Hydrogen-Fueled Rotary Engine

Prediction and analysis of nitric oxide emissions

For the completion of my Master’s degree at the Catholic University of Louvain, I chose a thesis topic aligned with my future career interests, the automotive industry. The university had discontinued a previous project involving the restoration and hydrogen conversion of a 1989 Mazda 13B Series 5 rotary engine.

The thesis was divided into two parts: the experimental work focused on hands-on skills for restoring the engine to run on gasoline and initiating its conversion to hydrogen fuel, while the theoretical work involved predicting and analyzing engine performance and nitric oxide emissions in the exhaust gases.

Context

Stored for more than 15 years, this engine was missing its electronics and required a complete restoration for gasoline operation before considering its conversion to hydrogen.

The objective for this engine was to compare its performance and emission characteristics using different fuels, and to benchmark it against conventional internal combustion engines through both simulation and testing.

Challenges Encountered

Throughout this project, I faced numerous challenges that were successfully overcome through creative problem-solving, innovation, and effective team communication during regular meetings. Since the project had previously been undertaken by other students, my first task was to thoroughly review and understand the work that had already been done on the engine. This deep investigation not only helped me get up to speed but also led to established the budget required for this project which I would need to manage throughout the project.

M84 ECU

The initial plan was to manage the engine using the MoTeC M84 ECU, which is capable of running rotary engines. However, the M84 does not support the oil metering pump (OMP) management required by the Mazda 13B engine.

First approach: developing a custom OMP control system using an Arduino, but synchronizing it with the high engine speeds of the rotary engine proved too challenging.
Final solution: invest in the more advanced MoTeC M130 ECU with the dedicated rotary engine package. The M130 not only supported OMP management but also accommodated different types of injection systems compatible with hydrogen fuel.

Part Replacement

One of the major challenges in restoring a vintage engine is the scarcity of spare parts. After careful individual component testing, some parts presented issues:

Oil Metering Pump (OMP): This pump lubricates the rotor inside the housing. We identified a malfunction in the spring that actuates the ECU-controlled valve. To resolve the issue, we disassembled the component for a detailed inspection, successfully diagnosing and repairing the fault—saving costs and preserving the project budget.
Starter: Early Mazda RX-7 models were known to have a starter that was too weak to reliably start the engine. To overcome this, we modified the internal electric motor to increase its starting torque.
Oil leaks: After circulating oil through the engine prior to start-up to ensure proper lubrication, we detected leaks. Using a borescope, we located the sources and subsequently shut down the engine to tighten loose bolts, resolving the issue.

Bench test construction

To ensure accurate measurement of the engine’s performance and emissions, we needed to build a test bench capable of handling the engine’s power while ensuring user safety.

Safety cabin: The engine is housed in a secure enclosure to protect users from potential hazards. This required routing the engine’s electronic cables and controls outside the cabin to allow safe operation.
Bench test: Based on existing engine test benches, we performed calculations to ensure the transmission shaft connecting the electromagnetic brake could handle the engine’s high power, which exceeded the capacity for which the original setup was designed.
Cooling system: The main challenge was managing water flow to maximize cooling around the combustion chamber walls. This was especially critical for hydrogen use, as exposure to hot surfaces could cause pre-ignition of the air-fuel mixture, reducing performance or even damaging the engine.

Learning outcomes

Practical

This project provided me with invaluable mechanical and electrical hands-on experience in component testing, engine assembly, and the design and initialization of electronic systems. The electronics and control systems had to be completely rebuilt using MoTeC's GPR Rotary package.

Technical

This thesis also offered a valuable opportunity to enhance my programming skills, particularly in Python. I used the language to simulate engine operation and model chemical reactions within the combustion chamber, focusing on predicting and converting nitric oxide (NO) emissions.

The foundation of this work was based on Ferguson’s research on piston engines, which I adapted for rotary engine applications. This required modifying the equations to account for the varying temperature distribution within the housing. Unlike piston engines—where intake, combustion, and exhaust occur in the same region and a constant temperature can be assumed—rotary engines separate these phases spatially, leading to non-uniform thermal conditions.

This adaptation led to an important insight: rotary engines may offer a significant advantage over piston engines in hydrogen combustion applications, particularly regarding NOx emissions. Because the combustion gases in rotary engines pass through cooler areas of the housing before being expelled, the formation of additional nitrogen oxides is reduced.

Leadership and team work

Throughout this project, I had the opportunity to lead and coordinate a team of seven members, including mechanics, electro-mechanics, and electricians. As this project served as my final year thesis, I was responsible for managing the team and assigning tasks following our weekly meetings. This process allowed me to appreciate the complexities of leadership, ensuring that deadlines were met and everyone contributed effectively. The experience can be considered a success, as strong relationships were maintained among all team members throughout the project.

Work organisation

Originally intended for a group of two or three students, this project was ultimately assigned to me alone, requiring significant organizational effort to stay on track with both the experimental and theoretical work. To manage this, I used an organizational tool like TRELLO, which was essential for meeting my deadlines and allowed me to quickly adapt to any last-minute challenges.

Budget management

This project came with an allocated budget that I was expected to adhere to as closely as possible. The budget was initially determined through an early evaluation of the engine's condition, which helped estimate the remaining work and identify the parts that needed to be ordered. However, mistakes made in previous years by other students revealed that some parts were unsuitable for the engine, requiring additional funding. Initially, these overages weren’t a concern due to the safety margin built into the budget. However, as more hidden issues with the engine were uncovered, further financial support from the university's research department became necessary. This situation gave me the opportunity to present and justify the additional expenses to the committee.