the bond reduction for both in-fire status and after-fire status. ... all tests a
significant reduction of steel-concrete bond was found as the result of fire
temperature.
TESTING STEEL - CONCRETE BOND IN FIRE CONDITIONS
1
2
Zoja Bednarek , Paweł Ogrodnik
1 2
The Main School of Fire Service, ul.Słowackiego 52/54, Warsaw, Poland,
[email protected]
The Main School of Fire Service, ul.Słowackiego 52/54, Warsaw, Poland,
[email protected]
Received
; accepted
Abstract. The article presents results from the research on fire temperature influence on steel-concret bond and on
the bond reduction for both in-fire status and after-fire status. In the paper has been described bond test and results for materials St35, 18G2 reinforce ment steel and C16/20, C40/50 concerete, both for in the fire and after-fire conditons. For all tests a significant reduction of steel-concrete bond was found as the result of fire temperature. It was proven, that significant bond propert differences exist between in-fire tests and after-fire tess, which wvidence that the bond is regained partially after the exposure.
Keywords: bond, concrete, fire conditions
1. Introduction The objective of our research was determining the character and data for the steel bond reduction in concrete under thermal conditions of normal fire, with the temperature-vs.-time relation defined by the formula T=20+345*log(8t+1)
(1)
whereas: T - temperature [oC], t - time [min]. and with a similar curve to that of the temperature distribution on concrete plate surface. Tests were made on two sample groups in order to compare results [6]. Tests were made while the samples were being heated, which means the conditions were the same as in fire. This type of tests will be hereinafter referred to as “hot tests”. Another set of tests was made on samples treated as in hot tests and then cooled down. Tests were made after the samples had been cooled down to approx 20°C. This type of tests will be hereinafter referred to as “cold tests”. The purpose for comparing tests on two sample groups was to explain whether the room-temperature tests of bonds weakened at high temperatures were reliable enough to evaluate the behaviour of concrete structures in fire. This is an important factor for the concrete structure strength as well as for firemen safety. The reason therefore is the chipping of concrete ceiling in a shorter
time than it should occur based on the concrete structure fire grade. On the other hand, cold tests are the basis for evaluating the bond strength reduction and the status of concrete structures after fire [4], [5], [7].
2. Description of sample materials Concrete of C16/20 grade and C40/50 grade was used for manufacturing the samples [8]. The characteristic compressive strength on day 28 was determined according to PN-EN 12350-1 [9]. The reinforcement was made of A-I class smooth steel bars and A-II class ribbed steel bars, both of 10 mm diameter. The purpose of using both smooth bars and ribbed bars was testing the impact of ribs on the bond reduction under high fire temperatures as well as explaning this phenomenon. Samples were made at the Institute For Construction Structures of the Warsaw Technical University.
3. Cold test Cold tests were made as rollers of 100 mm diameter and 150 mm height. Thermocouples were located in the sample middle where concrete contacts the rebar as well as on the sample surface. A small-diameter channel for the thermocuple was made in the course of pouring the concrete [1].
The thermal treatment was carried out in a furnace fitted with a programmer and temperature controller using the temperature-vs.-time curve as adopted for tests. The temperature distribution at 15 mm depth from the heated surface inside concrete was assumed as in standard fire according to the formula (1). After the required temperature had been reached, samples were kept constant at constant temperature for 30 minutes
(Fig. 1). At the same time, temperatures equalized on the sample surface and on the rebar-concrete contact surface (Fig. 2). After the heating, samples were cooled for 24 hours to reach the room temperature. Pull-out bond tests were made using a strength testing machine. The test was used to determine the maximum force required to move the rebar inside concrete.
1000 900 800 700
]C [° 600 er ut rae 500 p m eT 400 300 200 100 0 0
10
20
50°C 200°C 350°C 500°C Standard temperature-vs. Time curve
30
40
100°C 250°C 400°C 550°C
50
60
Time [min]
150°C 300°C 450°C 600°C
Fig. 1 Temperature distribution as assumed for cold tests with reach times for different set temperatures.
620
1 520
] oC [ ER420 UT AR EP320 M ET
2 3
220
120
20 0
10
20
30
40
50
60
70
1 - Furnace temperature as measured by control thermocouple 2 - Temperature on concrete surface
80
TIME [min]
90
3 - Temperature inside sample on the steel –concrete contact
Fig. 2. Temperature curves obtained in tests (St3S steel, C16/20 concrete)
3. Hold test It was decided, that the temperature distribution on the steel-concrete contact surface in the sample middle will be the basic temperature distribution for comparing the bond reduction in hot tests and in cold tests in the same thermal conditions.
