Engine rebuild overview: what actually fails, what it costs, where the DIY line sits

Every rebuild conversation starts with the same question: is this engine worth saving? The honest answer depends on which engine you have, why it came apart, and how much of the work you can do yourself. This page is the map — the individual jobs each get their own guide with the factory numbers.

What actually fails, per platform

N54 (2007–2016). The engine that keeps machine shops in business. The failure shortlist, roughly in order of how often we see it:

• Rod bearings on tuned, high-mileage engines. Cylinder pressure is the killer — the bearings see load the factory never spec'd for sustained big-torque operation. The e90post world has made "check your rod bearings" a ritual for good reason.
• Ring wear and oil consumption past 100k miles, especially on engines that lived on long drain intervals. Direct injection and a tight ring pack do not forgive neglected oil.
• Valve cover and oil filter housing gaskets, oil pan gasket — the perpetual weep list. Cheap jobs, but they mislead people into thinking the engine is unhealthy when it is just BMW's gasket philosophy.
• The crankhub friction stack and timing chain wear — covered in depth in our crankhub guides.

N55 (2011–2017). Same bottom end architecture as the N54 with a single twin-scroll turbo. Spun rod bearings on tuned cars are the headline failure; Valvetronic hardware and the valve cover integrated into the head add cost to top-end work.

S55 (2014–2020). The M3/M4 motor. Rod bearings again — but note the factory data has an important difference: the crankshaft journals are marked "no machining possible." A damaged S55 crank is a new crank, not a reground one.

B58 (2015+). The closed-deck block, forged crank and rods make this the strongest factory bottom end of the group. What actually brings B58s apart: oil consumption from valve stem seals and ring wear, coolant system pieces (water pump, heat exchanger), and rod bearings only under sustained abuse. Most "B58 rebuilds" are refreshes, not rescues.

The honest cost shape

Rebuild conversations go wrong when people compare a bearing swap to a rebuild. These are different jobs with different bills. Rough shape, in escalating order:

The economics flip depending on your labor rate. If you pay a shop $120+/hour for every step, a full N54 rebuild rarely beats a good used engine. If the labor is yours, the parts bill for a bearing-and-rings refresh is modest and the measuring is the whole game — which is exactly what the factory data gives you.

Where the DIY line sits

Yours in the garage

• Rod bearing replacement and clearance checks with plastigage
• Ring file-fitting and piston installation
• Head gasket and head bolts — a cleaning and sequence job, not a machining job
• Timing chain, tensioner, VANOS work (see the crankhub guides)
• Oil system priming and break-in

The boundary is not skill, it is equipment. You cannot honestly check your own work without the tool that measures it.

The machine shop's

• Boring and honing — and on an aluminum N54/B58 block, honing with a torque plate installed is non-negotiable for round bores under bolt load
• Decking surfaces flat
• Valve seat cutting and valve guide work
• Crack testing (pressure or magnetic particle) — see our block-off and pressure-testing guide for what the factory procedure looks like

The baseline: know what healthy looks like

Before an engine comes apart — and after it goes back together — the factory gives you two cheap health checks: compression and oil pressure.

A fresh rebuild should sit at the top of the band with near-zero spread. One hole 2 bar down is the factory's own definition of a problem.

Where to go next

• **Pistons and rings** — the file-fit gap numbers for N54, N55 and B58, straight from the factory data
• **Rod and main bearings** — clearances, the classification systems, and every torque stage
• **Head gasket and TTY bolts** — the stage tables and the retorque answer
• **Valvetrain refresh** — hydraulic compensation, cam data, and what mates to the crankhub guides
• **Oil prime and break-in** — the first twenty minutes that decide the whole job

Every number in this canon traces to the ISTA document ids cited on each page. Where the factory data is silent — and on failure rates it is always silent — we say so instead of inventing a statistic. And once the fresh engine has seated its rings, treat it as a new baseline for the tune too — the MHD flashing guide covers starting from a clean stock read.