Trench Settlement

[soilsman]

One year ago, a new sewer line was installed down the middle of an existing street. The total length of the trench is about 2 miles. The width of the trench was about 10 feet. The backfill material consisted of gravelly soils and compacted to about 90 percent according to the inspector. However, there was no full-time inspection during compaction. The top of the pipe is about 25 feet below street grade. Now the trench has settled in some areas as much as 12 inches. Has anyone come across this problem before, if you have, what were your recommendations for the repair of the trench settlement.

 

[dicksewerrat]

I installed miles of pipe in Mpls with little or no settlement of the trench. Compaction in the trench should be 90% Mod. Proctor to 5 feet from top. Tested every day. 95-100% from 5 ft. to grade. Tested every day. Test results brought back to job site by me every day. Just having the testing lab on site adds 3-6%to compaction effort. My guess is they pushed in the first 10 plus feet and gave the rest a lick and a promise. Of course this takes longer but the street stays where you put it. Inspector showed up for an hour once in awhile.

Richard A. Cornelius, P.E.

http://WWW.amlinereast.com

 

[moe333]

If the sewer was bedded in open graded gravel, there is a potential for the backfill to migrate into the voids in the gravel bedding, resulting in surface settlement. The potential is higher if there is groundwater to carry the finer backfill materials into the gravel. A layer of filter fabric over the gravel surface can reduce this potential.

 

[emmgjld]

You say gravelly soils. Could you be a little more descriptive, with attention to the fine fraction? Also, 90% of which proctor?

Your situation sounds very much like a few problems I have investigated. The typical scene was a moderate to deep (10' to 20') backfill was placed, compacted and tested. Ground water was very close to the bottom of the trenches and surface waters were usually not well controlled. Within a few months, Settlement was about 5% to 8% the total fill depth.

After the initial finger pointing and gnashing of teeth,there was very little doubt that the fill was placed to the original project requirements, 90% of Proctor. The contractor complained the soils were wet and slightly pumping. The majority of the test results were slightly dry of 'Optimum Moisture and Density was usually less than 94%. Due to observations of others (not the contractor or the testers), it is not believed the contractor was severely 'cutting corners' or the testing was compromised.

However the job compaction requirements were were not well considered. The 90% referred to the Std Proctor, ASTM D-698. The soils were 'Gravelly' but had clayey and silty fines. The fines are often sandy CL-ML (My practice has been blessed with an exceptional amount of CL-ML in the area). When the Modified Proctor Test ASTM D-1557 was performed, the Optimum Density increased 7 to 11 lbs/ft3 and the Optimum Moisture dropped 4% to 6%.

When the repairs were made, the contractor favored placing the fill slightly wet of the 'Optimum Moisture' and the compaction test results were mostly in the 94% to 96% range, without a lot of effort. The soils did not have a tendency to 'slightly pump' during compaction. The Dry Unit Weight of the fills increased about 6% to 12%. Fills are good after 2 to 4 years.

I know that at least one of my cases is a rather exceptional soil for my area, a gain of 11 pcf and drop of 6% moisture when going from a D-698 to a D-1557.

Just a suggestion.

 

[fndn]

Unless there was a backhoe with a long reaching bucket equipped with a vibrating plate or you had a long ramp where a vibratory roller descended, you could pretty much expect the bottom 10 to 15 feet were loosely placed.

 

[DRC1]

Even wh well compacted backfill some settlement can be expected. If the trench was 25 feet deep and they drove sheets, expect some settlement on removal of the sheets. when we do trench work in the street, we typically patch to grade, then mill and pave the following season. This allows the trnch to settle over the winter. 12 inches is a lot though. I would patch the road one more time and wait untill the end of the season to be sure it is okay, then mill and pave

 

[civilperson]

Pressure grout installed at a 4 foot grid in the top of the trench backfill width is one method of repair. Another is to remove pavement and compact the trench with ponding or puddling as well as mechanical compaction. The grout is very expensive, up to $190.00/cu.yd. plus associated traffic control and patching costs.

