ISSN 0794 - 0378

THE NIGERIAN JOURNAl.;

OF
PURE AND APPUED SCIENCES

Nig. 1. Pure & Appl. Sci. Vol.

Printed in Nigeria

22 (2007): 2129 - 2135

07940378
2007 Faculty of Science Unlv, of Ilorin, Nigeria.

A CONTINUOUS COLUMN DESIGN FORMULA AT ULTIMATE STRENGTH FOR

AXIALLY LOADED IROKO (MllicJa exce/sa) TIMBER COLUMN

A. A. Jimoh
Department of Civil Engineering University of Ilorin, llorin, Nigeria.
ABSTRACT: A continuous column design equation for axially loaded Iroko (Botanical name: Milicia excelsai wood
column or strut is presented. The procedure simplifies the usual tedious design method of categorization of columns into short.
intermediate and slender according to the values of slenderness ratio. To derive the design equation, an experimental program was
carried out on direct axial compressive strength test on lroko specimens, to obtain compression stress and strain data. The data
were subsequently analyzed to derive the wood typical stress-strain equation in compression. This equation and the Euler Engesser stress formula were used to obtain values of critical stress and corresponding slenderness ratios. From these values.
analysis was carried out to derive a continuous column design equation to fit the theoretical ultimate loads with the experimental
values.
.
,
INTRODUCTION

1.

The timber material of interest in this report is Iroko, of botanical name Milicia excelsa. The tree is up
to 48 m high and 9 m .in girth, deciduous, with straight and cylindrical bole and grey to dark brown or
blackish bark. lroko is a very common large forest trees with dark green foliage and distinctive leaves
with pale to dark brown wood having a fairly open grain. Iroko is called in local language in Nigeria
as follows, Hausa: loko, Yoruba: iroko, Ibo: oji. It is wide spread in tropical Africa, from the Ivory
Coast to Mozambique. Its main habitats are the rainforest and forest outliers in savanna woodland areas.
It is very durable, resistant to decay, termites, and marine-borer attack, and very stable, with little
shrinkage and movement once seasoned. It is the most sought-after construction timber in West Africa
and used in boat building, laboratory benches, domestic flooring and marine work (Keay et aI., 1964;
NCP2, 1973; Jackson, 1978; Desch, 1991).
However, there has been no established design procedure for Iroko column, which can be used for its
rational application as a column or strut. A related design information on wood where a column is
categorized into short, intermediate and slender is given in a reference (Jozsef and Benjamin, 1982)
where each column category has its own peculiar design formula. However, it has been reported that
division of the slenderness ratio into short, intermediate, and slender categories with subsequent
analysis is quite involved and may not be necessary. Instead of categorization into these ranges, a
continuous column formula is appropriate (Mazur, 1965). At ultimate, when stress is not proportional to
strain, tangential modulus can replace modulus of elasticity in Euler's equation (Borislav, 1975;
William, 1977), giving Euler-Engesser stress formula, This equation and the stress-strain equation in
. compression of the material are used to obtain the critical stress and corresponding slenderness ratio.
These processes are carried out on Iroko material, as already analysed for Ayin (Anogeissus leiocarpus)
column, (Jimoh, 2005), to derive the column design equation.

2. Theory
The intention here is toobtain values of Iroko column compression stress against slenderness ratio.
Euler's critical load for a strut of elastic material is given by

1[2EI
P =-cr.
L2

(I)

From the equation, the critical stress (O'er) is given as:

1[2E

cr

=---

cr

(;

(2~

)2
2129

Nig. J. Pure & Appl. Sci. Vol.

A.A. Jimoh

At ultimate, stress is not proportional to strain .and therefore. Young's modulus E in eqn.(2) is replaced
by tangential modulus E, and becomes

0'

cr

n2E
t

(3)

(~)2

This is the Euler- Engesser critical stress formula.

Eqn.(3) can be written in terms of slenderness ratio as

L=nJE,
r

(4)

O'er

In order to solve eqn.(4), pairs of val~es of E, and O'er are obtained from stress-strain equation so that
after substitution for them in the equation, corresponding values of slenderness ratio (Ur) can be found.

