Tailings pipeline laing on the ground 1

[MISA]

Our client want from us to analise tailings line CS(A 106 Gr.B) line 10" OD 1/2" rubber lined using Caesar II. Pipeline is 9 km long , heat traced and insulated in HDPE jacket. 3/4 Pipeline laying on the graveled ground. Pressure is 1,000 psi. Is anybody expirienced modeling line laying on the ground (constantly supported)? Any guidelince or advice?
Tx in advance.

 

[Thomasjl]

It looks like a very expensive line. How do you maintain the rubber lining at the (welded??) joints? What is the ground look like, sand, gravel, turf, soil??? What are tyhe min and max design temperatures??
What is the required life span? Supporting a heat traced and HDPE outsode line is very costly too. Your solution is unorthodox, but if the ground is a rather soft it might be an excellent cheap solution. Why not.

Good luck, Thomas.

 

[schwabe]

Hi MISA,
Sorry in advance; this is going to be a bit lengthy.

Let me assume that you have been through the discussions of how to protect the insulation and jacket from the graveled ground as the pipeline moves, and that you have dismissed the shoe & discrete support option.
Obviously, you need to consider friction between gravel and jacket. The pipe wants to elongate under temperature increase and pressure, and the friction forces (force per length) want to prevent that. As you have a "sandwich" system with three interfaces ( gravel-jacket; jacket-insulation; insulation-pipe), you need to establish the weakest link first to check whether the friction forces can make it all the way to the pipe, or whether failure (=slippage) will occur at any one interface.
Assuming that the interfaces are strong enough to transfer the friction forces and that the accumulated friction forces (force per length x length of straight line) fully restrain the pipe, you can now check whether the pipe wall thickness is sufficient to take the combined longitudinal plus hoop stress. B31.11, para 1119.6.4(b) describes the relevant equation. This is a hand calculation. Adjust the pipe wall if required.
Now you may start the CAESAR model. You only need to model the pipe bends to the virtual anchor lengths on each side, because the fully restraint straight sections were qualified earlier by the hand calculation.
I would not use the underground modeler because it requires the input of biaxial spring values for horizontal restraints, which in your case represent static and dynamic friction between gravel and jacket, and where will you get that from? Rather, model the pipe as being supported at short intervals (say, every 5m) and add the friction coefficent between jacket and gravel. Of course, you may ignore the sustained stress calculation because there are no bending moments due to weight.
Check the stress at the bends in accordance with B31.11, 1119.6.4(c) and that's it. You will likely find that:
- you need 5D bends rather than elbows to reduce the SIFs,
- the displacements at the bends are large
- bends with angles below 60 deg are overstressed and require anchors

Check the settings in CAESAR to ensure suitability for pipeline analysis (Bourdon effect ON, hoop stress calculated for OD, etc.). Do not use the "corroded wall thickness fields but just run new pipe wall and corroded thickness in separate models because there is a bug in CAESAR in the stress calculations.

It should take you about 100 hours to do the job, including report. Happy analyzing!

 

[richay]

Schwabe, you stated "Do not use the "corroded wall thickness fields but just run new pipe wall and corroded thickness in separate models because there is a bug in CAESAR in the stress calculations".

Could you elaborate on this, as I am unaware of any errors in CAESAR II with respect to corrosion.

Richard Ay
COADE, Inc.

 

[schwabe]

Richard,
Let's take the CAESAR discussion offline. I'll contact you at COADE next week when I get a minute.

Cheers,

 

[richay]

Schwabe, see 1119.7.3.c.

Richard Ay
COADE, Inc.

 

[smckennz]

Hi MISA
you may want to talk to your principals, as what they are proposing will be a pig to install and service.
Our normal practice is to install steel tails line on concrete sleepers with steel hoops; 2 per 12m length (450NB, 250 psi). To assemble flanged joints on direct laid steel pipe can be quite difficult as nothing ever lines up and dirt gets in the way even if the ground under the flanges is scooped out. The chances of achieving a line load along the bottom of the pipe are zero because the ground cannot be graded with sufficient precision, and even if you could, settlement would prevent it in the long term. With concrete sleepers the contact points are clearly identified, although again, settlement ensures contact at only some sleepers.
Direct ground laying is OK with HDPE pipe because the pipe is flexible and the pipe can be lifted easily at connections to make and break joints. HDPE is normally snaked on the ground to permit movement, with some form of side restraint to stop it going for a walk.
I think your client may be thinking of how HDPE pipe is installed and assuming this practice is also suitable for steel pipe.
You can do it I wouldnt advise it.

Cheers

Steve