Saab Had Around 100 BMW N18 Engines in Trollhättan: Peter Dahl Reveals How Far Project 540 Had Progressed

When Saab and BMW announced their engine agreement in September 2010, the press release described a future supply arrangement. BMW would begin delivering a 1.6-liter turbocharged four-cylinder gasoline engine in 2012, adapted to Saab’s requirements and intended for the company’s next generation of vehicles.

That announcement did not reveal how much work had already taken place.

In the fourth installment of Alrik Söderlind’s Saab Fenix video series, former Saab powertrain project leader Peter Dahl explains that cooperation with BMW began during 2009. The first development engine arrived in Trollhättan during the winter of 2009–2010, and Saab had it installed in a drivable 9-3 development car only three weeks later.

By the time Saab Automobile entered bankruptcy on December 19, 2011, Dahl estimates that roughly 100 BMW engines had reached Trollhättan. They were used in Project 540 development mules, engine test cells and continued calibration work.

The planned series deliveries never began. The engineering deliveries unquestionably did.

Video: Peter Dahl Explains Saab’s BMW Engine Program

In Part 4 of Alrik Söderlind’s Fenix series, Peter Dahl explains the BMW N18 program, Saab’s powertrain targets, the Koenigsegg negotiations and the surviving engines now being used by Griffin Cars. Anders Johansson also demonstrates the electronic bridge that allows the Project 540 engine to operate inside an existing Saab 9-3.

The BMW Agreement Did Not Start the Engine Project

The origins of the program predated Saab’s formal separation from General Motors.

Dahl had worked at Saab since 1994 and, between approximately 2006 and 2008, was involved in the development of a new 2.9-liter V6 turbodiesel for GM. Developed with Italian diesel specialist VM Motori, the engine was intended for a broad group of vehicles, including the Saab 9-4X, the second-generation Saab 9-5, Opel Insignia and a longitudinally installed version for the Cadillac CTS.

The program had consumed around two and a half years of engineering work and was approaching production readiness when GM cancelled it as its financial position deteriorated in 2008.

A revised version of the basic engine later appeared as a 3.0-liter diesel in models from Jeep, Maserati and other Fiat Chrysler brands. For Saab’s engineers, however, the cancellation demonstrated how vulnerable Trollhättan had become to decisions made elsewhere in the GM organization.

After returning to Sweden, Dahl spoke with Kjell ac Bergström, then head of Saab Powertrain. Bergström anticipated that GM’s difficulties could eventually leave Saab without access to suitable future engines.

The assignment was therefore not simply to identify an economical replacement for the existing 9-3 engine. Saab needed a powertrain that could support the planned successor to the 9-3 while reducing the company’s long-term dependence on GM.

That successor was Project 540, developed around Saab’s new Phoenix architecture. Our earlier reports examine both the broader Project 540 program and Griffin Cars’ revival work and the platform’s new front and five-link rear suspension systems.

More Than 1,000 Engine Variants Were Screened

Dahl’s team began with a global survey containing well over 1,000 engine variants.

That figure did not mean Saab seriously considered one thousand completely unrelated engine designs. It included different outputs and specifications of engines already in production. Even so, the scale of the initial survey shows that the choice was not made simply because BMW had a suitable MINI engine available.

The team reduced the field to approximately 10 to 20 plausible candidates, then selected four or five for detailed investigation. Six-cylinder engines were eliminated early. Saab wanted to replace and improve upon the turbocharged four-cylinder installation in the existing 9-3, not add more mass, cost and fuel consumption to the nose of the new car.

The requirements were tightly connected:

• lower fuel consumption and emissions;
• less powertrain mass;
• strong torque from low engine speeds;
• a broad and predictable torque plateau;
• reduced parasitic losses;
• controllable auxiliary systems;
• compatibility with Saab-specific calibration;
• and a credible route to E85 BioPower operation.

Dahl describes the existing GM-derived 2.0-liter turbo as effective in several respects, but difficult to keep comfortably inside future emissions limits. Saab’s engineers repeatedly had to invest substantial calibration work simply to maintain the necessary margins.

They wanted a cleaner starting point.

A lighter engine also mattered to the chassis program. Reducing mass ahead of the passenger compartment could improve steering response and make the work of the new suspension and braking systems less demanding. It allowed the entire Project 540 package to be optimized around a smaller and lighter powertrain rather than compensating for an unnecessarily heavy engine.

