Topic Options
#3273 - 07/15/05 03:55 AM OCCASIONAL STRESS
flexMen Offline
Member

Registered: 05/06/05
Posts: 32
Loc: Spain
This is my doubt.

Imagine a piece of piping with 3 cases of load:

T1= 500º
T2= 25º
T3 = 560º

If I realize the calculation of those 3 cases of load in 3 different files, the verificacion of occasional stress code, IS CORRECT.

But analysing the 3 cases together in the same calculating file, the occasional stress es higher than the 100%.

The reason of that,I believe is due to that the SH of comparison of occasional stress is taken for the higher temperature of the 3 cases.

The case that gives an error ( code compliance ) is T2. But this way, the Sh material should be Sh at 25º, in my opinion.

I wonder : If the Sh considered by the program is the SH maximun for the 3 cases or only Sh for each one of the 3 cases??????


I think CAESAR uses the value Sh for the higher temperature of calculation, that is, 560º, whatever the calculation is. And what I think it should do is to use the Sh in each one of the cases separately.

I beg of you to give me a due in such a matter.

Thanks a lot.

Top
#3274 - 07/15/05 06:51 AM Re: OCCASIONAL STRESS
John Breen Offline
Member

Registered: 03/09/00
Posts: 482
Loc: Pittsburgh, PA (& Texas)
Hello flexMen,

Please explain what the temperature of the piping system has to do with Occasional Load stresses.

Temperature is all about expansion (displacement) stresses (please look at B31.1, paragraph 104.8.3) and those are secondary stresses. Sustained (additive) loads (please look at B31.1, paragraph 104.8.1) result in primary stresses and Occasional loads (please look at B31.1, paragraph 104.8.2) also result in primary stresses. When you look at Occasional loading cases you do not use load combinations that include expansion/contraction (temperature).

Please provide a listing of all the load cases and combinations that you are running. That will be necessary if someone is to help you sort out your problem.

Regards, John.
_________________________
John Breen

Top
#3275 - 07/15/05 07:02 AM Re: OCCASIONAL STRESS
Richard Ay Offline
Member

Registered: 12/13/99
Posts: 6226
Loc: Houston, Texas, USA
In determining the allowable stress for an "Occasional load case" ( k*Sh ), CAESAR II uses the Sh corresponding to the highest temperature vector used in the load cases.

So, in your conditions above, if the load cases contain T1, T2, and T3, then Sh3 will likely be the one used. If the load cases contain only T1 and T2, then Sh1 will likely be the one used.

Sustained or Sustained+Occasional stresses are the result of primary loads. Ignoring non-linear effects for a moment, primary loads don't change with the thermal movement of the piping system. Therefore you must use the lowest value of Sh, which typically corresponds to the highest temperature.

This is why in a simple set of load cases:

1) W+T1+P1 (OPE)
2) W+P1 (SUS)
3) L1 - L2 (EXP)

The Sh used for case 2 is the Sh corresponding to T1, and not the Sh corresonding to ambient.
_________________________
Regards,
Richard Ay - Consultant

Top
#3276 - 07/15/05 07:26 AM Re: OCCASIONAL STRESS
flexMen Offline
Member

Registered: 05/06/05
Posts: 32
Loc: Spain
Hi, Richard and John.

I don´t understand the answers.

I will explain what I think about this matter.

If we look at B31.1----------->

Socc=0,75iMa/z + 0,75iMb/z + Pdo/4t < k SH


0,75iMa/z-----> Doesn´t depend on temperature
0,75iMb/z-----> Doesn´t depend on temperature
Pdo/4t -------> Doesn´t depend on temperature

At this point we agree.

But when we compare the addition of this values with SH,I don´t agree with you.

Richard, you say that the value chosen is the lowest one. Where have you read it in B31.1?. I think,the value that must be compared to, is the value of basic allowable material stress at the hot temperature BUT in the case that we are analysing.

I'm so confused,

regards.

Top
#3277 - 07/15/05 07:31 AM Re: OCCASIONAL STRESS
Richard Ay Offline
Member

Registered: 12/13/99
Posts: 6226
Loc: Houston, Texas, USA
Say you have these load cases:

1) W+T1+P1 (OPE)
2) W+T2+P1 (OPE)
3) W+T3+P1 (OPE)
4) W+P1 (SUS)
5) L1 - L4 (EXP)
6) L2 - L4 (EXP)
7) L3 - L4 (EXP)

Now, when evaluating case 4, what Sh do you suppose you should use? You must use the lowest value of Sh. The same argument holds when you address SUS+OCC.
_________________________
Regards,
Richard Ay - Consultant

Top
#3278 - 07/15/05 07:45 AM Re: OCCASIONAL STRESS
flexMen Offline
Member

Registered: 05/06/05
Posts: 32
Loc: Spain
Yes Richard I understand you, but for me, this a hugh simplification.

Obviously If you calculate one only sustained case, to be in the safest way you must choose the lowest SH.

But I think that there would be 3 sustained cases and every sustained case compared to corresponding Sh.

According to your opinion, if i do 3 separate calculations (one for T1, two for T2 and three for T3 ), the results of stress obtaineds will be different with yours.

Top
#3279 - 07/15/05 08:29 AM Re: OCCASIONAL STRESS
Richard Ay Offline
Member

Registered: 12/13/99
Posts: 6226
Loc: Houston, Texas, USA
Quote:
But I think that there would be 3 sustained cases ...
You're probably right. Check the Technical Reference Manual on how to propely setup the hot sustained load cases.

