- Email: [email protected]
Surface regularity
,, Surface
19
..
Regularity
Surface regularity has been the focus of great interest in recent years. For most of the twentieth century, floor designers either ignored surface regularity or tried to control it with specifications based on a 3-m (10-ft) straightedge. Those straightedge tolerances were rarely enforced; indeed, effective enforcem e n t was next to impossible because there was no standard test. So the situation m i g h t have remained, except for a d e v e l o p m e n t in the materials-handling field. In the early 1970s, forklift makers i n t r o d u c e d the very-narrow-aisle (VNA) turret truck. It differed from ordinary forklifts in having its fork m o u n t e d on a rotating turret. While ordinary forklifts needed aisles about 4 m (13 ft) wide, the new turret trucks could run in aisles only 2 m (6 ft) wide. Narrower aisles m e a n t greater storage d e n s i t y s o m e t h i n g m a n y warehouse users sought. The turret truck had one glaring drawback, however. It needed a flat and level floor. Warehouse users w h o tried to run turret trucks on ordinary 1970s floors soon learned that neither the 3-m (lO-ft) straightedge nor the floorlaying m e t h o d s t h e n in use were good e n o u g h for the new vehicles. To satisfy turret-truck users, floor designers and builders developed ways to build flatter and more level concrete slabs. Their efforts led to the superflat floor, discussed at the end of this chapter. Perhaps even more importantly, the sharper focus on surface regularity led people to question the flatness and levelness of all floors, n o t just those that supported turret trucks and needed a superflat surface. The result was radical new ways to specify and test floor surface regularity: the F-number system in America and the TR 34 system in the UK. But before we look at those systems, we need to consider two key distinctions: between flatness and levelness, and between r a n d o m and defined traffic.
312
Design and Construction of Concrete Floors
Flatness and levelness
These terms have o v e r l a p p i n g definitions in e v e r y d a y speech. But w i t h i n t h e field of concrete floors, flatness a n d levelness m e a n different things. Flatness is defined as planarity or lack of curvature. A flat floor is s m o o t h a n d free of b u m p s or dips. An unflat floor is b u m p y or wavy. Levelness is defined as h o r i z o n t a l i t y or lack of slope. A level floor is horizontal. An unlevel floor is sloped or tilted. A few examples m a y help illustrate the d i s t i n c t i o n b e t w e e n flatness a n d levelness. In n o r m a l use, a billiards table is b o t h flat a n d level. If we p u t bricks u n d e r o n e end, the table will still be flat b u t it will n o longer be level. O n a calm day, the surface of a p o n d is b o t h flat a n d level. If the w i n d rises t h e surface will cease to be flat, b u t it r e m a i n level. Some floors are designed to be unlevel. T h e y are sloped for drainage or to m i n i m i z e site grading. I have never hear of a n y floors designed to be unflat, t h o u g h m a n y t u r n out t h a t way. Flatness a n d levelness are b o t h desirable properties (except o n those floor with design slopes), b u t t h e y have different i m p l i c a t i o n s for t h e floor user. Flatness is critical where the m a i n issue is the b e h a v i o r of w h e e l e d vehicles. Flatness is also i m p o r t a n t where a p p e a r a n c e m a t t e r s - as in an institutional floor t h a t will be covered with vinyl tile. In contrast, levelness is critical where the m a i n issue is the installation of fixed e q u i p m e n t such as w a r e h o u s e racks or m a c h i n e tools. While exceptions exist, m o s t users find flatness m a t t e r s m o r e t h a n levelness. If a floor is unlevel, fixed e q u i p m e n t can be s h i m m e d or jacked u p to a c c o m m o d a t e it. But it is n o t so easy to a d a p t a w h e e l e d vehicle to a surface t h a t is n o t flat e n o u g h for it. Flatness a n d levelness also have different implications for the floorlayer. Flatness comes from finishing; it is d e t e r m i n e d m a i n l y by the finishing m e t h o d s , a n d to a lesser extent by the finishability of the concrete mix. Levelness comes from forms a n d strike-off; it is d e t e r m i n e d m a i n l y by the accuracy of the side forms and the m e t h o d used to strike of the concrete.
D e f i n e d versus r a n d o m t r a f f i c
Floors get two kinds of traffic: defined a n d r a n d o m . O n a defined-traffic floor, vehicles are c o n f i n e d to fixed paths. O n a r a n d o m - t r a f f i c floor, vehicles (or walkers) are free to roam, t h o u g h inevitably some traffic p a t h s see m o r e use t h a n others. The distinction matters because the two kinds of traffic call for different ways of m e a s u r i n g surface regularity. O n a defined-traffic floor, y o u can
Surface Regularity
313
m e a s u r e a c o n t i n u o u s , or nearly c o n t i n u o u s , profile d o w n every travel path. You can check effectively all of the floor profile t h a t m a t t e r s to t h e vehicle user. In contrast, a random-traffic floor has an infinite n u m b e r of possible travel paths. You could never m e a s u r e t h e m all, so y o u h a v e to fall back o n statistical sampling. You check selected points or lines a n d h o p e t h e y represent the w h o l e surface. Historically, the highest r e q u i r e m e n t s for surface regularity have b e e n f o u n d in the defined-traffic category. The reason is t h a t turret t r u c k s - t h e vehicles t h a t d e m a n d the flattest a n d m o s t level f l o o r s - r u n in defined paths. It w o u l d be wrong, however, to assume t h a t defined-traffic floors are always superior in surface regularity. T h o u g h defined-traffic floors are i m p o r t a n t , their random-traffic cousins are far more n u m e r o u s . Defined-traffic floors are f o u n d in very-narrow-aisle warehouses a n d in buildings w i t h a u t o m a t i c a l l y - g u i d e d v e h i c l e s - a n d a l m o s t n o w h e r e else. The great m a j o r i t y of industrial floors, as well as all n o n - i n d u s t r i a l floors, fall in the random-traffic category.
F-numbers Since its a d o p t i o n into ACI a n d ASTM standards in the late 1980s, t h e F - n u m b e r system has b e c o m e the m o s t c o m m o n way to specify a n d measure floor surface regularity in the U n i t e d States a n d Canada. It is also widely used in Latin America a n d Australia. Europeans, in contrast, are m o r e likely to use the TR 34 system or t h e G e r m a n standards DIN 18 202 a n d DIN 15 185. The F - n u m b e r system includes three separate n u m b e r s : 9 Fmin, w h i c h e n c o m p a s s e s b o t h flatness a n d levelness a n d is used o n l y o n defined-traffic floors; 9 FF, w h i c h controls flatness a n d is used o n l y o n random-traffic floors; 9 FL, w h i c h controls levelness a n d is used o n l y o n random-traffic floors. Defined-traffic floors are specified by the single n u m b e r Fm~n. R a n d o m traffic floors are n o r m a l l y specified by a pair of n u m b e r s , w r i t t e n as FF_/FL_. However, some s u s p e n d e d random-traffic floors are specified by F F alone. The three n u m b e r s have this m u c h in c o m m o n : 9 A h i g h e r F - n u m b e r always d e n o t e s a better floor. 9 The n o r m a l scale runs from 10 to 100. O n l y the worst floors exhibit F-numbers below 10, a n d o n l y a few of t h e best m e a s u r e above 100. 9 The scales are linear. If y o u double the F-number, y o u halve t h e m a g n i t u d e of the allowed deviations.
Design and Construction of Concrete Floors
314
Fmi n for defined traffic floors
The F-number system controls defined-traffic floors by means of a single number called Frnin. Fmin defines both flatness and levelness, and is measured over the wheel spacings of the vehicle that will travel on the floor. Testing is confined to the travel paths. W h e n you say a floor measures, say, Frown50, you are not making a general statement about the floor's flatness and levelness. You are saying that when a particular vehicle travels in a particular path, it will not encounter any slope or change in slope greater t h a n that allowed by the Fmin equations, shown below: In SI units" [6.55 (L +69) ~ - 4 8 ] Gin
dmax= elTlaX
33.3 F~in
where" m a x i m u m elevation difference between points separated by L, in rnm; emax m a x i m u m rate at which elevation difference between points separated by L can change with travel across the floor, in mrn/lOOmm; L distance between wheels (transverse) or axles (longitudinal), in rnm; G i n ~ the defined-traffic F-number (dimensionless).
dmax
In fps units:
dmax emax
[1.3 (L + 2.7)~
Fmin 4.00 Fmin
where:
dmax - m a x i m u m elevation difference between points separated by L, in inches; = m a x i m u m rate at which elevation difference between points emax separated by L can change with travel across the floor, in inches/foot;
Surface Regularity
315
L - distance between wheels (transverse) or axles (longitudinal), in inches; F m i n = the defined-traffic F-number (dimensionless). The tolerances dma x and emax are usually expressed to 0.1 m m or 0.001 in. The e q u a t i o n s for dma x (levelness) are good for wheel or axle spacings from 300 m m to 3 m (1 to 10 ft). The e q u a t i o n s for emax (flatness) are good for wheel or axle spacings from 1 to 3 m (40 in to 10 ft). Almost all vehicles used on defined-traffic floors have d i m e n s i o n s w i t h i n those ranges. Fm~n is normally measured with a differential profileograph set up to m a t c h the wheel and axle spacings of the defined-traffic vehicle. The profileograph measures the elevation difference between left and right wheels (the transverse-differential profile) and between front and rear axles (the longitudinal-differential profile). Both profiles are checked for compliance with the calculated limits for dmax and emax. Specified Fmin values range from 40 (rarely lower) to 100 (never higher). Fm~nlO0 is the traditional superflat specification, and is reserved for VNA warehouses with very tall racks or very high t h r o u g h p u t , or both. In recent years, Fmin60 has become popular for less-demanding VNA applications. With laser screeding, contractors can lay Fmin60 floors in large bays, eliminating the costly narrow-strip layout needed for Fm~nlO0. Fmi~ 40 is sometimes specified for the least d e m a n d i n g VNA applications, and also for low-rise automatically-guided vehicles.