Saab Approached PSA Before Turning to BMW

The engine that eventually reached the top of Saab’s list belonged to the BMW–PSA Prince family. Because versions of the engine were used by both PSA Peugeot Citroën and BMW, Saab initially spoke with PSA. The engineering team then concluded that much of the relevant design responsibility and technical authority remained with BMW.

Saab approached BMW and asked whether the German manufacturer would supply engines to Trollhättan.

BMW agreed.

Dahl does not believe BMW regarded Saab as a serious threat to its own sales. The expected volume may not have been transformative for BMW, but the supply project would contribute to the company’s external powertrain business and provide additional funding for engine development. Technical cooperation began around the middle or latter part of 2009.

This date materially changes the established timeline. The official agreement signed on September 29, 2010 was not the beginning of contact between the two companies. It formalized a program that was already producing hardware and running prototypes.

Under the official Saab purchase, supply and development agreement, BMW was to deliver four-cylinder 1.6-liter turbocharged gasoline engines from 2012, adapted to Saab’s requirements. BMW’s own announcement added that the engines would use Start-Stop and other fuel-saving technologies and would be assembled at the company’s Hams Hall plant in the United Kingdom. Neither side disclosed the financial value or contractual volume.

Contemporary reporting went further. Autocar identified the engine as the N18, reported that a joint Saab–BMW engineering team had been formed and stated that approximately 200,000 engines were expected during the production life of the new 9-3. That number should be treated as an informed contemporary report rather than a confirmed contract figure, because Saab and BMW explicitly declined to publish the agreed volume.

BMW Returned Three Weeks Later and Found a Running Saab

During the winter of 2009–2010, BMW sent the first development engine to Trollhättan. Three weeks later, BMW engineers arrived to inspect Saab’s progress.

Instead of showing them an unfinished engine bay or an installation model, Saab’s team took them out in a working development car(!). The N18 had already been installed in a Saab 9-3 mule and coupled to a GM F40 six-speed manual transmission. The installation was sufficiently complete for the car to be driven and evaluated.

Dahl recalls that the BMW engineers were impressed by the speed of the work:

“They thought it was fantastic that we had managed to get it drivable in three weeks.”

The episode says more about Project 540 than a list of planned specifications. Saab was not waiting for the final production body before validating the engine. Existing 9-3 bodies were being used as mules so that powertrain, chassis, calibration and electronic development could proceed in parallel.

Dahl also found BMW’s engineering culture unusually compatible with Saab’s. When the outcome of an idea was uncertain, both groups preferred to build or test it rather than reject it in advance.

That approach differed from some of Saab’s experiences inside the much larger GM organization, where an idea could be dismissed because another division had previously tried something similar or because the expected result had already been decided in a meeting.

Asked by Söderlind whether Saab and BMW were a good match, Dahl’s answer is unambiguous:

“It was perfect.”

Why Saab Chose the N18?

The selected engine was the BMW/MINI N18, a turbocharged 1.6-liter four-cylinder from the Prince family. It gave Saab several of the systems its engineers had identified during the selection process:

• variable valve timing and valve lift control;
• a switchable or demand-controlled coolant pump;
• a demand-controlled oil pump;
• turbocharging and direct injection;
• reduced auxiliary losses;
• and useful torque from relatively low engine speeds.

The displacement reduction from 2.0 to 1.6 liters did not represent a retreat from Saab’s established performance targets. The objective was to produce the required torque and output with lower mass, lower consumption and cleaner emissions.

This suited Saab’s long-established approach to engine rightsizing. The company did not need to increase cylinder count to deliver premium-car performance. It needed a compact turbocharged engine that could provide strong response through the part of the rev range used most often on real roads. The N18 also gave Saab a more favorable basis for its new chassis. A lighter engine reduced the load carried by the front axle and allowed the suspension, steering and braking systems to be developed without accommodating unnecessary powertrain mass.

Saab Planned 160, 190 and 220 HP Versions!

Dahl confirms that Saab planned three output levels:

• 160 hp;
• 190 hp;
• 220 hp.

The lower-output versions would have supported higher-volume Project 540 models, while the 220 hp version was intended for the more performance-oriented end of the range. Griffin Cars’ present-day specifications follow the same basic structure. Its official Project 540-based offerings list the N18B16 in 190 hp and 220 hp forms, with corresponding torque figures and short overboost functions.

Saab was not planning to install an untouched MINI powertrain and change only the engine cover. Dahl says the base engine was already highly optimized, so the team did not replace the turbocharger simply to create a different component specification.

