Clarification of Stress Classifications (Primary, Secondary, Peak etc.) 3

[stevo200]

Hi,

Im newly graduated and looking at a pressure vessel job and am trying to identify what stress to classify intensities I am finding. I am running off the AS1210 standard appendix H - stress intensity classification.

Im struggling to understand exactly what they mean by primary stress intensity, secondary stress intensity, peak stress intensity and local primary stress intensity.

Please correct my train of thought...

Primary stress intensity: Intensities that would be found along the wall of a pressure vessel due to the pressure. For example at the centre of a pressure vessel wall where the deflection would be the largest, would the stress also be the largest as it is the furthest point from any support structure/members and therefor at that centre point would be the highest stress that could be seen as an intensity, and be classified as primary due to the fact it results from the pressure?

Secondary stress intensity: Im not too sure about this one due o this comment: "Examples of secondary stress are bending stress at a gross structural discontinuity,and differential thermal expansion." Does this meaning it is a stress caused not by the pressure, but in a way caused by some sort of axial load on a member which induces bending in another member at the discontinuity between the two members (im thinking of 2 plates welded at a 90deg angle)

Peak stress: A combination of the primary stress due to pressure and the Secondary stress as i mentioned above combined (for example at the hatch of a pressure vessel where the hatch meets the vessel at 90 degree planes there will be the pressure on the walls combined with some form of bending at the 90deg discontinuity?)

Local primary stress intensity: I am a little unclear all together.

If someone could please clearly explain these stresses in simple terms (I dont understand what is written in the standard AS1210) and possibly explain some examples or cases I would find these stresses it would be very helpful and greatly appreciated!

Thank you.

Steve.

 

[TGS4]

stevo200 - Welcome to eng-tips. Normally, when engineers post questions regarding stress classification, I would provide them with this quotation from ASME Section VIII, Division 2:

5.2.1.2. For components with a complex geometry and/or complex loading, the categorization of stresses requires significant knowledge and judgment. This is especially true for three-dimensional stress fields. Application of the limit load or elastic-plastic analysis methods in 5.2.3 and 5.2.4, respectively, is recommended for cases where the categorization process may produce ambiguous results.

In your situation, though, I would suggest that you go through the archives of eng-tips (it's pretty vast - spanning almost 2 decades) and see what you can find. I would also suggest that you consider taking a course on the topic - search for "ASME Section VIII Division 2 Part 5 training course". Also, you will want to read ASME PTB-1, and probably WRC 429 as well. Finally, I would suggest that you find yourself a mentor. Since you are using AS1210, I am going to assume that you are in Australia - depending on the part of Australia that you are in, I could recommend a local consultant who is well-versed in this topic.

That said, right now you are unqualified and inexperienced to perform this work. That makes you dangerous. Pretty much everything that you wrote in your original post is incorrect. You need assistance before you think about taking anything like this on. Please find qualified and experience assistance before you continue this work.

 

[corus]

There are notes that used to accompany the British standard BS5500 (now renamed EN something) that gives diagrams, explanations and examples as to what the stress classification means. In general primary stresses are load dependent whereas secondary are strain dependent. Peak stresses are sue to a stress concentration, such as a hole or a fillet radius.

 

[NRP99]

Stevo200

I agree with TGS4. Following is an example which might be helpful for your understanding.

Imagine a long hollow cylinder with only internal pressure P closed at both ends by flat plates. Then the average stresses induced in mid section from ends will be free from bending and there will be only membrane stress. This membrane stress in the mid section can be classified as Primary membrane stress(P_m). Now towards the end section where the cylinder is closed by the flat plates, the expansion of cylindrical shell at the connection is restrained by the flat plates. So junction will impose loads in the shell section as well as flat plates. Hence there will be bending stresses near the junction in both end plates as well as shell section. Collectively the stresses at the connection(away from corner) of shell to end plate is called as primary + secondary stresses. Understand that Secondary stresses are introduced due to restraint and if we remove the restraint (in this case, remove end plates), there will be no secondary stresses. Hence secondary stresses are caused by restraint type of loading such as temperature loading and constraints.

