This is an imaginary situation, where someone wants to have a piping design pressure higher than the same allowed by tmin >= (PD/(2SEW+PY)+A)/0.875 in ASME B31.3 in elevated temperature service where creep & stress rupture strength governs ?

As per para 302.2.4(1), occasional variation of pressure is allowed.

Now, if someone determines u < 1 as per Appendix - V, is it possible to have sustained, not occasional, design pressure for small but safe service life in ASME B 31.3 higher than permitted?

regards,

sam

Hi Sam,

NO.

Sustained internal pressure causes primary stresses. Single applications of excessive primary stresses may cause failure (collapse, bursting,...). At temperatures less than those that cause creep, time is not a factor (except for pressure pulsation related fatigue). The B31 Codes don't really directly address continuous pressure pulsations that do not spike above the design pressure and there must be some reserve strength for that.

Consider all the loadings that must be accommodated by the allowable stress (Sh) of the material. Usually the piping thickness is designed for internal pressure and the circumferential (hoop) stress associated with it. Longitudinal pressure stress is half the magnitude of the circumferential stress and this leaves one half of the "remaining allowable" longitudinal stress for sustained weight and other sustained loadings. Note that B31.3 paragraph 302.2.4 refers to pressure variations as "OCCASIONAL" loadings, so you must consider the limit of paragraph 302.3.6. All the load-controlled primary stresses of sustained and occasional loadings must be accommodated by the basic material strength. Below creep temperatures the allowable stress (Sh) is 2/3 yield. Using an allowable (for occasional loadings) of 1.33 times Sh to address all the load-controlled primary stresses of sustained and occasional loadings puts you at 89 percent of yield. So the design margin is not as great as might be thought. Now consider the thermal (and other) displacement stresses that occur in concert with the primary stresses....

You do pose some interesting questions.

Regards, JOhn.

Excellent reply, John. I would like to add something here regarding solution to the imaginary situation.

Besides providing method for evaluating allowable PT variations, In my understanding, Appendix V also provide a basis for determining Design conditions (from among various operating/cyclic/surge conditions) according to which one should work out wall thcknesses or consider material of pipe from strength point of view. The reverse of it can be used for evaluation of correct design conditions only ............................. else, Consider replacing any surge pressure P(s)as design pressure for design wall thickness t(min) calculated as per original design pressure P(o), certainly it would fail the pipe in hoop stresses.

Designs in creep range are problematic. Small increase in pressure and expected life drops drastically. We have had cases where we have dropped the pressure just a bit to "buy" time to get replacement material. Also there are issues with material values. I do not know how good B31.1/B31.3 values have been but BS/DIN have used extrapolated allowables, which have turned out to be totally out.

I would not take too many changes in creep range.

Thanks to you all, Sirs.

I understand , in such a situation there is no other way than dropping the pressure to meet code requirement.

regards,

sam

John/Sam,,

Why cannot the decision be based on the service level and the intent like say A,B,C.D etc as outlined in Sec III? Even though the design may be as per B31.3, why these factors i.e. service levels and the intent be not considered? We can get substantial margin if Level D is considered , of course so long presure boundary is not violated.

Regards

Dear Anindya,

So far as I can remember Level D in NCA-2141 means faulted service limit which permits gross general deformation and some consequent loss of dimensional stability & damage requiring repair, which may require removal of component from service.

In the problem under discussion, as Breen-Sir commented, for thickness calculation due to internal pressure, the hoop stress is not self-limited; neither it is due to an occasional event like earthquake. Can you apply Level-D service limit for it ?

A proprietary design can have such freedom to disobey the relevant industry code alongwith a price tag of possible failure - financial, environmental & social.

regards,

sam

Hello all,

I am uncomfortable with "mixing the philosophies" of various piping Codes.

