Influence of Flow Coefficient on Ingress Through Turbine Rim Seals

Abstract

Rim seals are critical in terms of limiting the temperature of highly-stressed engine components but function with a penalty to the power output and contribute to entropy gain stemming from mixing losses in the turbine. Ingress through rim seals is influenced by the presence of rotor blades and stator vanes, and the mainstream flow coefficient in the annulus that determines the corresponding swirl. This paper presents an experimental study of ingress upstream and downstream of the rotor disc in a 1.5-stage rig with double radial clearance rim seals. Two rotor discs were used, one with blades and one without, and two platforms were used downstream of the rotor, one with vanes and one without. Tests were conducted at two rotational speeds and a range of flow conditions was achieved by varying the annulus and sealing mass flow rates. Concentration effectiveness, swirl and steady pressure measurements separated, for the first time, the influence of the blades and vanes on ingress over a wide range of flow conditions. Measurements on the downstream stator platform provide added insight into the complex interaction between the egress and the mainstream. Measurements of unsteady pressure revealed the presence of large-scale structures, even in the absence of blades. The number and speed of the structures was shown to depend on the flow coefficient and the purge flow rate.

A New Interpretation of Hot Gas Ingress Through Turbine Rim Seals Influenced by Mainstream Annulus Swirl

“…Measurements of effectiveness were conducted in the downstream wheel-space with a vaneless stator installed downstream of the rotor. The superposition effect on sealing effectiveness of a downstream vane row, and the varied influence with CF, was described in detail by Graikos et al [12] and does not feature in this study. A downstream vane will affect the annulus swirl at flow coefficients that correspond to a GTP-22-1355 Scobie 17 sweep (or arc) of absolute flow angle equivalent to an impingement on the leading edge of the downstream vane.…”

“…Effectiveness measurements were acquired in the downstream wheel-space at r/b = 0.958, as marked in Figure 9. This measurement position is considered representative for the wheel-space as previous publications [12] using this seal have shown only a minor variation in the radial distribution of effectiveness on the stator and rotating core. were collected at constant rotational speed with the flow rate in the annulus varied to generate a range of CF.…”

“…These low-order models also only predict a monotonic relationship between effectiveness and purge flow. Graikos et al [12] have shown that ingress is influenced by the interaction between the emergent purge from the rim seal and the impingement on downstream vanes and blades.…”

“…a rotating ring). The low-and high-reaction designs were used respectively by Patinios et al [15] and Graikos et al [12]. Note high and low are used to describe these rotors despite 0.3 < Λ < 0.47 not being considered a high degree of reaction in terms of turbomachinery.…”

“…Scobie 29 presented here at lower radius through the inner rim seal. The inner wheel-space would be fully sealed over a wide range of β at Φ0 = 0.018 [12].…”