Ever stricter global emissions control legislation and lower target CO₂ values not only characterize the development of drivetrains offered by automotive manufacturers currently, but will also be a dominant factor in the future. To ensure that all mobility needs are catered to, the combustion engine will continue to play a major role in the drivetrain mix in both the short and medium-term and in so doing make an important contribution toward achieving CO₂ emissions targets.
To this end, legislation in Europe underwent a fundamental paradigm shift in the last few years. Previously more stringent emissions laws were implemented by mandating lower limit values that were cross-checked in the laboratory by running a test cycle that had not been changed years. This so-called NEDC (New European Driving Cycle) only covered a relatively narrow spectrum of engine operation – a comprehensive set of engine speeds and engine loads was never measured. This has changed with the advent of the new Worldwide Harmonized Light Vehicles Test Cycle (WLTC), which significantly broadens the measurement testing range. Adding to this is the even more intriguing RDE (Real Driving Emissions) test, which involves measuring pollutant emissions in real, actual driving conditions on public roads.
As such, emissions certification covers a much broader range of engine speeds and loads. In addition, average fleet emissions of CO₂ must be reduced to 95 g/km by 2021 and by a further 30 percent before 2030. The changed constraints have led to new requirements for future combustion engine concepts.
• To minimize charge cycle losses, de-throttling continues to gain in importance.
• In this context, active emissions regulation controls must be able to quickly respond to changes in engine speed and load during transient operation.
• In light of RDE requirements, gasoline engines must be operated stoichiometrically across the entire engine performance map (λ=1); running a rich mixture under full-load conditions in order to protect componentry will no longer be possible.
Based on this requirements profile, fully variable valve train systems quickly prove to be the ideal technology for the job. As at each operating point, the correct air mass in the combustion chamber is actively regulated in the intake ports of the cylinders by way of valve port management, the throttle valves typically found in the induction tract of conventional gasoline engines can be fully opened or even be omitted altogether.
Fully variable valve trains can generally be subdivided into mechanical, electrohydraulic and electromagnetic systems based on their actuator control (see Figure 1).
Figure 1 Current fully variable valve train systems
Schaeffler introduced the UniAir technology as a production-ready solution back in 2009. The same year saw its first application in the Fire MultiAir engine from Fiat as used in the Alfa Romeo Mito. Schaeffler supplies the UniAir as a modular system comprising:
• Electrohydraulic actuator elements
• Control unit software module for engine management
• Application-specific calibration data set
Since the start of production, Schaeffler has manufactured more than 3 million units of the UniAir system as part of a global production network. Used in different engines, the system has continually been upgraded and improved with new functions. By ensuring compatibility of the UniAir software module with the engine management systems of established manufacturers such as Continental, Magneti Marelli and Bosch, Schaeffler realizes a high degree of flexibility when it comes to selecting the components of an engine.
The latest generation of the UniAir system not only makes it possible to reduce charge cycle losses, but also optimize combustion and achieve transient torque control via the air path.
Figure 2 illustrates the design of the electrohydraulic UniAir system. Instead of using rigid connecting elements between the camshaft and engine valve, a defined oil volume encapsulated in the high-pressure chamber transfers the cam (lobe) contour to the engine valve. A pump driven by the camshaft via a finger follower builds pressure in the high-pressure chamber.
Figure 2 Design of the electrohydraulic UniAir system
When the solenoid valve is closed, the oil pressure acts on the engine valve by way of a piston, and the valve opens. As soon as oil flows through the open hydraulic valve out of the high-pressure chamber, the force level acting on the engine valve against the valve spring assembly drops, and the engine valve closes. When the solenoid valve opens at cam lift, the oil flows out of the high-pressure chamber into a buffer known as the intermediate pressure chamber. When the high-pressure chamber is subsequently filled in the base circle phase, the oil returns. This minimizes the amount of energy required to establish oil pressure in the system.
The control system, in addition to the electro-hydraulic actuator technology, represents a key module of the UniAir system from Schaeffler. The control algorithms developed by Schaeffler are provided to automotive manufacturers as a software module, which is then integrated as a control module in the engine control unit (see Figure 3).
