Tieback soldier pile wall built from bottom up (in fill)

[cwrugrad]

I know these walls are more usually built by excavating, but...

Does anyone know how to analyze a wall like this for the temporary condition just prior to installation of the 3rd tieback, which is when the problem becomes indeterminate?

I used the approach described in Braja M. Das's Foundation Engineering text to analyze the wall as a cantilever wall, prior to installation of the lowest tieback, and as a wall with one row of tiebacks, prior to installation of the 2nd tieback from the bottom. But parts of the wall are tall enough to need a 3rd row of tiebacks.

The two approaches described in the 2004 AASHTO LRFD Section 11.9.5.1 Commentary (Tributary Area Method and Hinge Method) give very different results for the single tieback condition if the tieback is below the center of the pressure diagram, and the way they ignore everything below the lower groundline makes me uneasy.

 

[Panars]

The pressure diagrams for anchored walls in AASHTO LRFD are for top-down construction only. It says this at the very beginning of Section 3.11.5.7. I am guessing that you are trying to use these pressure diagrams because you state that they are ignoring everything below the groundline.

The FHWA manual on ground anchors (Geotechnical Circular No. 4, FHWA-IF-99-015) has one section on anchored walls for fill applications (page 122, section 5.11.5). If I read it correctly, it recommends using the same earth pressure diagrams that you would use for a cantilever wall. There are a number of other considerations it gives for bottom-up construction of anchored walls. Given all the concerns, it is evident why bottom-up construction of anchored walls is not done very often.

 

[PEinc]

Do you have a total fill wall above a flat existing ground surface or are you installing a tiedback wall at the toe of an existing slope?

Bottom-up tiedback walls are expensive, trouble, and not recomended. Tieback drilling and testing have to be done up in the air, off a high platform or suspended by a crane. Usually you also need to install temporary cribbing behind a tieback level in order to test the tiebacks. That is because the wall has no backfill at that level or else the soldier beams would be overloaded prior to the tiebcks being stressed and locked off.

Sometimes, a bottom-up, fill wall is better constructed with buried tie rods and deadmen as the fill is being placed, rather than with tieback anchors.

 

[TDAA]

I would recommend MSE. If you have room for tiebacks, you should have room for reinforcing. If part of the wall is in cut, you can do tiers with different wall types, or stack them straight. We have a bridge project with the cuts using secant piles with an MSE wall above.

 

[cwrugrad]

The wall will be installed at/near the toe of an existing roadway side slope, to widen the road. Maximum exposed wall height would be about 23 ft. The client would much prefer that the wall construction not interfere with existing traffic.

The tiebacks would have to be re-stressed to the correct level for the final loading condition following completion of the backfilling, and we would plan to have them tested at that point rather than during construction.

The Contractor would be required to install a couple of test tiebacks in each of the two different soil types encountered along the length of the wall, and test them to failure before starting the wall construction.

 

[TDAA]

At the interstate widening here, they performed a cut along the existing road, with a soil nail wall. They are now filling the new lane and constructing MSE.

 

[PEinc]

cwrugrad,

The approach you are considering will be a nightmare. You have described the classic example of how not to build a tiedback wall.

I would agree with TDAA about using temporary excavation support near the edge of roadway, digging down to subgrade, and then installing an MSE wall. The temporary support wall could be tiedback sheet piling, soldier beams & lagging, or soil nailing (if the material is not loose granular or wet). Whichever support wall you use, it would have to be located at or just beyond the ends of the MSE wall grogrids or metal straps. You will also need room behing the temporary support wall for your Jersey barrier for traffic control.

 

[PEinc]

Sorry, I meant geogrids, not grogrids.

 

[cwrugrad]

Thanks to TDAA for a great idea. If soil nails would work, overall cost would be very reasonable, and I would sleep better at night knowing the project is unlikely to result in a construction disaster.

My problem right now is that I am the checker and not the designer, so I'll need to sell the change.

I am told there are ways to install tiebacks from above, sitting on the fill on the high side of the wall - though that wouldn't work for re-stressing and testing the middle tieback in a 3-tieback-level wall.

