stress analysis of patchwork quilt piping system 1

[e123344]

- Client wants us to stress check a modified pipe
re-routing utilising as much of the existing pipe as possible. The client is going to free issue us with new pipe to replace sections that are rejected after inspection. Seems simple enough, however, the original system was designed to BS 3608 using BS materials, whereas the new analysis will be to EN piping code using ASME materials, the old pipe has lower allowable, thicker wall and the client has also managed to order new material with a bore mis-matched to the old.

The schedule is critical due to the work being on a power station, ergo the stress analysis has to be conducted prior to the pipe being cut and the material designation in the route being finalised.

Would it be a reasonable approach to take the lowest allowable stress and use that in conjuction with the greater wall thickness for the analysis, or can a safe design only be achieved using the actually installed configuration of old and new pipe? My engineering group are suggesting they do the whole analysis using the lowest allowable, but that to me woudl suggest greater flexibiility and hence not necessarily a more onerous design case.

Furthermore, can anyone advise if there would be any stress reduction factors applicable for the use of mis-matched CS materials in this fashion ? i.e. would the sigma for the material with the lowest allowable actually be the lowest
allowable in the design or would it be reduced further due to factors already described; welding of dis-similar materials with different allowables, taper boring that will be required.

Many thanks.

 

[saplanti]

To me the reasonable approach is to use both materials with their own allowables and thicknesses. In the flexibility calculation you may need the high allowable at certain locations. This approach may take you out of trauble most of the time.

At the connection of these two materials you will be seeing the stresses for both materials. In terms of modelling of these two materials you will not waste time or additional money. So, I suggest you to do the right thing at the first place.

Hope it helps.

Ibrahim Demir

Ibrahim Demir

 

[e123344]

thanks for the response, but the problem is that we won't know what the configuration of the piping will be until we have cut it out and inspected. The new piping has to be installed within 2 weeks, my stress guys are telling me they need 6 weeks for the analysis (of 2 lines), i don't believe south wales will tolerate having no electricity so we can get the analysis 100% correct.

 

[saplanti]

The best thing is to do the analysis appropriately. I do not believe the analysis will take more than a week including the verification depending on your ability for planning, however the stress engineer needs to prepare the existing piping model before hand and modify it for materials and thicknesses after getting the cut locations.

In case you do not have the stress isometrics of existing piping, you need to go to the site and prepare pipe isometrics to create the stress model of the existing piping before cutting.

If the existing piping is adequate for the thermal expansion with the existing support system there will not be any surprises for the thermal expansion under the same conditions if the new pipe material and thickness is selected appropriately and cut locations are selected free from the bending moment due to thermal expansion.

Hope it helps,

Ibrahim Demir

 

[MJCronin]

e123344....

I suggest that you repost on the "Piping & fluid mechanics" forum. You will get many more eyeballs reviewing your problem and probably get more suggestions.

Looks like that you were handed a dangerous mess that was a product of many years of management incompetence and (IMHO)possible criminal neglect. High-Energy steam lines have ruptured in the past and have killed people.

Your current MBA strawboss requires a quick fix....of course

Specifically, what are the mix of piping materials and what are the maximum operating temperatures and pressures ?

Do you have simple carbon steel piping (e.g. A53/A106-B) operating at temperatures above 775F ?

BTW "saplanti" gives good advice and I agree with his post.

The best advice on ASME piping is given by John Breen in that other forum I mentioned

My opinions only

-MJC

 

[e123344]

MJC,

thanks for the advice re. forum to be honest I went to the forum where the top guys in terms of number and quality of response posted according to the stats, you funnliy enough are one of the guys whom i hoped took the bait.

I wholeheartedly agree on the correct approach but alas the pipe needs to be installed this year, and is inaccessible until the time it will be cut out, engineering as you're no doubt aware is often about compromise and managing the inherent risks that entails.

thanks for the help.

 

[JohnBreen]

With all due respect, any job worth doing is worth doing "correctly". Correctly means risk management and that implies additional design margin.

"a bore mis-matched to the old".... OK, so you will have to have some transition spools machined to ameliorate the mismatch. If you leave the mismatch at the weld line you will have a incipient crack.

"can a safe design only be achieved using the actually installed configuration of old and new pipe?".... Ahaaaa you used the "S" word. Power Station piping is typically high pressure, high temperature. There is a reason for doing piping system stress analysis. If it is not accurate it is worthless as a risk management tool.

"...so we can get the analysis 100% correct"... Actually, we know the best analysis we can do under the best attainable conditions will not be "100 percent correct" even if they dither about for 6 weeks and use the right material properties and section properties. The fit-up and fabrication will introduce unknown residual stresses (both due to welding and due to forced fit-up (inadvertent "cold-spring") and the inevitable dimensional errors). The vagaries of material and component fabrication will also introduce inaccuracies. Typically, if we come as close as 10 percent on getting the loadings correct we are doing well. The final calculated stresses will be no closer than 10 percent to the real operating stresses (one way or the other). THAT is why the Codes and Standards have "factors of safety" (aka "indices of ignorance") written into them. So knowing that, if we are willing to "compromise" in the modeling then we need a greater "index of ignorance" - we have to use lower allowable stresses than the Code maximum allowable stress.

"My engineering group are suggesting they do the whole analysis using the lowest allowable, but that to me would suggest greater flexibiility and hence not necessarily a more onerous design case"..... This is puzzling. Anyone who has done much piping stress analysis using good software (Caesar II) will know that the amount of extra time needed to change the pipe properties to suit the actual configuration is trivial - your engineering group is being disingenuous at best. Your instinct regarding the implied error in flexibility is well founded.

"Furthermore, can anyone advise if there would be any stress reduction factors applicable for the use of mis-matched CS materials in this fashion?"..... I don't know where you would get the SIF's to use with the exception of that shown in the Code and those SIF's are included to address the geometrical issues wiht mismatch (of course you would avoid or reduce this by using machined transitions). The material mismatch at the weld line (and the differences of expansion coefficient) is not addressed by any Code that I am aware of.

"my stress guys are telling me they need 6 weeks for the analysis (of 2 lines)"....... There will be a period of construction where an "as built" drawing can be continuously developed and updated to reflect that which was installed/modified - if good data is communicated to the stress analysis engineer in a timely manner it will not take six weeks (or even two) to complete a good analysis.

"engineering as you're no doubt aware is often about compromise and managing the inherent risks that entails"..... Compromise and risk management are diametrically opposed. The more compromise you have in the engineering assessment the more you move away from risk management. The system should be modeled accurately with all the actual material properties and geometries (and pipe section properties) and the current Codes and Standards should be used. THAT is risk management. If you KNOW there are even "small" inaccuracies in the analyses models you better introduce an additional layer of conservatism (lower allowable stresses), one that is consistent with the level of risk. Model it as accurately as possible but reduce the acceptance criteria (allowable stresses) to which your calculated stresses will be compared. Again, the old cliche "when in doubt, build it stout". Lives may be at stake here.

Regards, John.

 

[e123344]

John,

thanks for the response. I fully accept that the best way is to model accurately, and it would be my very strong preference to have all the data available in order to do so.

For the record I am against the gung-ho approach i see being adopted and I have been using this forum as a double check to confirm my fears.

Thanks for the point re. lower allowable, i hadn't considered using a lower "stress not to be exceeded" as an additional saftey margin.

thanks again.