FF/F L for random-traffic floors
The F-number system controls random-traffic floors with a pair of numbers, F F and F L. The flatness number, F~, is based on the floor curvature over 600 m m (24 in) (see Figure 19.1). The levelness number, F L, is based on the floor slope over 3 rn (10 ft) (see Figure 19.2). ASTM E 1155 (fps units) and ASTM E 1155M (SI units) describe the standard test for F F and F L. A typical survey involves these steps: 1. Determine the m i n i m u m n u m b e r of 3-rn (lO-ft) slope readings. For test sections of 150 m 2 or more, the m i n i m u m is the area divided by 3. (In fps units, it's the area in ft 2 divided by 30, for sections of 1600 ft 2 or more.) 2. Lay out e n o u g h straight lines on the floor to provide at least the m i n i m u m n u m b e r of 3-m (lO-ft) slope readings. The n u m b e r of readings from each line is equal to the length in m, minus 2.7 m, divided
316
Design and Construction of Concrete Floors ii
Floor surface
7 J j
er"
/' ~
600 mm (24 in) curvature
j ~-- 300 mm --~ ~-- 300 mm --~ (12 in)
(12 in)
Figure 19.1 600-mm (24-in) curvature determines the flatness number, F~. TR 34 calls this Property il
by 0.3 m. (In fps units, the n u m b e r of readings from each line is simply the length in ft, minus 9 ft.) Figure 19.3 shows one possible layout, with all lines oriented 45 ~ to the test section's longest side. It is also permissible, where the test section is at least 7.5 m (25 ft) wide, to lay out runs parallel and perpendicular to the longest side. Under most circumstances, survey lines must avoid slab edges, construction joints and penetrations by at least 600 rnm (2 ft). 3. Measure point elevations on 300-ram (12-in) centres down each survey line. 4. Calculate the elevation difference between each pair of adjacent points. This value is the 300-rnm (12-in) slope. 5. Calculate the difference between each pair of consecutive 300-rnrn (12-in) slope readings. This is the 6 0 0 - m m (24-in) curvature.
Floor surface
I.~
3 m (10 ft)
~
m (lOt) slope
....
r
Figure 19.2 3-m (lO-ft) slope determines the levelness number, FL. TR 34 calls this Property IV
Surface Regularity
600 mm (24 in) ~1~
Floor edge or construction joint |
317
Profile runs
.....
tm" v
E ----~ E 0 0 CD
600 mm (24 in)
Figure 1 9 . 3
One way to lay out an F-number survey is w i t h all lines at 45 ~ to the slab's
longest side
6. C a l c u l a t e t h e F F n u m b e r f r o m o n e of t h e f o l l o w i n g e q u a t i o n s : In SI units:
FF (3Sq114+ q) where: F F = t h e r a n d o m - t r a f f i c flatness n u m b e r ( d i m e n s i o n l e s s ) ; q = 6 0 0 - m m c u r v a t u r e , in mrn; Sq = s t a n d a r d d e v i a t i o n of q; q = m e a n of q. In fps u n i t s :
FF
4.57 (3Sq + q)
where: F F = t h e r a n d o m - t r a f f i c flatness n u m b e r ( d i m e n s i o n l e s s ) ; q = 2 4 - i n c u r v a t u r e , in i n c h e s ; S q - s t a n d a r d d e v i a t i o n of q; q = m e a n of q.
Design and Construction of Concrete Floors
318
7. Calculate t h e s u m of each set of t e n c o n s e c u t i v e 3 0 0 - m m (12-in) slopes. This is the 3 - m (l O-fl) slope. 8. Calculate the F L n u m b e r from o n e of t h e following equations" In SI units: 315 EL -- (3S~ + z) where: F L -- t h e r a n d o m - t r a f f i c levelness n u m b e r (dimensionless); q-- 3-m slope, in m m ; S z - s t a n d a r d deviation of z; z - m e a n of z. In fps units" EL ~-
12.5 (3Sz+Z)
where: t h e r a n d o m - t r a f f i c levelness n u m b e r (dimensionless); z -- l O-ft slope; S z = s t a n d a r d d e v i a t i o n of z; z -- m e a n of z.
F L --
ASTM standards allow a variety of tools for m e a s u r i n g F F and F L, including optical levels, lasers a n d levelled straightedges w i t h d e p t h gauges. However, a l m o s t e v e r y b o d y uses an electronic i n c l i n o m e t e r - either t h e w a l k i n g t y p e w i t h feet 3 0 0 m m (12 in) a p a r t (the D i p s t i c k is t h e b e s t - k n o w n b r a n d n a m e ) , or t h e r o l l i n g t y p e w i t h w h e e l s s e p a r a t e d by t h e s a m e d i s t a n c e . Figure 11.11 in C h a p t e r 11 s h o w s a n e l e c t r o n i c i n c l i n o m e t e r . The resulting F - n u m b e r s will be practically t h e same w h e t h e r surveyed in SI or fps units. While no fixed relationship exists b e t w e e n F F a n d F L, the former averages about 40% higher t h a n the latter o n g r o u n d - s u p p o r t e d slabs. O n s u s p e n d e d floors the ratio of F F to F L tends to r u n higher. ACI 117 r e c o m m e n d s t h a t a two-tiered f o r m a t for r a n d o m - t r a f f i c F - n u m b e r specifications. The overall F - n u m b e r s apply to the floor t a k e n as a whole. The m i n i m u m local F - n u m b e r s apply to smaller sections of t h e floor. M i n i m u m local n u m b e r are always lower t h a n the overall n u m b e r s typically one third to o n e half lower. The two-tiered format, t h o u g h sometimes criticized as needlessly complex, has p r o v e n an effective tool for i m p r o v i n g floor flatness a n d levelness.
Surface Regularity
3119
It pushes builders t o w a r d h i g h e r n u m b e r s , w i t h o u t p u n i s h i n g t h e m brutally for the occasional bad day. The goal is to lay t h e w h o l e floor to the overall F-numbers. But if the floorlayers do worse t h a n that on an individual slab, or part of a slab, it can still be accepted provided it meets the m i n i m u m local F-numbers. Such an occurrence serves as a w a r n i n g t h a t the crew m u s t do better o n later slabs, to b r i n g the overall n u m b e r s up to the specified values. If a test section fails to m e e t even the m i n i m u m local values, it m u s t be repaired or replaced. But such failures are rare if the n u m b e r are c h o s e n w i t h care a n d if all parties u n d e r s t a n d t h e rules from t h e start. Table 19.1 shows overall a n d m i n i m u m local F-numbers for four classes of surface regularity. The classification comes from ACI 117, b u t the r e c o m m e n d e d applications are m i n e .
Construction joints ASTM E 1155 and ASTM E 1155M forbid measuring F F and F L within 600 m m (24 in) of a n y slab edge or c o n s t r u c t i o n joint. (There is an e x c e p t i o n for very small slabs.) Because the edges a n d c o n s t r u c t i o n joints are statistical anomalies, i n c l u d i n g t h e m in t h e calculations w o u l d distort the results. Table 119.1 Classification of floor flatness and levelness by F-numbers - for random-traffic floors (ACI 11 7)
Minimum F-numbers required Floor class
Application
Very flat
Warehouses with reach trucks over 8 m (26 ft) high, random-traffic areas in some VNA warehouses, air-bearing systems Offices with moveable partitions, warehouses with mobile racks or reach trucks up to 8 m (26 ft high), busy generalpurpose warehouses
FF50/FL30
FF25/FL15
FF301FL20
FFI 5/FLI 0
Warehouses and factories where surface regularity is not critical Foot traffic, garages
FF2OIFL15
FF151FL10
FF15/FL1 3
FF1 3/FL10
Flat
Conventional Straightedged
Bull-floated
Overall
Local
320
Design and Construction of Concrete Floors
The exclusion of construction joints from testing causes some problems, because the areas adjacent to construction joints are o f t e n - I would say u s u a l l y - the least flat parts of a floor. It is hard to justify a system t h a t excludes the spots where bad readings are most likely to occur. The 1996 editions of ASTM E 1155 and ASTM E 1155M offer a solution. Designers can specify a limit for 600-ram (12-in) curvature at construction joints. The usual tolerance is 3.0 or 4.0 m m (0.125 or 0.150 in). The standard test calls for at least one test for each 3-m (lO-ft) length of joint. Enforcement is normally limited to those construction joints that will be subject to traffic.
