Powertrain Design For Small Scale Parallel Hybrid-Electric Test Bench

Parallel hybrid-electric propulsion systems for small scale unmanned aerial systems (UAS) when tested with an internal combustion engine are susceptible to damage due to increased torque compared to all-electric configurations. The University of Victoria Centre for Aerospace Research has conducted testing and identified several locations in the system for potential upgrades. One of the largest issues identified was the electromagnetic clutch’s inability to handle the torque of the Corvid-50 engine. Thus, a new clutch and powertrain system was specified which is better rated for combustion operation. Similar findings are reported based on testing performed at Ł-Institute of Aviation, where a hybrid-electric powertrain stand experienced torque spikes. The spike amplitude was several times higher than the nominal momentum of the ICE, given in the specification sheets. In result, at some working modes, a strong slippage of the clutch has been observed. It is also of highest importance to propose and test potential methods to minimize momentum influence to the powertrain. The goal is to increase robustness of such hybrid systems, and decrease the overall mass of the system.


Introduction and Background
The Aviation sector is currently responsible for the production of approximately 2% of all human induced CO 2 emissions [1].This figure has remained relatively constant since the 1990s despite increased air activity due to better engine efficiency.This has lead to the production of NOx from increased combustion temperatures [2].Alternative propulsion systems need to be further researched and developed, as a sustainable way to decrease the impact of aviation on the environment.
Hybrid-Electric Propulsion Systems (HEPS) use both internal combustion engines (ICE) and electrical power sources which offers a viable option to mitigate the environmental drawbacks of using ICEs while addressing the challenges of using an all-electrical configuration [3].Although this hybrid approach will not completely eliminate the pollutant emissions it is a sustainable pathway to mitigate the harmful effects of aviation [4].
A parallel hybrid-electric propulsion system can be defined by power being produced by one or both of the sources while the torque is the combination of the two.Parallel configurations possess a higher degree of redundancy since each source is isolated from the other.The main disadvantages of this architecture when compared to series is the added complexity of the system and control methods [5,6].
In 2014, the University of Cambridge project SOUL was the first parallel hybrid manned aircraft, this mid-scale aircraft had a MTOW of 210kg [7].The majority of published work and commercial projects in hybrid propulsion are part of full scale aircraft projects.Multiple projects [8,9,10] involve retrofitting Cessna 337 as test platforms, which have all flown successfully.These projects demonstrate increased fuel efficiency and reduced maintenance cost, with some claiming a reduction of 50-70% and 20-50% respectively [8].However, when looking into small scale UAV projects very little quantitative results can be found on HEPs.There is a need for this research because smaller systems can be used for scale prototype aircraft and further push the possibility of applications for hybrid electric propulsion for future undeveloped aircraft.
The teams at the Center for Aerospace Research (CfAR) and Ł-Institute of Aviation in Warsaw (Ł-Ilot) have been developing test platforms for small scale parallel hybrid propulsion systems for UAVs.The findings and design process are discussed throughout this paper to demonstrate how these systems work and to provide a framework for further discovery in the sector.

Previous Work
The CfAR hybrid project first started in 2017 with the development of theoretical component models such as propellers, batteries, and electric motors.These early stage models were used in the development of a MATLAB framework which simulated three modes of operation: ICE only, EM only, and Series Hybrid.
In 2018 CfAR [5,6] developed version one of the hybrid test bench (HTB) which showed promising results despite having some problems caused by the unpredictability of the ICE and lack of control over the regenerative braking mode.This test bench was improved in 2020 with the implementation of the electric motors (EM), electromagnetic clutch, and load cells coupled to the motors [11].Another group of students built a dynamometer for the test bench which allowed for the implementation of complex loading patterns and power analysis without the use of a propeller.
Throughout 2022 [4] research into the performance of combustion-electric parallel systems was pursued.This consisted of building a graphical interface for the test bench, improving telemetry from the system, integration of a 50cc combustion engine, and gathering performance data.The mechanical system was modified to integrate the combustion engine which replaced one of the existing electric motors.After testing this configuration of the test bench it was clear a redesign and upgrade of the system was required to properly use the ICE and collect clean data for research.The redesign and upgrades performed to the HTB are discussed in detail throughout this paper and are fundamental in understanding how to design a test bench capable of harnessing hybrid-electric power.
The development of a parallel hybrid-electric propulsion system at Ł-Ilot began with the desire to create an unmanned aircraft capable of performing a variety of missions.To build a drone for the application of security and inspection of hard-to-reach areas, the accessibility of which was increased by the use of a hybrid propulsion system.The project is part of the aviation industry's efforts to reduce emissions and noise.Aircraft equipped with this propulsion system will be able to perform different mission profiles with increased reliability and reduced fuel consumption due to energy recovery during certain flight modes.The process began by identifying the necessary basic components which were used as a basis for developing conceptual and structural designs.These initial designs were further refined through practical observations during use to create a final design.
At Ł-Ilot work has been carried out on an unmanned model (1:4.4scale) inspired by the geometry of the PILATUS PC 6 aircraft.The model was equipped with a hybrid propulsion system consisting of an AXI 5345-18 BLDC engine, a 70cc DA-70 two-stroke petrol engine and a two-blade wooden XOAR propeller.Constructed from wood, it features a trussed fuselage with a detachable horizontal tail.The UAV's on-board system includes an autopilot equipped with sensors and radio for communicating with ground-based radio communication devices.