Quote:
... and every sustained case compared to corresponding Sh.
Don't think I'm going to agree with this statement. While you may have different boundary conditions to evaluate primary streses for the various positions/conditions of the system, using anything other than the lowest Sh is non-conservative.
_________________________
Regards,
Richard Ay - Consultant

Top
#3280 - 07/15/05 09:35 AM Re: OCCASIONAL STRESS
John Breen Offline
Member

Registered: 03/09/00
Posts: 482
Loc: Pittsburgh, PA (& Texas)
OK, I have to quit posting in the AM before I have had my coffee. From the above I finally understand your question. Its all about allowable stress limits. Sorry that I misunderstood.

But now that I understand "a little" I am left to wonder what is the "event" that you are evaluating as your "occasional" load. Can that event happen at only 25 degrees? Can that event happen at any time the system is operating?

Regards, John.
_________________________
John Breen

Top
#3281 - 07/15/05 10:20 AM Re: OCCASIONAL STRESS
flexMen Offline
Member

Registered: 05/06/05
Posts: 32
Loc: Spain
Hi John. The matter is that, in case the event happens or not, if the event takes place, the material stress will be the same as the stress of the material according to the temperature that it presents.

Top
#3282 - 07/20/05 05:12 AM Re: OCCASIONAL STRESS
flexMen Offline
Member

Registered: 05/06/05
Posts: 32
Loc: Spain
I am still thinking that Caesar does not well enough this verification,without the aim of beginning a technical argument.
What is the purpose of it, for the same piping, for the following four models, the results of the occassional stress are different?

Model 1: with T1
Model 2: with T2
Model 3: With T3

Model 4 : With T1, T2 and T3.

Best Regards!

Top
#3283 - 07/20/05 05:11 PM Re: OCCASIONAL STRESS
John C. Luf Offline
Member

Registered: 03/25/02
Posts: 1110
Loc: U.S.A.
because the allowable S value the code uses is based upon temperature try reading the code
_________________________
Best Regards,

John C. Luf

Top
#3284 - 07/21/05 04:07 PM Re: OCCASIONAL STRESS
Edward Klein Offline
Member

Registered: 10/24/00
Posts: 334
Loc: Houston, Texas, USA
flexMen,

Trust me, my well experienced and educated colleagues know their stuff. Regarding you latest post, where you ran four separate models - two of them would not be a valid analysis for the sustained stresses because you wouldn't be be checking against the correct Sh.

I think the part that you are not quite getting is that your system has to satisfy the sustained stress requirement *at temperature*. As such, your operating temperature is what governs the Sh value.

You make this comment:
"But I think that there would be 3 sustained cases and every sustained case compared to corresponding Sh."

Now, for simplifcation, lets assume that none of your supports lift off. In that case, your calculated sustained stress is going to be the same for each of your temperature cases. Comparing them in two cases to a higher Sh because the temperature is lower provides no useful information, becuase those same calculated stresses still have to be lower than the minimum Sh, which is set by the highest operating temperature. If you system can regularly see a temperature, your sustained stresses have to be good for that Sh.

Now, as for the business of Occasional stresses, you say as your opening line:

"If I realize the calculation of those 3 cases of load in 3 different files, the verificacion of occasional stress code, IS CORRECT."

Actually, no, it isn't, for the same reason I gave above with regard to the sustained stresses. Two of your three files are not valid for calculating the sustained and occasional stresses because they do not consider the highest temperature that your system will operate at.

Now, here's the big if that I think had been hinted at - if you can justify that your occasional load case can only occur when your system is at T=25°F, then I think you could make a case for checking your occasional stresses based on Sh at 25°F. Quite frankly, I can't think of what kind of loading that would be, certainly niether wind nor seismic. If this is your situation, I'd like to hear it in detail.

Otherwise, you need to consider Sh based on the highest temperature. Let's take wind - it is meaningless to say that your pipe won't be overstressed due to wind load when the pipe is 25°F when it will be overstressed at 560°F. The fact that is "passes" at 25°F doesn't make your design any good. If it fails at 560°F, then it is an inadequate design and needs to be better supported/restrained.
_________________________
Edward L. Klein
Pipe Stress Engineer

All the world is a Spring

Top
#3285 - 07/21/05 06:04 PM Re: OCCASIONAL STRESS
John C. Luf Offline
Member

Registered: 03/25/02
Posts: 1110
Loc: U.S.A.
Post a sign at the fence line of the plant... all occasional loads may pass only when the systems are at 25F!!!!!!
_________________________
Best Regards,

John C. Luf

Top
#3286 - 07/22/05 04:33 AM Re: OCCASIONAL STRESS
flexMen Offline
Member

Registered: 05/06/05
Posts: 32
Loc: Spain
First of all, thank all of you for your answers.

I must say I am not agreeing with your opinions and I will say why:

I calculate my seismic cases as it follows:

Seism 1 = OPE 1 + Seism 1 - OPE 1(Case T high)

Seism 2 = OPE2 + Seism 2 - OPE 2 (Case Environment temperature)

In the case of piping and its butts for seism, those two are with enough gap for a good work in occasional cases.

By this way, the seism 1, will be the obtained load for from the dilatation point of the pipe till the millimetres that rest to arrive to the butt.

Meanwhile, in the seism 2 that won’t have dilatations, it will have a seism 2 bigger due to the butt is very far and the espace run should be bigger.

According to that, the stress caused by a seism for the case 2 is much bigger than in the case 1. On comparing with the Sh smaller, the pipe will fail for such situation.

For all of that, I think that for the stress Sh, it should be the correspondent for each case of operation, since I do not see that the Code indicates what Sh be the higher T of operation of the line, but THE OPERATING TEMPERATURE, that in the case 2 will be the environment temperature.

Best Regards.