Suspended floors Random-traffic F-numbers apply to suspended floors, but with some restrictions. American standards let us enforce the flatness number, FF, on any type of suspended floor. We need to recognize, however, that F F values above 30 can be very hard to achieve on suspended slabs. While making a suspended slab flat is hard, m a k i n g it level is even harder. In view of that fact, ACI 117, the m a i n American standard for concrete tolerances, prohibits the e n f o r c e m e n t of F L tolerances on susp e n d e d floors, with the exception of fully shored slabs that are tested while the shores are still in place. U n s h o r e d floors, including most slabs on metal deck, are excluded from levelness testing. If a suspended floor needs to be very flat or very l e v e l - luckily most do n o t - the best choice is usually a t o p p i n g or screed.
The TR 34 system The Concrete Society's Technical Report 34, Concrete Industrial Ground Floors, c o m m o n l y called just TR 34, includes a thorough, well-thought-out system for specifying floor flatness and levelness on industrial floors. The TR 34 system shares m a n y features with F-numbers - a fact that stems from their c o m m o n origin in research carried out by Allen Face in the early 1980s. Like F-numbers, TR 34 controls b o t h flatness and levelness. Again like F-numbers, TR 34 has different specifications for defined traffic and r a n d o m traffic, which TR 34 calls "defined m o v e m e n t " and "free m o v e m e n t " . Perhaps the most significant difference between F-numbers and TR 34 lies in the way tolerances are presented. F-numbers are continuously variable.
Surface Regularity
321
The designer must choose a n u m b e r and can, within practical limits, choose any number. While ACI 117 does classify floors (see Table 19.1), designers remain free to specify other values, and m a n y do so. TR 34, in contrast, divides floors into a small n u m b e r of standard categories, each having its own set of tolerances. The designer does not set a tolerance, but only picks a category. I confess a slight prejudice in favour of the F-number approach, while recognizing that standard categories have merit, too.
TR 34 properties At the heart of the TR 34 system are four properties, identified by R o m a n numerals I-IV. Property I (see Figure 19.4) is the elevation difference between two points separated by 300 m m (12 in). This is a levelness property. It applies only to defined-traffic floors, where it is measured parallel to the direction of travel. Property II (see Figure 19.1) is the arithmetic difference between a pair of consecutive Property I readings. This is a flatness property. It applies both to defined-traffic and random-traffic floors. On defined-traffic floors it is measured parallel to the direction of travel. On random-traffic floors it is measured in two directions. Property II is the same as 600-ram (24-in) curvature in the F-number system. Property III (see Figure 19.5) is the elevation difference between the left and right wheelpaths in which a vehicle runs. This is a levelness property. It applies only to defined-traffic floors, where it is measured in the direction perpendicular to travel, with readings taken on 300-ram (12-in) centres. It is the same as transverse levelness in the F-number system.
Direction of travel
t
Floor
Property I
i-~
Figure 1 9 . 4
TR 34 P r o p e r t y I
9
300 mm (12 in)
,,-
322
Design and Construction of Concrete Floors
Floor surface Load axle of vehicle
Property I!1
I_ i-~
Vehicle wheeltrack
Figure 19.5 TR 34 Property III
Property IV (see Figure 19.2) is the elevation difference between two points separated by 3 rn (10 ft). This is a levelness property. It applies only to random-traffic floors, where it is measured on a 3-m (10-ft) grid. It is the same as 3-m (10-ft) slope in the F-number system.
TR 34 for defined-traffic floors
TR 34 divides defined-traffic floors into three categories of surface regularity, called Superflat, Category 1 and Category 2. Each category is associated with limits on Properties I, II and III (see Table 19.2). Table 19.2 TR 34 tolerances for defined-traffic floors Tolerance Property I
Property II
Property III Wheeltrack up to 1.5 m (60 in)
Up to 1.5 m (60 in)
Category
95%
100%
95%
100%
95%
100%
95%
100%
Superflat
0.75 mm 0.030 in 1.5 mm 0.059 in 2.5 mm 0.098 in
1.0mm 0.039 in 2.5mm 0.098 in 4.0mm 0.157in
1.0mm 0.039 in 2.5mm 0.098 in 3.25mm 0.126in
1.5mm 0.059 in 3.5mm 0.1 38 in 5.0mm 0.197in
1.5ram 0.059 in 2.5mm 0.098 in 3.5mm 0.138in
2.5mm 0.098 in 3.5mm 0.1 38 in 5.0mm 0.197in
2.0mm 0.079 in 3.0mm 0.118in 4.0mm 0.157in
3.0ram 0.118in 4.5mm 0.1 77 in 6.0mm 0.236in
1 2
Surface Regularity
323
Note that TR 34 imposes two limits on each property. No readings are allowed to exceed the 100% limit, but up to 5% of the readings m a y exceed the 95% limit. The use of two limits represents an a t t e m p t to control not just the extreme values, but also the distribution of readings below those extremes. TR 34 ties the r e c o m m e n d e d category to the lift height of the truck, as follows: 9 Superflat for lift height over 13 rn (43 ft); 9 Category 1 for lift height of 8-13 m (26-43 ft); 9 Category 2 for lift height under 8 m (26 ft). That's useful, but it oversimplifies the situation. The need for surface regularity depends not only on lift height, but also on vehicle speed, vehicle type and the overall d e m a n d s of the floor user. Since the TR 34 properties are based on readings taken at 300-ram (12-in) intervals, floors specified in this way can be tested with same electronic i n c l i n o m e t e r s used to measure r a n d o m - t r a f f i c floors in t h e F - n u m b e r system. In practice, however, e v e r y b o d y uses differential profileographs like those used to m e a s u r e Fmi n. But while a n Fmi n profileograph is set up to m a t c h the wheel pattern of the vehicle, a TR 34 profileograph is set up to m a t c h just one axle of the vehicle (normally the wider of the two axles, if they differ). This takes care of Property III. Trailing wheels 300 m m (12 in) b e h i n d the axle allow m e a s u r e m e n t of Properties I and II. All this m a y be changing, however. The latest version of TR 34 (Concrete Society, 2003) presents, in Appendix C, an alternative m e t h o d for definedtraffic floors. The alternative m e t h o d is m u c h closer to the way definedtraffic floors are measured in the F-number system. It relies on full wheel-pattern testing, with tolerances adjusted for the distance between wheels. Floors are divided into three categories, called DM (for definedm o v e m e n t ) 1, 2 and 3. The classes correspond roughly to the current categories Superflat, 1 and 2.
TR 34 for random-traffic floors
TR 34 divides random-traffic floors into three categories of surface regularity, called FM (for free-movement) 1, 2 and 3. Each category is associated with limits on Properties II and IV. (see Table 19.3). As with the defined-traffic floors, each Property is subject to b o t h a 95% and a 100% limit. For a floor to meet its specification, no more t h a n 5% of the readings can exceed the 95% limits, and no readings can exceed the 100% limits.
324
Design and Construction of Concrete Floors
Table 119.] TR 34 tolerances for random-traffic floors Tolerance Property II
Property IV
Category
Application
95%
100%
95%
100%
FM 1
Where very high standards of flatness and levelness are required Warehouses with stacking over 8 m (26 ft) high, transfer areas in VNA warehouses Warehouses with stacking up to 8 m (26 ft) high, factories, retail stores
2.5 mm 0.098 in
4.0 mm 0.157 in
4.5 mm 0.177 in
7.0 mm 0.276 in
3.5 mm 0.138 in
5.5 mm 0.217 in
8.0 mm 0.315 in
12.0 mm 0.472 in
5.0mm 0.197 in
7.5mm 0.295 in
10.0mm 0.394 in
15.0mm 0.591 in
FM 2
FM 3
Despite the fact that Properties II and IV are identical to the properties that determine F F and F L in the F-number system, they are n o t surveyed in exactly the same way. TR 34 calls for testing random-traffic floors on a 3-m grid. Property IV is measured between the grid points. Property II is measured along the some of the lines that c o n n e c t the grid points, chosen so the total line length in metres equals or exceeds one t e n t h the floor's area in square metres. Unlike an F~/FL survey, a TR 34 random-traffic survey normally calls for two separate instruments: an optical level to check Property IV, and a rolling or walking gauge to check Property II. Despite the differences in testing, we can still reasonably compare the FM categories to F-numbers (see Table 19.4). Interestingly, British and American specifications are fairly close on flatness but differ substantially on levelness. TR 34 imposes levelness requirements t h a t American floorlayers would consider onerous. One reason for this difference m a y be that TR 34 explicitly addresses industrial floors, w h i c h are usually g r o u n d - s u p p o r t e d and are often laid to high standards. American standards, in contrast, deal with a wider range of floors, including susp e n d e d slabs. Table 119.4 Random-traffic specifications compared TR 34 category
FM 1 FM 2 FM 3
Equivalent F-numbers
FF30/FL46 F~22/626 FF16/FL21
Surface Regularity
325
TR 34 on overall elevation
Properties I-IV do n o t directly control the floor's overall elevation. For m o s t uses, overall elevation matters less t h a n local flatness a n d levelness - b u t t h a t is n o t to say it does n o t m a t t e r at all. TR 34 r e c o m m e n d s an overall tolerance of + 15 m m (+ 5/8 in) from d a t u m , for all floor classes.
