This article is a continuation of the Whats This Report For?

series, based on a technical session sponsored

by ACI Committee E702, Designing Concrete Structures. In keeping with ACIs mission to provide knowledge and
information for the best use of concrete, the articles will be posted on the ACI Web site (www.concrete.org/
education/edu_online_CEU.htm) and, along with sample reports and multiple-choice questions, be used for
educational materials.

The Floor Flatness

Report
What the designer needs to know

by Mark A. Cheek

loor profile finish quality has traditionally been

specified by limiting the gap under either an unleveled
or leveled 10 ft (3 m) straightedge. Some specifications
still take this tolerance approach, even though there
is no nationally accepted standard either for taking
measurements or for establishing compliance of a floor
profile. In many specifications, slab finish quality is not
addressed at all. Use of a nonstandard test procedure and
failure to specify floor profiles often lead to conflict and
litigation. For example, if the project specification calls
for a 10 ft (3 m) straightedge to determine the quality of
the finished floor and the test area is 100 x 100 ft (30 x
30 m), a technician can place the 10 ft (3 m) straightedge
at a single location and measure the gaps between it and
the floor. Operating without a standard, the technician
could simply use the measurement from that one location
as representative of the entire test area. It may (by
some chance) be representative of the whole floor
but most likely is not; thus, the results obtained are
essentially useless.

Profiling Standard

The technology for measuring floor profiles has

developed in response to the need for a standard method
to evaluate them. This technology, called the F-number

system,1 provides a welcome alternative and a solution to

the generally recognized inadequacies of the 10 ft (3 m)
straightedge to describe and define floor profiles.
Floor flatness (FF) and levelness (FL) numbers
determine whether a floor is sufficiently smooth and
level, respectively, as constructed. Floor flatness can
affect flooring installation, ride quality and safety in
warehouses, and drainage. Floor levelness can affect
shelf placement and design and a slabs drainage plan.
For example, the levelness of the floor in a warehouse
could limit how high pallets of goods can be safely
stacked. Typical FF and FL values for different applications
are shown in Table 1 and Fig. 1.

Appropriate flatness and levelness

As in any other specification, the engineer should

determine what is good enough for the application rather
than impose a standard that is unnecessarily exacting
and costly. Both overall flatness and levelness numbers
should be specified, along with the local minimum values
(typically 60% of the overall numbers). The test should
be conducted within 72 hours of finishing the slab, as the
curing process could cause the slab to curl and deviate
from the flatness achieved by the finishers. Obtaining
these numbers within 72 hours also allows the contractor
Concrete international

/ January 2011

35

to make adjustments to the procedures, if necessary, while

the floor is still being placed. Both flatness and levelness
numbers can be determined on shored decks, but only
flatness numbers can be determined on unshored decks.

Measurement

The FF number is an indication of how bumpy or wavy

the slab surface is, demonstrating the quality of the initial
strike off and finishing process. The F-number system
uses floor surface curvature calculated from elevation
differences over 24 in. (600 mm) increments as a measure
of flatness (Fig. 2). The FL number is an indication of how
level the slab is, demonstrating how level the forms were
set. The floor slope is measured over a distance of 10 ft
(30 m) (Fig. 3).

ASTM E1155, Standard Test Method for Determining

FF Floor Flatness and FL Floor Levelness Numbers, is a
quantitative method of measuring floor surface profiles to
obtain estimates of the floors characteristic FF and FL
numbers. Each slab requires a number of individual
sample measurement lines (test runs). The quantity of
test runs is determined by the area to be tested. The
greater the area, the more test runs are required; thus,
more data are accumulated and processed to determine
F-number values for the slab.
In accordance with ASTM E1155, the test area must be
organized into a test surface, test section(s), and test
runs (Fig. 4). After the number and length of test runs are
determined, the test runs can be laid out and the run
path swept clean. Once the test runs are laid out and

Table 1:
Typical flatness (FF ) and levelness (FL) numbers for various applications (ACI 302.1R)2
Composite flatness, FF

