Hello,

Can anybody address the gap between real situation and analysis results?

I have Desander connected to a wellhead. Desander consists of 12” pipe vessel and 3” nozzle at the end with two saddles anchored to a skid (it is a kind of anchor to anchor design; unbelievable!). Its design temp is 400 deg F (204 deg C). This is vendor’s design responsibility.

There is a 3m long 2” XS piping straightly running between the wellhead and Desander. Design /Install temp is 38 and -20 deg C, respectively. This is anchor to anchor; one anchor is wellhead and the other is nearest anchored saddle. I have the max allowable bending moment at the wellhead given by vendors. Piping should be straight run between them.

It looks simple but the anchor to anchor layout bothers me. Of course thermal force between two anchors causes a way higher bending moment at wellhead- Failure. Consequently, everybody is surprised and says “If the analysis is right, the past projects more than 3000 job should be failure” Main factor seems to be at design/install temp but I don’t want to play with the figures.

Any idea? Or any special consideration I would miss?

Many thanks in advance.

At 3m and a delta T of 58 degrees your pipe is growing around 2mm. Caesar applies a restraint stiffness of 10^12. In the real world nothing is this stiff. As well there is likely some safety factor in the maximum moment the well head can take. So the difference between the real world results of successful operation and the failure you are seeing in your model is likely a combination of conservative modeling and using up the safety factor in the real world.

If you can show a system operating in the real world you could apply 319.4.1 a) (assuming B31.3 is your code) Nor formal analysis is required if the piping "duplicates, or replaces without significant change, a system operating with a successful service record" The difficulty in using this clause is to get this documented. Ideally you would have the client sign off on something saying we have this arrangement at this site and has been operating without incident for x years. Getting someone to sign off on this and people saying "well its worked before" is a big step.

How have you modeled the well head. Are you just anchoring it in space, How are you accounting for wellhead growth? If you are just anchoring it in space and not modeling the well head that may give you overly conservative results

I don’t believe that somebody volunteerly is responsible for the past performance and I have not applied the clause 319.4.1(a) to a new system. Yes, it is big step.

Regarding the last one – wellhead growth.I don't quite understand it. Elevation change between the well head bottom at ground and nozzle is about 1.5m and there is little temp change up to the wing valves (almost amb temp.) The concern is bending moment at bottom anchor of wellhead. I believe that the result with anchor in space is not conservative. 1500# Christmas tree with several valves is quite stiff ( even less than 10^12 though). I got axial force and calc’d moment at the bottom.

Many thanks,

I work on mainly SAGD style projects so the steam and production wells are at high temperatures up to 300 C. The well head will grow up out of the ground from the thermal expansion from where the virtual anchor in the ground exists. These values usually have come from the client based on their passed experience.

Hi everybody,

Is there any one who can help on where to get the external loads on x-mas tree nozzles if vendor data is not available?
It's clearly stated in the API 6A section 4.2.1.3 that the standard does not cover it.

many thanks in advance.

B-Madi,

You have to contact the tree supplier to get the allowable load if it is critical.

When I got the loadings after bugging some people, I was surprised with the lower values than I expected; specifically the allowable bending moment is not that high.

Thanks,

NewPiper,

Could you please post the allowables you got from your vendor along with the size of flow line...

Thankyou in advance

Sorry for late response.

4" 600#
"Under 50,000 lbf of compression, a 7000 ft*lb bending moment is allowed"

thanks,