3.
EXPERIMENTAL INVESTIGATION
An experimental program was carried out to obtain stress and strain data and stress- strain equations. A
laboratory test was carried out on direct axial tensile and compression test parallel to grain and flexure
on Iroko wood material. Air-dried Iroko wood obtained from a timber market in IIorin town, Nigeria,
was used for the test. Compression specimen was cut into nominal dimension 20 mm by 20 mm in
section and 80 mm long, flexural specimen dimension was 25 mm by 25 mm in section and 380 111m
long. Dimension for the irregular shape tension specimen is shown in Figure I, all specimens are
shaped as in ASTM, (1985), but with.reduced dimensions to suit the testing equipment. Fifty specimens
were tested in each case using a Testometric Universal Testing Machine. Strain was monitored using
the downward movement of the compression plate.
10

~cb
~

30

20

/D

25

~~

~~

~
50

20

30

Figure I: Tension specimen nominal dimension (in mm)

At the end of the test, stress and strain data were analyzed, using Microsoft Excel 2003, to obtain the
average strength and strain and stress-strain relationship for each specimen. Polynomial stress-strain
equation of order 3 of the form (ax) + bx2 + ex ) was found appropriate because the least squared error
(R2) was close to I. The average of.each of the coefficients (a, b, and c), in the stress-strain equation was
obtained and used to derive the typical stress-strain equation in compression. The summary of the
result of analysis is' shown in Table I.

2130

Nig. J Pure & App/. Sci. Vol. 22

Column design formula for axially loaded iroko limber column

Table I: Experimental

data of Iroko material

Minimum
34.48
34.85
84.53

Average
51.5
71.38
123.80

Maximum
68.53
114.50
151.53

Strain

Minimum

Tension
Compression

0.019

Average
.0378
0.04636

Maximum
0.066
.0753

Strength

property

Compression
Flexure
Tension

0.0285

Standard deviation
9.9

22.16
22.26

Moisture content of specimens vary between 8.5 to 11.5 with an average of 10.5 %. Typical axial
stress-strain equations in compression and in tension are
.

crc

= -13844953'+

724182 -582

(5)

crt = 1601189 3 -156748 2 + 6492.3

(6)

respectively, where (J and E denote the stress and strain, respectively. The curves representing these
equations are shown in Figures 2a and 2b respectively.

fit = y = 1601189x3

156748x2

frt

= y = -1384495x3 + 72418x2 +

+ 6492.3x

E
.
z

~
II)
II)

~
II)

60

100

50

80

E
E

60

>:

40

III
III

-e

20

-20

40
30
20
10

III

o
~

582.07x

120

..

0.02

0.03

0.04

0.01

-10

strain (x)

--_._-_ .._----_ .._---Figure 2a: Typical tensile stress-strain curve

~

0
0.05

0.02

0.04

0.06

strain (xl

Figure 2b: Typical compression

stress-strain

curve

4. DERIVATION OF CONTINUOUS COLUMNEQUA TION

(a) Values for Stress and Slenderness ratio

The stress and slenderness ratio values were obtained from analysis of the typical stress-strain
5. The analysis was carried out as follow:
The tangential modulus E, is the first derivative of equation 5 and it is given by

= 8cr/8E

That is;

E =-3(1384495)2
t

equation

+2(72418)-582

(7)

2131

A.A. Jimoh

Nig. J. Pure & Appl. Sci. "01. 22

In order to solve equation 4, strain (8) values were substituted in equation 5 and 7 to obtain pairs of
stress (0') and tangential modulus (EJ Both results (0' and E,) are then substituted in equation 4 to
obtain the corresponding slenderness ratios, shown in Table 2. The data show that for increasing critical
stress, the slenderness ratios are decreasing. The limit for optimum critical stress is 50 N / mm2 when
the slenderness ratio value is 10,

Table 2: Stress versus slenderness

Strain

Stress from
Eqn.(5)

Tangential
modulus
Et

ratio results for Iroko

Slenderness
ratio
Llr

Stress
from fit
Eqn.(8)

Ratio
of
Fit col
5

factual
stress
col.2
(I)
(2)
(3)
(4)
0.0078
8.289042249
1459,0928
41.697874
0.0079.8.435350351467.055441.4473095.193029
0.00795
8.50880196
1471.0056
41.323547
0.008
8.58245056
1474.935
41.200ne
0.009
10:09519115
1549.1617
38.93281

