Thermal and Flow Phenomena Associated With the Behavior of Brush Seals in Aero Engine Bearing Chambers

Abstract

Due to the increasing fuel cost and environmental targets, the demand for more efficient gas turbines has risen considerably in the last decade. One of the most important systems in a gas turbine is the secondary air system, which provides cooling air to the disks and to the blades. It also provides air for sealing of the bearing chambers. The amount of secondary air that is extracted from the compressor is a performance penalty for the engine. In aero engines, bearing chambers are in most cases sealed by the most traditional type of seal, the labyrinth seal. Bearing chambers contain the oil lubricated components like bearings and gears. In order to avoid oil migration from the bearing chamber into the turbomachinery, the seals are pressurized by secondary air; thus, a pressure difference is setup across the seal, which retains the lubricant into the bearing chamber. Oil loss can lead to a number of problems like oil fire or coking with the probability of an uncontained destruction of the aero engine. Oil fumes can also cause contamination of the air conditioning system of the aircraft thus cause discomfort to the passengers. Beside labyrinth seals, other types of seals such as brush seals and carbon seals are used. Both the latter are contact type seals, that is, they may be installed with zero gap and lift during operation when they get pressurized. Brush seals particularly may be installed having an overlap with the rotating part. An original aero engine bearing chamber was modified by MTU Aero Engines to run with brush seals in a simulating rig in Munich. Two types of brush seals were used for testing: (a) a brush seal with bristles made of Kevlar fibers and (b) a brush seal with bristles made of steel. Both types were installed with an overlap to the rotor. The targets set were twofold: (a) to measure the transient temperatures in the rotor and particularly in the contact zone between the bristles and the rotor and (b) to calculate the heat generation by the seals which could enable predictions of the heat generation in future applications (i.e., scaling to bigger rotor diameters). For the heat transfer calculations, numerical models using ansys cfx were created. Additionally, a coupled computational fluid dynamics (CFD) and finite element analysis (FEA) approach was applied to simulate flow and bristle's behavior. In order to obtain the transient temperature measurements with high fidelity, a new pyrometric technique was developed and was applied for the first time in brush seals as reported by Flouros et al. (2013, “Transient Temperature Measurements in the Contact Zone Between Brush Seals of Kevlar and Metallic Type for Bearing Chamber Sealing Using a Pyrometric Technique,â€‌ ASME J. Gas Turbines Power, 135(8), p. 081603) and Flouros et al. (2012, “Transient Temperature Measurements in the Contact Zone Between Brush Seals of Kevlar and Metallic Type for Bearing Chamber Sealing Using a Pyrometric Technique,â€‌ ASME Turbo Expo 2012, Copenhagen, Paper No. GT201268354). This technique has enabled positioning of the pyrometer (SensorthermGmbH, www.sensortherm.com) into the bristles pack of the seal adjacent to the rotating surface. The pyrometer could record the frictional temperature evolution in the bristles/rotor contact zone during accelerations or decelerations of the rotor. The sealing air demand can be reduced up to 97% with brush seals compared to traditional three fin labyrinth. It has been estimated that this can result in a reduction in fuel burned up to 1%. Further, the reduction in air flow has additional potential benefits such as a possible simplification of the bearing chamber architecture (vent less chamber). Even though the rotor was accelerated up to 19,500 rpm, the temperature induced overshoots in the seal/rotor contact zone have caused no deterioration in either the materials or the oil.