International Conference o n S h o t Peening and Blast Cleaning

CONCRETE FILLED STEEL TUBULAR COLUMR
Dr. P. Suryanarayana
Maulana Azad College of Technology, Bhopal, India

INTRODUCTION
Concrete filled steel tubular columns consist of a well compacted high grade concrete (minimum M25 grade) within mild steel tubes. Column sections utilizing square and circular shapes are s h o w n in Fig. I .

Fig. I . Concrete filled steel tubular columns

Filled steel tubes offer structural advantages due t o efficient shape of t h e steel section (which offers a large moment of inertia) and full composite action of steel and concrete. The tubular section contributes a large part of the flexural rigidity of the column. The concrete being confined within the tube has an enhanced crushing strength due t o triaxial effect of confinement. The column needs neither reinforcement nor formwork. The tube protects the concrete f r o m impact and abrasion. Concrete filled steel tubular columns have an advantage over reinforced concrete columns and concrete encased steel sections E that the load n carrying capacity is n o t reduced when carrying long term load. The elastic modulus of concrete is reduced under long term loading and hence the load carrying capacity is also reduced in case of columns where the concrete contribution t o flexural rigidity is

Concrete filled tubes are also liable t o fatigue failure at the pier. This assumption. due t o development of design methods and codes. Concrete filled steel tubular columns are efficient structural members that can be used as bridge piers. This assumption is valid for bridge piers where the load is transferred through capping plates and the eccentricity of the load can be accurately estimated.K. high performance concrete (upto 1 0 0 MPa).large.S. Interest in this type of construction is being revived in U. was constructed using 42" (10 7 0 m m ) diameter tubes w i t h core concrete of M 3 0 grade and steel thickness of 5 0 m m . k n o w n as proportional loading case is made in case of pin ended columns. The combination of load and moment causing failure can be calculated assuming that the eccentricity 'e' (momentlload) is constant. admixtures and self compacting concrete. The latest Eurocode EC4 published in 1 9 9 7 uses a modified approach consisting of a bilinear interaction diagram in place of the diagram used in ECCS method. It is shown in this paper that in case of filled tubular columns the load carrying capacity is not reduced under long term loading.A. Hollow tubes supporting o f f shore platforms are vulnerable t o fatigue caused b y wave forces. A t the junction the tube is t o be strengthened either b y studs or shotpeening. DESIGN METHODS European codes have adopted design method originally proposed b y Basu & Sommerville later modified by Virdi & Dowling. In 1 9 6 6 t h e four level motor w a y bridge at Almondsbury in U.K. and U. The construction is greatly facilitated due t o development of high strength. This method is incorporated in the Model Eurocode of 1 9 8 1 published b y European Committee of Constructional Steel Research (ECCS). .platform junction or where the capping plates are welded t o the tubes.

are as listed in Table 1 . In the ECCS method these equations provide for modified design stresses in concrete and steel t o account for t h e effects of confinement. SQUARE AND CIRCULAR COLUMNS The ultimate strength of concrete filled steel tubular columns depends upon the strength of concrete and steel. Basu & Sommerville showed that for square or rectangular columns the enhancement of the concrete strength due t o confinement is negligible. eccentricity of load 'el. i t is generally assumed that the moment is constant and n o t proportional t o the loading. The design equations f ~ circular columns. In case of EC4 these modification factors are given only for columns of m e d i u m slenderness and small values of load eccentricity. o n the other hand incorporate parameters r t o a c c o u n t for enhanced strength. Calculations for permanent loading are done assuming that the long term elastic modulus of concrete is half the value of short term modulus. Concrete core of M25 grade and steel tube of Fe250 grade ( w h i c h are the minimum permitted b y ECCS) are assumed. creep and plasticity. . These modified stresses are listed for columns of medium and high slenderness and medium and large load eccentricities. the tube thickness t length of the column 'L'. The plasticity of the steel is also t o be considered in the strength assessment. Therefore. The core size 'd'. The results presented in this paper are based on t h e assumption of proportional loading. COLUMNS ANALYSEB This paper presents the ultimate loads of 36 square columns and 36 circular columns. Results are presented for t w o cases (1) assuming all t h e load is short term and (2) all load is permanent. The strength of concrete is enhanced due t o confinement but reduced due t o creep. creep of concrete and plasticity of steel. the ultimate loads predicted b y EC4 are lower than the values given b y ECCS.In case of columns in building frames where t h e moment is shared both by column and beams.

d. (All load permanent) Calculations by EC4.5 6.5 4. Ultimate Loads of Filled Tubular Columns Ps (kN) Pi (kN) Ps (kN) SQUARE SQUARE CIRCLE PI (kN) CIRCLE Pm(kN) CIRCLE t.0 0 2 5 0 3 7. cm No.5 0 3 7.5 0 3 7.0 6.0 3.5 6.Table 1: S .0 3.(mm) L(m) e .0 6.0 6.5 1586 1256 957 2781 2116 1557 2676 1985 1430 2534 1848 1313 3324 2560 1907 3205 2424 1775 3043 2276 1653 1550 1233 944 2767 2095 1537 2648 1953 1400 2486 1823 1301 3310 2546 1890 3182 2401 1755 3002 2252 1640 1153 1083 994 2822 2672 2460 2314 2139 1923 1954 1799 1610 3476 3270 2991 2828 2618 2358 2374 2192 1969 1115 1050 966 2799 2643 2431 2273 2098 1880 1904 1751 1562 3454 3240 2962 2794 2583 2321 2333 2154 1931 1070 1050 966 2108 2031 1929 1966 1919 1808 1832 1751 1562 2943 2743 2508 2333 2154 1931 2096 1935 1734 PS (All load short term) Pm (All load permanent) P.5 4.0 3.0 6.5 0 3 7.5 0 3 7.5 4. Calculations by ECCS .0 6.0 3.0 3.5 4.0 3.0 0 2 1545 1176 1533 1165 1374 1272 1358 1255 1157 1057 8 8 8 6 6 6 6 6 6 6 6 6 8 8 8 8 8 8 8 8 8 6.0 4.5 0 3 7.0 3. cm 20 20 20 30 30 30 30 30 30 30 30 27 3 0 30 28 30 29 30 30 31 30 32 30 30 33 30 34 30 35 30 36 Note : 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 20 20 6 6 3.0 4.0 6.