(5)
4.990.1691

(6)
0.60
0.62
0.62
0.63
0.76

5.2960262
5.4000055
7.6271894

Limit for good result ( slenderness ratio Llr not greater than 37)
0.01
0.015
0.0151
0.0152
0.0153
0.0154
0.0155
0.0156
0.0157
0.0158
0.02
0.021
0.022
0.023
0.024
0.025
0.025
0.026
0.0265
0.0266
0.0267
0.028
0.03
0.032
0.034
0.036
0.038
0.0385
0.0385

11.678005
1615.0815
36.960474
9.9890984
20.3524293'8
1820.0759
29.720852
21.052784
20.53453674
1822.0575
29.604874
21.250303
20.7168381
1823.956
29.489682
21.446859
20.89932517
1825.7715
29.37526
21.642456
21.08198963
1827.5039
29.261595
21.837102
21.26482318
1829.1532
29.148674
22.0308
21.44781751
1830.7195
29.036482
22.223556
21.63096431
1832.2027
28.925006
22.415377
21.81-125528
1833.6028
28.814234
22.606269
29.53264
1817.396
24.654619
29.875917
31.33799981
179f.9391
23.76570131.424989
33.1/374924
1758.1753
22.900847
32.918957
34.351581341716.1044
22.053893
34.364515
36.543189121665.7266
21.21892
35.768031
38.18026563
1607.0419
20.390072
37:13555
38.18026563
1607.0419
20.390072
37.13555
39.75450388
1540.0501
19.561354
38.472802
40.51546274
1503.4392
19.145079
39.131782
40.66542877
1495.8678
19.061595
39.26289
40.81463352
1488.2133
18.977949
39.393781
42.681237761381.145817.87828741.077811
45.2569:35
1189.0135
16.109255
43.621759
47.41513984
963.65336
14.168587
46.133217
49.08939652
705.06534
11.910925
48.61364
50.21324928
413.24944
9.0161601
50.962129
50.72024236
88.20566
4.144596
52.311126
50.74281852
1.7528587
0.5841317
50.78928
50.742818521.75285870.5841317
50.78928

2132

0.86
1.03
1.03
1.04
1.04
1.04
1.04
1.04
1.04
1.04
1.0 I
1.00
.0.99
0.99
0.98
0.97
0.97
0.97
0.97
0.97
0.97
0.96
0.96
0.97
0.99
.1.01
1.03
1.00
1.00

Column design formula for axially loaded iroko timber column

(a)

Nig. J. Pure & Appl. Sci. Vol. 22

Graph ofStrcss Versus Slcndcrncss Ratio

Graph for stress versus slenderness ratio results of Table 2 were plotted and is shown in Figure 3 as a
design chart. The graph' shows an initial optimum region with a critical stress of 50 N/mm2, This stress
has a slenderness ratio of 10, Beyond this value, the graph has negative gradients,

r--'----.-

60

Equation of fit

stress = 0.0013(Ur)3 - 0,1014(Ur)2

+ o 881(Ur) + 50.309
R2 = 0.9942

50

fit ------------------actual

40

I
I

20

10 .

o .~-------.-------.-------.
o
70
40

60

slenderness ratio (Ur)

---_._--_._._---_._----------------'

Figure 3: Stress versus slenderness

ratio for Iroko column

(b) Equation of Fit for Stress versus Slenderness Ratio

In order to have a handy equation to design Iroko column, statistical regression analysis was carried
out, using Microsoft Excel 2003, on critical stress and slenderness ratio results in Table 2, From the
regression analysis, Equation (8) was obtained,
.
Stress = 0.00 13(Ur)J - 0.10 14(Ur)2 + 0.881 CUr) + 50.309

(8)

The curve of tit for this equation is shown in Figure 3. The least square error for the tit is 0,99. Also
stress results from the experimental stress-strain formula equation (5) and the. stress from the tit
equation (8) are shown respectively in columns (2) and (5) in Table 2, The two column results show a
tit of least square error (R2) of unity,

5.

VERIFICATION OF DESIGN EQUATION

Verification of equation (8) was carried out by comparing the results of experimental compression test
carried out on iroko specimens and the result obtained with design equation (8), These results are
shown in column 5 and column 8 of Table 3. From the table, the ratio of theoretical stress obtained
using equation (8) to experimental stress ranged between 71% to 123% with an average of 100 %.

