Lou is right but does not suggest a number. Many people would say something like 25. It's entirely arbitrary and no theory can tell you how many are enough. 25 comes from the reasoning that, under most circumstances, 25 samples will make the sample s.dev. a somewhat reliable estimate of the population s.dev. I recommend 25 in the same way that I recommend 95% as a confidence level: no line is the best but if you insist on an arbitrary line you could do worse than this one. regards, yves Yves Dupuis Process Engineering Leitch Technology Phone: 416-445-9929 X3389 / Fax: 416-445-7927 25 Dyas Road, North York, Ontario M3B 1V7 > -----Original Message----- > From: Lou Hart [SMTP:[log in to unmask]] > Sent: Tuesday, July 03, 2001 9:36 AM > To: [log in to unmask] > Subject: Re: [TN] Sample size in computing cpk > > Ken, before calculating Cpk you need to show the process is in control > with > SPC charts. From the data you use to make the charts you can calculate > Cpk. Let me know if you need more information. Lou Hart > > -----Original Message----- > From: Ken Fong [SMTP:[log in to unmask]] > Sent: Monday, July 02, 2001 10:36 PM > To: [log in to unmask] > Subject: [TN] Sample size in computing cpk > > Dear Technetters, > Could anyone give some idea what the minimum sample size is when > calculating the > cpk of a process. And based on what theory to arrive at this minimum > sample > size? > > Thanks, > Ken > > ------------------------------------------------------------------------ > --------- > 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 the following message: SET > Technet NOMAIL > Search previous postings at: www.ipc.org > On-Line Resources & Databases > > E-mail 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 the following message: SET > Technet NOMAIL > Search previous postings at: www.ipc.org > On-Line Resources & Databases > > E-mail 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 the following message: SET Technet NOMAIL Search previous postings at: www.ipc.org > On-Line Resources & Databases > E-mail 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 ---------------------------------------------------------------------------------