Ultimate loads of square columns by ECCS method assuming (1) all the load is short term (2) all the load permanent. A circular section of 2 0 c m core diameter and 6 m m tube has concrete area 3 1 4 c m 2 and steel area 38. (The effect of concrete confinement. This conclusion can be expected t o be valid for square and circular columns of equal core areas as and Ic the former has larger values of steel area and larger IS (moment of inertia of steel and concrete). Steel areas of such magnitude are not possible w i t h RCC columns. It is seen that square columns have larger load capacity especially as the value of L/d increases. In case of circular columns there is enhancement of compressive s t r e n g t h o f core concrete w h i l e i n square c o l u m n s t h i s phenomenon is not present. creep and plasticity being predominant only in circular columns). The load predicted by EC4 method for the 3 m column is 1 5 percent less at 1 1 5 7 kN. Calculations are repeated by EC4 method for circular columns. In case of eccentrically loaded columns it is seen that the load carrying capacity greatly diminishes w i t h increase in eccentricity. DISCUSSION OF RESULTS Concrete filled tubular sections of large load carrying capacity can be designed w i t h compact cross sections. the girders directly rest on capping .The ultimate loads 'PI are calculated for the following cases. Square columns have larger values of concrete and steel areas while circular columns exhibit larger core strength. Hence it is necessary t o properly estimate and control the eccentricity of loading. Results are presented in Table 1. Hence it is interesting t o compare the ultimate axial loads of square and circular columns (of equal value of d and t). Ultimate loads of circular columns by the same assumptions. The reduction in the load carrying capacity due t o long term loading is negligible (as can be seen directly from the table) for both square and circular columns. In most cases of concrete filled tubular columns (such as bridge piers). The ultimate axial load of this column 1358 kN for 3 m height and 8 5 4 kN for 6 m height by ECCS method.8 cm2.

The ultimate loads decreases w i t h increase in the column length. / fmiJ As already mentioned. EC4 recognises the need for keeping the eccentricity l o w and specifies an upper limit of eccentricity = O. Since the columns are generally pin-ended w i t h girders resting o n cap plates.8 log f r N fatilgue life cycles fr stress range (fma. and Pm) can be seen directly from the table. The difference in t h e estimated ultimate loads b y both the methods (P. the ECCS method predicts a higher load .17 Hz Horizontal wave forces = 1 0 times wind force Life expectancy in sea = 40% of land structure There are 150 platforms in India more than 25 year old.plates and eccentricity can be realistically estimated and kept low. For larger eccentricities the ultimate load is calculated without using t h e strength enhancement parameters used in ECCS. the extent of reduction being directly seen f r o m the table. Stud fatigue formula log N = 22. square columns have a larger axial capacity than circular columns of same core width.ld. In case of eccentrically loaded columns the situation is different. It is seen that as many as 18 out of 24 cases.32 . C I d 4 r z o n e of concrete detachment pier I Fig. 2 Zone requiring strengthening ++-- FATIGUE LOADING 100 million cycles in 2 0 years Wave frequency = 0. there is no difficulty in estimating the effective length and t h e corresponding load.

ISBSE. 23. .. 27. 9. (Columns 2. .carrying capacity for circular columns. The Construction Press. Vo1. . P. Steel Consrtuction Institute. 6. REFERENCES 1 . it is not likely that in normal range of columns eccentricities of the order 0. Supplementary volume. pp 1 65-1 8 4 .. 30. . 24. Hence square columns may be more widely used compared t o circular columns. 2. The sections of tubular columns are more compact compared t o RCC columns for the same length and load carrying capacity.. 21. 15.. Square cross-sections are stronger compared t o circular sections under axial loading. 3. ECCS ( 19 8 1 ).25d used in this paper occur. CONCLUSIONS 1. Virdi.. EC4 ( 19 9 7 ) . . It is thus seen that in case of columns w i t h large eccentricity the enhanced strength of core concrete of circular columns outweighs the advantage of higher areas and moments of inertia of square sections. The comparison is made between square and circular columns of same height.. Proc.33. core width. the eccentricities of load for bridge piers are not likely t o be of a large magnitude t o warrant the use of circular cross-sections. tube thickness and eccentricity. p p 230-280. 29. 'Composite Structures'. .K. 3. ( 19 7 6 ) . 18. Concrete filled steel tubular columns offer both constructional and structural advantages. Design methods and codes are available for routine adoption of tubular columns as bridge piers.J. ' A Unified Design Method for Composite Columns'. Circular cross sections are stronger than square cross-sections when the eccentricity of the load is large. 27. .K. K. 33. (Column 3. 12. However. Section 16. 4. 30. . Institution o f Civil Engineers. 36). 24. Dowling. When the more conservative EC4 method is used the circular columns show a larger load carrying capacity than the corresponding square columns in 12 cases. London. 12. 4 . W.. 9. ( 19 6 9 ) . Composite Column Design t o Eurocode 4.. 36)..S. Sommerville.36-ll.. London. . Basu. Zurich. .. 5. 32. A. However. 18. 20. 2. . 'Derivation of Formulae for t h e Design of Rectangular Composite Columns'. 6.. 29. 3. U. 21.