12 January 2009
We have a string of 5-1/2" DP with VX-57 (old Omsco TM4-57) threads.
The string is 20 months old and being used on development wells some
of which are moderate ERD. In the last well we have seen two
separate washouts in the DP tube. One about 2ft from a pin TJ and
the other exactly the same distance from the box TJ. The wash is
about 3-4cm long x 1cm think (like a slot). It seems to coincide
with an internal upset (tube ID is 4.6" TJ ID is 4" and there is an
taper in the last 2ft of the tube either end tube.
We will be performing material tests and additional inspections on
the two joints in question. The fact that we have two separate
incident so close together obviously gives cause for concern. The
pipe is being inspected to DS-1 cat.5 every ~4 wells on our
development. This equates to ~60,000 ft or 1000 hrs approx. Some
wells are more challenging than others.
Anyone have this kind of experience before and can add
Appreciate the help.
Devon Energy, Brasil
NOTE FROM MODERATOR : Please see photos, which have been loaded up
in the "Response" section of this discussion
For what you describe it seems like a typical failure of the internal upset ("Miu" too steep).
Please, refer to Paper IADC/SPE Nippon Steel FEA for optimum internal Drill-Pipe taper: Nippon Steel provides the finite element analysis and full-size fatigue tests to define the right geometry of the above internal taper.
To check it easily, it must be done BEFORE welding the Tool Joint. After TJ installation checking the "Miu" is more complicated. However, TUBOSCOPE had a tool for it (Sunny Rodgers` gadget).
The Quality Assurance of the Tubular Goods Provider should cater for that. The Quality Control of the Buyer should ensure the convenient taper.
I need an explanation about this Drillstring damage from Anyone have this kind of experience before and can add
Happy New Year old collleague (If you still remember me from EO).
Wash outs are fatigue cracks and the best way to prevent them is to
watch the energy that is put into the drillstring and used
efficiently in combination with monitoring and immediately
preventing energy wasting atctivities such as stick slip, vibration
etc. The well profile is sometimes a given based on target
objectives, casing setting depths and other (well) reasons.
Unfortunately at presently no decent software exists to take all
surface and downhole parameters every second or five second
(maximum) and evaluate all parameters to detect ineffective
drilling. So the best things you can do are:
a. prevent vibration, stick-slip, bit bounce and if it occurs change
drilling parameters and/or drilling system
b. calculate torque and drag and if actual are much higher you using
the energy inefficient which will go somewhere that may fatigue your
drillstring. Evaluate sliding/orienting, hole cleaning, reaming,
check tripping to get friction what it supposed to be
Good luck Dave
Peter de Rooij
I think the previous gentlemen have the main points covered, however
I did some studies a while ago to see the effect of different well
trajectories and dog-leg sizes versus depths to see the effect of
these variables on pipe used to drill and subsequent casings run.
The casing is a different story, but for the drill pipe I would say
a wash-out is just the easily observed secondary result of a fatigue
crack in drill pipe with sufficient toughness that prevents sudden
pipe body fracture which has an obvious catastrophic consequence.
In drill pipes with less than ideal upsets, the probability is
higher in the upset transition zone due to the geometric stress
concentration factor. Otherwise the highest probable area is about 2-
3 feet below the box because the upset weld zone X-sectional area
is, by design, at least 10% bigger in X-sectional area. Even without
the jaw mark effect, this region in whipping and rotational bending
see higher stress due to the effect of X-Sectional (Stiffness)
In concert with dog-legs in the top well section, when the tool
joint is riding on the high point, the section right below tool
joint also sees both high bending load and axial pull. This is where
pipe and tool joint box can be damaged through relatively low cycles
due to high side-load stresses (sometimes also accompanied by heat
especially if circulation is interrupted in, for example, a reaming
Fatigue effect is accumulative through service life. Below 1/3 of
yield stress, the fatigue life is almost infinite but deteriorates
rather quickly when stress is elevated.
There is however another source of stress concentration, namely
impurity in the steel from manufacture and the other pre-delivery
processes.. Therefore, if the pin end pipe body happens to have most
of the adverse factors (i.e. larger inclusion in thin wall section
with corrosion pit) then the pin end pipe body could still see wash-
out before box-end pipe body. (The probability is just lower).
Best regards”¦”¦hendo (aka Chris Henderson).
We have received the following from Ray Procter of Fearnley-Procter
QUOTE : Based on the info given - your inspection interval is twice
what it should be. 60000ft!!!
Refer to NS-2 which has a fully explained inspection frequency
system. The inspection interval for DP is 30000ft Max. ERD may
I would expect washouts in the DP in exactly this location if this
is exceeded by a lot - which is what has been done. If the string
is 20 months old and there have been no failures prior to this then
properties are probably ok - so that is a waste of cash as an
Inspect more frequently.
Regards, Ray Procter
Here's photos of the Drillstring washout to which this discussion
I agree this sounds like typical fatigue failures. Inspection may
not be effective in picking these up ahead of failure. Do you have
high string weights with high doglegs near surface?
The only way to prevent fatigue failures is to reduce the stress or
reduce the number of cycles. If the stress is below the threshold
values there is no problem, if you have high doglegs near surface
(i.e. where tension or stress is highest) then reducing the cycles
may be easiest solution - consider using a downhole motor to reduce
TH Hill DS-1 volume 2 has a good section on fatigue design and is
worth reading through.
Looks like washouts originating from fatigue cracks. My past
experience has often seen that MIU length can be a contributory
issue. Having MIU length greater than 3" will greatly increase
resistance to fatigue and having a good inspection regime to spot
developed cracks will also help. I am not sure how DS1 handles MIU
lengths but NS2 would flag MIUs less than 3" as a concern and
recommend not for use in ERD.
Head of Performance Services