Control of motor-assisted shiftings in dedicated hybrid transmissions

Several trends could be observed in the world of automotive drivetrains over the past few years: Firstly, for EVs there is a trend to develop transmissions with two or even three speeds for increased system efficiency and torque capacity. Secondly, multi-step power-split transmissions, which are well known for off-highway and mobile-machine applications, were ‘rediscovered’ for passenger cars and recently labelled as dedicated hybrids, DHTs.

Shifting a two-speed EV-transmission with big ratio-steps turns out to be rather difficult in terms of control. In order to see how a DHT could make sense considering these points, the depicted transmission can be used as an example.

Fig. 1: DHT concept and corresponding speed diagram

The shown hybrid transmission structure with two electric motors is suited for front transverse applications. It supports three possible shift combinations, two of which can also incorporate an internal combustion engine. In first gear both EMs can be used to propel the vehicle and the torque distribution between the EMs can be varied. In second gear EM2 and an ICE could be used for driving and the load can be distributed among these two. When the two brakes are opened, the speeds of the input shafts can be chosen according to the ladder diagram in Fig. 1. Both EMs and the ICE can be used to drive the vehicle in this mode.

The most obvious benefit of this mode is the possibility to freely adjust operating points to input speeds with optimal efficiency or maximum torque capacity. A less obvious benefit is the possibility to enable speed synchronisation using the electric motors during gear shiftings. Figure 2 gives an example of a low power upshift with this transmission.

Fig. 2: Power-on upshift 1-2


The first gear has an additional kinetic degree of freedom allowing to control a torque hand-over phase before B1 is opened (t1 to t2 in Figure 2). This means that B1 is unloaded before it is opened. During this phase an additional power loss is generated as EM1 is switched from motoring to generating and there will be a power flow from EM1 via the inverters to EM2. After that, B1 is opened and the transmission is in the third mode as all shift elements are now open. The target speed of the closing element can now be synchronised using just the E-motors (t2 to t3 in Figure 2). After synchronisation the transmittable torque of B2 is increased and E1 is turned off as this shifting ends. Note that no power dissipation due to friction occurred and no pressure control of the shift elements was needed. To evaluate efficiency, the additional loss energy due to circulating powers during the shift has to be compared to the dissipated energy in the shift elements during a conventional power-shift. The underlying control laws can be generated by doing a kinematic and kinetic analysis of the transmission.

However, the power-on upshift can only be realised in the described way, if the EMs have enough capacity left to unload the B1 before it is opened. At very high loads the output power will be reduced for a very short time. During that time B2 can support EM1 to minimize the power interruption at the cost of added dissipated energy. During such a shifting the kinetic energy saved in the EMs, which are rotating at high speeds, is utilised to partly fill the power gap during shifting.


Authors of the article

Mick Jordan M.Sc., Daniel Kupka M.Sc.,Prof. Dr.-Ing. Peter Tenberge
Ruhr-University Bochum