We had a 4.5" Vam Top 12.6lb/ft CWSD tubing leak on a recent completion, identified after running a leak detection tool. We kept the leaking connection in tact on completion pull and pressure tested it showing the mill end to be leaking not the field end.
The Mill end was torqued at the mill to 106% optimum torque, the tong jaws grip pipe to pipe so torque goes through field and mill end. Has anyone seen something similar, particularly with Cleanwell Semi-Dry?
There are a number of other issues and human factors here but no conclusive outcome as yet. Been good to hear from the industry about similar situations, although I wouldnt wish it on anyone.
I have seen something similar (leaking tubing - mill end connection ) once.
End 2007 (Malaysia), we ran completion (including 7-5/8" VAM TOP 33.7 # SC L80 28 Cr).
We ran / tested completion OK - all good tests - no issues.
Next day we observed pressure on A annulus which we 'traced' to a mill end coupling / connection. The 28 Cr tubing had been remachined / recut (under VAM licence) and new couplings installed 'locally' (Labuan). We ended up with a time consuming 'challenging' WO.
My 33-years experience with the subject of centralizers and stop collars, taught me a couple of things:
FIRST The spring centralisers must be over a stop collar. NEVER between!. I.e. Bowspring centralisers MUST BE mounted around stop collars+
SECOND : Most of the stop collars available from centraliser suppliers are very short of requirements. AGIP (now ENI) requires some 25 Tons resistance to displacement whilst the normally supplied centraliser stop collars barely reaches 6 Tons.
I saw only 4 potential acceptable designs, albeit not used in the OCTG application
1. SHAFT-HUB connectores, used to fix boat propellers to the propulsion shafts.
2. Johnson Filtration Systems,use a reliable stop collar to fix sand screens on OCTG
3. Western Well Tools Inc. Central stop collar
4 SHRINK FIT stop collars could be a solution provided we discover a way around OCTG OD variations...
A few public references:
1982 Burnie Simpson test: ; stop collars too soft! ROBERT Gordon´s University, tests by Shell UK´s Burnie Simpson at the then RGIT 1990 Halliburton centralisers x RGIT; stop collars; S4 x B2?
Conventional Stop Collars Weatherford JSH better than the conventional ones but still vey weak Gemoco, between the two-> B&W friction lock weakest
1986 “Drilling Contractor”: New stop collars for Heavy Weight casing, NAM
1989 Johnson Filtration Systems stop collar?*
1990 Stop Collars Weatherford JSH > Gemoco >>> B&W friction lock
W/F centraliser on JS stop collar couldn’t rotate; resistance force large.
Western Well Tools Inc. Central stop collar
1990 Stop collars targets resistance to displacement : AGIP 25 tons; NAM 44 tons; API = Tensile Strength of casing.
1992 Centralisers OVER stop collars a MUST. OTHERWISE... Pulled casing w/ difficulty; lost all 15 centralisers + 30 stop collars from Must casing and 6 centralisers + 6 stopcollars
Worked 9 5/8” casing free with aid of acid pills. Lost centralisers and stop collars.
*1996 Austoil, stop collar sliding resistance .
WEATHERFORD: shrink fit stop collars.Stop collar for collarless OCTG and CRA. Non-rotating cement plugs; shrink-fit stop collars;
Darron stop collar weak
1996 API Spec 10D (5CT, Q1), free to spin; stop collar sliding resistance = Tensile Strength; Centraliser API RP 10D TOP-Co; stop collars drag resistance AGIP 25 kips; NAM 44 kips; JS 8 kips! Non Rotating Drillpipe/casing Protectors. Lost/Cored one; Corrosion of Al stop collar;
During my tenure with Shell International , we detected a flaw in Quality Control of OCTG preparation.
The mill-ends were seldom "phosphated" - only the "field ends"!
Hence, we experienced galling tendencies, incomplete make-up and leaks on the connection pipe x coupling,
Work with your supplier and ensure proper Third Party Inspection, until you are convinced the supplier´s QA/QC works to your satisfaction.
When I was at Shell Expro in the 80s, we experienced similar issues to the ones that Steve most eloquently described.
The direct cause was swarf in the mill end, with the root cause being inadequate quality control and cleanliness in the area where the coupling was bucked up to the mill end thread of the pipe.
I hope it helps ( a bit).
Decades ago I had a problem with leaking VAM tubing. We ran in, pressure tested OK, ran in some more, had a leak. Pulled back to the previous good pressure test, leaked. I ended up at the steel works where our tubing was manufactured and followed the full process from blast furnace to tube extrusion, machining, inspection. The final step took the machined tube on rollers to a leather stop pad where it rolled off to a rack and inspectors looked at it. I took a close look at the pin end and noticed scratches on the seal (this was new VAM) of the just-inspected joints. There were scratches deep enough to catch your fingernail (rule of thumb; must be more than four thou deep to catch). There was swarf embedded in the leather stop pad which was scratching the seal areas and the inspectors had passed all of these joints. For your mill end connections you might - or might not - be able to see damaged seal area if you back them out. It might be worth going to the shop where the collars are bucked on and taking a look at that whole process to see if there is anything compromising the integrity of the seal before or during make-up.
Good luck - Steve.