Keystone MSE Wall Design and Seismic Applications

• 4/5/16 to 4/7/16
 Topics

1. Product Offerings
1. Extensible - Geosynthetic
2. Inextensible – Steel
3. Research, Development and Testing
2. Design Basics and Introduction to Methodologies
1. Differences
2. Comparison
3. Introduction to Seismic Design Principles
1. Discussion of the current state of design practices
4. Keystone Wall Design Software (Interactive)
1. KeyWall
2. KeyDraw
3. KeySystem I Spreadsheet
 Keystone Introduction

• First to market and #1 Structural SRW in the World

• Headquarters in Minneapolis, MN– Worldwide
Distribution network
• Leader in Engineering Development of SRW systems
• First segmental retaining wall to market (1986)
• 30 years of innovation: 180+ Patents / Patents Pending

• Keystone Departments
• Engineering
• Marketing
• Sales
• Product development and production
Structural Product Offerings
 Products – Country Manor

– Entry Level Structural Product

– Create small freestanding walls, parapet

walls, pilasters, columns, and retaining walls.

– 3 unit system with 7 unique face dimensions.

– Units are packaged together as a system to

create a random, natural look.
1. 2. 3.

– Three textured sides on each unit.

Piece # 1 2 3
• Used as an exposed end unit or a 90°corner.
Height 150 150 150
Width 400/350 300/250 150/100
– Shouldered pins give multiple setback
positions. Depth 250 250 250

• Create vertical or setback walls Weight (kg) 27 18 11

Pins Shouldered Pins
Note: Unit colors, dimensions, weight, and availability vary by
manufacturer.

3 pc. System (1;4;6)

System availability varies by manufacturer. Contact directly for details.
Versatility – 3 Sides Textured
Country Manor

Unique
Face
Texture
Country Manor – Parapet Walls
Country Manor Water Feature
 Product – Standard Unit

Keystone
Standard
Piece # 1 2

Height 200 200

Width 455 455

Depth 533 533

Weight (kg) 52 56

Pins Straight Pins –

2/unit
0.091 m2 per unit
1. TRI-PLANE 2. STRAIGHT
Note: Unit colors, dimensions, weight, and availability vary by
manufacturer.
 Product – Standard Unit

• First Segmental wall unit developed

• Early gravity applications, still best option for gravity wall
applications
• Later geogrid introduced for taller walls
• 30 years of unit evolution
• Development of specialty applications for taller gravity
walls
• ICC-ES report ESR-2113, ICC-ES is a subsidiary of
International Code Council.
– Evaluation service for independent verification of compliance to
the International Building Code (IBC)
 Keystone Standard™ Evolution

Standard I Unit Standard II Unit

Standard III Unit

 Standard I → III
• Wider pin receiving hole for more alignment flexibility
• Vertically aligned cores for ease of core filling
• Individual unit weight decrease of 4.5 to 5.5 kg
• Superior connection strength
• Greater construction flexibility
Gravity Applications
 Gravity Applications

• Single Width Unit Maximum Heights (Good Soil Conditions)

– Battered (no surcharge) – 2 m
– Vertical (no surcharge) – 1.5 m

• Interlocked Back to Back Standard Units (Good Soil

Conditions)
– Battered (no surcharge) – 2.8 m
– Vertical (no surcharge) – 2.2 m

• For most gravity application batter always recommended

 Keystone
Compac IV
1. Straight Face Split 2. Tri - Face Split

Piece # 1 2
Height 200 200

Width 455 455

Depth 305 305

Weight (kg) 37 34

Pins Straight Pins -2/unit

Face Area 0.091 m2

Note: Unit colors, dimensions, weight, and availability vary by
Bottom View manufacturer.

Of Compac IV
 Compac Unit

• Workhorse of Keystone – probably 75% of walls Nationally

& Internationally are with Compac unit designs.
• When utilized with geogrid reinforcement, walls of all
height can be designed.
• Highway Innovative Technology Evaluation Center (HITEC)
– U.S. DOT highway geogrid wall system evaluated for long term
connection
• ICC-ES report ESR-2113, ICC-ES is a subsidiary of
International Code Council.
– Evaluation service for independent verification of compliance to
the International Building Code (IBC)
 KeySystem II HITEC

• Evaluation published
April 2012
• Evaluation based on
AASHTO LRFD 5th
Edition, 2010 and NHI
FHWA 2009
• Keystone Compac II
Units
• Mirafi Geogrid
 Keystone Compac™ Evolution
Compac II Unit