Instead, Saab worked on the engine-management software to alter throttle response, torque delivery and the way power built through the rev range. The aim was not to manufacture exaggerated turbo lag. It was to retain the recognizable Saab quality of strong low- and mid-range acceleration, supported by a relatively flat torque curve.

E85 Development Had Already Started

An E85-compatible version was also under development.

BMW did not offer an equivalent ethanol application, but was prepared to support the program if Saab covered the additional engineering cost. According to Dahl, that work had already started.

This was an important part of the Project 540 powertrain strategy. BioPower had become a meaningful Saab engineering program rather than a decorative badge. Earlier production cars used engine-management calibration to take advantage of E85’s higher octane rating while retaining the ability to operate on ordinary gasoline.

A future Saab gasoline range without an E85 development route would have abandoned work Trollhättan had already carried through multiple production applications. The N18 therefore offered Saab two things at once: a modern, efficient base engine and enough technical access to create versions BMW itself did not sell.

The First Emissions Test Changed the Development Baseline

The first emissions test produced one of the clearest technical arguments for the N18.

Dahl explains that earlier prototype installations could initially produce results at 110 to 120 percent of the applicable legal emissions limit. Engineers would then begin the lengthy process of calibration and hardware optimization required to create a comfortable production margin.

The first N18-powered mule started from a very different position.

When Saab placed the car in its emissions laboratory, the result was already close to the company’s normal internal development target  well below the legal threshold.

That gave the team space to work on refinement, drivability, fuel consumption and Saab-specific response instead of first correcting a fundamentally marginal emissions installation.

The lower displacement and more advanced auxiliary systems also helped fuel consumption, while variable valve control allowed the 1.6-liter engine to retain enough bottom-end torque to move the car convincingly from rest.

For Project 540, “smaller” did not mean accepting a weak engine and recovering performance through aggressive boost. Saab had selected a unit whose basic airflow, valve control and friction characteristics gave the calibration team a much stronger starting point.

Around 100 BMW Engines Reached Trollhättan

As the program expanded, BMW supplied additional engines.

Some went into Project 540 mules built from existing Saab 9-3 bodies. Others were installed in engine test cells at Saab’s powertrain laboratory. Different vehicles and test environments allowed development work to continue on installation, emissions, cooling, calibration, durability and electronic integration.

Dahl estimates that around 100 engines were in Trollhättan when Saab went bankrupt. This requires precise wording. Saab did not build one hundred unique “Saab-BMW engines.” BMW supplied development engines that Saab used and modified as part of the Project 540 program.

Some of the surviving engines were later acquired by Anders Johansson and Gabriel Palmenäs, the former Saab engineers behind Griffin Cars. Those engines now make it possible to construct roadgoing cars using genuine hardware from the unfinished program rather than a modern engine selected only because it has similar specifications. The engines are not symbolic artifacts placed beside a display vehicle. They are functioning development components that have returned to the road.

The Court Case Confirmed That the Deliveries Were Real

The physical scale of the program was later documented during BMW’s legal action against Saab Automobile Parts. BMW claimed that a substantial quantity of engines and standalone components had been ordered and delivered between the autumn of 2010 and the summer of 2011, together with engineering work that Saab Automobile never paid for.

In August 2012, BMW sought approximately €2.6 million, then equivalent to slightly more than SEK 20 million, from Saab Automobile Parts. BMW argued that the surviving parts company was also bound by the 2010 agreement. Saab Parts, later renamed Orio, maintained that it had not ordered the engines or components supplied to its former parent company.

The Nyköping District Court rejected BMW’s claim in January 2014. The court concluded that Saab Parts was responsible only for its own orders and that BMW had not shown that the parts company had ordered or received the disputed material. BMW was also ordered to cover approximately SEK 2.57 million in legal costs.

BMW appealed, but the Court of Appeal reached the same conclusion on March 3, 201 Orio had no payment liability for the engines BMW delivered to Saab Automobile 5. Orio had no payment liability for the engines BMW delivered to Saab Automobile under the 2010 agreement.

The litigation therefore provides independent confirmation of a central point in Dahl’s interview:

The series-production contract died with Saab, but BMW had already supplied engines, components and development work.

Making the Project 540 Engine Communicate With an Existing 9-3

The second half of Söderlind’s video moves from Saab’s original program to Griffin Cars’ current engineering work. Installing an N18 in a second-generation Saab 9-3 is not primarily a question of fabricating mounts and connecting fuel lines. The engine and the existing car use different electronic architectures.