Top
#3287 - 07/22/05 06:55 AM Re: OCCASIONAL STRESS
Edward Klein Offline
Member

Registered: 10/24/00
Posts: 334
Loc: Houston, Texas, USA
flexMen,

I'm afraid I'm having trouble understanding some of the terms you are using as they are very different from what I'm used to hearing here in Texas. I'm making some guesses at what you're trying to say. Can you clarify these terms:

butts (I think your meaning is restraints, line stops, guides)

dilations (thermal expansion?)

escape run (I can't figure this one out from the context)

However, assuming that butts does mean restraints and dilations refers to thermal expansion, then you are saying pretty much exactly what I said with this part:

"Now, here's the big if that I think had been hinted at - if you can justify that your occasional load case can only occur when your system is at T=25°F, then I think you could make a case for checking your occasional stresses based on Sh at 25°F. Quite frankly, I can't think of what kind of loading that would be, certainly niether wind nor seismic. If this is your situation, I'd like to hear it in detail."

It sounds like your seismic2 case won't have thermal expansion (dilations) becuase it won't be at a high temperature and you therefore want to be able to use the Sh at the lower temperature to check you occasional stresses for this case.

The only dispute that I have, and I think Mr. Luf agrees, is, how can you justify that your seismic2 case won't occur when the system is hot? There's not usually any warning for a seismic event like there would be with a hurricane, such that you could have an orderly shutdown of the unit and get the temperature down.
_________________________
Edward L. Klein
Pipe Stress Engineer

All the world is a Spring

Top
#3288 - 07/22/05 08:09 AM Re: OCCASIONAL STRESS
flexMen Offline
Member

Registered: 05/06/05
Posts: 32
Loc: Spain
Mr. Klein:

I think you have already understand my dispute,and the reason of using the Sh in each case.

According to your last sentence in your post, and the doubt of if the seismic2 case can occur when the system is hot, surely, the answer is YES, but if it happens it wouldn't be the seismic case 2 but the 1, in other words,the hurricane can happen in whatever case,when the pipe is cold and also when it is hot.If the seism happens when the pipe is cold, then, there is no reason to use the Sh smaller,since it is taken into account that the seism can occur when the pipe is hot in a second state of load. It is in this moment when it should be correct the use of a Sh smaller.

Regards,

Top
#3289 - 07/25/05 06:19 AM Re: OCCASIONAL STRESS
flexMen Offline
Member

Registered: 05/06/05
Posts: 32
Loc: Spain
Some other opinion?

Regards

Top
#3290 - 07/25/05 09:19 AM Re: OCCASIONAL STRESS
Edward Klein Offline
Member

Registered: 10/24/00
Posts: 334
Loc: Houston, Texas, USA
flexMen,

I think we're getting closer. Let's say that your occasional load case looks something like this:

W+P1+SEISMIC1

Now, unless you've got lifted supports that need to be removed, you are going to get the same calculated values of stress no matter what T you use in an operating case. You agree that the load needs to be checked against the lower Sh value due to the higher operating temperature.

So you then ask why not use the higher value of Sh when you run, possibly in a different file, with a lower operating temperature.

My question is - why run this case at all? Since you have to run this case against the lower Sh value, running it again against a higher Sh value adds nothing to your analysis and gives you no useful information.
_________________________
Edward L. Klein
Pipe Stress Engineer

All the world is a Spring

Top
#3291 - 07/25/05 11:16 AM Re: OCCASIONAL STRESS
Dave Diehl Offline
Member

Registered: 12/14/99
Posts: 2382
Loc: Houston, TX, USA
Watch it with that "W+P1+SEISMIC1" load case.

I believe most piping codes calculate stress from "W+P1" and add that stress to the values from "SEISMIC1". No signs come through this. In CAESAR II we call this SCALAR summation.

Using stress from your combination would be ALGEBRAIC summation.

For example, if "W+P1" gives a moment of -10 and "SEISMIC1" gives a moment of +5, the algebraic stress would be based on (-10)+(+5) or -5 while scalar summation would yield (10)+(5) or 15.
_________________________
Dave Diehl

Top
#3292 - 07/25/05 12:12 PM Re: OCCASIONAL STRESS
flexMen Offline
Member

Registered: 05/06/05
Posts: 32
Loc: Spain
Dear Edward and Dave:

Thank you so much for your answers.
I know all that you have said but I am not agree with it.

Evidently, if we value an only case W+P1+ Seismic1, we have no other chance to calculate in a suitable way, that to compare this value with the lower Sh, but, in my opinion, there are several "mistakes" in what you have post:

1) There is not only a sustained W+P1 but 3, one for each case.

2) There is non only a Seismic1 but 3.

3) I could agree to that the 3 sustained cases be equal(forgetting about non-linear), but seismic 1, seismic2, seismic3, never won't be the same for different states of temperature, since the moment that the Seismic1,2 and 3 are calculated in the following way:

-Seismic1 = (Ope1+seismic1)-Ope1
-Seismic2 = (Ope2+seismic2)-Ope2
-Seismic3 = (Ope3+seismic3)-Ope3

In conclusion:

Sus1+Seismic1 is different from Sus2+Seismic2 and also different from Sus3+Seismic3.

For that, being different, I insist on it, each one should be compared with its correspondant Sh.


Best Regards.

Top
#3293 - 07/25/05 06:39 PM Re: OCCASIONAL STRESS
John C. Luf Offline
Member

Registered: 03/25/02
Posts: 1110
Loc: U.S.A.
the calculated sustained stresses assuming no non-linearity calculated for T1 T2 and T3 with P being the same for all three temperatures should be identical, however the allowed code stress varies as a function of the metals strength at the 3 different temperatures. The higher the temperature the lower the code allowable.

So if the sustained stresses calculated are the same then the % of the code allowable varies. Thus for instance a system may not be over-stressed @ T2=25Deg.... it is possible with the same calculated stresses to be over-stresed at T1 and T3.

Now if it is possible for the plant to be operating during a seismic event (most probable) then the calculated Occasional stress level must be below the code allowable stresses for all the possible temperatures the event may occur at...

