Bearing temperature control in stacked geared engine architectures
The thermal envelope of a modern geared turbofan depends on a chain of small hydrodynamic decisions that, taken together, decide whether a bearing survives its rated life or quietly succumbs to scuffing and brinelling. Nowhere is this more evident than in stacked gearbox configurations, where two reduction stages share a common housing and the available volume for scavenge pumps, sumps, and supply lines compresses dramatically. Engineers designing for power density above 200 kW/kg quickly discover that the lubrication system is no longer a peripheral utility but a structural element of the drivetrain.
Recirculating oil jets sit at the heart of this thermal problem. The jet must first reach the bearing cage with enough momentum to displace hot oil already clinging to the rollers, then recover and return to the sump without aerating or foaming. When the jet geometry is wrong, the bearing can run ten or fifteen degrees hotter than its thermal limit even when the bulk oil temperature reads healthy. Conversely, a well-tuned jet can drop bearing temperature by twenty degrees or more under identical load and speed, opening headroom for higher ratings or smaller, lighter components.
The OPTIMIZE Project has invested considerable effort in mapping these interactions across the full operating envelope, using design-of-experiments methods that span materials, surface treatments, and geometry. Earlier work examined how a DoE diversity study revealed coupled effects between case-hardened steel grades and macro-geometry, and the same methodology now applies to jet placement. Australian researchers in Adelaide and Melbourne have been directly involved in the test campaigns, contributing to instrumentation protocols and post-test inspection at facilities such as those operated by Defence Science and Technology Group.
For Australian aerospace engineers, the relevance is immediate. Local airframers and engine component manufacturers from Brisbane to Perth are responding to new Defence Strategy Review commitments and growing demand for unmanned platforms that operate in the country's harsh northern climates, where ambient temperatures at low altitude routinely push gearbox cooling systems beyond their design point. CASA airworthiness expectations for powerplant cooling, captured in the relevant chapters of the Air Navigation (Aircraft Engine Standards) Orders, make it essential to demonstrate thermal margin with test data rather than purely analytical claims.
How recirculating oil jets cool high-speed bearings
The fundamental job of a recirculating jet is to deliver a controlled volume of cool oil at a known velocity onto the inner race of the rolling element bearing, displacing hot oil before it can degrade the lubricant film. In a stacked gearbox, the jet usually enters through a dedicated nozzle in the shaft or in a stationary sleeve, angled obliquely to the raceway so that the oil spreads along the rolling direction rather than running straight into the next downstream component. The jet also performs a secondary duty: it scavenges itself, in the sense that the returning oil carries away heat and any wear debris generated in the contact.
The effectiveness of the jet depends on three coupled parameters: nozzle diameter, supply pressure, and impingement angle. A larger nozzle at the same pressure delivers more mass flow but at lower velocity, which can cause the oil to cling to the nozzle exit rather than penetrate the cage window. A smaller nozzle gives sharp, focused impingement but is easily starved during transient manoeuvres such as a rapid climb-out from a regional Australian airport on a hot afternoon. The impingement angle, often overlooked, determines whether the jet spreads along the race or simply ricochets into the next gear mesh.
Researchers have found that small changes in standoff distance produce disproportionately large temperature shifts. Pulling a nozzle back by just two millimetres can raise bearing temperature by eight to twelve degrees at 20,000 rpm, because the jet loses momentum through air resistance inside the housing before it reaches the critical zone. Pushing the nozzle closer is not always the answer, as impingement pressures that are too high locally can erode the race surface and accelerate white-etch cracking in the carburised case.
Stacked gearbox geometry and thermal constraints
In a stacked arrangement, the upper stage typically sits directly above the lower stage, sharing structural bolts and a common lubrication circuit. This compactness is the whole reason for adopting the architecture in the first place, but it also means that the oil leaving the upper bearing must travel through a confined return path before it can rejoin the supply circuit. Anything that impedes this return path, whether a poorly placed rib in the housing or a sharp bend in a scavenge line, will raise the pressure at the bearing exit and force hot oil back into the inlet of the jet.
The geometry of the oil return channel has received detailed treatment in a recent piece of work on sump scavenging, where the return channel geometry study showed that sweeping fillets and gradual expansions reduce local pressure losses by roughly forty percent compared with abrupt transitions. That finding translates directly into bearing temperature performance, because a lower return-side pressure means less back-flow into the bearing cavity and a more stable inlet condition for the jet.
Some of the most sensitive geometric variables in a stacked configuration include:
- Jet nozzle standoff distance from the inner race
- Bearing cavity free volume between the two stages
- Scavenge pump inlet placement relative to the upper bearing
- Local rib and boss geometry that disrupts return flow
- Shaft shoulder geometry that interferes with the jet cone
Each of these interacts with the others. A change to the scavenge pump inlet may look like a hydraulic decision in isolation, but in practice it alters the pressure field around the jet and can either help or hinder thermal performance depending on the bearing cavity volume.