Figure 3 Interlinking of actuator technology and engine control unit
The electronics utilize sensor signals quantifying the position of the camshaft and crankshaft as input factors. This information is augmented by data on the oil temperature in the UniAir module. The requirements defined by the engine management system are then implemented, and the solenoid valves of the UniAir system are energized as needed. The on and off times of the solenoid valves are used by the control system to monitor the cylinder activation of the engine valves, and the system compensates for tolerance and age-specific changes in system response as well as balances out the effects of fluctuating oil quality.
A few years ago, Schaeffler embarked on a collaborative project with a development service provider to leverage the benefits of the UniAir system in practical operation. The objectives of the venture were to define and establish a solid standard development method for integrating UniAir in existing engine-vehicle combinations as well as conduct follow-up investigations on vehicles that have been modified in-house by Schaeffler. Figure 4 provides an overview of the current UniAir application process. After a suitable test vehicle is selected, all vehicle parameters that are critical to the investigative work are quantified on the roller dynamometer. The engine is then analyzed in detail on the thermodynamic engine dynamometer. The result serves as the benchmark for the UniAir system application. To this end, actual engine data is integrated in a GT-Power model and UniAir-specific adaptations are made. It is on this basis that the reduced fuel consumption is plotted across the engine map during simulations and in some cases verified in dynamometer test runs. This proven application process also allows Schaeffler to help customers looking to integrate UniAir into their engine system by offering additional development services for adapting the engine components and software parameters to the fully variable valve train system.
Figure 4 Application process
Figure 5 plots a load curve on the performance map of a gasoline engine with forced induction. The areas, or ranges in which UniAir can bring about considerable benefits under different loads in steady-state engine operation appear as dotted lines:
• Under low-load conditions, throttle losses can be reduced (point 1 in Figure 5).
• By leveraging strategies with early intake valve closing (EIVC) or late intake valve closing (LIVC), the knock limit can be shifted higher in the load range without having to take worsening combustion into account (point 2 in Figure 5).
• When operated under load conditions, EIVC or LIVC, together with adapted forced induction, elimination of wide-open throttle enrichment is possible. Subsequently, the engine can be operated throughout the entire performance map with λ= 1, which coincides with the optimal range for the three-way catalytic converter (point 3 in Figure 5).
Figure 5 Load curve on the performance map of a gasoline engine with forced induction
Practicing the development method outlined, Schaeffler carried out investigative exercises in steady-state engine operation for the above three load scenarios utilizing EIVC and LIVC technology. For the performance map range involving low engine loads, the characteristic operating point of 2,000 rpm and 2 bar of mean boost pressure was examined. Figure 6 reveals the results of the simulation for specific fuel consumption, friction intake manifold pressure and combustion duration.
The green line represents the EIVC and the orange line the LIVC strategy. As is indicated by the plotted consumption curve in the upper left corner of Figure 6, LIVC reduces fuel consumption by around 2 % compared with the production standard, while EIVC generally increases fuel consumption. This is particularly remarkable given the fact that engine de-throttling improves with early intake valve closing as opposed to late closing, as the intake manifold pressure in the upper right corner of Figure 6 indicates. The lower right quadrant of Figure 6 provides an explanation: The combustion duration with EIVC is much longer and therefore less fuel efficient than with LIVC. Early intake valve closing also leads to higher friction levels (lower left quadrant of Figure 6). These two factors overcompensate the benefits of reduced throttle losses with EIVC and are responsible for allowing the LIVC strategy to reduce fuel consumption. The situation changes if the intake valves have masking, or additional covers at the valve seat. Figure 7 shows the results under otherwise identical constraints as for the previous simulation.
Figure 6 Results of engine tests at an operating point of 2,000 rpm with 2 bar of mean boost pressure, without masking, Figure 7 with masking
Now the EIVC strategy considerably reduces fuel consumption as the masking leads to a much shorter combustion duration and the improved de-throttling response really comes into its own thanks to the early intake valve closing. Despite the still higher friction levels of early as opposed to late intake valve closing, EIVC is the better choice. Generally speaking, the effect of the charge motion is key to combustion efficiency and cannot be overlooked during the operative application. Optimal results can be achieved if the overall engine is tuned to the new operating conditions when the UniAir system is integrated. This includes the engine valve springs, which can generally be lowered with the UniAir system and lead to a further reduction in fuel consumption, particularly in the low load range. Schaeffler currently is investigating the potential of this design measure in simulation exercises as well as on the test stand.
