Wafflemat System Design Considerations

start SMART
build SMART
stay SMART
DESCRIPTION
The Wafflemat System is a variation of the typical post-tensioned ribbed
foundation, with some significant advantages. It possesses the greatest
floor stiffness of any system in its class, with sufficient strength to resist
differential swelling resulting from landscaping practices, surface d ra i n a g e,
or flooding. In addition , the Wafflemat System does not re q u i re pre s o a k i n g
u n d e rlying soil pads, and there is no need for additional footings.
Over 6.5 million square feet of Wafflemat System slabs have been poured
since 1993, without a single failure or structural callback.
The Wafflemat System is intended for use in moderate to cri t i ca lly expansive
soils. The design calculations for a Wafflemat System slab foundation
adhere to the PTI method, with minor modifications described herein.
Dimensions & materials
The individual Wafflemat form (Wafflebox) is a 19 inch square, with a
height of 8.5 inches. It is made from injection-molded polyethylene. The
forms are designed to be used in arrays of up to four forms per void unit,
typically with 6-wide beams spaced 3-8 on center. Wafflemat System
foundations are formed and poured on grade, with no excavation required.
DESIGN FEATURES
The simplest foundation for residences is slab-on-
grade. In perfect soil conditions, a relatively thin steel-
reinforced concrete pad is adequate for a residential
foundation. However, in many areas the land available
for new housing construction is poor quality, and builders
have i n c re a s i n g ly relied on thicker pads to ach i eve
a c c e p t a b l e performance. In many instances, builders have
taken to excavating poor quality soil and replacing it with on ly
m a r g i n a lly better fill in order to improve site soil conditions.
Increasing the thickness of the concrete pad involves increased costs
for materials and handling. Excavating and refilling takes time, labor,
and energy, and often results in a tremendous volume of spoils that must
be disposed of.
The original intent of the Wafflemat System was to devise a ground-level
foundation that could be used on less than optimal soil, and that would
save time and money in materials, labor, and ground preparation. After it
was designed and implemented, other advantages became apparent.
The concept of waffle form floors is not new. Waffle slabs have been used
for years in above-ground floors, where they are preferred because of their
s t i f f n e s s , s t a b i l i ty, and stre n g t h . A b ove - g round waffle slabs are tra d i t i on a lly
poured using expensive, reusable forms that must be disassembled,
cl e a n e d , and pre p a red for the next pour. This would obv i o u s ly be
i m p ra c t i ca l for slab-on-grade construction.
One approach to making waffle slabs-on-grade is to carve waffle forms
into the soil. This involves tedious excavation and is highly dependent on
prevailing weather conditions.The subsequent waffle grid is vulnerable to
crumbling, backfilling, and damage or flooding from precipitation.
In Australia and New Zealand, builders have been pouring on-ground
waffle raft foundations for 1 and 2 story buildings for many years. The
waffle raft is structurally equivalent to the Wafflemat System, except the
voids are formed using solid expanded polystyrene blocks, or pods. The
popularity of waffle rafts persuaded the Australian Standards Committee
to specifically include them in their building code (AS 2870) in 1988.
ADVANTAGES
While the waffle raft is a good idea, the Wafflemat System is even better.
For instance, waste polystyrene is difficult to recycle, whereas the
Wafflemat System uses polyethylene, which is 100% recyclable. Also, the
p o lys tyrene p o d s take up a lot of space because they are solid.
Transportation and on-site storage of Wafflemat System components is
much more efficient because the forms can be delivered in smaller loads
for the same void volume, and stacked inside one another on pallets until
needed. And, in windy conditions, a stack of Wafflemat forms is much
less likely to blow away.
Other advantages of the Wafflemat System are revealed during the form-
ing process. Wafflemat forms are strong enough to be walked on, which
makes layout go faster, without risking damage to the forms. Also,
Wafflemat forms fit together with clips which act as spacers for beams
and supports for reinforcing tendons or steel.The Wafflemat System is an
integrated package specifically designed to reduce labor costs and time.
