Shear keys are not an effective means of horizontal shear transmis- sion. Push-off tests(1) have demon- strated that a considerable amount of slip at the interface is required to this actuate the shear keys, and movement is large enough to break the concrete-to-concrete bond. There- fore, the effects of shear keys and bond are not additive. "It should be assumed in design that the entire shear is transferred either by bond or by shear keys.”(9)
Steel ties
The function of the steel ties across the interface is two-fold: (a) to pre- vent vertical separation caused by secondary tensile stresses, and (b) to resist horizontal slip caused by hori- zontal shear stresses.
A small quantity of steel is requir- ed to "tie down" the cast in-situ slab to the precast beam. A minimum of 0.15% of the area of the interface is recommended. Greater quantities of steel will improve the efficiency of the concrete-to-concrete bond, thereby increasing the horizontal shear re- sistance of the interface.
Exploratory tests carried out to investigate the performace of steel stirrups as shear connectors(10) have indicated that stirrups which offer direct resistance to the horizontal shear but do not tie the two com- ponents together are ineffective. Futher tests(11) have shown that monolithic action is lost when the vertical separation exceeds .001". It is important that steel ties be firmly anchored at both ends, preferably in the form of closed stirrups, and dis- tributed at a close spacing.
Similar to shear keys, steel ties can- not develop their direct shear resis- tance until an appreciable amount of slip has occurred at the interface. It has been generally agreed that this amount of slip is large enough to break the bond at the interface. For example, Samuely(12) stated in 1952: "If projecting steelwork is used, it is equivalent to assuming that separa- tion may take place between the pre- cast and in-situ concrete, and that the steel reinforcement is strong enough to take all stresses which would occur."
Bennett(13)
Similarly, Evans and suggested in 1958 that "projecting re- inforcement should be designed on the assumption of total failure of the bond, and should develop sufficient shear resistance to balance the shear- ing force at the connecting surface." In the Standard Specification for Bridge Beams, issued by the Division of Roads and Road Traffic (Southern Rhodesia) in 1960,014) it is stipulated that "mild steel stirrups shall be in- corporated in the beams to carry the total shear acting at the beam-topping interface."
Design Recommendations
Except in short-span composite beams subjected to heavy concentrat- ed loading, a rough bonded interface is an adequate connection. To im- prove the efficiency of the connection
50
50
TABLE 1
ULTIMATE SHEAR STRESS IN COMposite beaM
with shear keys and stirrups
Cube strength Type of
Ultimate
of in-situ
contact
horizontal
Remarks
Investigator
concrete
surface
shear stress
(lb. in.2)
Smooth
(lb./in.2)
150
19
Samuely(2)
500
2300
310-345
Graf & Weil(3)
2300
365
with shear keys at beam ends
2500-3100
120
Meyer(4)
2500-3100
105-170
with shear keys
-1-
2300
215
"
Gessner(5)
1700
240-285
with .12% steel
6500*
98-208
Ozel](6)
6300*
114
with .82% steel
5
4100*
55
280+
with .34% steel
3500*
340‡
with .46% steel
Hanson(1)
3700*
Rough
320‡
3000*
455‡
with .34% steel
3700*
4301
with .46% steel
"
Kaar et al.(7)
4000*
385
with .41% steel
Mattock & Kaar(8)
3650*
310
with .17% steel
*
‡
converted from cylinder strength.
at a horizontal slip of .005”.
N.B. The quantity of steel across the interface is expressed as a percentage
of the area of the interface.
and to prevent any probable vertical separation, a small quantity of steel (not less than 0.15%) should be pro- vided across the interface, in the form of firmly-anchored and closely-spaced stirrups, along the entire span of the beam. Shear keys are unnecessary. unless the horizontal shear exceeds the shear capacity of the rough bond- ed surface: and even so, steel stirrups effective as mechanical
are more
shear connectors.
On the question of permissible horizontal shear stresses, there have been diverse opinions and specifica- tions:
(i) Samuely(2) suggested the follow- ing "safe working shear stresses": On smooth interface without mechanical anchorage
40 lb/in2 On smooth inter- face with castellations and steel ties 125 lb. in,2
(ii) Rühle(15) recommended the fol- lowing limits for horizontal shear stresses on a smooth in- terface:
Concrete strength
(lb.in.2) 3200 4250 6400 Shear stress at
working load
(lb. in.2)
85
115
130
140
Shear stress at ultimate load (lb. in.2)
170 200 (iii) Hanson(1) estimated the maxi- mum shear strength of a smooth interface as 300 lb. in.2 and that of a rough interface as 500 lb. in.2 To these values, 175 lb./in.2 may be added for each percent of steel across the interface.
(iv) Saemann and Washa(16) found from their experimental investi- gation that the shear strength of the connection is a function of the percentage of steel across the interface and the ratio of the shear span to the effective depth.
The following empirical formu- la was propounded:
S=
2700 X + 5 where s =
X
33-X
+300P (x2 +6x+5
the shear strength of the connection,
the ratio of shear span to effective depth.
Р
www
the
percentage
of
steel across the in- terface.
Far East Architect & Builder February, 1967