Part Average Analysis PAA Rev. 2.0 / 2005 - PowerPoint PPT Presentation

1 / 4
About This Presentation
Title:

Part Average Analysis PAA Rev. 2.0 / 2005

Description:

3 = 10 ki 100; 9 = ki 100. Relevance of observed parameters (type) ... In case, that there is no valid reglementation it is allowed to ship the product ... – PowerPoint PPT presentation

Number of Views:50
Avg rating:3.0/5.0
Slides: 5
Provided by: HKRU
Category:

less

Transcript and Presenter's Notes

Title: Part Average Analysis PAA Rev. 2.0 / 2005


1
Part Average AnalysisPAA Rev. 2.0 / 2005
P
A
A
  • 2. Mathematics part BCalculation rules for
    fault risk assessment and cpk-monitoring

2
Risk figure Rk
  • Calculate the risk figure Rk
  • Significance of anomalies
  • 1 6 lt ki lt 103 10 lt ki 1009
    ki gt 100
  • Relevance of observed parameters (type) (U, I,
    t, ....), depends on the circuit design
  • 9 Clear indication of technical defect
    (Leacage current, Voltage Drop)3 High
    probability to indicate technical defect
    (default value) 2 Low probability to
    indicate technical defect 1 Less
    relevance
  • The relevance should be selected by the
    development engineers

3
Risk figure Rk
  • Rk calculation, select component if RK ?
    9
  • Statistical Process Analysis (SPA)cpk
    monitoring for systematic failure detection
  • The focus of the PAA is the detection of
    stochastical failures in standard processes.
    Systematical failures in manufacturing processes
    can be observed by a wide range of parameter
    distribution (?) or continuous parameter drifts
    which lead to a reduction of the process ability
    characteristics cp and cpk. (VDA 4)
  • For instance, ceramic capacitors can break by
    mistakes in the SMD assembly or in thermal
    processes with which the signal / noise ratio
    decreases and thus the distribution of the
    characteristic voltage values shows burglary
    cp-and cpk values. In this case the decreased cpk
    value with lt1.67 is an unequivocal indicator for
    a preinjured product.
  • With local cpk values lt 2 and a symmetrical
    parameter distribution the quality of abnormal
    parts detection by PAA will be reduced. Notice,
    it has turned out suitable to calculate the local
    cpk value in supplement to the PAA, and to submit
    to a monitoring as in the following is described.
  • The following definitions are based on standard
    processes with N ? 400 and Mlokal ? 20 for all
    measuring parameters

4
Risk figure Rk
  • Calculation of the local gliding cpk value
  • cpk (local) MIN(cpk, USL (local) cpk, LSL
    (local))
  • For cpk (local) lt 1.67 index (red)Process
    ability is insufficient corrective action items
    should be defined. The product must be seperated
    in a stock, and can be requalified as good part
    by screening methods or by risk evaluation of the
    quality engineers, when the real cause for the
    bad cpk value has been found out.
  • For cpk (local) ? 2.0 index (green) Process
    ability is sufficient.
  • For 1.67 lt cpk (local) lt 2 index
    (yellow)Process ability is poor corrective
    action items should be defined. In case, that
    there is no valid reglementation it is allowed to
    ship the product to the customer. But it is
    neccessary to sign up the product for
  • a good tracebility in prevention of possible
    later reclaims. When the quality is recorded
    please give a hint, that the PAA observation
    level
  • was reduced.
Write a Comment
User Comments (0)
About PowerShow.com