http://www.americanlifelinesalliance.org/Products_new3.htm#BuriedPipe

Also, look the IBC or UBC for seismic loads

Thank you very much for your reply!

I am a new person to learn how to design piping system. I read the piping handbook, but it seems not specific enough for a new person to start over to learn it. Would you please provide any source, from where I can learn how to design piping system by myself?

Thank you very much and Have a nice weekend!

supprise to know that you are an electrical expert, but answer mechanical question :)!

http://www.kineticsnoise.com/seismicbook/index.asp

Also has some very good information.  

I used to be intimately involved in pipe systems design at a previous job.

The Alyeska (Alaskan Oil) pipeline was larger (48") but designed for seismic conditions.  It had both above ground and buried portions.
You would be wise to study how this line was designed.

http://www.alyeska-pipe.com/pipelinefacts.html
click on "Pipeline Design"

If nothing else it is interesting reading.

Have you identified active fault crossings?

BigInch-born in the trenches.
http://virtualpipeline.spaces.msn.com

That is definately getting into some touchy areas.  I've worked quite a few pipelines with seismic considerations, as has been pointed out, probably the most famous is TAPS.  But there are plenty of others.  There are many considerations, not the least of which is will any of your pieline be in frozen ground?  You need to get an expert in this.

For some basic guidance, pipeline fault crossings generally utilize three basic fault-crossing trench configurations.  This will depend on what type of stress the pipeline will be in - strike/slip, compression or tension.  A depth of soil cover not exceeding one meter is generally required for all special fault-crossing trench configurations.  The 3 configurations used to mitigate the 3 types of stresses are: 1) sand shading, trapezoidal trench, and geotextile lining.  The angle of the pipeline with regards to the fault is also critical as is the length the pipeline traverses the fault, sometimes a long length is best, sometimes short, depends.

Liquefaction may also be a consideration and that adds yet another layer of complications.  I know in many of the pipeplines I've worked on, a higher yield pipe is used for the fault crossings.

The special construction measures generally must extend over the length of pipeline that might possibly experience transverse (horizontal or vertical) movement.  The width of crossing, Wc, on each side of the pipe must be included in the length of special construction measures, because the surface expression of any future fault movements could occur anywhere within the crossing width.  A depth of cover not exceeding one meter is normally specified for each fault-crossing zone and whatever depth is calculated, it should be maintained within reasonable tolerance over a length of crossing.  Small variations in cover over short distances will not be detrimental, but significant or frequent deviations should be avoided.  

There is allot that goes into the design, you must look at the seismicity of the pipeline route, check local tables and contour maps for peak acceleration, short period spectral accelerations, and l.0-second period spectral accelerations for various applicable event return periods.

If frozen soil is a possibility, then the performance of a buried pipeline at fault crossings depends on maintaining a relatively low-strength soil medium around the pipeline in the zone of soil failure.  What happens is during wintertime temperature extremes at the higher elevations, the soil will freeze from the surface downward, sometimes to a depth that will serve as a nearly rigid encasement above and to the sides of the pipe.  If the pipeline is unable to produce local soil failure in response to fault displacements, pipeline structural integrity may be compromised for even relatively small displacements.  It is critical to limit the depth of soil freezing within the fault crossing sections.

In summary, get an expert.  I've worked quite a few of these and have some general knowledge, but wouldn't even think about tackling it.

Greg Lamberson
Consultant - Upstream Energy
Website: www.oil-gas-consulting.com

At the Denali fault, engineers designed the pipeline to move with the earth. It was laid close to the ground on a gravel berm and is supported by shoes that slide on beams. That construction allows movement of 20 feet horizontally and 5 feet vertically. The shaking broke five aboveground cross beams that support the pipeline and at least two vertical support members. Nine anchors that restrict the pipeline's horizontal movement were tripped, and a honeycomb of insulating material was crushed in several places, but the pipeline was not dented.

Here's some more details,

View USGS' Denali Earthquake reconnaissance here,

http://walrus.wr.usgs.gov/geotech/denaliposter/intro.html

I've uploaded these to rapidshare, download them (free) here,

Download the Performance of TAPS to Denali quake,
http://rapidshare.com/files/14666408/Denali_Performance.pdf.html

Download the TAPS Response to Denali quake,
http://rapidshare.com/files/14665941/Denali_Response.pdf.html

BigInch-born in the trenches.
http://virtualpipeline.spaces.msn.com

In addition to horizontal and vertical (downward movement and upward breakout), you also need to consider longitudinal strains.

For longitudinal forces, if you have experience in driving piles, it's very similar.  Longitudinal restraint represents the pull-out resistance along the pipeline and is very simlar to skin friction for piles.  The load-deformation relationships developed for load transfer at pile/soil is very similar for buried pipelines.  