One feature of the Wafflemat System that offers unique advantages over any other slab
foundation system is the integrated void spaces under the concrete pad. There are three
major aspects. First, since the void forming components are water-tight polyethylene,
they provide an excellent moisture barrier. In soil with high water content, water cannot
leach up through the bottom of the concrete pad. This is significant for protecting the
structure from water damage and mold.
Second, the void spaces in the Wafflemat System accommodate differential
soil expansion without excessive uplift force on the slab itself. In expansive
soils, edge or center uplift is a major cause of foundation failure. A typical
P/T slab on grade (ribbed or uniform thickness) is subject to the entire uplift
force, and can be prone to cracking and failure. The Wafflemat System, on
the other hand, is designed to allow expanding soil to flow upwards into the
void spaces. Since the entire weight of the foundation and structure rests on
the bottom surface of the beams, the uplift force from expanding soil on the
foundation is considerably less, because total uplift force is a function of the
a rea of the foundation in direct contact with the soil. In addition , t h e
concentrated point loading of the entire structures weight onto the surface
area of the beams increases resistance to soil swelling.
Third, the void spaces can be interconnected and vented. This is useful in
areas where hydrostatic pressure is significant, because water would be free to
collect in the void spaces, and then be pumped or allowed to drain out from
under the foundation . A l s o, for basements and areas where ra d on gas is a
c on c e rn , the interc onnected polye t hylene-lined void spaces could act as
gas-impermeable collectors that would then be safely vented to atmosphere.
FOOTINGS AND CUTOFF WALLS
Post-tensioned concrete foundations, designed in accordance with the PTI
method, qualify as designed footings in accordance with the IBC 2003
Chapter 18, Soils and Foundations, and UBC 1997 Chapter 18, Foundations
and Retaining Walls. As such, post-tensioned slabs-on-ground are exempt
f rom UBC 1997 Se c t i on 1806.7.2. [ Po s t - Te n s i oning Institute, 2 0 0 4 ,
Design of Post-Tensioned Slabs-on-Ground, Third Edition, Phoenix, AZ]
Additional footings are not necessary with the Wafflemat System, and are not
recommended.
Vertical moisture barriers may be used to reduce the soil support parameters
( e
m
and y
m
) . Su ch barriers must be designed to virt u a lly stop moisture migra t i on
to or from the foundation area on a permanent basis, around the entire
perimeter of the foundation.
SMART SMART SMART start build stay
An embedded vertical moisture-impermeable membrane such as visquene should be used where
impedance to moisture migration is desirable. The depth that the vertical barrier must extend below
adjacent ground surface in order to be effective should be determined by the soils engineer.
In cold cl i m a t e s , f ro s t - p rotected shall ow foundation design (FPSF) techniques using rigid foam
insulation have proven effective for preventing frost heave.
Conversion for PTI calculations
OVERVIEW
Section 1816 of the 1997 Uniform Building Code describes the structural
design procedure for slabs on expansive soils based on specifications from the
Post-Tensioning Institute (PTI). Following is the re c ommended pro c e d u re for
designing Wafflemat System foundations according to PTI methodology.
PTI recommendations for ribbed slab design include spacing limitations for the
ribs and limits on the rib depth and width.
Rib spacing S shall be a maximum of 15 ft. A minimum rib spacing
S of 6 ft shall be used in the design of ribbed slabs; however the actual
spacing may be less than that if desired. (PTI Section 4.5.2.1)
...the total rib depth h shall be in no case less than 11 in., and the rib
must extend at least 7 in. below the bottom of the slab ( h " t +7in.)
