Two distinctive materials behind anchored soldier pile wall 5

[gre2408]

I am designing anchored soldier pile walls for State DOT. Their boring logs show two very distictive materials behind the wall location. Mostly 18 feet of silt (N=10+/-) is sitting on hard granite type of fresh rock. The height of the wall is about 35 feet and I am trying to use FHWA design recommendation for possibly two rows of anchors supporting the wall. Since the FHWA design method uses trapizoidal shaped pressure diagram (not triangular distribution), I have no idea how to model it with these two layers of materials. Only thing I can think of is somehow conservatively averaging those two strength parameters (e.g. friction angle) and using it for the full height. Does anyone have an idea how to model the pressure diagram with these two very distinctive materials?

Thanks!

 

[STVU]

I would not use cohesion. Keep phi angle low (30 Deg)unless you have a direct shear test results. For your case, it is appropriate to use 0.5, 0.5, 0.25 trapezoid. The 0.5 section could be loaded with active earth pressure of (0.65xKaxGamma) ; where gamma is not the dry or maximum dry density but the moist unit weight of the soil averaged throughout the depth.

Check with AASHTO manual on specific applications to DOT work.

 

[STVU]

I had a typo earlier, the trapezoidal pressure should read, -0.25H, 0.50H, 0.25H.

Find similar problem to yours solved on page 469 of Foundation Engineering, 2nd ed, 1974 by Peck, Hanson & Thornburn.

 

[gre2408]

Wouldn't the friction angle of 30 degree for full depth be too conservative? As I mentioned above, the quality of the rock is excellent (RQD value is 90~100). The upper silt layer actually has some sand in it and was classified as ML or SM depending on boring locations. I do understand the trapizoidal pressure diagram you mentioned and am not trying to use any cohesion in design. What I am really curious about is, like triangular pressure diagram, how harder layer with less Ka value underlying upper soft layer influences its pressure diagram in section (e.g. reducing lateral pressure right below the interface due to less Ka). I do apprecite your comments though.

 

[fattdad]

The trapozodial stress distribution of earth pressure is a 30 percent increase over the corresponding active state. You have two distinct active earth pressures. Go ahead and determine the envelop of active earth pressure for each of these layers, compute the areas for each and convert these areas into the appropriate trapozoid.

I'd consider cohesion for the case of the granite, unless there are significant secondary discontinuities.

It's much easier to reply to these threads when we don't have boring logs and core descriptions as we can guess and all. . .

RQD=100 can still be 5-in long core pieces with clay infill.

f-d

¡papá gordo ain’t no madre flaca!

 

[InDepth]

What is the reasoning to convert to an equiv trapezoid? If the materials are very different, the pressure distribution should be different. In my mind if you are creating an "equivalent trapezoid" you may not capture the load distribution to the tiebacks appropriately. i.e. A ficticious load case. But I do recognize the need to increase the load to account for arching at the tieback restraints (origin of trapezoid distribution). Note: If the soils were similar you may be able to get away with an equivalent trapezoid.

Can the rock standup while the face is excavated? I would expect that you would only need soldier piling and lagging for the upper 18 feet. For the lower 17 feet of excavation can you go with a rock bolted and shotcreted face?....with this said...is a soil nail wall more appropriate for the upper 18 and a rock bolted face for the lower 17?


Also, is there any concern about a perched water table?

Other questions:
What about the passive? Is the rock compentent enough to act as a point restraint (socketed)?

 

[BigH]

Maybe its the time of day - very early, but why are you using a soldier pile wall to "support" fresh hard rock granite? I would think that your wall need only go down to the granite - use the pressure diagram in the silt. Would you not then need to blast to excavate the granite? Presplit holes - use rock bolts (dowels) where needed for the granite and shotcrete the face for support. Or, am I missing something?

 

[gre2408]

fattdad and InDepth, thanks for your comments

Since there are significant utility issues in upper 10 feet of the wall, DOT actually requires anchored soldier pile wall. So, soil nailing for the entire wall height is not an option for this project. Originally I have thought about the same thing InDepth suggested, using rock bolts or soil nails for lower 17 feet. However, I couldn't make myself fully understood on how passive systems like rock bolts or soil nails would work in conjunction with active systems like anchors. I just thought that it would be a relative stiffness problem for the whole system, if I see it as one wall trasfering loads through H-pile. I understand that it could be a lot simpler if considered as two different walls. However, on the other hand, I want to design the wall more cost effectively, of course while maintaining an adequate factor of safety.
Please take a look at the wall section attached herein and let me know if you guys have any comment or idea. Thanks!

 

[PEinc]

gre2408,

I agree with what BigH just said. However, your last posting with the sketch seems to indicate that there probably isn't enough room to sit the sheeting a few feet back, away from the face of the rock cut (that is, to leave a rock bench of about 3 feet wide in front of the sheeting wall). Therefore, if that is the case, you will need to drill the soldier beams down to below the proposed subgrade OR you will need to have a much smaller, line drilled, rock bench and then drill in and grout some other vertical steel elements, such as a #18 rebar at each soldier beam, which can be welded to the lower portion of the soldier beam in order to support the soldier beam weight and the downward component of the tieback anchor(s) so that the small rock bench does not shear off. Be careful. An inadequate rock bench is one of the more common causes of sheeting wall failures. Besides being tricky to design and detail, these vertical supports for soldier beams often are constructed with poor workmanship.