Testing
Running the CfAR test bench with the Corvid-50 internal combustion engine verified the layout and capabilities of the system configuration (Figure 1).Initial low power and RPM tests where executed to verify the integrity of the components and check that telemetry was being captured.Once the system was operating correctly the first higher powered test was performed.The combustion engine ran at up to 40% throttle with the electric motor having segments of augmentation and regeneration (Figure 2).Unfortunately, the clutch and pulleys began to slip once the programmable load was activated at 15A (Figure 3).The upgrades and improvements in the last section should remove these issues and allow for more robust and insightful testing.As shown in the figures below the system was only able to produce 780W at 3000RPM with no load, exceeding this speed and power caused power loss due to slippage.With a lower power of 500W the clutch slipped when a 15A load was applied to the system.These issues did not occur when the system was previously tested in all electric mode, which exceeded 2kW and 4000RPM, verifying that the combustion torque spikes are the cause for pulley and clutch slippage.Testing was stopped to prevent further damage to the system since no further meaningful data could be collected.The laboratory test stand at Ł-Ilot is shown in Figure 4 while the drive system is described in the next section (Figure 8).Two types of laboratory tests were conducted: performance and SIH mode (Simulation In Hardware).These laboratory tests were conducted to determine the performance range of the hybrid propulsion system and verify that the prototype operated correctly in simulated conditions.The safety of the system under these tests were defined by the operating parameters.Testing determined the minimum response time required by the electric motor in the event of a combustion engine failure.Additional tests were also performed on the algorithms to check the feature of automatic ICE startup, system component synchronization management and detection of ICE malfunction during flight.The SIH test used simulation environments to validate all cooperating components in the fully equipped PILATUS aircraft; avionics including autopilot and hybrid drive are working correctly in each phase of operation.A virtual trial flight commenced in electrical mode and subsequently transitioned to combustion mode after the completion of the assumed manoeuvres for automatic mission goals (flight path).The virtual flight featured segments of combustion only, electric only, hybrid, and regeneration modes to collect robust data.The performed tests had demonstrated that the aircraft was working correctly in all planned modes of operation in simulated flight mission conditions providing greater level of confidence before real field flight testing.

Design Improvements and Experimental Setup
The main challenge during testing of the HTB at CfAR was the electromagnetic clutch slipping under load.Consulting the team at Ł-Ilot we found that the combustion engines produce substantial torque spikes which cause the slipping.The clutch that was used meets the nominal torque requirements of the ICE, but not the spikes of up to 40Nm.The second largest issue from integrating the engine was pulleys slipping on the shafts, which also caused tensioners to loosen during operation.
The previous electromagnetic clutch having a maximum torque rating of 5Nm was not adequate to harness the 40Nm spikes created by the ICE.Finding upgrades proved to be challenging because of the high speed requirements of the system.After numerous months of research only a single clutch met the systems requirements.The Mayr Type 500.304.0 has a nominal rating of 40Nm with a tolerance of +50%/-12% and 7000RPM max rotational speed, both of which are at the limits of what the system requires.
Mayr indicates that the clutch needs to be ran for 3 minutes at 50RPM with 8 volts applied to be properly bedded-in.Using a speed reduction pulley system to properly drive the system this run-in procedure yielded a nominal torque of 57Nm, which exceeds the expected torque spikes created by the combustion engine.To improve the operation of this clutch the axial and planar alignment tolerances were tightened by installing a second support bearing between the two plates (Figure 7) and locking the shaft with end caps.With the clutches ability to transmit power the pulleys, belts, shafts, and tensioners required upgrading.The belts were upgraded from 9mm rubber belts to 12.7mm fiberglass reinforced belts.These were combined with aluminum keyed pulleys, adding keyways greatly improves the torque rating of the system and reduced the probability of the pulleys slipping on the shaft.
Lastly, the belt tensioner systems were redesigned from 3D printed parts to machined aluminum plates with idlers.These tensioners are substantially stronger and maintain tension during operation which provides cleaner data.These drivetrain upgrades meet all the expected specifications required to harness the power of the ICE and operate substantially smoother with less resistance compared to the previous configuration.
The powertrain architecture under development at Ł-Ilot is a parallel hybrid system featuring a singular engine shaft.The system comprises of an internal combustion engine, which is connected to the electric motor's shaft via a clutch.The transmission of torque to the propeller depends on the engagement of the clutch.The hybrid propulsion test bench consists of an aluminium base containing the propulsion system connected to a set of two strain gauges and measurement equipment.Force is measured by a strain gauge beam which is deflected when the engine produces thrust.This assembly is mounted to a pendulum shaft which enables torque measurement through a second strain gauge which allows for slight movement around the propeller's axis of rotation.Throughout the test instrumentation records and processes physical data, specifically torque and static thrust against time.The electronic speed controller software is used to record the generator's data.
The primary issue identified was a lack of sufficient static torque transmission from the initially chosen clutch.At this stage we encountered an issue with the concentricity of the clutch discs which destroyed the clutch linings due to heat and friction.This slipping resulted in the combustion engine not starting and the electric motor to malfunction.The ESM1-15-24 friction clutch was replaced with an EZM1 electromagnetic tooth clutch with a static torque of 50 Nm.This toothed clutch requires both sides of the disc to be equalized before coupling, which in practice necessitates reducing the engines speed to nearly zero before coupling and re-engaging.Also, the main shaft was not strong enough to transmit the coupled power.FEA analysis revealed that the component required a redesign and a change of material to enhance its strength.This led to an Inconel shaft being strengthened through machine undercutting which facilitates the proper transfer of torque from the engines to the propeller.
The powertrain has four operating modes, which include electric mode, combustion mode, hybrid with electric motor priority, and hybrid with combustion motor priority.In electric mode, the clutch is disengaged and the electric motor torque is controlled by the throttle joystick.When entering combustion mode with the electric motor running the ignition system is started and the clutch is safely engaged once the electric motor is braked.The idle parameter in the application sets the time after which the internal combustion engine turns off automatically when not in use.If the battery packs are not fully charged, they are recharged with low current.In electric priority hybrid mode the combustion engine charges the battery packs if they are not fully charged (90%).In hybrid mode with the combustion priority, the joystick controls the engine up to its mid-range.The electric motor takes over beyond this point.

