Soldier Pile Wall on slope

[reshmislal]

We are proposing a soldier pile wall with drilled caisson for a slope repair project. The height of the wall is 13'. The pile embedment is 30.5' in soil. No ground water table encountered.
The soil parameters for the fill: phi- 30, unit weight - 120 pcf
The soil properties below subgrade: phi- 36, unit weight- 130 pcf
Slope- 1.5:1.
Geotech subconsultant did deflection check in PY wall software and recommends W18 x283 fpr soldier pile section. They used a 0.45 p-multiplier to account for the reduction in the passive pressure due to slope. The section seems to be huge. Have anyone come across a similar situation. Please let me know your thoughts.

Thanks

 

[msquared48]

How about a tieback to help?

Mike McCann, PE, SE (WA)

 

[oldestguy]

What is the classification of the fill as well as the original ground? No cohesion? The data for the fill seems weird. Well graded granular fill of that density and a very low friction angle??? What spacing of the piles? Has a global stability analysis been done?

 

[GeoPaveTraffic]

Depends on the spacing of the piles, the height of the slope, depth to rock, and other factors.

From similar designs that I've done, W sections are rarely the best way to reinforce these types of drilled shafts. Much more efficient to use rebar cages.


Mike Lambert

 

[reshmislal]

Thanks for the responses. The pile spacing is 6'-0". Fill is actually original ground. The proposed plan is to excavate 13' and construct soldier pile wall. In the borings they have sand for top 13' and sand and gravel for bottom layer. No rock is encountered. I am not convinced with the p-multiplier used. Have anyone done deflection check in PY-wall?

They did slope stability analysis. Tie back would be the next option if this doesn't work.

Please see the attached sketch.

Thanks

 

[GeoPaveTraffic]

Based on your sketch you got all kinds of problems. Not only does the slope continue uphill from the wall, thereby increasing the forces on the wall. The slope also continues downhill of the wall, thereby reducing the effective embedment of the wall.

Frankly I would be surprised that the W section you listed at 6 foot centers would work at all.

Mike Lambert

 

[oldestguy]

GeoPave is right. Those Ka values are for level backfill. Kp on the down hill also would be for level area there, unless you make a reduction. In addition, just how you use those numbers against piles at spacing is another question.

Figure 10-3 from my copy of the Navy Design Manual for geotechnical shows the Ka as 0.8 for that slope. Your flat area will change that some.

Kp is about 1.5 on the downhill side, not the usual .

Have you looked at making that flat area all in cut, rather than fill. It might have less risk.

We are not here for designing things, but my comments are that, instinctively, I don[t like the idea at all.

 

[Riggly]

I agree with others, that this design is very risky. The downhill slope is 1.5:1, that is quite steep. Will you rely on this slope being stable over the long-term? There might be shallow slides/sloughing in that region. I would ignore the passive resistance of at least of few feet of that slope totally - I would be that conservative in this case. The upper edge of the slope is only about 9 feet from your wall. Do you think that amount of soil is able to resist these rigid drilled shafts?

I would expect you looked at slope stability at a few places to, including above and below the wall. Also to fully mobilize passive pressure you need some movements, so even with that, a significant reduction makes sense.

You said that the wall is 13 feet, but is the w-section extending to the bottom of the hole. In other words, do you have an concrete encased w-section below ground line, and exposed above ground, making a total pile length of 43.5 feet? Or are the W-sections connected via a base plate at the top of the piles? I guess either of these methods would affect the size, but I agree the W-section is a little hefty. I would would on piles w-sections encased in concrete of similar height, but on flat area to withstand E-80 train loads, and the sections were much lighter. I would check the geotech structural design parameters to ensure the were using reasonable structural analysis.

How big are the caissons?

 

[reshmislal]

Thank you all for your responses. The passive pressure for 3 ft below ground is ignored per Pennsylvania code. This is in PA. The pile is encased in concrete making a total pile length of 43.5 ft. Their slope stability analysis have a safety factor of 1.6. I will look into tie back design. Thank you for your inputs.

 

[oldestguy]

Perhaps some one will clarify it for me, but the usual conditions I am familiar with are that the "dumped" slope of clean granular materials (as in stock pile) is approximately the angle of friction. On that basis, before anything is done to the slope it sits at a stability safety factor of 1.0, not 1.6.

 

[aeoliantexan]

I agree with oldestguy. There was a slope of sand standing at its angle of repose, approximately equal to the friction angle. Someone made a road by cutting and filling, steepening both the uphill and downhill slopes. This is often done for road construction on mountainsides, and the result is continuing maintenance problems. The overall stability is tenuous, and the downhill slope is having a hard time keeping itself there without accepting the additional force of the wall. Long tiebacks to transfer those forces well back into the mass may work.

Remember that the angle of repose for a saturated cohesionless slope is one-half the friction angle. You can't afford to have stormwater seeping into a ditch on the uphill side of the road.

Be careful. You have a tiger by the tail.

 

[fattdad]

we use soil nails a lot for slope repair. The drawing seems to show a slope redesign though!

I agree that the top and toe slopes need to be carefully considered by a real geotechnical engineer. There are also arching considerations assumed by lateral load on pile calculations. These arching considerations may or may not be invalidated by the toe slope. In other words, I hope that the p-y calculations and ultimate load safety factors consider the toe slope.

I hate the term, "Angle of repose." I don't allow cohesion in my slope analysis. Cohesion doesn't hold up over time and the freezing and thawing, wetting and drying, etc. of the design life will reduce cohesion to zero. As such, the safety factor of slope (in infinite slope failure mode) is tan(phi)/tan(beta). You can't return a 1.5:1 slope with a safety factor of 1.3, irrespective of the drilled shaft design.

I agree, the design has problems.

f-d

ípapß gordo ainÆt no madre flaca!