Straightedge tolerances Some specifiers continue to specify floor flatness by reference to a 3-m (10-ft) straightedge (see Figure 19.6). This m e t h o d has two serious drawbacks, however. The first is the so-called w a s h b o a r d l o o p h o l e (see Figure 19.7). C h e c k e d w i t h a straightedge b o t h profiles in Figure 19.7 are equal. Yet for a l m o s t a n y use, the profile o n top is clearly the better of t h e two. The second drawback is the lack of a s t a n d a r d test. W h i l e ACI 117 a n d BS 8204:Part 2 b o t h offer a bit of g u i d a n c e in straightedge use, n e i t h e r comes close to p r o v i d i n g a c o m p l e t e test m e t h o d like those available for F-numbers a n d TR 34 tolerances. No s t a n d a r d answers these questions: 9 W h e r e is the floor measured? 9 How m a n y tests n e e d to be m a d e ? 9 Are a n y failures allowed - a n d if so, h o w m a n y ? W i t h o u t answers to those questions, a n y o n e trying to enforce a straightedge specification is likely to face a dispute.
3 m (lOft)
Figure 19.6
The straightedge test
326
Design and Construction of Concrete Floors
3 mm (1/8 in) f
\
3 m m (I18 in)
I-~
3 m (lOft)
Figure 19.7 The washboard loophole
Factors that affect surface regularity These are the m o s t i m p o r t a n t factors: 9 9 9 9 9
structural deflection; curling; w i d t h of pour; accuracy of side forms; striking off a n d finishing.
Structural deflection Suspended floors all deflect, a n d m o r e t h a n a few deflect excessively. Deflection is h a r d to control a n d predict. Deflection has a h u g e effect o n levelness, a n d a smaller b u t still significant effect o n flatness. Deflection is a serious p r o b l e m for s u s p e n d e d floors t h a t n e e d to be level. ACI 117 recognizes this by s i m p l y e l i m i n a t i n g the F L levelness r e q u i r e m e n t o n m o s t s u s p e n d e d floors. It says "levelness tolerance shall n o t apply to slabs placed o n u n s h o r e d form surfaces a n d / o r shored form surfaces after the r e m o v a l of shores". In effect, the ACI standard is telling us n o t to use o r d i n a r y s u s p e n d e d slabs w h e r e levelness matters.
Surface Regularity
327
That is good advice, to the extent we can follow it. But some s u s p e n d e d floors do need to be level, no matter w h a t ACI 117 tells us. W h a t are our options? One answer is to reduce the deflections. Chapter 6 contains some ideas on that. Another answer is to lay a t h i n topping over the suspended structural slab. But a topping w o n ' t stop the floor deflecting under loads applied later. Deflection is seldom a problem on g r o u n d - s u p p o r t e d floors.
Curling This is a major cause of poor surface regularity o n g r o u n d - s u p p o r t e d floors, concrete toppings a n d slabs cast o n metal deck. Testing flatness a n d levelness during c o n s t r u c t i o n usually fails to detect curling, because the floor curls afterwards. Chapter 14 discusses curling a n d presents some ways to m i n i m i z e its effect o n floor flatness and levelness.
Width of pour The narrower the slab, the easier is it to m a k e the floor flat a n d level. The m a i n reason for this is t h a t a n a r r o w p o u r allows a m o r e accurate strike-off, at least w i t h traditional (before the laser screed) m e t h o d s . Superflat floors are almost always laid in n a r r o w strips no m o r e t h a n 6 rn (20 ft wide). In wide-strip construction, where the floor is struck off w i t h a vibrating screed, floor levelness tends to deteriorate as the pours get wider. Some vibrating screeds can span 24 rn (80 ft), b u t y o u c a n n o t expect t h e m to m a i n t a i n a very level surface over t h a t distance. Laser screeding reduces the c o n n e c t i o n b e t w e e n slab w i d t h a n d surface regularity. With a laser screed, some floorlayers can achieve good flatness and levelness o n pour as m u c h as 30 rn (100 ft) wide.
Accuracy of side forms Side forms m a t t e r because t h e y d e t e r m i n e the elevation to w h i c h the floor is struck off. Forms have a big effect on levelness, b u t m u c h less effect o n flatness.
328
Design and Construction of Concrete Floors
Laser screeding does not eliminate the importance of accurate side forms. Though the laser screed itself does not rely on the forms, its action stops about 0.5 m (2-3 ft) from the slab edges. The accuracy of the forms determines the surface regularity of that narrow strips between the slab edge and the laser-screeded surface. With ordinary levels, it is hard to set forms closer t h a n +3 m m (+1/8 in). But that is good enough for most floors. Superflat floors need better forms. Some floorlayers who specialize in superflat work use special forms with an adjustable top edge. Others use what look like ordinary steel or timber forms, but they take extra care in setting them. It is not e n o u g h to set the forms level. They need to stay level till the concrete is finished. Forms need strong bracing to stand up to a heavy vibrating screed.
Striking off and finishing The m e t h o d of strike-off has a big effect on levelness, but only a smaller effect on flatness. Finishing methods work the other way around, heavily affecting flatness but doing little for levelness. Here are the main tools used for striking off, listed in order of decreasing levelness achieved: 9 9 9 9 9 9 9
h a n d straightedge riding on forms (narrow strips only); full-size laser screed; vibrating screed on narrow strips; roller screed; vibrating screed on wide strips; lightweight laser screed; h a n d straightedge on wet screeds.
W h e n it comes to finishing, the main things to consider are the a m o u n t of straightedge work and the use of pans for floating. Working the floor with h a n d straightedges, including highway straightedges, dramatically improves flatness. Floorlayers aiming for superflat tolerances normally straightedge the concrete m a n y times in two directions, scraping and filling as the concrete grows harder. The narrow-strip layout almost universal for superflat floors makes the straightedge work more convenient. Straightedges are less effective on wide-strip and large-bay pours. There floorlayers aiming for tight tolerances often choose pans for floating.
Surface Regularity
329
P a n s - especially the big ones used with riding trowel m a c h i n e s - flatten a floor better t h a n float shoes, and far better t h a n h a n d floating. Dry shakes tend to make floors less flat (but not less level) because they interfere with the finishing steps that enhance flatness. Phelan (1989) suggests that we avoid dry shakes if the required degree of flatness is higher t h a n FF45.
Superflat floors There is no universally accepted definition, but in general a superflat floor is one near the upper limit of the achievable. In the TR 34 system, the term superflat comes with a specific set of tolerances. American usage tends to be looser, but by anyone's standards an Fm~nlO0 floor would be considered superflat. In any case we are talking about a floor well over twice as flat and level as the typical concrete slab.
Figure 19.8 Superflat construction -filling a low spot that the highway straightedge has revealed
330
Design and Construction of Concrete Floors
The first recorded superflat f l o o r - for Boots in N o t t i n g h a m , E n g l a n d was laid in very narrow strips (about 2 m or 6ft wide), with screwadjustable forms. Modern practice is m u c h closer to normal narrow-strip construction, but with some extra precision and straightedge work. These steps are typical: 1. Lay out the floor in strips no more than 6 m (20 ft) wide, with construction joints in non-critical areas. 2. Set forms as accurately as possible with an optical level.
Figure 19.9 Superflat construction - marking a new slab for grinding, after testing with a profileograph
Surface Regularity
0
o
5. o
7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.
18. 19. 20.
331
Check forms with an electronic inclinometer (see Figure 11.11 in Chapter 11). Re-check forms just before pouring concrete and adjust as needed. Pour concrete at a slump of 100-125 m m (4-5 in) with minimal variation between batches. Compact concrete with poker vibrators. Strike off with a h a n d straightedge riding on the forms. Strike off again with a h a n d straightedge riding on the forms. Strike off with a 3-m (10-ft) highway straightedge at right angles to the forms. Lap each pass by 1.5 m (5 ft). Strike off third time with a h a n d straightedge riding on the forms. Strike off again with a 3-m (10-ft) highway straightedge. Wait for bleed water (if any) to form and then disappear. Power float, making passes across the slab width. Right after floating, scrape with a 3-m (10-ft) highway straightedge. Fill low spots with material scraped from high spots (see Figure 19.8). Wait as long as possible. Power trowel. If possible, scrape again with a 3-m (10-ft) highway straightedge. Filling low spots will probably not be possible at this stage, but scraping high spots may be. Continue power trowelling to produce the desired finish (usually burnished). Survey the floor and mark out-of-tolerance spots (see Figure 19.9). Grind minor defects with a 250-ram (10-in) d i a m o n d disc.
As with any list of floorlaying steps, this one should be regarded as descriptive, not prescriptive. Some floorlayers get good results from somewhat different methods. On the other hand, following these steps to the letter will not guarantee a superflat floor.
