Statistical Process Control for Precision Machining - Part 1

G

Guest

Mr. Doering,

Could you share with me how you created the hi/low InfinityQS chart please? We are using Proficient and I can't seem to find my way to recreate anything that resembles the example. Thanks much! Tamarlyn
 

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
Tamarlyn;bt590 said:
Mr. Doering,

Could you share with me how you created the hi/low InfinityQS chart please? We are using Proficient and I can't seem to find my way to recreate anything that resembles the example. Thanks much! Tamarlyn

I will post how to do it. I need to do a step by step, so give me a little time. Thanks for your interest!
 
G

Guest

Mr Doering - thank you very much for taking the time to reply and for posting the step by step instructions for Infinity. I'm looking forward to the description. Best Regards.
 

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
Tamarlyn;bt590 said:
Could you share with me how you created the hi/low InfinityQS chart please? We are using Proficient and I can't seem to find my way to recreate anything that resembles the example. Thanks much! Tamarlyn

I made a new blog with the technique. Please let me know how it works for you!
 
G

Guest

Hi Bob, I'm interested with the technique but can't find the new blog. Can you help me please? Thanks in advance.
 

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
Erick Mirabel;bt750 said:
Hi Bob, I'm interested with the technique but can't find the new blog. Can you help me please? Thanks in advance.

The technique for using IninityQS Proficient for the X hi/lo R chart is here. Thanks for your interest!
 
Last edited:

optomist1

A Sea of Statistics
Super Moderator
Hi Bob,

Thanks for the info...I do have a couple of questions, the first may be semantics.

Regarding Correct SPC parts 1 & 2, when you reference roundness I assume this is the same as circularity.

I like the Hi /Lo chart...where or at how many points along the length of the "air bag plug" did you measure roundness/circularity?

Was or is barreling/waisting or taper a concern?

What device/gage did you use to conduct the roundness measurment?

Thanks again....good material
Marty
 

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
optomist1;bt817 said:
Regarding Correct SPC parts 1 & 2, when you reference roundness I assume this is the same as circularity.

Yes, but to be even more specific, true GD&T circularity can not be measured as the difference between the high and the low diameter. In fact, it is related to radii, so it will be double. But, for control (if using the correct gage for the applicable lobing), the minor error of this technique and the permitting the factor of two to keep from additional math, will still represent the condition adequately to show the signals we need to see for process decisions. (I hope that didn't muddy that water too badly!)

optomist1;bt817 said:
I like the Hi /Lo chart...where or at how many points along the length of the "air bag plug" did you measure roundness/circularity?

For this particular chart, this is specifically one specific "slice" of the diameter - in fact great care must be taken to ensure the "slice" is taken on the same location of each sample to avoid the effect of taper on the data.

optomist1;bt817 said:
Was or is barreling/waisting or taper a concern?

There is an advanced technique that I have that employs taking 3 "slices" per part, and using that data to control the taper variation. I have not gone into that technique here, as I have tried to focus on the basics first.

optomist1;bt817 said:
What device/gage did you use to conduct the roundness measurement?

Depends on the tolerance. Typically I have used indicators, air gages, bore gages, and even sometimes the dreaded micrometer. The key is to have an ndc>10, when calculated using the control limits (UCL-LCL) rather than PV.

Hope that helps!
 

optomist1

A Sea of Statistics
Super Moderator
Bob,

yes thanks for the responses and insight....I guess at some point (not necessarily in this case/part) the more slices one measures or takes the more likely one is to select a different geometric control say circular or total runout, but then you have to establish a Datum Axis...
 
Top Bottom