Built S55 Engine Guide: What Each Stage Needs, and When

Most built S55s don't start with a plan. They start with a noise.

Someone turns the boost up on an F80 M3, F82 M4 or M2 Competition, a rod bends, and the engine comes apart on our bench. From there the questions are the same every time: what does this engine need to be safe again, and what does it need for the power the owner actually wants?

We don't sell fixed engine packages. Every S55 we build is specced around two things: the damage it came in with, and the power it will make afterwards. This guide walks through how we make those decisions, stage by stage, and which of the parts we sell belong at each step. Just as useful, it covers the parts we think you don't need.

How a stock S55 usually gets hurt

The most common S55 rebuild we see starts with a bent connecting rod, and it doesn't take big turbos to do it. A rod can bend on the stock turbochargers.

The pattern is familiar. An owner raises the boost and chases torque. The small factory turbos spool hard and make a lot of boost at low rpm, which puts a huge amount of torque through the rods at exactly the point where they are least able to take it. BMW's own figures show how early it arrives: even stock, the S55 in the M3 and M4 makes its full 406 lb-ft from just 1,850 rpm. If the tune goes a little too far and the engine starts to detonate, it can take only one or two track days before a rod bends.

A word on port injection kits, because they come up. We wouldn't blame them every time, but a lot of the engines that come to us with detonation-bent rods have one installed. With a tuner who knows what they're doing and a kit that works properly, there shouldn't be a risk. But a second fuel system adds complexity, and complexity means you have to be very sure of what you're doing. For a customer who wants a genuinely fail-safe engine afterwards, we often recommend taking it out.

Caught early or caught late

When a rod first bends, it cocks the piston slightly in the bore. Owners often hear it: a light rattle or shake from the engine when they lift off after accelerating. If you shut it off right away, the rod may only be slightly bent, and the engine can be rebuilt without replacing too many parts.

Keep driving it and the story changes. A badly bent rod can seize the piston in the bore and then break. That damages the crankshaft, often puts a hole in the block, and turns a rebuild into a much more expensive job. These blocks are not cheap.

Metal debris caught on the oil pump pickup. Once metal is circulating, everything it passes through is at risk.

The first decision is the bore

Before anyone talks about rods or horsepower, we have to know what the cylinder walls look like, because the S55 is unusual here.

The S55 block is aluminium with no iron cylinder liners. Instead, a thin layer of steel is sprayed onto each cylinder wall. BMW calls the process LDS; Ford uses a similar process on some of its engines. That steel layer is only about 0.010" thick, so it doesn't take much damage to go straight through it into the aluminium. The piston is part of that system too: on the S58, which uses the same kind of bore, BMW specifies a Grafal-coated piston skirt specifically because the bores are LDS-coated.

That gives us three outcomes, and the bore decides which one you get:

• No damage: the bore stays at the stock 84.00 mm with a new set of pistons.
• A very minor scratch: we clean it up with a light hone and fit custom 84.05 mm pistons. For an 84.05 build we almost always use CP, built to a custom spec we trust.
• Anything deeper: through the coating and into the aluminium, the block gets sleeved. We bore it out, press in a steel liner, and bring it back to the stock 84.00 mm. We sleeve all six cylinders, every time, for consistency. That's our S55 sleeving service.

A sleeved S55 block. Once damage goes through the factory coating, this is the repair.

Notice what isn't on that list: going bigger. We don't overbore turbo engines. Our S55 builds stay at 84.00 or 84.05 mm.

Sleeve it, or find a used block?

When a block needs sleeves, owners often ask whether a used block would be cheaper. Almost always, it isn't.

Sleeving the block you already have is usually less expensive than buying a used one. A used block will often need honing or boring anyway, which lands you in the same place. You never really know its condition until it's in the machine shop being measured. The main line bore, where the crank sits, could have a problem. And it's worth asking why that block is for sale in the first place.

The exception is a catastrophic failure: a hole in the side of the block and damage to the head as well. Then a good used long block makes sense as the starting point for a new build. Tear it down and, with luck, it only needs a bore and hone. But we wouldn't go shopping for a used block when we could sleeve the one in front of us.

Stage 0: a stock rebuild, at the lowest cost

Stage 0 is for the owner who has no interest in more boost. They want a stock-ish engine back in the car at the lowest cost, and they'll run the stock tune.

Rods. With one bent rod and a very tight budget, we can replace just the damaged rod. We find another stock rod that is very close in weight, then send all six to our local balance shop. They remove material from the big and small ends until the set is evenly matched. It works, but it's often not worth the hassle, and we'll usually suggest a set of Carrillo rods instead.

Pistons. Almost every engine we build gets a new set. All six have to match in weight, brand and size, so mixing old and new doesn't work. On a sleeved block back at 84.00 mm, a Mahle Powerpak set is an option alongside CP.

The cylinder head. At minimum it gets a valve job and a careful crack check, which is what our Stage 1 head service covers. Cracking around the spark plug is very common on this engine when it has seen detonation, and engines that have been pushed hard often come in with cracks in every cylinder. Very minor cracking can sometimes be welded and repaired. Whether to weld or replace depends on what the owner does next. If they'll keep running high boost, we recommend spending the extra on a good used head with no cracks, because a repaired head is never as good as an undamaged one.