Now imagine you drilled a hole in the same cylinder at mid section. There will be stress concentration due to hole since cylinder has to resist the same internal pressure with reduced cross section now. The average stresses near opening can be classified as the local primary membrane stress(P_L) due to their local nature and local effect of hole(But make sure that these are really local and die out rapidly away from the hole). The total average membrane + Bending stresses are classified as primary+secondary stresses(away from hole edge). Here bending stresses are now introduced due to hole as restraint. At the edge of the hole there will be another component of stress apart from the primary and secondary i.e. peak component. This component comprises the most part of the total stress at that location and is mainly due to stress concentration effect. So at hole you have Primary+Secondary+Peak components of stress.

Note that these three stress categories are different components of total stress in section. There classification (you can say availability) will depend on the location in the cylinder. Primary stress-at mid section, Primary+Secondary stress-at shell to flat plate junction and Primary+Secondary+Peak stress-at hole.

Hope that this helps you in understanding the stress categorization which is bit tedious but very much logical.

 

[Mech2325]

NRP99: You have explained the topic in a brilliant way. Seems you have good understanding of it. I was once doing a stress analysis of a tower in which a Window was to be cut in its top section. The client wanted to have stress analysis if it was feasible as cutting a Window at top section may cause benfing due to high wind loads in the region. While i was doing the analysis ( i was just a fresh hire and was just acting upon instructions by my supervisor) i noticed that yhe software showed a "peak stress" at the edge of the opening but my supervisor said it was just a peak stress and dint take it seriously. Why is it that peak stresses are not to be taken seriously??

 

[NRP99]

Mech2325
Thank you.
First thing, Peak stresses should be taken seriously every time you encounter them. Engineering judgement and logic is required to decide whether they can be ignored or should be taken into account for further consideration. Also failure mode of component under examination helps you to decide whether to ignore them or not.
Wind pressure is type of static loading.(Here I am assuming negligible fluctuations in wind velocity. Also generally wind loads are occasional loads) Peak stresses at the edge of opening are due to local effect of opening. When full loading applied, over period of time material will undergo local yielding near edge due to high stresses and thereby deforming geometry by providing more area thus reducing stresses in the edge. This will be permanent change for the tower. And tower will resist the same force for entire its life.Since through thickness yielding i.e. plastic collapse is not happened and stress will be redistributed to lower value actually than the values shown by software, we can safely ignore these peak stresses.
Now imagine wind velocities are fluctuating with considerable range. The wind pressure is no longer static but dynamic one. Wind pressure will vary with time. There will be again local yielding of the material at edge. Fluctuations in wind pressure will cause the fibres at these locations undergo repeated straining and crack will form if the stresses are unbearable. This crack will then propagate through thickness and column will fail by fracture. Here we need to take peak stresses seriously and we should find out the fatigue failure life of tower using the peak stress at these edges since failure mode is now fatigue.

 

[Mech2325]

NRP99: Thanks once again for a thorough explanation

 

[NRP99]

I request forum experts to intervene and share their view on stress classification as well as their understanding of the concept. It will be helpful to members as well me also.

 

[TGS4]

NRP99 - have you read the references that I listed to the OP in my original post? Lots of good information there. This is a complicated subject and one that many now-senior engineers spent their careers learning and developing. First, learn from their writings, and then look to your peers (eng-tips) for answers.

 

[NRP99]

TGS4
I fully agree with you that stress classification is very complicated and new engineers should be extremely cautious while using it. Otherwise best way they should leave this to their more experienced seniors or get trained for the subject by expert mentors. Although I have not read the references that you mentioned, I will definitely go through them for better clarity of the subject. Thank you for the suggestion.

But sometimes experience teach you something which can not be explained or understood by reading books. At least my experience taught me this. My intention to request experts was to have their view and experience on subject matter. What is your and other expert's experience?

 

[TGS4]

If you want my unvarnished opinion it's this: DO NOT use elastic stress analysis for demonstrating Protection Against Plastic Collapse and the subsequent considerations of stress linearization and classification.

That's 20 years of experience speaking.

 

[NRP99]

Thank you TGS4 for sharing your experience. You are expert in this field considering the vast experience you have. I will definitely follow your guideline.

What other experts have to say on this? Please share your experience.