( PTI 4.5.2.2 ) & the rib width used in section property calculations
must be limited to a range of 8 to 14 in. ( PTI 4.5.2.3 )
The Wafflemat system uses forms to create a waffle slab with a set geometry, with ribs typically spaced
at 3-8 on center. Calculations showed that the Wafflemat System is equivalent to a post-tensioned
slab on grade that complies with PTI re q u i re m e n t s , with ribs spaced at 6- 0 on center. T h e
p ro p e rt i e s and post-tensioned tendon layouts for the Wafflemat slab were similar to the PTI Ribbed
slab. The method d e s c ribed below complies with the re q u i rements of the PTI methodology and is the
design adaptation of the PTI method to the Wafflemat system.
The design process involved creating a design for the Wafflemat slab and another for the Ribbed slab,
using guidelines for selecting initial values of geom e t ry of the slab. If either of these designs exc e e d e d
a ll owable stress v a l u e s , then an itera t i ve process was used to bring the slab(s) into com p l i a n c e . Both
designs had to com p ly with all owable values and must also have had similar geom e t ric section pro p e rt i e s
and pre s t ressed tendon qualities (effe c t i ve prestress, tendon locations, etc.)
PTISlab, version 2.06, from Geostructural Tool Kit, Inc (GTK) was used to
perform structural calculations for a Wafflemat System and its equivalent
PTI-conforming design. The results are presented below.
(This design process should be con s i d e red on ly an aid to perf o rming numeri ca l
calculations. In no event will Pacific Housing Systems or MKM &
Associates, its officers, owners, or employees be liable to anyone for any
unfavorable conditions occurring from the use of this design methodology.
Engineering judgment must be used to interpret the results of this design
methodology. The engineer-of-record for a project should thoroughly
rev i ew the results and must take re s p on s i b i l i ty for the use of the final
calculations.)
Materials Properties
Concrete
Compressive strength was set to 4500 PSI. (Minimum 2500 PSI is
recommended for PTI method designs.)
Unit Weight was set to 145 PCF
Tendons
1/2 diameter, 270 KSI tensile strength
Slab Properties
Rectangle Geometry
Slab thickness was initially set at 5.
A rectangular slab footprint (34 ft x 45 ft) was used.
Slab Reinforcement
A minimum prestress of 50 psi was selected.
(This is the minimum prestress value recommended by PTI.)
b
w
, b
r
s
w
, s
r
Figure 1 - Slab and Beam Dimensions.
Beam Properties
The number of beams in each direction should be the slab dimension
divided by rib spacing, plus one.That is, the number of 3-8 or 6-0 wide
spaces (depending on which analysis is being performed) that can fit into
the slab, plus one, to reflect that there is also a rib at each edge.
The depth of the beams is the slab thickness plus the depth of the
Wafflemat form, for example: 5 + 8.5 = 13.5.
The width of the interior beams was specified as 6 for the Wafflemat
System slab, and 9.8 for the PTI Ribbed Slab. The relationship between
the beam width (b
r
) and beam spacing (s
r
) of the Ribbed analysis and the
beam width (b
w
) and beam spacing (s
w
) of the Wafflemat analysis is
The width of the edge beams was set to 12, the typical minimum width
for edge beams in the Wafflemat System.
One tendon was placed in each beam.
The beam tendon cover (from the bottom of the beam to the tendon) was
i n i t i a lly set to 9 to reflect default tendon positioning in the Wa f f l e m a t
System. (If design analysis required repositioning of the tendons, the beam
tendon cover setting was adjusted accordingl y.)
b
w
s
w
b
r
s
r
=
.
b
w
s
w
b
r
s
r
=
Initi al Desi gn Parameters
The foll owing table provides suggestions for initial values to be entered based on the
a n a lysis type and the Wafflemat form depth. These were the initial values used in the
e x a m p l e analysis. If either of the designs do not work using these parameters (see the
RESULTS section), modify the geometry of the system until design requirements are met.
