Minimum Temperature for Underground Pipeline

Where does the -5ºC ambient soil temperature criterion come from?

If the soil ambient is really that low, there may be a period during start-up before the pumped oil is hot enough to guarantee +8ºC pipe temperature.

Depends on where you are in the world.

Probably best to assume realistic worst case values (-5 deg C in this case). I am not an expert in this field however so there may be a way around that due to a number of reasons I don't have enough knowledge on to say (such as temperature transition changes from pipe empty to pipe running hot crude raising the pipe and surrounding ground temperature up and re-leaving the temperature differential stress).

Where is the pipe located and what's the expected ground temperature?

Thanks for your advice. Pipeline is installed in western Canada and construction will start in August. Installation temp at restrained soil is 8 deg. Up to 5 ft depth from ground -5 degC has been measured during winter according to some statistics.

Thanks

Western Canada is cold, in which case I would say you would have to go with the -5 deg C value for ground temperature (being the realistic worst case for installation and non-running pipe temperatures in service), but if the customer only gave you an installation temperature figure (8 deg C) where did the ground temperature come from?

Others here may have a better answer and reason, I really am not sure if installation pipe stress matters in the case of a crude oil pipe if the fluid is almost always flowing and hence almost all of the time the pipe will be at a higher temperature.

"To evaluate the thermal conditions of a underground heated pipeline, the first thing to do is to calculate unstable temperature field around the pipeline. The variations in temperatures of static oil and soil were investigated during pipeline shutdown in both summer and winter, in which some important parameters of the soil and crude oils of a Northeast & Northwest pipeline are employed. Usually, source - convergence method is used for this purpose. The method regards oil pipeline's heat transfer in common transmission as a stable process, therefore it has usage limit. According to the heat transfer character of usual buried heated pipeline, the heat transfer course of it is described with two dimensions unstable heat transfer equations, The effect of ground surface temperature change and pipe diameter on pipe temperature drop is considered in boundary conditions, and mathematical model of soil temperature field is established, the finite difference theory is used for calculation. A mixed mesh method is presented, that is, rectangular mesh is used in the inner field, and normal line mesh is used near the pipe surface. Though the method is approximate one, if the mesh is very smaller the solution accurate enough still can be gotten. A software in VC visual studio has been programmed to input necessary data, finish iteration and draw equal temperature lines around the pipe on different time. The establishment of this model has laid the scientific base for later research on breaking down and restarting of pipeline."

Do you have a link you can provide as the source of that quote? Please.

I think that -5 °C surroundings temperature could cause a tensile rupture of the pipeline when not under flow, as the contents may also cool to that temperature, if the cold snap is extended (as in Western Canada).

If the pipeline is good only to 8 °C, OP had better get some more expensive liability insurance, and also set aside a fund from which future penalty payments may be extracted by the Canadian government.

Or OP has the option of going back to the contracted company, stating the discovered obvious problem, and requesting a material good to below the coldest temperature at depth on record. Otherwise - somebody will be in for a world of hurt at a later date.

Western Canada is a big area. Here is a link to some data for the frost line.

http://www.urecon.com/applications/municipal_ambient_below.html

In the coastal and southern parts of western Canada the 1.5 meters puts the pipe line below the freeze level.

Like some of the others where the -5 degC comes from.

Also 1.5 meters is closer to six feet not five. That extra foot can make a difference.

1.5 meters is less than 5 feet.

It sounds like you are setting design criteria for the installation.

The first question is whether you are designing to "never" fail, fail after so may years, or fail regularly because the consequences and recovery are acceptable.

The overstressed condition you describe needs to relate to an expected failure rate and failure mode. For some work I have done, overstressed was defined as 2/3 of yield strength or 1/3 of ultimate failure strength. In that case, exceeding 2/3 yield may not reach the fatigue endurance limit for the material and it could be expected to "never" fail. In other cases, the fatigue limit would apply and you could expect the assembly to fail after so may years and so many thermal cycles, but have that failure anticipated to be well past the lifetime of the system. In other cases you might design the system to act as a mechanical fuse and break before something worse happens, like the brass shear pin on the output shaft of an outboard motor.

Being an enlightened sensitive type, I would suppose that you "never" want to fail and that when you say overstressed, you mean stress at 3/3 of ultimate failure strength, so on a -5C day, the pipe will break. If so, you will want to set an ambient design temp somewhere around -10 or -15C to get to the statistical 99.9999% not getting that cold confidence level and design for those conditions. Once upon a time I certified in an HVAC design system and was surprised to note that the planned design criteria made the system fully adequate only 50% of the time. I suppose sweating or shivering grumpy office proles was not considered to be an unnecessarily adverse operating condition and that the .00001% probability of freezing or broiling an office prole to death was met.

Probability only that high? They would die of a paper cut first.

The .000001% is a disposal issue which may require a permit and an OSHA investigation. Otherwise, no impact.

I am curious as to what pipe materials and what piping codes require any allowable stress reduction at -5ºC (+23ºF).

Last I knew, ASME pressure vessel code assumes full strength from -20ºF to +650ºF; and even US Coast Guard allows full strength down to 0ºF.

It appears that the issue is not a material strength reduction, but rather a stress level change when a boundary condition changes, similar to the growth or contraction of railroad tracks with ambient temperature changes. The pipe acts like a beam with built in ends and thermal changes applied.

You get the Gold Star on that answer!

I know that the Trans-Alaska Pipeline was built above ground for a reason - heaving, yes, but also they were able to incorporate expansion loops (more like zig-zags), and insulated and lagged the pipe for its entire length. That gives better insulation value than the soil, the pipeline is 100% available at all times, with few exceptions.

Any leak on above ground pipeline is clearly detectable, and easily diagnosed.

It may be that the part of Western Canada in question does not have Muskeg soil, and permafrost as does Alaska.

Well, Ahem, Er, Ah, mostly above ground. I stand corrected by Mr. Whitephone, who is apparently standing upon his box of Tide this morning. Good moaning, Mr. Whitephone, how are you today!

"Any leak on above ground pipeline is clearly detectable, and easily diagnosed."

As well as repaired.

See? We can agree on some things. Casper for Congress 2018!!!

Congress???

Is that a compliment or a dig?

Sorry about that! I still cling to the antiquated notion that being a member of Congress is an honor, a job not taken lightly, and not a way to line one's pockets on the blood, sweat, and tears of the electorate.

Here you go: This new sensor network can sense the strain in a pipeline.

sensor for strain during seismic events

Even though designed for seismic events, the sensor could also be employed in other ways, so I believe. If the pipeline is getting too cold, maybe the operator could arrange to increase flow of hot oil down the line.