ZETA
Safety System - Riser Monitoring & Safety System
Failure Modes of Well Intervention
Stacks
There are basically two ways an intervention stack can
fail.
Buckling
Buckling occurs due to instability of the structure. A
long slender column will fail when a compressive “buckling
load” is applied to the column. The classical method
of calculating this buckling load is known as the Euler
buckling calculation. There are models in the industry
that use Euler buckling calculations to try to determine
if an intervention stack is safe. There several problems
with this calculation technique:
A ”K” factor is required
for the Euler calculation, which is dependent upon
the end conditions of the column.
Real world intervention stacks often have more complicated
end conditions than those provided for with theoretical
Euler buckling. Often the K factor is misapplied.
Fortunately, the K factor error is usually on the side
of caution.
The Euler models apply to a straight, constant diameter,
weightless column. An intervention stack has large
BOPs, small lubricator, larger valves, all of which are far from
weightless! As a result, the predicted critical buckling
loads from the Euler model are higher than the actual
buckling loads. This can lead to a catastrophic buckling failure
during field operations.
These Euler models cannot address
the complex loading conditions and supports found in
a typical intervention stack.
Side loads such as coiled tubing reel back tension, bending
moments applied by off-center loads, guy wire
or chain supports attached at various locations along
the stack,
and the bending and dynamics of moving wellheads
and platforms (SPARs and TLPs) cannot be considered in these models.
With all of these shortcomings of the Euler approach,
why don’t we hear of more catastrophic buckling failures?
NOV CTES has analyzed many different intervention stacks as
part of this effort. We found that even though accurate
modeling had not been previously performed, most stacks
had buckling loads greater than the expected working loads.
While we’ve focused on catastrophic buckling events
to this point, there are other failure modes that can cost
significant time and money.
Bending
Bending of a stack component is considered a “failure” in
engineering terms, but often doesn’t result in a
catastrophic event such as the collapse of the stack or
a release of well pressure. Thus, bending failures are
often not counted as failures. This type of failure mode
may damage equipment and result in operational flat time,
but is more likely to be counted as a “near miss”,
if it is counted at all. Yet a bending failure is much
more common than a buckling failure, and increases the
cost associated with the well intervention.
Until this Zeta model was completed, there were no commercially
available models which determined the stresses in an
intervention stack due to pressure, bending, axial
load and dynamic
forces. A tall intervention stack may have relatively
low combined stresses when it is static, but when it
begins
to sway dynamically due to reel-back tension surge, wellhead
movement and/or platform movement, the stress levels
in the stack can increase dramatically.
Results from the Zeta Model
have also provided some additional insights that
may
not be immediately obvious. For example, bending
failure due to wellhead movement may be worse with lower
compressive loads than with higher
loads. This is counter to the concept of a buckling
load, which is worse as the load increases. Often a lifting
frame
or compensated support structure is used to apply tension
to the stack, in an attempt to prevent buckling, when
no buckling was eminent! Instead, these additional supports
may increase the risk of bending failure. Additional
cost
and complexity is being added, and the job risk is
being increased!
