance engine should be built.
Before selecting pistons, deciding on an overbore, ordering bearings or planning machine work, you first need to understand exactly what you have in front of you.
That means disassembly, cleaning, inspection and measurement.
Recently, we tore down a European-spec BMW S50B32 at VAC Motorsports. With the block, cylinder head, crankshaft and pistons separated on the workbench, it provides an excellent opportunity to look at what makes the Euro S50 different—and, more importantly, explain the process that should happen before any performance engine goes back together.
Whether the finished engine will become a restored street engine, high-compression naturally aspirated build, track engine, or forced-induction project, the philosophy is the same:
Take nothing for granted. Measure everything. Build around what the engine actually needs.
The S50B32 is one of the most significant naturally aspirated inline-six engines from BMW M’s 1990s era.
Introduced during the evolution of the E36 M3, the S50B32 displaced 3,201 cc and produced 321 hp at 7,400 rpm in European specification. BMW also introduced continuously variable control of both the intake and exhaust camshafts—double VANOS—on this engine.
That was more than 100 horsepower per liter from a naturally aspirated production engine in the mid-1990s.
The S50B32 was used in European-market E36 M3 applications and also powered versions of the BMW Z3 M Roadster and M Coupe outside North America. BMW records the Z3 M application at the same 321-hp output.
But understanding the Euro S50 requires clearing up something that remains confusing.
At a glance, a North American E36 M3 engine and a European S50 may appear closely related.
They are BMW inline-six engines from the same vehicle generation and share a similar fundamental architecture.
Look more closely, however, and the differences become substantial.
For the 1995 North American E36 M3, BMW used the 3.0-liter S50B30US, producing 240 hp. BMW described the U.S. engine as a specially developed version of its regular production inline-six architecture, featuring dual overhead cams, four valves per cylinder, and a single throttle body.
Beginning with the 1996 model year, the North American M3 moved to the 3.2-liter S52, which retained the 240-hp rating but increased torque.
BMW’s own historical documentation explains that the North American approach was deliberately less elaborate and less expensive to manufacture than the European M engine.
So when enthusiasts casually compare the “US S50” with the “Euro S50,” there is considerably more separating them than displacement or horsepower.
| Feature | North American E36 M3 | Euro S50B32 |
|---|---|---|
| Primary engine designation | S50B30US / later S52 | S50B32 |
| Market | North America | Primarily Europe and other non-US markets |
| Displacement | 3.0L initially; 3.2L from 1996 | 3,201 cc |
| Factory output | 240 hp | 321 hp |
| Throttle architecture | Single throttle | More motorsport-oriented S50 architecture |
| VANOS | Simpler U.S.-market arrangement | Double VANOS |
| Valvetrain philosophy | Production BMW-derived architecture | Solid-lifter, M-specific architecture |
| Cylinder head construction | Conventional integrated cam-carrier approach | Separate cam-box-style architecture |
The important takeaway isn’t that one engine is “good,” and the other is “bad.”
They were designed around different goals.
The North American engine delivered strong performance, drivability, durability, and a lower vehicle cost.
The Euro S50B32 pursued a considerably more specialized high-performance philosophy.
Once the S50B32 cylinder head is sitting on a workbench, some of those differences become easier to understand.
During our teardown, Tony from VAC pointed out that the Euro engine uses an architecture reminiscent of earlier BMW Motorsport engines such as the S14 and S38.
One obvious example is the upper cylinder-head and camshaft arrangement.
The European engine uses a separate cam-box-style assembly that installs over the cylinder head.
On the North American engine, the cam carriers—or “cam trays,” as they’re sometimes called—are incorporated differently into the cylinder-head assembly.
That’s not merely a visual difference.
It reflects two different approaches to cylinder-head and valvetrain design.
Another significant characteristic of the S50B32 is its factory solid-lifter valvetrain.
For a performance engine, valvetrain control becomes increasingly important as engine speed increases.
At thousands of revolutions per minute, every valve, spring, retainer, follower, cam lobe and associated component is being accelerated and decelerated incredibly quickly.