19
..
Regularity
Surface regularity has been the focus of great interest in recent years. For most of the twentieth century, floor designers either ignored surface regularity or tried to control it with specifications based on a 3-m (10-ft) straightedge. Those straightedge tolerances were rarely enforced; indeed, effective enforcem e n t was next to impossible because there was no standard test. So the situation m i g h t have remained, except for a d e v e l o p m e n t in the materials-handling field. In the early 1970s, forklift makers i n t r o d u c e d the very-narrow-aisle (VNA) turret truck. It differed from ordinary forklifts in having its fork m o u n t e d on a rotating turret. While ordinary forklifts needed aisles about 4 m (13 ft) wide, the new turret trucks could run in aisles only 2 m (6 ft) wide. Narrower aisles m e a n t greater storage d e n s i t y s o m e t h i n g m a n y warehouse users sought. The turret truck had one glaring drawback, however. It needed a flat and level floor. Warehouse users w h o tried to run turret trucks on ordinary 1970s floors soon learned that neither the 3-m (lO-ft) straightedge nor the floorlaying m e t h o d s t h e n in use were good e n o u g h for the new vehicles. To satisfy turret-truck users, floor designers and builders developed ways to build flatter and more level concrete slabs. Their efforts led to the superflat floor, discussed at the end of this chapter. Perhaps even more importantly, the sharper focus on surface regularity led people to question the flatness and levelness of all floors, n o t just those that supported turret trucks and needed a superflat surface. The result was radical new ways to specify and test floor surface regularity: the F-number system in America and the TR 34 system in the UK. But before we look at those systems, we need to consider two key distinctions: between flatness and levelness, and between r a n d o m and defined traffic.
312
Design and Construction of Concrete Floors
Flatness and levelness
These terms have o v e r l a p p i n g definitions in e v e r y d a y speech. But w i t h i n t h e field of concrete floors, flatness a n d levelness m e a n different things. Flatness is defined as planarity or lack of curvature. A flat floor is s m o o t h a n d free of b u m p s or dips. An unflat floor is b u m p y or wavy. Levelness is defined as h o r i z o n t a l i t y or lack of slope. A level floor is horizontal. An unlevel floor is sloped or tilted. A few examples m a y help illustrate the d i s t i n c t i o n b e t w e e n flatness a n d levelness. In n o r m a l use, a billiards table is b o t h flat a n d level. If we p u t bricks u n d e r o n e end, the table will still be flat b u t it will n o longer be level. O n a calm day, the surface of a p o n d is b o t h flat a n d level. If the w i n d rises t h e surface will cease to be flat, b u t it r e m a i n level. Some floors are designed to be unlevel. T h e y are sloped for drainage or to m i n i m i z e site grading. I have never hear of a n y floors designed to be unflat, t h o u g h m a n y t u r n out t h a t way. Flatness a n d levelness are b o t h desirable properties (except o n those floor with design slopes), b u t t h e y have different i m p l i c a t i o n s for t h e floor user. Flatness is critical where the m a i n issue is the b e h a v i o r of w h e e l e d vehicles. Flatness is also i m p o r t a n t where a p p e a r a n c e m a t t e r s - as in an institutional floor t h a t will be covered with vinyl tile. In contrast, levelness is critical where the m a i n issue is the installation of fixed e q u i p m e n t such as w a r e h o u s e racks or m a c h i n e tools. While exceptions exist, m o s t users find flatness m a t t e r s m o r e t h a n levelness. If a floor is unlevel, fixed e q u i p m e n t can be s h i m m e d or jacked u p to a c c o m m o d a t e it. But it is n o t so easy to a d a p t a w h e e l e d vehicle to a surface t h a t is n o t flat e n o u g h for it. Flatness a n d levelness also have different implications for the floorlayer. Flatness comes from finishing; it is d e t e r m i n e d m a i n l y by the finishing m e t h o d s , a n d to a lesser extent by the finishability of the concrete mix. Levelness comes from forms a n d strike-off; it is d e t e r m i n e d m a i n l y by the accuracy of the side forms and the m e t h o d used to strike of the concrete.
D e f i n e d versus r a n d o m t r a f f i c
Floors get two kinds of traffic: defined a n d r a n d o m . O n a defined-traffic floor, vehicles are c o n f i n e d to fixed paths. O n a r a n d o m - t r a f f i c floor, vehicles (or walkers) are free to roam, t h o u g h inevitably some traffic p a t h s see m o r e use t h a n others. The distinction matters because the two kinds of traffic call for different ways of m e a s u r i n g surface regularity. O n a defined-traffic floor, y o u can
Surface Regularity
313
m e a s u r e a c o n t i n u o u s , or nearly c o n t i n u o u s , profile d o w n every travel path. You can check effectively all of the floor profile t h a t m a t t e r s to t h e vehicle user. In contrast, a random-traffic floor has an infinite n u m b e r of possible travel paths. You could never m e a s u r e t h e m all, so y o u h a v e to fall back o n statistical sampling. You check selected points or lines a n d h o p e t h e y represent the w h o l e surface. Historically, the highest r e q u i r e m e n t s for surface regularity have b e e n f o u n d in the defined-traffic category. The reason is t h a t turret t r u c k s - t h e vehicles t h a t d e m a n d the flattest a n d m o s t level f l o o r s - r u n in defined paths. It w o u l d be wrong, however, to assume t h a t defined-traffic floors are always superior in surface regularity. T h o u g h defined-traffic floors are i m p o r t a n t , their random-traffic cousins are far more n u m e r o u s . Defined-traffic floors are f o u n d in very-narrow-aisle warehouses a n d in buildings w i t h a u t o m a t i c a l l y - g u i d e d v e h i c l e s - a n d a l m o s t n o w h e r e else. The great m a j o r i t y of industrial floors, as well as all n o n - i n d u s t r i a l floors, fall in the random-traffic category.
F-numbers Since its a d o p t i o n into ACI a n d ASTM standards in the late 1980s, t h e F - n u m b e r system has b e c o m e the m o s t c o m m o n way to specify a n d measure floor surface regularity in the U n i t e d States a n d Canada. It is also widely used in Latin America a n d Australia. Europeans, in contrast, are m o r e likely to use the TR 34 system or t h e G e r m a n standards DIN 18 202 a n d DIN 15 185. The F - n u m b e r system includes three separate n u m b e r s : 9 Fmin, w h i c h e n c o m p a s s e s b o t h flatness a n d levelness a n d is used o n l y o n defined-traffic floors; 9 FF, w h i c h controls flatness a n d is used o n l y o n random-traffic floors; 9 FL, w h i c h controls levelness a n d is used o n l y o n random-traffic floors. Defined-traffic floors are specified by the single n u m b e r Fm~n. R a n d o m traffic floors are n o r m a l l y specified by a pair of n u m b e r s , w r i t t e n as FF_/FL_. However, some s u s p e n d e d random-traffic floors are specified by F F alone. The three n u m b e r s have this m u c h in c o m m o n : 9 A h i g h e r F - n u m b e r always d e n o t e s a better floor. 9 The n o r m a l scale runs from 10 to 100. O n l y the worst floors exhibit F-numbers below 10, a n d o n l y a few of t h e best m e a s u r e above 100. 9 The scales are linear. If y o u double the F-number, y o u halve t h e m a g n i t u d e of the allowed deviations.
Design and Construction of Concrete Floors
314
Fmi n for defined traffic floors
The F-number system controls defined-traffic floors by means of a single number called Frnin. Fmin defines both flatness and levelness, and is measured over the wheel spacings of the vehicle that will travel on the floor. Testing is confined to the travel paths. W h e n you say a floor measures, say, Frown50, you are not making a general statement about the floor's flatness and levelness. You are saying that when a particular vehicle travels in a particular path, it will not encounter any slope or change in slope greater t h a n that allowed by the Fmin equations, shown below: In SI units" [6.55 (L +69) ~ - 4 8 ] Gin
dmax= elTlaX
33.3 F~in
where" m a x i m u m elevation difference between points separated by L, in rnm; emax m a x i m u m rate at which elevation difference between points separated by L can change with travel across the floor, in mrn/lOOmm; L distance between wheels (transverse) or axles (longitudinal), in rnm; G i n ~ the defined-traffic F-number (dimensionless).
dmax
In fps units:
dmax emax
[1.3 (L + 2.7)~
Fmin 4.00 Fmin
where:
dmax - m a x i m u m elevation difference between points separated by L, in inches; = m a x i m u m rate at which elevation difference between points emax separated by L can change with travel across the floor, in inches/foot;
Surface Regularity
315
L - distance between wheels (transverse) or axles (longitudinal), in inches; F m i n = the defined-traffic F-number (dimensionless). The tolerances dma x and emax are usually expressed to 0.1 m m or 0.001 in. The e q u a t i o n s for dma x (levelness) are good for wheel or axle spacings from 300 m m to 3 m (1 to 10 ft). The e q u a t i o n s for emax (flatness) are good for wheel or axle spacings from 1 to 3 m (40 in to 10 ft). Almost all vehicles used on defined-traffic floors have d i m e n s i o n s w i t h i n those ranges. Fm~n is normally measured with a differential profileograph set up to m a t c h the wheel and axle spacings of the defined-traffic vehicle. The profileograph measures the elevation difference between left and right wheels (the transverse-differential profile) and between front and rear axles (the longitudinal-differential profile). Both profiles are checked for compliance with the calculated limits for dmax and emax. Specified Fmin values range from 40 (rarely lower) to 100 (never higher). Fm~nlO0 is the traditional superflat specification, and is reserved for VNA warehouses with very tall racks or very high t h r o u g h p u t , or both. In recent years, Fmin60 has become popular for less-demanding VNA applications. With laser screeding, contractors can lay Fmin60 floors in large bays, eliminating the costly narrow-strip layout needed for Fm~nlO0. Fmi~ 40 is sometimes specified for the least d e m a n d i n g VNA applications, and also for low-rise automatically-guided vehicles.