Once allowable strain limits are developed, a finite element analysis of the pipeliensection should be done to determine the strain critiria.  That will drive in part your selection of pipe grade.  One recent project we had to do wdie plate testing to establishe critical weld defect size with project-specific materials and welding methods.

There are a couple other references that might be handy:

ASCE (AMERICAN SOCIETY OF CIVIL ENGINEERS), 1984.  Guidelines for the Seismic Design of Oil and Gas Pipeline Systems, Technical Council on Lifeline Earthquake Engineering, Committee on Gas and Liquid Fuel Lifelines, New York, 473 p.

HANSEN, J.B., 1961.  “The Ultimate Resistance of Rigid Piles against Transversal Forces,” Bulletin 12, Danish Geotechnical Institute, Copenhagen, Denmark.

HONEGGER, D.G., 1999.  “Field Measurement of Axial Soil Friction on Buried Pipelines,” Proceedings of the 5th National Conference on Lifeline Earthquake Engineering, American Society of Civil Engineers, August.

Greg Lamberson
Consultant - Upstream Energy
Website: www.oil-gas-consulting.com

I been reading about the TAPS and seen some techical papers on the EQ design.

One paper refered to EQ piping design softwear developed for the project. Does anyone know if this is aviable for others to use?

Thanks

Gentlemen,

The first step is to determine what regulatory design document (Code) you are required to design the piping system to.  The next step is to READ that Code.

A good seismic design document (by George Antiki) can be found at:

http://www.americanlifelinesalliance.org/pdf/Seismic_Design_and_Retrofit_of_Piping_Systems.pdf

Regards, John

thanks very much for so many people's reply! Really appreciate that a lot! It seems not an easy topic to do, right? Especially for a new engineer to do this!:) I am not sure what is "fault crossing"?

My design is within a dam. There is gallery within the dam concrete. The pipe is within the gallery to let water pass througth the dam from upstream side to downstream side. The total length of the pipeline is about 6 m. there is a control valve to maintain the water volume per hour. The valve is heavy due to the pressure and temperature sensor within the valve. I am doing the piping support strong enougth for seismic effect.

So the working condition is environmental temperature, not freezing condition I think, the pressure will be the maximum reservoir water level. That is about 20 m height of water.

If further help, I will appreciate a lot!

cheers!

hello2006

Sorry, a fault crossing is where there is a discontu=inuity in the rock caused by tensional forces (normal fault) opr by compressional forces (reverse fault).  It's basically the point where the ground is pushing or pulling against itself and eventually something has to give.

If you do not have fault crossings, and are in a seismic area, you have faults nearby and liquefaction may be something you need to look at.

Liquefaction of granular soils or sediments can be a major seismic threat to pipelines.  Liquefaction does not occur randomly in natural deposits but rather in a narrow range of geologic and soil environments, river channel, flood plain, deltas to name a few.  Sediments most susceptible to liquefaction are granular soils that remain loose and uncemented after deposition.  Liquefaction ocurs only in saturated sedements (sedimants in shallow ground water table).  

You'll need ot check topo maps of the reas for evidence that ground water levels are near or within a few meters of ground surface along the pipeline route.  If the groundwater levels are sufficently high, and the area has seismic activity, it may allow liquefaction to occur.
  
Obviously proximity to seismic sources will influence the likelihood of liquefaction.  If the seismic hazard maps indicate there are trending faults capable of generating suffecient magnitude earthquakes an the pipeline is within close enough proximity, then you may have to design for these levels of seismicity.

I would recommend having a geotech survey done.

Greg Lamberson
Consultant - Upstream Energy
Website: www.oil-gas-consulting.com

You had us thinking it was a pipeline, but 6 meters is a bit short for that.

Your problem will be more of a structural design nature and you should post to a structural engineering forum.

BigInch-born in the trenches.
http://virtualpipeline.spaces.msn.com

I missed that, I read it as 6 km!!

Greg Lamberson
Consultant - Upstream Energy
Website: www.oil-gas-consulting.com

Yes, the pipe is only about 6 m long. Originally I think that is a mechanical design task. Now I am not sure I should ask more help from civil people or not. I will think and talk to civil people about liquefaction problem to see if that is the case for my project.


Appreciate everybody!

Shouldn't be a problem if you're building a dam there.  I hope I hope I hope!

BigInch-born in the trenches.
http://virtualpipeline.spaces.msn.com

Hello, Everybody,

I have a question what is "The horizontal PGA with a mean AEF of 1/1,000 was 0.44g and that with a mean AEF of 1/10,000 was 0.88g"

I do not understand what is "AEF"?

Thanks for any reply!