allowable combined stress value? 1

[picasa]

Consider a combination of following loads on a horizontal drum being designed per ASME Section VIII Div 1.

Internal pressure + External mechanical loading (moments and forces in 3 directions: longitudinal, circumferential, torsion)

What is the maximum allowable combined stress value? 1.5*Sm or 3*Sm?

 

[jte]

picasa-

Sorry, but the answer is both. Seems that (you're using Sm) you're pulling the limits from Div. 2 App. 4, Figure 4-130.1. That used to be the usual way to go with a Div. 1 vessel (but using the Div. 1 S instead of Sm) since Div. 1 didn't address these situations. In the last few years, Div. 1 has more or less incorporated the Div. 2 guidance officially: See UG-23(c).

What you need to do is distinguish the primary from secondary stresses. Once you've done that, follow the rules as spelled out in UG-23(c).

Just don't tell me that this question involves a particular 4" LWN nozzle on a 1/2" shell!

jt

 

[picasa]

NO. I have enough going on with that 4" LWN dude.

This one is a different nozzle 10" 150# on a 7/8" thk SA516 Gr. 70N plate.

Let us for a second think like an engineer. Keep the code aside. We are talking about combined stress generated by external mechanical loading + internal pressure. This stress is present when the vessel is in operation. How can you use an allowable of 3*Sm on that? You are crossing yield by a long way!

 

[CRG]

Is the mechanical loading self limiting? In other words, will the external load diminish with a small amount of localized yielding?

 

[SnTMan]

The 3*Sm is for combined primary + secondary stresses, and also combines membrane and bending stresses. It is implicit that a portion of the loading is self-limiting. See Div. 2, AD-140 and 4-120 & 4-130. I caution that my understanding of Div. 2 is less than comprehensive.

My usual practice,in the absecence of REAL info on the source of nozzle loads, is to limit allowable stresses to a fraction of yield at design temp, 80% to 95%, for a WRC-107 analysis. I rarely look at stresses at the edge of a reinforcement pad.

If I know that a large component is thermal (self-limiting) I may use the Div. 2 multiple against Div. 1 allowables, 3*So. This case is more rare.

 

[jte]

picasa-

If we set the Code aside for a while to think like engineers... Then be sure to put your nonlinear thinking hat on! By definition, material yield is a nonlinear function. Yet, your calculations are... Linear! Therein lies the quandry! Another perspective is that for decades the 1.5*S and 3*S limits have been used without significant failures. For what its worth, the 3*Sm limit for Div. 2 brings you not only past yield, but up to the specified minimum ultimate strength! We should be popping shells all over!

To take this issue a few steps further, pull your copy of Div. 2 and take a look at Appendix 5, Figure 5-110.1. By using this figure, I could justify taking a component fabricated of SA516-70 and subjecting it to an alternating stress of 400 ksi. This would be roughly 10 times yield. But Code legal and defendable, and provided you did your analysis correctly, entirely safe and reliable. Just don't cycle it more than 10 times.

In essence, the Code has provided you tools/methodology which allow you to cheat by using linearly calculated stresses in something which is inherently nonlinear. There are some papers out there which get into the discussion a bit, one by George, Seipp, and Morrison regarding the reasonableness of secondary stress classification published in 2004 comes to mind.

http://store.asme.org/product.asp?c...Papers&category_name=&product_id=PVP2004-2613

You are asking good questions! Get into it and you'll find some answers and more questions.

jt

 

[solinar]

A hydrotest is a leak test primarily for welds. As long as your clip is not being installed over an existing through wall weld (such as a circ or long seam) then a hydrotest would not really do anything for you. I would guess you could probably get the AI to buy off on not rehydrotesting.

 

[solinar]

Whoops, replied to the wrong thread. Sorry

 

[TGS4]

I completely agree with Jaan (I mean jte...) - that is in fact an excellent paper, which discusses some o fthe issues that picasa is dealing with.

I would also recommend getting your hands on the "Criteria of the ASME Boiler and Pressure Vessel Code For Design By Analysis in Secions III and VIII, Division 2" document. It is likely out of print, but there always seems to be copies floating around.

Cheers,
TGS4