The Project 540 engine controller communicates using the newer GM Global A architecture, while the production 9-3’s braking, body and comfort systems use the earlier electrical platform. A start request, torque message or system-status signal sent in one format cannot automatically be understood by the other.

Anders Johansson solved the problem by designing a new body control module capable of communicating with both sides of the car. The module acts as an electronic interpreter between the Project 540 powertrain and the older 9-3 network.

Johansson designed the circuit board, selected its processors and power electronics, wrote the software and created a reproducible unit rather than a temporary development box. The BCM handles functions including ignition requests, lighting and other body systems while translating messages between the two vehicle architectures. It also supports newer functions, including Bluetooth, Wi-Fi and the use of a smartphone as the vehicle key.

Without that bridge, Griffin Cars would have a usable BMW-Saab engine and a usable Saab 9-3 body that could not operate as one vehicle. With it, the car is registered, drivable and capable of demonstrating how several Project 540 subsystems behave together.

Not a Replica of the Lost Saab, but a Working Engineering Demonstrator

The Griffin Cars vehicle is not a visual recreation of Jason Castriota’s unfinished production design. Its importance lies elsewhere.

It combines an existing Saab 9-3 body with substantial elements of the powertrain, front suspension, rear suspension and electrical thinking developed for the model that was supposed to replace it. That makes it a working engineering demonstrator rather than a styling tribute.

The original Project 540 development program used old-body mules for the same reason. Engineers did not need the final exterior to test an engine map, steering response, emissions behavior or communication between control units. Griffin Cars is now continuing that method with hardware that survived the collapse.

Peter Dahl on Koenigsegg’s Saab Negotiations

The Video also returns to the failed Koenigsegg Group acquisition of Saab in 2009. Dahl was not one of the principal negotiators, but occasionally provided technical support. He recalls Christian von Koenigsegg and Bård Eker presenting proposals while facing a much larger GM legal team on the other side of the table.

According to Dahl, GM progressively restricted the intellectual property that would be transferred with Saab. What initially appeared to be a potentially workable acquisition became increasingly difficult to justify. He believes Koenigsegg was right to withdraw once the required investment became too large relative to the reduced package of technology and rights. Dahl nevertheless thinks the plan could have worked under different terms. Saab had the new 9-5 sedan in production, the 9-5 SportCombi approaching production, the 9-4X prepared for launch and Project 540 already far beyond the stage of an untested proposal.

His estimate is that 70 to 80 percent of Project 540’s development lead time had already been invested.

That should not be misread as a claim that every production tool, supplier contract and certification test was 80 percent complete. Dahl was describing the amount of scheduled engineering time already consumed by the program. For additional background on the sale process, see Christian von Koenigsegg’s later assessment of whether Saab could have been saved and our broader investigation, “Who Killed Saab Automobile?”.

Peter Dahl’s Career Did Not End With Project 540

The engine program also adds context to Peter Dahl’s later work in Trollhättan. After the Saab era, he became one of the central figures behind the NEVS Emily GT, another fast-moving vehicle program built by engineers with extensive Saab experience. SaabPlanet has previously covered Dahl’s account of the last Saab and NEVS 9-3 vehicles, his work on Emily GT’s possible production future and the advanced systems that did not reach the existing Emily prototypes.

The same development pattern appears repeatedly: assemble a small team, reduce the time between decision and physical prototype, and use a working vehicle to resolve questions that remain abstract in a conference room. The N18 mule that surprised BMW after three weeks belongs to that same Trollhättan engineering tradition.

Project 540 Was Not Waiting for Its First Engine

SaabPlanet’s archived report, “Saab Receives Gasoline Engines From BMW,” preserved the public announcement of the agreement. Peter Dahl’s interview now provides the missing engineering record.

By the time the agreement was announced publicly:

• Saab and BMW had already begun cooperating;
• the first engine had already reached Trollhättan;
• a running mule had already been built;
• the N18 had been combined with a GM manual transmission;
• emissions testing had produced encouraging results;
• 160, 190 and 220 hp versions were planned;
• E85 development had started;
• and further engines were arriving for vehicles and laboratory work.

The failure of Project 540 was not caused by indecision over its powertrain. Saab had selected the engine, integrated it, calibrated it and driven it. The production body and full vehicle program were still unfinished, but the BMW-powered successor to the 9-3 was no longer a paper proposal.

Approximately fifteen years later, surviving N18 engines, Fenix suspension parts and newly developed control electronics have brought part of that program back onto public roads. The car is not the Saab that would have entered showrooms in 2013. It is the clearest working evidence yet of how close Saab came to building it.