Wick drains 1

[jdonville]

We are working on a new roadway that will have several new spans over existing roads and watercourses. Several of the approach embankments (on the order of 20 to 30 feet high) will be founded on compressible soils that are expected to undergo considerable settlement (on the order to 6 to 12 inches, depending on the structure).

Wick drains have been recommended in the geotech reports for locations where 1) waiting for 90% consolidation is expected to have significant impact on the construction time, or 2) staged construction is anticipated due to stability concerns.

I have downloaded and perused FHWA RD-86-168 "Prefabricated Vertical Drains" Volume 1: Engineering Guidelines (the other 2 volumes are not available for download). Based on the guidelines presented and the available lab data, we can probably get the spacings worked out and conservatively estimate the pore pressure relief as a function of time.

However, there does not seem to be much guidance available on the extent of the drains required below the embankment (i.e., how far back from the abutments and how far from the centerline). I would appreciate your thoughts or references to standard practice in any jurisdiction.

Regards,

Jeff


Jeffrey T. Donville, PE
TTL Associates, Inc.

http://www.ttlassoc.com

The views or opinions expressed by me are my own and do not necessarily reflect the views or opinions of my employer.

 

[rchanke]

I'd try talking to the design engineers at Nilex Corporation.

 

[BigH]

Jeff - the extent of the wick drains, both longitudinally and laterally, would be dependent on a number of things, as I see it:
1. The need for increasing the Su value in order to ensure stability - in conjunction with the stage loading. Close spaced wicks - faster achievement of the 90% of so normally used before next stage allowed.
2. The criteria of allowable embankment settlement that will be permitted - not only of the magnitude but also with respect to distance from abutment. It is most important that settlements directly behind the abutment be minimized to prevent the all-knowing 'bump' that occurs with an yielding embankment and unyielding abutment. As you get further from the abutment, more settlement would be tolerable and still maintain the basic road crossfalls/vertical grades. In this vein, you might consider using 1.2 or 1.5m spacings directly behind the abutment, but further away - say 100m or so, you would start to space them out to 1.8 to 2.2 or so - this would build out settlement but at a slower pace (a transition zone) - to match better that part of the embankment where you don't use such ground improvement.
3. Use of wicks (or other vertical drains) is also good if given time for you to preload (surcharge) directly behind the abutments to offset pavement structure loading. Otherwise it is a cat chasing its tail situation.
3. Lateral extent of the wick drains will depend on the thickness of the compressible layering. See Poulos and Davis Elastic Solutions for a handle on the induced vertical stresses due to an embankment. You may find, if your soils are weak enough, that you may need to put in stabilizing berms not for the purpose of counter weight but for the purpose of increasing very low undrained shear strengths.
4. The depth of the installation needs also to be considered. We found on our subcontinent job that the equipment we had could only install the wicks to where the Std Penetration Test N value was in the order of 7 to 8. If greater than 8, the equipment could not advance the wicks any further. This was fine in our case as the particular stratum when this occurred was a suitable stratum "as was" with respect to stability. Our wicks were for the purpose of stability, not of settlment, per se.

On my subcontinent job - the main concern that became cognizant once construction started was stability. We had 9 to 11m high retaining walls and similar height embankments scheduled to be built on 6 to 8m of very soft (Su in the order of 20 kPa) clay overlyling firm (Su in the order of 35 kPa) clay. Stability computations indicated that only 4 to 5 m of the RE Wall and about 5 to 6m of embankment could be built while maintaining non-seismic stability of unity. We therefore used wick drains for those areas where the height of the wall was greater than 4 m. If less, wick drains were not used. We used 1.5m spacings and 90% consolidation was achieved in the order of 40 to 50 days. We ended up with upwards to 800mm of settlement (30 inches) of settlement - as alluded to in other posts, we didn't have the time or mandate to use surcharge - yes, we had some bumps at the end - regulating asphalt layers. We found, too, that the settlements were not too bad up to a height of fill of about 3m - so that concern of mismatching settlements was not all that bad - of course, in all the cases, regulating layers of asphalt will be needed periodically to maintain the RFLs.

 

[BigHarvey]

Jeff

Normally wick drains are installed only directly under the embankment. The main reason being that nobody cares generally how much time the adjacent ground will take to settle.

 

[Panars]

My masters thesis project involved a test embankment on 20 feet of water treatment sludge. One half of the test embankment had wick drains installed underneath it. The lateral extent of the wick drains was 20 feet; the distance was equal to the depth of the soft deposit. STS consultants designed the test embankment, so I don't know how they arrived at that distance. Just something for you to consider.

 

[BigH]

BigHarvey - I agree with your comments with respect to settlement, but disagree if you need to improve the strength of a significant zone outside the embankment for stability. Depending on geometry and stratigraphy, you may find it critical to improve the strength of the original soils out beyond your embankment. As you observe the critical slip surfaces nearly 50% of the surface is beyond the toe of the embankment. I point out, again, that if you need to improve the strength outside the embankment proper, the wick drains won't help much unless you surcharge the area with a stabilizing berm - provide a positive loading (or use of a vacuum system). On our project, we went some 15 ft (4.5m) beyond the RE Wall and/or embankment (6 to 8 m very soft deposit). In some cases I was able to provide a bit of weight for the outside wicks; in other cases I wasn't successful.

 

[jdonville]

All,

Thanks for your input. Many of the considerations here become clear when the embankment-induced stress distribution and/or stability-critical geometry is considered.

Jeff


Jeffrey T. Donville, PE
TTL Associates, Inc.

http://www.ttlassoc.com

The views or opinions expressed by me are my own and do not necessarily reflect the views or opinions of my employer.

 

[BigHarvey]

BigH

I totally agree with you, if there is no surcharge then no need for drains. That's why I wrote I would limit the drains to the extent of the embankment. I would also like to point out that drains are great to deal with settlements, but i believe it's not the best method to increase strength.

 

[rochplayer]

Another consideration in where to terminate the wickdrains laterally along the embankment is differential consolidation rates creating a 'bump' between the un-wick drained area and the areas with wick drains. At a bridge approach were settlement magnitude may be most critical, use a denser spacing, with increasing spacing back from the abutment until you get to a point where the longer term settlement can be accommodated. If there is a bridge approach slab, I generally use the tightest spacing from the abutment back to just behind the approach slab, then start tapering off from there.

Another issue with wick drains is that they can increase the amount of secondary consolidation that occurs. Basically, they allow dissipation of pore pressures so quickly, that some of the rearrangement of the soil skeleton that would normally occur during primary consolidation does not have a chance to occur. To address this, wick drains work best when coupled with an engineered surcharge program that is designed to take out primary + secondary consolidation of the unsurcharged embankment well before t90.