Sinkholes

[muuddfun]

I am looking for some ideas on what may be causing manholes and sewer mains to sink. I was recently called to check out the repair operation of a manhole and replacement of the sewer line where the line had been videoed and the pipe was so badly seperated near the manhole that the video camera could not pass the seperation. The sewer line is a 27 inch main. The depth of the sewer line is about 22 feet. The soils are clay (CH) and also some clayey silts / silty clays ML/CL that fall along the A line. The clays were fairly soft. Groundwater is about 6 to 10 feet below grade. There are occasional lenses of sand, though in the area of the sewer it appered to be mostly the clay that was present. The sediments are lake deposits. The clays are generally slightly overconsolidated. The sewer pipe is clay pipe encased in about 12 inches of unreinforced concrete. There was no other bedding material below the pipe. The sewer system is old probobly installed between 1900 and 1930. The area is subject to many earthquakes and strong ground shaking. This particular manhole was not in a high traffic area.

I am thinking about liquefaction, bearing failures of the clay, or infiltration of the fines into the line and errosion of soils surrounding and supporting the line. What else should i consider?

 

[civilperson]

Underground caves, karst formations, water induced settlement of sands and gravels, clay layer consolidation.

 

[muuddfun]

Thank Civilperson, the sediments are very deep, there is no Karst or caves in this area, so that eliminates those causes, but consolidation may be possible, but why such an abrupt differential?

 

[epongra2]

Any construction activities in the area, i.e. pile driving, basement or underground parking excavations, dewatering wells, etc.?

 

[civilperson]

Alluvium clay soils and loessal,(wind deposited), soil are liable to collapse or consolidate large percentages. Increase in moisture content for loess, and surcharge and subsequent loss of pore pressure in alluvium can cause multiple feet subsidence. I would guess ex-filtration as a primary source of water. Piping is a failure mechanism for embankments where small particles are carried away by water flow, (unlikely without a void nearby).

 

[dgillette]

Interesting forensic problem, but since it's a 27-inch sewer main, they'll never give you the budget to investigate it enough to satisfy your curiousity.

Are the manhole and pipe actually settling downward relative to a known datum? They probably aren't significantly heavier than the soil they "replace," or maybe not even as heavy, which makes me think it may not be a clay consolidation problem, particularly since you mention the clay being lightly OC. If not for the concrete encasement, the pipe would be a lot lighter than the soil it replaces, even when full. Any chance it's a flotation problem? Do you know the geometry of the separation (pipe higher than manhole, pipe lower than manhole, pipe pulled away from manhole approx horizontally)?

The title is "Sinkholes," but you didn't describe them. Where, when, and what did they look like? If material was lost into the sewer, that might create voids around the pipe.

Water wells near by?

Any idea on when the separation occurred?

 

[rconner]

Perhaps elaborating a little more on e.g. basically what epongra2 has mentioned , though I don't know if this possibility in the present case I believe deep surface or underground mining, and/or movement of groundwater associated with mining has been a factor in some past significant movements of soils and/or structures.

 

[muuddfun]

I got some more information from the client for the sewer problem. The sinkhole was noted at the surface around the manhole, distress in the pavement, etc., and it had sunk about a foot. I don't know exactly when it was discovered but it must have been within the last few months. When they examined the pipe insitu they found the crown was crushed, so it looks like it was an infiltration problem. They did not have any information about what differential movement if any occured with the manhole / pipe etc. This is not the only place they have detected problems of this sort within the City. They are finding several other places with problems as they investigate. But this one is one of the most immediate.

What has me puzzled now is why they would get a break in the pipe that was encased in about 12 inches of concrete. It was heavy clay pipe, and should be able to withstand significant loading. They also told me that this section currently under repair, is a bypass put in place to fix an adjacent line that had previously failed as well. They did not have any info on when that occured but suspect it was a long time ago, more than 50 years.

I wander if maybe the concrete encased pipe and manholes were such a rigid system that during earthquakes when the waves are passing by it was enough to break the pipe from relative motions of the soft ground around the stiff sewer system? Maybe pore pressures built up enough to float portions of the pipe as well? Maybe your on to something Dgillette. Maybe the pipe and concrete encasement cracked enought to allow infitration of the fine soil to occur and we ended up with a piping problem? What are your guys thought on any of these theories? I am getting some pictures of the repair operation and will try to post some of them when they come in. Thanks guys for your help.

 

[rconner]

For whatever reason the latest description "crown crushed" sounds quite different than the original "badly separated" (maybe we should wait to get some pictures to see what the damage really consisted of). It does now sound like the subsidence could indeed be from infiltration of soils into the sewer near the broken or gapped manhole location (maybe odds are external causes wouldn't necessarily selectively seek out the manhole location?) In any case I agree a brittle, rigid pipe (in turn cemented/joined rigidly to a large rigid structure like a manhole?) might arguably not be the best strucural situation when things get to moving and shaking in earthquakes!

