There are a few select methods of determining the requirements for an expansion loop – both for design and layout. These range from the simplistic to the exhaustive; it depends on the intent, thus alluding to my query.
For a rough check: a simple built-in beam or guided cantilever method would suffice. This entails reducing the piping configuration and your expansion loop to a series of free body diagrams about the guides and anchors.
For a less rough check: using “Spielvogel” piping stress charts / shapes would yield a satisfactory indication of where you are heading in terms of your strategy to replace the EJ.
For a pretty accurate check: use formal stress analysis software to determine code requirements, loadings; thus being exhaustive, responsible and safe.
You may glean a bit more from these references:http://www.eng-tips.com/viewthread.cfm? ... 902&page=7http://thermacor.com/downloads/document ... -4_001.pdfhttp://books.google.co.za/books?id=mvaJ ... &q&f=false
The reason for asking whether an expansion loop is necessary alludes to the oft, over-reliance on flexibility to provide unnecessary solutions. When considering thermal expansion and secondary stresses there are three routes a) if permissible, let the equipment simply accept the load, b) provide additional flexibility to reduce the load, c) provide additional stiffness to transfer the load to supports, civils or structures. The last would invariably include correctly placed anchors and stops, and heavier support structures.
Not being aware of the equipment, it could be assumed that the expansion joint was originally installed to alleviate all loading on the nozzle. If this were the case even additional flexibility i.e. a loop or offset would only reduce the level of loading
I would suggest that you may consider four, rather general, guiding principles when analysing brownfield piping modifications:
1. Don’t tear down a farm fence until you know why it was put up and who put it up in the first place.
2. If it's not for NASA and you’re designing necessarily to 3 decimal places, in practice it aint going to work.
3. Any new component in a piping run, however trivial, equals a new system entirely i.e. the piping's personality changes.
4. Keep the centroid (center of area) of your piping arrangement as far away (offset) from the line drawn between your nozzles
Hopes this helps