| |
| Confucius |
|
Guest
|
| Posted: Wed Sep 17, 2026 11:08 am Post subject: Cpk for Test Parameters |
| |
Can someone explain to me what's Cpk in the context of test parameters? Is this the same as Cpk in normal processes? Is the computation for Cpk in test the same?
Confucius |
|
| Back to top |
|
| testeng99 |
|
Joined: 20 Jun 2026 Posts: 20
|
| Posted: Thu Sep 18, 2026 10:19 am Post subject: |
| |
Confucius,
I think Cpk is still the same when used for test. The USL and LSL will be your test param's max and min limits (suggest you use the prod limits, not QA). The mean and sigma will be based on your actual test data.
Test Engg 99 |
|
| Back to top |
|
| testeng99 |
|
Joined: 20 Jun 2026 Posts: 20
|
| Posted: Thu Sep 18, 2026 10:20 am Post subject: |
| |
I forgot to mention that you can only use this for params with normal distributions.
Test Engg 99 |
|
| Back to top |
|
| Guest |
|
|
| Posted: Tue Mar 22, 2026 3:12 pm Post subject: |
| |
CPKs are a good indication of where your mean is located with respect to the upper and lower spec limits. the higher the value of the cpk, the farther the mean is from the spec limit of interest.
CPK_hi = (usl-mean)/3sigma
CPK_lo = (lsl-mean)/3sigma |
|
| Back to top |
|
| Guest |
|
|
| Posted: Thu Apr 21, 2026 3:57 pm Post subject: |
| |
I just have a correction. CPK_lo should be computed as shown below.
CPK_lo = (mean-lsl)/3sigma
In addition to the explaination above, cpk's are computed using the three factors:
a. Limits
b. mean
c. The standard deviation(sigma)
Assuming that the limits are known, the cpk's would then be :
1. directly related on how far the mean is from the limits(USL or LSL)
2. and indirectly related to the standard deviation(sigma)
= this means that the cpk would be reduce if readings on a certain parameter is erratic(resulting to a large Standard Deviation)
Asmoth |
|
| Back to top |
|
| new wave |
|
Guest
|
| Posted: Fri Apr 22, 2026 9:44 pm Post subject: |
| |
Con-fuse-use,
wag ka nang maghirap pang umintindi sa mga notations sa expalanations ng mga to.
just go to www.google.com and research about Cpk and statistics.
or go to a good, I mean a GOOD engineering library. I suggest you go to UP library, its the only good engineering library.
masyado ka kasing outdated. upgrade yourself.
Ka Totoy |
|
| Back to top |
|
| Guest |
|
|
| Posted: Mon Jun 13, 2026 10:16 am Post subject: |
| |
| Cpk is a measure of how capable your process is. So if you are talking about the Cpk as applied to a test parameter, then the question is what process are you considering? It is either you are measuring the capability of your test process ( tester,test program, etc ) or the capability of the fabrication of the die itself. In a product engineer's point of view, I would think the Cpk which is of more importance will be the capability of the wafer fab process as this translates to yield,cost,reliability,etc. This does not mean that your test setup should not be given importance as well. |
|
| Back to top |
|
| Choose Display Order |
|
| User Permissions |
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
Skin Created by: Sigma12 Powered by phpBB © 2001, 2002 phpBB Group
|
|