 

[muuddfun]

Here are some pictures.

 

[muuddfun]

http://files.engineering.com/getfil..._repair_27_inch_sewer_line__Aug._2008_023.jpg

http://files.engineering.com/getfil..._repair_27_inch_sewer_line__Aug._2008_019.jpg

http://files.engineering.com/getfil..._repair_27_inch_sewer_line__Aug._2008_026.jpg

 

[muuddfun]

http://files.engineering.com/getfil...6-46b8-a806-68cf17b2fd52&file=Picture_170.jpg

http://files.engineering.com/getfil...a-4972-95f5-c7cfd1f1bc4a&file=Picture_172.jpg

http://files.engineering.com/getfil...c-49f5-a52f-4e75573a952e&file=Picture_174.jpg

http://files.engineering.com/getfil...4-4928-a256-6b41312d1e01&file=Picture_175.jpg

 

[muuddfun]

http://files.engineering.com/getfil...c-4765-beba-ccd56bd06563&file=Picture_208.jpg

http://files.engineering.com/getfil...e-418d-86f7-04a537bef981&file=Picture_246.jpg

http://files.engineering.com/getfil...d-4d51-8b2f-fc05b1d82c2a&file=Picture_257.jpg

 

[howardoark]

I wouldn't think it was an earthquake. It might be that the pavement opened and allowed water to get into the backfill which then settled. At one point, there was a 22-foot deep trench there that they probably backfilled by end dumping soil (given the era it was constructed) into the trench. That would explain why the pipe crushed - it was suddenly subjected to the full prism load of the fill. Of course, most of the trench was already saturated, so it was only the top 6 to 10 feet of fill tht would have been impacted.

It looks like you're in California, it could also be that the current drought caused the fat clays to shrink - but you'd think that would have impacted the whole street.

 

[msucog]

i've dealt with a few problems around storm pipe manholes so maybe my thoughts will help. first of all, quite often the structure itself is founded on really firm material but then the pipe and everything above has more fill associated with it. and being in deep holes, it doesn't always get compacted as good as it should. so the manhole structure itself doesn't settle but the surrounding backfill does. secondly, the sinkholes i saw were in structures from 10-30 feet deep. we determined that there was likely a small area right around the manhole that the contractor didn't compact as well as everything else (i'd say less than about a foot or so all the away around). for my case, since it was at the bottom of the slope catching water in a big parking lot, the water that made it in to the base course stone built up on the side of the structure until it finally ran along the face of the structure downward. small cracks around pipe penetration grouting and structure lift hole grouting allowed the water to penetrate in to the structure. the soil began piping material in to the structure and after 5 years or so, it had slowly eaten away from the bottom up and expanded outward. most sinkholes i saw with not more than say 3-5 feet around the structure. one of the sinkholes was maybe 50' across and swallowed a car but that was probably due to the massive amount of water hitting the structure from all sides at the bottom of a steep slope...the connection grouting had actually completely cracked, broken out and pushed in to the manhole.

i am in a low seismic area so that can be ruled out. it was just the slow patience of water over a long enough period of time.

 

[aeoliantexan]

There was a similar problem in Texas where the sewer was installed in sandy alluvium along a river. A few years later the river was dammed to create a shallow lake, and the water table rose to well above the sewer line. Manholes sank 1 to 3 feet. I believe that water infiltated the connections between the pipe and the manhole and brought in sand from beneath the manhole. The process is slow, but eventually creates a void under the manhole, which then sinks.

The CL-ML and sand lenses you describe are apparently erodible. Look at the slurry flowing in the excavation. You said the sewer is at 22 feet and the water table at 6-10 feet. That is a lot of differential head to drive infiltration, boiling and piping. The bottom of the excavation in your photos is heaving and unstable. It's tough to build any thing well under those conditions. Probably wasn't much better 50 years ago, so it's not surprising that there was not a water tight seal at the pipe-manhole connection.

You need a water tight installation, and you will probably need to dewater the soil to stabilize the excavations so the contractor can get high quality construction. The soils are now disturbed, and replacement construction can be expected to settle some, so you need flexible connections at the manholes and gasketed pipe joints to tolerate movement without creating leakage. Geotextile around the pipe and manholes to filter any moving soil can serve as a back-up measure.

You need a really good dewatering contractor and good sewer contractor. Good luck.

 

[LCruiser]

Was the pipe repaired before? It appears the encasement may have been a repair. The pipe may be old, but when was the previous repair?

The patches in the street indicates this is not the first time this has happened - is this an ongoing thing? Has there been recent construction downstream? Has the groundwater table been lowered locally for some other construction recently? Has the grade been raised there recently and the manhole extended?

Interesting situation.