An S55 cylinder head, combustion chamber side.

The crank hub. We fit our keyed crank hub even on Stage 0 builds. The factory hub holds engine timing with friction alone, and when it slips the valves can hit the pistons. Usually it's the exhaust valves that get damaged, though how bad it gets depends on how far it slipped. Having just rebuilt the engine, it makes no sense to leave that risk in place. Our S55 crank hub guide covers the failure and the fixes in detail.

Stage 1: forged rods and pistons

Stage 1 is Carrillo rods and CP pistons, and it's where most S55 owners who want more power end up. It lets you push meaningfully past stock with a large margin. Carrillo rates these rods at 1,500 hp with the standard WMC bolts. CP rates the pistons at 1,000 hp in standard form.

With every Stage 1 build we also recommend aftermarket head studs and the keyed crank hub. Factory head bolts stretch under serious cylinder pressure; studs don't, and they can be reused.

For bearings we default to ACL rod bearings and ACL main bearings. ACL and King bearings sometimes come in slightly different sizes. If ACL doesn't give us the fit we want on a particular engine, we use King.

Many customers are perfectly happy with the stock cylinder head at this level, so we often leave it alone beyond the inspection and valve job.

Stage 2: the cylinder head

The big change at Stage 2 is the head. We start recommending our Stage 2 head at around 650 horsepower.

It adds two things:

• A Supertech valve spring kit. The springs raise seat pressure from 56–58 lb stock to 82 lb, and from 124–130 lb to 150 lb at 0.400" lift. That keeps the valves seated under high boost. When an S55 floats its valves, the valve pounds on its seat and damages it.
• Ferrea valves. They're made from a stronger material than stock, so they take more abuse from the extra boost.

There's an economic argument too. If an engine comes in with valve damage on one or two cylinders, buying BMW valves for those cylinders costs about the same as a full aftermarket kit for all six. At that point, upgrading every valve is the obvious choice.

The bottom end doesn't change much between Stage 1 and Stage 2. Carrillo rods and CP pistons cover both. Because our CP pistons are built to order, we can add strengthening features such as DLC-coated or larger wrist pins, and rods can be specced beyond Carrillo's off-the-shelf part. We don't spell those out as a package. We recommend them case by case, based on the car and how it's used.

Stage 3: more power at the same boost

Our Stage 3 head has the same Ferrea valves and Supertech springs as Stage 2, plus CNC porting of the intake and exhaust ports and CNC work in the chamber to unshroud the intake valves.

The factory head deliberately shrouds the intake valves. It reduces reversion and carbon build-up on a direct-injected engine, but it costs a lot of intake flow. Opening that up lets more air move freely in and out of the head at the same boost pressure, which means more power without asking the turbochargers to work any harder. It pays off most on engines with larger turbos. The original Stage 3 cylinder head write-up has the full gallery.

The biggest builds: 1,000 whp and beyond

At the top of the ladder, every weak point gets addressed:

• CARR rod bolts. Carrillo rates its rods to 1,800 hp with CARR bolts, up from 1,500 hp with the standard WMC bolts. That's the CARR option on our Carrillo set.
• Upgraded wrist pins. CP rates its pistons to 1,600 hp with HD wrist pins, and recommends them for anything over 1,000 hp.
• Nickel-plated head studs. They're the same steel and strength as the burnished set; the plating adds corrosion protection for engines that come apart often.
• Sleeves and a copper O-ring. Our sleeving service can include a copper O-ring set into the block deck, which presses into the head gasket and adds sealing pressure between cylinders. We recommend it at around 1,000 whp. It runs with the stock BMW head gasket.
• The keyed crank hub. Non-negotiable at this level.

The most powerful S55 we've built made about 1,200 wheel horsepower. It had fully upgraded CP pistons with DLC-coated wrist pins and fully upgraded Carrillo rods with many of the options both companies offer.

What we don't sell: billet crankshafts

We almost never replace the stock BMW crankshaft on an S55. It is very strong. We've never seen one fail on its own, only as a result of another failure such as a bent rod, and we've seen these engines make well over 1,000 horsepower on it.

A billet crank is very expensive and solves a problem the S55 doesn't really have. Even if your crank is damaged, a good used crank, a new forged crank from BMW, or our S55 crankshaft repair service costs far less. That service grinds out the damaged journal, welds it up, regrinds it to original size, then heat treats and polishes it.

No head can save you from a bad tune

The Stage 2 and Stage 3 heads make more power and keep the valves under control. They do not make the head more resistant to cracking from detonation.

The weakness is in BMW's design. The direct injector, the spark plug and four valves all share a small chamber, so the holes in the head sit very close together. That leaves less material between them than we would like, and there isn't much a machine shop can do about it. The only real protection is a tune that never makes the engine suffer through detonation.

