Private Forums
Drill Pipe backout from TD saver sub and twisted at the connection between HWDP FH and the rest of the BHA at about 5600 ft below
06 September 2019

While backreaming out of hole, pumping at 1150 gpm with 120 RPM and max 30K torque, experienced total losses. Regained flow by varying rates to 200 gpm but could not backream out. String could still go down but not up. RPM was then set at 160 rpm and torque limit at 40K, string stalled and could not go up, taking a 100 k overpull. After about 5 mins, Observed string backout from saver sub and dropped into the well.

Fished the string successfully, but in two. One part was string to Jar stand crossover the other part from the HWDP to remaining of BHA. The boxend of the HWDP was seen to have some thread of the jar stand pin indicating a twist off.

What must have gone wrong?

Who has experienced this before?

17 answer(s)
Drilling Consultant
PT Drilling Services
Total Posts: 64
Join Date: 15/09/14
1) The cause of the losses could be due to packing off. 
2) Circulating at 200 gpm is not useful in moving cutting out of hole. And continue backreaming up will only compressing the pack-off stuff harder and harder. 
3) Many years ago, there was a chart distributed in our company mentioning about combination of overpull and torque. When you overpull, the torque limit of the pipe will be greatly reduced. This explained the reason for the twisted off eventhough it is still below the torque limit. I don't know the total pull on the string, eventhough it is mentioned 100kip overpull. 

In pack-off situation, where there is no returns, pipe can go down but not going up, one should not try to come up, but keep going down. Try to start with low circulation first, say 200 gpm. Sometimes we even have to go down a couple of stands to regain circulation.Then stage up the pump until full circulation or 80% of full circulation. Then try to circulate at least 1 bottoms up or until shakers are clean. If any concern of sidetracking, go to bottom and circulate. 
Until the hole is reasonable clean, then try to pull up, either by pumping out or backreaming out. 
I wouldn't want to pull out on elevator looking at hole condition as above. 

I have been working as a Drilling Supervisors for over 36 years, been through hundreds of pack-offs. But only 1 time that I could not get out of it, that was due to hole collapse. But that was in 1997, 22 years ago, due to hole situation and lack of experience to react to the situation.  

While I hate backreaming because it is time consuming, messy, especially with OBM and I have to spent more time on the Rig Floor because disaster can happen faster during backreaming. However I still prefer to backream in many situation, especially before logging or running casing or slotted liner.
In many wells, the hole is not like a gun barrel, thus even circulating at 2000 gpm at the bottom won't clean the hole completely. There are bound to be some areas that formed some cutting bed due to oversize hole or high angle hole. Many people believe in MWD's reading of ECD or T& D, but they are just indicators. And many operators tend to believe too much in those numbers which result into hole cleaning problems.    

Drilling supervisor
Sonatrach Petroleum Corporation
Total Posts: 7
Join Date: 11/02/16
Hi okenyi First of all I think the specialistes have said almost every thing about your case. Though, I want to share some lessons learned from the cases I ve experienced especially in deep and deviated wells. I Think the root cause of your problem is an excessive and erratic torque while drilling. That may be caused by 1- Hole profile : the more dog legs you have the higher the torque generated. Therefore, consider reaming each stand drilled. 2- Hole cleaning : the cuttings accumulated arround the BHA could generate extra torque and cause drill string stuck. 3- Bear in mind that the torque at the bit is much higher than in the surface especially if there was stick slip. Therefore if you increase the torque limite above 80- 85% DP make up torque that could lead to an overtorqued connections and even twist-off. Best regards
Drilling Specialist/Well Engineer/Training Consultant
Kingdom Drilling
Total Posts: 416
Join Date: 10/01/05
Case study late 80’s. First ERD well from offshore platform. 
Every 171/2 section was backreamed as apparent standard  best practices prior to this.
only had 5 in drillpipe in these days.
1. we changed connection practices 
2. We controlled Rop so well was being cleaned as it was drilled
2. We afforded more energy at bit to assure bit and cuttings removal efficiency and effectiveness.
3. We had no issues drilling a 6000ft high angle tangent 17 1/2 section. 
4. At td we assured Hope was clean.
5. Poh from td to casing shoe on elevators only. 

TP’s remark ‘first wellbore we have every pulled out of without backreaming’ 

Despite being longest section we were also way ahead of the curve.

lesson learned was we needed bigger pipe to afford more energy to system to be able to drill at higher Rop and clean hole while drilling.

We drilled two wells in similar fashion never having to backream at all. 

We have had similar experience where backreaming is exercised as a matter of course. Corrected failings that existed drilled a far better cylinder controlled drilling to assure we prevented vs had to cure problems and ended up well ahead of previous days vs depths curves. 