For tests conducted at the knock limit of the engine, a low-end torque point of 2,000 rpm and 13 bar under high-load and low operating speed conditions was chosen (point 2 in Figure 5). In Figure 8, the blue line represents the intake valve closing of the standard engine, while the diagram values to the left and right result from early and late intake valve closing, respectively.
Figure 8 Results of engine tests at an operating point of 2,000 rpm with 13 bar of mean boost pressure
Both approaches reduce the effective compression, thereby shifting the knock limit of the engine further upwards. Fuel consumption at the operating point in question can be considerably reduced with early as well as late intake valve closing. In the process, late closing offers slightly better performance (with up to 4 % less fuel consumed) than early closing (3.6 % at the optimum point) (top left diagram in Figure 8). One reason for the fuel savings is the better center of combustion as shown in the lower left quadrant of Figure 8: For the standard engine, the 50 % conversion point is at an unfavorable 16° Crank Angle (CA) after TDC. Utilizing EIVC and LIVC, this point approaches the theoretical optimum of 8° CA at approximately 12° or 9.5° CA, respectively. The center left diagram in Figure 8 plots the differential pressure in the intake and exhaust tract. A positive scavenging gradient can be observed in the gas-exchange loop during early and late intake valve closing. Even when pumping losses are factored in as shown in the lower right quadrant of Figure 8, EIVC and LIVC yield better performance over the standard engine. The position of the wastegate flap as depicted in the upper right quadrant of Figure 8 plays a key role at different engine operating points. While this flap is fully open on the standard engine, with the EIVC setup, the incoming air mass is cut off by the intake valve control system, whereas with the LIVC configuration, this mass is partially displaced so that the wastegate flap does not need to be opened all the way. This also means that the turbocharger in the overall engine system must be tuned to the specific operating requirements of the EIVC or LIVC strategy, whereby the engine must be run at higher boost pressures to ensure that the necessary air mass enters into the cylinder.
An operating point of 5,000 rpm and 22 bar of mean boost pressure were chosen for testing under full-load conditions (point 3 in Figure 5). In many cases, the maximum exhaust temperature under high-load conditions represents a limiting factor because engine components must be protected when exposed to extreme thermal loads. For the standard engine analyzed, an exhaust temperature limit of 950 °C to protect components was considered. Under a full engine load, the air-fuel blend must be considerably enriched to cool the combustion chamber and meet the 950 °C exhaust temperature limit. As Figure 9 shows, the standard engine is operated with λ= 0.88 at the operating point in question.
Figure 9 Air-fuel blend at an operating point of 5,000 rpm with 22 bar of mean boost pressure
Figure 10 Specific fuel consumption at an operating point of 5,000 rpm with 22 bar of mean boost pressure
Figure 11 50 % conversion point at an operating point of 5,000 rpm with 22 bar of mean boost pressure
In Figure 9, the intake valve closing is superimposed onto the ordinate; LIVC displaces the value upward, while EIVC displaces it downward. Both approaches can be used to configure a quasi-stoichiometric combustion recipe while taking the maximum temperature into account. Figure 10 shows the resulting effect on specific fuel consumption.
With a starting point of 330 g/kWh, fuel consumption can be reduced by 20 % to 260 g/kWh with EIVC or LIVC. In addition to the quasi-stoichiometric combustion, the better center of combustion during early and late intake valve closing also contributes to the fuel savings achieved. As a result of the fuel enrichment-induced cooling required, the 50 % conversion point of the standard engine shifts to an unfavorable 22° CA after TDC. Active control of the effective compression enabled by the UniAir system ensures lower compression temperatures such that the engine remains further from the knock limit. As such, the ignition point and center of combustion can be advanced. Instead of 22° CA after TDC, the 50 % conversion point moves to 16° CA after TDC (Figure 11). As is the case with the low-end torque point observed, here, too, the turbocharging system must be adapted for the system application so that the increased air mass required by the EIVC or LIVC setup can be accommodated.
In light of the WLTC and RDE legislation, the transient response of a combustion engine will play an ever-increasing role when it comes to evaluating fuel consumption and emissions. In conventional gasoline engines, highly dynamic transient load shifts are typically controlled by momentarily altering the ignition sequence, such as is the case when shifting gears in a dual-clutch transmission. With the cycle-specific control logic, the UniAir system opens up new possibilities for altering torque output via the air path as the air mass can be adapted almost as quickly as the ignition point. Examples of using the air path for fast torque control include not only rapid load shift and gear changes, but also engine idle speed, inertia fuel shutoff and cylinder deactivation.
