Design of a 10m cantilever wall 2

[kellez]

This is the design of a 10m cantilever wall which i was assigned to check and confirm.
From my calculations i found that the wall is failing in sliding in the seismic loading.
I want to take this step by step with you guys just to get your advice, since i am in a bit of a conflict with the engineer that actually designed this.
Below are my own calculations.

As you can see i am using Rankine for now since i am not sure how to apply coulombs on a cantilever wall.

My main concerns about the parameters used are:
1) Angle of friction of backfill - too high, they are not conservative at all
2) Angle of friction of base soil- too high, they are not conservative at all
3) Angle of ground rising above wall - this angle may not be achieved and/or the soil at the edge will not be compacted enough due to the difficulty of the slope.
4) Unit weight of backfill is a bit conservative so that one is ok

I will first carry the calculations with what they assumed for the parameters and then start making changes.

First step is the Total Sliding force shown below

 

[kellez]

My concern in the calculations below is the vertical component of Pa, Pav. This makes a big contribution to the sliding resistance.
I have also noticed that this value can greatly decrease if the angle of the ground rising above the wall decreases even by 13 degrees.
Pav is greatly depended on the slope of the backfill, do you think i should be conservative with this and reduce the value?

 

[EireChch]

Without looking at any calculations, the slope above the wall is too steep. You should do a stability analysis to check but I am pretty sure it will fail. Even if i could get a slope stability analysis to show it passes, I still wouldnt accept it. Its not maintainable.

If you have to maintain that slope angle above the wall you will need and MSE method or soil nail method. Or tiered retaining wall above.

Also, why are their parameters conservative? If they know its Engineered fill to 90% it could easily be 35 degrees.

 

[kellez]

You are right about the slope above the wall thats why i will also carry out calculations with reduced slope, lets say 20 degrees slope.

For the engineering fill/base soil which was compacted and also plate tests were carried on the fill, 35 degrees would be ok but what about the backfill sand? can it easily reach 35 degrees if its not compacted properly? or graded properly?

 

[kellez]

The initial calculations i posted above do not take into account any Partial factors of safety

Therefore below i have posted the results taking into account Eurocode 7 Combination 1 and 2 partial factors

Eurocode Combination 1 uses a partial factor of 1.35 on Permanent Unfavourable loads (ONLY), in this case "The Total Horizontal Thrust from Soil, Pa;dh"
Eurocode Combination 2 uses a partial factor of 1.25 on material properties (ONLY), in this case, the angle of friction of the backfill soil and angle of friction of base soil.
as a result combination 2 produces a much bigger value of the Rankine active earth pressure coefficient


Eurocode 7 - combination 1 - Results

Eurocode 7 - combination 2 - Results

 

[kellez]

As you can see above the wall is failing in sliding in Combination 2.

Furthermore, Eurocode 8 which is for seismic design suggests using an additional seismic load together with Eurocode 7 - Combination 2 partial factors for materials
Therefore imagine if the wall is failing now, without a seismic load, imagine what happens when i add that.

 

[EireChch]

Failure in sliding and slope instability of the slope above the wall are two different things that need to be considered.

You can redesign your wall to not fail is sliding and you will still have the instability problem in the slope above the wall.

I have never used EC8, however I always thought that for accidental loadings such as an EQ, all partial factors are 1 (i.e. unity).

 

[Mad Mike]

I think it looks doable- I would imagine some testing was carried out to obtain the friction angle of that re-compacted sand?

The 1:1,5 (V:H) back-slope should be checked by a stability analysis- it may be difficult to vegetate and some wrap-around geofabric could be considered for erosion protection.

The geotechnical parameters aren't alarming; what would be a concern though, were if this wall was being designed without a single material test being carried out and the parameters were all thumb-sucked...as noted in the other thread you're commanding, the friction angle of sand relates very much to its grading and density. I would want shear box tests on a few re-compacted sand samples to support the design.

All the best,
Mike

 

[EireChch]

Also, a cantilever RC wall is probably an uneconomical choice of wall type. A MSE wall would be much better.

 

[kellez]

Failure in sliding and slope instability of the slope above the wall are two different things that need to be considered.

