Pipe Stress Analysis and Existing Systems

[Robster1us]

I would like to thank everyone who responded to my previous thread on B31.3 Pipe Stress Analysis. It was quite informative.

I have received and been reading both the MW Kellogg and Peng and Peng books that this community suggested. They both contain a hug amount of information that will take me a little while to absorb, but I'm working on it.

For sustained loads, I am trying to understand what design criteria I can use not only for the new system, but to allow the owner of the system to make use of mechanical integrity inspection data on wall thickness to judge the health/remaining life of the system.

Typically in the system design process (if it were something new), I would find a preliminary wall thickness based on pressure (which is usually a max of 300 psi in systems I commonly deal with). I would find that Sch 40 pipe had significant excess wall, and that's what I would pick (service is below 250°F). That's industry standard for what I do (except in 1-1/2" and below we Sch 80 for the structural strength in case of impact or something like that). However, no one has criteria for what is OK after the system has been running for a few years. By that, I mean that you can take all of the pipe thickness measurements you want of an existing system, nobody can tell you what the minimum thickness is before you need to replace the piping (should it be corroded because of poor insulation maintenance, etc.)

It seems that, in the absence of wind/seismic forces (and of course assuming adequate flexibility for thermal stresses), one can simply say that wall thickness in pipe (and fittings with the stress intensification factor considered) can be given a minimum thickness equal to that calculated to withstand longitudinal pressure stress and stress due to bending from weight of pipe, contents, insulation, etc. (for those of you with Peng's book, the stress equation at the bottom of page 54), plus some safety factor, say, 10%.

Now, I understand some more about pipe stress than I did a month ago, but this seems like too simple a criterion. I would hate to suggest that this is the proper way to determine a minimum thickness for system piping (for a particular pipe size, MAWP, and configuration of supports) if it's not true. Please tell me where I'm being dangerous, as this appears to be a perfect learning opportunity to avoid someone else's mistakes.

BTW, since the piping has to be rated for full vacuum as well, are there any insights on what is typically the controlling case?

 

[kacarrol]

I believe you would use the Boiler Pressure Vessel Code Section VIII Div 1 UG-28 (Thickness of Shells and Tubes Under External Pressure) to calculate the wall thickness of piping that undergoes vacuum conditions.

K

 

[DSB123]

For fitness for service of existing piping systems revert to API 579.

 

[Robster1us]

OK, maybe I need to rephrase. I am talking about just the stress analysis portion of determining minimum wall thickness, which can be done during new system design to give some guidance to the PSM contractor years down the road on interpreting results. API 579 would not, I believe, tell you anything different on what the minimum wall thickness should be versus the ASME B31 codes. I'm not talking about a fitness for service analysis, which takes into account factors like how quickly the wall began thinning, estimating remaining useful life. I'm just talking about a basic criterion for minimum wall thickness, or if you like, how much excess the new system is designed with. Can anyone shed some light on this (please read my original question with this slant on it, as I guess it wasn't clear to begin with).

 

[pipesnpumps]

Robster, I literally have a headache trying to understand what you are asking.

I think what you are asking is how you can pre-calculate the minimum required thickness for any given service, and give this thickness to the owner/operator for this future reference.. For example, lets's say you have 150# steam with a 0.05" corrosion allowance and schedule 40 carbon steel piping with a 250 psig design pressure.

So in our example case you'd calculate the max. sustained stress in the system for that steam line (as it will be installed), and then you'd back figure the EOL thickness for this. Is that about right?

The problem with that is you seem to be implying that you'd calculate this required thickness (EOL) only for sustained loads.

What you will find is that the thickness required is alot smaller than is reasonable. I had this argument with a stupid supervisor years ago who insisted my specification requiring schedule 80 condensate piping was wrong, and that the company had been doing it for 20 years using schedule 40. I pointed out that the piping less 12.5% mil, less thread depth, less a reasonable corrosion allowance didn't leave any thickness for mechanical strength (bumping into the piping, someone stepping on it, hitting it with a hammer, etc.)

I think once you run the numbers you will find the same it true. What you really need to be looking at is EOL service thickness, and the above posters are sending you in the right directions.

