The performance of the last stage of a Low Pressure (LP) steam turbine is strongly coupled with the downstream exhaust hood performance. In particular, the effect of the diffuser within the exhaust hood on the pressure recovery is very important in retrofitting existing machines, which dictate many geometric constraints. Alstom’s in-house Exhaust Design System (EDS) simulates the three-dimensional flow in the exhaust hood by coupling the last stage blades and the exhaust hood. This EDS system can be used to design an LP diffuser in the exhaust hood and to achieve the required performance targets. In the first part of this paper, the EDS system is validated against measurements within model turbines, which represent both a standard machine as well as a retrofit machine. In the second part of this paper, an LP diffuser was redesigned to improve the performance using the EDS method. To begin with, an axi-symmetric diffuser was designed using numerical simulations of a passage in the last stage turbine as well as a slice of the diffuser and the exhaust hood. By carefully controlling the profile of the diffuser casing, the flow separation at the original casing walls was reduced significantly and this, in turn, improved the performance of the turbine substantially. Then, the full geometry of the exhaust hood was modeled in order to investigate the effect of the three-dimensional flow features. Based on the examined flow features, an asymmetric change was introduced to the diffuser casing to improve the three-dimensional flow structure. This new asymmetric diffuser was found to maximize the exhaust performance.

This paper presents the design and analysis of a multipurpose combined function magnet for use in accelerators. This magnet consists of three corrector magnets: (i) skew quadrupole, (ii) horizontal dipole, and (iii) vertical dipole magnets, along with the main sextupole magnet. The strength of the corrector magnets is smaller than that of the main sextupole magnet. The strength of all the four magnets can be varied independently. The excitation strength required to produce skew quadrupole gradient and the presence of various multipole components in the magnet are estimated using first order perturbation theory. The experimental data for the variation of the sextupole strength and its higher order multipoles in the presence of skew quadrupole excitations are presented and compared to the theoretical predictions. Simulation using two-dimensional fine element code, Poisson, is also done. Results obtained from all the above three methods are found to be in good agreement with each other. The variations of skew quadrupole gradient for different sextupole excitations are also measured. The validity of this theory is also checked for various combinations of excitations including the case where magnet gets saturated. The excitation strengths required for producing the horizontal and vertical dipole fields are estimated analytically along with the presence of various multipoles. Theoretical predictions of permissible multipoles are compared to the results obtained from simulation.

The introduction of longer last stage blading in steam turbine power plant offers significant economic and environmental benefits. The modern trend, adopted by most leading steam turbine manufacturers, is to develop long last stage moving blades (LSMBs) that feature a tip shroud. This brings benefits of improved performance due to better leakage control and increased mechanical stiffness. However, the benefits associated with the introduction of a tip shroud are accompanied by an increased risk of blade flutter at high mass flows. The shroud is interlocked during vibration, causing the first axial bending mode to carry an increased, out of phase, torsional component. It is shown that this change in mode shape, compared to an unshrouded LSMB, can lead to destabilizing aerodynamic forces during vibration. At a sufficiently high mass flow, the destabilizing unsteady aerodynamic work will exceed the damping provided by the mechanical bladed-disk system, and blade flutter will occur. Addressing the potential for flutter during design and development is difficult. Simple tests prove inadequate as they fail to reveal the proximity of flutter unless the catastrophic condition is encountered. A comprehensive product validation program is presented, with the purpose of identifying the margin for safe operation with respect to blade flutter. Unsteady computational fluid dynamics predictions are utilized to identify the mechanisms responsible for the unstable aerodynamic condition and the particular modes of vibration that are most at risk. Using this information, a directed experimental technique is applied to measure the combined aerodynamic and mechanical damping under operating conditions. Results that demonstrate the identification of the aeroelastic stability margin for a new LSMB are presented. The stability margin predicted from the measurements demonstrates a significant margin of safety.

We have performed two- and three-dimensional, relativistic, electromagnetic, particle-in-cell simulations of an axially extracted virtual cathode oscillator (vircator). The simulations include, for the first time, self-consistent dynamics of the anode foil under the influence of the intense electron beam. This yields the variation of microwave output power as a function of time, including the role of anode ablation and anode–cathode gap closure. These simulations have been done using locally developed particle-in-cell (PIC) codes. The codes have been validated using two vircator designs available from the literature. The simulations reported in the present paper take account of foil ablation due to the intense electron flux, the resulting plasma expansion and shorting of the anode–cathode gap. The variation in anode transparency due to plasma formation is automatically taken into account. We find that damage is generally higher near the axis. Also, at all radial positions, there is little damage in the early stages, followed by a period of rapid erosion, followed in turn by low damage rates. A physical explanation has been given for these trends. As a result of gap closure due to plasma formation from the foil, the output microwave power initially increases, reaches a near-flat-top and then decreases steadily, reaching a minimum around 230 ns. This is consistent with a typical plasma expansion velocity of ∼2 cm/μs reported in the literature. We also find a significant variation in the dominant output frequency, from 6.3 to 7.6 GHz. This variation is small as long as the plasma density is small, up to ∼40 ns. As the AK gap starts filling with plasma, there is a steady increase in this frequency.

No abstract