What BMW changed for the S58

The S58 replaced the S55 in the new M3 and M4, and BMW's technical training material for it reads a lot like our upgrade list. Here is what BMW M reinforced, and the S55 part that does the same job:

• A timing drive that can't slip. The S58's crank drives the timing chain and oil pump through gears, with no friction-fit hub to let go. On the S55, that's the job of our keyed crank hub.
• Forged rods from a bigger engine. The S58's connecting rods are carried over from the S63 twin-turbo V8. On the S55, Carrillo rods do that job.
• Coated wrist pins. The S58 runs 23 mm DLC-coated pins. On our biggest S55 builds, that's the HD wrist pin upgrade in the CP pistons.
• A stronger head seal. BMW moved to a four-layer steel head gasket on the S58 because of its higher combustion pressure. On the S55 we add head studs, plus a copper O-ring at around 1,000 whp.
• Less compression. 9.3:1 on the S58, against 10.2:1 on the S55, which leaves more room for boost.

In other words, BMW reinforced the same places we do. Our S58 build article goes through the two engines side by side.

One more S55 detail from the same material: the S55's oil pan started out as magnesium, and BMW switched it to cast aluminium during production. If you're sourcing a used pan, it's worth knowing which one you have.

Why we measure everything

The parts list gets the attention, but the measurements decide whether a build lives. These are a few of the calls we make on every engine.

An S55 bottom end on Carrillo rods with CARR bolts, the setup our biggest builds use.

Bearing clearances. BMW's maximum main bearing clearance is 0.0018". For a high-performance engine that's quite tight, so we often build a little looser, which is not usually a problem. The danger runs the other way. Too tight, combined with oil that's too thick, and the bearing can actually touch the journal. For rod bearings our target is usually 0.002–0.003", depending on the oil the engine will run.

Main bearing shells laid out in order, block side.

Bore sizes. The machinist bores the block with a torque plate bolted on, which simulates the distortion of a head being torqued down. We check bores in our own shop too, but without a torque plate they won't measure perfectly round, so we trust the machinist's numbers over ours.

Piston-to-wall clearance. When our measured clearance doesn't match CP's spec, there are a few ways to handle it:

• Too tight: easy. The block goes back to the machinist to hone a little more off the cylinder wall.
• Loose: often deliberate. A high-boost or high-rpm engine needs extra clearance for when the piston heats up and grows.
• Far too loose, because of a machining error or a bore that's simply too big for the piston: we send the pistons out for a thick skirt coating to take up the gap, then re-hone to the clearance we actually want.

Ring gaps. Rings sometimes arrive already gapped looser than our target. When they do, we gap the second ring a little bigger than the top ring. That's normal on high-boost builds, which run a looser piston-to-wall clearance anyway. The whole ring package is designed to work together.

Oil jets. Jets in the block spray oil at the underside of each piston to keep it cool, and they only do their job if they're aimed correctly. We check oil-jet alignment with a dedicated tool on every build.

Spark plug gap. At very high cylinder pressure, the stock 0.029" plug gap causes misfires. Closing it to 0.022", sometimes tighter, stops the spark from being blown out under boost.

Pistons in, head gasket on. We use the stock BMW multi-layer steel gasket, even with an O-ringed block.

Outside the engine: tune, fuel and cooling

Every engine goes back on a stock tune. A lot of people rebuild an engine assuming their old tune was fine, even though it may be the reason the engine came apart. So every car leaves us on stock software, and we ask the owner to be very careful about what goes back on it. If you want more power, have it tuned properly. We do dyno tuning in house on our Mustang dyno, and custom bootmod3 tuning for the S55's Bosch MEVD17.2.G engine computer.

Fuel is often overlooked. More boost needs more fuel, and the system has to keep up. BMW's own answer for the S58's extra power was more fuel pressure: up to 350 bar of direct-injection pressure, up from 200.

Cooling is overlooked even more often, especially on track cars. Most cars are tuned on a dyno, where heat soak isn't really an issue. Then they go to the track, get driven hard in the heat lap after lap, and the temperature keeps climbing. Not every tune compensates properly. Ideally, the tune pulls boost and backs off timing more aggressively as heat builds. If yours doesn't, that's worth knowing before a track day.

How we put them together

None of this is visible once the engine is back in the car, so here is some of what happens before it goes in.

Sealing the bedplate. The green is sealant from the bedplate sealing process, showing at the small holes it's pushed through.

The lower timing assembly (left) and the full timing chain cassette that runs up to the head (right).

The short version

• The bore decides the build. A scratch means 84.05 mm. Damage through the coating means sleeves, all six, back to 84.00 mm. Never an overbore.
• Every build gets new pistons and a keyed crank hub.
• Carrillo and CP are the core of every build that makes more than stock power.
• The head comes in at about 650 hp (Stage 2) and for bigger turbos (Stage 3).
• Keep the stock crank.
• A bad tune will break any of it. Every car leaves us on a stock tune.

Related S55 work

• S55 crank hub failure: which fix actually works
• F80 M3 bent rods: building the S55 for big boost
• S55 Stage 3 cylinder head
• Building an 800+ hp S58: a G80 M3 engine build
• Blown engine on your BMW? We can help