Discussion
The testing and simulation studies for Ł-Ilot's design confirmed the proper function of the hybrid propulsion system's various components.A fully compatible test stand integrated with the actual unmanned aerial vehicle system enabled the validation of the system's programmed control algorithms for different operating modes.These tests showed that the system is capable of outputting up to 285N of thrust with 22.23A of current for regeneration if desired.These results were further solidified using the hybrid PILATUS platform which completed several sequences, ran error-free and in a short time, which is a key aspect of real flight.The real system simulation flight execution facilitated unmanned aircraft behaviour verification and reduced potential flight failure risk.The data collected from these tests further increase the development of small scale hybrid systems at Ł-Ilot.
The introduction of combustion power into the all-electric parallel test bench at CfAR exposed some weak points in the design.Initial testing showed that the system was only capable of holding roughly 780W of power at 3000 rpm, once that speed was exceeded the clutch and pulleys began to slip.At a lower power of 500W the clutch also slipped when a 15A load was applied via the dynamometer.This clutch and pulley slippage is due to torque spikes of approximately 40Nm from the engine.The slip that occurs in the drivetrain is not experienced in all-electric mode which exceeds 2kW and 4000RPM, confirming the torque spikes of the ICE are the cause.To harness the torque a Mayr clutch was installed that can transfer 57Nm, ten times that of the previous clutch, along with keyed pulleys and wider fiberglass reinforced belts.These upgrades will allow for higher power testing on the test bench and a better understanding of how small scale parallel hybrid-electric systems work.

Future Work and Conclusions
The work on the parallel hybrid propulsion system at Ł-ILot allowed us to summarise our achievements and reach conclusions about the potential for further development of the test bench.The propulsion controller software functions properly and ensures the safe execution of the mission in multiple modes of operation; electric, combustion, and hybrid.The complete range of planned tests were conducted on the laboratory bench and simulated on the equipment during the virtual flight.The system was able to run at 6571RPM producing 285N of thrust and 22.23A of current which can be used for regeneration.These results drive the team to continue development of their hybrid system and further push research in the field of small scale hybridization.
The test bench at CfAR was not capable of transmitting power from the combustion engine, with system slipping and becoming damaged at 780W and 500W with a 15A load.The system was redesigned and will be tested later this year to verify the stronger clutch, pulleys, and belts.Once operating correctly the system will be used for research into scalability, optimization, and higher degrees of hybridization.The principles learned on this test bench will be used to create an integrated package that is flight ready allowing the center to collect real world data that increases understanding of small scale hybrid-electric propulsion in UAVs.

Figure 5 .
Figure 5. Diagram of RPM, electric motor current and thrust registered during test in configuration C

Figure 8 .
Figure 8. Final design of the Ł-Ilot Parallel Hybrid-Electric Powertrain

Table 1 .
Test results in Configurations A/B/C/D