The bottom line is if Dave, Ed or myself checked your worked you would be quickly be re-doing it... the last time I did a literature review on the codes or flexibilty analysis I did'nt see a "flexmen" authoring any papers nor did I see "flexmen" as a member of any code comittee so read this final reply.

Take a look through the member list of B31.3 and finally I quote Plato...

"Wise men speak because they have something to say; Fools because they have to say something."

http://en.thinkexist.com/quotation/wise_men_speak_because_they_have_something_to_say/218003.html
_________________________
Best Regards,

John C. Luf

Top
#3294 - 07/26/05 04:46 AM Re: OCCASIONAL STRESS
flexMen Offline
Member

Registered: 05/06/05
Posts: 32
Loc: Spain
Dear Mr.Luf:

It is incredible to read your technical dissertations and also your magisterial lessons for everything.

You should have to take into account the sentence “Stupid is the one who makes stupidities” and also another, perhaps not of the same cultural level you have but really true: “There are not stupid questions but stupid who do not ask questions”.

It has nothing to do with me if you belong to one or several committees or not, in the same way, I have not said if I belong or not to a certain seismology committee. What is certain is that you cannot answer in such a pride way to whatever doubt of anybody only for disagreeing with you.

I do not doubt you know a lot about piping, but you should have to take more lessons in life, to treat people not like stupid without knowing them.

In no order of life does exist an absolute truth.

Plutarch quotes: “The real wise is only severe with himself, with the others, he is kind”

And also, the great genius Albert Einstein, suggested once: “It is a great relief to know the one’s limitations”

My best regards.

Top
#3295 - 07/26/05 06:59 AM Re: OCCASIONAL STRESS
SUPERPIPER Offline
Member

Registered: 08/13/03
Posts: 405
Loc: Europe
Some understanding of the code is required here.
The code is not a 'Manditory' legal requirement.
The code is just an opinion on what is reasonable
engineering practise, which has been put together by a bunch of "very" clever engineers.

It is up to YOU if you want to comply with the code or not. But, if you do want to comply,(most companies insist)then you have to follow the rules wether you like it or not.
I am sure somewhere in B31.3 it states that the Calculated code stress has no relationship to the actual stresses whitnessed by individual fibers etc.

The Code had been constructed on the foundations of skill and commitment brandished by the likes of the people who post replies here.(Breen,Luf,Deil,)etc, It is silly to dismiss their opinions on the codes, You need to have your analysis agree with these opinions and rules for code compliance, wether you like it or not.

Personaly, i thing the whole (code stress)thing is Nuts, But it works well, Ensures nobody gets hurt, and pays my wages.

Good Luck
_________________________
Best Regards


Top
#3296 - 07/26/05 07:29 AM Re: OCCASIONAL STRESS
John C. Luf Offline
Member

Registered: 03/25/02
Posts: 1110
Loc: U.S.A.
The code is a little bit nuts... it is the product of many people of a wide variety of talents and background. It is a set of guidelines written by comittee. But it has been quite successful overall and B31.3 is an internationally recognized code.

As for others as well as this particular subject I have said all I care to... advice has been given whether it is used or not its not my concern!
_________________________
Best Regards,

John C. Luf

Top
#3297 - 07/26/05 07:51 AM Re: OCCASIONAL STRESS
flexMen Offline
Member

Registered: 05/06/05
Posts: 32
Loc: Spain
My purpose on writing it is to clarify that in no moment I treated to put into doubt the Code but the contrary. I only wanted to clarify a personal doubt about occasional loads.

I am always paying attention to the other's opinion and the reading they make about the Code, moreover coming from a person like John C.Luf, who is involved in the developping of it.

Regards,

Top
#3298 - 07/26/05 11:10 AM Re: OCCASIONAL STRESS
John Breen Offline
Member

Registered: 03/09/00
Posts: 482
Loc: Pittsburgh, PA (& Texas)
Hello all,

I have been lurking here around this thread since it was started and I think it is sad that this discussion seems to have drifted away (from the technical issues) to where some of us are showing disrespect for others. I would like to congratulate everyone who participated in the discussion of this thread. I think it is good for all of us to try to understand and “be comfortable” with what we are doing for a living. I especially thank those who, although English is not your first language, have made the effort to participate in this English speaking discussion board. This was very unselfish of you and by doing this you have brought to the forum, your ideas which we would not have had the benefit of considering if you would not have posted. This discussion has been truly international and that is a good thing.

But back to the point at issue. After thinking about the statement of the question, perhaps flexMen may simply be saying that he understands that when calculating stresses the Code stress limit for the material at the highest design temperature will a limiting factor in the final design of any system. However, I think it may be that flexMen would prefer that CAESAR II do a different comparison for each load (combination) for each temperature (T1, T2, and T3). That would show that the seismic event would have to occur at one of the higher temperatures before Code stress limits would be exceeded. From the COADE point of view I can see where they could be reluctant to do that because that might be confusing if some person who is not a piping engineer were to just quickly look through the calculated tabular data. But then again, maybe I simply do not understand flexMen's point.

As for what I am reading from my esteemed colleagues in this thread, my sense of the discussion so far is that many of us are thinking the following. flexMen is designing one system (not three) and the materials of construction will be chosen only one time and the system will be built (with the understanding that it can accommodate all credible loading conditions). That piping system will be expected to provide a reliable service life of some finite number of years. My understanding of the service scenario is that the piping system at issue will from time to time be operating at its “normal operating temperature and pressure” (something less severe than the “design conditions”) and it will be subject to normal “in-service” deterioration. For the sake of discussion, allow me to speculate that the highest calculated bending stress in the system (at a branch connection of course) is at 75 percent of the Code stress limit. Given enough years of service without “upsets” the system would (since the Code covers fatigue fairly well) likely eventually fail due to corrosion (I won’t be around that many years). But, due diligence dictates that the “upset” conditions be considered and included in our analysis even though it is at least likely that the piping system at issue will never experience a seismic event.