But maybe 2 tieback rows will work, if the Contractor braces the soldier piles from the low side at the start (there's a bike path on that side), so the bottom row of tiebacks can be at a more normal height that would be typical of top-down wall construction.

 

[PEinc]

A3' high wall should need only 1 tier of tieback anchors about 8' down from the top of the wall. However, that is still about 12' above original grade at the toe of slope.

You can install tiebacks from behind the sheeting wall but this requires work area which would probably interfere with traffic.

Possibly you could drive soldier beams or sheets and then tie back the wall with a tieback drill mounted on top of a dump truck to get up to the required height of the single tier of tiebacks. You would have to look at a cross section through the work area.

Even with soil nailing, you will need a work bench about 20 feet wide at each nail level for the drill to travel on. With the existing sloped embankment, it could be hard to get a wide enough work bench.

You need to decide what type of new, permanent wall is most appropriate. Then, you leave the design of the temporary support wall up to the contractor (unless the contractor is your client). Most jobs just indicate where the designer thinks a support wall is needed. Then the contractor figures it out. You should not try to tell the contractor how to build the temporary support wall. You may not be familiar with the contractor's equipment, experience, preferences, etc.

 

[cwrugrad]

PEinc,

I don't think a single tier would work for the taller portions of this wall, since we have to design for the new AASHTO LRFD traffic-barrier impact loading, in this case Test Level 4 (54 kips horizontal force at 2'-8" above roadway surface at top of wall).

A temporary excavation-support wall located right next to the traveled way would also have to handle that impact force, because the client won't be happy if we show temporary traffic barrier set back from the shoring wall into the existing right lane.

You are right that we should not specify Contractor means and methods, but the cost estimate is affected by what we judge to be feasible, and soil nail walls are usually quite inexpensive compared to other alternatives like tieback soldier pile or sheeting walls.

The available work space between the existing roadway and the permanent wall would vary from about 10' to about 16', but in the area where we think we need 2 rows of tiebacks the minimum would be 14'. Of course, some of that would have to be deducted for orange barrels or temporary concrete barrier. I know that's pretty tight, but is it doable?

 

[DRC1]

I have had tie backs installed with 10 feet of clearance to the railroad using small rigs. slow but doable. We had a cantilever wall along a railroad that had to be modified to put a large crane at the face of sheeting. Installed 2 rows of tiebacks off scaffolding. Top down you might get 23 feet to work, but with barrier load, it is doubtful. I would think two would work. Design the first 10 to 12 feet as a cantilever. I would use crushed stone 3/4" for backfill. Does not need compaction, will drain and will give you lower lateral pressure if wide enough. I would locate the first anchor 3-4 feet below the top of the fill. Check load using triagular distribution as a first tie and rectangular as a bottom tie. I would do normal proof and preformace tests and lock off at 80 - 100% of working load. I am missing the piont for coming back to do testing and lock off. Continue filling and anchoring in the same manner for the next level.
Another option would be to use a king pile system to cantilever 23 feet, possibly supporting the barrier on an independent pile foundation.
Another option may be a concrete gravity wall. You could always put part of the foundation below the bike path.
Hope this helps

 

[PEinc]

cwrugrad,

If you use a soil nail wall, you will not be able to support the factored 54 kip load. I have seen many engineers or DOT's specify that the temporary soldier beams extend above grade a few feet to act as a crash barrier. This is a bad idea. That's what Jersey barriers are for, not temporary sheeting walls. For permanent tiedback walls, I've mounted a permanent barrier on top of the wall facing and soldier beams. I haven't had to use the 54 kip Test Level 4 load yet but it sounds pretty heavy even considering that it is factored. You must have some heavy truck loading or military loading. I believe for ASD, there was only a 10 kip impact load in the same loaction. Can you take the impact load with a gravity slab that has a parapet barrier mounted on top of the slab? This type of barrier system is frequently used with MSE walls.

If you are using the new FHWA loading diagrams for tiedback walls, there will be no triangular loads, even for a one tier wall. You may need to use the newer, unsymmetrical trapezoidal diagrams which vary with the tieback locations.

