tr in its various locations in Sect VIII

[BigTank]

This question is general in a sense. I mean to ask it for wherever a required thickness is needed for opening reinforcement, UG-45 nozzle neck thickness, the calculation of an unstayed flat head coefficient 'C', or for coincident ratio, et al.

What equation (UG-27 or Appendix 1), what temperature/pressure conditions (design or ambient), what joint efficiency (if not specified), and if it's pipe and you're using UG-27, is the wall undertoleranced?

I understand that there might not be a single answer to all situations, but what I've seen some softwares calculate don't seem to follow any common guideline (i.e. conservativism).

--------------------------------
Fitter, happier, more productive

 

[BigTank]

Are the crickets chirping to indicate my question is a matter of Code interpretation?

--------------------------------
Fitter, happier, more productive

 

[metengr]

The crickets are chirping for me because your question is not clear. Are you are asking if t_r that is referenced in say UG-34 is the same as t, minimum required thickness as referenced for a shell? The answer is it could be based on shell versus head minimum wall calculations.

 

[BigTank]

I think you might be overthinking it metengr. I'm asking if tr is to be calculated the same way, using the same equation, with the same design info (allowable stress, joint efficiency, etc) unless otherwise specified, in all cases where tr exists for a shell: UG-45, UG-37, UG-22, et al.

I'm trying to 'pull back' a bit from the specificity and understand how the Code implies its use. If you look at the individual definitions/declarations of tr in the locations I mention, and really think about what it is this quantity will accomplish in the calculation, you'll start to see what I mean.

In short, this is not a specific question with a specific answer (unless your interpretation lends itself to one). This is probably more a discussion of Code implemention of its inherent vagueries.

--------------------------------
Fitter, happier, more productive

 

[SnTMan]

BigTank, I have been trying to formulate and get to posting an answer for a couple of days, but the darn work keeps interrupting!

In my experience the definition of tr is kind of context sensitive, but in general, tr is calculated at design rather than ambient conditions, excludes corrosion allowance and tolerances.

tr when used in other calculations, such as nozzle reinforcement is calculated per the applicable formula for the component, i.e., UG-27 or UG-28 for cylinders, UG-32 of UG-33 for formed heads, UG-34 for flat heads. Joint efficiencies are also per the applicable formulas, EXCEPT where the calculation in question states differently such as in UG-37 where they are commonly taken as 1.0.

UG-45 is a bit of a different animal, as the minimum thicknesses are NEW, UNCORRODED thicknesses. But the UG-45(b)(4) thickness is subject to pipe tolerance per footnote 26.

Then, UW-16 requirements, at least in the software I am familiar with, disregards pipe tolerance, it is based on nominals only, I generally do these at uncorroded dimensions to be on the conservative side, with respect to Code compliance.

As to software peculiarites, sheesh...

Regards,

Mike

 

[BigTank]

Mike,

One of the more disturbing peculiarities of some software that I'm seeing is in the choosing of the allowables when a pipe shell is specified.

In some parts of a calculation-i.e. UG-37 nozzle reinforcement)-the allowable stress used to find required thickness used in the reinforcement area is the reduced stress (x0.85) divided by 0.85...that is almost full strength. But the calculation detail for the shell cylinder itself uses the reduced stress (x0.85) unmodified. This results in more 'extra metal' when considering the reinforcement: unconservative. I understand that the reinforcement calculations do contain some verbage relating this calculated thickness (per UG-37) being the local vessel thickness, but how about consistency across a calc set? This practice spills over into the proportioning factors of relative allowable strength.

As I mentioned earlier in the week, I suppose I'm just having a bad code week. I wish the committees would put some effort into clear language ('a seamless shell', 'the shell for pressure'...considering the geometry of YOUR shell, but with an allowable from a different line in II-D than that used in the UG-27 calc...oh, and drop the mill tolerance...include corrosion, but only sometimes using an etherial motivation) and organization (make sure you consult UG-40, UW-16, UW-12 UG-41, UCS-66 and UHA-51, and sometimes UG-37...for a single component under (1) loading only), rather than vagueries...but then again I suppose there motivations might be strong for keeping it the way it is (read: non-specifism is easier to litigate).

After all of that...thank you for your thoughts! Ha.

--------------------------------
Fitter, happier, more productive

 

[jte]

Mike et. al. -

Note that the 2010 edition UG-45 was completely rewritten in an effort to reduce the confusion inherent in the previous version. The pipe undertolerance is now inherent in the minimum thicknesses, the thickness is specified in inches rather than "standard wall", the wording is reduced to an algorithm type approach.

