An Experimental Study of Gas Turbine Rim Seals

Student thesis: Doctoral ThesisPhD


Gas turbines are one of the most widely-used power generating technologies in the world today. In the face of climate change and continued global financial pressures placed on industries, one of the biggest challenges facing engine designers is how to continually improve turbine efficiencies.Rim seals are fitted in gas turbines at the periphery of the wheel-space formed between rotor discs and their adjacent casings. These seals reduce the ingestion of hot gases that can cause catastrophic damage to some of the most highly stressed components in the engine. In gas turbine engines this ingestion is principally caused by circumferential pressure asymmetries in the mainstream annulus, radially outward of the rim seal. A superposed sealant flow, bled from the compressor, is used to reduce or, at the limit, prevent ingestion. As the use of this sealing air can reduce the cycle efficiency, it is important to know how much flow is required to prevent ingestion and to understand the associated fluid dynamics and heat transfer when ingestion occurs.This thesis presents experimental results from a specifically designed research facility which models an axial turbine stage with generic, but engine-representative, rim seals. The test section featured stator vanes and symmetrical rotor blades. Measurements of pressure, CO2 gas concentration and swirl ratio are used to assess the performance of different seal designs. Although the ingestion through the rim seal is a consequence of an unsteady, three-dimensional flow field, and the cause-effect relationship between pressure and the sealing effectiveness is complex, the experimental data is shown to be successfully calculated by simple effectiveness equations developed from a theoretical orifice model.Effectiveness data were collected at the design condition for a datum radial-clearance single seal, and compared with a double overlap equivalent and a further derivative with a series of radial fins. The benefit of using double rim seal configurations was demonstrated, where the ingested fluid was shown to be predominately confined to the outer wheel-space between the two sets of seals. The radial fins increased the level of swirl in this outer wheel-space, rotating the captive fluid with near solid body rotation. This improved the attenuation of the pressure asymmetry which governs the ingress, and improved the performance of the inner geometry of the seal. A criterion for ranking the performance the different seals was proposed, and a performance limit was established for double seals, in which the inner seal is exposed to rotationally induced ingress only. Experiments were also performed at off-design conditions, where the effect on ingress of varying the flow coefficient (CF) was demonstrated for both under-speed and over-speed conditions. The correlated effectiveness curves were used to predict the required levels of sealant flow to prevent ingestion, and the variation with CF was in mainly good agreement with the theoretical curve for combined ingress, which covers the transition from rotationally induced to externally induced ingress. Departure of the measured values from the theoretical curve occurred at very low values of CF for all the seals tested. This was attributed to flow separation at large deviation angles between the flow and the symmetric turbine blades.The effectiveness measurements determined from gas concentration were then used to establish a new effectiveness based on pressure. A hypothetical location on the vane platform was assumed to exist where the measured pressures would ensure consistency between the two definitions. Experimental measurements for a radial clearance seal showed that as predicted, the normalised pressure difference across the seal at this location was linearly related to the pressure difference at an arbitrary location on the vane platform. When compared to the original concentration effectiveness measurements, good agreement was found with the values of effectiveness determined by the theoretical pressure model. It was shown in principle how parameters obtained from measurements of pressure and concentration in a rig could be used to calculate the sealing effectiveness in an engine.The design of a novel 1.5-stage facility, complete with representative turned rotor blades, is then described. The rig experimentally models hot gas ingestion in a downstream, as well as an upstream wheel-space. The methodology behind the design process was outlined, and details were given on the proposed design operating conditions. Experience gained from conducting experiments in the previous facility heavily influenced the design of the new rig. The instrumentation capabilities have been summarised and an explanation of the intended measurements given.
Date of Award26 Jun 2014
Original languageEnglish
Awarding Institution
  • University of Bath
SupervisorGary Lock (Supervisor) & Michael Wilson (Supervisor)

Cite this