Again, backfilling with crushed stone is no good for a top-down tiedback wall; and a bottom-up tiedback wall is a bad idea.

 

[cwrugrad]

PEinc,

The geotech says the soils aren't dense enough for soil nails. So much for that idea.

FYI, the barrier crash test levels in AASHTO LRFD go up to TL-6, which is a tank, and the horizontal design load for that is 175 kips!!! Take a look at Table A13.2-1 on page 13-17 of the AASHTO LRFD. It's amazing. They seem to be serious. See page 13-23ff for deck overhang design requirements.

A moment slab would have to be awfully big and hefty to resist 54 kips; seems to me it will be cheaper to install extra top-row tiebacks using a horizontal waler.

What troubles me the most about both designing a tieback wall for such a big eccentric force as the 54-kip impact force, and about bottom-up tieback soldier pile wall construction, is the uncertainty about how the forces would be distributed in a wall with more than one level of tiebacks.

The essentially rectangular earth pressure diagrams I keep seeing for multilevel tieback wall design seem not to address eccentric-loading situations.

I wonder if the client would be willing to accept a half-bridge concept...

 

[BigH]

Can you use a deadman anchoring system? You can construct the wall from the base up but use a few levels of deadmen to provide the passive lateral resistance to keep your wall in place. Might be easier than, say, normal tie-backs.

 

[PEinc]

Density is not the key to soil nail walls. For soil nail walls to be economical, the soil needs to stand unsupported vertically, about 5 feet high, for a reasonable time so that the excavated face can be shotcreted and the soil nail drill holes should ideally stay open without casing.

AASHTO & FHWA do not have rectangular earth pressure diagrams anymore for multi-tier tiedback walls.

AASHTO and/or FHWA seem to be getting out of hand with these increasingly large vehicle loads.

 

[BigH]

cwrugrad (I take it that that is Case Western of Cleveland!) - for the eccentric loading due to the crash barrier, have you considered a friction slab/jersey barrier combination such as that given in BS:8006 (I believe that is the correct number). The friction slab can also be tied down to offer more overturning resistance. Tanks should be on flatbeds anyway for transport!

 

[DRC1]

If this is a permenant wall which it sounds like, you could drive some sheets into the toe of the slope, excvate doen and build a conventional concrete retaining wall which could incorperate the barrier. If the top of footing is below grade, you could put the bike path over the top of it.

 

[DRC1]

Another thought is a precast bin wall such as Doublewal. No foundation is required, although you do need a pair of leveling strips about 12" wide and about 9" deep. Ste the bins on the strip, backfill with gravel, fill in behind the barrier,add another lift, repeat until you get to the top, set the crash barrier parrapet and pour concrete in the top unit to lock the barrier to the wall. Permenant attractive gravity wall, no sheeting or tie backs required. They will give you stamped drawings.

 

[cwrugrad]

BigH,

I agree with you about the tanks, plus if they're going to drive drunk it shouldn't be the taxpayers' problem, right?

It seems likely to be cheaper to just add extra top-row tiebacks to the wall rather than to build a whole moment slab with tiedowns that crisscross with the wall tiebacks. But to do it right, we'd have to increase the number of piles too, to ensure all the tiebacks in the row do their fair share (tiebacks not located at piles would have extra "give" due to deflection of the waler), so a friction slab/moment slab might end up a good idea - I'll have to run the numbers and see.

P.S. - yes, you guessed it. Are you a fellow alumnus?

DRC1,

The apparent problem with a conventional concrete wall is that there isn't much room. We'd have just enough space (for a reasonably-proportioned design) if the Contractor excavated on a vertical face within a foot or so of the existing traveled way, and since the geotechnical engineer says the soils aren't suitable for soil nailing, the Contractor would probably have to build a temporary soldier pile wall in order to build the permanent wall. The temporary wall would also have to be designed for the barrier collision load, since there wouldn't be room for a setback for the temporary traffic barrier.

I will definitely visit Doublewal's website, thanks!

If they'd just let us raise the bike path 10' or so, too, then we could split the wall into two walls of lesser height, and MSE walls would fit into the space available with minimal hassle. But I'll go out on a limb here and guess that they will politely decline that idea...