Do do the right things. 
Get things right first time.
Accept the evident physical ( parts) people and paperwork failings that exist. Correct these and far safer effective and efficient wellbores can rest assured be drilled. 

9 time out of 10 backreaming results because warning signs are not recognised analysed and identified early enough with correct evaluation and determination resulting.

My first reaction on my first top drive rig when I first saw the pipe rotating upwards was.
where is the physical evidence that proves this is a good thing to do for ones wellbore.
i then analysed this from a conditional physical mechanical hydraulic stability hole quality objective view that remains firm today. 

All the above evidence facts quite clearly states and  prove  that Downward rotation cleans a well several times more effectively than upwards rotation. 

In the 80’s 90’s good drillers also knew what to do. We did not need to afford them several pages of supposed best practice operators procedures to take a silk purse and turn it into ragged bag.

We evidently experience this today far too often. Ie Supposed best practices with little supporting facts figure and evident substance to back them up.     


One further point:

  • Mud was kept in spec.
  • Not one low or high viscous sweep was pumped on these two ERD wells. 

That at the time set the Horizontal/Vertical boundary record in this operating area. 

Well Site Drilling and Completion Supervisor
Greene's Energy Group
Total Posts: 5
Join Date: 12/10/19
First of all, I must commend all those who took their time to provide answers to this challenge.

Your analysis is apt and on point.

But then, I join those who inferred that everything was wrong with the drilling practices being implemented here.

Though the case did not give us information about the section of the hole that was just drilled. I infer that it is either 12.5" or 8.5" section. It is common practice in drilling ERD wells to complete your inclination build and turns in one section minimizing your sliding sections.

From the available information, it can be inffered that this section of open hole was not well drilled. But then, I want to say that the practice of backreaming with pumps on while tripping is totally wrong.

 First of all, prior to tripping, the driller is expected to circulate and condition the well prior to tripping. You are supposed to pick up the string, rotate and reciprocate the string while pumping at max. possible flowrate until shakers are clean. Note your mud ECD effect on the formation stability in choosing your flowrate. Not less than 1200gpm for this horizontal section.

As you start tripping, pick up the pipe without rotation or pumps on and monitor your up/down drag. If pipe is not free, continue circulation walking the pipe up and down. 2 joints per stroke. Then pump low density low viscosity pill. Note that the low density pill will not cause an underbalance situation. Then followed by a High viscosity pill to remove the dislodged cuttings. Then start tripping without pumps or rotation.

The pumps will only build dunes in the hole. Walking past the 60°-80° section. Rotate the pipe after tripping every 10 joints without the pumps or any over over pull. Ensure that the pipe is free to move up/down. Then continue tripping. At tight spots, assume it is cutting bed. Run in hole at least 3 joints or back to TD to circulate and condition the hole while rotating and reciprocating the string. Restart tripping. If tight spot has moved. It was cutting bed. But if tight spot is still there. It is a ledge or keyseat.

Was your pipe applying over pull in stages. Your over pull limit should not exceed half the BHA weight. If hole is still tight, run hole 2 joints and start rotating the pipe without the pumps. Then pull up again rotating the pipe with your TD.

After backreaming 80RPM (3 joints per hour) like 9 joints. Run in hole back to TD. You may need to ream back to TD. Circulate and condition the hole until shakers are clean before you resume tripping. A short trip(wiper trip) at the casing shoe is necessary.
Project Quality
SPREAD Associates
Total Posts: 108
Join Date: 06/05/09
You almost certainly packed off.  The only SPE paper in my files is no 92429, titled "Rotating While Packed Off ..."
Drilling Advisor
SPREAD Associates
Total Posts: 13
Join Date: 13/07/14
Backreaming with total losses happened to Vico in E.Kalimantan in Indonesia in 2004 experienced in the string parting due to severe heat checking.  This has been documented in the literature.  Essentially with total losses, there is no cooling where the tool joint/string section contacts the sand face.  In turn sufficient heat is generated to part the string.
Total Posts: 4
Join Date: 08/09/15

Nice discussion. All responses have been helpful. I have some additional information. The well was a directional well, inclination builing from around 5700 ft to 82 deg at 11000 ft, creating a bucket between 7200 ft and 8500 ft. The picture as attached shows the trend of events just before it happened.

Also from the pictures of the box end that back out from the TDS saversub, no noticeable damage. The picture from the box end downhole that parted also indicates some thread coatings of a pin on it, making us to believe it was twisted off.

Backreaming at full drilling rate has been the normal practice in this field for these kinds of wells. Should this continue?

Documents uploaded by user:
SPREAD Associates
Total Posts: 140
Join Date: 05/03/08
Hi  Kingsley,

Firstly, I agree with the other posters - if you are having to backream out of the hole, then you have a problem with the way the hole was drilled in the first place - i.e. at the very least, improper hole cleaning practices during drilling. So you have to examine your methodology for how you actually drill the hole section in the first place.

