FAQ for calibration

This discussion began on the Ohio Bay Specialists (OBS) LinkedIn site. So since I have the ability to edit/ clean things up here, I would like to build a decent FAQ for calibration. Hopefully, it can increase some traffic to the site.

Note: This is not just my endeavor; this is a group thing. Many of the points below and such were added by others at OBS.

• What is calibration?
  
  The comparison of a measurement system of unverified uncertainty to a measurement system of quantified uncertainty to detect or correct any deviation from required performance specifications using standards traceable (in SI units) to an international recognized standards lab.
  
  
• Calibration is not to be confused with adjustment:
  
  Adjustment of a measuring system is the practice of adjusting and standardizing critical measurement equipment performance (accuracy) to appropriate traceable standards.
  
  
• What does “traceability” mean?
• What is the difference between Calibration and Verification?
• Does equipment used in the Design/Development process need calibrating or only for final inspection?
• Can I calibrate my devices in-house? What is involved?
• Why the 4:1 or 10:1 rule?
• What is "Uncertainty" and how does it influence the decision to accept/reject the Instrument based on Calibration result?
• What are the appropriate intervals between calibration periods?
• How often does it need to be calibrated?
• Why are most calibration frequency intervals kept to an annual cycle ?
• What does it mean when an instrument fails? What steps should be taken?
• I have a Go/NoGo gauge. Must I calibrate or verify ?
• Can a matching part be used as a valid verification traceability ?
• When do I need to have a procedure for calibration, verification ?
• What is the relationship, if any, between calibration and MSA
  
• What role do records have in an effective calibration program?

I think once we have a good listing of the top issues, we can start writing up sections for each.

I look forward to your feedback/input!

 

Why calibrate/verify an instrument when it is not used for final inspection?

 

Does equipment used in the Design/Development process need calibrating?

 

Andy/ Jennifer, I took your more detailed suggestions and combined into one. I removed the more generic/general one.
Thanks for the feedback.

 

Hello, Mark! Thanks for contributing.
You make a couple of good points here. First, I'm not sure how separate it would need to be from 17025. My intent (at least from my perspective) is to create a general approach to the field of calibration. Many people get thrown into roles (QA/ metrology/ responsible for calibration dept.) usually before an audit or before their organization is planning to maintain compliance to an ISO specification. And... they really don't know where to start. How do they approach what should be calibrated? Do they need it "calibrated" every week; or is every five years OK?

etc. So to your point, part of this endeavor is to instruct them how important application and risk is, and that they will need to account for such things.

 

I added a definition for calibration. Let me know your thoughts.

 

You have cracked the code. Verifying feature conformance to requirements using reliable instrumentation has always been a risk-based pursuit to keep cost of poor quality low and customer needs met.

Some standards require a control plan, to list (among other things) the verifications done at places in the process in order to ensure product meets requirements.

We can otherwise define it ourselves in an effort to reduce wasted materials, human time, machine wear and overhead costs that can be impacted by cycle time/time to market. This latter set of considerations can provide justification, given a review of relative costs, for calibrating/verifying production equipment too.

 

It absolutely is a risk-based thinking approach - arguably one of the best because this type of risk is easy for most of us to understand and acknowledge as real.

By the way, I put a quality cost calculator in the Resources forum.

 

I know of someone who got a major nonconformity from a CB (to TS 16949) because of calibration in the design process. While we might debate whether it was appropriate to list it as a major (I did not see the writeup) it makes sense to align the instrumentation's performance at both design and inspection so as to reduce the chance of product nonconformity to specification and all the resulting actions and costs that entails.