I have read some of the threads about this topic and also the related article by Mr. John Luf. I just want to have some confirmation or otherwise regarding my understanding from what I have read :

1. Check for lifted supports in the operating cases.
2. Using a +Y support at the lifted node, check for thermal stresses and sustained stresses.
3. Check "hot sustained" using either
a) proper load combination (e.g. OPE-T)
b) make a separate file and remove all lifted supports (check only the sustained stresses)
4. and so on...

In general, the correct thermal stress is with the +Y and not without it. Am I in the right path ?

Thanks,
Stress Rookie

A minor point...

In your Step 3b, when running sustained loads alone with +Y's that lifted in operation removed, remaining +Y's should not be allowed to lift. You might want to replace them with Y supports. I think 3a is a better way to go but it may require explanation to others.

Yes, your thermal stresses are fine with the +Y's when you do the recommended (operating - installed) load case.

thanks for sharing...

This is a follow-up question:

I followed both steps 3a and 3b in my system and the calculated hot sustained stresses are the same? Both failed with the same percentage. But it failed at a point that in my opinion should have not failed.
Imagine this system: A vertical 24" steam pipe at 399^C with base support, then a bend, then a horizontal pipe to the EAST, then another bend, then a horizontal pipe to the west with 3 vertical supports and 2 guides, On this horizontal pipe there are 3 PSV's connected by a weldolet. The 2 supports lifted up during operation and so I checked the hot sustained, I was surprized to see that the system failed at the weldolets. I placed nodes near the weldolets but those are not failing.
To add more...the PSVs discharged to the atmosphere and the vent pipe is guided all-around.
When I removed the guides of the pipe vent the system becomes OK.
I noticed that the weldolet is failing during the HOT SUS case because the guides of the pipe vent are restricting the downward-east movement of the PSV piping during W+P case (using step 3b).
This is a case where guides affect the hot sus stress output. Is it OK to remove guides just for hotsus check ?

I will comment on your model and results if you send it to techsupport@coade.com.

Tell 'em Dave sent you...

Dave,

This is off topic, but still relevant - I think it would be a useful upgrade for the community for us to be able to add attachments to forum posts. I'm not sure what software is used, but I've been on other forums of different topics where it is possible to add an attachment to a post so that others can grab it.

Naturally, there's always the possibility of abuse, but I would expect you could restrict the attachments to speicifc types (i.e. .A and .JPG files)

You're right. The vendor is working on a new version and hopefully this capability will be provided.

Thanks, i will send it to you right away.

Your(dave's) observations would make intresting reading (Case study anyone?)

You have yourself an interesting situation here. I don't think the B31.3 interpretation asking for sustained stress in all expected support configurations works here and neither does the (Operating-Thermal) trick.

The deadweight sag causes rotation on the vent stack but the guide up higher on the stack resists that bending and causes a moment and overstress on your o-let. For those of you trying to picture this: Vertical stack connects to long horizontal header supported by +Y restraints. Other end of the header drops down. Vertical thermal growth at this end lifts header off of +Y's at this end. The +Y's under the vent stack at the other end are still active. Code interpretation asks that you evaluate sustained loads without inactive supports so you run (W+P1) with those inactive +Y supports removed. So the pipe sags and the resulting header rotation, in conjunction with the guide above on the stack overloads (overstresses) the weld-o-let.

In my opinion, such a collapse load would not exist in the operating position. I would probably ignore the guide in this calculation.

Here might be an easy way out... You modeled your weld-o-let kind of like how some nuclear pipers would - with a rigid element from the intersection of the centerlines, up to the header wall and then specify an SIF on the branch piping at the header wall. If I recallm this is a 24x8 connection. But the nuclear boys also introduce a flexibility factor at the intersection. B31.3, on the other hand (and this is what you are using here), says the SIF is at the intersection. I would want to see what the "calculated" stress would be if you called out your weld-o-let at the intersection point and ignored the rigid element. The added flexible length may drop your sustained load bending moment enough to drop the stress under the allowable limit.

You might say I'm playing games here but it is an interesting game. The mantra is - adding flexibility reduces expansion stress and adding stiffness reduces sustained stress. Since this doesn't apply here I again state that you have an interesting situation here.

The stresses dropped dramatically at those points. Thanks again for your comments Mr. Diehl.
For my last question, what can you say about the layout? Is it OK if the PSV header lifted off (10mm at lifted support, but with 1 active support near the PSVs) as long as the OCC stresses due to PSV reaction forces are within code allowables ? Is that not a common practice ? Or do I really have to re-route to make the header sit (client may not like it)?

