In this article we’re going to cover all things calibration. What it means, why it’s important, when to do it and how it should be done.

The difference between calibration and accuracy

Calibration and accuracy checks are two different things.  Accuracy checks are when you check if a piece of equipment is working as it should – you’re checking to see if it’s accurate.

Calibration is more than an accuracy check – sure, it requires an accuracy check to see if it’s accurate, but if it’s not, then calibration will make the equipment accurate again.  So, if a set of weighing scales do not read 100g when a 100g weight is put on them, then calibration will adjust the reading, so it does read 100g.

Confusion

Although calibration and accuracy checks are different, the BRC Standard tends to call them both calibration. Unfortunately, they also talk about accuracy checks in product inspection section – which is confusing.  During this article, we will only refer to calibration, when it really is calibration.  Accuracy checks, will be referred to as accuracy checks!

Calibration

Calibration is a very skilled task, which is carried out less frequently than accuracy checks. Calibration checks that the measuring equipment is providing an accurate result, and where it isn’t, the calibration method can adjust the equipment so that it’s accurate again.

Accuracy checks

Accuracy checks are carried out more frequently than calibration, as it is a less skilled task. Accuracy checks will tell you if the equipment is reading the correct result or not. Accuracy checks cannot adjust the equipment so it’s accurate again. Therefore, where the accuracy determines that the equipment is inaccurate, it will need to be calibrated to adjust it, so that’s it’s accurate again.

Recognised standards

You should carry out calibration to recognised standards. This means the method that you use to calibrate a piece of measuring equipment must have been validated to prove that it will work.  Where possible it should be a certified recognised standard.  This means that the company that is doing the calibration have been audited and it has been verified that they are working to recognised standards, and doing it consistently.

But what if there isn’t a recognised standard for what you’re doing?  Well, then you can create your own method.  But – you need to validate that the method works and keep evidence of the validation.

How often?

You need to work out how often you’re going to do your accuracy checks and how often you’re going to do your calibration checks. This should be determined, based on risk.  To keep this simple you should use something that the following hazard analysis method:

Severity: What would happen if the measuring equipment goes wrong?  Would it impact safety, legality, integrity or quality?

Likelihood:  How often does the measuring equipment fail? How likely is it to get damaged?

The more severe the outcome and the more likely it is to fail, then the more often it should be checked for accuracy and calibrated.

You should also refer to the equipment manufacturers guidelines, for frequency.  Plus, they should be able to provide you with the most accurate methods for checking the equipment’s accuracy.

Register

All the measuring equipment that requires calibration and therefore, accuracy checks, should be listed on a register.  The register should also detail where the equipment is, the serial or ID number of each piece of equipment and the frequency of calibration and accuracy checks.


If you have any questions about this subject, or you can add any more detail to the article, please pop it into the comments box below.

Have your say…

9 thoughts on “Everything you need to know about calibration

  1. Hi Kassy
    Could not agree more with you. I have called Accuracy checks verification checks when explaining to peers . Anyway on accuracy documents are you required to put tolerances along with serial no and location of equipment?
    Please advise

    1. Hi
      Do you mean the inaccuracy tolerances? If so, it would be a good idea to put them on the register. The acceptable tolerances for the accuracy checks definitely need to be on the accuracy check records.
      Thanks
      Kassy

  2. BRC Clause 6.4.2 & 6.4.3
    If our company has 10 moisture testers, can we just calibrate one of the testers (by method that can be traceable to a recognized standard or international standard) and use it as a reference equipment/master equipment (Clause 6.4.3 = calibration) to check the accuracy of the other 9 moisture testers (Clause 6.4.2 = accuracy test)?

    1. Hi Kong
      Yes you can do that because that means the 9 moisture testers are still traceable back to recognised standards. You must keep the reference equipment safe so it stays in calibration. You must also risk assess how often the reference equipment must be calibrated.
      Thanks
      Kassy

      1. Hi Kassy,

        Thank you very much for your quick response. May I Know, how should we compare the reference equipment with the other 9 testers? What is the common acceptable criteria here? Does it base on Standard deviation and other parameters?

        1. Hi
          I would use both testers to test the same material at the same time. The material you test should be at the top and at the bottom of the range that the tester would normally check samples for.
          Compare it to a temperature probe. You would use the master probe and the other probe, and put them both in ice water. They should read the same +/- the tolerance you’ve set. Then you would repeat for boiling water.
          Does that help?
          Kassy

          1. Hi

            Our scenario here is, moisture tester for rice product and paddy.
            Example: This is a rapid moisture tester and normally we do it three times then the result are:

            Reference Tester: 12.6%, 12.5%, 12.9% (Average = 12.7%)
            Tester 1: 13.0%, 12.9%, 13.0% (Average = 13.0%)

            With these results, we do not have a strong reason to say that Tester 1 is still accurate or not (not significant difference) or if different at what standard error & can we note on Tester 1 about the error so that whenever it is used the standard error is considered during record taking.

          2. I can’t comment on the error I’m afraid, as I don’t know the context or the acceptable tolerance for the equipment by the manufacturer. I would check that first. My advice would not be to use the average though, as this could average out an unacceptable result.

Share your thoughts…

Your email address will not be published. Required fields are marked *

We've tagged this article as:

The icing on the cake

We've got a range of products for organisations of every size. Our clients agree that they really do put the icing on the cake…