Consider the piping system at a time when the temperature is 25 degrees F (I do not know if this is an operating condition or a shut-down condition but no matter). The system then experiences an earthquake of a non-trivial magnitude. The resulting calculated bending stress are well below the Code limit of 1.33 x Sh at the 25 degrees F temperature. The highest bending stress at our branch connection is 125 percent of Sh and there is no yielding of the material at the branch connection or at any other location in the piping (and I hope that no hangers were broken). Should the designer say “ah ha, I was right, the earthquake happened and my piping system is not damaged”?

A year later the same piping system is operating at “normal operating temperature and pressure” and let us (again, for the sake of discussion) assume the temperature is 550 degrees F. Stretching probability (and credibility) “a bit”, the piping system experiences an earthquake of the same magnitude as the event of the previous year. This time although the calculated bending stress at our branch connection is (just for discussion) exactly the same at it was due to the previous event, that bending stress is 138 percent of Sh (at temperature) or 104 percent of yield this time and so there is yielding of the material at the branch connection and maybe at other locations in the piping. Now remember, these are primary stresses and after the earthquake “rings down” there is still a pressure loading at the same temperature (no shakedown to elastic response for these loadings). What if it were thin wall pipe and the hoop (circumferential) stresses (twice the longitudinal stresses in our sustained case) were up at about 80 percent of allowable? With the material now yielding due to bending stresses....OOOps! Now, considering the above scenario, what Code occasional stress limit should the piping system be designed for? 1.33 times Sh at 25 degrees or 1.33 times Sh at 550 degrees? So, if I am understanding the responses of my esteemed colleagues, that is the essence of the answer you are posing for flexMen’s question.

I would be interested in knowing if I have gotten the question right (if you will comment please, flexMen – please expand upon my statement of the scenario) and I would be interested in knowing if I am understanding the answers correctly (please comment my esteemed colleagues).

And, if you are evaluating stress range per the new B31.3 Appendix P, what do you do with the above scenario? Hey, just wondering.

Regards, (and play nice), John.
_________________________
John Breen

Top
#3299 - 07/26/05 02:31 PM Re: OCCASIONAL STRESS
John C. Luf Offline
Member

Registered: 03/25/02
Posts: 1110
Loc: U.S.A.
God help us as far as Appendix P is concerned....

People have a hard enough time following the code methods that have been around for over 50 years as this and other threads indicate... and yes I helped with Appendix P and no I add no technical comments here at this pointin time.
_________________________
Best Regards,

John C. Luf

Top
#3300 - 07/26/05 03:39 PM Re: OCCASIONAL STRESS
flexMen Offline
Member

Registered: 05/06/05
Posts: 32
Loc: Spain
Dear all:

According to John's Breen post,I'll try to explain in other words.

1) The code calculation for this concrete situation is ASME B31.1

2) The situation I am referring to will be better reflected if we treat a case of a process pipe. This process pipe would be a steam pipe, which runs from the boiler till the turbine, including a bypass valve that will give service to the condenser.

There are 2 different possible situations where the pipe will operate.

a) The first one, where all the pipe, from the boiler to the turbine, will be at the highest temperature, and the pipe that goes to the condenser will be at environement temperature.

b)The second case, would be when the bypass valve appears in scene, so that the pipe that runs to the condenser will operate at a certain high temperature and the zone of turbine entry will be cold.

c)The third case won't be described, since it would be a similar case to the first one, where a second boiler would appear.

Said that, we can understand easily that the pipe of turbine entry may be very hot (Case a, 560 ºC),or at environement temperature (25ºC, case b)

The verification of this part of pipe is my discussion issue.

Do not forget what I have said previously in my post, the loads due to seismic event (Seismic1,2=Ope1,2+Seismic-Ope1,2(Algebraic), for that, we will have three seismic values instead of only one. Evidently, if there is only a value, the question has no sense).

After all, I am not convinced of when we verify the Socc,in the case b., CAESAR II is comparing against a temperature which is not the one of the pipe when the event happens, but it would be the temperature if the event happened when this part of pipe were working in the case a. Due to this fact, I am not totally agree with what the CAESAR II calculates.

I hope, I had explained better my doubt and, you , Mr Breen could give me your point of view.


Yours sincerely,

Top
#3301 - 07/27/05 07:30 AM Re: OCCASIONAL STRESS
P Massabie Offline
Member

Registered: 06/14/04
Posts: 50
Loc: Toronto Ontario
Flexmen,

When I was in the university one professor told us that material (any material) stress analysis had two components, and that these components were in both opposite sides of the stress equations. On one side are the applied loads, your calculated stresses; on the other side are the material properties. The prevailing condition is that applied side should be always lower than material side (unless we are talking about plastic conformation like extrusion, bending or so). You can play with the loads to reduce the calculated stresses, but you cannot change the material side unless you change the material itself, OR, you conduct a series of experiments and calculations on the material and convince everyone else that the material in that particular case is more resistant than expected. In other words, you will play with safety factors and material strength.
Said that, your situation is LIKE trying to convince everybody that we are using the wrong material side of the equation. The conservative point of view (wich is always preferred) is saying that we should use the lowest SH at calculation. You can’t go wrong with that, if the strength is higher then better: you have a “hidden” safety factor.
Regarding you specific topic. I’m not sure if the material can remember that it was at certain temperature (ergo at certain SH and SA) and then changed to another temperature (to another SH and SA), but every calculation points into that direction: The material is able to remember that it was at higher temperature and his strength is related to that particular temperature, even though, is not anymore at that temperature. And I think this is why you should consider always the lowest SH (highest temperature in your calculations).
Now, YOU are the engineer (or the responsible of this design) you can choose to manually change the SH for your calculations. The code gives permission to do so, but be aware that you’re playing with fire. If after the earthquake something noticeable happens to your system the insurance companies are going to look for someone to blame in order to avoid any payment. If someone goes into your stress files then you would have to explain WHY you decided to use a higher stress limit. And of course they are going to call an specialist, if happens to be one of the supporters of the lowest SH they’re so going to nail you… look, they have all they need: a failed system with not so defendable stress calculations, all your explanations will have a strong evidence against it.
If I can give you and a word of advice, play safe, use the lowest SH do some changes here and there (on the applied loads of the equation) and move on. In this way if something ever happens to your system you will be able to say that you followed the codes to the best of you knowledge and nobody would be able to blame you.