The requirements become even more demanding when an engine is modified with:
That doesn’t mean a solid-lifter design automatically makes an engine better.
It means the engine was engineered around a different performance philosophy.
And it is one of the reasons you should never treat a European S50B32 cylinder head as simply another version of the U.S. S50/S52 head.
The S50B32 also introduced BMW M’s double-VANOS system, providing continuously variable timing control of both the intake and exhaust camshafts.
BMW announced the 3.2-liter M3 engine in 1995, producing 321 hp at 7,400 rpm, and highlighted the addition of exhaust-cam control to the existing intake-cam adjustment strategy.
Variable cam timing allows engineers to optimize valve events across a much broader range of engine operating conditions.
Instead of choosing cam timing that is ideal only at low or high RPM, VANOS can adjust camshaft position as operating conditions change.
That can influence:
For an engine builder, it also adds another layer of complexity.
Camshaft selection, cylinder-head work, compression ratio, piston-to-valve clearance, VANOS condition, and ECU calibration need to work together.
Performance engine building is a system, not a collection of unrelated parts.
One of the biggest lessons from this S50B32 isn’t actually specific to the S50.
It applies to almost every performance engine we build.
You cannot accurately plan a rebuild until you know the engine’s condition.
A used engine has a history.
Maybe it has 60,000 miles.
Maybe 160,000.
Maybe it spent its entire life on the street.
Maybe someone tracked it.
Maybe it overheated.
Maybe the oil wasn’t changed frequently enough.
Maybe someone rebuilt it 15 years ago, and nobody knows what clearances they used.
Maybe a previous machine shop already bored the cylinders.
Maybe one cylinder has suffered damage that isn’t obvious until the engine is apart.
The engine doesn’t care what the seller told you.
The measurements tell the story.
The engine needs to come apart far enough to allow accurate inspection of the major components.
On this S50B32, that means separating components such as the:
During disassembly, an experienced engine builder is already looking for evidence.
Bearing appearance can provide clues about lubrication.
Piston skirts can reveal abnormal contact.
Cylinder walls may show scoring.
Combustion chambers can reveal differences between cylinders.
Deposits can suggest oil consumption or combustion problems.
Fasteners can provide clues about previous work.
The teardown itself is part of the diagnostic process.
There is a reason professional machine shops don’t evaluate dirty engine components.
Oil, carbon, corrosion, sealants, and decades of contamination can hide what you actually need to see.
The S50B32 cylinder head shown in our Tech Tuesday video went through VAC’s blasting process, leaving the aluminum casting looking remarkably close to new.
But appearance isn’t the real objective.
Inspection is.
A properly cleaned casting makes it easier to identify things such as:
Cleaning also prepares components for accurate measurement and subsequent machining.
Vapor blasting—sometimes called wet blasting—uses abrasive media suspended in liquid rather than relying solely on dry abrasive media.
When used appropriately, it can provide an excellent finish on aluminum castings.
That’s why a properly processed BMW cylinder head can emerge looking dramatically better than a greasy, oxidized casting that has spent decades inside an engine compartment.
But this isn’t simply cosmetic restoration.
A clean surface allows the person inspecting the engine actually to see the material.
It is also important to understand that not every component or machined surface should be blasted indiscriminately.
Critical surfaces, oil passages, bearing locations and other precision areas require proper preparation, masking, cleaning and post-process inspection.
Like most engine-machine work, the process matters as much as the equipment.
Now we get to one of the most misunderstood parts of an engine rebuild.
The original S50B32 cylinder bore is approximately 86.4 mm, as Tony explains in the teardown.
That does not mean we automatically order 86.5 mm pistons.
It does not automatically mean we bore it to 87 mm.
And it definitely doesn’t mean we install new standard-size pistons into the old cylinders simply because the engine ran before it was disassembled.
First, the cylinders need to be measured.
A cylinder that looks round to your eye may not actually be round.
A bore can develop:
The cylinder diameter can vary from top to bottom.
Areas exposed to greater combustion pressure and ring loading can wear differently than areas lower in the cylinder.
