A bit of advice please

[DekDee]

I have spent quite a while trying to get my head around this but still struggling so will try posting and see if LI, BI or one of the other knowledgable pipeline engineers can help.
Please correct me if I am wrong but my understanding is the reason buried pipelines have lesser wall thickness and a less stringent acceptance criteria for welds is because the pipeline is covered with dirt so it has a natural shock absorber !
What happens if the pipeline is above-ground - does it require wall thicknesses similar to the refinery piping it will connect to ?
What then is the acceptance criteria for welds on an above ground pipeline - ASME B31.8/ API 1104 or ASME B31.3 NFS ?

The point I am trying to make is why if you have a pipeline supplying a refinery with a line break/spec break just outside the fence do you have two different acceptance criterias for the welds either side of the line break/spec break valve ?
As soon as the valve is opened the welds are subjected to the same medium, same pressure, same temp etc so why different acceptance criterias ?
Cheers,
Shane

 

[zdas04]

What I call the "risk density" (i.e., the combination of risk and the proximity of one component that contributes to the risk profile to another component that contributes to the risk profile) is very large inside of a facility, and relatively small outside of the facility. Inside a facility you have a more or less fixed surrounding-population at risk of being impacted by a failure. Outside the facility, the line is in proximity to a varying population density. Therefore ASME B31.8 has an emphasis on population density ("Location Class" in B31.8 vernacular) and the occurrence of risk increaseers (e.g., railroad crossings, river crossings, etc.) and ASME B31.3 has a emphasis on component interaction and sympathetic failures. Should that change in risk-drivers happen at the plant fence? It has to happen somewhere, the plant fence is as good a place as any since it is generally some distance from the plant fence to the first risk-increasing component within a facility.

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[brimmer]

It's more so due to population, proximity of population and consequences of a failure as already well pointed out above. Remember you don't need to conform only to minimum code requirements, if you think there is a risk outside a fence the same as inside a fence, do a study to determine what you would consider a safe distance and put more stringent standards in that area... Nothing stopping you from increasing design/acceptance criteria above minimum code requirements in a specific area.

 

[zdas04]

Also there is nothing to keep you from defining the spec break some arbitrary distance outside the fence, you are the engineer after all.

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[DekDee]

Thanks guys - appreciate the responses.
Cheers,
Shane

 

[LittleInch]

my understanding is the reason buried pipelines have lesser wall thickness and a less stringent acceptance criteria for welds is because the pipeline is covered with dirt so it has a natural shock absorber !

Hmmmm,

There are a significant number of differences in the design approach between pipelines and refinery piping and the fact that pipelines are normally buried is one of them, but not, IMHO, anything to do with it being a shock absorber.

The key issues for me are:
Pipelines are normally buried and are normally straight, with no other forces other than internal pressure or thermal stress. They are fully uniformly supported and restrained in all directions. Hence the only stress they normally need to be concerned with is hoop and axial stress.

Piping on the other hand is predominantly above ground, supported at discrete points, is subject to bending stresses from the many branches, elbows and bits of equipment hanging off it like valves, instruments etc.

Pipeline design recognizes that the line pipe is a very significant part of the material budget and hence optimising and reducing that element is good value. Refinery piping on the other hand is a minor component of costs compared to valves, flanges, tees, vessels, steel work etc and hence there is little value in optimizing the pipe thickness.

The pipeline design codes utilise SMYS as the only consideration in pipe design, wheareas piping design uses both SMYS and UTS to create a stress value to use in B 31.3. This penalizes use of higher strength steel as the UTS does not increase at the same rate as SMYS hence there is little benefit (~10%) coming from an increase in SMYS of 90% (Grade B to X 65). The result is that no one uses higher strength steels in piping design, but this is very common on pipeline design.

I'm no expert in the different welding acceptance criteria, but I've not heard previously that pipeline acceptance criteria is significantly lower than piping, but I'll stand to be corrected on that.

In terms of a code break, having an artificial break at or near a fence is a complete nonsense. The pipeline design should, as a minimum, follow through to the end of the pipeline up to and including a pig trap, within a facility. I've seen breaks at fencelines and they cause nothing but grief for very little benefit. Piping codes don't handle buried pipe very well at all and there is a potential for greatly increased wall thickness and reduction in ID.

Pipelines above ground normally have a slightly reduced design factor, but can be designed and analysed for stress using point support loads and hopefully 3 or 5D bends and not elbows.

So my point is that the pipeline should be designed to pipeline codes which are well established and optimal all the way to the end of the pipeline, whether this is in it's own fence or within a refinery. That might help.

Makes sense?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[DekDee]

Thanks LI for the detailed response.
As to the acceptance criterias that is where I was bemused (see below).
I have been a pipewelder and then an inspector on both pipelines and refinery piping and it always had me puzzled.
Pipeline comes out of the ground at 45 degree angle with an overbend on it.
Weldneck flange is welded on to the end of the line and a valve is installed.
Refinery piping (generally B31.3 Normal Fluid Service) is then connected to the pipeline.
API 1104/ B31.8 allows incomplete fusion up to 25 mm long
API 1104/ B31.8 allows incomplete fusion due to cold lap up to 50 mm long
B31.3 NFS does not allow any lack of fusion.
So, one weld either side of a valve with totally different acceptance criterias (I have just used Incomplete Fusion as an example) and yet when the valve is opened those welds are exposed to the same medium, same pressure, same temperature. ???
Regards,
Shane

 

[LittleInch]

Yup, that's what you get sometimes. The pipe strength though in the pipeline is probably twice what it is for the refinery piping.

Comparing welding inspection codes would be interesting, but outside my area of interest - I just specify it and let the inspectors get on with it....

For me a pipeline comes out of the ground and continues some distance into a pig trap or similar then reverts to B 31.3.

Piping is generally designed to be a little more robust and able to cope with people hitting it, crawling on it, vibrating it etc. Pipelines are a bit more delicate.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.