The trend to equalize the temperature distribution on the steel-concrete contact surface has led to smaller sample diameters for hot tests. The diameter was reduced by 30 mm, the height remained unmodified. This modification allows to obtain temperature distribution curves being very similar to one another as seen in Fig.3.
740 660 580
o
]C500 [ ER420 UT AR EP340 M ET 260 180 100 20 0
10
20
30
40
50
60
Temperature inside sample on the steel-concrete contact surface – cold test Temperature on concrete surface – cold test Temperature inside sample on the steel-concrete contact surface – hot test Temperature on concrete surface – hot test
70
TIME [min]
80
Fig. 3. Actual temperature distributions compared for both test types
FFig. 4. Hot test diagram. Determination of critical temperature for bond reduction whereas:
1 – temperature on the steel-concrete contact surface 2 – force applied to the sample (pulling the rebar from the sample)
The hot tests were made according to the curve as in Fig. 4. Samples were loaded with a constant force over the test time. At the same time, samples were heated according to the temperature-vs.-time curve as adopted with the temperature being measured on the sample surface and on the steel-concrete contact surface. The test objective was to determine the critical temperature, Tkr, where the bond force reduction is equal to the sample load. The test example is seen in Fig. 5. The bond loss event was defined as the rebar movement with relation to the concrete, which results in a sudden force reduction. The tests were made on a test stand consisting of: a furnace, strength testing machine, temperature metering devices, rebar movement measurement devices, data recorder.
5. Test results
Statistical calculations were made for hot test results and for cold tests results in order to determine the critical temperature for bond loss at constant parameters as adopted (concrete grade, steel grade). Fig. 6 shows the comparison of diagrams that describe the bond reduction in hot tests at low temperatures for St3S steel and C16/20 or C40/50 concrete grades. This diagram was the basis for the conclusion regarding the concrete grade impact on bond and on the bond reduction at higher temperatures for smooth rebars. Fig. 7 shows diagrams describing the bond reduction at fire temperatures for 18G2 rebars and C16/20 or C40/50 concrete grades as obtained from cold test results. Fig. 8 shows bond reduction diagrams for St3S smooth rebars and C16/20, C40/50 concrete grades at high temperatures as obtained in hot tests. Fig. 9 shows bond reduction diagrams for 18G2 ribbed rebars and C16/20, C40/50 concrete grades [1], [2], [3].
8
600
3 7
500
temp. wew.-1
loaded-3
temp. zew.- 2
400
]C ge d[ ER300 UT AR EP 200 M TE
6
2
5 4
1
3 2
100
0
1
0
10
20
30
CZAS [min]
40
50
60
70
0
Diagram for sample loaded with P = 7.00 [kN] Fig. 5
. Hot test example
whereas: 1 – temperature on the steel-concrete contact surface 2 – temperature on the sample surface
]C °[ ec arf us tc at no c et er cn oc -r ab re no reu ta re p m eT
Comparison of curve inclination Evaluation of bond reduction rate for St3S steel and different concrete grades
600 500
y = 6,148x + 22,152
400
y = 5,762x + 21,916 300 200 100 0 0
20
40
St 3S steel C16/ 20 concret e
60
80 Bond force reducti on [%]
St 3S steel C40/ 50 concret e Linear (St3S st eel C16/20 concret e) Linear (St3S st eel C40/50 concret e)
Fig. 6
. Critical temperatures for bond reduction for smooth rebars (cold tests)
100
]N k[ P D ED A O L
800
] C °[ ec arf us tc at no c tee rc no c-r ab re no reu ta re p m eT
Comparison of curve inclination Evaluation of bond reduction rate for 18G2 steel and different concrete grades
700
y = 203,51Ln(x) - 81,234
600 500
y = 8,325x + 19,252
400 300 200 100 0 0
20
40
60
18G2 steel C16/ 20 concret e
80
100
Bond force reducti on [%]
18G2 steel C40/ 50 concret e Linear (St 3S steel C40/ 50 concret e) Log. (18G2 steel C40/ 50 concret e)
Fig. 7
Comparison of curve inclination Evaluation of bond reduction rate for St3S steel and different concrete grades
600
ec arf 500 us tc at no 400 c et er cn ] 300 oc C -r [° ab er 200 no reu ta 100 re p m eT 0
. Critical temperatures for bond reduction for smooth rebars (cold tests)
y = 5,981x + 5,385
y = 5,294x - 19,009
0
20
40
St3S st eel C16/20 concret e St3S st eel C40/50 concret e
60
80 Bond force reducti on [%]
Linear (St 3S st eel C16/ 20 concrete) Linear (St 3S st eel C40/ 50 concrete)
Fig. 8.