 

[rconner]

muudfun, with regard to the observation you provided in your 3rd post, "they found the crown was crushed", recently (when I was, shall we say, "indisposed")I took a small short cardboard "roll" that was handy and inserted it up a short distance into stiffer short stiffer pipe nipple that ahd an I.D. slightly larger than the O.D. of the cardboard roll. I then bent the length of the short encased "paper pipe" downward relative to the larger pipie, then removed it from the encasing pipe end to observe the damage. I observed the upper part of the paper roll end was crushed or buckled inward, whereas the bottom of the paper roll end was basically undamaged. I believe I demonstrated exactly the following:

1. The top end portion of at least a small paper roll inside a stiffer, larger encasement will be damaged inward when there is sufficient bending (negative bending moment) of an initially very close-fit (not allowing rotation or joint deflection) connection.

While I'm not sure of any relevance to whatever you are dealing with, this may kind of beg the question, was the crown crush by any chance in the basic location where "the pipe that was encased in about 12 inches of concrete", and/or are you by any chance talking about the general location where the pipe end was rather rigidly grouted into a larger hole in the large manhole structure?

If so, I think it is possible a more flexible connection and or stronger, more ductile pipe (like ductile iron) might be advantageous where there might be some relative vertical movement of a structure like a manhole and the connecting piping.

 

[muuddfun]

Great comments guys. Sorry it has taken a few days to get back to this, I have been moving from one of our offices to another one. For this particular piece of sewer the chance of it being related to the construction makes a lot of sense to me. I took a close look at the trench neerer the surface when it was safe to enter, and the clay fill soils were not very good. So what you guys are pointing out sounds resonable to me.

One bit of confusion is that I am getting some of the information on what has occured second or third hand, so I am not completely sure at which location some of what I have been hearing occured. Through out the city there is more than one location with similar problems so some of the info on how the pipe was crushed, or offset, etc. may be for some of the different locations. One reason the city is so keen on trying to figure out what is going on. We may have more than one cause.

The design engineer is planning on using an in-situ liner to rehabilitate some of the area of some nearby streets with degraded pipe that maybe not completely sinking but is deteriorating to some point. They videoed the line and it needs rehab. How good would you think a liner would be if my first hypothesis about pipe stiffness relative to ground stiffenes were the cause for some of the distress, and how good do you think it would be at rehabing areas that may have problems caused from the backfill, assuming they can get it past any obstructions? I understand that these liners are steam cured in place, so how ductile/ brittle will they be?


Msucog
- What kept the area of distress around the manhole from not expanding beyond about 3 or 4 feet, was it just due to the depth?

Lcruiser
- I believe the area in the pictures was a repair for a bypass line from the original failed one. But don't know when, that is why I think maybe they used concrete but maybe they broke the bypass line first, then they tried to repair with concrete? But the concrete encasement extened from the manhole all along the pipe to the next manhole so it looked to me like it was by design for some reason. I don't believe this is an ongoing thing. So there is some uncertainty in my mind as to why they have a concrete encasement. Any if this concrete encasement if only for this bypass line, or if maybe they used similar type encasement for other sewer lines in the city. I also don't have any data on the groundwater fluctuations, but suspect that it has been fairly steady at 6 to 10 feet for the last 100 years.

aeoliantexan
- what you are describing with the "flow failure" of a sand lense sound intersting to me. I did CPTs along several of the sewer lines in the city that they are looking to repair and definetly would go through some cleaner sand lenses, or some buried channels. That is one reason why I was suspecting some differential settlement / floation may have influenced the system in the past. If we had a large increae in pore pressures from liquefaction then you could speed up the piping or flow of the sand into a sytem preaty quickly. Or maybe just enought to get the process started, and then the rest slowly degrade over time with the hydraulic differences between groundwater head and the deep sewer. This area had major earthquakes in 1940, and 1979.

Rconner
- I am not certain of the exact location of the failure in relation to the manhole. But I am definetly looking to introduce more flexiblity into the system.

Howardoak
- the streets in the area are very poor so they may be getting torn up all the time by shink and swell of the clays in the area, so I think this was something different they noticed. Your pointing out the level of construction quality seemed to hit the target. I could just see them dumping a full load in the trench when the concrete was still green and something just tearing up the pipe.

Thanks everyone let me know what else you think.

 

[msucog]

i suspect it was probably due to the fact that most contractors don't pay quite as much attention immediation around manholes when they're predominantly using large compaction equipment. also, since this is where the small leak initially started, the water followed the path of least resistence and simply eroded around the manhole. the one big sinkhole was due (i think) since it was at the bottom slope of a very large parking lot so the water that made it to the base course eroded a much larger area as the sinkhole expanded. that big one easily swallowed a car (could have taken an eighteen wheeler). all the grout around the three pipes that entered the structure had cracked and fallen out too at all three spots.