Upheaval Buckling of Buried Pipeline 1

[Johnrem]

Dear Specialist;

I’m facing a problem of upheaval buckling of a gas injection buried pipeline. Originally the pipeline was buried 1meter below the ground (01 meter from ground to the top of the 10” pipeline).
As you can see in the attached photo, due to thermal expansion, the pipeline is coming out from its trench. I’m requested to provide a solution but I have no idea how to proceed.
1) Please help by giving your advices.
2) As you know, we used to provide expansion loops in order to absorb a thermal expansion of above ground pipes.
If we provide an underground loop for an underground pipeline, can someone tell me if it will absorb thermal expansion or soil lateral bearing effect will not allow. In other words, is it usual to provide underground loops for buried pipes?

Thanks in advance

 

[rneill]

This is more typically a problem at overbends where the pipe may loose full restraint ... I haven't seen an issue like this is straight flat terrain but then I've never worked in a desert ...

I assume that the sand is unable to provide lateral restraint to "fully restrain" the pipeline so it is buckling.

Options might include ...

1) providing anchor blocks to full restrain the pipeline at periodic intervals (probably not a practical solution).
2) providing buried expansion loops or offsets to allow the movement. However, this would typically require that you have soft backfill material behind the bends so that the pipe can move relatively freely otherwise the expansion loop won't work. A product from Dow called "Ethafoam" is something we've used for this purpose.
3) providing deeper burial so there is more weight of material above the pipeline to prevent lifting.

I recommend you retain a good pipeline stress engineer to review the situation and prepare a formal study / recommendation for you.

 

[Johnrem]

Thanks Rneil;

1) What do you mean by ''Problem at Overbends''? Is it different from upheaval buckling?
2) Do you think that your third solution (providing deeper burial) will not require cutting the pipe?

Many thanks

 

[rneill]

An overbend would be a location where there was a hill or a valley and the pipe was bent during installation to follow the ground contour with the bend pointing out of the ground. An underbend or sag bend would be at the bottom of a valley where the pipe is bent to follow the ground contour with the bend pointing into the ground. A typical problem at overbend locations is the pipe popping out of the ground.

A contractor can excavate the pipeline, temporarily hold it out of the ditch, deepen the ditch, place the pipeline into the new ditch, and then backfill. Whether all this can be done with the pipeline in service and/or without any cutting at all is up to the contractor to advise on.

Take a look at API 1117 "Recommended Practice for Movement of In-service Pipelines"

 

[MikeHalloran]

If the pipe had buckled laterally, you probably wouldn't have heard about it.

So I'm wondering if it's possible for a pipeline contractor to introduce a few subtle lateral S-bends while he's got the pipe out of the trench...



Mike Halloran
Pembroke Pines, FL, USA

 

[SNORGY]

I'm not a pipeline guy like the posters above, but I wouldn't be fixing this one while it was under pressure.

I like rneill's suggestion of installing foam-packed bends at this location...by the looks of things you would need to excavate quite a distance back from each point of soil exit. My first reaction is to excavate a trench parallel to the pipeline, then cut the pipeline across from each end of the new trench, install a foam-packed z-bend at each end, lay the old pipeline segment (if salvageable - look for wrinkles and other defects first) into the new trench, and weld it back up.

Regards,

SNORGY.

 

[JT100]

MikeH is correct in that if it was a lateral buckle you wouldnt see it, however in my experience lateral buckles tend to appear on non-buried lines and arent as highly stressed as UHBs due to the deflection being concentrated into one location (in lateral buckling the restraint is constant). i tend to deal with subsea pipelines, and am yet to come acorss UHB in action, it is always designed against though. and i have read about several cases occuring on-land in the Middle East and Africa (sandy soils). has the pipeline been subject to cyclic loading? maybe of pressure and temperature as ramping up production took place (ratcheting or pipeline walking may be to blame)? how long after the line came into servce did this occur?
as far as i am aware the only solution is to rebury the line. i would then run a stress analysis (FEA) based on the as-laid survey, and possibly install anchors at all overbends.

 

[StephenA]

With pure dry sand small grain (i.e. wind blown) I am wondering if vibration had a hand in the UHB with the particles moving down past the pipe as it rises.

Stephen Argles
Land & Marine

http://www.landandmarine.com

 

[BigInch]

A deeper burial depth is usually specified when there is a potential for upheaval buckling, provided no overstress is incurred. A deeper burial provides more restraining weight from the soil above. This is more likely the result of having little to no cohesion in the sand and thus a less than usual total effective restraining weight provided by the sand above. For sand you get very little effective width and instead of using the width of the trench, you might have to use 0.6 x the width when calculating the weight of soil above the pipe.

**********************
"The problem isn't finding the solution, its trying to get to the real question." BigInch

http://virtualpipeline.spaces.live.com/

 

[HotBend]

Judging from the photo it is definetly an up heaval buckling problem. The "over bend" above ground includes a field joint, which is not typical pipeline installation practice.

Before you can fix the problem, you first need to understand why this has happened. There is no mention of operating temperatures. Typically gas pipelines will operate around 3-5 deg C unless the gas is hot or the flow is isolated. Installation temperature is also a consideration and it is possible to intentionally pre-stress a pipeline to help overcome potential UHB problems during operation if the flowing gas is hot.

