TECHNET Archives

March 1997

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
David Anderson <[log in to unmask]>
Date:
Tue, 18 Mar 1997 21:49:45 -0600
Content-Type:
text/plain
Parts/Attachments:
text/plain (64 lines)
Ed and others,

In my experience, I have used the supplier representatives as a starting
point for identifying process variables.  They often have technical data
sheets that identify items that should be controlled.  However, some
variables in the field may not be identical to the variables in the lab. 
Air sparger geometry on plating baths may be one example.  

While the suppliers may do an excellent job characterizing their product,
it would probably be unrealistic to expect them to have parameters for
every situation (environment, piece of equipment, etc) where their product
is used.  It would be nice, but I have yet to see it!

In the ideal world, one should run a screening experiment (or several of
them) to identify the major sources of variation.  This DoE should include
as many expected sources of variation as possible.  With fractional
factorials, Taguchi arrays, etc., it usually doesn't take that many runs. 
The responses for the DoE should be the product characteristics important
to the customer (the external customer or the down-line processes).  Many
times, a majority of the suspected sources of variation (and interactions)
will be identified as insignificant (if you're lucky), and these variables
can be controlled at arbitrary conditions.  

There have been occasions where I was struggling to control a parameter to
a tight specification, and the DoE showed that it was not important.  There
have also been occasions where I wasn't controlling something that was very
significant source of product variation.

For the remaining variables that are significant (from the screening
experiments), an optimization (response surface) DoE can be performed to
identify target values.  In addition, the experiment can identify the
process window and operating (spec) limits which provide acceptable
product.  The spec limits may or may not be the same as the vendor
recommended limits.

Once the optimization is complete, capability studies and run charts for
each significant variable can be started to identify control limits and
determine if the process is capable of producing acceptable product.

There are also gauge R&R studies, FMEA's, etc. that should be done to
support the SPC efforts (probably before the experiments are performed). 
When all of the experiments, etc. are completed, one usually has learned a
great deal about the process that he/she is working on.

In summary, while the suppliers are usually excellent sources of
information, it is probably unrealistic to rely on them as a sole source
for critical process variable identification.

That's my 2 cents (whew!),

Dave Anderson

***************************************************************************
* 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]      *
***************************************************************************



ATOM RSS1 RSS2