Calibration of measuring equipments

[Ganesan]

In our registrar audit the auditor requirements is to determine the calibration acceptance accuracy by our company.We are sending our instruments for third party involving in calibration and the certificate stated whether the instrument intended to use or no.
It specifies the acceptance error reference to international standard.
Please provide your expertise whether company should have accuracy limits before accepting the calibrated instrument.
The file attached for your comments.

 

[Manoj Mathur]

Same question from my side also.

Can any one of you tell me what should be accuracy of calibration? Is it one fifth of tolerence (Just a Thumb Rule what I read from Somewhere/)

Manoj

 

[Icy Mountain]

At Least 25%

Ganesan and Manoj,
This is from ANSI/NCSL Z540-1-1994: Section 10.2 b) "..the collective uncertainty of the measurement standards shall not exceed 25% of the acceptable tolerance..."

Meaning: If you have a voltage meter that needs +/- 5% accuracy, then the power supply (your voltage standard) used to calibrate the voltage meter needs to be at least +/- 1.25% accurate. That's where your 1/5 rule of thumb comes in. Good labs will use a +/- 1% standard to calibrate +/- 5% equipment to give themselves some room for error.

I believe that ISO/IEC Guide 25 (I don't have a copy in my file, oops) has superceded the above standard. I sure one of the others here can make sure this hasn't changed.