As we have discussed here for years, “flanged joints” are usually MORE difficult to manage than are welded connections. However, somehow the rumor got started that flanged connections can be done by “Billy Joe Jim Bob.” Nothing could be further from the truth! And guess which connections most often leak? Yep, flanged connections. There are many reasons for this higher leak rate, but in most of my experiences, the flange leaked because it was NOT assembled correctly. Bottom line, our flange connections MUST have the same level of QC/QA as our welded connections.
In some of my more hazardous processes, some of which we classified as Category M Service (i.e., “Lethal Service”), our flanged connections had to meet the requirements of ASME PCC-1, specifically Section 14.2, Joint Assembly Records. In this article, I want to share this practice and explain the VALUE it brings to the assembling of our PRIMARY CONTAINMENT SYSTEM that utilizes FLANGED CONNECTIONS.
Consideration should be given to the preparation of a joint assembly record for each assembled joint, particularly those that are deemed to be in critical service or deemed to be prone to leakage.
Joint assembly records may include the following information:
(a) joint location or identification
(b) joint class and size
(c) specifications and conditions of flanges, fasteners, washers (including nut- or washer-bearing surfaces), and gaskets
(d) date of activity (assembly, disassembly, pressure test, etc.)
(e) Names of assemblers/workers
(f) name of inspector or responsible person
(g) disassembly method
(h) adverse disassembly conditions, such as the presence of nut seizing or thread galling
(i) leak history
(j) bolts, nuts, and washers used
(k) flatness measurements, when made (see Appendix D)
(l) assembly procedure and tightening method used, including applicable Target Prestress values as per the indicated tightening method
(m) Unanticipated problems and their solutions
(n) tool data such as type, model, pressure setting, and calibration identification
(o) tool access or safety issues
(p) Recommendations for future assembly procedures and joint maintenance and repairs
Examples of three (3) different types of joint assembly records are shown in Figures 5 through 7. An example of a joint leakage record is shown in Form P-3.4.
Often, the first step in maintaining joint assembly records is to uniquely identify each joint. The unique identifier should ideally be permanently fixed to the joint using a marking or tag system that enables an assembler to verify that they have the correct joint prior to working on it. Such a system enables storage of joint assembly records against the unique joint identifier and minimizes the potential hazard associated with disassembly of the wrong joint.
Joint assembly records have the following two (2) purposes:
– They facilitate joint assembly quality control by recording the individual responsible for a particular portion of the joint assembly process.
– They provide a record of the joint assembly parameters that were used and the lessons learned.
The first purpose is achieved by increasing the responsibility of the individual involved in assembling the joint by maintaining a record of their actions and identity. The second purpose is achieved by documenting the actual parameters used and observations made during joint assembly. This information will be useful if the joint leaks during operation, or for guidance during future assembly of the joint. It is essential that any deviations, comments, observations, or changes to the assembly procedure are fed back into the joint assembly record system, such that the assembly procedure may be updated prior to the next joint maintenance activity.
Users should decide what level of detail to record based on their own needs and resources. Due to the large numbers of pressure-boundary bolted joints assembled during construction and maintenance activities, it may not be practical to maintain a record of all joint assembly parameters (i.e., the example record shown in Figure 7).
Therefore, this sort of record is often completed and archived only until the next assembly for joints that are the most critical or most likely to leak. The shortened record (i.e., the example records shown in Figures 5 and 6) typically does not include all joint assembly parameters or observations and is usually discarded after unit start-up. This means that the second purpose is not achieved, but the more important effect of improving joint assembly quality control is still attained.
There is typically also a register of joints that is maintained, in addition to the joint assembly records, which tracks the progress of each joint, whether there is an assembly record or not. The register may incorporate the use of joint assembly records by requiring visual confirmation of a completed record before updating the joint register to indicate that the joint may be returned to service.
An additional note of guidance is that the tendency to provide or request too much information on the joint assembly record should be avoided. It should not be considered a substitute for written assembly procedures or assembler qualification.
Source: ASME PCC-1