No need to convert to "E".  (Is that Young's Modulus you're talking about ?)

What kind of pipe material ?  If you know that you can obtain good tables of minimum and maximum cover from the pipe maufacturer which are usually based on full scale testing.

Do you plan to use the native material as backfill?  Unless you can tolerate a lot of long term settlement it doesn't sound like very good material to use and will be difficult to compact.

Don't forget to add live loads if the pipeline is in an area subject to wheel loads.  It will see some wheel loads during construction in any event.

Sorry, but I disagree with you entirely.

I have spent the last 8 years on a committee producing a national standard for the design and installation of buried flexible pipelines. The approach used in design isnt just opening a manufacturer's brochure and reading a table. There is agreat deal of science involved in designing a combined soil/pipe structure.

One has to consider the relationship of the modulus of commpacted fill to that of the native soil, consider the depth of the water table, vacuum and pressure conditions, surge, pipe material, trench dimensions and loads. Loading may include live, dead or impact loads as well as the load from the trench backfill. The specified compaction needs to be taken into account along with the confidence level arising from the degree of inspection and QA.

Once armed with this data one can model the trench to determine levels of stress, strain, deflection, resistance to combined loading and most important resistance to buckling.

The only bit of data I am missing is the soil modulus. Not being a geotechnicval engineer or soils scientist I dont know how to convert the information I have been given to the information I need. Hence the question.

I have received some advice from a geotechnical engineer that the shear stregth given is representative of a soft clay and the use of E=3MPa would be acceptable.

http://www.usace.army.mil/inet/usace-docs/eng-manuals/em1110-1-1904/

This link will probably tell you how to "estimate" the undrained Young's modulus (E.u) from the undrained shear strength (c.u). Check out Appendix D if you need a quick estimate. Is the soil saturated? Do you need drained Young's modulus (E.d) or undrained Young's modulus in your analyses? You can relate them thru the formula of shear modulus, because the undrained and drained shear moduli are the same (G.u= G.d).

stanier:

Use of an E/c ratio of 50 to 100 should get you in the ballpark; but that still leaves a big concern:

You are "fine tuning a coarse tuning knob."

Before I continue, let me first say that the effort to develop such a procedure is generally commendable, and I am a firm believer in the use of soil/structure interaction analyses as a part of the design process.  Having said that, however...


I'm cynical about any new design method that uses soil modulus as a design input.  The behavior of backfill depends on so many different factors - many of which cannot be known in advance and/or cannot be controlled in the field - that I have to wonder about the wisdom of developing a standard that relies on the Young's modulus, E_Y, or shear modulus, G, of "engineered" backfill as an important design factor.  Hell, many (most?) geotechnical engineers don't have a good handle on the range in elastic modulus of the predominant soil / rock in their area!  (Which is probably why your geotechnical consultant gave you the answer that brought you to this forum...)

If the procedure that you are using is sensitive to a doubling - or halving - the modulus, then it's a waste of time.  Pipelines are linear projects that often extend for tens - even hundreds - of miles; most of the experience with elastic behavior of soil is limited to a localized condition (i.e. one building) that is no more than 300 feet square.  The experience extrapolates rather poorly to your type of problem, and gives the impression that one can determine "the" E_Y for soil along a given alignment.  You also have two rather large variables: the modulus of the soil around the trench as well as the trench backfill itself.  Obtaining a reasonable in situ measure (not estimate) of both of these soil moduli will require equipment and procedures that are relatively rare in the geotechnical engineering business: pressuremeter and/or dilatometer.  You can achieve similar results with cyclically loaded consolidated-drained triaxial compression tests - which are also expensive, but a little easier to find.  And all this for a product that (in much of the country) seldom has a thorough geotechnical study done.  Unless there is a failure and lawyers are involved.

[Focht3 now descends from his soapbox]

I hope that your method really works - but be damn careful!

A question:
Your previous message indicated that you have been serving at least eight years on a committee developing this procedure; what are the names of the civil engineers (particularly geotechnical engineers) on the committee?  [In my opinion, at least one-third of the committee should have included geotechnical engineers.  After all, you can specify pipe parameters within fairly narrow tolerances, but soils - well, they don't work that way!]

Hi Focht3

Thanks very much for your insight into this phenomena.

The standard committee memebers are not geotechnical engineers or soils scientist however a number come from water authorities that access to such specialists. The economic reality these days is that standards committes tend to be stacked with vested interests so the pipe material manufacturers are there to make sure competing products dont get an edge in the standard.

The values used for modulus in the standard are very conservative and thus cater for the variation in measured or perceived soil modulus. In practice the ratio of native to embedment modulus is used as in the Spangler formula in the AWWA standards.

The soil pipe design has to be conservative particularly when considering buckling when the pipe has exposure to live and dead loads.

I dont see the general piping industry moving away from cone penetrometers, plates and othe crude measurements that use empirical relationships to come up with a modulus. Thus I see that we will continue to build pipelines that never fail but are grossly over designed.

This doesnt happen in mechanical engineering.

I'm not surprised that no geotechnical engineers were a part of the process; in fact, I would have been surprised if there had been even a single participant.

I understand your comment about continuing "to build pipelines that never fail but are grossly over designed" - but please recognize that someone, somewhere will find a way to exceed the boundaries of your design envelope, and a pipe will fail.

It's sad, really - wrapping a new design method in 'sophistication' by using 'advanced' parameters, but still having the same old "garbage in - garbage out."

Perhaps what you are really looking for is a design parameter known as "modulus of soil reaction". There is an excellant paper on the subject on the net.

http://www.ceworld.org/ceworld/Presentations/TransportationInfrastructure/JeyapalanWatkins.cfm

Anyone who designs buried flexible pipe should take a look at this paper as it presents a very good background on the subject.

Nope - k (modulus of soil reaction) and E_Y (Young's modulus) are quite different.  stanier was very specific in his use of terminology.  Besides, k has no meaning in treating backfill and native soil simultaneously...at least in a trench context.

I thought the discussion was on designing a buried pipeline using the Spangler (or modified Spangler) equation. According to the authors of the above mentioned reference the modified Spangler equation contains a term symbolized by E' (that's E prime) which is in fact the Modulus of soil reaction, and not Young's modulus.

steve1,

If you are talking about Spangler's theory of earth loads, E is the modulus of elasticity of the soil.  As Focht3 mentioned this is different from the modulus of soil/subgrade reaction.  I must agree with focht3 again that the main issue here is that E of soil is difficult to define and values can vary significantly... and your design is only as accurate as your input parameters.  

regards