Dear all,

In our company, stress analyis engineers have been taking Xmas Tree vertical displacements (for oil and gas wellheads in onshore installations) in doing analysis. The value taken is normally 50mm upwards. Due to this movement, we have to give spring support at very first support near the wellhead. What is the source of this 50mm displacement, nobody knows.

Recently a question was raised, why this vertical movement? What is the basis? How can we calculate the Wellhead Christmas Tree displacement? And should we consider any displacement at all?

One of my seniors who has very good experience in off-shore piping, says it is not required at all. Please give your opinion or share your experience with us on this.

The oil / gas well is not just a hole in the ground, it's a tubular, cemented into the strata way down below. The upward movement is due to thermal and pressure expansion of the well string. Dependent on the depth of the well, expansion can be around 300mm. Ignore this at your peril. Note that in certain circumstances there can be a net downward movement.

MoverZ,

Thanks for reply and noted that it should NOT be ignored!

But what exact displacement should I take? Should the source for this infomration be client only? Can I calculate this displacement myself? If yes, what is the method? Or should I give "SOME" value based on judgement? Unfortunately, on this we cant apply our "ENGINEERING JUDGEMENT"! Or can we?

Shahid,

You could estimate well expansion if you can get the actual or proposed well string data and cementing profile. Similar to a buried pipeline, you would be looking for a point of fixity, or a 'virtual anchor' location down the string, above which expansion will take place.

The best source of data is the Operator or their drilling / cementing contractor, who should supply the authorised information that you would use for design.

Dear Shahid Rafiq,

I ever involved in analysis line pipe from water producer well to filter. Christmas Tree is located in the same small platform with water producer well. Our scope was analyzing the modification of water producer line.

We decided the displacement of well head of water producer well = {Displacement measured of first support} - {displacement of modeled line pipe for the first support in OPE case (we got it very small)}. It convinced us that the displacement of first pipe support affected by well head displacement.

We did modification typically for 8 clusters. Surprisingly there were also some water producer well has first pipe support resting not lift off in actual condition! why be like this?

Then I asked my friend who worked in well service in division of cementing about the configuration of well. No data/literature for the calculation to have well head displacement we got. May be someone have it in this forum. But at least from the configuration these parameters will affect i think (opinion): friction between soil and cement, length of tubular (deep of drilling), temperature of fluid, and internal pressure (see the change direction on well head and imagine when the force by fluid when pass it).

I just want to compare the behavior of water producer well and Christmas Tree which is very near each other this way (picture attached).

Dear Shahid and other forum mates,

Its very interesting discussion and knowledge sharing being done on this topic. In fact, this is increasing our interest level in the field of stress.

Regards,

This is what I am hoping! Someone may throw some light on it and may guide me to some reference, book, method etc.

There is another point here. Which by the way, arises from your post that this displacement varies with each well. In some cases it may be ZERO . And as MoverZ said, 300mm in some extreme cases. Unique in all the cases!! So one may not be able to calculate it. Also, we may say that if the well is old (say 15 years as in my case), there may not be ANY displacement????

Seniors, please correct me if I am wrong!

Wellhead movement as discussed is related to cementing, reservoir depth, fluid temperature and pressure. I don't believe in can be accurately calculated, at least not during the detail design phase of a project. Most operators rely on similar experience elsewhere when coming up with a design value, and my experience is that the number is always more than reality when actually measured (mostly North sea). I don't believe the age of the well has any particular effect, unless the flow has reduced to practically nothing, and thus negligable thermal and pressure effects.

The well head displacement is normally available and this can be obtained from platform OIC. They record the movement and average movement is obtained. Don not assume as it. Yes, some of the platforms there will not be any movement, there maight be negligible horizontal movements depending upon how the conductor is choked.

PKU,

All manner of useful data can be obtained after the fact, and site evidence often proves up-front design data to be conservative. The original question was raised concerning DESIGN data and my responses were similarly phrased .... "at least not during the detail design phase of a project".

Well displacement data is needed to size loops and springs. If you expect the information to be obtained from a flowing well, you have created a chicken and egg situation.

MoverZ and PKU,

I thank you for replies. It is evident that it cant be calculated. Period. Yes, sometimes, client asks us to consider horizontal movement as well.

But your replies are for offshore(Many things apply to my case as well), the wellhead in question is onshore.

Greetings everyone.