Compac III Unit Compac IV Unit

 Compac I → IV
• Wider pin receiving hole for more alignment flexibility
• Vertically aligned cores for ease of core filling
• Individual unit weight decrease of 4.5 to 5.5 kg
• Superior connection strength
– 40% to 100% increase in geogrid connection
• Greater construction flexibility
Compac IV
 Keystone Wall Advantages

• Universal Facing Unit

– One facing unit used to construct wall system
– Universal facing unit allows on-site alignment changes
without delays of casting specialty panels
– Facing units are field cut for pipe penetrations or other
obstructions
– Facing unit can be field cut for slip joints in excessive
settlement conditions
– Face batter adjustable from 1H:8V to 1H:64V
– Various color and texture options available
 MBW Connection and Alignment

• Keystone Fiberglass Pins

– Quick & easy alignment for stacking units
– Provides various degrees of setback
– Unit shear connector every 12”
– 20% better connection strength 12” O.C.
– Allows geogrid to be pre-tensioned
– Ensures grid is attached to wall
– Non-corrosive
– High strength 6,400 psi short beam
 Components

• High Density Polyethylene (HDPE) Geogrids

– Behaves like plastic
– Uses lower % of ultimate due to creep
– Extruded and stretched structure
– High dimensional stability
– Newer versions have lower soil interaction values
 Components

• Polyester (PET) Geogrids

– High strength uniaxial polyester geogrid
– Woven and coated with a PVC coating
– Broad ultimate tensile strength range
– Excellent durability properties
 Products – KeySteel

• KeySteel® Soil Reinforcement

200 mm
(7.8 in.)

200 mm
(7.8 in.)
 Components – Face Unit

• Keysystem I Unit
– Compressive Strength
• 4000 psi
– 82 kN Ultimate
reinforcement connection
strength @ 15 mm Displ.
 KeyStrip™/Facing Connection

KeySystem I Unit/KeyStrip Connection

18600 plf @ 0.5 in. displacement 19600 plf Ultimate Connection load
25000
Connection Capacity (lbs)

20000

12000 plf @ 2” deformation

Max Keystrip design load is
15000 9000 lbs (FS conn >2)

10000

5000

0
0 0.5 1 1.5 2 2.5

Displacement (in)
 Components – Soil Reinforcing
• KeyStrips
– 9mm, 10mm, 11mm Wire
– Crossbars at 300mm & 450mm
– Hot Dipped Galvanized
– 75 to 100 year design life
– Connection capacity limited to 42.4 kN,
after factors of safety applied
 Components - Pins

• W24 – 14.3mm Steel

Connection Pins
(9/16” diameter)

• 12.7mm Fiberglass
Alignment Pins
 Wide Range of Applications

• Heavy Construction
– Industrial walls
– Heavy Highway
– Railway Design
• Transportation
– Bridge Abutment
– Development Roadways
• Seismic
– Better Seismic Performance under heavy loads than
extensible geogrid
 KeySteel

Waikato Expressway Construction

KeySteel
Design Methods and Seismic
 Successful Walls

Construction

Soils Design

Products

Require attention to four items.

 Soil Summary

Soil Types
• Granular Soils - Sand & Gravels
• Fine Grained Soils - Silt & Clays
• Other - Organic, Peat

Preferred Soil Gradation

Sieve Size % Passing
2 inch 100-75
3/4 inch 100-75
No. 40 0-60
No. 200 0-35
PI < 15 LL <40
 Soil Summary

Design Properties
• γ, Moist Unit Weight
• φ, Effective Shear Strength
• c, Cohesive Strength

Minimum Compaction Density

• 95% of Standard Proctor
• 92% of Modified Proctor
• Moisture +0%, - 3%
 SRW Design

The art of balancing driving and resisting forces.

External Stability Internal Stability

• Base Sliding • Reinforcement Tension
• Bearing Capacity • Pullout
& Settlement
• Overturning
• Global Stability
Facing Stability
• Connection
• Shear/Bending
• Overturning @ Top
 Gravity Walls

Low Permeability Soil Finished Grade/Backslope

Keystone Units

Setback
Batter

Wall Height Retained Soil Zone

Limit of Excavation
Finished Grade

Drainage Fill

Leveling Pad Drainage Collection Pipe

Foundation Soil Zone

Gravity walls rely on their mass

and batter to resist overturning.
 Gravity Wall Failure Mode

Overturning

Weight of Wall
Earth
Pressure

Pivot Overturning
Simple Overturning
Weight of Wall
Earth
Pressure

Pivot
Settlement and Overturning

Gravity walls typically fall over

when built too tall for unit size
 Reinforced MSE Walls

The creation of a reinforced mass or

geocomposite, made of soil,
geosynthetics or steel reinforcement,
and concrete facing units,
of sufficient size to resist
the imposed forces.

