January/February 2024 | E-Mobility Engineering 26 stacking factor, as well as continuous performance at 300 C without any adhesion degradation. Rotor performance For its rotors, H3X receives 0.1 mm laminated iron sheets (either cobalt or silicon iron, depending on the application), as well as laminated and segmented permanent magnets at its facility. It then stacks the sheets, installs the shaft through the middle of the stack and inserts the magnets into the assembly, all using its proprietary adhesive for bonding. Lastly, some outer-diameter grinding and balancing is performed by third parties, although H3X plans to bring these processes in-house. “We spin-test our rotors to verify their performance at room temperature and in a temperature chamber at 300 C,” Liben says. “Our rotor is not actively cooled; it has to reject its heat into either the stator through the air gap or out through the bearings. That means the rotor can get even hotter than 265 C, sometimes up to 300 C, given that it runs inside the stator, farther away from the housing and its cooling jackets.” While the magnets and laminations tolerate 300 C without issue, the polymer adhesive imposes a temperature limitation. Structurally, the rotor primarily depends on the metal lamination geometry, and secondarily on the adhesive, to not burst apart. Traditional aerospace e-machines wrap a carbonfibre sleeve around the rotor to compress the rotor components together, which is particularly useful for surface permanent magnet (SPM) architectures. “So, again, the polymer adhesive is key to the rotor’s performance at thermal limits. It doesn’t play a huge role structurally in the final component since, mechanically, the laminations themselves provide far greater structural strength, but breakdown of the adhesive due to aggressive operating temperatures would still be unacceptable,” Liben notes. Magnetic attraction To create a motor that can operate efficiently even while extremely hot (without needing direct rotor cooling), samarium cobalt magnets were deemed the best choice. “Past about 180 C, samarium cobalt gives you more air-gap flux density than neodymium. It’s expensive, but because of our high stator current, our power or torque relative to magnet mass [or magnet utilisation] is very good, so we don’t need a large quantity of magnetic material to produce 12 kW/kg of specific power,” Liben explains. Additionally, the interior permanent magnet (IPM) structure achieves a higher air-gap flux density than SPM rotors with a sleeve. A typical sleeve acts similarly to air, creating distance between the stator and rotor, reducing the flux density, which can cross the air gap to create electromotive force and torque. “We didn’t go the IPM route strictly for power density,” says Liben. “We determined that using an SPM rotor with a sleeve and spinning the rotor much faster would net even more power output, which could suit future turbogenerator or turbocharger applications, but that’s not optimal for primary propulsion or primary power generation at the MW scale.” The IPM arrangement has been chosen to sacrifice significant speed capability in order to gain torque. If an e-motor must drive a propeller, the propeller’s blade-tip speeds will rapidly lose efficiency if forced to run close to the speed of sound, and the rotors will similarly lose efficiency if run too quickly. A gearbox can help to get around these limitations (one is integrated into the HPDM-250). H3X has determined that to maximise power density in e-powertrains running at tens or even a couple of hundred kilowatts, it makes sense to run the e-motor at high speed and use a gearbox to convert speed into torque. Meanwhile, the MW-class powertrains can be run with slow enough shaft speeds and high enough output torque for direct drive operation. “The necessity of a gear reduction at lower power stems from maintaining a reasonable aspect ratio of the e-machine to minimise the impact of end leakage on the air-gap flux density, while avoiding modal issues that can stem from very large or very small rotor L/D aspect ratios,” explains Liben. He estimates that 320 kW is the approximate cut-off point above which a gearbox starts to become unnecessary and prohibitively large for increasing output torque relative to speed. The mechanical simplicity of direct drive is also seen as a significant benefit, especially for high-reliability MW-class applications. Dossier | H3X In the HPDM-250, the SiC inverter boards are laid in a hexagonal arrangement around the inner housing
RkJQdWJsaXNoZXIy MjI2Mzk4