By Steve Gillott
European Sales Director, Intergraph CADWorx & Analysis Solutions
Commonly, piping networks are modelled using beam type models with the stresses and flexibilities at discontinuities (for example, branch or tee connections) being calculated using simple empirical relationships, such as those found today in ASME B31.3 Appendix D. The simple formulae found in this and other international design codes* are based upon work carried out in the mid 1900’s by a pioneer in the field A.R.C. Markl, a gentleman most respected by engineers worldwide. The data in Appendix D and other design codes used by pipe stress engineers provides the relevant Stress Intensification Factors (SIFs) and flexibility factor (k’s) for the discontinuity at hand.
Our knowledge, expertise and technology has developed almost exponentially over the last few decades and we now know there are many limitations in the accuracy of these formulas which, depending on the situation, can lead to either grossly over or under conservative results. Within Appendix D itself and in many other locations in the design codes, these inadequacies are acknowledged and the engineer is instructed by the code to use ‘more applicable data’ if available; see extract below.
In 2007, ASME awarded a contract to Paulin Research Group (PRG), known as the 07-02 Project, to develop this ‘more applicable data’ for use in design code calculations.
More Applicable Data
Based on results of the latest analysis, research, and testing conducted for the ASME committee in the 07-02 Project, this “more applicable data” is now widely available within the new** ASME B31J Document. This improved data is especially relevant for piping systems…….
• With large D/T ratio
• Containing sensitive or rotating equipment
• In cyclic operation
• In ‘tight’ configurations and with short or stiff thick-walled piping
• Where the solution is dependent on the SIF for a small bore connection
• Where torsion is significant - even though ASME B31.3 gives a torsional SIF term in the stress equations, there is no guidance on the value to use and the engineer is advised to use a value of 1.0 ‘in the absence of more directly applicable data’.
…………and many other instances.
Application of these newer values can result in significantly lower system forces/moments and stress levels which, in turn, can result in more economical designs. For some more critical situations it can even result in more realistic stress values which are higher; this allows for a reduction of risk for failure in such cases.
The publication of the new** B31J document is extremely significant as:
• It provides, for the first time, a tangible reference for the ‘more applicable data’ that the design code requires.
• Using this data will enable pipe stress engineers to have total confidence that a given design is carried out in accordance with the intent of the code using the latest, fully tested and approved more ‘accurate’ data, without unnecessary over or under conservatism.
• Using the better and ‘more applicable data’ helps to ensure safer systems, fewer or zero failures whilst facilitating a more efficient project workflow with fewer unnecessary system changes (i.e. to resolve problems which don’t exist in practice) – peace of mind and saving money.
• It helps to streamline project workflows whilst ensuring safe, compliant designs
Using the ‘More Applicable Data’
The relationships in B31J, whilst improving the accuracy of the analysis, can be quite complex in their application. Analysts can manually apply the data in B31J to their pipe stress analysis calculations with the associated risk of transposition errors coupled with a large increase in the time required to create a given analysis ‘model’.
However, during the 07-02 Project conducted by PRG for ASME, software called FEATools™ was developed which incorporates the ‘more applicable data’ as published in B31J. FEATools seamlessly couples with CAESAR II®, the standard pipe stress analysis solution - the model is built in CAESAR II, imported into FEATools where the SIFs and k’s are updated automatically in an improved CAESAR II model.
FEATools also provides the ability to conduct accurate Finite Element calculations on each component (e.g. tee, bend, nozzle connection….) for ‘state of the art’ accuracy. The Finite Element model is created automatically using the data obtained from CAESAR II combined with the technology developed by PRG in their proprietary software; optional additional data may be provided by the engineer as required (e.g. specific bend attachment data).
This is by far the easiest and most reliable way to incorporate the ‘more applicable data’ into a piping system stress analysis as required by ASME in the code and as seen in the B31J Document. Results review is extremely easy and intuitive. No longer does an engineer have to open multiple files, export to Excel, create side by side tables etc. in order to compare Model 1 with Model 2 to see what has changed and where. With FEATools multiple derivatives of a given model can be viewed on graphical comparison charts to see at a glance where displacements, forces, moments or stresses have changed at all locations in the piping system. It’s easy to see how the ‘more applicable data’ derived by FEATools has influenced and improved the CAESAR II model.
Further Refinement of Pipe Stress Models
FEATools Version 3.00 not only provides the ‘more applicable data’ in B31J but also optionally provides the pipe stress engineer with additional better data to further refine any CAESAR II analysis model with:
• Accurate nozzle flexibility and SIF data
• Accurate bend flexibility and SIF data (with or without attachments)
• Allowable loads on supports (when used in conjunction with NozzlePRO)
• Accurate API 661 nozzle data
Finally, if a given model still demonstrates non-compliance with the piping codes, any single component can be assessed separately using a ‘design by analysis’ Finite Element approach.
Safe Designs and Peace of Mind
In summary, you can use CAESAR II and FEATools to incorporate the intent of the design code and the ‘more applicable data’ in B31J. Following this workflow is quick and easy but also ensures greater accuracy without compromising designs whilst providing engineering companies with the opportunity to reduce project time scales and costs.
Owner Operators are under pressure to carry out Projects with reduced costs, maintain safety and deal with smaller margins. The B31J ‘more applicable data’ and its implementation within CAESAR II via FEATools provides the perfect combination to achieve these objectives.