SIF FOR MITERED ELBOW WITH REINFOSMENT 2

[gguerrageci]

Im involved on a flexibility analysis of a hot gase duct that operate a external pressure .

the mitered elbows have circunferencial reinforsment.

how can I calculate the SIF incluiding that ribs.

 

[JohnBreen]

Hello gguerrageci,

Well there are only two ways to do it - by testing such that you can develop an S-N curve or by modeling the component with FEA software such as FE/Pipe (Paulin Research Group). Then you would use these calculated SIF's as input to the beam theory software you are modeling with. There are Welding Research Council Bulletins that tell you how to proceed in either case. But the calculated stresses might be approximate at best.

But before you go to the trouble of doing that please tell us the diameter and wall thickness you are addressing. Remember that SIF's are associated with beam theory and if you have a duct section with a D/t ratio of 100, beam theory will not predict the largest stresses in the ducting. If this is the case, you might be able to use a beam theory program to determine the overall distribution of the forces and moments over the system, but the calculated stresses will not be accurate (or valid for a structural evaluation). At locations of concentrated stress (such as mitered elbows, supports, branch connections, etc.) you would have to model the component using FEA method (for a distance of at least 5 diameters away from the stress riser) and used the forces and moments calculated in the beam theory model as input to the FEA model. In addition, depending upon the method of fabrication, you might have to make an FEA sub-model to capture the nuances of weld beads and other geometrical discontinuities for the purpose of calculated the true elastic stresses (of course you would use the methods and allowable stresses from the ASME B&PV Code, Section VIII, Division 2). If the gross FEA model predicted yielding, a non-linear, elastic-plastic analysis (geometry updates as a function of distortion over a time period) would be necessary.

How many duct designers do this? Not very many. If your duct has a D/t ratio greater than 100 or a OD greater than 72 inches, try to find a copy of AWWA Design Manual M-11 for the design of large diameter penstocks and read the section dealing with reinforcement of miters and branch connections. These empirical designs seem to work for relatively low pressure systems that do not have great ranges of temperature excursion.

Regards, John.

 

[JohnBreen]

Hello again,

Having re-read your question (I overlooked the external pressure aspect) I would direct you to the ASME B&PV Code Section VIII, Division 1, Paragraphs UG-28 and UG-29 for the design of stiffening rings. This does not directly address the mitered bends and it does not address B31.1 type SIF's but again, SIF's are associated with tha ASME Code for Pressure Piping, B31, and SIF's are associated with beam theory solutions and this might not be the approach you will need to take.

Regards, John.

 

[gguerrageci]

Thanks John

the duct is 4000 mm diameter and have 8mm thikness D/t = 500
the external pressure is 100 mBar and the temperature is 760°F

what is beam teory program , I have pipepak and I think use this teory ??

 

[gguerrageci]

Aditionaly the duct have reinforentd and the elbows and a TEE also

 

[DSB123]

gguerrageci,
Just a point, you say "What is beam theory program, I have pipepak and think use this theory". If you do not know what beam theory is and also are not sure what basis the software you are using is based on you need to ask yourself " Should I be really be tackling this problem and do I need to obtain a little more training on the software to understand the basis on what it assumes and how it works?"
I am not trying to be negative but there are too many people in engineering at present who tend to use software "due to the user friendliness of the packages" without knowing the basics of the subject or the basis of the software or having the adequate training.