micrometer gauge, whilst the wall thickness measurements  for 2 mm cylinder was done using Ultrasonic thickness measurement gauge。 The wall thickness was measured at eleven different equi- distance points along the axial length of the cylinder。 This was then repeated along 36° spaced along the circumference of the cylinder, resulting in 11 ~ 10 ¼ 110 measuring points。 The nominal tnom, average tave, minimum tmin and maximum tmax wall thickness and the corresponding standard deviation for all specimens   are

shown in Table 2。 Next, both the inner and outer diameters of the cylinders were measured at five equally spaced diameters using digital vernier caliper at the top and bottom ends respectively。 The average mid-surface diameter at the top and bottom and the corresponding standard deviation for all cylinders are shown in Table 3。 Lastly, the axial lengths of all cylinders were measured using digital Vernier caliper at eleven equidistance point。 The corresponding average lengths and standard deviation are also shown in Table  3。

Finally, all cylindrical specimens (CY1_t0。5, CY1_t1。0, CY2_t1。0, CY1_t2。0 and CY2_t2。0) were subjected to axial compression using Universal Instron Machine。 Specimen was placed between the platens of INSTRON machine without any clamping plates, as shown in Fig。 3a for 2 mm cylindrical specimen。 Then, the movable crosshead is gently adjusted until a contact is established between the specimen and the instron platen, as depicted in Fig。 3b。 It is assumed that the platen of the INSTRON machine will helps to provide the desirable boundary condition。 Though it is practically impossible to achieve fully clamped condition without rotation, but the effect of rotation on the magnitude of buckling load of cylindrical shell has been proven to be marginal [17]。 The same experimental set up was repeated for 0。5 mm and 1 mm cylinder。 Incremental axial load was applied to one side of the cylindrical specimen at the rate of 1。0 mm/min。 This is the same rate of loading used to obtain the material properties of the mild steel plate from which the cylinders were made。 The compression ex- tension and the corresponding load at each increment were measured by the machine。 Fig。 4 shows a typical curve of load versus compression extension for cylinder with different radius- to-thickness ratio, R/t。 It can be seen from Fig。 4, that the load deflection curve is nearly linear up to the failure load。 Again, it is apparent from Fig。 4, that increasing the thickness of the cylinder

results in an increase in the stiffness of the   material。

Table 3

Measured average data and standard deviation of mid-surface top and bottom diameter and axial length for all tested cylinders。 Note: D ≡average mid-surface diameter, ΔD≡

standard deviation for mid-surface diameter。

Dtop Δ¯Dtop

Dbottom Δ¯Dbottom Lave Lstd Dtop /tave Lave/ Dtop

(mm)

CY1_t0。5 102。02 0。963 101。55 0。732 113。06 0。0954 204。04 1。108

CY1_t1。0 101。54 0。146 101。04 0。067 112。4