您还没有绑定微信,更多功能请点击绑定

XY方向精度的过程能力定义

In the plant i work for, the scanner is used to check the accuracy of the position of the hundred thousand of holes all drilled and therefore, the huge amount of raw data couldn’t be kept for long and so I hope that some experts here could show me some ideas: how to use statistics parameter mean/standard deviation of X/Y to infer the process capability according to the specification.

!(http://sheldon.sundns.com/temp/spec.jpg)

!(http://sheldon.sundns.com/temp/accuracy.png)

thanks in advance!
对“好”的回答一定要点个"赞",回答者需要你的鼓励!
已邀请:

pengshanju (威望:0) (上海 闵行)

赞同来自:

Measurement offsets in both X and Y direction are assumed to be normal distributed and independent to each other: X<- N(u1,s1) and Y<- N(u2, s2) ; the mean and standard deviation of X and Y were stored in the scanner after each measurement. Specification for registration is R = sqrt(X^2 + Y^2) <= 75um, zero point for offset calculation is based on the two-alignment hole in the corner. The distribution of value R couldn’t be regarded as a normal distribution, which leads to that no common process capability index could be obtained.

2 个回复,游客无法查看回复,更多功能请登录注册

发起人

扫一扫微信订阅<6SQ每周精选>