Table 4 The result of Taguchi confirmation experiments

Optimal parameters A1B3C2D1E1F3G3H1

calculated by the following equation。 The optimal conditions are used for five confirmation experiments to calculate   CI,

  where F0。05;1,13 =4。67; Ve =4。85; and rn =5, then

Prediction S/N −3。233±3476 (−6。709~−0。243)dB

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

temperature, and injection speed。 The factor levels are shown in Table 1。 The experimental results are shown in Table 2, in

¼ 4:67 × 4:85 × 

3 þ 5¼ 3:476ðdBÞ ð5Þ

which the lens form accuracy is the average value of five parts。 The response graph in Fig。 4 is obtained according to the S/N ratio of the STB characteristic。 It is observed that the optimal parameter combination is A1B3C2D1E1F3G3H1, i。e。, the melt temperature of 230 °C, injection pressure of 80 MPa, filling- to-packing switchover position of 6。12 mm, cooling time of 10 s, packing time of 2 s, packing pressure of 85 MPa, mold

where Fα;ν1,ν2 denotes the F value of α at a level of signifi- cance; the level of confidence is 1−α; v1 is 1; v2 denotes the degree of freedom (DF) of the pooled residuals; Ve is the variance of pooled residuals; neff is the number of effective observed values; and rn is the number of validation  experiments。

  Total run of  experiments ðN Þ

temperature of 80 °C, and injection speed of 80 mm/s。

The analysis of variance (ANOVA) is usually used to identify  the  influence of  control  factors。  Table  3  is the

neff¼ 1 þ Degree of freedom for significant factors ðDFÞ18

ANOVA table of the lens form accuracy S/N ratio, in which

the significant influence of factors are mold temperature and packing time under a confidence level of 95 %。

¼ 1 þ ð2 þ 2Þ ¼

4。2 Confirmation  experiment

The optimal parameter combination can be obtained by the Taguchi experiment with an L18(21 ×37) orthogonal array。 However, the results must be experimentally verified because the number of Taguchi experiments is too small。 In the L18 Taguchi experiments, the average value of the S/N ratio    is

−10。233 dB。 Based on significant factors in ANOVA, i。e。, mold temperature, the predicted value of the S/N ratio of the optimal conditions is estimated to be

Therefore, the predicted S/N ratio of the optimal parameter

combination is −3。233±3。476 dB (−6。709~−0。243 dB)。 The results of five confirmation experiments are shown in Table 4。 It is observed that the S/N ratios of all experimental runs are within the confidence interval, proving that Taguchi experi- ment’s results are reliable。

4。3 Influence of sensor position

This section discusses the mounting height of the top surface of the pressure sensor in the runner, observing the  pressure

change and hoping to obtain the mounting height approximat- ing the cavity pressure profile。 The sensor top was mounted