A cylinder can measure differently depending upon the direction in which it is measured.
Instead of being perfectly circular, it may become slightly distorted.
Foreign material, lubrication problems, piston damage, broken rings, or other issues can damage the cylinder wall.
Even without obvious scoring, the original cylinder-wall finish changes over time.
Never assume you’re looking at an untouched original block.
A previous owner may already have bored, honed, decked, or otherwise machined the engine.
Measure it.
This particular engine illustrates the decision perfectly.
The original bore is 86.4 mm.
Depending on what measurement reveals, the block might potentially clean up at approximately 86.5 mm.
If wear, taper, scoring, or other conditions require additional material removal, an 87 mm piston may make more sense.
But notice the sequence:
We inspect the block first.
Then:
We decide what piston and finished bore make sense.
Not the other way around.
There is another misconception worth addressing.
If an 87 mm piston fits, why not just bore every S50B32 to 87 mm?
Because an engine builder generally shouldn’t remove perfectly good cylinder-wall material without a reason.
Increasing bore can add a small amount of displacement, but that isn’t automatically worth sacrificing material that may be useful during future rebuilds.
The goal should be to arrive at the correct finished bore for the application and condition of the engine, not simply the biggest bore available in a catalog.
If an engine cleans up properly with minimal machining, that can be desirable.
If it requires more material removal to produce a straight, round, properly finished cylinder, then the engine needs what it needs.
Again:
Let the measurements determine the build.
Let’s say we’ve decided the engine needs forged pistons.
People sometimes think an “87 mm piston” goes into an “87 mm hole.”
That’s not how precision engine machining works.
The actual piston must be measured at the piston manufacturer’s specified measuring location.
Then the cylinder is machined to provide the required piston-to-wall clearance for that particular piston design and application.
That clearance depends on variables including:
A forged performance piston changes dimension as it reaches operating temperature.
The cylinder must accommodate that behavior.
If piston-to-wall clearance is insufficient, the piston can expand as temperature rises and lose the clearance it needs.
Potential consequences can include:
A clearance that appears insignificant when measured in thousandths of an inch can determine whether the engine lives or dies.
More isn’t always safer.
Excessive piston clearance can contribute to:
There is a target.
The machine shop’s job is not to make the hole “big enough.”
The job is to produce the correct geometry and clearance.
These terms are often used interchangeably in casual conversation, but they describe different machining operations.
Boring removes material from the cylinder and establishes the approximate new cylinder size.
If an S50B32 block is being significantly increased in diameter, boring may be part of the process.
Honing brings the cylinder to its precise final dimension and creates the surface finish required by the piston rings.
That surface finish is critical.
Piston rings require an appropriate cylinder-wall texture to seat properly and control combustion pressure and oil control.
So when someone says:
“Just throw a hone through it.”
The correct question is:
What does the cylinder actually measure?
A hone cannot magically correct every damaged, tapered, or out-of-round cylinder while maintaining whatever size someone hopes to keep.
Cylinder-wall preparation and piston rings work together.
The final hone needs to be appropriate for the ring material and application.
After the cylinders are finished, piston-ring end gap also needs to be checked and set according to the build.
Again, engine building is a system.
Your piston and ring package, finished bore, piston-to-wall clearance, and cylinder-wall finish are all related.
Ignoring one because you bought expensive parts doesn’t make the others less important.
For a simplified overview, the process looks something like this:
Understand what is actually inside it.
Remove the contamination that interferes with inspection and measurement.
Look for visible wear, damage, and previous repairs.
Cylinder bore, taper, out-of-round, and other critical dimensions need to be evaluated.
Only now can you intelligently decide whether the block remains near its current bore or requires an oversize piston.
Choose the piston based on:
Never machine an engine from the number printed on a box alone.
Establish the required piston-to-wall clearance and cylinder geometry.
Cylinder finish matters.
Engine assembly should involve measurement, not assumptions.
Only after all of those steps are complete should final assembly begin.
That is a very different process from:
“I bought forged pistons. Can you put them in my engine?”