Regards
_________________________
P Massabie

Top
#3302 - 07/27/05 09:30 AM Re: OCCASIONAL STRESS
flexMen Offline
Member

Registered: 05/06/05
Posts: 32
Loc: Spain
Dear Mr Massabie:

I am TOTALLY AGREE with you, in everything you have mentioned.

Although I am still thinking that the strict calculation must be done with the correspondant
Sh, any engineer would be quiet if he is on behalf of safety. But referring to Mr Liang-Chuan Peng, at what point can go the safety factor against what economically can mean the excedent of such safety?

A teacher at my university said "If an engineer takes too much factors of safety in his calculations, perhaps, this factor of safety is due to his own unsafety".
I say that safety is a very important factor but the economical to assure you such safety it is also important.

Thank you very much for your comments.
Best Regards.

Top
#3303 - 07/27/05 08:11 PM Re: OCCASIONAL STRESS
John C. Luf Offline
Member

Registered: 03/25/02
Posts: 1110
Loc: U.S.A.
Perhaps I am overly conservative, it seems so are the majority of opinions expressed by various people in this thread your opinion seems to be the only voice supporting your position. However in order to clarify this issue as I understand once and for all for the community of people all around the world engaged in the design and analysis of piping systems I propose the following. “Flexmen” can send in a code inquiry to the B31.1 and B31.3 code committees. We each will be meeting soon this September so perhaps this can be settled soon.

I suggest you read the chapter of each code book on how to submit an inquiry…. Basically you send a question, which reflects your point of view and a Yes or No response. (see link below) If I don’t see an inquiry by the end of August I will submit an inquiry to both committees as follows…

Proposed Inquiries:

Question1:
Does B31.X require that all piping systems be evaluated for Occasional loads such as seismic acceleration forces, wind forces or Pressure relief valve unbalanced relieving loads, non-concurrently with each other…

Answer1:
Yes, occasional loads shall be considered see para. XXXX

Question 2:
Does B31.X when evaluating occasional loads such as seismic acceleration forces; wind forces or Pressure relief valve unbalanced relieving loads require that the Sh value used for the code allowed value be based upon the maximum operating temperature that the system will be operated at?

Answer2:
Yes see para. XXXX

This will help all parties concerned and the community as a whole… (Interestingly B31.3 and I believe B31.1 recently balloted something concerning seismic design)

An email sent out to the contact information below should suffice.

So the next move is yours FLEXMEN but then again if you do not submit said inquiries I will submit them and then will post the results on this thread, afterwards I will ask Coade to lock this thread.


Link for preparation of inquiry per B31.3 http://cstools.asme.org/csconnect/pdf/CommitteeFiles/3565.PDF

Contact information for B31.1
James N. Shih
The American Society of Mechanical Engineers
Three Park Avenue
New York, NY 10016
Phone: (212) 591-8539
Fax: (212) 591-8501
shihj@asme.org


Contact information for B31.3
Noel Lobo
The American Society of Mechanical Engineers
Three Park Avenue
New York, NY 10016
Phone: (212) 591-8540
Fax: (212) 591-8501
lobon@asme.org
_________________________
Best Regards,

John C. Luf

Top
#3304 - 09/11/05 10:35 AM Re: OCCASIONAL STRESS
John C. Luf Offline
Member

Registered: 03/25/02
Posts: 1110
Loc: U.S.A.
Well I looked over the agenda for the upcoming B31.3 meeting and I do not see any inquiries on this subject so I will submit the following request for interpretations to these comittees....

B31.1
Inquiry:
In accordance with ASME B31.1-2004, should piping systems exposed to wind be designed to withstand wind load?

Reply:
Yes, see paragraph 101.5.2
-------------------------------------------------
Inquiry:
In accordance with ASME B31.1-2004, Should piping systems in an earthquake-affected area consider earthquake lateral load acceleration forces?

Reply:
Yes, see paragraph 101.5.3. Also if the piping may experience another type of occasional load such as wind or pressure relieving forces those loads do not have to be considered as acting concurrently.

-------------------------------------------------
Inquiry:
In accordance with ASME B31.1-2004, Should piping systems that become loaded by a pressure relief event consider the loads and their calculated stresses as an occasional load?

Reply:
Yes, see paragraph 102.3.3
-------------------------------------------------
Inquiry:
In accordance with ASME B31.1-2004, when a piping system operates at multiple sets of coincident operating temperatures and pressures, and the system may at any random moment be loaded with an occasional load such as earthquake, wind, pressure relief, or any other type of occasional load that may occur during operation, should the occasional load stresses combined with the sustained stresses (ref: paragraph102.3.2 and 102.3.3) be compared against the lowest value of Sh used in paragraph 104.8.2 (12A) (12B) as determined from all of the operating temperatures?

Reply:
Yes, see paragraph 102.3.2(C) and 102.3.2(D)
-------------------------------------------------


--------------------------------------------------------------------------------------------------
B31.3

Inquiry:
In accordance with ASME B31.3-2004, should piping systems exposed to wind be designed to withstand wind load?