Thanks X 10^3

IMHO... lift off of piping systems at supports other than those supports which are clearly intended for maintenance during shut downs is poor practice.

Why? Well if things are satisfactory code stress and sag wise, without the supports being active why put them in? Also these lifted off supports may cause consternation in the field.

One other caveat as well if your line operates in the creep range the Caesar II calculated sag would be actually larger in the real world. The B31 code rules require evaluation for code stresses with the cold or ambient temperature modulus of elasticity this number changes significantly in the creep range.

Spring hangers were invented for a reason…. The same thing applies to hold down supports as well.

You say "The stresses dropped dramatically at those points". I'd rather you said the CALCULATED stresses dropped. It's up to the engineer/designer/technician/user to confirm that the model reflects the real-world system and the piping code intent. I'm not wagging my finger at you; I just want to remind everyone that there's more to it than simply saying this number is less than that number.

Regarding the layout, I will not say more than this - I would be concerned about the added dynamic response if you allow those valves to open while the header is not solidly supported. Perhaps, too, the thrust loads may run down your riser and do some disturbing things below. Remember that you have the piping code issues ( usually stress) and you have your mechanical or structural issues that are independent of Code recommendations.

Don't just play the numbers game...

Regarding John's point on lift-off. I know of some (older, wiser?) engineers who automatically say lift-off is poor design. Most, however, while they appreciate that software can analyze these situations, they cannot offer a quantitative limit on what is OK and what is too much.

I really appreciate your replies. I have learned a lot through this forum. Thanks.

I take serious issue with Dave's post "Regarding John's point...". I am in full agreement with JCL until somebody can prove to me the lift off representation is a reasonable representation of real life for cool down and subseqent cyles.
I'm not saying Dave is wrong. I just haven't heard a convincing answer. While I accept that pipe stress is more opinion than most would like, if you can't figure out whether something is correct or not correct or at least
bracket the "definitely right" - "definitely wrong" then something is seriously wrong. Is it possible perhaps that the software cannot analyse this situation?

Caesar can calculate beam deflections of over 90 deg if you let it and not report an error (or certainly used to). Ability to calculate doesn't always mean a reasonable answer.

My post of May 15, 2005 - How do you put a link in.

This posts listed topic is really only part of the issue. As the famed radio commentator (Mr. Paul Harvey of the U.S.A.) says now the rest of the story.... Foglamp (HID?) mentions "On this horizontal pipe there are 3 PSV's connected by a weldolet."

Well frankly the idea of PSV piping loosely supported I find disconcerting. Especially if the PSV's start to simmer (pop open slam shut pop open slam shut etc., etc. ......)

These lines should be held down at least in a few spots relatively securely. Please review the non-mandatory chapter on relief piping found in ASME B31.1

Also a factor to be considered is the ever popular and mysterious flow induced vibration. Some people insist on full countoured welded insert to help reduce the likelihood of such phenomena.

So sustained, occasional stresses, and support lift-off not withstanding you have other issues to consider at the very least look at B31.1!

Andrew - to put in a link:

1) Click below on [Full Reply Form]
2) Once on that page, use the [URL] button at the bottom and follow the prompts.

This refers to final remarks of L.C.Peng's recent writing in Feb'06 issue of Chemical Engineering on the topic of hot sustained run:
"The first priority of the analysis is to accurately determine the sustained weight stress at hot operating condition. This is not to say that expansion stress is unimportant. A good analysis shall calculate as accurately as possible both sustained and expansion stresses."

Now, my question is : although we have vessel nozzle load allowables generally apportioned in 1:2 for dead weight:thermal expansion load, how many of us meet this division religiously ?

Mostly, we meet the whole allowable under any possible operating condition - sustained load & hot operating load - isn't it ?

Even in WRC-107 or Nozzlepro/FePipe check, sustained & expansion load allowables are different as per ASME SEc-VIII Div-2 Elastic Analysis.

regards,

sam

The B31.3 committee has handed down more than one interpretation that states that the sustained stresses shall be evaluated in all conditions.

If you are evaluating a line per B31.3 you are required to evaluate your sustained stresses Hot Cold or in-between.

Is this done? My guess would be that 80% of the time it is done incorrectly and people when told they performed the work incorrectly automatically blame the computer software (such as Mr. Peng does). The last time I checked I found the computer did only what I told it to do! In this particular case for instance the designer asked about support lift-off on a system that should be strapped down in all 6 DOF at least at one point. The computer does not know these things humans are supposed to and direct the computer towards a correct solution….

8 out of 10 analysts probably miss this…..