The mandate to the “sitting committees” of B31.3 has always been to provide the most simple methodologies for design/analysis, fabrication, erection, examination and testing consistent with an appropriate level of safety. The original B31.7 Nuclear Piping Code departed from this "simplified" approach slightly to attempt to introduce an additional level of rigor to assure the safety of the ecosphere against the "new" challenges of nuclear contamination and to assure the safety of the population against the "new" challenges of exposure to radioactivity. Subsequent Nuclear Codes have been driven by what we have learned, perceived safety issues and a federal “what if” mind-set. The B31.7 Code was withdrawn and the responsibility was transferred to the ASME B&PV Code, Section III where all the other nuclear Codes reside.

B31 Pressure Piping Systems are designed to be operated. Look to B31.1 or B31.3 Chapters II. The loading conditions described in Chapters II will not likely result in a damaged (“faulted”) piping system. If a B31 piping system is subjected to an occasional loading as described in Chapters II, it may remain in service (albeit, due diligence would in some cases indicate a thorough examination – earlier this month, on this board, we discussed unlikely events that might result in multiple layers of occasional loadings and the probability of permanent damage). The concept of a “faulted case” analysis is built upon containment (for as long a period as is needed for a “safe shut-down”) even if the system is subjected to a certain level of PERMANENT damage. A faulted (damaged) ASME Section III piping system will be raised and rebuilt and not placed back into service.

A “non-nuclear” piping system might be subjected to a seismic event in concert with a dynamic flow event (e.g., water hammer, steam hammer, slug flow, etc.) such that, despite the apparent continued containment of process fluids, the possibility of severe ratcheting (and other) damage is present. The structural integrity of piping systems subjected to such unlikely (or rare) loadings would be suspect. Due diligence would indicate a very thorough post-event examination of such systems. Post-event examinations MIGHT detect permanent deformation of piping but on the other hand ratcheting damage (the most likely life-limiting damage due to seismic loadings) might be very hard to detect especially in insulated piping systems. Upon the determination that there is significant permanent post-event damage (e.g., gross plastic deformation), the owner might conclude that unseen ratcheting damage is likely present and consider the piping system “faulted” (not to be returned to service). Perhaps the new ASME Post Construction Codes will address these issues (remember the B31 Codes are “new construction” Codes).

Returning to the original question of “can we “trade-off” some (presumably) “unused design margin” to allow sustained pressures that exceed Code maximum allowable pressures, for systems that are (by design) expected to be used only for a short time” – again I say, in my opinion, “no”.

Regards, John.

This is a very important statement from Sir-Breen. "I am uncomfortable with "mixing the philosophies" of various piping Codes".

Even though the code says "A designer capable of applying more rigorous analysis shall have to latitude to do so....", this does not mean we can take provisions from other (stringent or lenient) codes whenever we are in trouble. This certainly needs prior written approval from the owner and must be well documented. The owner may not concede to such requests unless it is absolutely justified and warranted.

Well John has about covered this from my end Jouko's observation are exactly on point.


I guess all I have to add is philosophy. We all will or have experienced "pressure" to say somethings OK when well maybe its not....


Each code has been developed and customized for a specific industry... representatives from all points of view have contributed to that code, designers, fabricators, owners governmental representatives and so on.

Sometimes it takes more than a keen trained mind to be a good engineer, sometimes it takes heart or courage to take an unpopular stand.

I have clearly mentioned in my response that a pressure boundary violation has to be avoided.What I have tried to say is that if a situation which can be considered tantamount to a faulted situation, i.e.the system need not remain operational during or after the event but pressure boundary will be retained,why the provisions of Sec III will not be followed? We are not doing anything technically incorrect or violating the code requirement( in the sense the B31 code is not forbidding us to use Sec III ).The higher allowable for Pm as given in Sec III definitely ensures aviodance of pressure boundary violation.

I am too small a person to make this comment, but I feel that where w/o compromisiing on safety if we can come out with solutions where cost of design can be minimized, may be by mixing of codes , why not?

Of course the additional QA requirements of SEC III should be addressed.

Regards