Composite levelness, FL

Typical applications

20

15

Noncritical: mechanical rooms, nonpublic areas,

surfaces to have thick-set tile, parking structure slabs

25

20

Carpeted areas of commercial office buildings or

lightly-trafficked office/industrial buildings

35

25

Thin-set flooring or warehouse floor with moderate or

heavy traffic

45

35

Warehouse with air-pallet use, ice, or roller rinks

>50

>50

Movie or television studios

Fig. 2: The flatness is calculated from elevation readings over

24 in. (600 mm) increments

Fig. 1: Typical flatness and levelness requirements for various

applications2

36

January 2011

/ Concrete international

Fig. 3: The floor levelness is calculated from elevation readings

over 10 ft (3 m) increments

cleaned, data can be collect using a Dipstick floor profiler

(Fig. 5) or equivalent. Minimum sampling requirements are
discussed in ACI 117.2

Reviewing the report

A typical report includes a description of the test

surface, test section(s) and location of test runs, the
overall FF and FL numbers for the slab, the individual
FF and FL numbers for each test run, and whether any
required local minimum was violated. A graph of each
test run may be included. The graph (Fig. 6) shows the
change in elevation versus distance for the surface.
When reviewing a report, you should first verify that
the overall FF and FL values meet the specified requirements.
For example, suppose your project specification calls for
a minimum FF of 25 and a minimum FL of 20. The corresponding minimum local values are typically 60% of these
values, or 15.0 and 12.0, respectively; these should be
spelled out in the specification. Looking at the example
data in Table 2, you can see that the overall flatness and
levelness requirements have been met.
After checking the overall flatness and levelness values
against the specification, check the values of FF and FL for

Fig. 4: Sample layout for

flatness and levelness
measurement of a floor

Concrete international

/ January 2011

37

the individual test runs. Figure 6 shows an example test

run. The vertical scale is exaggerated to show the profile
more clearly. In this case, FF = 26.71, higher than the
overall requirement and higher than the required local
minimum of 15.0, and FL = 19.42, lower than the overall
requirement but higher than the required local minimum
of 12.0.

Fig. 5: A Dipstick floor profiler is used to collect flatness and

levelness data

Fig. 6: Profile for one test run (vertical scale is exaggerated)

Table 2:
Example flatness and levelness data for an office building
Test run

Flatness, FF

Levelness, FL

1WE

35.13

24.70

2EW

24.71

11.42*

3WE

44.75

25.96

4NS

39.75

31.92

5SN

39.63

28.39

6NS

26.71

19.42

Overall (25/20)

35.11

23.80

Does not meet the local minimum value

38

January 2011

/ Concrete international

The last step is to check to see whether any local

minimum values have been violated. Reviewing the data
in Table 2, you can see that Test Run 2EW violates the
minimum local value of FL because it is only 11.42, less
than the minimum local value 12.0. All other test runs
meet the minimum local values for both FF and FL.

Remedies

If the overall FF and FL values exceed the minimum

specified requirements and the minimum local values
have not been violated, there is no need for remediation.
However, ifas in the examplethe overall values meet
the specification and the minimum local values dont,
the surface will need remediation in the areas where
the minimum local values were out of spec. Additional
testing will be required to determine the entire area for
remediation. If the specified minimum overall numbers
are not met, the entire surface or selected areas should
be remediated and the surface retested. Remediation
methods vary greatly in surface preparation, application
effort, and cost so the selected remediation method
varies from project to project.
Some reports may include measurements of the entire
slab to quantify a slab that has been found to be out of
spec. Different modeling programs can be used to aid
in selecting a remediation method. For example, a
mesh diagram (Fig. 7) can be very helpful in evaluating
a slab surface.
If the results do not meet the specifications, remedial
measures may be needed and a reduction in payment as
previously agreed upon may be called for. Remedial
measures for slabs-on-ground might include grinding,
planing, surface repair, retopping, or removal and
replacement. For suspended slabs, remedial measures
are generally limited to grinding or use of an underlayment
or topping material. Contract documents should clearly
spell out the penalties to be imposed should the specified
tolerances be exceeded. Generally, they will not mandate
the remedial measures to be taken, as the Engineer of
Record needs to make judgments about the appropriate
action(s) in each individual case. In an office that is to
be carpeted, a floor leveling compound may provide a
sufficiently level surface for the carpet; for a warehouse
floor, grinding the high spots may be preferred.

References
1. Face, A., Floor Flatness and LevelnessThe F Number
System, Construction Specifier, V. 40, No. 4, Apr. 1987, pp. 24-32.
2. ACI Committee 302, Guide for Concrete Floor and Slab
Construction (ACI 302.1R-04), American Concrete Institute,
Farmington Hills, MI, 2004, 76 pp.
3. ACI Committee 117, Specifications for Tolerances for Concrete
Construction and Materials and Commentary (ACI 117-10),
American Concrete Institute, Farmington Hills, MI, 2010, 76 pp.