FF/F L for random-traffic floors
The F-number system controls random-traffic floors with a pair of numbers, F F and F L. The flatness number, F~, is based on the floor curvature over 600 m m (24 in) (see Figure 19.1). The levelness number, F L, is based on the floor slope over 3 rn (10 ft) (see Figure 19.2). ASTM E 1155 (fps units) and ASTM E 1155M (SI units) describe the standard test for F F and F L. A typical survey involves these steps: 1. Determine the m i n i m u m n u m b e r of 3-rn (lO-ft) slope readings. For test sections of 150 m 2 or more, the m i n i m u m is the area divided by 3. (In fps units, it's the area in ft 2 divided by 30, for sections of 1600 ft 2 or more.) 2. Lay out e n o u g h straight lines on the floor to provide at least the m i n i m u m n u m b e r of 3-m (lO-ft) slope readings. The n u m b e r of readings from each line is equal to the length in m, minus 2.7 m, divided
316
Design and Construction of Concrete Floors ii
Floor surface
7 J j
er"
/' ~
600 mm (24 in) curvature
j ~-- 300 mm --~ ~-- 300 mm --~ (12 in)
(12 in)
Figure 19.1 600-mm (24-in) curvature determines the flatness number, F~. TR 34 calls this Property il
by 0.3 m. (In fps units, the n u m b e r of readings from each line is simply the length in ft, minus 9 ft.) Figure 19.3 shows one possible layout, with all lines oriented 45 ~ to the test section's longest side. It is also permissible, where the test section is at least 7.5 m (25 ft) wide, to lay out runs parallel and perpendicular to the longest side. Under most circumstances, survey lines must avoid slab edges, construction joints and penetrations by at least 600 rnm (2 ft). 3. Measure point elevations on 300-ram (12-in) centres down each survey line. 4. Calculate the elevation difference between each pair of adjacent points. This value is the 300-rnm (12-in) slope. 5. Calculate the difference between each pair of consecutive 300-rnrn (12-in) slope readings. This is the 6 0 0 - m m (24-in) curvature.
Floor surface
I.~
3 m (10 ft)
~
m (lOt) slope
....
r
Figure 19.2 3-m (lO-ft) slope determines the levelness number, FL. TR 34 calls this Property IV
Surface Regularity
600 mm (24 in) ~1~
Floor edge or construction joint |
317
Profile runs
.....
tm" v
E ----~ E 0 0 CD
600 mm (24 in)
Figure 1 9 . 3
One way to lay out an F-number survey is w i t h all lines at 45 ~ to the slab's
longest side
6. C a l c u l a t e t h e F F n u m b e r f r o m o n e of t h e f o l l o w i n g e q u a t i o n s : In SI units:
FF (3Sq114+ q) where: F F = t h e r a n d o m - t r a f f i c flatness n u m b e r ( d i m e n s i o n l e s s ) ; q = 6 0 0 - m m c u r v a t u r e , in mrn; Sq = s t a n d a r d d e v i a t i o n of q; q = m e a n of q. In fps u n i t s :
FF
4.57 (3Sq + q)
where: F F = t h e r a n d o m - t r a f f i c flatness n u m b e r ( d i m e n s i o n l e s s ) ; q = 2 4 - i n c u r v a t u r e , in i n c h e s ; S q - s t a n d a r d d e v i a t i o n of q; q = m e a n of q.
Design and Construction of Concrete Floors
318
7. Calculate t h e s u m of each set of t e n c o n s e c u t i v e 3 0 0 - m m (12-in) slopes. This is the 3 - m (l O-fl) slope. 8. Calculate the F L n u m b e r from o n e of t h e following equations" In SI units: 315 EL -- (3S~ + z) where: F L -- t h e r a n d o m - t r a f f i c levelness n u m b e r (dimensionless); q-- 3-m slope, in m m ; S z - s t a n d a r d deviation of z; z - m e a n of z. In fps units" EL ~-
12.5 (3Sz+Z)
where: t h e r a n d o m - t r a f f i c levelness n u m b e r (dimensionless); z -- l O-ft slope; S z = s t a n d a r d d e v i a t i o n of z; z -- m e a n of z.
F L --
ASTM standards allow a variety of tools for m e a s u r i n g F F and F L, including optical levels, lasers a n d levelled straightedges w i t h d e p t h gauges. However, a l m o s t e v e r y b o d y uses an electronic i n c l i n o m e t e r - either t h e w a l k i n g t y p e w i t h feet 3 0 0 m m (12 in) a p a r t (the D i p s t i c k is t h e b e s t - k n o w n b r a n d n a m e ) , or t h e r o l l i n g t y p e w i t h w h e e l s s e p a r a t e d by t h e s a m e d i s t a n c e . Figure 11.11 in C h a p t e r 11 s h o w s a n e l e c t r o n i c i n c l i n o m e t e r . The resulting F - n u m b e r s will be practically t h e same w h e t h e r surveyed in SI or fps units. While no fixed relationship exists b e t w e e n F F a n d F L, the former averages about 40% higher t h a n the latter o n g r o u n d - s u p p o r t e d slabs. O n s u s p e n d e d floors the ratio of F F to F L tends to r u n higher. ACI 117 r e c o m m e n d s t h a t a two-tiered f o r m a t for r a n d o m - t r a f f i c F - n u m b e r specifications. The overall F - n u m b e r s apply to the floor t a k e n as a whole. The m i n i m u m local F - n u m b e r s apply to smaller sections of t h e floor. M i n i m u m local n u m b e r are always lower t h a n the overall n u m b e r s typically one third to o n e half lower. The two-tiered format, t h o u g h sometimes criticized as needlessly complex, has p r o v e n an effective tool for i m p r o v i n g floor flatness a n d levelness.
Surface Regularity
3119
It pushes builders t o w a r d h i g h e r n u m b e r s , w i t h o u t p u n i s h i n g t h e m brutally for the occasional bad day. The goal is to lay t h e w h o l e floor to the overall F-numbers. But if the floorlayers do worse t h a n that on an individual slab, or part of a slab, it can still be accepted provided it meets the m i n i m u m local F-numbers. Such an occurrence serves as a w a r n i n g t h a t the crew m u s t do better o n later slabs, to b r i n g the overall n u m b e r s up to the specified values. If a test section fails to m e e t even the m i n i m u m local values, it m u s t be repaired or replaced. But such failures are rare if the n u m b e r are c h o s e n w i t h care a n d if all parties u n d e r s t a n d t h e rules from t h e start. Table 19.1 shows overall a n d m i n i m u m local F-numbers for four classes of surface regularity. The classification comes from ACI 117, b u t the r e c o m m e n d e d applications are m i n e .
Construction joints ASTM E 1155 and ASTM E 1155M forbid measuring F F and F L within 600 m m (24 in) of a n y slab edge or c o n s t r u c t i o n joint. (There is an e x c e p t i o n for very small slabs.) Because the edges a n d c o n s t r u c t i o n joints are statistical anomalies, i n c l u d i n g t h e m in t h e calculations w o u l d distort the results. Table 119.1 Classification of floor flatness and levelness by F-numbers - for random-traffic floors (ACI 11 7)
Minimum F-numbers required Floor class
Application
Very flat
Warehouses with reach trucks over 8 m (26 ft) high, random-traffic areas in some VNA warehouses, air-bearing systems Offices with moveable partitions, warehouses with mobile racks or reach trucks up to 8 m (26 ft high), busy generalpurpose warehouses
FF50/FL30
FF25/FL15
FF301FL20
FFI 5/FLI 0
Warehouses and factories where surface regularity is not critical Foot traffic, garages
FF2OIFL15
FF151FL10
FF15/FL1 3
FF1 3/FL10
Flat
Conventional Straightedged
Bull-floated
Overall
Local
320
Design and Construction of Concrete Floors
The exclusion of construction joints from testing causes some problems, because the areas adjacent to construction joints are o f t e n - I would say u s u a l l y - the least flat parts of a floor. It is hard to justify a system t h a t excludes the spots where bad readings are most likely to occur. The 1996 editions of ASTM E 1155 and ASTM E 1155M offer a solution. Designers can specify a limit for 600-ram (12-in) curvature at construction joints. The usual tolerance is 3.0 or 4.0 m m (0.125 or 0.150 in). The standard test calls for at least one test for each 3-m (lO-ft) length of joint. Enforcement is normally limited to those construction joints that will be subject to traffic.