Tileco Keystone Engineering Seminar

 Reinforced Walls

Low Permeability Soil Finished Grade/Backslope

Keystone Units

Reinforced Soil Zone

Setback
Batter
Wall
Height Geosynthetic Limit of
Reinforcement Excavation

Finished Grade Retained Soil Zone

Drainage Fill

Leveling Pad Drainage Collection Pipe

Foundation Soil Zone

Reinforced walls rely heavily on

soil strength for the structure.
Design Guidelines
 Design Methodology

• Rankine and Coulomb design using allowable stress

design (ASD) and factor of safety (FS) methods.
𝑅
≥ 𝐹𝑆
𝑃
R = Resistance (stabilizing forces)
P = Load (destabilizing forces)

• AASHTO Load Resistance Factor Design (LRFD) and

Capacity Demand Ratio (CDR)
𝜑𝑅
𝐶𝐷𝑅 = > 1.0
(𝛾1∗𝑃1)+(𝛾2∗𝑃2)
𝜑 = resistance factor
R = resistance (stabilizing forces)
γ = load factor for a certain load type
P = Load of a certain type (destabilizing force)
 Earth Pressure Theory

H 
 R
Pa

 

Coulomb (1776) Theory

 Earth Pressure Theory

H 
R


Pa

Rankine (1857) Theory

 Earth Pressure Theory

0.3H Coherent Grav. - Ko + OT

Simplified - 1.7 to 2.5Ka

Coherent Grav. - Ka + OT
Simplified - 1.2Ka

Inextensible Theory
 K/Ka Ratio (Proposed)

*Unofficial Document
Currently in AASHTO sub-committee,
awaiting review and approval.
 Formulas – Earth Pressure

Coulomb
Pa = 1/2 g H2 Ka

Sin2 (+)
Ka = 2
Sin2  Sin (-) 1 + Sin(+)Sin(-)
Sin(-)Sin(+)
Rankine
Pa = 1/2 g H2 Ka

Ka = Cos() Cos() - Cos2()-Cos2()

Cos() + Cos2()-Cos2()
 Failure Plane Location

For level surcharge and infinite slope conditions

• Coulomb - Coulomb failure plane varies as a function of the wall geometry and friction angles for both the soils and
the soil wall interface.

− tan 𝜑 − 𝛽 + tan 𝜑 − 𝛽 tan 𝜑 − 𝛽 + cot 𝜑 + 𝑙 {1 + tan 𝛿 − 1 cot 𝜑 + 𝑙 ]

tan 𝜌 − 𝜑 =
1 + tan 𝛿 − 𝑙 [tan 𝜑 − 𝛽 + cot 𝜑 − 𝑙 ]

where:
𝜑 = angle of internal friction
𝑙 = batter of wall measured from vertical (α - 90°)
β = slope angle above the wall
δ = angle of friction at back of wall

• Rankine – Where 𝜌 is fixed and measured from horizontal under all design scenarios, which is only technically correct
for level surcharge applications and minimal wall batter.

𝜑
𝜌 = 45° +
2

• In theory, the Rankine failure plane varies under backslope conditions. However, it is customary to fix the failure
plane at the equation above in earth reinforcement design, thus best representing the curved failure surface and
locus of maximum stress points for a reinforced soil mas
 Design Methodology Comparisons

• Major difference between Rankine and Coulomb

– Coulomb model and equations account for friction between
the back of the wall and the soil mass as well as wall batter.
– Rankine equations more conservatively assume no wall
friction at the soil-wall interface and a vertical wall structure
which greatly simplifies the mathematics of the problem.
– The friction at the back of the wall face and at the back of
the reinforced zone for external stability computations,
provides an additional resisting force component that helps
support the unstable wedge of soil.
– Because of the additional resisting forces, lateral earth
pressure in Coulomb is generally less than Rankine method.
 Design Methods

• The limitations of closed

form solutions, such as
Coulomb and Rankine, is
that only simple level and
infinite sloping surcharges
with uniform loadings can
be analyzed.
• It is necessary to look at a
“trial wedge” or
“approximation method”
when attempting to analyze
broken back slope or other
slope/load combinations.
 Design Methods

• AASHTO and NCMA suggest an approximation method for

broken-back slope conditions that defines equivalent
design slopes for the external analysis.
– However, the internal analysis is not well defined for unusual
slopes and loading conditions and the designer is expected to use
engineering judgement with the simplified methods.
• Keywall “Trial wedge” analysis used is consistent with the
fundamental assumptions of the applicable coulomb and
Rankine theories by setting δ=β.
– “Trial wedge” results match the equation solutions for the level
and infinite slope conditions, but will determine the “correct”
internal and external values for broken slope conditions and offset
live and dead loads.
*Note AASHTO LRFD use the AASHTO “Simplified” method for calculating internal pressure
and the trial wedge for calculating external loading conditions. MSEW utilizes Trial Wedge.
 Design Methods

• Which method do I use?

