Restrained and Unrestrained Section of buired pipe

[GPB005]

ASME B31.8 has clear definition provided in Para 833.1 as
1. Piping in which soil or supports prevent axial displacement of flexure at bend in "Restrained" which can have straight section of buried piping or bends /pipe in stiff soil.
2. Piping that is freed to displace axially or flex at bends is "Unrestrained" which may be bends in soft or unconsolidated soil.

My question is 'most of the engineers call entire buried section of the Pipeline as restrained in the Software. I differ this approach. One should see the buried pipe displacement and only those sections where axial movements are found zero should be called as Restrained and rest other (especially buried bends, Above ground below ground Transition area) should be called unrestrained as they move axially.
Is this a correct understanding?
This has a considerable inpact on the buried lines having temperature above 80 C as thermal stresses are added in the Restrained section.

 

[DSB123]

Your understanding seems to be correct

 

[LittleInch]

It is essentially correct, but in reality these are two extremes of state - fully restrained and unrestrained. Buried pipe can exist in these states but also in intermediate conditions, which is where you need to know what impact this designation has on "the Software" you mention. what "software"?

There is also the difficulty in determining whether the soil is "soft" or "stiff" as 31.8 gives no numbers or guidance.

Most stress software simply requires knowledge about whether the pipeline is buried or not and it works out whether the pipe actually moves or not or by how much.

The restrained / unrestrained section in 31.8 simply allow you to do some simple calculations rather than a full stress analysis, but can't determine e.g. bending forces at your bends as you can't easily work out how much force / axial movement you get in a buried situation. Above ground anchored pipes and supports are easier, but adding in support friction can often significantly affect calculated movements and stresses.

in the intermediate zone there are thermal stresses, but maybe not quite as high as the fully restrained sections, but there might be other forces due to that movement.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[SNORGY]

GPB005

Are you (they) using software that allows you (they) to manually create or input a soil model or models that can be used to simulate various soils properties along specific sections of the pipeline?

I agree with you.

For what it's worth, my experience has been that at 80 C (depending on the installation temperature at time of initial restraint) it is challenging to get a stress model to "pass". My rule of thumb has been to take the SMYS of the pipe in MPa, divide by 6, and see if that number is less than the change in temperature (installation to hot) in degrees C. If it is, then getting a model to pass becomes rather challenging because the combined longitudinal stresses computed in accordance with the Code begin to govern over the considerations of hoop stress in the determination of minimum wall thickness, and sometimes just throwing more metal at it (increasing wall thickness) doesn't help you all that much.

 

[GPB005]

Thanks all for input which gave me confidence in my understanding...
The software I am using is CAESAR II. It allows me to crate own soil stiffness. I am using American Lifeline Aliance method to generate soil spring stiffness. The thumb rule of SMYS / 6 seems to working ...
The pipeline thickness is just sufficient for manual combined stres calculation (Hoop, Bending, Thermal and longitudinal due to pressure). But actual model shows failure in straight section of pipe (bends are safe) and only increase in thickness and reducing difference in installation and design temperature is making model SAFE. I tride changing bend radii.
Do you think there is any other way passing to reduce stresses.....

 

[LittleInch]

It's odd that the manual way is giving different results than the analysis. Suggests something is wrong / different in the inputs - check all the data in - it's easy to make an error in either.

Normally you need to either increase the laid temperature, decrease the actual temperature experienced at that point, strength of pipe or for the locked in section accept that it is overstress to 31.8, but do a local buckling check to see if this stress is real or not.

hot lines above 80C are an issue when designing to 31.8 and I agree with SNORGY on this. Not come across the SMYS/6 ROT before though - I'll have to file that one away for future use.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[SNORGY]

LittleInch,

I derived that one by algebraically combining and rearranging the equations in CSA Z662 Paragraphs 4.6.5 thru 4.7.2 assuming a sour gas pipeline in a Class Location 1.

 

[BigInch]

There are partially restrained segments of pipeline; the points that move less than the calculated expansion distance for any given temperature change, however analyzing one and the other, fully restrained and unrestrained, is the conservative approach.


you must get smarter than the software you're using.