As others have said, this is not a number you can readily guess or calculate. We often deal with onshore wells that have vertical design growth of 300mm or more, I think the largest wellhead displacement I have been asked to consider in my designs has been 500mm (20").

When I have had to design piping from wellheads where I am unsure of the actual wellhead vertical growth (usually the clients provide this information) I usually end up making a conservative estimate, and then state clearly on the drawing that the piping has been designed assuming a well head growth of Xmm, and that if the actual well head growth is larger than this to contact the engineer for a design check. In the end you are limited as to what you can do if you are not given the information required.

Dear Shahid Rafiq,

The title of our project that time is "water injection and producer wells reactivation". The activation done to water producer wells by installing new REDA pump in tubular in order to increase the pressure since its nature pressure already low after operated for 30 years. The variation of the displacement value of first support I mentioned above may be affected by the variation of flowrate and pressure.

Dear All,

Thanks for your help and guidance!

If I have to sum up what I learned from above discussion, it would be like this:

1) Wellhead displacement should NOT be ignored.
2) It cant be calculated and has to be based on judgement, prior experience with similar conditions.

Thanks again!

MoverZ
In such situation you may get the data of existing wells nearby and it is not difficult to find, also, if you use constant hanger, It will solve the problem. I have seen many flow lines in North sea, the assumed data from nearby platforms has worked.

PKU,

So you have moved away from .... displacement is normally available and can be obtained from platform OIC.

In thirty odd years I've also dealt with a large number of North Sea flowlines. I could quote many occasions where a 'nearby' field has entirely different well characteristics to a neighbour.

A constant hanger will most definitely NOT "solve the problem" if the flowline or a connector is over-strained by poor design. You are not in a position to make decisions about fundamental design criteria such as wellhead growth. It is a specialised area for experts.

Dear PKU,

BTW, how to get the data of existing well head displacement as reference mentioned by you? from last piping calculation which satisfy or measure it using such tools?

Dear all,

Regarding this discussion, I have recently analysed well head (christmas tree) lines considering 50mm displacement upwards as well as downwards. It eventually gave "for operating case with upward displacement, lift off in first support" and "for operating case with downward displacement, heavy load developed downwards in first support and failed with code compliance".

In this case, i suggested spring support in first support location and designed the system with a spring. now, the stresses, loads on supports & terminals, displacements all were satisfactory and i prepared the report.

But then, it was strictly told by the client that no spring support will be used in this project. so, i redesigned the system with no rest but guides only (i prepared support detail for first support location with a maintenance rest support which will be located with 60mm gap below the line. so that during downward displacement the line will not press down the rest support and create more load and stress). since the horizontal length of line was less than 6mtr, maximum sagging (displacemnt) was 6mm with no rest. so i decided this is ok. i found the loads also within limit. now i have prepared the report.

But, i still wanted to know from you experienced engineers if the design i have done with no rest but only guides arrangement (note that i have given maintenance rest support which is in no effect during operation and sustained after installation)is desirable when i cant go for spring support. Please advise me on this. I have to release the report only after your replies.

(note that the max stress ratio is 37% and that is in sus case. the stresses in other cases are low. the max deflection in sus case is 6.2mm and the max vertical load is 3.5KN on terminals. the line is 4" dia, A333 gr6, 43C operating temperature, 30mm insulation)

thank you all in advance.

It seems like you have a solution here. 50mm upward movement should not be a problem to deal with as you propose. Personally I would doubt the realism of 50mm downward movement, I cannot imagine how that could occur. However if you have incorporated it in your displacement stresses, you are on the right side of the line, safety wise.

Shan,

Check the Christmas tree well head movement in the design basis. Because i hope that +50mm is ok. But -50(downward)is not at all possible as there will be a support at the bottom of Christmas tree arrangement. I am attaching sample arrangement where you can see the arrangement.

If any one have better clarity on this issue , is most welcome.


regards

Sha.

Sha,

Trees do not normally have the support shown in your diagram. Commonly the tree is supported on the conductor (outer pipestring) flange and other casing hangers are supported by it, via the tree.

Downward movement can occur ... but it is rare and usually not of any great magnitude compared to the usual upward growth. Downward movement in my experience is limited to (water) injection wells, where the injected fluid is significantly lower temerature than the ambient strata temperature in which the well was initially drilled and cemented. This causes a thermal contraction of the wellstring. It may occur in dual producer / injection wells too.

