Dave,

I agree with your findings. Many shops trying to apply SPC  techniques in their facilities usually end up with a lot of charts showing data points that are "statistically" out of control. The problem is compounded when the best effort applied cannot bring things statistically in control. I have concluded over the years that standard SPC practices cannot be applied in a normal PCB shop unless you have a large staff to manage it and are prepared to maintain a mountain of  information. Basically, most of our processes are depleting in nature. Meaning, the results vary based on production thru puts and requirements. This type of process is difficult to manage if you try to apply "all the rules" of statistics. Many companies loose sight of the fact that SPC applications should be there to support us, not for us to support SPC. 
Based on the above, I developed a course of few years ago designed specifically for the practical application of SPC techniques in the PCB manufacturing environment. I have provided this course to several shops throughout the country with a great deal of success and satisfaction. Although I no longer have the time to provide training courses, if you contact me off line and provide me with your address, I will be happy to supply you with a copy of the text. Maybe it will help.

Regards,

Ed Cosper
Director, Quality Assurance and Engineering
Graphic Electronics, Inc.
[log in to unmask]

----------
From:  Dave Willhard[SMTP:[log in to unmask]]
Sent:  Monday, March 03, 1997 1:58 PM
To:  [log in to unmask]
Subject:  SPC for PCB fabrication

We've found the implementation of SPC to be difficult in our circuit
board manufacturing facility.  Due to the wide of variety of board
designs we produce, I think the best way to use valid SPC techniques would
be to use a standard test panel.  The dissadvantage is that this can be
somewhat expensive and intrusive into normal prod'n (you have to stop
running normal product so the test panel can be run).

Another option is to use standard test patterns that can be printed on the border
of regular production panels.  This also has a disadvantage since
we've found that patterns on the borders of the panel often come out
of a process with different characteristices than features on the
interior area of the panels.  For example, for controlling line widths
in etching processes, we've tried to use test patterns (since different
lots will have different line width specs) but have found that due to 
spray pressures, circuitry density, etc. the measurements taken on the 
panel border don't necessarily correlate well with actual circuit measruements.

If anyone has struggled with these problems, we would appreciate some
information.

My E-mail address is [log in to unmask]

Thanks!

Dave Hall

P.S. - I submitted a request for information on HAL SPC several days ago and received
some very helpful responses.  Thanks!


***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To subscribe/unsubscribe send a message <to: [log in to unmask]>   *
* with <subject: subscribe/unsubscribe> and no text in the body.          *
***************************************************************************
* If you are having a problem with the IPC TechNet forum please contact   *
* Dmitriy Sklyar at 847-509-9700 ext. 311 or email at [log in to unmask]      *
***************************************************************************


***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To subscribe/unsubscribe send a message <to: [log in to unmask]>   *
* with <subject: subscribe/unsubscribe> and no text in the body.          *
***************************************************************************
* If you are having a problem with the IPC TechNet forum please contact   *
* Dmitriy Sklyar at 847-509-9700 ext. 311 or email at [log in to unmask]      *
***************************************************************************