– Each methodology is fundamentally different
– Understand the design methodology for a particular project
– Project Specific
• Public / DOT – AASHTO / Rankine
• Schools – Coulomb
• Private – Rankine
– The most important issue is that the designer understand
and be comfortable with a design methodology, its
limitations and follow the methodology in its entirety.
 Design Methods

Advantages / Disadvantages
Coulomb Rankine
• Provides lowest calculated • No assumption has to be
earth pressure by taking all made with regard to friction
beneficial components into
account between the wall structure
– Wall Batter and retained soil mass.
– Wall Friction • Simpler formula and failure
• Reinforcement lengths plane definitions
significantly longer at the top
of wall than the bottom due to • Due to the higher earth
flatter failure plane pressure coefficient,
• Reduced earth pressure may stronger reinforcement may
permit vertical spacing of be necessary at the bottom
reinforcement in lower walls of wall.
that exceed the wall facing’s
stability during construction
 External Stability – Reinforced Wall Forces

NCMA/Coulomb does not include vertical forces (Allowed in 3rd Ed.)

Rankine/AASHTO does include vertical forces
External - Sliding Analysis
 External - Sliding Analysis

Coulomb Rankine - AASHTO

• Driving Forces • Driving Forces
(Pa+Pq) cos δ (Pa+Pq) cos β
*
• Resisting Forces • Resisting Forces
(Wf+W1+W2) tan φ (Wf+W1+W2+Pav+Pqv) tan φ
weaker soil (reinforced or weaker soil (reinforced or
foundation) as the resisting force foundation) as the resisting force

Note: Live load does not contribute to resisting forces.

*Rankine includes vertical earth load components. NCMA 3rd Editions now
permits the inclusion of vertical earth load components at the designers option.
External - Overturning Analysis
 External - Overturning Analysis

Coulomb Rankine - AASHTO

• Driving Moments • Driving Forces
(Pa+Pq) cos δ, are the driving (Pa+Pq) cos β, are the driving
forces at there respective forces at there respective
moment arms of H/3 or HS/3 moment arms of H/3 or HS/3
and H/2 or HS/2 up from the toe and H/2 or HS/2 up from the toe
• Resisting Forces • Resisting Forces
(Wf, W1, W2) at their respective (Wf,W1,W2,Pav,Pqv) at their
moment arm from the toe to respective moment arm from
each center of gravity as the the toe to each center of gravity
resisting moment. as the resisting moment

Note: Live load does not contribute to resisting forces. The live load surcharge is
included as a driving force and not as a stabilizing force. Only permanent forces
within the wall are included as stabilizing forces.
 Reinforced Wall Analysis

Movement

Movement

Bearing Capacity Settlement

Bearing and Settlement

are geotechnical issues
 External – Bearing Capacity / Settlement

• No calculation differences
• Differences in the Factor of Safety
– F.S. > 2.0 NCMA
– F.S. > 2.5 AASHTO ASD
– CDR > 1.0 AASHTO LRFD (For Bearing be careful as a reduction
factor of 0.65 is applied, which under ASD would be a 2.0 factor of
safety.)
• Settlement, particularly differential settlement should be
evaluated by a qualified engineer.
• Maximum allowable differential settlement for reinforced
soil systems
– 1% NCMA
– ½% FHWA
 Internal Stability Analysis

Internal stability is the ability of the reinforced mass to

maintain its structure and resist the applied loads without
deforming or failing.