Moverz & sha,

Thanks for your reply. I dnt have clear idea how there can be a downward displacement. if anyone have a clear idea of it, please clarify. I dnt think temperature can cuase this much displacement. because, in our case the water depth is around 40m only. so the displacement for a carbon steel pipe would be (40m*0.2mm/m) 8mm max. it cant be 50mm or 300mm as said in the previous posts. i feel the reason for this displacement is some thing else. i would appreciate if anyone knows the reason.

And i feel moverz is right regarding the downward displacement. in our wellhead drawing i couldnt find any steel support (refer the drawing below).

and also refer the support arrangement i have made. the attachment isometric is one of the small line from wellhead in the whole system. I will appreciate any further valuable discussion and your views regarding this issue.

thanks to all of you and keep it going with more exciting innovatine ideas......

keep the discussion. i have holidays for 2 days. i will be back after 2 days. cheers...

Dear Mr moverz,


Thank you very much for the reply.I understand from you that x-mas tree is supported on the conductor pipe.It means that even after the thermal growth, the tree will not move either upward or downward.Then how the interconnection between x mastree and the main pipe which has displacement(thermal growth)during operating ?Is there any special connection so that the displacement associated with x mas tree is accomodating the thermal growth.

regards

S.Rajinikanth

Many thanks to MoverZ in introducing water injection behavior here...

Sha,

The tree will move with thermal and pressure effects in the well.

You can visualise the mechanism as several pipes in parallel, all connected at one point, the tree, and each anchored by cementing at different levels down to the producing formation. The inner will be the smallest diameter, the longest and usually the highest temperature since it carries the well fluid. The outer, the conductor usually the lowest temperature and largest diameter. There may be 4 or 5 intermediate strings. Each is really a spring of differing stiffness, length and temperature. The net wellhead movement results from the solution to this stiffness puzzle.

Dear Mr Moverz,

Thank you very much for the valueable information.I understand the concept almost. I am trying to imagine how it is in reality.

regards

Sha

Dear Mr Moverz

If you have materials on conductor pipe and well head thermal growth mechanism, can you please send it.

regards
Sha.

hi moverz & sha,

In my last reply, i had mentioned that the sea water depth in our project is 40 meters and this will not cause such a huge thermal displacement.

i am sorry, I forgot that the oil rock will be far deeper from the sea bed. and the production pipe will be of 600 meters or more in length. now it sounds little bit clear to me. then, i collected some sketches which have given me more details.

In the attached sketch for well head arrangement, the conductor is of several pieces with varying dia & legth. these pieces are cemented with the sea bed wall. It shows how these pieces will act as a spring like moverz has told before.

And in the sketch no.2 of jacket assembly, it can be seen that the conductor will have only guides above the sea bed (there is no rest point with steel plate in the structure. the conductor & production pipe is rested with cementing in sea bed only). so the spring action (up & down)will be transfered to the whole assembly including christmas tree which will have further effect on connected pipe line too.

it can be seen clearly in the 3rd sketch for christmas tree. hope now it is more clear regarding how the displacement occurs.

Now, i need a help from you all. Can anyone clarify if a spring support at first support location is mandatory to absord this displacement?
or shall i make no rest support arrangement as the attached isometrics in my previous reply shows? (in this case, i have given no rest supports. i have taken both the connecting ends to well head and manifold as anchors. And a Cnode at the well head end anchor with displacement +-50mm. in the result of analysis, my sagging is less than 6mm and the terminal loads are within limit. in the isometrics you can see the support arrangement is with a gap below pipe. the support wnt rest during operation. only when dismantled during maintenance, it will be rested in the support).

have anyone done such a arrangement before? can anyone provide me a suggestion? i need you people's openion in this regard please.

Shan,

A spring support is not mandatory. Use a spring only if it is necessary. You have criteria for maximum allowed sustained stress in the pipe, maximum allowed moment on end connections and any other joints, and maximum allowed support spacing, limited by deflection, support load etc., whether that support is a solid or a spring. If you can satisfy all the criteria without a spring, don't use one.

In some circumstances you may be concerned to reduce stress due to cyclic loads by using a spring support, but this usually does not apply to a wellhead.

Moverz,

Thanks for your valuable reply. I have considered all the criterias as you said.