Measured outcomes from instrumented bearing tests
Instrumented tests on representative cylindrical roller bearings have generated repeatable trends that engineers can use to bracket their design space. In a series of runs at 18,000 and 22,000 rpm, the bearing housing temperature was logged both at the outer race shoulder and at a point fifty millimetres upstream of the jet nozzle, allowing the team to separate the heat picked up in the bearing from the heat already carried in the supply oil.
Key findings from the test matrix include:
- Optimal jet velocities cluster in a narrow band between 35 and 50 m/s at the nozzle exit
- Bearing housing temperature rises by approximately 0.8 °C for every 1 °C of bulk oil temperature above 110 °C
- Aerated oil at the jet inlet raises bearing temperature by an additional 5 to 8 °C independent of bulk temperature
- Shaft tilt of more than 0.05 mm at the bearing seat shifts the impingement point and degrades cooling
The aeration finding is particularly important for stacked gearboxes, because the upper bearing tends to draw air down from the breather, especially during descent when ambient pressure rises faster than internal cavity pressure. Foamy oil has lower bulk modulus and a reduced capacity to carry heat, so even a small volume fraction of entrained air can wipe out the gains made by careful jet placement.
Interaction with sump scavenging and return channels
The thermal performance of a recirculating jet is meaningless without an efficient scavenge system to remove the heated oil. In a stacked gearbox, the scavenge pump must overcome not only the bearing cavity pressure but also the hydrostatic head of oil sitting in the return line during ground operation. When the engine is shut down on the apron at Sydney or Adelaide airports, the upper bearing cavity can remain flooded for several minutes as the residual heat continues to soak into the housing.
Designing for this ground-soak condition requires a different mindset from designing for steady-state flight. The team has explored auxiliary scavenge paths that activate only at low shaft speeds, drawing oil out of the upper bearing cavity through a dedicated channel. Early results suggest that a correctly sized auxiliary path can reduce bearing temperature during the critical first ten minutes after shutdown by up to twelve degrees, which has direct implications for post-flight inspection intervals and overall maintenance burden.
The Australian context reinforces this point. Royal Australian Air Force and civilian operators flying from hot-and-high airfields such as Mount Isa, or operating through the humid tropics around Cairns, routinely experience bearing soak-back scenarios that European test campaigns rarely replicate. The OPTIMIZE Project's collaboration with the University of South Australia and RMIT has deliberately incorporated these conditions into its test matrix to ensure relevance to local operational realities.
Implications for Australian aerospace programmes
Australia's defence and commercial aerospace sectors have a growing stake in advanced gearbox technology. The upcoming MQ-28A Ghost Bat programme, Boeing's Loyal Wingman derivative, and a series of unmanned cargo demonstrators all rely on compact, high-power-density propulsion systems. Local manufacturers such as Rosebank Engineering in Melbourne, CNC Design in Perth, and Magellan Aerospace in Brisbane already supply gears and shafts into global supply chains, and thermal performance data of the kind generated by OPTIMIZE directly informs their ability to bid on next-generation components.
The legislative backdrop also matters. Defence Trade Controls Act 2012 governs the export of controlled goods and technology, including certain gearbox design data, so any Australian partner participating in the OPTIMIZE Project must navigate both ITAR-equivalent restrictions and DSTG export protocols. CASA Manual of Standards Part 33 sets the airworthiness expectations for engine installations, including cooling system substantiation, and CASA's recent engagement with EASA on integrated propulsion systems means that local designers must increasingly harmonise their thermal margin arguments with European and American practice.
The combination of validated test data and practical design levers gives local engineers a head start when responding to tenders that demand demonstrated thermal margin. In practice, Australian engineers are applying a set of conservative design rules: jet nozzle diameters are specified within the validated 1.6 to 2.2 mm band, standoff distance is held to within ±0.3 mm of the design value through precision assembly fixtures, scavenge lines are routed with swept bends rather than sharp elbows, de-aeration elements are fitted at the scavenge pump inlet, and bearing soak-back performance is documented as part of the maintenance programme. This conservative approach also reduces the risk of late-stage design changes when the gearbox is integrated into an airframe programme that has its own thermal and packaging constraints.
Getting involved with the OPTIMIZE Project
Engineers and programme managers who want to explore how the findings on recirculating oil jets and stacked gearbox thermal management could apply to their own platforms are invited to contact the project team. Active test campaigns are expected to continue through the next design cycle, and opportunities exist for collaborative research with Australian universities and for supplier engagement through existing industry liaison programmes.
Early involvement typically shortens the learning curve considerably, particularly for organisations that are new to high-speed gearbox integration. The project can share validated test methodologies, instrumentation guidance, and post-test inspection protocols that would otherwise take a new entrant several years to develop independently. This is especially valuable for small and medium-sized Australian manufacturers looking to break into the global aerospace supply chain, where demonstrating thermal margin with traceable test data is now a baseline expectation rather than a differentiator.
Those interested in following the project's progress can also access the published articles and conference papers through the website, and subscribe to updates on upcoming test campaigns and workshop events. The combination of open methodology and controlled partner engagement has been a deliberate feature of the project from its inception, reflecting the consortium's belief that pre-competitive research on power reduction gearboxes benefits the entire aerospace community, not just the immediate programme partners.