 

[BigH]

I'm not a fan of puddling - there are "times" it will work but . . . If the soil is anything but a fine to fine to medium sand and clean, I doubt puddling would ever be a proper answer - especially gravelly soils. Also, you are relying on 'seepage' pressures to rearrange the soil particles and unless you have a head and some positive quick drainage, again, . . .

 

[Tippys]

Doesn't seem too many have addressed the repair. You didn't say if the pavement is PCC or asphalt but I am assuming asphalt. At this point there is not much you can do about correcting the source of the settlement - just hope that it is complete. Remove the pavement surface in the settled area and mill the remainder as required for an overlay. Depending on the pavement design, raise the grade in the settled area with fill, aggregate or bituminous base and place your thickness design for this road. Apply an overlay over the whole thing and hope it doesn't settle again. Since it settled a foot already I doubt a lot remains if any effort was made to compact it. There are many ways to compact the existing trench but none I know of are practical. Better to just deal with it if more settlement occurs. A very good way to avoid the settlement is to use a flowable fill for the trench, or at least for the top few feet - if flowable fill is available in your area for a reasonable price. If you have groundwater within several fee of the surface don't use a clean granular fill - it tends to collect water and fill like a bathtub. Your pavement will alligator quickly and the project will cost more to get the premium fill. This does not apply of course if water can get out of the "tub".

 

[mitchelon]

12 inches seems like a lot. I would check for possible infiltration (cracked pipe?). Just an idea.

 

[apmssoriano01]

We have experienced similar problems with poorly backfilled sanitary sewer trenches. The repair practice to-date has consisted of repeatedly leveling the settled areas with compacted asphalt materials (i.e. - cold patch, PermaPatch, hotmix, etc.). Patch thicknesses are typically 1- to 2-inches, but have been performed 5 to 7 times. Simple math shows that are settlements are similar to yours, but over a period of approximately 3- to 5-years. Probably like you, we do not have the funding to remove the poorly compacted backfill and replace. Therefore, we are planning to bridge the trench settlement with a stabilized base material.

The repair alternative that we are incorporating starting in 2oo7 is to reclaim the asphalt pavement section to a depth of approximately 12-inches and stabilizing the resulting mixture with approximately 6-percent (by weight) Portland Cement. After shaping and compacting the mixture, we will allow the material to cure (i.e. - hydrate) for 1- to 3-days, depending on the ambient and roadway temperatures, and then overlay with 2- to 3-inches of hotmix asphalt. Hope this helps.

 

[dirtsqueezer]

There's another problem with the practicality of soils testing in a trench. OSHA safety standards specify that none may enter a trench greater than 4'(I believe that is the figure). Therefore, trenches over 4' are not tested, at least by me. That leaves your 25' trench 84% untested.

I have seen this recently on a project with the same order of depth. The contractor pretty much took advantage of this, filling the trench in one big lift about halfway, then to 4', then 2' then subgrade. For some reason, with the proctor we were using, this seemed to work- this is a new company for me, and, well, it wasn't my proctor. The godsend was it was just temporary fill; it was to be reexcavated to install a local sewer line about 5' above the installed pipe. Didn't make me feel that much better about it. -not a very salient entry; sufficed to say contractors can take advantage if you aren't able to test at depth.

 

[PEinc]

If there was no full-time inspection, then most of the backfill was probably done with excessively thick lifts that were either not compacted or improperly compacted.

Who paid for the inspector? What was the inspector's instruction or requirement for being on site? Can the inspector be held responsible for letting the contractor backfill improperly?

You can dig a test hole or do a test boring to check the backfill density. Or, you can dig up the trench and backfill it again with proper compaction (not likely to happen). Or, you can compaction grout the backfill material above the pipe. Or, you can keep repairing the pavement for several years. The question is, Who pays for the repair or maintenance?

 

[BigH]

dirtsqueezer - You are partially right on the 4 ft (some jurisdictions use 5 ft) - this is for straight sided trenches - but depending on the class of soil(s), you can have compound excavations (4 ft vertical at bottom and 1:1, say above. Some depth and other restrictions apply. Or, you can shore - or use a trench box dropped in so you can go down and test, etc.