Suspended floors Random-traffic F-numbers apply to suspended floors, but with some restrictions. American standards let us enforce the flatness number, FF, on any type of suspended floor. We need to recognize, however, that F F values above 30 can be very hard to achieve on suspended slabs. While making a suspended slab flat is hard, m a k i n g it level is even harder. In view of that fact, ACI 117, the m a i n American standard for concrete tolerances, prohibits the e n f o r c e m e n t of F L tolerances on susp e n d e d floors, with the exception of fully shored slabs that are tested while the shores are still in place. U n s h o r e d floors, including most slabs on metal deck, are excluded from levelness testing. If a suspended floor needs to be very flat or very l e v e l - luckily most do n o t - the best choice is usually a t o p p i n g or screed.
The TR 34 system The Concrete Society's Technical Report 34, Concrete Industrial Ground Floors, c o m m o n l y called just TR 34, includes a thorough, well-thought-out system for specifying floor flatness and levelness on industrial floors. The TR 34 system shares m a n y features with F-numbers - a fact that stems from their c o m m o n origin in research carried out by Allen Face in the early 1980s. Like F-numbers, TR 34 controls b o t h flatness and levelness. Again like F-numbers, TR 34 has different specifications for defined traffic and r a n d o m traffic, which TR 34 calls "defined m o v e m e n t " and "free m o v e m e n t " . Perhaps the most significant difference between F-numbers and TR 34 lies in the way tolerances are presented. F-numbers are continuously variable.
Surface Regularity
321
The designer must choose a n u m b e r and can, within practical limits, choose any number. While ACI 117 does classify floors (see Table 19.1), designers remain free to specify other values, and m a n y do so. TR 34, in contrast, divides floors into a small n u m b e r of standard categories, each having its own set of tolerances. The designer does not set a tolerance, but only picks a category. I confess a slight prejudice in favour of the F-number approach, while recognizing that standard categories have merit, too.
TR 34 properties At the heart of the TR 34 system are four properties, identified by R o m a n numerals I-IV. Property I (see Figure 19.4) is the elevation difference between two points separated by 300 m m (12 in). This is a levelness property. It applies only to defined-traffic floors, where it is measured parallel to the direction of travel. Property II (see Figure 19.1) is the arithmetic difference between a pair of consecutive Property I readings. This is a flatness property. It applies both to defined-traffic and random-traffic floors. On defined-traffic floors it is measured parallel to the direction of travel. On random-traffic floors it is measured in two directions. Property II is the same as 600-ram (24-in) curvature in the F-number system. Property III (see Figure 19.5) is the elevation difference between the left and right wheelpaths in which a vehicle runs. This is a levelness property. It applies only to defined-traffic floors, where it is measured in the direction perpendicular to travel, with readings taken on 300-ram (12-in) centres. It is the same as transverse levelness in the F-number system.
Direction of travel
t
Floor
Property I
i-~
Figure 1 9 . 4
TR 34 P r o p e r t y I
9
300 mm (12 in)
,,-
322
Design and Construction of Concrete Floors
Floor surface Load axle of vehicle
Property I!1
I_ i-~
Vehicle wheeltrack
Figure 19.5 TR 34 Property III
Property IV (see Figure 19.2) is the elevation difference between two points separated by 3 rn (10 ft). This is a levelness property. It applies only to random-traffic floors, where it is measured on a 3-m (10-ft) grid. It is the same as 3-m (10-ft) slope in the F-number system.
TR 34 for defined-traffic floors
TR 34 divides defined-traffic floors into three categories of surface regularity, called Superflat, Category 1 and Category 2. Each category is associated with limits on Properties I, II and III (see Table 19.2). Table 19.2 TR 34 tolerances for defined-traffic floors Tolerance Property I
Property II
Property III Wheeltrack up to 1.5 m (60 in)
Up to 1.5 m (60 in)
Category
95%
100%
95%
100%
95%
100%
95%
100%
Superflat
0.75 mm 0.030 in 1.5 mm 0.059 in 2.5 mm 0.098 in
1.0mm 0.039 in 2.5mm 0.098 in 4.0mm 0.157in
1.0mm 0.039 in 2.5mm 0.098 in 3.25mm 0.126in
1.5mm 0.059 in 3.5mm 0.1 38 in 5.0mm 0.197in
1.5ram 0.059 in 2.5mm 0.098 in 3.5mm 0.138in
2.5mm 0.098 in 3.5mm 0.1 38 in 5.0mm 0.197in
2.0mm 0.079 in 3.0mm 0.118in 4.0mm 0.157in
3.0ram 0.118in 4.5mm 0.1 77 in 6.0mm 0.236in
1 2
Surface Regularity
323
Note that TR 34 imposes two limits on each property. No readings are allowed to exceed the 100% limit, but up to 5% of the readings m a y exceed the 95% limit. The use of two limits represents an a t t e m p t to control not just the extreme values, but also the distribution of readings below those extremes. TR 34 ties the r e c o m m e n d e d category to the lift height of the truck, as follows: 9 Superflat for lift height over 13 rn (43 ft); 9 Category 1 for lift height of 8-13 m (26-43 ft); 9 Category 2 for lift height under 8 m (26 ft). That's useful, but it oversimplifies the situation. The need for surface regularity depends not only on lift height, but also on vehicle speed, vehicle type and the overall d e m a n d s of the floor user. Since the TR 34 properties are based on readings taken at 300-ram (12-in) intervals, floors specified in this way can be tested with same electronic i n c l i n o m e t e r s used to measure r a n d o m - t r a f f i c floors in t h e F - n u m b e r system. In practice, however, e v e r y b o d y uses differential profileographs like those used to m e a s u r e Fmi n. But while a n Fmi n profileograph is set up to m a t c h the wheel pattern of the vehicle, a TR 34 profileograph is set up to m a t c h just one axle of the vehicle (normally the wider of the two axles, if they differ). This takes care of Property III. Trailing wheels 300 m m (12 in) b e h i n d the axle allow m e a s u r e m e n t of Properties I and II. All this m a y be changing, however. The latest version of TR 34 (Concrete Society, 2003) presents, in Appendix C, an alternative m e t h o d for definedtraffic floors. The alternative m e t h o d is m u c h closer to the way definedtraffic floors are measured in the F-number system. It relies on full wheel-pattern testing, with tolerances adjusted for the distance between wheels. Floors are divided into three categories, called DM (for definedm o v e m e n t ) 1, 2 and 3. The classes correspond roughly to the current categories Superflat, 1 and 2.
TR 34 for random-traffic floors
TR 34 divides random-traffic floors into three categories of surface regularity, called FM (for free-movement) 1, 2 and 3. Each category is associated with limits on Properties II and IV. (see Table 19.3). As with the defined-traffic floors, each Property is subject to b o t h a 95% and a 100% limit. For a floor to meet its specification, no more t h a n 5% of the readings can exceed the 95% limits, and no readings can exceed the 100% limits.
324
Design and Construction of Concrete Floors
Table 119.] TR 34 tolerances for random-traffic floors Tolerance Property II
Property IV
Category
Application
95%
100%
95%
100%
FM 1
Where very high standards of flatness and levelness are required Warehouses with stacking over 8 m (26 ft) high, transfer areas in VNA warehouses Warehouses with stacking up to 8 m (26 ft) high, factories, retail stores
2.5 mm 0.098 in
4.0 mm 0.157 in
4.5 mm 0.177 in
7.0 mm 0.276 in
3.5 mm 0.138 in
5.5 mm 0.217 in
8.0 mm 0.315 in
12.0 mm 0.472 in
5.0mm 0.197 in
7.5mm 0.295 in
10.0mm 0.394 in
15.0mm 0.591 in
FM 2
FM 3
Despite the fact that Properties II and IV are identical to the properties that determine F F and F L in the F-number system, they are n o t surveyed in exactly the same way. TR 34 calls for testing random-traffic floors on a 3-m grid. Property IV is measured between the grid points. Property II is measured along the some of the lines that c o n n e c t the grid points, chosen so the total line length in metres equals or exceeds one t e n t h the floor's area in square metres. Unlike an F~/FL survey, a TR 34 random-traffic survey normally calls for two separate instruments: an optical level to check Property IV, and a rolling or walking gauge to check Property II. Despite the differences in testing, we can still reasonably compare the FM categories to F-numbers (see Table 19.4). Interestingly, British and American specifications are fairly close on flatness but differ substantially on levelness. TR 34 imposes levelness requirements t h a t American floorlayers would consider onerous. One reason for this difference m a y be that TR 34 explicitly addresses industrial floors, w h i c h are usually g r o u n d - s u p p o r t e d and are often laid to high standards. American standards, in contrast, deal with a wider range of floors, including susp e n d e d slabs. Table 119.4 Random-traffic specifications compared TR 34 category
FM 1 FM 2 FM 3
Equivalent F-numbers
FF30/FL46 F~22/626 FF16/FL21
Surface Regularity
325
TR 34 on overall elevation
Properties I-IV do n o t directly control the floor's overall elevation. For m o s t uses, overall elevation matters less t h a n local flatness a n d levelness - b u t t h a t is n o t to say it does n o t m a t t e r at all. TR 34 r e c o m m e n d s an overall tolerance of + 15 m m (+ 5/8 in) from d a t u m , for all floor classes.