Figure 12 plots the progression of a WLTC test for a standard engine. The phases during which the ignition sequence is retarded to realize dynamic load shifts are highlighted in gray. You can easily see that these phases occur very frequently during WLTC testing. By retarding the ignition, the 50 % conversion point (shown in blue in Figure 12) is also shifted from typically 10° CA after TDC back to up to 70° CA after TDC. At this time, combustion is almost 100 percent inefficient for a brief period (bright green progression plotted in Figure 12). This has a measurable negative impact on overall fuel consumption. The black progression at the top in Figure 12 indicates the momentarily increased fuel consumption caused by the inefficient combustion sequence, while the black progression below cumulates the values. In the WLTC test cycle, this increased fuel consumption adds up to a noteworthy 50 g.
Figure 12 Impact of altered ignition sequence on fuel consumption during WLTC testing
By regulating the torque output via the air path, the ignition point and center of combustion can be maintained in the optimal window of operation when in transient operating mode. The air mass is regulated by the EIVC control similarly to the aforementioned processes in steady-state operation. This gives rise to three fuel-saving potentials in the WLTC test cycle (Figure 13).
Figure 13 Fuel-saving potentials afforded by transient air path regulation in the WLTC test cycle
At idle speed and under low-load conditions, the throttle valve linked to conventional engine control systems cannot throttle back the air as much as the engine would actually require, and the ignition is delayed accordingly. Due to its fine-tuned air regulation logic that lifts the valve by only 0.5 mm, the UniAir system also does not need to alter the ignition sequence under low-load conditions, such as at idle speed. This, in turn, reduces fuel consumption by up to 1.8 % in the the WLTC cycle. Adding to this are the benefits of the aforementioned torque reserve, which can be accessed during rapid transient load requests or gearshift changes without moving the center of combustion to a negative operating point. This reduces fuel consumption by 1.0 % in the WLTC test cycle. The third effect is the purging of the catalytic converter following periods of deceleration. On conventional engines, fresh air is pumped through the cylinders into the catalytic converter when the vehicle is coasting. When combustion is restarted, the engine must first be run on an extremely rich mixture with λ = 0.8 to 0.9 for 5 to 10 seconds in order to remove the excess oxygen from the catalytic converter; otherwise, it will not be able to chemically neutralize the pollutant emissions. As UniAir is able to close the intake valves during periods of deceleration, no pump effect is encountered nor does oxygen enrichment take place in the catalytic converter that then needs to be removed during subsequent firing events by running a rich combustion mixture. In the WLTC test cycle, this single effect reduces consumption by approximately 1.0 %. Combining all of the potential savings with aforementioned measures in transient operation, a total of 3.8 % reduction in fuel consumption can be achieved.
Schaeffler has carried out a series of simulation exercises to illustrate the overall benefits of UniAir for a three-cylinder turbocharged engine subjected to the WLTC test cycle (Figure 14). Due to the reduced throttle losses, the system minimizes fuel consumption by 3.6 % as compared with conventional throttle valve regulation. If one utilizes the higher knock limit afforded by the UniAir system to increase the compression ratio, the fuel burn is improved by 6 %. Together with the improvements in transient operation, the total reduction in fuel consumption of 8.4 % in the WLTC test cycle can be achieved, depending on the respective application scenario and application-specific constraints.
Figure 14 Fuel-saving potentials afforded by the UniAir system in the WLTC test cycle
The potential of the UniAir system for improving a combustion engine in an electrified drivetrain depends on the hybrid concept utilized. Figure 15 shows the performance map results in the WLTC test for the three different configurations
• 100 % combustion-engine drivetrain
• P0/P1 hybrid (starter-generator)
• P2 hybrid (hybrid module on the crankshaft with mechanical disconnection from the combustion engine)
The green dots in Figure 15 indicate the dwell time at a given operating point. Without hybridization, medium and high loads are run in particular (Figure 15, top). P0/P1 hybridization shifts operation to lower load levels for the combustion engine as the recuperated energy is routed through the electric motor into the drivetrain to “soften” peak engine loads. Even if the combustion engine is run in a less favorable range in the WLTC test, the fuel consumption of the overall system is reduced thanks to recuperation, or regenerative braking. The benefits of de-throttling under low and medium-load conditions as offered by the UniAir system are particularly effective in this hybrid setup as the combustion engine frequently enters this operating range.
