Abstract
With the increasing demand of the oil & gas industry, many pump companies are developing multiphase pumps, which can handle liquid-gas flow directly without separating the liquid from a mixed flow. The see-through labyrinth seal is one of the popular types of non-contact annular seals that act as a balancing piston seal to reduce the axial thrust of a high-performance centrifugal pump. The see-through labyrinth seal also generates reaction forces that can significantly impact the rotordynamic performance of the pump. Multiphase pumps are expected to operate from pure-liquid to pure-gas conditions. Zhang et al. (2019) conducted a comprehensive experimental study on the performance (leakage and rotordynamic coefficients) of a see-through labyrinth seal under mainly-gas conditions. This paper continues Zhang et al.’s (2019) research and studies the performance of the see-through TOS (tooth-on-stator) labyrinth seal under mainly-liquid conditions. The test seal’s inner diameter, length, and radial clearance are 89.256 mm, 66.68 mm, and 0.178 mm, respectively. The test fluid is a mixture of air and silicone oil (PSF-5cSt), and the inlet GVF (gas volume fraction) varies from zero to 12%. Tests are conducted at an exit pressure of 6.9 bars, an inlet temperature of 39.1 °C, three pressure drops PDs (27.6 bars, 34.5 bars, and 48.3 bars), and three rotating speeds ω (5 krpm, 10 krpm, and 15 krpm). The seal is always concentric with the rotor, and there is no intentional fluid pre-rotation at the seal inlet.
The air presence in the oil flow significantly impacts the leakage as well as the dynamic forces of the test seal. The first air increment (increasing inlet GVF from 0% to 3%) slightly increases the leakage mass flow rate, while further air increments steadily decrease the leakage mass flow rate. For all test conditions, the leakage mass flow rate does not change as ω increases from 5 krpm to 10 krpm but decreases as ω is further increased to 15 krpm. The reduction in the leakage mass flow rate indicates that there is an increase in the friction factor, and there could be a highly possible flow regime change as ω increases from 10 krpm to 15 krpm. For ω ≤ 10 krpm, effective stiffness Keff increases as inlet GVF increases. Keff represents the test seal’s total centering force on the pump rotor. The increase of Keff increases the seal’s centering force and would increase the pump rotor’s critical speeds. Ceff indicates the test seal’s total damping force on the pump rotor. For ω ≤ 10 krpm, Ceff first decreases as inlet GVF increases from zero to 3%, and then remains unchanged as inlet GVF is further increased to 12%. For ω = 15 krpm, Keff first increases as inlet GVF increases from zero to 3% and then decreases as inlet GVF is further increased. As inlet GVF increases, Ceff steadily decreases for ω = 15 krpm.
Swirl–nozzle interaction experiment: quasi-steady model-based analysis