I have one problem here. the bending moments on the manifold nozzle (4", 900#)are nearly 4 times than that of allowable (say, allowble is 1.8 KNm and actual is 7.9 KNm). still in this case, the stresses, displacements and the terminal forces all are well within limit. Even the torsional moments are ok. but only the bending moments are high.
In my design, there is no moment acting at well head nozzle. the moments are zero at well head. but it acts more on manifold side.
(Please find the attachment for your reference)
I tried with possible routing change and support location change. But it had very little effect on bending moments. i tried to find any write up in our forums regarding this issue. how to reduce terminal moments... but i could not find one.
My doubt here is , considering all other criterias, can i allow this high bending moment? will it affect the nozzle after few cycles?

could you please refer me any material or previous forums on nozzle moments & loads? Thanks in advance moverz.

And i will appreciate very much if anyone else also takes interest and share your point of view and experience in this. i thank you all in advance.

Shan,

1.8kNm ? That's a small bending load, equating to about 19MPa in 4" Sch 160 pipe. What exactly is limiting your bending to this value ?

Assuming the direction ofthe moment shows it is due to well movement and not deadweight, to reduce it will mean adding some pipe flexibility in an appropriate plane. In your case I would add pipe in the Z direction

You should discuss this with your lead engineer or client ... you might get your imposed moment accepted.

Shan,

It is not possible to have zero moments at a wellhead connection. Either you have modelled this nozzle as an X,Y,Z force restraint only (should be an anchor), or as a nozzle with displacements, but left the rotational displacements free (these should be zero).
I agree with MoverZ that 1.8 kNm is a small moment for this size and pressure rating of pipe, and the manifold should be well capable of handling this moment.

Thank you Moverz. you are right that for a 4" 160sch pipe, 1.8KNM bending load is equalant to 19MPa. Moreover, my pipe is 4" XXS sch. so it should be nearly 16MPa. very less.

But, the problem is 1.8 KNm is the allowble given by design basis for pressure vessels (commonly given by client for all pressure vessels). But, in my system, the actual bending moment i am getting is 7.9 KNM . that is nearly 70 MPa. I couldn't say if it is very high for a 4" 900# nozzle.

And moreover, I dnt have idea whether for a manifold i should consider strictly the allowables given for pressure vessels.

Can i consider this 7.9 KNm bending moment as acceptible? in your experience, how do you consider for a manifold nozzle? please spare your valuable time and clarify me moverz. thank you so much.



And Mike, thanks for your interest. I have considered the wellhead nozzle as an anchor point only. But i have given +/- 50mm displacement with a Cnode at anchor point. hope it clarifies you now how zero moments occurs at well head nozzle. please do share with me your further view also.

I will appreciate if anyone else would like to advice me on this. thanks to all.

Shan,

Which code is your manifold designed to ?

If it's ASME VIII or similar it's a vessel. If it's B31.3 or similar it's a pipe system. If it's a pipe, 70MPa is a perfectly reasonable figure for pipe, but you should check out any related flange stresses.

If the limit of 1.8 kNm refers to a vessel, speak to the vessel designer. It must be very heavy wall and in that case, 1.8 or even 8 kNm is nothing. Get the vessesl guys to check out local stresses with 8 kNm.

Thank you very much Moverz.
The manifold in our project is a piping system designed to 31.3. As you said, I have sent these forces & moments to the manifold manufacturer too. I appreciate your timely help and thanks alot moverz.

hi gurus and masters,

i have same exact case like these...client are not yet providing the actual vertical growth on the xmas tree but my seniors told me to consider 3ft?

this is insane right... making a big loop around the xmas tree and use of constant spring on the first support doesn't make any sense at all,still failing on code compliance.

what is the most realistic thermal growth you can suggest to me so i may not over design the system, as i read here 300mm and 500mm only..

please advice asap.

thanks.

Figure out what the rationalization of 3 ft. If the 3 ft is measured, use the 3 ft. If it's a wild guess, then offer options.

"I can give you x displacement with the simplest design."

"I can get you this much more if I add the spring."

"I can get you this much more with a loop."

"I can get you this much more with a loop and a spring."

"Do you want me to keep going? Others typically don't see this much displacement."

We deal with steam injection wellheads that can easily have about 3 feet of vertical growth. What we do is use three swivel joints in an A-frame piping arrangement. Very large amounts of growth can then be accomodated depending on the pipe lengths used between the swivel joints.