Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Register Log in
  • Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Metrology fundamentals 1

Status
Not open for further replies.
cervantes

cervantes

Mechanical
Aug 3, 2006
85
Hello,

just to be sure:

is a thesis that "measurement should be done with an instrument which resolution is 10x greater than measured eg. length" is a "golden rule" or is it described in any norm?

example: to avoid measurement error analysis for length 23,5 I should use caliper with resolution 0,01,
or for length 23,54 I should use caliper with resolution 0,001?

thanks in advance
 
Replies continue below

Recommended for you

Sort by date Sort by votes
This probably comes back to the convention to allow 10% of the dimension tolerance to be used for inspection/gaging tolerance.

Might be in ANSI B4.4M (not sure if that's an ASME now).

Posting guidelines faq731-376 (probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​
 
Upvote 0 Downvote
Keep in mind that resolution is very different from accuracy or repeatability. Just because a display on a digital caliper will show you 1/100 ths of a millimeter does not mean that the measurement is accurate to 1/100 ths of a millimeter. As an example, if you follow this link Link to the Mitutoyo website for some of their calipers and scroll to the bottom you will see a table that contains values for both accuracy and LCD resolution. This is a basic example, things can get even worse with more complicated inspection systems like 3D scanners or articulating arm CMMs.

Doug
 
Upvote 0 Downvote
That is an incorrect misstatement. describes "test accuracy ratio" (TAR) where the ratio of the allowable test uncertainty is divided by the uncertainty of the test equipment. MIL-HDBK-52B lists "high" TARs as 4:1, 10:1. 10:1 TAR was the default requirement in military contracts for many decades.

Note, however, just because one specific tool has a TAR of 10:1 does not necessarily mean that an analysis can be ignored. There may be other factors that might magnify errors in other parts of the system.

TTFN
faq731-376
7ofakss

Need help writing a question or understanding a reply? forum1529


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com:
 
Upvote 0 Downvote
Typical military, they define it the exact opposite of accepted industry standards. Gauge repeatability and reproducibility (GR&R) is normally required to be 10%.

We just had a TS16949 audit and the auditor insisted that every gauge had to have at least 10 increments in either direction from nominal to the upper or lower limit. We got dinged because we were measuring a +/-8 micron tolerance with a 1 micron resolution gauge.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.
 
Upvote 0 Downvote
That seems to be overkill. The 1 micron resolution implies 0.5 micron repeatability, so he's actually asking for 16:1

TTFN
faq731-376
7ofakss

Need help writing a question or understanding a reply? forum1529


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com:
 
Upvote 0 Downvote
Isn't it also important to record the measured value with the correct number of decimal places on the inspection report? For example, if a drawing dimension is shown to 2 decimal places, shouldn't the inspection report also show the measured value to 2 decimal places? This prevents rejection of some parts for being out of tolerance due to rounding effects. For example, if I measure a part dimension shown on the drawing to 2 decimal places using a caliper with a readout to 4 decimal places, and the drawing dimension is 0.25 +/-.03 while the caliper readout shows 0.2801 which is what gets recorded on the inspection report, then the part would get rejected for being 0.0001 oversize. But if the caliper readout was only 2 decimal places, then the value recorded on the inspection report would be 0.28, which would be acceptable.
 
Upvote 0 Downvote
thank you for all your thoughts!
I didn't get clear answer but they motivate me to look at the problems from other angle.
 
Upvote 0 Downvote
Drawing dimension tolerances are absolute limits, so if it measures .2801 (and TAR or GR&R is allowed for), it should be rejected.

"Know the rules well, so you can break them effectively."
-Dalai Lama XIV
 
Upvote 0 Downvote
Ultimately it is a matter of the odds of accepting bad material and the cost of rejecting acceptable material. The more closely a feature approaches the border between accepting and rejecting the greater the quality of the measurement needs to be to discern it.

If I wanted to be very sure that a +/-.25 tolerance was met, but had a tool that was +/-.03 I'd change the inspection limits to no larger than +/-.22. To get to .+/-.249, the instrument accuracy would have to be +/-.001.

Like I said, it depends on how things will be if you don't know what you are paying for and what you are throwing away. The closer to the edge, the more it costs and the more careful you need to be.
 
Upvote 0 Downvote
I agree that tolerance limits shown on drawing are absolute. But the reason I ask is I have seen a situation where an FAI report was rejected because the inspected values listed did not have the same number of decimal places as the drawing dimension. The source inspector simply recorded every linear measurement to the same number of decimal places he saw on the CMM display. And there were a couple entries where the 4 decimal place measurement was slightly over/under, but he assumed they were OK if rounded off to 2 decimal places. Our receiving inspection rejected the FAI report as being non-conforming, and gave a corrective action to record the inspection values on the FAI report using the same number of decimal places as the dimension on the drawing. A couple days later we received a revised FAI report and it was accepted by our QA.
 
Upvote 0 Downvote
tbuelna - I would reject those too, it's not at all unusual to measure to more decimal places than the drawing. If I have a dimension of 20+/-0.1 I would expect it to measured to at least 2 decimal places if not 3. If the gauge only read increments of 19.9, 20.0 and 21.0 it could not possibly have an acceptable GR&R.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.
 
Upvote 0 Downvote
Inspection dimensions should never be rounded. That's a way to accept non-conforming material. They should include the width of the 3sigma error in their measurement ability, but they never do.
 
Upvote 0 Downvote
Some companies do that, occasionally, but having to resort to the benefit of rounding suggests that the process is wandered off track or the design has insufficient margin.

TTFN
faq731-376
7ofakss

Need help writing a question or understanding a reply? forum1529


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com:
 
Upvote 0 Downvote
In looking at MIL-HDBK-52, I find that it is cancelled without replacement. It referenced MIL-STD-45662, which has also been cancelled and superseded. I found this page which sums up my feelings on the waste of money and affect on companies when the DoD was directed to exit the standards business. From that page "One of the sad results of Mil Spec reform is the fact that expensive industry standards, like the two documents above, have replaced the inexpensive (or free) military documents that companies around the U.S. used to reference extensively. Because of the high cost of industry standards, many of our customers have reduced or eliminated standards from their business practices. "
 
Upvote 0 Downvote
"waste of money and affect on companies when the DoD was directed to exit the standards business"

This was the effect of the false economy that certain pols have continued to preach, even now.

TTFN
faq731-376
7ofakss

Need help writing a question or understanding a reply? forum1529


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com:
 
Upvote 0 Downvote
Status
Not open for further replies.

Similar threads

V Thanki
  • Locked
  • Question
Need Help with Fundamentals 8
Replies
13
Views
227
mfgenggear
mfgenggear

Part and Inventory Search

Sponsor

Back
Top Bottom