04-08-2017, 01:30 PM

04-08-2017, 03:27 PM

Hi Bas,

There are several problems with your model.

First the following Warning has been given: 'Iteration criterion not satisfied' meaning that there is no equilibrium yet. If you enlarge the number of iterations to be performed, you will see that equilibrium is reached after 93 iterations.

But the main cause of instability is the specification of free axial displacements in the joints of 500 mm, both for tension and compression. This is also physically impossible for 2 joints in a pipe section of about 1 m long!

If you change the 500 mm to 50 mm (more a real value), then you will receive reasonable results in 26 iterations.

There are several problems with your model.

First the following Warning has been given: 'Iteration criterion not satisfied' meaning that there is no equilibrium yet. If you enlarge the number of iterations to be performed, you will see that equilibrium is reached after 93 iterations.

But the main cause of instability is the specification of free axial displacements in the joints of 500 mm, both for tension and compression. This is also physically impossible for 2 joints in a pipe section of about 1 m long!

If you change the 500 mm to 50 mm (more a real value), then you will receive reasonable results in 26 iterations.

04-08-2017, 03:39 PM

Hello Rien,

Thanks for your answer. I'm aware of the big displacement I input there.

I also put a MaxRot of 90 degrees in the same input table.

The real numbers are about 14 mm inwards and outwards displacement.

And a maximum rotation of 3 degrees.

But with these numbers I got a maximum rotation of 3 degrees, with a STOP.

If I use, for example, a MaxRot of 10 degrees, I get a resulting rotation of 10 degrees, with a STOP.

If I use 20 degrees, I get a rotation of 20 degrees, with a STOP, etc..

All the way up to the (unrealistic) numbers in the model I attached.

This seems unexpected.

In this test-model I've only used sand, and there is not even SET-Z.

So you would expect minimal displacements of the pipe.

And so you would also expect minimal rotations in the joints.

Thanks for your answer. I'm aware of the big displacement I input there.

I also put a MaxRot of 90 degrees in the same input table.

The real numbers are about 14 mm inwards and outwards displacement.

And a maximum rotation of 3 degrees.

But with these numbers I got a maximum rotation of 3 degrees, with a STOP.

If I use, for example, a MaxRot of 10 degrees, I get a resulting rotation of 10 degrees, with a STOP.

If I use 20 degrees, I get a rotation of 20 degrees, with a STOP, etc..

All the way up to the (unrealistic) numbers in the model I attached.

This seems unexpected.

In this test-model I've only used sand, and there is not even SET-Z.

So you would expect minimal displacements of the pipe.

And so you would also expect minimal rotations in the joints.

07-08-2017, 12:01 PM

Good afternoon Bas,

The results may be unexpected for you, but are quite reliable.

Indeed, if you specify a maximum rotation of 3 degrees and for instance 50 mm free displacements inwards and outwards in the joints, the you obtain STOPS in joint el. 60 and 133. See attached view of the displaced pipe (displacement factor 10).

To explain the results you obtained I made a simple example in PLE.

Due to the internal pressure of 1.725 bar a thrust force ("spatkracht") of about 24 kN in the upper and lower bend is generated more or less perpendicular to the oblique pipeline section of about 1 m long. The horizontal and vertical soil springs are approximately 0.01 N/mm3.

So the simple example model is a 1 m long horizontal pipe with free ends and soil springs as mentioned. Same diameter and wall thickness as the "zinker". At the one end a force of 24 kN upward and at the other end 24 kN downward is applied.

The results are as follows: A displacement at the ends of about 500 mm (in opposite directions) and a rotation of 1.256 rad = 72 degrees.

So if the max. rotation is less, the a STOP results for tensile resistant joints.

The results may be unexpected for you, but are quite reliable.

Indeed, if you specify a maximum rotation of 3 degrees and for instance 50 mm free displacements inwards and outwards in the joints, the you obtain STOPS in joint el. 60 and 133. See attached view of the displaced pipe (displacement factor 10).

To explain the results you obtained I made a simple example in PLE.

Due to the internal pressure of 1.725 bar a thrust force ("spatkracht") of about 24 kN in the upper and lower bend is generated more or less perpendicular to the oblique pipeline section of about 1 m long. The horizontal and vertical soil springs are approximately 0.01 N/mm3.

So the simple example model is a 1 m long horizontal pipe with free ends and soil springs as mentioned. Same diameter and wall thickness as the "zinker". At the one end a force of 24 kN upward and at the other end 24 kN downward is applied.

The results are as follows: A displacement at the ends of about 500 mm (in opposite directions) and a rotation of 1.256 rad = 72 degrees.

So if the max. rotation is less, the a STOP results for tensile resistant joints.

14-08-2017, 08:16 AM

For ease of use, the Word documents contains the following picture (only):

[attachment=896]

[attachment=896]