Reply:
Yes, see paragraph 301.5.2
-------------------------------------------------
Inquiry:
In accordance with ASME B31.3-2004, Should piping systems in an earthquake-affected area consider earthquake lateral load acceleration forces?

Reply:
Yes, see paragraph 301.5.3. Also if the piping may experience another type of occasional load such as wind or pressure relieving forces those loads do not have to be considered as acting concurrently.

-------------------------------------------------
Inquiry:
In accordance with ASME B31.3-2004, Should piping systems that become loaded by a pressure relief event consider the loads and their calculated stresses as an occasional load?

Reply:
Yes, see paragraph 301.5.5
-------------------------------------------------
Inquiry:
In accordance with ASME B31.3-2004, when a piping system operates at multiple sets of coincident operating temperatures and pressures, and the system may at any random moment be loaded with an occasional load such as earthquake, wind, pressure relief, or any other type of occasional load that may occur during operation, should the occasional load stresses combined with the sustained stresses (ref: paragraph 302.3.6) be compared against the lowest value of Sh as determined from all of the operating temperatures be used as the basic allowable stress in paragraph 302.3.6.

Reply:
Yes

-------------------------------------------------


I hope these will clear up this matter and take opinions out of the equation although I suppose the answers will only be the opinions of both comitees.
_________________________
Best Regards,

John C. Luf

Top
#3305 - 09/12/05 06:58 PM Re: OCCASIONAL STRESS
John C. Luf Offline
Member

Registered: 03/25/02
Posts: 1110
Loc: U.S.A.
comments window closes tonight....
_________________________
Best Regards,

John C. Luf

Top
#3306 - 09/12/05 09:56 PM Re: OCCASIONAL STRESS
John C. Luf Offline
Member

Registered: 03/25/02
Posts: 1110
Loc: U.S.A.
RFI's have been emailed in so my "conservative" opinion or others questionable opinions will be moot when these comittess add their opinions to the issue I will update this thread with the results.

B31.3 will render an opinion by the end of of this month... B31.1 I am unsure of but I believe they will be a bit longer....
_________________________
Best Regards,

John C. Luf

Top
#3307 - 09/22/05 07:16 PM Re: OCCASIONAL STRESS
John C. Luf Offline
Member

Registered: 03/25/02
Posts: 1110
Loc: U.S.A.
Well flexmen our opinions on this have been put to one side by the collective opinion of the B31.3 comitee... If the work you have performed as you proposed to do it was under the jurisdiction B31.3 your questionable method does not meet the rquirements of the B31.3 code, B31.1 has been given the following interpretation and will render their verdict at a later date, but I doubt that it will be significantly different in philosophy....

B31.3 05-13065

In accordance with ASME B31.3-2004, when a piping system operates at multiple sets of coincident operating temperatures and pressures, and the system may at any random moment be loaded with an occasional load such as earthquake, wind, or pressure relief, should all coincident operating pressures and temperatures expected during operation be considered and shall the basic allowable stress used in para. 302.3.6 be based on the applicable temperature for each condition?

Reply Yes.

TG B originally turned down the request as a consulting question... in other words the question was the responsibilty of the designer (who should be qualified as outlined in B31.3) but after looking over this thread and asking me why I sent in the RFI TG A provided the question and response!
_________________________
Best Regards,

John C. Luf

Top
#3308 - 10/11/05 03:41 AM Re: OCCASIONAL STRESS
Ross Sinclair Offline
Member

Registered: 09/06/01
Posts: 33
Loc: Auckland, NZ
Reading through all of these postings there seems to be a common view that you can’t check your occasional stresses at T2=25degC unless you can ABSOLUTELY guarantee that the occasional event won’t occur under the SUS+T2 situation.

However I don’t believe flexMen was trying to say this – rather he was trying to say that each case should be treated individually i.e.

Case 1 – calculate OCC stresses from OPE1+SEIS1-OPE1 using Sh for T1
Case 2 – calculate OCC stresses from OPE2+SEIS2-OPE2 using Sh for T2
Case 3 – calculate OCC stresses from OPE3+SEIS3-OPE3 using Sh for T3

What seems to have been overlooked in the ensuing discussion is that the stresses arising from the OCC loads in each of these cases can be different.

To illustrate this consider the following (hypothetical) case as per flexMan’s original example – a steam line to a turbine fitted with a bypass to a condenser.
Case 1 T1=500 degC (steamline to turbine hot)
Case 2 T2 = 25 degC (system on bypass, steamline at ambient)
Case 3 T3 = 560 deg C (steamline to turbine hot)

Assume the following situation:
- the pipeline runs along the x-axis
- there is a linestop installed on the pipeline with:
- gap back (-ve x direction) = 5mm
- gap forward (+ve x direction) = 100mm
- movements under thermal load at linestop are:
- Case 1 dx @ T1 = 80mm
- Case 2 dx @ T2 = 0mm
- Case 3 dx @ T3 = 90mm
- SEIS load acts in +ve x direction
- SEIS load has sufficient magnitude that from any operating condition it will close the forward gap to zero

So the displacements arising from the OCC case would be as follows:
- Case 1 = SEIS1 = 100mm gap – 80mm dx = 20mm
- Case 2 = SEIS2 = 100mm gap – 0mm dx = 100mm
- Case 3 = SEIS3 = 100mm gap – 90mm dx = 10mm

Assuming stresses are proportional to displacements in this case, then the OCC stresses for Case 2 will be much higher than the other two cases. It seems entirely reasonable that for Case 2 the code check should use Sh=25degC to take advantage of the higher allowable. This assumes that all OCC cases are combined with each OPE case.

Practically speaking it will often be a large task to generate all of these load combinations and I suspect that most users will prefer to use the current (conservative) CAESAR II methodology whereby the lowest value of Sh is used for checking all OCC cases.