Figure 15 Performance map results for a 100 % combustion engine drivetrain and hybrid systems in the WLTC test
Compared with a P0/P1 hybrid, the electric motor in a P2 hybrid typically has a higher rated output. This means that a large portion of the low-load operating range can be covered using electrical energy only, whereby the kinetic energy harnessed during regenerative braking is channeled back into the drivetrain so that the combustion engine can be switched off. The combustion engine is largely responsible for propelling the vehicle in the higher load range (Figure 15, bottom). Here, the UniAir system can also be used to optimize the engine under these conditions. Not only can the system de-throttle the engine in the medium to high-load range; it also reduces consumption under full-load conditions by shifting the knock limit and center of combustion as previously described.
In-house analyses at Schaeffler point to the practical potentials of the UniAir system in a WLTC simulation for a 48 V hybrid drivetrain as used in a P0/P1 and P2 configuration (Figure 16).
Figure 16 Fuel-saving potentials afforded by UniAir in a hybrid system
The starting point for the tests was a conventional 1.0-liter three-cylinder gasoline engine without hybridization or a fully variable valve train. By applying UniAir system and tweaking the compression from 10 to 11.7, a fuel savings of 4.9 % could be achieved (Figure 16, top). If the compression is increased still further to 13.5 %, fuel consumption can be reduced by an entire 6.3 %. The graph at the center of Figure 16 shows the potential of the UniAir system for a P0/P1 hybrid powertrain. Here, the starting point is the engine without the UniAir system but which has been fitted with a hybrid module. By integrating the UniAir system, together with the compression ratio of 11.7 %, 5.8 % less fuel was burned. A high compression ratio of 13.5 % reduces fuel consumption by 7.1 % as compared with the base version. The graph at the bottom of Figure 16 plots the results for the P2 hybrid. Here, too, the technological leap afforded by the UniAir system, together with a higher compression ratio, yields a fuel consumption savings of between 3.9 % (11.7 % compression ratio) and 5.1 % (13.5 % compression ratio).
Schaeffler has been mass producing the fully variable UniAir valve train system since 2009 and has since delivered over three million units to customers around the world. The system has also been continually improved and optimized with more functions. Schaeffler has devised a development method that also makes it possible to quickly and easily integrate the UniAir system in existing engine designs. In this context, in order to fully exploit the performance capabilities of the UniAir system, it is important that the overall engine with turbocharging system be adapted to the requirements of the fully variable valve train. With this setup, Schaeffler was able to reduce fuel consumption by a minimum of 8.4 % during WLTC testing. Particularly in transient engine operation, the cyclically variable air path regulation logic of the UniAir system considerably lowers fuel consumption not only during cycle-based measurements, but also in operation on public roads.
When it comes to future drivetrain concepts involving P0/P1 hybridization, UniAir represents a key module for coordinating combustion to the specific requirements of the various concepts. Additional engine tests at Schaeffler also point to the fuel-reducing capabilities of the UniAir system when the combustion engine is restarted, which are realized by not having to recondition the three-way catalytic converter. The pronounced flexibility of the UniAir system likewise facilitates cost-effective modular concepts as the engines can be easily adapted to different applications by fine-tuning the engine software.
Preliminary tests are currently being run on a UniAir system fitted with an additional pressure accumulator. The goal with this configuration is to reduce the hydraulic losses associated with EIVC concepts. Instead of simply dissipating the oil pressure in the UniAir actuator via the return line, the pressure accumulator utilizes the excess energy to tension a spring, which then releases the pressure back to the hydraulic system during the next cycle. Initial results indicate that this improvement can reduce hydraulic friction levels by upwards of 30 %.
 Haas, M.: UniAir – The First Fully-Variable, Electrohydraulic Valve Control System. 9. Schaeffler Kolloquium, Baden-Baden, 2010
 Haas, M.; Piecyk, T: Valve Trains for Implementing Innovative Combustion Strategies. 10. Schaeffler Kolloquium, Baden-Baden, 2014