Apologies if I mis-understood what you were asking, if you could focus on one question rather than jumping around to so many topics it might be easier to get a decent answer

 

[Robster1us]

Thanks Pipes. You did get to the heart of my question. I wasn't trying to jump around on topics so much as give background to why I'm asking. I've learned in this forum that many people will read something that seems like areasonable question and put no thought about what the person is really asking into their flippant answer. What I was giving were my question and some underlying assumptions so that I wouldn't get the "you haven't given enough information" response. Sorry if it was confusing.
Anyway, thank you for your answer. This forum can be very informative, but I find in most of the threads I have read, the poster seems to be asking one question and the respondants answering a different question that they sort of made up themselves. I don't know if it's because they didn't take time to thing about the poster's intent, or that they just can't help but put their own slant on it. It's a tough problem that I have tried to compensate for with additional background. Seems it doesn't always help.

 

[MJCronin]

Robster,

Congrats on your initiative on trying to learn Piping Engineering

But.....Do you understand the purpose and function of Piping Line Specifications ? (also called piping specs)

Take a look at the sample specification on this website:

http://www.pip.org/practices/practicelist1.asp?p=6.

Note how this "group of requirements".....called a piping spec is meant to cover many service conditions at a wide variety of temperatures and pressures..

Does this help ?

 

[Robster1us]

Hi MJ. Thanks for your input. I've written many pipe specs and purchase requirements for piping systems, both for B31.3 normal fluid service and B31.5 systems. That's not really what I'm after.

Knowing that a pipe provides the proper strength, and properly specifying it per the code (with respect to thickness and material), is not where I'm lacking, especially in my industry where almost all carbon steel piping is specified the same or similarly, partly due to industry standards, partly due to code requirements. That includes use of Sch 80 small bore piping, as I mentioned above, and other practices to make sure there is adequate strength other than just for pressure and weight.

Where I'm falling short is not on the new system, but on the system after it's run for a while, the plant has allowed the insulation to fail in places, hasn't kept up with the paint, and the pipe wall has become pitted in places on the outside. This is a typical mode of degradation rather than from internal corrosion or erosion, due to properly-specified materials and pipe sizes.

It would be of value to be able to say to the owner of the pipe system, after they've had a mechanical integrity inspection (or even provide as part of the original design, since that's when calculations are done), what to do with all the thickness readings they just had taken. How thin is too thin? I guess it's difficult for me to see how this is only possible through the use of API 579. The system was originally designed with some minimum thickness calculated, based on the stresses that the pipe would see (or at least could be calculated). This doesn't change over the years (outside the creep range, pressures are not terribly high, and material is not subject to significant aging). I'm not speaking to how quickly the system will deteriorate. That is clearly covered by API 579 and requires multiple data points of wall thickness, to my understanding.

What I'm after is this. Practically speaking, it's good to be able to tell an owner, "you're getting very close to a reasonable minimum wall thicness on this pipe. You should price out replacing it." How do you do that?

Niether Peng nor MW Kellogg seems to speak to this. They also don't seem to give any guidance on how to model the stress of the impact of the hammer mentioned in an earlier reply, or someone standing on it. At least they don't speak to it where I'm understanding it. Am I reaching for the stars in trying to understand how to calculate this minimum wall thickness prior to replacement? If it's not possible or practical, OK. But I have to think that you don't have to be an API 579 expert or have a CAESAR II license in order to come up with this basic criterion.

Is it maybe as asimple as saying something like, the original design had a 1/16" corrosion allowance? If you have burned through that, and wish to continue using the system without replacing the piping, you need to talk to someone who can analyze it per API 579? Is that a viable option? This doesn't really help if it wasn't your design to begin with and you don't know the corrosion allowance.

Anyway, I know everyone is trying to help, and I am very appreciative. I do not have access in my daily work to someone who has this knowledge, and this is basically the only place I can ask questions. Thanks to all.

 

[amarmona]

I want to calculate the pressure and velocity in case of waterhammer in a pipe having valve at one end and reservoir at other end. The valve is closed abruptly and the diameter of the pipe is reducing towards the valve. I need a MATLAB program. Can anyone help me.
Thanks

 

[Robster1us]

Amarmona,

If you have a new question, please start a new thread.

Thanks

 

[pipesnpumps]

Robster1us, ok now I "get" what you are after.

Structural minimum thickness is not in the code anywhere as you have found, and is subjective based on what you think the pipe will be subjected to.

Below are some numbers I was given about 10 years ago by my supervisor (who supervised the author of Caesar II many years before that). He did alot of his stress analyses by hand using an ancient HP calculator.