M o d i f i ca t i ons may include changing the slab thick n e s s , number of beams, number of
t e n d on s , or tendon cove r. If the number of beams (beam spacing) and beam widths are
e d i t e d ,the relationship must still be satisfied.
ANALYSIS SLAB DEPTH OF BEAM BEAM TENDON BEAM
THICKNESS WAFFLEMAT DEPTH, WIDTH, COVER SPACING
[ IN ] FORMS b s
[ IN ] [ IN ] [ IN ] [ IN ] [ FT ]
RIBBED 5 8.5 13.5 9.8 9 6
WAFFLEMAT 5 8.5 13.5 6 9 3.67
Soil Properties
Actual values from a geotech n i cal re p o rt were used for modera t e, h i g h , and cri t i ca lly expansive
soil types. In addition, the critically expansive soil condition was enhanced by increasing the
Edge Moisture Variation Distance (e
m
) in order to challenge the default design. The Soils
parameters were as follows:
MODERATELY HIGHLY CRITICALLY INCREASED
EXPANSIVE EXPANSIVE EXPANSIVE e
m
ALLOWABLE BEARING 2000 2000 2000 2000
VALUE [ PSF ]
MODULUS OF SUBGRADE 100 100 100 100
REACTION [ PCI ]
EDGE LIFT [ y
m
] 0.5 1.0 2.5 2.5
CENTER LIFT [ y
m
] 0.375 1.0 3.6 3.6
EDGE MOISTURE 2.5 2.5 2.5 3.2
VARIATION DISTANCE [ e
m
]
CENTER MOISTURE 4.75 4.75 4.75 6.4
VARIATION DISTANCE [ e
m
]
(When identifying the edge Moisture Variation Distance, e
m
, and the Differential Soil
Movement, y
m
, verify with the geotechnical engineer that these values have been derived
using the appropriate edition of the PTI methodology. In this example, the 2nd edition was
used.)
SMART SMART SMART start build stay
Load, Deflect & Prestress
Loading conditions, deflection criteria, subgrade friction selections, and prestress loss values
that were used are shown in the table below. (For more information on these values, see the
PTI publications.)
U n i f o rm Superimposed Load: 40 psf
Perimeter Total Load: 1000 lb/ft
Edge Li ft minimum load: 500 lb/ft
Center Lift Deflection: L / 3 6 0
Edge Li ft Deflection: L / 7 2 0
P re s t ress Loss: 15 ksi
Line Load (long di re c t i o n ) : 500 l b/ft.
RESULTS
Design Compliance
Both the Ribbed and Wafflemat designs worked for the moderate, high, and critically
expansive soils conditions without modification. Some adjustments were necessary in order
to get the designs to work for the critically expansive soil with increased Edge Moisture
Variation Distance (e
m
).
Simple adjustments for increasing slab design performance include reducing beam tendon
cover (lowering the beam tendons), increasing slab thickness, adding more slab tendons, and
thickening edge beams.
The table below shows the initial designs, and modifica t i ons re q u i red to meet the
design cri t e ria for cri t i ca lly expansive soil with increased e
m
. In this ca s e, l ow e ring the
beam tendons 2 (Ribbed design) or 1/2 (Wafflemat design) was sufficient.
RIBBED RIBBED WAFFLEMAT WAFFLEMAT
MODIFIED MODIFIED
SLAB THICKNESS 5 5 5 5
# SLAB TENDONS, 3 3 0 0
SHORT DIRECTION
# SLAB TENDONS, 3 3 0 0
LONG DIRECTION
SLAB TENDON, 2.25 2.25 - -
COVER, SHORT DIRECTION
SLAB TENDON, 2.25 2.25 - -
COVER, LONG DIRECTION
# BEAM, 9 9 13 13
TENDONS, SHORT DIRECTION
# BEAM, 7 7 10 10
TENDONS, LONG DIRECTION
BEAM TENDON 9 7 9 8.5
COVER, SHORT DIRECTION
BEAM TENDON 9 7 9 8.5
COVER, LONG DIRECTION
As can be seen from these values, although both designs were similar, the Wafflemat slab
re q u i red less adjustment than the Ribbed slab in order to meet the design re q u i rements for
c ri t i ca lly expansive soil with increased e
m
. C ove r s e ly, g i ven the same modifica t i on s , a
Wafflemat System design would be expected to alw ays perf o rm at least as well as the
c o r re s p onding Ribbed design.