• In soil reinforced wall system, it is the tensile and pullout

capacity of the reinforcing elements and inter-unit
shear/connection capacity that holds the potential wedge
of soil in place.
• The retained soil mass, or structure, is composed of the
Keystone units at the face combined with reinforcing
elements extending back beyond the Coulomb or Rankine
failure plane.
 Reinforced Wall Analysis

Units Active Wedge Passive Zone

Connection Tension Pullout

Internal Stability
Tension, Connection and Pullout
 Reinforced Wall Analysis

Overturning

Bending
Bulging

Shear

Local Stability
Overturning, Bending, Shear
 Reinforced Wall Analysis

Earth pressure
resisted by
top reinforcement

Earth pressure
resisted by
2nd reinforcement

Earth pressure
resisted by
3rd reinforcement

Earth pressure
resisted by
4th reinforcement

q a
Base Shear

Geogrid Load = (σq + σa) • Tributary Area ≤ Tal

Pullout Force Analysis

Hov

Pullout capacity of
top reinforcement
Le1

Pullout capacity of
2nd reinforcement
Le2

Pullout capacity of
3rd reinforcement
Le3

Pullout capacity of
4th reinforcement
Le4

Pullout = (2 Le)(γ Hov)(Tan φCi)

FS = Pullout / Geogrid Load
 Internal Stability Analysis

There are no differences in the internal formulas between Coulomb and

Rankine methods. The only difference is ASD vs. AASHTO LRFD and the
application of load and resistance factors in the internal calculations.

The Elements of Internal Design are to ensure:

1. The tensile elements do not exceed their working stress or factored
resistance limits.
2. The tensile element have adequate connection capacity to the
Keystone units.
3. The tensile elements have adequate anchorage beyond the potential
failure plane to hold the wedge of soil in place.
4. There is not a potential surface where the mass can shear internally.
5. The facing is stable against potential shear, bulging and overturning.
 Design Comparison

• Design Scenario
– Wall Height 20’
– Retained and Found. soil zone parameters 28°, γ=19 kN/m3
– Reinforced Zone Foundation 32°, γ=19.6 kN/m3
– Infinite backslope 2h:1v, (26°)
– Level Toe Slope
– Near Vertical Wall batter
– Compac II, Mirafi Geogrids
Design Comparison - Rankine
Design Comparison - Coulomb
 Design Comparison - Difference

Rankine Coulomb

Failure Plane
Design Comparison – Vertical Comp.
 Design Comparison - Battered

Rankine Coulomb
Design Comparison – Vertical Comp.
MSE Wall Design and Seismic
 Seed & Whitman (1970) Quotes

"While all investigators have concluded that the

dynamic lateral pressures developed during
earthquakes exceed the static pressures on earth
retaining structures, a survey of a number of
engineering companies highway departments and
port authorities in California shows that .... it is
general practice to make no special allowance for
increased lateral pressures on retaining walls ....
due to earthquake effects. This also appears to be
the case in many other countries."
 Seed & Whitman (1970) Quotes

"It should be noted that the factor of safety

provided in the design of walls for static pressures
may be adequate to prevent damage or
detrimental movements during many earthquakes.
(...) Thus where backfill and foundation soils
remain stable, it is only in areas where very strong
ground motions might be expected, for walls with
sloping backfills or heavy surcharge pressures and
for structures which are very sensitive to wall
movements, that special seismic design provisions
for lateral pressure effects may be necessary."
Classical Earth Pressure Theory

• Coulomb
• Rankine
• Mononobe-Okabe
 Formulas – Earth Pressure

Coulomb
Pa = 1/2 g H2 Ka

Sin2 (+)
Ka = 2
Sin2  Sin (-) 1 + Sin(+)Sin(-)
Sin(-)Sin(+)
Rankine Note: Backslope can not be
greater than phi angle
Pa = 1/2 g H2 Ka

Ka = Cos() Cos() - Cos2()-Cos2()

Cos() + Cos2()-Cos2()
Mononobe-Okabe (1929)
 Mononobe-Okabe (1929)

• Backslope MUST be < φ of retained zone

• Let’s look at the critical portion of the equation

sin(𝜑 − 𝜃 − 𝑖)

𝜑 = angle of friction of soil

𝐾ℎ
𝜃= 𝑎𝑟𝑐 tan( )
1−𝐾𝑣
𝑖 = backfill slope angle.
 Mononobe-Okabe (1929)

• Example
𝜑 = angle of friction of soil = 28°
𝐾ℎ
𝜃= 𝑎𝑟𝑐 tan( ) where, Kh=0.20g, Kv=0
1−𝐾𝑣
𝜃 = 11.3°
𝑖 = backfill slope angle = 3h:1v = 18°

sin(28 − 11.3 − 18) = sin (-1.3) = -0.023 / Neg. Value

Mononobe Equation doesn’t solve
 Mononobe-Okabe (1929)

• How are we working around this?

