4 Buckle Triggers

Subsea Pipelines Global Buckling DesignWhat happens in a buckle, and how to get one

ScopeWhat happens in a buckle.

Pros and cons of engineered buckle triggers.

Naturally occurring buckle triggers and how to analyse them.

A bit of large-scale reality.

Buckle shape / modeMode how many lobes:

Buckle typeTwo

Upheaval

Lateral

(OK, three):Upheave that falls over

What happens in a bucklePreviously covered what a buckle isSnap BucklingTaylor Gan solutions

Failure modeNeed to exceed bending resistance of pipe as well as frictional resistance of soil (for a lateral buckle):

Link to snap buckling

Sense checks 1Snake lay buckles should initiate at

Post buckle force snap / not should smoothly decrease after buckle:

Acceptable model length?

Always look at the pictures!

Sense Checks 2Strain / displacement plots:

Sense Checks 3Moment plots:Much the same as strain.Equivalent moment / curvature should collapse on to moment curvature plot, analogous to stress / strain.

Sense Checks 4Trends:Expect critical buckle force to decrease with sleeper height.Three out of four isnt good enough:

Reality Check 1Can you actually install what you want?Sleeper size / weightSeabed topography at sleeper landing siteSnake lay radiusField layout limitationsBuoyancy

Reality Check 2Can you ensure you will get what you want?Racing Line through a snake lay corridorPotential settlementThe offshore effect...Sh*t happens! How much can you deal with.

Reality Check 3Rock dumpingAnchoringDo you really (I mean, really) need them?

Engineered Buckle Triggers Vertical Sleepers 1Proven in the fieldLow peak loadsElevated buckle crown reduced uncertainty of buckle restraintIntroduce spansVIVSlugging issuesInstallationOTC15310 Harrison et al

Engineered Buckle Triggers Vertical Sleepers 2OTC15310 Harrison et al

Engineered Buckle Triggers Modified Sleepers 1

Engineered Buckle Triggers Modified Sleepers 1Lowest buckle initiation force.

Can direct your buckle.As per sleepers.

With added installation issues.

Engineered Buckle Triggers - Buoyancy ModulesVery low buckle initiation forceBenign bucklesInstabilityDegradation over timeUncertainty over effects on lateral restraintLimited application

Engineered Buckle Triggers Snake LayCheapReliable and well-provenLow through-life loadsLikely to know buckle directionNot suitable for very soft soilsSpace constraints

Counteracts

Natural Buckle Triggers - BathymetryReliable already installedRough seabed lots of potential buckle initiatorsCan complement engineered triggersProving what will happen.

Natural Buckle Triggers As-Laid Out-of-Straightness 1The more buckles the betterCan control to some extent with lay tolerancesNo idea on what you will get until installationClient nervousness over rogue bucklesJustified with e.g. midline Tees, pipeline crossingsSAFEBUCK require you to consider variance in route bends as well as nominally straight pipe

Natural Buckle Triggers As-Laid Out-of-Straightness 2Two paths:Take your pipe and predict what may occur during worst-case installation manoeuvres.

Natural Buckle Triggers As-Laid Out-of-Straightness 3Care must be taken pipeline should be similar in all ways to existing dataset bending stiffness / soils / lay detailsORTake existing survey data and try to second-guess based on historical trends.Dominant aspect of potential feature:RadiusChange of angle

Natural Buckle Triggers As-Laid Out-of-Straightness 4What shape to select:Arc / Half sineSineSAFEBUCK give non-dimensionalised reference lengths for sine curves

RecapBuckle terminology.Consideration of all potential buckle triggers:Initiation forces.Post buckle behaviour.Sense check of FEA results.Reality check of potential design.

******Have a read through Cumming and Rathbone, OMAE2010 20262 Euler Buckling of Idealised Horizontal Pipeline Imperfections

*Model should be long enough that the buckle does not cause the model ends to twang and suddenly limit feed-in to buckle. Check force profile at end of step change of force at buckle and there should be a small amount of fully restrained force still evident, else the model is not long enough. Adequate model length can therefore be estimated by taking post-buckle force, buckle initation force, and the rate of force build up due to submerged weight and axial friction. If you can get between the post buckle force and the buckle initiation force in less than, or at most exactly, half the model length you are OK.***(0.75m Sleeper plot does not lie between 0.5m and 1m results).

*Witness the famous Pluto sliding sleepers. Also, for one of them, the slope of the seabed was so great that the seabed was 6m lower at one end of the 36m structure that the other.

*The Eni Blacktip project, installed by Saipem using the Castoro Otto specified snake lay with such loose installation tolerances that you could lay straight through the middle and put no bends in at all!

***Pluto sleeper again.

*Picture 1 Use an ROV actuated tool to push pipe sideways on sleeperPicture 2 Sloping Sleeper to cause pipe to naturally fall down to create the OOS*Picture 3 As per Rosa attach buoyancy around a pulley to pull pipe sidewaysPicture 4 Zero radius lay. Pipelay continue in a straight line until touchdown point just past structure, then lay barge changes heading with no associated pull to bend pipe around vertical post on structure**Courtesy of JPK KL

*Courtesy of JPK KL

*****Reference Rathbone and Cumming OMAE2010 20136, A holistic design approach for considering rogue buckle formation due to pipelay-induced out-of-straightness.*Using existing survey data is is SAFEBUCK approach and worked reasonably well on Pluto. Still have to be careful what that data actually means, and that the data you use as a basis is similar in terms of pipeline size / weight, water depth, underlying soil type, and installation method.

*Sine curve will tend to buckle at minimum force (intercept) as found from previous chart. This is because peak radius is not long enough, so benign growth against friction happens until feature is big enough to buckle pipeline as a column*

4 Buckle Triggers

Documents