As mentioned by others, burial depth, or lack of cover is another contributing factor. From the photo it looks as though the pipeline has been buried in eolian sands (fine wind blown sand) that typically form the transient secondary dunes. It is possible that depth of cover has been reduced due to wind erosion and the upward forces due to expansion have overcome the backfill weight and friction forces. Normally in sand dunes, we install pipelines deeper into the medioum-coarse sand s which form the stable primary dunes. These dunes are not subject to movement and minimum depth of cover is maintained.

There are several options for remedial work, however I would need more information relating to current operating conditions before providing further advice.

 

[Johnrem]

Thank you a lot;

After investigations it came out that this UHB problem occurred near a summit of sand dune.
In the attached file you can see the location where this problem occurred.
Design and Service conditions of this pipeline are as follows:
Design Pressure = 371 barg.
Design Temperature = 85°C
Operating Pressure = 343 barg
Operating Temperature=20~70°C
Regarding installation temperature, I don't have exact information: It may vary from 11°c in winter up to 45°C in summer.

I agree with you that solution will be to bury again the pipeline but deeper this time.
My concern is the calculation that I shall carry out to convince the Client.
Does someone have any idea about such calculations? Any document to share?

Many thanks

 

[DSB123]

Johnrem,
So it was at an "overbend". All you need to do is calculate the required soil coverage to ensure this will not occur again at the overbend. Overbends and horizontal bends are the locations where this sort of thing happens.
There are standard calcs for determination of coverage requirements for overbends.

 

[Johnrem]

Many thanks DSB123;

Standards calculation for determination of coverage requirments for overbends is exactly what I'm looking for.
Do you have these standards calculations? Do you know the folmulas? Do you have any doc to share?
Do you thinks that the calculation will require software juste or just hand calculation?

Thanks

 

[rconner]

I am a little curious, (could you provide just a little bit more information) at what precise time in the life cycle of the pipeline etc. did this occur?

 

[Johnrem]

rconner;
Sorry I don't have such information;

 

[Eyeoftheneedle]

Johnreb,
The 'standard calculations' mentioned by others will be Hobbs-type formulation checks, which in themselves give you an idea of what's going on but are not detailed enough for what you need. Your best bet is to go to a reputable Design House and get them to do the calcs for you using FEA software such as SAGE.
You should also keep in mind that this pipeline has already buckled, so you will need to go the 'extra mile' to show that it won't happen again the next time you start up the line.
Fortunately the pipeline is in sand (not giving much resistance) so you probably don't have any excessive strain in the buckle but that's also something that the Design House would be able to advise you on.

For those that seem a little confused as to why it buckled up as opposed to sideways, it's a 'path of least resistance scenario'; i.e. there's more soil beside the pipeline to resist movement but the pipeline wants to move because of the driving force in the system due to internal temperature and pressure and up is the easiest way to relieve that force.

 

[athomas236]

Looks to me as if the membrane stresses in parts of the pipe have been stressed beyond the elastic limit. My concern would be whether the pipe is suitable for continued service.

regards,

athomas236

 

[Eyeoftheneedle]

Membrane stresses...elastic limit..come on..step out of NON-Limit State Codes for a second...help this guy out with something that can actually help him, don't scare him with issues that are overly conservative.

Pipelines these days are (and should be) designed specifically to limit states; i.e. tolerant of plastic strain. The shape shown initially looks nothing like a plastic deformation and is standard procedure for a pipeline with relatively high temperature and pressure in an undulating desert where local codes were not good enough to give adequate cover as they were based on lower temprature and pressure scenarios. There have been quite a few similar cases recently in very similar environments but they are generally not a problem if assessed properly by reputable companies who do this for a living.

 

[Johnrem]

Dear Specialists;

It's me again and I'm still with UHB problem.
Finally I could find some formulas to check if UHB phenomena will occur or not in a pipeline for given conditions. Some calculations are based on OTC Paper 6335 by A.C.Palmer.
In this paper, the required down load, Wi, to prevent upheaval buckling of pipe is a function of pipe flexural rigidity, EI, and the axial force, B, due to the longitudinal pressure and temperature loads.
Equation 12 of OTC paper 6335 gives the required down load Wi as follows:
Wi= [1.16 – 4.76 x {(EsI Wg / di)0.5 / B}] x B x (di x Wg / EI)0.5

where :
Wi= required down load to prevent upheaval buckling, kN/m
EsI = pipe flexural rigidity, kN-m2
di = imperfection depth, m
B = pipe axial load, kN
Wg = weight of pipe + contents, kN/m

I could understand all the parameters of the above equation except 'di' (imperfection depth').
I found in some papers that values of 0.1m, 0.2m, up 0.5m were assumed for 'di', then check if UHB will occur or not.

For me such values of 'di' are too small and can't be applicable to hills or dunes. Such small 'di' maybe related to imperfect trenching operation not to hills or dunes.

My questions are as follows:
1) Can 'di' accept big values (10m or 20m) which correspond to difference of elevation between hill crest and flat area elevation? Is the Palmer formula still applicable for big values of 'di'?
2) How to lay a pipeline on large dunes or hills during construction? Is it by creating a trench which radius shall be equal to elastic bend radius (around 500 time diameter of the pipe)?
3) Let us suppose that the solution for my pipeline to increase the burial depth up to 02m. Can someone teach me what shall the extent of the pipeline part that need to be reburied under 02m of soil? Is is the 'Imperfection Length' discussed in Palmer paper?


Many thanks in advance

 

[Johnrem]

Dears;
This time I attached some sketchs and photo to let you understand better my concern.