• When sin(𝜑 − 𝜃 − 𝑖) is negative this portion of the
equation is often set to 0
• Other forms of analysis is important
– Slope stability programs using cohesion
– Finite element analysis
– Displacement Method?
• The problem is that we are trying to solve something
that we don’t completely understand or are
attempting to put 2-D numbers to a 3-D solutions.
 Coulomb vs. Rankine Question

How do we use M-O -Coulomb seismic

analysis if Rankine earth pressure is the
prevailing theory for MSE wall design?

Coulomb = Rankine if delta angle is set

equal to backslope angle and batter is set
equal to zero.
 MSE Seismic Analysis

khW2
khWs
W2(1-kv)
Ws(1-kv)
khW1 H2
H
 or 
W1(1-kv) 
Eae

L
Inertial Force Active Wedge Force

Neglecting inertial forces

is a common mistake
MSE Seismic Analysis Application

khW

Effective Edynamic@ 50%

H' zone Ev
 or 
H2 Eh
khW1 Estatic
khW Ev
H Eh
W1 0.6H2
HÕ
/3
cL
e
d R
.5H2 Tan  x 0.5H
L H2 = H +
(1 - 0.5 Tan)

Kdynamic = Kae - Ka
 Seismic MSE Internal Analysis

 0.3H

AmW3
Am W3

AmW2 W3 W3
HÕ
/2 Am W2
H'
Am W1 W2 AmW1
H H W2

W1 HÕ
/2 W1 Le
Le
45+/2

d d
B B
Extensible Reinforcement Inextensible Reinforcement

Le of reinforcement level
Add'l Load/reinf = (Pi)
Sum of Le for all reinforcement levels

Pi = Am (W1 + W2 + W3)
 Displacement Analysis

The Mononobe-Okabe equation for earth

pressure is still used widely for design,
although actual conditions during
earthquake shaking of retaining structures
are quite different from those assumed in
developing the equation…

The proposals of Richards and Elms have

simulated use of design methods based
upon allowable permanent displacement.

Robert V Whitman (1990)

Displacement Analysis

Displacement Controlled Design

Richards and Elms (1979/1990)
Displacement Analysis
 Displacement Analysis

0.087 V2 1/4
N=A [ dAg ]
Where:
N = Design Cutoff Acceleration

A = Peak Design Acceleration

V = Peak Velocity
d = Allowable Displacement
g = Gravitational Acceleration

Maximum Acceleration, Kh
Richards and Elms
 Displacement Analysis

kh = 0.74 As [
1/4
]
As
d
Where:
kh = Horiz Acceleration Coeff.

As = Design Acceleration
d = Allowable Displacement (in)
(1” - 8” range)

Simplified Acceleration, Kh
Kavazanjian et al.
 Displacement Analysis Summary

“For most design purposes, it has been shown (Elms

and Martin, 1979) that a design value of Kh =
0.50A is adequate, provided that the wall can
accommodate an outward displacement of up to
about 250A mm”(10A in inches).

Geotechnical Earthquake Engineering by the

Federal Highway Administration (1998)
 What PGA, A, or kH to use?

What is the design earthquake?

• 2% probability in 50 years
• 10% probability in 50 years
• 40% probability in 50 years
• 7% probability in 75 years
• All or some of the above

This is an Owner driven criteria based on the

importance of the structure.
 New Zealand PGA Determination

• NZ Transportation Agency’s Bridge Manual

– 6.0 Site stability, foundations, earthworks and retaining walls
• Loads determined from section 6.2.2
• Other NZ design references for MSE walls
– Road Research Unit Bulletin 84
• Provides a basis for seismic design
• Shall be compiled with
– NZTA research report 239
• Provides additional guidelines
NZ Bridge Manual Section 6.2.2
NZ Bridge Manual Section 6.2.2
Addendum 6A Table 6A.1
NZ Seismic Maps
 NZ Bridge Manual Section 6.6

• 6.6 Earth Retaining Systems

– Numerous codes can provide guidance
 NZ Bridge Manual Section 6.6

• 6.6.9 Design performance of earth retaining structures

and slopes
– “Retaining structures and slopes may be designed to remain
elastic under the design earthquake load specified in 6.2.2 or
to allow limited controlled permanent outward displacement
under strong earthquake shaking.”
– “The displacement likely at the design ultimate limit state
seismic response, and under the MCE (maximum considered
event), shall be assessed using moderately conservative soil
strengths consistent with the anticipated strain and
Newmark Sliding Block displacement approach.”
Section 6.6.9
 Current US Highway Seismic Criteria

Retaining walls around bridges have been

designed for a 2% exceedance in 50 year event.

Retaining walls away from bridges have been

designed for a 7% exceedance in 75 year event.