Secondly, look at your APWD data in the recovered tools memory. If you drilled to TD without losses and then suffered them while backreaming, the odds are that the hole packed off while backreaming (probably by a cuttings bed avalanche down the hole) creating a brief pressure spike (which may not show up on surface data) and the pressure build up below the pack off which fractured the formation.

Thirdly, if you can rotate, go down, but not up, then that is typically evidence of a keyseat in the hole somewhere.

This is always at a change in OD of the drill string, be it a DP tool joint, HWDP - DC X/O or even between different OD's of DC's if you are running a mixed string. Look at the side forces on your drill string to see where the modeling predicts them to be at their highest and look for any harder formation beds in that area - that might indicate where the problem lies.

Fourthly - I agree with the phrase 'brutal' to describe those reaming practices. The drill string whipping & banging around at those speeds will be beating the living daylights out of the side of your hole and just make things worse.

If you want a visual analogy of what the drill string is doing, take plum bob, or a length of string with a small weight tied on the end (so you have an equivalent of the DP and BHA).

Hold the weight in place while twisting & tightening the top of the string around as much as you can. Then let the weight go while holding the top still (i.e. the equivalent of the TDS stalling). See how much the string whips around as the weight unwinds. Feel what the top of the string is doing in your fingers. And then note what happens when the string is fully unwound - the weight keeps on rotating and winding the string up the other way.

I'm willing to bet an awful lot of money that if you examine the data from the MWD, it will show the string has turned backwards downhole.

And to take the analogy further, if you have 5,000' of 5" DP and the string is 0.1" in diameter, then to be the equivalent length of the drill string, you need a piece of string that is 100 feet long..

Good luck and let us know how it turns out.

All the best!

Drilling Specialist/Well Engineer/Training Consultant
Kingdom Drilling
Total Posts: 416
Join Date: 10/01/05
Losses resulted because you likely fractured well. 

Best  practice is typically run back in hole until normal torque and drag is observed and circulate a bottom up.

Then attempt to poh with no rotation walk before one runs and if required rotate out.

if you have to backream you have failed to drill a good cylindrical hole 

pulling 100k is also the wrong thing to do as you got stuck pulling and rotating out pulling another 100k just got you well and truly stuck likely around bha and within all the force present the string twisted off.

i would question best practices as there a few exhibited in the evidence been given.

it of feasible to drill high angle 171/2 wellbore control Rop so Hole is getting cleaned. Make connections without the.need to backream and poh on elevators! Any deviation from this evidently indicates things ain’t as right as we would like them to be?   
Drilling Specialist/Well Engineer/Training Consultant
Kingdom Drilling
Total Posts: 416
Join Date: 10/01/05
The facts presented indicate another failure event that invites learning where several things are likely wrong in your well.

Suggestion is the need to go back several steps and reveal through due and diligent latent cause analysis investigation or something similar?  all the physical people and paperwork evidence that all contributed to the end and consequential results.

We can assume this is an inclined well?  
Drilling Specialist/Well Engineer/Training Consultant
Kingdom Drilling
Total Posts: 416
Join Date: 10/01/05
The facts presented indicate another failure event that invites learning where several things are likely wrong in your well.

Suggestion is the need to go back several steps and reveal through due and diligent latent cause analysis investigation or something similar?  all the physical people and paperwork evidence that all contributed to the end and consequential results.

We can assume this is an inclined well?  
Drilling & Workover Liaisonman / Supervisor
Saudi Aramco
Total Posts: 7
Join Date: 15/11/18
Studer gave the correct answer. Backreaming at 120RPM is not a good drilling practice. Maximum backreaming RPM is 100, 80RPM is ideal.

The twist off was inevitable, firstly, the backreaming RPM that was high already was later increased to 160RPM.

Secondly, the TQ limit of 30kft-Ib already on the high side was also increased to 40kft-Ib. Even if the DP is 5-1/2" HT-55 that can withstand up to 49kft-Ib, such TQ limit is way too high for reaming down not to talk of back reaming.

What is the build/ turn rate of your well trajectory. Even if the build rate is less than 2deg/100ft, the turn rate might have made the hole tortious.

In conclusion, twisting off at the saver sub is an indication that your TQ limit is overkill.

General Manager
Total Posts: 20
Join Date: 27/02/16

Thanks for the added information.

How much of the 5000ft long 17 ½” OH was distributed between 20-40° / 40-65° inclination and above?

What was the BHA design (STABs amount and space out / DC size and amount), and how tortuous was the wellbore drilled (average dog legs through build & slant intervals)?

Indeed, one can understand the need to ream at 120rpm above 65° to manage / clean any cutting bed through a long slant 80° section, ERD type well design.