Fig. 7: Mesh diagram showing the profile of the surface. This type
of diagram is useful in determining the appropriate repair method

Note: Additional information on the ASTM standard discussed in

this article can be found at www.astm.org.

Mark A. Cheek, FACI, is Vice President of

Beta Testing & Inspection, LLC, New
Orleans, LA. He has over 20 years of
experience in construction materials
testing and inspection. He is a Past
President of the ACI Louisiana Chapter, and
is a member of the Chapter Activities
Committee, Certification Programs Committee,
Convention Committee, and ACI Committees
C610, Field Technician Certification; C620, Laboratory Technician
Certification; 228, Nondestructive Testing of Concrete; 214,
Evaluation of Results of Tests Used to Determine the Strength of
Concrete; and E702, Designing Concrete Structures. He is
also a member of Honors and Awards Committee and Chair of the
ACI Young Member Award for Professional Achievement. Cheek
received his BS in civil engineering from the University of New
Orleans and is a licensed professional engineer in Louisiana
and Mississippi.

Selected for reader interest by the editors.

eLearning

Build Your
Success Online
Now Available:
Online Training for Concrete Field Testing Technician
Grade I Certification*
Full course (6 modules): $149, $119 for ACI members
Individual modules: $30, $24 for ACI members
Online Training for Concrete Strength Testing
Technician Certification*
Full course (4 modules): $99, $79 for ACI members
Individual modules: $30, $24 for ACI members
Concrete Fundamentals
Full course (3 modules): $99, $79 for ACI members
Individual modules: $40, $32 for ACI members
Concrete Basics
1module: $30, $24 for ACI members
Additional courses coming this spring. Visit the Web site for details.
Separate written and performance exams, administered through an
ACI Local Sponsoring Group, are required to receive certification
from ACI.
*

Visit our Web site: ACIelearning.org

Concrete international

/ January 2011

39

Letters
The Floor Flatness Report

While we applaud the effort to make CI readers aware of

the contents of floor flatness reports, we are concerned
about a number of issues in The Floor Flatness Report
from the January 2011 CI (V. 33, No. 1, pp. 35-39). The
article may cause some building owners and others interested
in floors meeting flatness and levelness tolerances to have
unreasonable or misguided expectations regarding reported
floor flatness and levelness results.
Some statements included in the article therefore require
correction or clarification.
Quoting from p. 37 on taking measurements with a
Dipstick floor profiler or equivalent:
Minimum sampling requirements are discussed in
ACI 117.
In fact, the requirements for the minimum number of
10 ft elevation difference readings per the test section, Nmin,
are discussed in Section 7.6 of ASTM E1155, Standard Test
Method for Determining FF Floor Flatness and FL Floor
Levelness Numbers. ACI 117, Specification for Tolerances
for Concrete Construction and Materials (ACI 117-10) and
Commentary, provides only minimum sampling requirements for the manual straightedge method (Section 4.8.6.2).
Also quoting from p. 37:
When reviewing a report, you should first verify that
the overall FF and FL values meet the specified requirements. For example, suppose your project specification calls
for a minimum FF of 25 and a minimum FL of 20. The
corresponding minimum local values are typically 60% of
these values, or 15.0 and 12.0, respectively; these should be
spelled out in the specification. Looking at the example
data in Table 2, you can see that the overall flatness and
levelness requirements have been met.
The inclusion of the word minimum in the second
sentence is problematic. Minimum FF and minimum FL
could be confused with minimum local values for flatness
(MLFF) and levelness (MLFL) as described in ACI 117,
Section 4.8.5.3, so better wording would have been:
suppose your project specification calls for a specified
overall value for flatness (SOFF) of 25 and a specified
overall value for levelness (SOFL) of 20. The corresponding
minimum local values for flatness (MLFF) and levelness
(MLFL) are typically 60% of these values, or 15.0 and 12.0,
respectively.
A similar issue occurs on p. 38:
If the overall FF and FL values exceed the minimum
specified requirements and the minimum local values have
not been violated, there is no need for remediation.
This should be phrased as, If the overall FF and FL values
meet or exceed the specified overall values and the minimum
local values have not been violated, on any minimum local
6

march 2012 Concrete international

areas, there is no need for remediation.