Note: The PGA of a 7% probability event is around

half of a 2% probability event.
 Is Seismic Analysis Necessary for MSE Walls?

11.5.4.2 AASHTO – Extreme Event I, No Analysis

“A seismic design shall not be considered mandatory for

walls located in Seismic Zones 1 through 3, or for walls at
sites where the site adjusted peak ground acceleration,
As, is less than or equal to 0.4g, unless one or more of
the following is true:”

*AASHTO LRFD Bridge Design Specifications, 7th Edition, 2014, 2015 Interims.
 Is Seismic Analysis Necessary for MSE walls?

Cont’d
• Liquefaction induced lateral spreading or slope failure, or
seismically induced slope failure, due to the presence of
sensitive clays that lose strength during the seismic
shaking, may impact the stability of the wall for the design
earthquake.
• The wall supports another structure that is required based
on the applicable design code or specification for the
supported structure to be designed for seismic loading and
poor seismic performance of the wall could impact the
seismic performance of that structure.

*AASHTO LRFD Bridge Design Specifications, 7th Edition, 2014, 2015 Interims.
 Is Seismic Analysis Necessary for MSE Walls?

Cont’d
• In Seismic Zones 2 and 3
– Exposed wall height plus average surcharge depth is > 30’
– Tiered walls the sum of the exposed height of all the tiers
plus the average soil surcharge depth is > 30’
– The wall has abrupt changes in its alignment geometry (e.g.,
corners and short radius turns at an enclosed angle of 120
degrees or less)
– For gravity and semi-gravity walls, the wall backfill does not
meet the requirements of Article 7.3.6.3 of AASHTO.

*AASHTO LRFD Bridge Design Specifications, 7th Edition, 2014, 2015 Interims.
 Why the change?

Observed Seismic Performance for Walls

• Good performance of MSE Walls in Seismic Events

– 1995 Kobe Earthquake; masonry and concrete gravity walls
collapsed due to weak soils, heavy soil surcharges, or
structural failure, mainly where As>0.6g; MSE Walls had
some damage but did not collapse even up to 0.8g
– 1999 Izmit Earthquake (As>0.40g); Rigid Structure Collapse
MSE Structures remained in place
– 2001 San Salvador Earthquake (As>0.30g); Example wall
shown
1995 Kobe Earthquake Japan

GRS Wall - 1992 Before

Tatsuoka et al.
1995 Kobe Earthquake Japan

GRS Wall - 1995 After

Tatsuoka et al.
 1995 Kobe Earthquake Japan

GRS Wall - After RCW Wall - After Tatsuoka et al.

1999 Izmit Turkey Earthquake

Mark Aschheim et al.

1999 Izmit Turkey Earthquake

Mark Aschheim et al.

1999 Izmit Turkey Earthquake

Mark Aschheim et al.

 Why the change?

Laboratory Seismic Studies

• Two seismic forces are out of phase
– Dynamic earth pressure was at its maximum, the wall inertial force was
at its minimum or very close to 0
– When the wall inertial force was at its maximum, the total seismic earth
(Pae) was close to its static value.
• Nakamura, S. (2006). “Re-examination of Mononobe-Okabe theory of gravity
retaining walls using centrifuge model tests.” Soils Foundation 46(2), 135-146
• Al Atik, L., and Sitar, N. (2010). “Seismic earth pressures on cantilever retaining
structures.” Journal of Geotechnical and Geoenvironmental Engineering, 136(10),
1324-1333
• Seismic earth pressures appear to not develop until As>0.4g
• Al Atik, L., and Sitar, N. (2010). “Seismic earth pressures on cantilever retaining
structures.” Journal of Geotechnical and Geoenvironmental Engineering, 136(10),
1324-1333
 Why the change?

Seismic Earth Pressures on Cantilever Retaining

Structures
Linda Al Atik and Nicholas Sitar, 2010
Abstract in the Journal of Geotechnical and
Geoenvironmental Engineering – ASCE

Laboratory Centrifuge Experiments

 Al Atik and Sitar (2010)

*Al Atik and Sitar (2010)

 Why the change?

*Al Atik and Sitar (2010)

 Al Atik and Sitar (2010)

Dynamic Earth Pressure near maximum

Seismic Behavior of Wall –
Backfill System

• Comparison of dynamic
moment increments,
dynamic earth pressure
increments and wall
inertial forces.