Structural Minimum thickness, CS :
0.5 to 3": 0.1"
4 to 5": 0.11"
6 to 10": 0.15"
12 to 24": 0.19"

I have never done what you are asking about, but I have considered it some more. First, it would need to account for all loads (occasional, thermal), not just sustained loads as pointed out earlier. The purpose would be to calculate the minimum thickness that is still within code allowables.

You are looking for a retirement thickness without the benefit of an actual stress analysis based in the installed geometry; it would be more for a pipe specification. Perhaps you could make some broad sweeping assumptions about the stress levels from the various sustained, occasional, and thermal cases, and then document those assumptions along with the retirement thickness (which corresponds to those stress level assumptions).

Otherwise the 'retirement thickness' could be misinterpreted by others later on down the road to be the more traditional fully corroded thickness which usually still has some meat left in it (margin above code minimum thickness in all parts of the system)

If you do that, make sure the retirement thickness is at least the thickness listed above for basic mechanical strength. I don't think you have to worry about full vacuum in your piping; you might want to double check large diameter very thin piping, its been many years since I first decided it would never be an issue. The poster above referring to ASME code for external pressure has me baffled and wondering. The only time I use the BPVC is when a hot tap welded tee will be hydrotested (e.g. before the coupon is cut, the hot tap branch is hydrotested at 220 psig which acts on the outside of the main). This was a client requirement to check the allowable external pressure, and to my recollection was never an issue.

 

[kacarrol]

Pipesnpumps,

The reason why I wrote the BPVC is because the one time that I did a vacuum calculation on piping (for piping following a steam out of a system) I was using B31.3 which states:

“304.1.3 Straight Pipe Under External Pressure. To determine wall thickness and stiffening requirements
for straight pipe under external pressure, the procedure outlined in the BPV Code, Section VIII, Division 1, UG-28 through UG-30 shall be followed…..”

And if you are using B31.1: “100.1.3 This Code does not apply to the following:….building heating and distribution steam and condensate piping designed for 15 psig [100 kPa (gage)] or less, or hot water heating systems designed for 30 psig [200 kPa (gage)] or less”

Vacuum pressure is a funny thing, cylindrical vessels hold up much better under internal pressure than they do with external pressure. You can get away with thinner vessel wall when there is only internal pressure. There are loads of examples on the internet where pressure vessels have collapsed under the slightest vacuum.

To give this a practical sense just look at a pop can. Under normal circumstances the internal pressure is around 25-35 psi (brand and temperature dependant). But if you take the same can, boil a tablespoon of water in it, and quickly turn it upside down in a bath of ice water it will crush really really quickly under the relatively small vacuum created when the steam condenses (maximum 14.7 psi vacuum). That's half the pressure!

Hope that helps someone!
K

 

[kacarrol]

Ha ha, sorry about that, put in the wrong section of code for B31.1 that quote should read:
"104.1.3 Straight Pipe Under External Pressure. For determining wall thickness and stiffening requirements for straight pipe under external pressure, the procedures outlined in UG-28, UG-29, and UG-30 of Section VIII, Division 1 of the ASME Boiler and Pressure Vessel Code shall be followed."

 

[pipesnpumps]

kacarrol,
I agree pressure vessels can easily collapse, but why do you suggest he needs to check for this? under what conditions would a pipe designed for say, 150 psig, collapse under a full vacuum?

Like I said, it has been years but I think I ran the numbers and even schedule 10 was plenty thick enough for FV.

I remember using jacketed piping with schedule 40 stainless steel core piping and 200 psig in the jacket pipe, so that is 200 psig of external pressure. Still plenty of margin left at 13X full vacuum.

Now I am curious what the breakpoints are.

 

[kacarrol]

pipesnpumps,

I don't know where the break points are either, but I always like to know where the thresh hold is. Granted, the wall thickness calc in B31.1 or B31.3 may produce the thickness required to hold up to the vacuum conditions but the code says that there is a different calc for pipes under external pressure. The OP was writing about pipes that have been in service for a while and have a reduced wall thickness, maybe sch 10 is okay but maybe the pipe has reduced to a point where it's not okay.

The OP seems keen on learning piping and knowing where the boundaries lie is something I always want to know, I always get nervous when people say "that's the way we've always done it" because you never know when you will get one of those cases that lies outside the beaten path.

K