John Luf
I note that your B31.3 RFI wording was slightly different to that posted on 12 Sept. The phrase:

“…should the occasional load stresses combined with the sustained stresses (ref: paragraph 302.3.6) be compared against the lowest value of Sh as determined from all of the operating temperatures be used as the basic allowable stress in paragraph 302.3.6.”

became:

“….should all coincident operating pressures and temperatures expected during operation be considered and shall the basic allowable stress used in para. 302.3.6 be based on the applicable temperature for each condition?”

So with the B31.3 committee response of “Yes”, my interpretation is that the methods proposed by flexMan are correct – provided that he checks all OCC cases against each OPE case.

This is where I become confused – this seems at odds with your statement “If the work you have performed as you proposed to do it was under the jurisdiction of B31.3 your questionable method does not meet the requirements of the B31.3 code.”

Can you explain how you came to this conclusion given the committee's response?

regards
_________________________
Ross Sinclair

Top
#3309 - 10/11/05 12:47 PM Re: OCCASIONAL STRESS
John C. Luf Offline
Member

Registered: 03/25/02
Posts: 1110
Loc: U.S.A.
It appeared to me perhaps incorrectly that flexmen was trying to avoid looking at all or at the very least the most significant OCC case.

As you have stated the gaps may vary on a system based upon temperature and this is why the algebraic differences between operating cases (with U and without) and then a scalar addition with the SUS stresses is the way to handle the mathmatics for a non-linear system.

Flexmen in this post seemed to be refuting what the comittee stated ....

posted July 27, 2005 11:30 AM
--------------------------------------------------------------------------------
Dear Mr Massabie:

I am TOTALLY AGREE with you, in everything you have mentioned.

Although I am still thinking that the strict calculation must be done with the correspondant
Sh, any engineer would be quiet if he is on behalf of safety. But referring to Mr Liang-Chuan Peng, at what point can go the safety factor against what economically can mean the excedent of such safety?

A teacher at my university said "If an engineer takes too much factors of safety in his calculations, perhaps, this factor of safety is due to his own unsafety".
I say that safety is a very important factor but the economical to assure you such safety it is also important."
-------------------------------------------------
This post about conservatism led me to believe he was agreeing but disagreeing. The B31 codes allow a designer to be more conservative in their aproach however for a designer to become less conservative requires some justification...

And the justification should be agreeable to the apropriate parties.

B31.3 has stated unequivocally that all cases must be examined I suspect B31.1 will render a similiar RFI This removes all speculation out of this issue.
_________________________
Best Regards,

John C. Luf

Top
#3310 - 10/12/05 05:38 AM Re: OCCASIONAL STRESS
YOGESH Offline
Member

Registered: 09/25/05
Posts: 7
Loc: Mumbai, India
I agree with everyone who has the opinion that sus stress should be checked with lowest Sh values (i.e. Sh of highest temperature)
While adding occasional stress with sustained stress I also compare them with k*Sh where Sh is of the highest temperature. However i would like to put one point here.
Say in my case there are 2 temperatures
T1 - Design Case. (this is eithre a start-up temperature or the temperature the line may see if there is some malfunctioning in the process)

T2 - Operating case. (This is the temperature at which the system will be running for the max time of the plant life.)

with the above two definations we can consider that T1 is a sort of occasional case. So if we compare the stress of SUS + OCC(Say Seismic) with the Sh of T1 then it would mean we are adding 2 occasional cases and the design will be too conservative which is not called for.

So in some exceptinal case if the system is showing failure in SUS+OCC stress when compared with Sh of highest temprature then can we create seperate model for occasional case with only one temprature i.e. T2 and compare sus+occ stress with Sh of T2?
Will this be in line with the code requirement?

Regards
_________________________
Yogesh

Top
#3311 - 10/12/05 06:24 AM Re: OCCASIONAL STRESS
John C. Luf Offline
Member

Registered: 03/25/02
Posts: 1110
Loc: U.S.A.
The 31.3 RFI seems clear, if your T1 is an operating temperature then it must be considered....

"B31.3 05-13065

In accordance with ASME B31.3-2004, when a piping system operates at multiple sets of coincident operating temperatures and pressures, and the system may at any random moment be loaded with an occasional load such as earthquake, wind, or pressure relief, should all coincident operating pressures and temperatures expected during operation be considered and shall the basic allowable stress used in para. 302.3.6 be based on the applicable temperature for each condition?

Reply Yes."
_________________________
Best Regards,

John C. Luf

Top
#3312 - 10/20/05 11:30 AM Re: OCCASIONAL STRESS
Gimini55 Offline
Member

Registered: 05/25/05
Posts: 33
Mr. Luf,

I apologize but was reading also B31.1 and I was looking for any exact or same meaning of your statement to power piping code. I want t know if this is the case too in B31.1. Thank you.
_________________________
Gimini55

Top
#3313 - 10/20/05 11:37 AM Re: OCCASIONAL STRESS
John C. Luf Offline
Member

Registered: 03/25/02
Posts: 1110
Loc: U.S.A.
The RFI I sent to the B31.1 comittee has not been answered yet, however I did furnish them this RFI we rendered in B31.3 so if I had to guess their response will be similar in nature....
_________________________
Best Regards,

John C. Luf

Top
#3314 - 10/20/05 12:37 PM Re: OCCASIONAL STRESS
Gimini55 Offline
Member

Registered: 05/25/05
Posts: 33
Thank you very much Mr. Luf.

Gimini55
_________________________
Gimini55

Top



Moderator:  Denny_Thomas, uribejl 
Who's Online
0 registered (), 115 Guests and 3 Spiders online.
Key: Admin, Global Mod, Mod
May
Su M Tu W Th F Sa
1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31
Forum Stats
12065 Members
14 Forums
16973 Topics
75151 Posts

Max Online: 303 @ 01/28/20 11:58 PM
Top Posters (30 Days)