Wall mass inertial force near zero or negative

 Al Atik and Sitar (2010)

Observations and Interpretations

“…when the inertial force acts in the active direction, the

total earth pressure is equal to or less than the static
earth pressure. Dynamic earth pressure increment is at
its maximum when the inertial force is close to zero (i.e.
static case) or when the inertial force acts the passive
direction.”
 Al Atik and Sitar (2010)

Observations and Interpretations

“In contrast, the limit equilibrium assumption inherent in

MO theory means that the earth pressure increases
when the inertia force is loaded in the active direction
and stability analyses of retaining are usually conducted
for maximum dynamic earth pressures and inertia
forces.”
 Al Atik and Sitar (2010)

Shows earth pressure distribution is triangular, indicating resultant at h/3 and less
than 65% of M-O earth pressure.
 Conclusions

1. The experimental and numerical analysis results

consistently show that the maximum dynamic earth
pressures increase with depth and can be reasonably
approximated by triangular distribution analogous to
that used to represent static earth pressures.
Consequently, there seems to be no basis for the
currently accepted position of the point of application
the dynamic earth pressure force in dynamic limit
equilibrium analyses at 0.6 to 0.67 H and, instead, the
point of application should be at 1/3H, as originally
suggested by Mononobe and Matsuo (1932).

*Al Atik and Sitar (2010)

 Conclusions

2. An important aspect of the dynamic interaction

between the cantilever retaining walls and retained
soils is the fact that the maximum dynamic earth
pressures and maximum wall inertial forces to not
tend to occur simultaneously. As a result, the current
design methods based on the MO theory were found
to significantly overestimate the recorded dynamic
earth pressures and moments.

*Al Atik and Sitar (2010)

 Conclusions

3. The relationship between the back-calculated seismic

earth pressure increment coefficient (∆Kae) at the
time of maximum dynamic wall moment and peak
ground acceleration obtained from our experiments
suggests that seismic earth pressures on cantilever
retaining walls can be neglected at accelerations
below 0.40g.

*Al Atik and Sitar (2010)

 Conclusions

4. The analytical results show that the FE analysis is able

to capture quite well the essential system responses
observed in centrifuge experiments. However, the
veracity of the numerical analyses is strongly
dependent on access to high quality experimental or
field performance data for model calibration and,
therefore, field performance predictions using
numerical models should be approached with
caution.

*Al Atik and Sitar (2010)

 What does this mean?

• Take everything with a grain of salt

• Theory is that M-O Seismic design is overly
conservative
• MSE Walls that have been designed for Static
Conditions at:
– Reinforcement Lengths = 0.7 * H
– AASHTO Select backfill
– Proper vertical spacing and strengths design for internal
stability
Will perform well in seismic applications under 0.40g
without Seismic design.
 What does this mean?

• Does this mean that seismic design can / should be eliminated below
0.40g?
– Use your judgement
– Slopes are still a big issue, especially high backslope low friction angle soils
– Likely 20 states or so have A > 0.40g
– Most seismic states will require a design even if it is below 0.40g.
Kho = Fpga x PGA = As, For PGA > 0.50g -> Fpga = 1.0, site class B, C, D
Kh = 0.5 x Kho (or As)
• The below 0.40g criteria was broadcast generally to all wall types
including gravity and semi-gravity.
– The idea that segmental gravity walls including large gravity block walls (no
geosynthetic reinforcement) can withstand seismic below 0.40g without
toppling over seems aggressive. Especially considering the large rigid
structures failing in seismic conditions.
• Review counter points
– Leshchinsky, D., and Vahedifard, F., and Shahrokhabadi, S., “Does No-seismic
Design in AASHTO violate AASHTO’s rules?” 2015 Geosynthetics Conference,
February 15-18, Portland, OR
 What does this mean?

*Leshchinsky et. al., (2015)

 Keystone Resources

• Design Software
– KeyWall®
– Excel spreadsheets
• Technical Notes
• Standard Drawing Details
• Design Manual
• Construction Manual
• Specifications
 Speaker Profile

Dan Tix, P.E.

Engineered Structures Technical Manager

Keystone Retaining Wall Systems, LLC

Dan has been working in the segmental retaining wall industry for 12 years. Dan
has over 9 years of MSE Wall design experience working in Keystone’s engineering
department. His current position is Engineered Structures Technical Manager,
where he coordinates and manages Keystone’s DOT product submittals, is
responsible for managing structural wall product offerings nationwide and teams
with Keystone’s region managers and international partners to provide technical /
engineering support for project sales and promotion. Prior to working for
Keystone, he worked 2 years as a geotechnical consulting engineer in Michigan.
Dan graduated from Michigan Technological University with a B.S. in Civil
Engineering in 2001. He is a licensed Professional Engineer in the state of
Minnesota since 2007.