There isn’t a universal parts list because the correct answer depends on the engine’s condition and what you’re trying to accomplish.
A restoration-oriented street engine has different requirements from a 9,000-RPM competition engine.
A naturally aspirated engine has different priorities from a high-boost turbo build.
Depending on the project, the rebuild may involve evaluation or replacement of components including:
The important word is evaluation.
Don’t replace a part merely because somebody on a forum says every S50 needs it.
And don’t reuse something simply because it doesn’t look broken.
Understand the application and measure the engine.
Before ordering internal components, we want to know what the customer expects the engine to do.
Questions should include:
What is the power goal?
What RPM will the engine see?
Street, track, or competition?
Naturally aspirated or forced induction?
What fuel will be used?
How often will the engine see sustained high RPM?
Is drivability important?
What compression ratio makes sense?
What camshafts are planned?
What cylinder-head work will be performed?
Those questions matter because an engine isn’t “built” simply by containing forged components.
A great street engine and a great race engine may use very different combinations.
This is perhaps the most important lesson in the entire article.
You can purchase:
And still build a bad engine.
If the cylinder isn’t straight…
If piston clearance is wrong…
If bearing clearance is wrong…
If ring gap is wrong…
If the deck isn’t correct…
If the cylinder-head work isn’t correct…
If the engine isn’t clean…
The logo on the piston box isn’t going to save it.
Performance parts only function properly when the supporting foundation is correct.
The factory cylinder bore is approximately 86.4 mm.
That does not mean every used S50B32 block will still measure exactly 86.4 mm throughout all six cylinders.
That’s why measurement comes first.
Potentially, if the block can be properly machined to the required finished dimensions and piston-to-wall clearance.
Whether 86.5 mm is appropriate depends on the actual condition of your block.
An 87 mm bore is a common performance rebuild option, but whether it makes sense for your particular engine should be determined after the block has been inspected and measured.
The question isn’t simply:
“Can we make it 87 mm?”
The better question is:
“What finished bore does this particular block need for the engine we’re building?”
We would not recommend simply dropping new forged pistons into worn cylinders without properly inspecting, measuring,g and preparing the bores.
New pistons need the correct cylinder geometry, surface finish and piston-to-wall clearance.
No.
Although they belong to the same E36 M3 era and share basic BMW inline-six heritage, their architecture and performance design differ considerably.
BMW itself described the 1995 North American 3.0-liter M3 engine as a simpler, single-throttle interpretation, designed in part to reduce cost, while the European 3.2-liter S50B32 produced 321 hp and incorporated double VANOS and a more specialized BMW M architecture.
BMW rated the European S50B32 at 321 hp (236 kW) at 7,400 rpm from 3,201 cc.
It isn’t one piston.
It isn’t one connecting rod.
It isn’t one cylinder head.
And it isn’t one magic collection of expensive parts.
It’s the process.
Disassemble.
Clean.
Inspect.
Measure.
Engineer the combination.
Machine it correctly.
Measure it again.
Then assemble it.
That’s the philosophy we apply whether we’re working with a classic BMW S14 or S38, an S50B32 like the engine shown here, an S54, or one of BMW’s modern turbocharged platforms.
Every engine tells a story once it’s opened.
The job of the engine builder is to listen to what the measurements are saying before deciding what comes next.
At VAC Motorsports, we’ve been building, machining, and developing BMW performance engines for decades—from street restorations to race engines and high-horsepower forced-induction combinations.
If you’re planning an S50B32 rebuild, don’t start by asking:
“Which piston should I buy?”
Start with:
“What does my engine actually need?”
That’s where a properly engineered build begins.
Take nothing for granted. Measure everything. Build it right the first time.
In our latest VAC Motorsports Tech Tuesday, Tony walks through this disassembled European S50B32 and explains the block, cylinder head, solid-lifter architecture, cleaning process, factory bore size,e and what happens before the engine moves into machining and assembly.
Have an S50B32, S50, S52 or another BMW performance engine you’re planning to rebuild? Contact VAC Motorsports before ordering parts. We can help determine the right combination for the engine, application, and performance goal.