I wish the committees would put some effort into clear language...

I suppose there motivations might be strong for keeping it the way it is (read: non-specifism is easier to litigate).

I suspect you'd be surprised at the level of effort required to get even a seemingly minor change such as the UG-45 modification through. This isn't a situation where one person writes something and three others agree and it gets published. There are a lot of experienced people who will look at any change with a critical eye and try to see how it could be misinterpreted, misapplied, etc, and offer suggestions for making the proposed change better. The fun begins when a particular change gets contradictory suggestions for improvement.

Please attend a few code committee meetings prior to – even light heartedly and perhaps in jest – implying that the folks writing codes try to make them difficult in order to profit from being expert witnesses in litigation. Throwing rocks is all too easy. BigTank, I strongly encourage you contribute to the vessel engineering community by attending a committee meeting or two. Seriously. You clearly have an interest and enthusiasm for the Code. Coming up next… August in DC.

http://cstools.asme.org/csconnect/pdf/CommitteeFiles/28314.pdf

see page 2 for meetings after the August one. [Now, if you would wish to throw rocks at meeting location planning – hot areas in the summer, cold in the winter, I’m open to that!] The “Aw shucks, I’d love to but I can’t afford it” excuse is not valid. Attendance is free. I’m guessing the DC meeting is no more than a 6 hour drive. If you were to pick a particular day to attend, I’d vote for Tuesday, based on what I perceive to be your interests – in particular, I’d recommend Subgroup on Design. If you truly cannot attend, but have solid ideas – as in you have a change which you’ve written up and would like to see considered – then submit it to the committee secretary and I assure you, it will get due consideration. You’ll find contact information at

http://cstools.asme.org/csconnect/CommitteePages.cfm?Committee=N20110000

End rant. Don’t mean to sound mean/negative/etc. I just find it odd that people are quick to volunteer for local charitable organizations but neglect their volunteer powered professional organizations.

jt

 

[SnTMan]

BigTank, about nozzle reinforcement:

tr thicknesses ahould be per UG-27 / 28 except joint efficiencies should almost always be 1. This can give tr's smaller than then UG-27 calculation. For the nominal thicknesses, corrosion allowance should be deducted, pipe tolerance should be deducted for shell, but not for nozzle walls. See 37(a), tn.

Allowables shoud always be as tabulated in Sec II, Part D at design temperature. The only time an allowable should be divided by 0.85 is for allowable LONGITUDINAL stress in a welded tube or pipe, and then only when Notes G3 and G24 are present. I would very much mistrust any software doing differently.

As for Code complexity, as the Code itself states, it is not a handbook. And, in part that is what allows me to make a living

Regards,

Mike

 

[BigTank]

jte,

I have been aware that 2010 ed. is going to be significantly re-written. To what quantifiable extent I am not sure. I wasn't aware that UG-45 is being (or has been) re-written. That is good news.

Sincerely, I don't take your comments as negative. I whole-heartedly agree with you. You're right, DC is relatively close to me, and the cost of a hotel room isn't much of a deterrent in itself. I simply cannot lose 2 days at this point in this particular year. I was aware of this meeting and it's proximity, and was considering it, but I just can't do it at present.

I am going to make an effort however to put together some thoughts with some semblence of concision in the effort to, at the very least, begin the process of becoming active. I thank you for making that option apparent!

And I'm with you that throwing rocks is easy. I am passionate about 'rules' of any kind, and HATE myself for my lack of participation in their change or development. I am completely and solely at fault for this. In this particular case, spending the time I have (and do currently) with my nose in the book, I am beginning to feel comfortable enough to become active. After all, to make a significant contribution, especially to something this complicated, it is wise to learn as much about history and motivation before gelling a real opinion/interpretation or anything of the sort. Most of my posts on eng-tips are to this effort. Sometimes it's difficult for me to find the patience to listen, but I believe in the experience of others as being the most valuable resource for knowledge.

Thank you for the encouragement!

--------------------------------
Fitter, happier, more productive

 

[metengr]

jte;
If you are going to be at the ASME Code meetings in August, I will be there, as well (Section I activities, however). Perhaps we can chat over beer.

 

[jte]

BigTank-

We're on the same page, then. If we should be so fortunate as to be able to attend the same Code Week, I'll buy you a beer. For what its worth, I won't be in DC either. But I did attend the February and May meetings and will be there in Vancouver.

metengr-

I'd be happy to get together over a beer in Vancouver. I forgot if I've purchased tickets already, but would expect to arrive late Monday morning and depart Wednesday late afternoon or Thursday morning.

jt