In terms of setting unreasonable expectations, the
following statement from p. 38 is the most questionable:
However, ifas in the examplethe overall values
meet the specification and the minimum local values dont,
the surface will need remediation in the areas where the
minimum local values were out of spec. Additional testing
will be required to determine the entire area for remediation.
This statement is true if minimum local values are not
met in a local tested area or areas. Unfortunately, the cited
example provides only data from a single run, not a complete
test section (minimum local area). Thus, the reader is led to
believe that a single run with results below the minimum
local value requires remediation for violating the minimum
local tolerance value. This is not true.
An individual run can be used solely as datanothing
more. The data might indicate a possibility of an issue with
flatness and/or levelness in an area around a particular run,
but the data cannot be compared against a specified value.
As ACI 117, Commentary Section R4.8.5.3, states, Acceptance
or rejection of a minimum local area requires that data
collection within the minimum local area in question meet
the requirements of ASTM E1155. ASTM E1155, Section 7.2.1,
specifies that no test section shall measure less than 8 ft on
a side, nor comprise an area less than 320 ft2. Because a
single test run represents data from a line (not an area), the
data in the example cannot be compared against specified
MLFF and/or MLFL values.
More information on why one sample measurement line
of flatness/levelness cannot be used as a means for rejecting
a floor is provided in the July 2008 Concrete Q&A
(Rejecting Floors Based on One Sample Measurement
Line, Concrete International, V. 30, No. 7, pp. 83-84).
Darrell L. Darrow, Allflat Consulting, Norfolk, VA
Bryan M. Birdwell, Birdwell and Associates, Lakeland, FL
Editor and authors response
The writers comments are appreciated. We agree that
minimum sampling requirements are not discussed in
ACI 117 and the reference should have been to ASTM
E1155. The use of the word minimum when referring to
specified overall values (SOF), albeit possibly considered
problematic, is a correct term in the sense that the SOF is
the lowest value tolerable by the specifier. The error on p.
38 was the result of an unfortunate attempt by the Editor to
get more value out of a table. A single test run cant be
compared against an SOF, so the clause, as in the example,
should not have been included in the discussion.
Rex C. Donahey, American Concrete Institute, Farmington
Hills, MI
Mark A. Cheek, Beta Testing & Inspection LLC, Gretna, LA

Rejecting Floors
Based On
One Sample
Measurement Line
Q.

Weve just been informed that the owner is rejecting

a floor weve recently completed. The basis of the
rejection: flatness/levelness measurements taken along one
sample measurement line are below the specified minimum
value. When minimum FF /FL values, as defined in ACI 117-061
or ACI 117-90,2 are specified for random-traffic floor surfaces,
do those minimum values apply to each sample measurement
line, as defined in ASTM E1155, 3 or do they apply only to
the composite of the measurement lines?

A.

Lets first consider what is meant by minimum

local values, which are defined only in ACI
documents. Section 4.8.5.1 of ACI 117-06, Specifications
for Tolerances for Concrete Construction and Materials,1
states that specified overall values for flatness (SOFF) and
levelness (SOFL ) shall conform to one of the floor surface
classifications shown in Table 1, unless noted otherwise.
Section 4.8.5.3 of ACI 117-06 states:
Questions in this column were asked by users of ACI documents
and have been answered by ACI staff or by a member or members
of ACI technical committees. The answers do not represent the
official position of an ACI committee. Only a published committee
document represents the formal consensus of the committee and
the Institute.
We invite comment on any of the questions and answers published
in this column. Write to the Editor, Concrete International, 38800
Country Club Drive, Farmington Hills, MI 48331; contact us by fax at
(248) 848-3701; or e-mail Rex.Donahey@concrete.org.

Minimum local values for flatness (MLFF ) and

levelness (MLFL ) shall equal 3/5 of the SOFF and
SOFL values, respectively, unless noted otherwise.
Section R4.8.5.3 (the Commentary) of ACI 117-06 states:
Acceptance or rejection of a minimum local area
requires that data collection within the minimum
local area in question meet the requirements of
ASTM E1155.
This means that the answer to your question can be
found in the data collection requirements of ASTM E1155,
Standard Test Method for Determining FF Floor Flatness
and FL Floor Levelness Numbers.3 Sections 7.1 through
7.3 of this document define test surface, test section, and
sample measurement line as follows:
7.1 Test SurfaceOn any one building level, the
entire floor area of interest shall constitute the
test surface...

