It is reported in a journal that in 2004, out of 1.8 billion medical prescriptions made in US, half were incorrectly written. As one in 20 adults in US can't read English, many patients may misunderstand the prescription also.

In the case of todays' mass-produced piping stress analysis support load marking in piping stress iso for subsequent pipe support design, similar situations occur. Many times I have seen an insignificantly low value of support load for a reasonable size of pipe, directly lifted from restraint report of Caesar-II or similar piping analysis software. What could be the possible reaction from the pipe support designer? If he uses a standard pipe support standard, he will have some minimum rating based on pipe size & class & thus, any serious problem will be avoided. But, if he adopts the design by analysis procedure, he may provide a very weak support just suitable for the specified load. If the support designer also designs the support for a minimum default support stiffness (for rigid ones) as usually followed in nuclear NC/ND piping, this could be avoided. Otherwise, for support load, based on size & class, a minimum load (say, twice the NB (mm) in kg for verical load & 1.0 time the NB(mm) in kg in lateral/axial horizontal load) needs to be considered for support detail & structural design, just like retirement thickness in piping specification for small-bore piping.

I wish to learn from the forum members how do they counter such a situation so that no failure occurs due to inadequate support design due to too-accurate load marking?


regards,

sam

Depending on the size of the load, I will generally round up to at least the next highest 100lbs increment for any load that I report.

If you are marking something like Fy = 213lbs on a stress ISO, I can guarantee that you will not be well thought of as a stress engineer by those who evaluate your work.

Shoes and spring assemblies have mass that we typically don't account for. Pipe can have a +/- 12.5% variation in thickness. Fluid and insulation densities are approximate. Instrument people are going to use your line to hang conduit and instruments on. After you're done, the client maintenance guys are going to hang other small field runs lines from your pipe. Some dude is going to let his forklift get away from him and bump your control station.

I know that weight is critical for the offshore platform guys, so they seem to want to design them with aircraft criteria. However, for those of us in the downstream world, supports should be designed robustly.

I recently went through an exercise where a job is being constructed in the field. One area involved a packaged boiler where the piping design was done by the vendor and our pipe met theirs at an interface. The difference between our design and theirs was markedly different and I was called out to examine the vendor piping for the client. It passed all the stated requirements, but only just.

I suppose there is a balance that must be achieved. The piping system needs to look and be robustly restrained/supported, but you also don't want to end up with the Statue of Liberty pedestal under your base support, either.

Round up figures to suit.

To suit 'who'is the question.

I would argue to round up the figures to the nearest reasonable integer

Anybody designing to just 100% force values should be shot. even aircraft use FOS of 1.1

The person using the figures to design supports etc, should understand where the forces came from
and that the forces are subject to interpretation.
The forces are absolute, the use of them is not.

So for me the answer is (nearly) NO.
Just round out the figure sensibly.

eg
1189N=1200N

113N=115N

23450N=23500N

etc etc.

What is the reaction from Coade ? Can You provide some rounding up option for restraint report in next versions, if majority of us - CAESAR II users feel the need?

Till then, TJN's 'The forces are absolute, the use of them is not.' can be our justification.

regards,
sam

Edward and Superpiper,

Of course it really depends on who you are trying to impress. Really "accurate" stress calcs are written out to the nearest newton with wall thickness to mm+4 decimal plates.

Old duffers like me copy results to the nearst kN (0.5 kN for very small lines or 100 lbf in old money as Edward suggests). The problem with this approach is that it shows a lack of respect for accuracy or puts you in the old school of guided cantilever and TubeTurns rather than at the cutting edge.

I always smile when I think of the poor apprentice stamping a batch of spring cans with 15724N, 26981N, 897N, 12075N, 1721N, 19873N. He is surrounded by a stack of mis-stamped labels, has spent hours rivetting on new labels and then can't find the No 3.

As I have said before, best policy is to go with the flow. Its the same rate for writing 19873N as it is for 20KN. In fact, it is more likely to be a better rate for writing 19873N.

Still, no consensus has emerged - will we round up or not; if not, if support designer doesn't, will it not be unsafe?