 

[kslee1000]

I am intrigued by the words "hard granite type of fresh rock". Does "fresh" mean the rock formation has not yet completed? Then how could it be "hard", shouln't it be "dense" instead? Do not mean to argue about the terminologies, just try to grab some ideas by reading off the information provided.

Another thing puzzles me is, if the particles are still largely in a seperated condition (dense soil, nearing rock), couldn't the earth pressure coefficients be evaluated by its friction angle? And, if the rock has already formed as a cohesive mass, then unless there is sliding/weak plane, would it exert any pressure on the sheetings? Did I misunderstand something? Please comment.

 

[DRC1]

My 2 cents - Along the lines of PE Inc, the lateral pressure will apply only to the silt. Granite with fresh breaks and RQD's of 90- 100% is not transfering lateral load. I am not sure of the origin of the FHWA loading diagrams and not a big fan of them. Typically the pressure envolope for multiple supports is .65 ka gama H, where H is the height. A significant change in ka or gama (the unit weight) would cause a bump in the diagram.
To solve your problem, I would try to bring your first anchor up a little, and perhaps flatten it to 15 degrees to see if it would work as a soil anchor. I would then place my second anchor just above the rock and angle it back at 45 degrees to get a good rock anchor. Be sure tah you have a good toe on the soldier beam or add a tie near the bottom (which would not have a very high load) to secure the bottom. But since we don't know that much about the project specifics,you should review it and run numbers before using any of our suggestions.
Note that shoring design is tricky, even for problems that may appear simple, so if you do not have a lot of experience designing sheeting, have someone who does review it with you.
Good Luck

 

[gre2408]

Thanks for all the comments!
As PEinc said, we don't have the luxury to have 3 ft of rock bench along the wall. If we did, this could have been a lot easier to work with. Basically the wall needs to be plumb from top to bottom. So, I am thinking to drill down and have soldier pile socketed about 3 ft below bottom of wall (See my sketch) like PEinc mentioned earlier. DOT engineers would not allow us to expose rock face without any protection. So basically my idea was having one anchor at top for soil (ML or SM) retention and soil nailing with shotcrete face for rock cut below it. Then I was further thinking how to design H-pile section required for soldier pile against bending moment between the anchor and bottom of wall. Since soil nails are passive elements, they will require some degree of wall movement to be activated. Although boring logs show very high REC and RQD values, it could still be 4 inch long core pieces with manganese seams (as fattdad said above) and rock mass would want to slide out as cut is made. As usual, we don't have any information on orientations of faults, joints or etc. I just thought that 25 ft span below the anchor down to bottom of wall is quite significant length and would need more careful consideration prior to finalizing my design. Thinking that somehow making soil nail system more stiffer would help to reduce bending moment along the H-pile (e.g. using more stiffer steel bars for nails and tightening nuts on shotcrete face with more torque like using air wrench). How do you guys think about this concern?

 

[kslee1000]

If the top row of anchor is adequate to prevent lateral movement, than from structural point of view, you have a beam with two supports and a long cantilever, which is loaded by the silt. By such loading scheme, the beam segment tends to bent (push) towards the rock rather than bowing out. Make sense?

To avoid the tendency of rock movement, rebar dowels with epoxy grout should do the trick, and can be utilized to fasten the lower wall.

 

[gre2408]

I do understand what you are talking about. However, my point is that what elements (either rock bolts or soil nails) are used at rock face, unless those are pre-loaded somehow, it is almost impossible to prevent H-pile section from being negatively influenced by bending. If so, how we can quantify the impact. It seems to me that it is requiring more careful consideration on it.

 

[kslee1000]

The system I mentioned can be pre-tensioned. We had use it in tunnel works for rock stabilization. The rate of sucess depends on the competency of the rock mass.

 

[DRC1]

If you have a high quality granite, odds are it will stand. If not I doubt the H-piles will prevent movement. You would probably be better off implementing a rock bolting scheme than relying on soldier piles. If you were able to bench the rock and found unfaverable jointing, you would have to rock bolt anyway.

 

[InDepth]

Maybe think of it this way
1. On the lower portion: If you place rock bolts either side of the soldier pile, then all the load is taken by the rock bolts if the rock moves...the H pile is just going for the ride. Its a stiffness compatibility problem, the bolts in tension are probably stiffer than H pile in bending. i.e. The H pile is not loaded, but continually supported and merely displaced.

2. Thus, model the pile will a load due to the silt, and then add an assumed lateral displacement to the pile in the rock. What you will probably find is some additional load going to the last anchor in the silt. - Pin at tiebacks and at bottom of H pile. If you performed finite element analysis you would probably see a stress concentration/ bearing failure at the rock/silt interface and the H pile.


OR just add on rock bolt each side and place an angle over the rock bolt and H pile - a saddle support....then it is really continually supported and not acting as a beam element but a shear element.