2133

ig. J. Pure & Appl. Sci. Vol. 2

A.A. Jimoh

SINo

Width
b
(mm)
I

I
2
3
4
5
6
7
8
9
10

17.40
17.20
17.00
,17.20
17.20
17.00
17.00
17.40
17.00
17.00

Table 3: Experimental verification of design eqn.(8) for Iroko column

Exper.
_ Slendernes
theoretic.
height L
Radius.
Exper.
ultimat
ult.stressultimate.
of
s ratioUr
c
strcss
Gyratn r
eqn.(8)
Load
(mm)
(mill)
(kN)
(Nzmm ')
(Nzrnrn ')
(mm)

Depth
d

2
17.10
17.20
17.40
17.00
17.00
17.20
17.20
17.00
17.40
17.00

57.3
57.77
57.85
57.9
58.9
57.75
50.45
59.11
57.12
58.77

13463
16084
15023
1)258
15220
12605
15868
11508
11851
19806

45.25
54.37
50.79
52.18
52.05
43.11
54.27
38.90
40.06
68.53

4.94
4.97
5.02
4.91
4.91
4.97
4.97
4.91
5.02
4.91

7
11.61
11.63
11.52
11.80
12.00
11.63
10.16
12.04
11.37
11.98

Theorl
Exp.
(Co1.8
/coI.S)

8
48.91
48.88
48.99
48.72
48.52
48.88
50.16
48.48
49.13
48.55
average

6.
DISCUSSION
The design equation (8) obtained here is for air-dry Iroko material with moisture content ranging from
8.5% to 11.5 %. Usually for design of wood, the maximum slenderness ratio recommended is 30. this
gives for a Iroko column, a design stress (ae,) of20 N/mm2.
The stress-slenderness ratio curve for Iroko is identical to that of wood given in a reference (Jozsef and
Benjamin, 1982), because for wood the limit of slenderness ratio (Llr) for short column is eleven while
II is also obtained here when the optimum stress is 50 N/mm2 However, the continuous column
equation obtained here has removed the difficulty usually encountered in categorization in the design of
a timber column.

7.
CONCLUSION
From the analysis, it can be concluded that:
(I) The stress-slenderness ratio curve for Iroko is identical to that for other woods and the stress
slenderness ratio relationship for air dry lroko with moisture content not more than 11.5% is
Stress = 0.00 l3(Llr)) - 0.10 14(Lld + 0.881 (Llr) + 50.309
(2) Both the equation (8) and stress-slenderness ratio curve in Figure 3 can be used to design an Iroko
column when the slenderness ratio is not more than 37.

REFERENCES
Annual Book of American Society for Testing and Materials Standard, (ASTM), (1985). Wood,
Volume 04. 09, section 4, Philadelphia, USA pp 55-56.
Borislav, D. Z. (1975). Inelastic behaviour of wood Beam- Columns. Journal of the Structural Division,
ST2. pp. 416-434.
Desch, H. E. (1991). Timber, its structure, properties and utilization, 6tl) Edition, revised by
Dinwoodie ,J. M., Macmillan Education, London
Jackson, N. (1978j. Civil Engineering Materials, English language book society and Macmillan Press
Ltd, London.

2134

9
1.08
0.90
0.96
0.93
0.93
1.13
0.92
1.25
1.23
0.71
1.00

Column design form II la lor axially loaded iroko timber column

Nig. J. Pure & Appl.

sa.

Vol,.22

Jimoh, A.A. (2005). Ultimate Strength Design of Axially Loaded Ayin iAnogeissus leiocorpusy Timber
Columns, Journal of Applied Science and Technology (Ghana), Vol. 10, Nos. I & 2, pp, 2934.
Jozsef, B. and Benjamin. A. J, (1982), Mechanics of Wood and Wood Composites, Van Nosttrand
Reinhold Company, New York.
.
Keay, R. W.J., Onoche, C.F.A. and Stanfield, D.P. (1964). Nigerian Trees, Vol. II, Department of
Forest Research Institute Ibadan, Nigeria. pp 188-189.
Mazur, S. J. (1965). On a rational column formula Symposium on Timber and Timber Structures,
Transactions of the Engineering Institute of Canada, P.aper No. EIC -65 -BR&
STRI3, Bridge and Structural, Vol. 8, No, A-16. pp 11-12.
Nigerian standard code of practice on building and construction part 2 (NCP2) (1973), Federal
department of forest research lbadan , Nigeria, pp 13-24
William, A. N. (1977). Theory and Problem of strength of Materials, Schaums Outline Series, New
York: McGraw Hill Book Company, pp 264-278.

2135