Xiaoying Li

Liquid film phase change is a critical factor in the performance degradation of mechanical seals. However, experimental data regarding the evolution of phase change and groove-type differences for typical hydrodynamic grooves such as inner herringbone and inner spiral grooves remain scarce. In this study, phase change experiments were conducted on mechanical seals with these two groove types using high-temperature hot water as the working medium. The effects of rotational speed and medium temperature on phase change evolution were systematically investigated, the temperature distribution on the back surface of the stationary ring was synchronously measured, and the influence of film thickness on sealing performance was analyzed using a numerical model. The numerical predictions show good agreement with the experimental data. The results demonstrate that, compared with the inner spiral groove seal, the inner herringbone groove seal exhibits a 6.44%–33.61% higher phase change rate, an 18.4%–35.4% lower leakage rate, and an average back-surface temperature up to 4.88 K lower. Phase change in the inner spiral groove seal concentrates at the low-pressure inner-diameter side and propagates outward along the spiral grooves, whereas in the inner herringbone groove seal it is additionally distributed on the leeward side of the grooves. This difference arises from the coupled effects of non-uniform liquid film thickness between the groove and dam regions and localized low-pressure zones. The phase change rate decreases with increasing rotational speed but increases significantly with medium temperature. Film thickness analysis reveals a stable operating interval and a phase-change-sensitive interval. The dimensionless friction torque gradually attenuates and plateaus as film thickness increases, while the leakage rate exhibits a continuous upward trend.

Liquid film phase change critically degrades mechanical seal performance. Experimental data on phase change evolution and groove-type differences for typical hydrodynamic grooves remain scarce. In this study, phase change experiments were conducted on mechanical seals with these two groove types using high-temperature hot water as the working medium. The effects of rotational speed and medium temperature on phase change were investigated, back-surface temperature was synchronously measured, and film thickness effects were analyzed numerically. The numerical predictions agree well with experimental data. The results demonstrate that, compared with the inner spiral groove seal, the inner herringbone groove seal exhibits a 6.44%–33.61% higher phase change rate, an 18.4%–35.4% lower leakage rate, and an average back-surface temperature up to 4.88 K lower. Phase change in the inner spiral groove seal concentrates at the low-pressure inner-diameter side and propagates outward along the spiral grooves, whereas in the inner herringbone groove seal it is additionally distributed on the leeward side of the grooves. This difference arises from the coupled effects of non-uniform liquid film thickness between the groove and dam regions and localized low-pressure zones. The phase change rate decreases with increasing rotational speed but increases significantly with medium temperature. Film thickness analysis reveals a stable operating interval and a phase-change-sensitive interval. The dimensionless friction torque gradually attenuates and plateaus as film thickness increases, while the leakage rate exhibits a continuous upward trend.

In cryogenic liquid-film mechanical seals, phase change readily occurs, altering the flow and heat transfer behaviors and thereby inducing thermal deformation. Such coupling between liquid-film phase change and end-face deformation significantly degrades sealing performance. To address this, a thermo-fluid-solid coupling model incorporating phase change is established for an inner spiral-groove liquid-film mechanical seal. After validation against experimental results, the model is employed to systematically examine how taper, rotational speed, temperature, and film thickness influence liquid-film phase change and sealing performance. The results reveal that a positive taper enhances the load-carrying capacity while reducing the friction torque and phase-change rate, but at the expense of increased leakage and a higher fluid film temperature rise. Furthermore, rotational speed significantly alters the temperature distribution and deformation characteristics. The end-face temperature shifts from a local drop to an overall rise, and the thermal deformation changes from negative to positive as speed increases. Additionally, the thermal state of the seal surface transitions from an overall temperature rise to a uniform temperature decrease with increasing temperature, whereas the sealing performance parameters are highly sensitive to film thickness variations within a narrow critical range.

To investigate the dynamic response of liquid film seals under strong vibration conditions, an experimental setup capable of generating excitations with various amplitudes and frequencies was designed. Excitation force was applied to the axially movable static ring, and the time varying force, acceleration and displacement were measured using an impedance head and eddy current sensors. Based on the single degree of freedom vibration equation, the axial dynamic stiffness and damping coefficients were identified, and the evolution of their dynamic characteristics with rotational speed, seal chamber pressure, external excitation frequency and amplitude was systematically studied. The main conclusions are as follows: Under low frequency excitation, the average and median values of the axial dynamic stiffness and damping coefficients decrease with increasing rotational speed, and the decreasing rate gradually slows down and stabilizes. Meanwhile, the effect of rotational speed on both coefficients weakens as the external excitation frequency increases. Both coefficients increase approximately linearly with rising seal chamber pressure. As the excitation frequency increases, the axial dynamic stiffness coefficient first decreases slightly and then increases sharply, reaching a minimum at 100 Hz. Under normal operation with low frequency excitation, the two coefficients increase approximately proportionally with excitation amplitude. However, when the excitation amplitude is large enough to cause contact between the rotating and stationary rings, the axial dynamic stiffness coefficient decreases or even becomes negative, whereas the damping coefficient shows no obvious discontinuity.