About Torsional inertia: several 17 ½” STABS and 9 ½” DCs/NMDCs/MWD size collars likely present in the BHA and these are heavy, when spinning them at 160rpm, they build up lots of centrifugal energy. Your say indicates string stall-out at 160rpm. So the energy built up in the high speed rotating string needs to dissipate all of a sudden, when stalling, explaining what you experienced: TDS SS backed off and string twist off.

Furthermore, think of the linear speed at the edge of the STAB blades, the side force and associated torque. Your DrillString above winds up, as a spring, storing further torsional energy (Stribeck type stick & slip vibration can develop along the string).

In terms of BHA design, for enhanced string integrity, your Jar OD size should not cross-over to a different tubular sizes or connection types. Ideally JAR & ACCELERATOR should form an homogeneous built, with same tubular size and connection, above, between and below these critical string components. This should be assured upfront with your down hole tool provider. You can cross-over a stand or more below JAR and Above ACCELERATOR, but not in between as this would likely be a weak link. Tensile strength and Make Up Torque design shall be checked through advance TnDnB analysis, involving stiff string style models, and investigating load cases such as back-reaming, accounting for wellbore tortuosity (effect of cutting bed would be more challenging).

Your case worth post analyzing thoroughly to identify limiters that put you in that situation: ROP vs hole cleaning, ie: Mud pumps capability? Bit-BHA design, i.e. drill string components compromises ?

Hope this helps

Well Engineering Team Leader
Daleel Petroleum
Total Posts: 2
Join Date: 05/07/18
How long did you circulate before start pulling out? did you analyze the torque and drag while pulling out? it seems that you are pulling with high cutting bed. you should stop run back 3/4 stands pump high weighted pill with max RPM & flow to make sure hole is well cleaned than continue POOH.
Directional Driller
Total Posts: 8
Join Date: 25/08/15
Kingsley - what happened to the standpipe pressure throughout, especially just before they first took losses? Was there any SPPA increase (or ECD data), suggesting solids pack-off downhole, thereby resulting in the total losses?

Even up at 1150gpm, in 17-1/2" Section the annular velocities (~100ft/min) are around 50-66% the nominal rate for efficient hole cleaning. Was there increased breakover torque or hookload observed coming out of slips on the previous few stands? Tripping broomstick plot?

It would be informative to know your BHA specifics and inclination where the jar twisted off e.g. 40-60deg avalanche zone? Given the trajectory profile to horizontal in that hole size, it is plausible that you have pulled the BHA into a cuttings bed and the turbulent flow around the larger OD collars was insufficient to propel the cuttings away and up the hole. Tight hole may have occurred at the BHA/SZRs but given the Jar (mandrel) weak point, the string has severed there up in the HWDP (~50k M/U TQ?).

Was the backreaming scheduled or did you try to pull on elevators first and fail? Once back-reaming has started it must be continued until low inclination zone. An On-Demand MWD frame (enabled after TD depth reach) which repeatedly sends up only ECD/Shock data at a high update rate is useful to monitor solids build-up whilst reaming out of hole.

Finally were they seeing much come across the shakers during the earlier backreaming? Easier to be a Monday Quarterback of course, but tripping in deeper 2-3 stands... away from the first tight hole/losses zone... may have allowed a clean-up cycle to remove some of the solids. Though if higher flow rates lead to unsustainable losses it may have been too little too late.

Torsional inertia -
Layman's terms - twist a beer bottle hard enough and hold the cap strong enough and eventually the top (or something weak) will come off :) Cheers
Total Posts: 4
Join Date: 08/09/15

Apt analysis.

The reaming rate was guided by hole cleaning procedures... it was needed to properly clean the hole - 17-1/2" hole section, 5000 ft long moving from 20 degrees to 82 degrees

If you can, please explain torsional inertia and how it affect design window of drill string (Tensile strength of drill string, Make up torque) Pipe at saver sub, 6-5/8 FH. Pipe twisted at bottom 5-1/2 FH. Pipe strength S135

General Manager
Total Posts: 20
Join Date: 27/02/16
Although there is a lack of context explained to give you some detailed answer, what you describe hints at rough drilling pratices:
- Backreaming at 120RPM: there is no need for that unless you were already in a desperate situation while POOH (tight hole)
- If not, BR at 120rpm is calling for troubles, damaging the wellbore walls, wiping out your mud cake you spent time to build-up, and triggering losses through losses zones you may spent significant curring.
- Then 160RPM reaming/back-reaming is just brutal, and likely increasing borehole destabilization, owing to drill string high vibrations, torque stalling (borehole destabilized) would then back off at TDS as a result of high tortional inertia, and twist-of your string down-hole for same reasons. 
Jump to top of the page