In piping stress analysis, for each calculated value we have an upper limit fixed by code, equipment vendor etc to keep within, but do we follow some lower limit too, to avoid unnecessary over-engineering?

It may be possible in the past, people could afford over-engineering, not now anymore, like newspapers nowadays show their daily visitor count in their website as their circulation figure to the advertisers instead of daily paper copy counts. We may have to forget much of past processes, borrow some & learn new processes for the activities we are paid to perform.

Some may laugh at this topic, but very soon this 'future' will be the 'present', when people will not move by cars to workplace, instead money & work will move by optical fiber already laid.


regards,
sam

Sam,

I'm not quite sure that I understand your meaning by a lower limit. As you say, we have upper limits of loads, stresses, deflections, etc defined by codes, vendors, clients, etc. Often times, the engineering required to satisfy these limits feels excessive, but they are what they are.

Now, if your suggesting that, when presented with a pump system, I analyze and find that I'm only at 20% of the allowable for a given force, that I work to tighten up the piping to get that force up to 80% or so?

My answer to that would be no, I would not. Quite frankly, as tight as budgets continue to grow for engineering work, I would count my blessings for having a competent designer, stamp it and used those saved hours for another problem.

I've been to the field enough times during construction to see how the process works. Unless you're working with some really exotic material or really big lines, there's no value in trying to engineer out an elbow or a few feet of pipe from the design.

Keep in mind that all those limits that we follow aren't the only considerations. I've had cases with systems that are well below allowables because a more flexible design is required to allow proper access. Or a loop that is bigger than it needs to be becuase it is simpler to run the loop all the way to the edge of the rack and catch the stringer vs. dropping a special trapeze to catch a "correct" size loop in the middle of the rack.

Dear Edward,

You have to understand the present situation, too.
With near $70/barrel crude price & competition from cheap labor & products elsewhere, can we afford the extra any more?

Just imagine, with the price I pay to a chinese clockmaker, I am expecting the life of a swiss grandfather clock!

Moreover, in some EPC acivity, these extras have to be factored & accounted for; if some client pays for it, it's okay! Otherwise, whatfor these tools of analyses are there if we don't cut down the number of supports so that resonably high, but within sustained allowable stress is attained; similar is the case for occasional & expansion stress allowables? If cutting down the flab meets the code & fattens my purse, why willn't I do it?

regards,
sam

Uhm, aren't the tools of analysis to make sure
whatever we build doesn't blow up and kill millions of people? At least in part.

But, I think the part you're missing is it's
a balance. If Engineering costs $100/hour
and Structural steel costs $1/lb, for example
(I have no idea what structural steel costs)
if all your extra engineering saves
$1 million and costs $2 million, then you
don't win. Also, typically plants don't get
smaller, they get bigger, so if you designed
that pipe rack to the nines, somebody added
4 or 5 extra pipes to it, and a stiff wind
came up and it broke, you still haven't won.

But hey, if you have a client that will pay
for engineering beyond a bare minimum I bet
most of the people on this forum would love
to know who it is.

:-)

Shannon

Dear sam，

What you said are reasonable,but it depends,
the lay out of piping depend on many things, such as enough space, limited time for re-analysis,macro lay out consideration for a project as a whole,it's not an easy things to optimize everything.Taking those most important into consideration seams more wise.

Dear Shanon,

I appreciate your 'feel-good' sense in 'typically plants don't get smaller, they get bigger' term.

But, in a professional culture, no transaction can be really a 'WIN-WIN' situation, to let someone 'WIN', the counterpart has to lose.

So, todays's consultants don't expect some favor from client, just the price in the prevalent market-it can be a penny to some & the heart for the other- all is fair in a market & client, too, doesn't expect anything extra or free. That is why I was blunt or provoked enough to write that 'if some client pays for it, it's okay! Client has to pay for his safety, too!

We, too, make nukes & missiles - but don't use against any one. We also get education, obey law & order, have the best-of-World network connectivity & bandwidth infrastructure. We may not spend so much on innovation, but on applications to solve customer's problems with cheap but honest labor - that's the difference.

regards,
sam

Dears,

I am sorry, if I had offended any of you.

We're brought up to hate Individualism, when used in self-destruction. When we network, we sincerely believe that it helps. We seldom fear to lose in public in a game without compromising on principle, as most have seen parents doing the same for collective win. But, with prosperity, our identity is getting lost, we, too, are becoming individualistic!

regards,

sam

Well, Sam, I would say that the goal is "design to the limit"
The term Over-engineering looks quite confusing in the terms that you originally posted. It looks much like the term "conservative design", which is what engineers normally tend to do. Our tendency is to add a safety factor over numbers (limits) that already have safety factors. In fact, adding few elbows more or making more rigids supports to reduce the loads to half of the allowables is "safer", but is expensive in terms of men-hours, materials, construction, etc. This attitude will probably create delays in finishing dates ( product of the summation of the hours here and there adding safety to our design).
In the other hand, reducing the number of elbows or run of pipe to make things reach the load limit, as mr. Klein points, is a nonsense too.
Constructability is something to consider in this process and should be allways in our minds: "How are they going to build this?" is a question that should be answered every time that we are "stressing" a pipe.
So yes mr. Sam, "design to the limit" where is safe, constructible, fast and cheap. And please, do it at once!

Best regards,

P Massabié

Yes and No,

Yes we probably over engineer what we engineer. I have never known a pipe failure on the areas we cover and spend hours discussing. eg. Pipe, tees, elbows, flanges. I am sure people know or such failures but the point I trying to make is that they are rare. However, I've known many failures on areas we do not cover. e.g. Welded shoes and attachments. A few years ago, I was in a metallurgist's office on a refinery. He had a sample cut from a failed attached on a steam line. I politely asked if I could borrow it to photograph. I'd criticised the detail without success for 25 years. "Why do you want to borrow it, take as many as you like. There are 10 more outside. We put more in the skip / dumpster a few months ago". I photographed the failure, wrote a note to the CE explaing why the "design" should be changed. "Its been a standard for 50 years and we are not going to change it now". doh! The same detail appears in every company standard I have seen in nearly 30 years.

