In Europe, in 2001 the new standard EN 1591 for strength and tightness proofs of bolted flange connections (BFC) of floating type flanges was released. In addition, the German nuclear code was revised regarding the calculation of BFC. With this standard not only the floating type but also the metal-to-metal contact type of flanges (MMC) can be treated. Additionally, the ASME code is the basis for the flange calculation in the European standard EN 13445, which is the standard for unfired pressure vessels. In compliance with the goal of the calculation, the different calculation codes can be used. There must be a differentiation between the design of the components, the determination of the prestress values for assembly, the stress analysis and the tightness proof of the BFC. First, all parameters which influence the function of the bolted flange connection are considered. In a second step, the range of use of the different standards and the calculation algorithm are discussed.

To study the effect of temperature on the tightening characteristics of gaskets, leak rate tests were performed using a servo-hydraulic test equipment (short term temperature exposure). According to the tests the tightening behaviour of gaskets made of fibre based sheet material improved with temperature due to additional deformation of the gaskets during the test. In tests with camprofile gaskets with graphite layers there was no significant additional deformation of the gasket thus the leak rates showed no difference between room temperature and elevated temperature. Regarding long term exposure to temperature it is expected that with a lot of gasket materials the leak rates increase after a certain aging period. It is too expensive to study these effects with complex servo-hydraulic test rigs. Therefore a more simple test rig was developed that can be loaded in hydraulic test rigs (to determine leak rate vs. gasket stress curves) and that can be exposed to temperature in an oven or by a special heating device.

No abstract

Testing techniques were developed to investigate the long-term behaviour of stuffing box packings under real service conditions. This equipment meets all requirements for the simulation of high temperature applications with gaseous fluids, hot water and steam: • controlled pre-tightening (gland load); • variation of the stiffness of the assembly (e.g. Bellville springs); • determination and monitoring of packing stress - respective gland load; • control of stroke and stem force; • heating-up, pressurization and temperature control; • conditioning of water and steam chemistry. All relevant influencing parameters on the long-term behaviour of varied stuffing box packings based on flexible Graphite, woven PTFE fibres and special packing designs were taken into consideration: • assembly procedure (packing stress); • stroke and number of cycles; • effect of Bellville springs; • medium and temperature. In detail, friction, relaxation and leakage tests were carried out. On the basis of these results an evaluation of the long-term performance of the different packing types based on Graphite and PTFE is possible.

To consider special demands, two gaskets floating between the flange faces are used for example in flanged connections of pressure vessels, valves, pumps etc. Tightness control can be one of the reasons to choose this design. The inner (primary) gasket has to tighten the internal pressure; the outer (secondary) gasket tightens possible leakage of the primary gasket. By the use of two identical gaskets in the same plane the procedure for the tightness proof is the same as in connections with one gasket. In many cases there are used two gaskets with different characteristics (deformation and tightness characteristics) in two tightening planes. The behaviour of this connection depends not only on the gasket characteristics but also on the tolerances of axial dimensions. The distribution of prestress on both gaskets is determined by the elastic-plastic deformations of both gaskets and tolerances of axial dimensions of the connection. The decrease of gasket stress in operation depends on the plastic deformations of both gaskets in operation, too. Depending on the demanded tightness class, the possible gasket stress during assembly and the gasket stress in operation determine the selection of suitable gaskets (with necessary gasket characteristics). It is necessary to take into account the internal pressure and possible pressure between primary and secondary gasket, too. In the paper the design of gaskets for this type of connection is discussed. Originally, it was used: - nickel ring as primary gasket, - asbestos packing as secondary gasket. Both gaskets have suitable deformation characteristics for the existing type of connection. The secondary gasket has low stiffness for the adjustment of an optimal stress on the primary gasket after prestressing. The primary gasket has a good relaxation behaviour. However, the nickel ring needs a very high local stress to achieve suitable tightness behaviour. This stress leads to plastic deformation of the flange surface during assembly, too, and this can induce leakage after the next assembly (with new rings) as operation experience showed. The use of asbestos packing is not allowed any more due to health reasons. This invoked a redesign of this gasket. The paper deals with gasket proofs and gasket characteristics for selection of suitable gaskets. Calculations of gasket stress distributions of both gaskets (elastic-plastic deformations) are presented. The analysis of the stress distribution allows reviewing the tightness class in operation incl. relaxation behaviour of the connection. Other recommendations are listed like the increase of connection safety by a reduction of bolt stiffness (better relaxation behaviour of the connection and higher bolt elongation), more accurately bolt elongation measurement, new assembly procedure, etc. Main target of the re-design is life time extension of connection, too.