Private Forums
T&D Analysis - Buckling
05 July 2019
Opinion Needed:
Performed T&D on an S shaped well on Well Plan on an 8-3/8" BHA, where buckling limit is predicted near the depth the well starts to become vertical. [ Max inclination ~ 15 deg]. Weight to buckle the string helically ~ 5 T in the analysis, on which drilling is operationally not feasible. To double check the results, SLB & HLB were asked to perform the analysis, where SLB results showed the onset of helical buckling at 15 T [ significantly higher value], but HLB results were same as our's [maybe because they use the same software i.e. Well plan] Is it because of different algorithms used for buckling in both the software? What might be the reasons?
Inclination profile and analysis results are attached
P.S All three analysis used the Stiff String model and same friction factors
9 answer(s)
Loïc BRILLAUD
Business Development
DrillScan
Total Posts: 1
Join Date: 25/08/17
Hi Umair,

As Elman mentions, our WellScan software allows to calculate the post-buckling shape of any string (drilling, completion, casing, liner...) and allows to quantify the consequences of buckling in the well.

The helical (and sinusoidal) buckling criteria usually used in the industry are too simple, and don't take into account the effects of Torque, Rotation and Doglegs. Our model allows a much deeper analysis.
We have published many papers showing the added value of this model (SPE-112571, SPE-119861, AADE 11-NTCE-9, SPE-151279 and many more)

It would be very interesting to play with the data on your specific well to see the differences with our model.
Can you please share more data about this case?
elman
SDE
BP
Total Posts: 20
Join Date: 01/09/12
Umair, DrillScan has been helpful in those cases.
umairbaig
Drilling Engineer
Pakistan Petroleum Ltd
Total Posts: 26
Join Date: 02/08/15
Dear All,

Thank you for the detailed responses, appreciated.

Elman,

Our well is near vertical and becomes vertical at a point, so that might be an issue. Also, which software does BP use then for vertical wells?
mhayes
Consultant Driling Engineer / ERD Advisor
Stanfield-Hayes Consulting
Total Posts: 41
Join Date: 25/03/11
Buckling, particularly Helical buckling is subject a number of different models factors dependent on author(s) of the original papers and how models are applied in the various available software's. 

As an overview, helical buckling is usually defined along the lines of:

Fh = fFs

were
Fh = Compression Force for the onset of helical buckling
Fs = Compression Force for the onset of sinusoidal buckling
f = buckling constant

Whilst there are limited calculations for the calculation of Fs (though there is more than one), there are multiple definitions of the value of f based on the author in question and ranges from -2.4 through to -5.66. Wellplan I believe uses ±2.83 in it's loading model. It's unloading model uses the SQRT(2).

But this isn't the only variable, there can also be a Buckling Limit Factor applied to Fs prior to the buckling constant, this is used to calibrate the buckling model for wellbore quality, tortuosity, etc. This is  Again various authors have published various values, Wellplan by default uses He-Killingstad, which is conveniently 1.0, but there published range is between 0.848 and 2. (This is all detailed in the Well Plan help pages)

As to where SLB Drillplan lies in the application of buckling criteria, it is not (from what I have observed) publicly published, so it is best to ask their representative how the number is calculated. In the past there has always been variation between providers in the models applied within their software solutions, and undoubtedly this remains a valid concern (this wasn't only a SLB vs Landmark variation, most providers applied a slightly different solution and had different results).

While it isn't hugely helpful to have different software applying different models that create different results it does highlight the importance of knowing the models being applied and their limitations rather than taking results on face value. 
elman
SDE
BP
Total Posts: 20
Join Date: 01/09/12
It is known to us that WellPlan's buckling calculations in vertical and near vertical conditions are quite conservative. We use different software to model buckling under those conditions and no issues so far.
HB uses WellPlan as well so you would get same results.
Dolman
Technical Director
OILSERV
Total Posts: 4
Join Date: 19/06/16
Make sure that you are using loading and not unloading in the calculations.  Unloading also called post buckling...it tells you how much you need to slack-off to unbuckle rather than load to buckle. Load to buckle is always significantly higher.  I dont think SLB has an unloading model.

Also check there hasn't been tortuosity applied to any of them as that can influence the results also.
sandeepdhawan
Principal Well Engineer
WellPerform Aps
Total Posts: 8
Join Date: 17/05/13
I have discussed this issue with Umair....Purpose of this post is to share the findings with My-Spread community.

Note: Only WellPlan case was looked at.

Two reasons for components buckling are:

a) At buckled component depth - Axial force buoyancy is ca -17.5 kips due to high pr. differential (internal - external) which is higher than the buckling threshold.

This is largely attributed to high flow rate in high mud wt. 2.0 s.g. environment. 

This also means that even when the string is off-bottom these components remain buckled with pumps on @ 450 gpm.

Switch off the pumps OR reduce flow rate to say 300 gpm - buckling disappears.

b) Buckling is also a function of stiffness. Some components especially Mud Motor have larger ID i.e. less stiffness hence less buckling threshold.

Conclusion: You have to live with it if circ. @ 450 gpm

Potential Fatigue and Twist-off risks mitigation to ALARP were discussed offline.

Hope this helps!

Note: Umair, you may share this post with SLB.
IainDD
Lead ERD Advisor
Merlin ERD Limited
Total Posts: 9
Join Date: 04/05/16
Umair,

I wouldn't say "maybe" for the fact you and Halliburton are getting the same answer. It should be expected. The fact that SLB are getting something different may point towards how buckling is actually calculated within DrillPlan. So then the question arises what bucking equations do the different software packages use.

The simple Paslay & Bogy equation for critical buckling has a multiplier of "sin inclination" and as such in a vertical hole the critical buckling limit is zero. Not an expert in all the different buckling equations, but I wouldn't be surprised if there is a difference between WellPlan and DrillPlan.

And in addtition to Daniel's list of inputs - have you included tortuosity? You are showing a pure vertical hole in your well plan. Maybe SLB have added tortusoity which has impacted their answer.

dr2019
CWI Engineer
SPREAD Associates
Total Posts: 11
Join Date: 11/05/19
Hi Umair,

If I were you I would check the entries are consistent across different simulators. I would look as a minimum to the following entries :

  • trajectory
  • drill string and BHA including connection dimensions etc 
  • casing and open hole configuration
  • mud weight and rheology
  • blocks weight
  • friction factors in open and cased hole ( as a note here I would run sensitivities for FF, also CH and OH FF are seldom the same) . Furthermore on Schlum soft I noticed they've used same FF for sliding and rotating..not quite common.
  • wob, tq etc 
  • calculation spacing is the same ( eg both simulators performed calculations every 30', or 100' or XX') 
Finally, if the results still won't match, I would get all parties involved together and clarify it.
 
Hope this helps.

Daniel 
Posted by

Umair Baig

Drilling Engineer

Pakistan Petroleum Ltd

Total Posts: 26
Jump to top of the page