Leafs seals are an advanced sealing technology to separate high and low pressure parts in turbomachinery. They consist of small metal sheet elements arranged in a housing around the shaft. During operation the seal should be non-contacting to minimize wear and friction. The static pressure on the surface of a single leaf appears to be the main parameter. The higher axial stiffness in comparison to brush seals allows greater differences in pressure and a hang-up effect can be averted. Assessing the discharge behaviour of the sealing is important, because leakage is resulting in less efficient turbomachinery.
In this paper a new approach for an analytical calculation of the pressure distribution inside the seal is presented. It is based on a leakage flow model inside the side plate gaps. Furthermore the flow is assumed to be laminar and incompressible using simple pipe flow correlations with specified hydraulic diameters. The complete seal geometry is parameterized. The results in pressure distribution, leakage flow and static pressure on a single leaf (lift-up, blow down) are compared with the data of Nakane et al. (2004). The choice for their design is based upon their publication of the complete geometric data of the seal. Also the works of Ortner (2009) and Bischelmaier (2011), who made further investigations of this seal design, are taken into account. A good qualitative match was found, so in a second step, variations of the seal design were calculated. The influence of upstream and downstream side plate gaps ratios, heights of rotor side plate gap, numbers of leaves for constant rotor diameter and interleaf gap-gradients are investigated. For validation of the model, the results are compared as far as possible to published data. Also calculations are made for patented designs using leaves with non-rectangular cross-section to estimate, if further and more detailed research is reasonable.
The leakage flow model provides a simple approach for the investigation of leaf seals regarding pressure distribution and seal-leakage with qualitative good results. Due to this, it is a good tool for fast and easy design variations compared to 2D or 3D CFD simulations.