Barry-

Its not always clear in the books, but there are 2 different kinds of DOE methodology: classical and Taguchi.

Classical DOE will work when your experimental population and real life population are of the same distribution.  If you are concerned about warped BGAs or boards, you hope that your experimental sample has the same properties as you will see every day.

Taguchi DOE forces you to test your process at the extremes of what would ordinarily be random variation.  If you are concerned about the occasional warped BGA or board, you look for process variables that will produce the most consistently good joints, whether the parts are flat as a pancake or warped like a potato chip.  The Taguchi terminology is "robustness in the face of noise".

The Taguchi terminology makes it a little difficult to absorb by just reading a book.  Have your boss send you to somewhere nice for a seminar.  I went to Detroit in February.

For software, I use Statistica by Statsoft.  I have no idea what most of its capabilities are used for, but the DOE module is nice.

I'm guessing you will find statistically little difference between gel flux or paste using classical DOE and small sample sizes.  However, we have found paste to be more robust in the face of BGAs or PCBs that are warped before or during the reflow process.  Also, some pastes and fluxes are better than others in the face of marginal solderability.

Larry Jindra
Mfg Engr Group Lead
TRW Radio Systems
[log in to unmask]
w) 858-592-3424
f)  858-592-3940



-----Original Message-----
From: Barry Gallegos [mailto:[log in to unmask]]
Sent: Wednesday, May 15, 2002 12:51 PM
To: [log in to unmask]
Subject: [TN] D.O.E


Good day folks.

I have never conducted a Design of Experiments, and my illustrious boss has
mentioned that maybe I should perform one on a BGA flux vs. paste rework
scenario.

is there a software or can some one lend some free advise on this.

If necessary please feel free to contact me off line.

Barry Gallegos
Assoc. Process Engineer
Western Electronics
1550 South Tech Lane
Meridian, Idaho 83642
Phone: 208/955-9771
Fax: 208/955-9755

---------------------------------------------------------------------------------
Technet Mail List provided as a free service by IPC using LISTSERV 1.8d
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives
Please visit IPC web site http://www.ipc.org/html/forum.htm for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
---------------------------------------------------------------------------------

---------------------------------------------------------------------------------
Technet Mail List provided as a free service by IPC using LISTSERV 1.8d
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives
Please visit IPC web site http://www.ipc.org/html/forum.htm for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
---------------------------------------------------------------------------------