Table 1:
Floor surface classifications in ACI 117-061
Floor surface
classification

Specified overall
flatness SOFF

Specified overall
levelness SOFL

Conventional

20

15

Moderately flat

25

20

Flat

35

25

Very flat

45

35

Super flat

60

40

Concrete international

/ july 2008

83

7.2 Test SectionA test section shall consist of

any subdivision of a test surface satisfying the
following criteria:
7.2.1 No test section shall measure less than 8 ft
[2.4 m] on a side, nor comprise an area less than
320 ft2 [29.7 m2]
7.2.2 No portion of the test surface shall be
associated with more than one test section.
7.2.3 When testing a concrete floor, no test section
boundary shall cross any construction joint.
7.3 Sample Measurement LineA sample
measurement line shall consist of any straight line
on the test surface satisfying the following criteria:
7.3.1 No sample measurement line shall measure
less than 11 ft [3.3 m] in length.
7.3.2 No portion of any sample measurement line
shall fall within 2 ft [0.6 m] of any slab boundary,
construction joint, isolation joint, block-out,
penetration, or other similar discontinuity.
Now lets assume, for example, that levelness, and
specifically MLFL, is the value of interest. ASTM E1155
requires a minimum number (Nmin) of elevation difference
readings per test section. Nmin is calculated as follows
Nmin = 2 A for 320 ft2 A 1600 ft2
(in.-lb units) (1)


Nmin = 6.56 A for 30 m2 A 150 m2
(SI units)
or

8.2.3 The sample measurement lines within each

test section shall be arranged so as to blind the
test results (to the extent possible) to any surface
profile anisotropies resulting from the floors
method of construction. Accomplish this by
distributing the sample measurement lines
uniformly across the entire test section and either:
8.2.3.1 Orienting all lines at 45 to the longest
construction joint abutting the test section, (not
corner-to-corner diagonals), ... or
8.2.3.2 Placing equal numbers of lines of equal
aggregate length both parallel to and perpendicular
to the longest test section boundary.
Using results from only one sample measurement
line doesnt blind the test results as required. If sample
measurement lines were to be considered separately in
determining a floor levelness number, ASTM E1155
would be expected to require that the report include
results for each sample measurement line. ASTM E1155
requires only that the F-numbers and associated 90%
confidence interval be reported for a particular test
section. The rationale for not reporting results on
individual lines would seem to be that individual
sample measurement lines dont provide enough data
to satisfy the statistical requirements of ASTM E1155,
nor do they blind the test results as required. Thus,
the minimum values dont apply to each sample
measurement line. The same reasoning can be used
with regard to FF numbers.

N = A/30 for A > 1600 ft2

min

(in.-lb units) (2)

References

Nmin = A/3 for A > 150 m2

(SI units)

Construction and Materials (ACI 117-06), American Concrete

Institute, Farmington Hills, MI, 2006, 70 pp.
2. ACI Committee 117, Standard Specifications for Tolerances
for Concrete Construction and Materials (ACI 117-90) (Reapproved
2002), American Concrete Institute, Farmington Hills, MI, 2002,
12 pp.
3. ASTM E1155-96, Standard Test for Determining FF Floor
Flatness and FL Floor Levelness Numbers (Reapproved 2001), ASTM
International, West Conshohocken, PA, 2001, 8 pp.

1. ACI Committee 117, Specifications for Tolerances for Concrete

where A is the test section area in ft2 (m2). These

readings must be taken along sample measurement
lines that are at least 10 ft (3 m) long. Thus, the
minimum number of elevation difference readings
per test section is independent of the number of
sample measurement lines but varies with the size of
the test sectionlarger test sections require a larger
number of readings. Its common practice to define a
single concrete floor placement as the test section. The
single placement can be divided into smaller test
sections if desired or specified, but this results in more
measurements being required and usually higher
measurement costs. Whatever the test section area
chosen, the minimum number of readings must be
taken as required by ASTM E1155.
A further requirement in Section 8.2.3 of ASTM E1155
is as follows:

84

July 2008

/ Concrete international

ACI members can find this Q&Aand many

others related to concrete design and construction
in the Technical Questions section of the Concrete
Knowledge Center. Go to www.concrete.org, log in
using your ACI username and password, and click
on the Concrete Knowledge Center button on the
right side of the screen.