Yeah, I'm still waiting for the computer (or
software) that does what I want it to do
rather than what I tell it to do, Richard
any progress in that area yet? (grin)

-Shannon

Yes of course. It's already finished. We're just waiting for the appropriate time to activate it.

In all seriousness, John is correct in his statements.

In July 2006, at the PVP Conference in Vancouver, Dave Diehl will present a paper (PVP2006-ICPVT11-9310) titled "Longitudinal Stress Due to Sustained Loads in a Nonlinear World". It's pretty much right on the money. Make a note to find this paper in August.

What hope do we have if even the good guys can't agree?
Maybe the answer is irrelevent, and we are asking the wrong questions.

For myself, i am not the best at understanding the deepest technical and accademic formulee and parts that form the etherial world of pipe stress analysis,
I am good (i think) at the more practical side of the stress analysis.

I just can't reason for myself, why something which is so important as pipe stress analysis is left to the vagaries of personal opinion.
to me, it is either:
a)Correct and therefore safe
b)Incorrect and therefore unsafe

(a+b)/2 is a grey zone and chance is not my first option.

Example.

On recent post, D.Diehl commented that to push or pull a pipe with a bellows in it is irrelevent.
But if a system is 99.9% stressed before you pull it ,then bingo overstress.

Whats my point here?
I should be able to by analysis, ensure that my system is safe and meets a set of clear requirements. which in turn, gives the client confidence in his design

By all means, if costs are high, and you have the skill then push into the greyzone
I shouldn't have to wait till August to read a paper which is in dispute by someone who i would judge to be hugely more capable than myself.


(Steps of soapbox)

re Sam's comment on Section VIII, Div 2.

From the standpoint of the pressure vessel nozzle, external loads due to both mechanical loads and those due to thermal expansion of the piping are considered the same. Local stresses in the shell due to thermal expansion of the vessel itself, would be treated differently.

This is commonly misunderstood. The vessel nozzle does not know the source of the loads, and local deformation of the vessel nozzle will not in most instances significantly relieve the thermal expansion induced loads from the piping. Thus, from the nozzle view point, the thermal expansion induced load of the piping is a primary load.

An excellent point by Dr. Becht one which also to some extent is true in piping systems as well. If the location of the thermal displacement stresses (Or strain) were such that the component cannot adequately yield before it breaks then thermal displacement loads would be acting as a sustained load. This occasionally happens to people who have unbalanced systems where all the thermal strain energy only has one unfortuanate very small spot to go.

Superpiper.... its an imperfect, fuzzy gray world all we can do is do our best.... so hang in there.... my guess is that Daves response to you was based on his viewpoint which was more distant then yours.

This refers to Dr. Becht's comment

"From the standpoint of the pressure vessel nozzle, external loads due to both mechanical loads and those due to thermal expansion of the piping are considered the same. Local stresses in the shell due to thermal expansion of the vessel itself, would be treated differently."

Does it mean that Caesar-II WRC-107 module or FE/PIPE & Nozzlepro local stress check are not as per ASME Section VIII, Div 2 ? If so, M/s Coade should clarify as per which code it is evaluating WRC-107 local stress for sustained, expansion & occasional load respectively ?

regards,

sam

In Dr. Becht's opinion,

"from the nozzle view point, the thermal expansion induced load of the piping is a primary load".

Then, the way we treat thermal expansion load local stress checking in WRC-107 module of CAESAR-II & NOZZLEPRO/FEPIPE must be nonconforming to Dr. Becht's opinion. I wish our moderators from M/s Coade to express their opinions too, in this regard: Is piping thermal expansion load a primary load to a pressure vessel - why ?

I have seen a laptop, cellphone, car designed by FEA entirely conforming to a single code; then why will we conform a chemical piping/vessel/structural/ pumps to B 31.3, SEC-VIII Div-1/AISC/API-610 with separate factor of safety & approach to design ? Can't we conform to a single code of B 31.3 in near future, while simulating all such items together!

regards,
sam

Sam et. al.,

The unification of codes is a desire that will never be fullfilled....

The codes have diverged for sound reasons and although they share some points they also diverge in places.

This work remains complex and I suppose mysterious to a certain extent and always will. The need for sound engineering judgement is paramount and one set of rules will never fit all problems, see what happens when one tries to remove thinking from the equation....

http://www.coade.com/ubb/msg_images/coaltrain.pps


As for Daves PVP paper about sustained stress it is copyright protected material and as such cannot be posted for free dissemination. However heres my take on the topic....

http://www.coade.com/newsletters/jan01.pdf
and
http://www.coade.com/newsletters/jul01.pdf