Critical temperatures for bond reduction for smooth rebars (hot tests)
100
700
ec 600 far us tc 500 tan oc 400 et er cn ] oc C -r [° 300 ab er no 200 er ut 100 ar ep m eT 0
Comparison of curve inclination Evaluation of bond reduction rate for 18G2 steel and different concrete grades
y = 9,075x + 23,944 y = 6,454x + 36,206
0
20
40
60
18G2 steel C16/ 20 concret e
80
100
Bond force reducti on [%]
18G2 steel C40/ 50 concret e Linear (18G2 st eel C16/ 20 concret e) Linear (18G2 st eel C40/ 50 concret e)
Fig. 9.
6.
Critical temperatures for bond reduction for ribbed rebars (hot tests)
Conclusions
1. It should be stated based on test results that high fire temperatures are the reason for a significant reduction of steel-concrete bond. A conclusion can be drawn from the result analysis that: the bond reduction in cold tests (the bond tested after fire) and in hot tests (the bond tested in fire) are different over the entire temperature range up to 800°C. The bond reduction in the hot status is in all cases bigger than in the cold status. This fact is the evidence of a partial return of the bond after the sample was cooled; friction is probably in part responsible for that phenomenon. 2. The rebar type (smooth or ribbed) has a very significant impact on steel-concrete bond features not only at normal temperatures (room temperatures) – this being a well-known fact, but also at fire temperatures. This is related to the different types of contact surfaces and to the load transfer method from rebars to the concrete. 3. The concrete strength impact on the steelconcrete bond at fire temperatures is differentiated and related to the steel surface: the impact is small for smooth rebars, whereas it’s significant for ribbed rebars. It was determined for smooth rebars based on tests that: after the bond was reduced to zero at higher and high temperatures, rebars “slip out” from concrete without any significant damage to the concrete. For ribbed rebars: before the rebar could be moved, concrete is destroyed around the ribs, and lateral or slant cracks (in relation to the rebar centreline) are created. 4. For ribbed rebars: high-strength concrete grades tend to increase the bond reduction temperature threshold. In cold tests, the bond reduction for 19G2
ribbed steel – C40/50 concrete is in practice insignificant over the temperature range up to 500°C. Summary of conclusions: • In cold tests, while analysing the applicability of RC structures for further use, a reduction of rebar – concrete bond strength must be considered. • While determining the critical temperature that impacts the fire resistance of RC components (in particular: RC ceilings), the possibility of concrete chipping due to the bond reduction as well as a significantly increased temperature of uncovered rebars should be considered. • The bond reduction tests made at room temperature on samples cooled after fire treatment are not reliable enough to evaluate the bond reduction as caused by fire. 7.
References
1. Bednarek Z., Ogrodnik P., Study of the influence of fire temperatures on the fall of the bond between steel and concrete. IV International conference. Safety of fire building, Częstochowa 2002. (in polish). 2. Bednarek Z., Ogrodnik P., Study of the influence of thermal conditions during fire on the bond between steel and concrete. Contemporary problems of fire safety in buildings and environmental engineering, Koszalin – Łazy 2004. (in polish). 3. Bednarek Z., Ogrodnik P., Study of the influence of high temperatures of the bond between ribbed steel and concrete C40/50. XIX Conference scientifically – technical. Concrete and Prefabrication, Jadwisin 2004. (in polish). 4. Chih-Hung C., Cho-Liang T.: Time–temperature analysis of bond strength of a rebar after fire exposure, Cement and Concrete research, No 33, p. 1651-1654, 2003.
5. Fellinger J. H. H., Jołop A., Uijl D.: Bond of pretensioned strands in fire exposed concrete. Bond in concrete – from research to standards. Budapest 2002. 6. Morley P.D., Royles R.: Response of the bond in reinforcing at normal and high temperature. Magazine oif Concrete Research No 123, p. 67-74, 1983.
7. Reichel V.: How fire affects steel-to-concrete bond, Building Research and Practice, 1978. 8. PN-EN 206-1:2003. Concrete. Part I: Requiremen, properties production and compatibility. (in polish). 9. PN-EN 12350-1: Study of concrete mix. Part 1. Drawing of samples. (in polish).