Dear Andrew,

Pipe failure on the areas we cover and spend hours discussing in stress analysis don't occur due to the factor of safety or "conservative design" as per code or company standard as P Massabié has referred to.

But, this is regarding past & present. You can't be sure about the future.

Although we have FEA tools & material labs available to analyze & test the steam piping Welded shoes and attachments you discussed, nowadays many of us design the failures or short life-span for quick turn-around of customers. If someone buys a cheap toy, do you expect to have a FEA done on the toy to ensure that it will not break in the hand of a child & create unsafe situation at home ?

Maybe we should ask for some bundled insurance against failure & maintenance outages within lifetime assured from 'Wal-Mart' type EPC firms?

regards,
sam

trying to make sense of sam's philosophizing sure is tiresome. round-up and over-engineering? you must have a lot of time to ponder abstract ideas.

However, I'd like to know more about Mr. Weighell's steam shoes. What is the standard design and why is it a problem? Think you could start a new thread and share with those of us on the other side of the pond?

Thanks.
Richard

Uhm.. that's not "warm fuzzies". That's
"protect the public" (grin). However, my
main point was that it's a balance. Spending
2 dollars to save 1 dollar doesn't help the
client, nor does solving the current problem
without keeping the future at least in mind.

-Shannon

Hey, I suscribe what Mr. Richard Havard is proposing...
"...I'd like to know more about Mr. Weighell's steam shoes. What is the standard design and why is it a problem?..."

Regards,

P Massabié

Richard,
Nothing much to explain. Shoe cut from Tee section. Difficult to weld in situ because overhead therefore welds not always ideal. (I say difficult because pipe support welders are not as qualified or as controlled by weld procedures and inspections as pipeline welders). Poor welding rather than poor design is normally the reason given for failure. Acts as a cooling fin. Lots of delta T. High stress concentration. Cracks radiate from the end of the shoe. FE calcs show high stress at modest temp and load if you really needed to do a calc. Standard design proven by years of successful service so can't be changed! Then again, as Confusious said when he was an apprentice on the drawing board many years ago in China - its the same rate for rubbing out. (=erasing)

Andrew,
Good point about the shoe.
But for your information, you might be intrested...

We (ireland/Europe) are dipping our toes in PED.
One consequence is that ALL welded atachments have to be done by "qualified" welders. (6g?)

This is equating to the industry in ireland anhows switching to bolted shoes.

Although they cost more to make, the money saved in installation alone makes them worthwhile.

Although for critical stops etc, i still pescribe welded atachments.

Are you therefore infering that a welded shoe is acceptable if done by a qualified welder?

Superpiper,

My sentiments entirely. Being a simple sole, I try to avoid needing to solve problems. Very sensible approach you have in Ireland.

The poor welder often gets blamed for design problems. Welded shoes should ok if you check affects of local loading and design for realistic temperature differentials. Bottom of shoe cold etc. Temperature differentials during start-up are particularly interesting. You'll spend a several thousand dollars on FE and end up with..... No prizes for guessing.

Would you really want welded attachments to the pressure boundary of the pipe to be welded by an unqualified welder?

"One consequence is that ALL welded atachments have to be done by "qualified" welders. (6g?)"

I prefer clamp ons for many reasons.... however the WT has and will continue to be a popular item.

As for the blame game IMHO a poor designs field performance is usually blamed on bad fabrication and installation. Nobody ever questions the design because after all they have all that computer output!

Spot on except I mean't not as qualified rather than unqualified. IMHO in-field pipe support welds are not as well controlled or inspected as butt welds. Even the most qualified welder might perhaps not produce his best welds when the snow is coming down horizontally on the top of a structure. (and when the inspector has been called away on a rush job back in the site office to re-index his radiographs as once happened quite a few years ago).

Funny thing about bad weather and inspection....

They are mutually exclusive!

When B31.3 Iso Ized B31.3 for the PED I personally thought that the PED had some good points to ponder upon.

Dear Richard,

I feel glad that I made you tired by writing here - at least, you have read me!

But, if you term stressing for quality and merits in engineering as 'abstract and philosophizing', perhaps no engineer would have ever tried to improve upon!

regards,
sam

PS: Just for example, the chklists used by yesterday's experienced pipe stress engineers can be routinely filled up by young ones today. I have seen people using any low value irrespective of pipe diameter of expansion joint's lateral stiffness w/o back-up of vendor data or CAESAR II EXJ modeler. These people need to learn so that careless failure can be prevented.

Isoized.

Wonderfull phrase!.

In India, on 26th Jan - the republic day when we gave ourselves the constitution of a democratic nation, we saw a strange situation. Big Bazaar, indian incarnation of Wal-Mart Malls, as Wal-Mart has not entered India yet, was giving products at half price of max retail price (MRP) & who were buying & blocking the counters from ordinary consumers were the trading community itself, having only male children as females are systematically killed before or just after birth even today in this rich community. These people themeselves asked people to boycot the Mall sales.

Now, why has this happened ? The fat margins these parasite traders used to are now not realistic enough! Moreover, in a cruel market, who can really protect them & why? So far they had not allowed any competition, not from even other states - leave alone from foreign ones by force!

Unless we reduce the inefficiency & mediocracy we could afford with low gas price & no competition from globalization, market will not allow us to exist! To exist, we have to innovate & add value anyway!

regards,
sam