INNOVATIVE SHALLOW SUMP CUSTOMIZATIONS FOR AERO-ENGINE BEARING CHAMBERS

Aero-engines incorporate various bearing chambers and these typically contain bearings, seals, rotating shafts, stationary walls and struts, and sometimes gears. Oil is supplied for lubrication and cooling and is removed (scavenged) from the sump region of the chamber (note that in some parts of the world the entire bearing chamber is referred to as the sump). Depending on the location and function of the bearing chamber, the sump region may be deep or shallow. Effective oil removal is essential as unnecessary working of the oil can lead to excessive heat generation and reduced overall efficiency. Therefore the design of the scavenge region in a bearing chamber, as well as the ability to assess its performance is very important. Previous work, much of which was conducted at the University of Nottingham Technology Centre in Gas Turbine Transmission Systems (UTC) suggests that oil often does not flow cleanly into the off-take due to a combination of several factors: oil momentum, windage, three-dimensional air flow that blocks the off-take flow or transports oil away from the off take, and pooling because of separated air flow that acts on the oil once oil momentum is dissipated. Experimental research at the UTC found that scavenge performance is highly affected by the sump geometry, especially its depth. Variations of shallow sumps, although some are better than the others, cannot offer the same level of performance as a deep one. However space limitation in an engine often only allows for a shallow sump. This paper presents some experimental exploration on new design ideas. They are in the form of various inserts and attachments that were designed to improve scavenge performance of a shallow sump. These "custom" sumps were tested on the UTC's scavenge test facility at various flow settings (wall film/flying droplets, liquid flow rates, scavenge ratios, shaft speeds). The residence volumes were measured and compared to a baseline configuration with reduction in residence volume desirable. The inserts tested were a Grille Cover, a Stepped Spillway, a Perforated Plate and a Porous Insert. Both the Porous Insert and the Perforated Plate showed reduced residence volumes in the demanding droplet/windage dominated flow condition with the Perforated Plate offering the best improvement over baseline.