COMPARING RESULTS
COMPARING RESULTS
Section Properties
The Se c t i on pro p e rties of the two slabs were similar in both the long and the short dire c t i on s ,
as shown in the following table:
RIBBED - SHORT WAFFLEMAT - RIBBED - LONG WAFFLEMAT -
DIRECTION SHORT DIRECTION DIRECTION LONG DIRECTION
CROSS SECTIONAL AREA (IN
2
) 3,487 3,465 2,661 2,652
MOMENT OF INERTIA (IN
2
) 38,132 37,391 29,664 29,384
SECTION MODULUS, 9,477 9,371 7,281 7,242
TOP (IN
2
)
SECTION MODULUS, 4,024 3,932 3,147 3,112
BOTTOM (IN
2
)
BETA DISTANCE, 7.25 7.21 6.81 6.79
(RELATIVE STIFFNESS
LENGTHS) (FT)
EFFECTIVE PRESTRESS, 296.65 323.54 250.04 250.18
FORCE AT BETA (KIPS)
EFFECTIVE PRESTRESS, 85.07 93.37 93.98 94.34
AT BETA (PSI)
Similarity in the cross-sectional areas, moments of inertia, and section modulus (top and bot-
tom) are essential for making valid comparisons between the Ribbed and Wafflemat designs.
In this case, the differences between the Ribbed design and the Wafflemat design were all
within 2.5%.
Stress Analysis
By comparing the calculated stresses on each design, it can be shown that the Wafflemat slab
is similar to the PTI-compliant Ribbed design. The table below shows results for both
designs applied to moderately expansive soil conditions.
RIBBED - SHORT WAFFLEMAT - RIBBED - LONG WAFFLEMAT -
DIRECTION SHORT DIRECTION DIRECTION LONG DIRECTION
MAX. MOMENT, 3.57 FT-K/FT 3.10 FT-K/FT 3.41 FT-K/FT 2.96 FT-K/FT
CENTER LIFT
DIFFERENTIAL DEFLECTION 0.35 IN. 0.22 IN. 0.36 IN. 0.23 IN.
CENTER LIFT
SHEAR STRESS, 36 PSI 30 PSI 21 PSI 15 PSI
CENTER LIFT
MAX. MOMENT, 1.15 FT-K/FT 1.10 FT-K/FT 1.24 FT-K/FT 1.19 FT-K/FT
EDGE LIFT
DIFFERENTIAL DEFLECTION 0.13 IN. 0.08 IN. 0.14 IN. 0.09 IN.
EDGE LIFT
SHEAR STRESS, 37 PSI 38 PSI 36 PSI 37 PSI
EDGE LIFT
Note that the Wafflemat slab showed less differential deflection in both center lift and edge
lift modes. This is consistent with the increased stiffness characteristic of the Wafflemat
design, and further validates the use of the modified PTI method for designing Wafflemat
System foundations.
CONCLUSIONS
When the section properties for the Wafflemat and Ribbed slabs are equivalent, PTI
methodology can be used in the design of the Wafflemat slab. And, with over 6.5 million
square feet of residential living space poured since 1995 without a single foundation failure,
the Wafflemat System has proven to be a robust and reliable alternative to the standard post-
tensioned slab foundation.
APPENDIX - Wafflemat System Foundation
Design Details
Drawing 1 | Typical Foundation Plan.
