Available online at www.sciencedirect.

com

Procedia Engineering 14 (2011) 297–303

The Twelfth East Asia-Pacific Conference on Structural Engineering and Construction

Experimental Study and Evaluation Method for the

Buckling Strength of Scaffold
Hiroki Takahashi1a, Katsutoshi Ohdo1b, and Seiji Takanashi1c
1
Construction Safety Research Group, National Institute of Occupational Safety and Health, Japan

Abstract

The materials, structure and strength parameters of all the components of the scaffolds are regulated by safety
standards to ensure the safety of construction site workers. Whenever a new scaffold member is developed, its design
must follow the standard. Therefore, all scaffolds have the same structure. This study is fundamental to the
establishment of a new method to evaluate a scaffold’s performance in the buckling strength. Thus the buckling of
scaffolds is often caused by the excessive weight of vertical loads. In the buckling modes, there is member buckling,
as each story of the scaffolds curves, and total buckling, as the whole side of the scaffold caves in. Recent studies
confirm that total buckling is often caused when a stiffening member in the vertical frame is shorting. It is known that
shear rigidity of the horizontal frame can also influence buckling modes. However, the influence of shear rigidity of
both vertical and horizontal frames on the scaffold’s buckling modes has not been made clear. We therefore
performed a numerical analysis to investigate this, using the shear rigidity of the vertical and the horizontal frames as
parameters. In this study, a buckling experiment was executed in order to obtain fundamental data to help solve the
problem of buckling scaffolds.

© 2011 Published by Elsevier Ltd. Open access under CC BY-NC-ND license.

Keywords: Scaffolds, Vertical frame, Horizontal frame, Shear rigidity, Buckling.

1. Introduction

The prefabricated scaffolds are commonly used on construction sites in Japan. To ensure worker safety,
the scaffold structure is regulated under the Japanese Industrial Safety and Health Law (Roudou, 2009).
Based on this law, the materials, structure, and strength of the scaffolding members are regulated. The

a
Corresponding and presenter: Email: takahah@s.jniosh.go.jp

1877–7058 © 2011 Published by Elsevier Ltd. Open access under CC BY-NC-ND license.
doi:10.1016/j.proeng.2011.07.036
298 Hiroki Takahashi et al. / Procedia Engineering 14 (2011) 297–303

vertical load on these scaffolds consists of live weight; i.e., people, as well as materials. The prefabricated
scaffolds might also be used as timbering. Therefore, the prefabricated scaffolds can be subject to
excessive vertical loads that may cause buckling. The buckling modes of scaffolds are illustrated in Fig 1.
They include member buckling, as each story of the scaffold curves inwards and total buckling, where the
whole side of the scaffold caves in. The load required to induce total buckling is smaller than for member
buckling because the buckling length required to induce total buckling is greater. Recent studies confirm
that total buckling usually happens when a stiffening member of the vertical frame fails (Mori et al., 1962).
Therefore, it has been thought that total buckling is associated with deficiency in shear rigidity of the
vertical frame.
On the other hand, shear rigidity of the horizontal frame is an influencing factor on scaffolds buckling.
Therefore, it has been thought that the shear rigidity of the horizontal frame also influences the strength of
scaffolds.
The author of this study and his group have developed a numerical analysis, using the shear rigidity of
the vertical and the horizontal frames as parameters. From the analytical results, the relationship between
the two buckling modes, member buckling and total buckling, could be investigated (Takahashi et al.,
2009).
In this study, a buckling experiment was executed in order to obtain fundamental data to help solve the
problem of buckling scaffolds.

Load
Load

(a) Member buckling (b) Total buckling

Fig 1: Buckling modes of scaffolds

 Hiroki Takahashi et al. / Procedia Engineering 14 (2011) 297–303 299

2. Buckling Experiment of Vertical Frame

2.1. Experimental model and method

The buckling load of one vertical frame more or less equals that of the whole scaffold when member
buckling occurs. It is been thought, therefore, that the buckling load of scaffolds can be estimated from the
buckling load of one vertical frame. In the light of this, we used the buckling load of one vertical frame
for our experiment.

Load
Horizontal member
Pressure device (Diameter: 42.7™Thickness: 2.5)

Universal jack

Vertical frame

Leg member 1,500

1,700
(Diameter: 42.7™
Thickness: 2.5)

Stiffening member
(Diameter: 27.2™
Thickness: 2.0)

(Unit of size: mm) Universal jack

900

Fig 2: Vertical frame used in experiment showing pressure device

The vertical frame used in the experiment was situated in a general construction site. The vertical
frame used in the experiment and the pressure device are illustrated in Fig 2. The upper and lower edges
of the leg member were set the universal jack of the pin joint. The load was set on the upper leg member
by the pressure device. The load was measured by the load sell in the pressure device.

2.2. Experimental results

The buckling load of the vertical frame in the experiment Pe is 115.3 kN. The equation for the
calculation of the buckling load of the vertical frame is that proposed by Mori and his group (Mori et al.,
1962) and we used that equation to evaluate our experimental result:
S 2 EIe
Pse (1)
h02
Where, ʌ is the ratio of the circumference of a circle to its diameter;
300 Hiroki Takahashi et al. / Procedia Engineering 14 (2011) 297–303