It reminds me about the following I read recently:

"When Hurricane Katrina struck New Orleans, we thought we had the engineering prowess to build a city on a silt-based river delta, interrupting the natural deposition cycle and lulling hundreds of thousands of people into the false sense of security. Couple that with extracting oil and gas from below the delta and add to it years of draining the natural wetlands and coastal marshes, it was a recipe for disaster. We knew all that, but a 1998 programme to restore Louisiana’s coastal marsh system was never adopted. Why? The US$14 billion price tag was too high.

That cost seems negligible compared to the US$125 billion damage that Katrina caused, the US$50 billion repair bill, and the 1,000-plus priceless lives lost. - By Dr Chris Hails is Conservation Programme Director at World Wide Fund for Nature International, based in Gland, Switzerland. "

Had we decided to do what was needed, we could have saved so much! Where we know that our practice leads to unsafe situation, we need to speak up & change.

regards,

sam

I used to dump about 50 lbs. One time I met another lead who would prefer adding the same 30-50 lbs and we discussed this with structural engineers and we found out that even if we give them 100 lbs less they are still dumping extra huge amount of load that results in a structure overdesigned most often. If you go back in calculation from say the sizes of steels they have specified, you end up realizing such steel could accomodate 10x of the load you gave. So right now I give the load like in the nearest tens only. The thing is, a certain size of steel, WF for example has a bit wide range of allowable loads in any aspect, bending, torsion, etc. before going into another size. Meaning that if even if your off by 100 lbs, it won't matter seriously. However I think of the scenario where one could give a high extra load and it might jumped up to the next bigger size of steel and ifthe occurence is frequent i a project, it might results in uneconomical budget. With the errors mentioned by Edward Klien I am opt to agree on his criteria: nearest 100 lbs.

To quote Sam, "Still, no consensus has emerged - will we round up or not; if not, if support designer doesn't, will it not be unsafe?"

Jeepers creepers folks! If rounding up or not rounding up support loads is going to make or break a structure, we're all in a heap of trouble.

It is not our job to set the safety factor for the support designer. It's common knowledge that even with the most diligent engineering effort the accuracy is only within +/-10%, and with pipe stress analysis, I doubt it's even that close. You had better not set foot in a plant if the supports will fail at 1 lb more, 10 lbs more or even 10% more than the support loads submitted for design.

Some engineers staple the Caesar Restraint Report to the stress iso. I usually mark the loads and round up to the next 0.1, .5 or 1 kip depending on the line size and load magnitude. With preliminary routings and large lines, I generally add 10% and then round up because I know small configuration changes can affect the loads 10% or more. All of these methods are acceptable. If the support designer bumps the load you give again, great, if he doesn't, so what? On run-of-the-mill supports, forget the economics. Normally the minimum size members that the structural engineers would consider using will probably handle several times your load.

Let's not make a mountain out of a mole hill.

That's my 2¢,

Kevin Monroe is right to call us 'Jeepers creepers folks'! Mostly, we don't practise a very rationalized system of rounding up or factor of safety at every stage of design.

Recently Purdue researchers said that most car suspensions are "overengineered", as different component suppliers design their parts to be as rugged as possible, with liberal use of factor of safety in each stage.

If we perform integrated design of civil structures, mechanical equipments,equipments, pipings and pipe supports all combined, we can optimize the overall design. When this can't be done, atleast a transparent rounding up or factor of safety at every stage of design should be used
so that duplication of factor of safety does not occur.

regards,
sam

Sam,

I don't think that makes sense in a world of standardized structural shapes, pipe diameters, and wall thickness, bolt sizes, etc.

Plus, there's not a very long chain in our business for factors of safety to build up. I give a rounded load to my Civil/Structural engineer and he throws that in with all his other loads and criteria and picks his shape. In our world, it's going to be a W8 minimum anyway.

I'm i going to tell you why I round up: To avoid reissue drawings to structural again and again!
When you are doing piping desing, chances are that pipes are going to change along the design process. When this changes affects the stress calculations that you already have made, the last thing you want to do, is go to the structural guy and tell him that you have to change some loadings on the supports. If you do so, next thing you know is that you are holding half of the structural department because you haven't been able to submit the loads on the structure, and they are waiting for you. (every body knows that is not the complete truth, but hey, who cares it works!).
So I have to tell you that you should give yourself some reasonable maneuvering margin with loads and round up. How much is reasonable, well: It depends on you!

Best regards,