The small squares inside the foundation perimeter re p resent the Wa fflemat void units, made up of
one, two, three, or four individual Wa fflemat forms (Wa ffleboxes). The standard beam width
between forms is 6 inches. Foundation plan dimensions will often re q u i re wider interior beams for
reasons of spacing the void forms. The wider beams should be positioned under interior load-bearing
p a rtitions, if any, to take advantage of the extra support .
Detail 1 | Typical Wa fflemat System Perimeter Cro s s - S e c t i o n .
Post-tensioning tendons are located in the beams at about the same
elevation as the top of the void form s .
F I N I S H
G R A D E
P A D
G R A D E
V O I D
T Y P I C A L
6
8 3 - 0
S E E
P L A N
3 - 2
10 MIL VISQUENE
1/2 DIA.PT TENDON
TOP OF
SLAB
2-#4 BACK-UP BARS
(CONT. L.E. 4 D.E.)
Detail 2 | Garage Entry Cro s s - S e c t i o n .
The Wa fflemat garage floor is cast on top of the pad grade.
The driveway is added later, on top of the finished grade.
Detail 3 | Typical Reinforcement for a Prefabricated Metal Fire p l a c e .
A heavy masonary fireplace will normally have its own separate foundation.
Detail 4 | Typical Reinforcement for a Prefabricated Metal Fire p l a c e .
If additional strength is needed, fill the space with steel-re i n f o rced concre t e .
EXPANSION JOINT
P A D
G R A D E
DRIVEWAY SLAB
(BY OTHERS)
CURB BEYOND
S E E
P L A N
6
V O I D
T Y P I C A L
S L O P E
3 - 2
DETAIL AT GARAGE ENTRY
POST & BEARING WALL ON SLAB
P A D
G R A D E
#3 X 3 - @ 12
O.C. FULL LENGTH
OF WALL IF LOAD < 500
2 #3 CONT. @ 12 O.C.
UNDER WALL
FILLED W/CONC.
IF LOAD E
#4 X 3 - @ 12 O.C.
EACH WAY U.O.N. ON PLAN
FILLED W/CONC. FOR LOADS
> 5 0 0
REMOVE WAFFLE & FILL
WITH CONCRETE @ POST
ONLY. NOT REQ D. AT
BEARING WALL U.O.N.
ON PLANS
S E E
P L A N
#5@12 O.C. EACH
WAY, TOP & BOT.
F I N I S H
G R A D E
P A D
G R A D E
#5 X8 - @ 12 O.C.
TOP.
Detail 5 | Bearing Wal l Over Beam.
R e i n f o rce beams directly under load-bearing interior walls with re b a r.
Detail 6 | Brick Ledge Integrated Into the Wa fflemat Foundation.
BEARING WALL ON RIB
1-#5 FULL LENGTH
OF WALL
2X SILL PLATE
5/8 DIA. AB S @ SPACING
SHOWN ON PLAN.
1/2 DIA. TENDON
P A D
G R A D E
BRICK OR
STONE VENEER
1 -
#4 @ 12 O.C.
1 -
F I N I S H
G R A D E
V O I D
T Y P I C A L
P A D
G R A D E
S E E
P L A N
3 - 2 6 6
# 4
LEDGE AT PERIMETER
Detail 7 | Clearance for Slab Opening.
Recommended minimum clearance for an opening in the slab between tendons is 6 inches.
N o rm a l l y, the void form is removed from the location, so as to not obstruct the opening.
TYPICAL TENDON LAYOUT
AT CORNER
6 MIN.
10 MAX.
EDGE OF
S L A B
1 / 2 0 / T E N D O N
Detail 8 | Typical Tendon Layout at Corn e r.
Edge beams are usually 12 to 18 inches wide. The outermost tendons are placed
6 to 10 inches from the outer edges.
PLAN-MIN.CLEARANCE FOR
AN OPENING IN SLAB
1/2 TENDON
6 MIN.
SLAB ON
G R A D E
O P E N I N G