Straightedge tolerances Some specifiers continue to specify floor flatness by reference to a 3-m (10-ft) straightedge (see Figure 19.6). This m e t h o d has two serious drawbacks, however. The first is the so-called w a s h b o a r d l o o p h o l e (see Figure 19.7). C h e c k e d w i t h a straightedge b o t h profiles in Figure 19.7 are equal. Yet for a l m o s t a n y use, the profile o n top is clearly the better of t h e two. The second drawback is the lack of a s t a n d a r d test. W h i l e ACI 117 a n d BS 8204:Part 2 b o t h offer a bit of g u i d a n c e in straightedge use, n e i t h e r comes close to p r o v i d i n g a c o m p l e t e test m e t h o d like those available for F-numbers a n d TR 34 tolerances. No s t a n d a r d answers these questions: 9 W h e r e is the floor measured? 9 How m a n y tests n e e d to be m a d e ? 9 Are a n y failures allowed - a n d if so, h o w m a n y ? W i t h o u t answers to those questions, a n y o n e trying to enforce a straightedge specification is likely to face a dispute.
3 m (lOft)
Figure 19.6
The straightedge test
326
Design and Construction of Concrete Floors
3 mm (1/8 in) f
\
3 m m (I18 in)
I-~
3 m (lOft)
Figure 19.7 The washboard loophole
Factors that affect surface regularity These are the m o s t i m p o r t a n t factors: 9 9 9 9 9
structural deflection; curling; w i d t h of pour; accuracy of side forms; striking off a n d finishing.
Structural deflection Suspended floors all deflect, a n d m o r e t h a n a few deflect excessively. Deflection is h a r d to control a n d predict. Deflection has a h u g e effect o n levelness, a n d a smaller b u t still significant effect o n flatness. Deflection is a serious p r o b l e m for s u s p e n d e d floors t h a t n e e d to be level. ACI 117 recognizes this by s i m p l y e l i m i n a t i n g the F L levelness r e q u i r e m e n t o n m o s t s u s p e n d e d floors. It says "levelness tolerance shall n o t apply to slabs placed o n u n s h o r e d form surfaces a n d / o r shored form surfaces after the r e m o v a l of shores". In effect, the ACI standard is telling us n o t to use o r d i n a r y s u s p e n d e d slabs w h e r e levelness matters.
Surface Regularity
327
That is good advice, to the extent we can follow it. But some s u s p e n d e d floors do need to be level, no matter w h a t ACI 117 tells us. W h a t are our options? One answer is to reduce the deflections. Chapter 6 contains some ideas on that. Another answer is to lay a t h i n topping over the suspended structural slab. But a topping w o n ' t stop the floor deflecting under loads applied later. Deflection is seldom a problem on g r o u n d - s u p p o r t e d floors.
Curling This is a major cause of poor surface regularity o n g r o u n d - s u p p o r t e d floors, concrete toppings a n d slabs cast o n metal deck. Testing flatness a n d levelness during c o n s t r u c t i o n usually fails to detect curling, because the floor curls afterwards. Chapter 14 discusses curling a n d presents some ways to m i n i m i z e its effect o n floor flatness and levelness.
Width of pour The narrower the slab, the easier is it to m a k e the floor flat a n d level. The m a i n reason for this is t h a t a n a r r o w p o u r allows a m o r e accurate strike-off, at least w i t h traditional (before the laser screed) m e t h o d s . Superflat floors are almost always laid in n a r r o w strips no m o r e t h a n 6 rn (20 ft wide). In wide-strip construction, where the floor is struck off w i t h a vibrating screed, floor levelness tends to deteriorate as the pours get wider. Some vibrating screeds can span 24 rn (80 ft), b u t y o u c a n n o t expect t h e m to m a i n t a i n a very level surface over t h a t distance. Laser screeding reduces the c o n n e c t i o n b e t w e e n slab w i d t h a n d surface regularity. With a laser screed, some floorlayers can achieve good flatness and levelness o n pour as m u c h as 30 rn (100 ft) wide.
Accuracy of side forms Side forms m a t t e r because t h e y d e t e r m i n e the elevation to w h i c h the floor is struck off. Forms have a big effect on levelness, b u t m u c h less effect o n flatness.
328
Design and Construction of Concrete Floors
Laser screeding does not eliminate the importance of accurate side forms. Though the laser screed itself does not rely on the forms, its action stops about 0.5 m (2-3 ft) from the slab edges. The accuracy of the forms determines the surface regularity of that narrow strips between the slab edge and the laser-screeded surface. With ordinary levels, it is hard to set forms closer t h a n +3 m m (+1/8 in). But that is good enough for most floors. Superflat floors need better forms. Some floorlayers who specialize in superflat work use special forms with an adjustable top edge. Others use what look like ordinary steel or timber forms, but they take extra care in setting them. It is not e n o u g h to set the forms level. They need to stay level till the concrete is finished. Forms need strong bracing to stand up to a heavy vibrating screed.
Striking off and finishing The m e t h o d of strike-off has a big effect on levelness, but only a smaller effect on flatness. Finishing methods work the other way around, heavily affecting flatness but doing little for levelness. Here are the main tools used for striking off, listed in order of decreasing levelness achieved: 9 9 9 9 9 9 9
h a n d straightedge riding on forms (narrow strips only); full-size laser screed; vibrating screed on narrow strips; roller screed; vibrating screed on wide strips; lightweight laser screed; h a n d straightedge on wet screeds.
W h e n it comes to finishing, the main things to consider are the a m o u n t of straightedge work and the use of pans for floating. Working the floor with h a n d straightedges, including highway straightedges, dramatically improves flatness. Floorlayers aiming for superflat tolerances normally straightedge the concrete m a n y times in two directions, scraping and filling as the concrete grows harder. The narrow-strip layout almost universal for superflat floors makes the straightedge work more convenient. Straightedges are less effective on wide-strip and large-bay pours. There floorlayers aiming for tight tolerances often choose pans for floating.
Surface Regularity
329
P a n s - especially the big ones used with riding trowel m a c h i n e s - flatten a floor better t h a n float shoes, and far better t h a n h a n d floating. Dry shakes tend to make floors less flat (but not less level) because they interfere with the finishing steps that enhance flatness. Phelan (1989) suggests that we avoid dry shakes if the required degree of flatness is higher t h a n FF45.
Superflat floors There is no universally accepted definition, but in general a superflat floor is one near the upper limit of the achievable. In the TR 34 system, the term superflat comes with a specific set of tolerances. American usage tends to be looser, but by anyone's standards an Fm~nlO0 floor would be considered superflat. In any case we are talking about a floor well over twice as flat and level as the typical concrete slab.
Figure 19.8 Superflat construction -filling a low spot that the highway straightedge has revealed
330
Design and Construction of Concrete Floors
The first recorded superflat f l o o r - for Boots in N o t t i n g h a m , E n g l a n d was laid in very narrow strips (about 2 m or 6ft wide), with screwadjustable forms. Modern practice is m u c h closer to normal narrow-strip construction, but with some extra precision and straightedge work. These steps are typical: 1. Lay out the floor in strips no more than 6 m (20 ft) wide, with construction joints in non-critical areas. 2. Set forms as accurately as possible with an optical level.
Figure 19.9 Superflat construction - marking a new slab for grinding, after testing with a profileograph
Surface Regularity
0
o
5. o
7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.
18. 19. 20.
331
Check forms with an electronic inclinometer (see Figure 11.11 in Chapter 11). Re-check forms just before pouring concrete and adjust as needed. Pour concrete at a slump of 100-125 m m (4-5 in) with minimal variation between batches. Compact concrete with poker vibrators. Strike off with a h a n d straightedge riding on the forms. Strike off again with a h a n d straightedge riding on the forms. Strike off with a 3-m (10-ft) highway straightedge at right angles to the forms. Lap each pass by 1.5 m (5 ft). Strike off third time with a h a n d straightedge riding on the forms. Strike off again with a 3-m (10-ft) highway straightedge. Wait for bleed water (if any) to form and then disappear. Power float, making passes across the slab width. Right after floating, scrape with a 3-m (10-ft) highway straightedge. Fill low spots with material scraped from high spots (see Figure 19.8). Wait as long as possible. Power trowel. If possible, scrape again with a 3-m (10-ft) highway straightedge. Filling low spots will probably not be possible at this stage, but scraping high spots may be. Continue power trowelling to produce the desired finish (usually burnished). Survey the floor and mark out-of-tolerance spots (see Figure 19.9). Grind minor defects with a 250-ram (10-in) d i a m o n d disc.
As with any list of floorlaying steps, this one should be regarded as descriptive, not prescriptive. Some floorlayers get good results from somewhat different methods. On the other hand, following these steps to the letter will not guarantee a superflat floor.