You can redesign your wall to not fail is sliding and you will still have the instability problem in the slope above the wall.

Yes i do understand, once i finish the design of the wall i will then look into the slope of the soil, and ofcourse revisit the wall design

I have never used EC8, however I always thought that for accidental loadings such as an EQ, all partial factors are 1 (i.e. unity).

Are you reffering to any particular Code? Eurocode indeed asks that partial factors for materials be used in the seismic loading.

 

[kellez]

I would imagine some testing was carried out to obtain the friction angle of that re-compacted sand?

There was a thorough site investigation which indicated that the top 1-10m was backfilled with construction waste materials and had to be removed and new enginering fill to be backfilled.
The only test carried out on the new engineering fill was plate tests to measure the bearing capacity of the soil.

Some of the plate test results were:


Bearing Pressure --> Settlement
50kPa --> 0.16mm
100kPa --> 0.39mm
150kPa --> 0.66mm
200kPa --> 0.90mm
250kPa --> 1.19mm
300kPa --> 1.38mm
-->
500kPa --> 2.12mm

 

[kellez]

I have never used EC8, however I always thought that for accidental loadings such as an EQ, all partial factors are 1 (i.e. unity).

All partial factors to 1.0 but what safety factor does the code requires?

 

[BridgeSmith]

Depending on the project/site requirements, it's sometimes allowed for a wall to slide under seismic loading. IOW, if it moves in an earthquake, that can be acceptable, depending on what the backslope supports. If sliding is acceptable for seismic loading, it should be noted in the design, along with the magnitude of the expected movement.

I agree with EireChch about the MSE being a far more economical choice for a wall of that size. Block-faced MSE walls are also far more forgiving when it comes to movement/settlement, especially differential movement. The MSE typically designs out a little wider (the 'heel width'), so if the width available for excavation is tight, that would be a reason to go with the semi-gravity wall type shown.

I recently completed a wall about that height where excavation was tight. We opted for a stem wall of precast double Tee sections, post-tensioned with unbonded high strength bars, and a CIP footing. We got a value engineering proposal from the contractor to substitute a precast panel solution with tapered stems and mild reinforcing embedded into the footing. We didn't accept the VE proposal due to a lack of successful installations of the system (the consequences of failure were too serious to risk the use of an unproven system).

Rod Smith, P.E., The artist formerly known as HotRod10

 

[kellez]

Well i have no experience with MSE walls and yes the excavation is a bit tight.

Do you have any pictures of the wall you finished, it would be very interesting to see

 

[BridgeSmith]

Mine hasn't been built yet; they're supposed to start on it this spring or summer. I'll try to get some pictures during construction to post once it's underway.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[r13]

i was assigned to check and confirm.

Usually codes, assumptions and reference of source materials are listed in a calculation, and they are the first to be checked in the reviewing process. You shall ask the originator for soil report. Without that, nothing can be "confirmed".

 

[kellez]

Usually codes, assumptions and reference of source materials are listed in a calculation, and they are the first to be checked in the reviewing process.

Well they only used a software for the design. The output is as good as the parameters you input right?

 

[r13]

No. The checker is not required to figure things out from calculation. Without title page citing code and reference, the calculation is considered incomplete, because it ensures there is no deviation and eliminates guess works. It can be a base for rejecting the calculation altogether, but it's up to you then.

 

[EireChch]

Quote (EireChch)
I have never used EC8, however I always thought that for accidental loadings such as an EQ, all partial factors are 1 (i.e. unity).

Are you reffering to any particular Code? Eurocode indeed asks that partial factors for materials be used in the seismic loading.

From memory EC7 says partial factors are 1 for seismic (accidental). HTURKAK, in your EC8 forum post also seems to indicate that EC8 says seismic partial factors are 1. To me I would use partial factor of 1 for sliding, overturning and bearing capacity checks for seismic.

Show me where EC7/8 says partial factors for accidental are not 1.

(Note, partial factors and material partial factors are different things)

 

[kellez]

You are right about the Partial Factors for Actions, EC8 specifies a partial factor of 1.0 for both permanent and seismic or Accidental loads.

However EC8 specifies a partial factor of 1.25 for materials (eg. backfill soil angle of friction). This directly affects the Force applied on the wall.