Packed stuffing-boxes are mechanical sealing systems that are extensively used in pressurized equipment such as valves and pumps. Yet there is no standard design procedure in use to verify their mechanical integrity and leak tightness. It is only recently that standard test procedures to qualify packing materials have been suggested for adoption in both North America and Europe. Nonetheless the assessment of the structural integrity of the housing requires a well-documented design procedure to insure safe use of packed stuffing-boxes. While the packing contact stress with the side walls is predictable using existing models there is no analytical methodology to verify the stresses and strains in the stuffing-box housing. This paper presents an analytical model that analyzes the stresses and strains of the stuffing-box components including the packing rings. The developed model is validated both numerically using FEM (finite element method) and experimentally on an instrumented packed stuffing-box rig that is specially designed to measure the structural integrity and leakage tightness of different packing materials.

No abstract

The analytical prediction of the contact stress in tube-to-tubesheet joints subjected to hydraulic expansion is conducted without any consideration to reverse yielding that can occur inside the tube. Most existing models consider the tube and tubesheet to unload elastically when the expansion pressure is released. These models are therefore less conservative as they overestimate the contact pressure. An analytical model that considers strain-hardening material behavior of the tube and tubesheet and accounts for reverse yielding has been developed. The model is based on Henckey deformation theory and the Von Mises yield criteria. The paper shows that reverse yielding that is present in tubes during hydraulic expansion unloading makes the joint less rigid and causes a decrease in the contact pressure depending on the gap clearance and the materials used. A good correlation between the analytical and finite elements results is obtained on different treated cases which gives confidence on the developed model.

Packed stuffing-boxes are mechanical sealing systems that are extensively used in pressurized valves and pumps. Yet there is no standard design procedure that could be used to verify their mechanical integrity and leak tightness. It is only recently that standard test procedures to qualify the packing material have been suggested for adoption in both North America and Europe. While the packing contact stress with the side walls is predictable using existing models there is no analytical methodology to verify the stresses and strains in the stuffing-box housing. This paper presents an analytical model that analyzes the stresses and strains of all the stuffing box components including the packing rings. The developed model will be validated both numerically using FEM and experimentally on an instrumented packed stuffing box rig that is specially designed to test the mechanical and leakage performance of different packing materials.

Stuffing-box packed valves which confine high-pressure fluids are often subjected to leakage failure. The lack of a design procedure and the vulnerability of packing ring sealing materials to withstand different operating conditions are the root cause of the problem. The sealing performance of valves with packed stuffing-box depends on the ability of the assembly to maintain a minimum threshold contact pressure between the packing and the stem and the packing and the housing throughout service operation. The distribution of the contact stresses in the packing materials is a key parameter to efficient sealing performance. This study presents a contact stress modeling study of two different design configurations that are helpful to produce a uniform distribution of the contact stress. The first model is based on the introduction of a variable gap between the packing and the side walls. The second model is based on a multistage loading of the packing rings. The two developed analytical models are validated by comparison with the numerical simulation using FE method and the results show a good agreement. The two design configurations can be used to improve valve sealing performance.