E is Young’s modulus;
Ie is the equivalent geometrical moment of inertia of the scaffolds;
h0 is the height of the vertical frame.
Ie is calculated to take into account the influence of the scaffold’s stiffening member and is shown in
the following equation:
§ h ·
Ie 2¨¨ I 0  I s s ¸¸ (2)
© h0 ¹
Where, I0 is the geometrical moment of inertia of the leg member in the vertical frame;
Is is the stiffening member’s geometrical moment of inertia;
hs is the height of the stiffening member.
The buckling load of the vertical frame is represented by Eq. (1) Pse is 105.2 kN. The ratio of the
buckling load represented by Eq. (1) to that of the experimental results Pse / Pe is approximately 0.91. The
buckling load represented by Eq. (1) is only 9% less than one of the experimental results which makes it a
good fit. Therefore, it was confirmed that the buckling load of the vertical frame can be calculated by Eq.
(1).
Pressure device Load Pressure device

Universal jack Horizontal frame

Universal jack
Scaffolds

1,200
1,700

brace
1,200
1,700

Vertical frame
Scaffolds
1,200
8,500

1,700

1,200
1,700
1,700

1,200

Universal jack 900 1,800 1,800 Universal jack

3,600

Fig 3: Cross-sections of scaffolds used in the experiment

 Hiroki Takahashi et al. / Procedia Engineering 14 (2011) 297–303 301

3. Buckling Experiment of Scaffolds

3.1 Experimental model and method

The buckling experiment was conducted in order to obtain fundamental data on the buckling load of
scaffolds in accordance with safety regulations. The scaffolds used in the experiment and the pressure
device are illustrated in Fig 3. The scaffolds selected were a 5-story and a 2-span. The horizontal frame
and the braces of the scaffolds were set identically for all stories and all spans. The upper and lower edges
of the scaffolds were set the universal jack of the pin joint. Stays were fastened to the scaffolds in
accordance with current regulations. The intervals between stays had to be not more than 8,000 mm in the
horizontal direction and 9,000 mm in the vertical direction. Stays were placed at intervals appropriate for
a 5-story (1,700 mm [length of vertical frame] × 5 = 8,500 mm) and 4-span (1,800 mm [length of
horizontal frame] × 4 = 7,200 mm) scaffold. When a vertical load was set on top of the vertical frame with
stays in position, it was assumed that the stays shared the load. However, we then set the vertical load on
top of the frame without stays in place to simulate a situation of greater risk. Stays were set in the vertical
frame at both ends of the fifth story of the scaffolds, as shown in Fig 3. The upper and lower edges of the
scaffolds were set the universal jack of the pin joint. The load was set onto the highest story of the central
vertical frame with the aid of the pressure device.

3.2 Experimental results

DĞŵďĞƌďƵĐŬůŝŶŐ
ĂƚƚŚĞĨŝĨƚŚƐƚŽƌLJ

ƚŚĞďƵĐŬůŝŶŐůĞŶŐƚŚŽĨƚŚĞƐĐĂĨĨŽůĚƐ
ĞƋƵĂůƐƚŚĞůĞŶŐƚŚŽĨĂŶŝŶƐƚĂůůĞĚĐƌŽƐƐ

Fig 4: Buckling of scaffolds in the experiment

302 Hiroki Takahashi et al. / Procedia Engineering 14 (2011) 297–303

The scaffolds in our experiment both displayed member buckling at the fifth story, as shown in Fig 4.
Buckling occurred along the installation length of the brace. The buckling load of experiment Pe is 221
kN. When a braced scaffold exhibits member buckling, the buckling length is less than one story because
the brace itself (hb) is shorter than the height of one story, as shown in Fig 4 and Fig 5. Therefore, the
buckling load of the braced scaffold is greater. When the buckling length of the scaffolds equals the length
of an installed cross brace hb, the buckling load of the scaffolds is shown as follows:
S 2 EI e
Pbe (3)
hb2
The buckling load is represented by Eq. (3) Pbe is 211 kN. The ratio of the buckling load by Eq. (3) to
that of the experimental results Pbe / Pe is approximately 0.95. Eq. (3) corresponds well to the
experimental result and so it is confirmed that the buckling load of the braced scaffolds can be calculated
by Eq. (3).

Load P Load P Load P Load P

hb(<h0)

hb(=h0)
h0

h0

(a) Case of hb<h0 (b) Case of hb=h0

Fig 5: Member buckling of scaffolds

4. Conclusion
In this study, we conducted a buckling experiment with the vertical frame of scaffolds to test and confirm
strength.
The buckling load of the vertical frame in the experiment Pe is 115.3 kN. The buckling load was
calculated taking into account the influence of the scaffold’s stiffening member Ie and it corresponded well
to the experimental result. Therefore, it was confirmed that the buckling load of the vertical frame can be
calculated by Eq. (1).
The scaffolds in our experiment were became the member buckling. Buckling occurred along the
installation length of the brace. The buckling load of experiment Pe
 Hiroki Takahashi et al. / Procedia Engineering 14 (2011) 297–303 303

221 kN. The buckling load was calculated taking into account the influence of the scaffold’s stiffening
member Ie and it corresponded well to the experimental result. We confirmed that the buckling load of
braced scaffolds can be calculated by means of Eq. (3).
In future these experimental results can be compared with other evaluation methods that investigate the
buckling strength of scaffolds and we can further examine the suitability of the evaluation method presented
here.

References

[1] Roudou Chousakai (2007). Handbook of Industrial Safety and Health Law. Tokyo: Roudou Chousakai, in Japanese.
[2] Mori, Y., Mae, I. and Kunimori, M. (1962). On the Load-carrying Capabilities of the Steel Tubular Vertical Frames which are
Used for Supporting the Concrete Bridge Mold, Research Report of the Research Institute of Industrial Safety, 3, pp. 1-8, in
Japanese.
[3] Takahashi, H., Ohdo, K. and Takanashi, S. (2009). Influence of shear rigidity in vertical and horizontal frames on strength of
prefabricated scaffolds, Safety, Reliability and Risk of Structures, Infrastructures and Engineering Systems, pp.3023-3030.