Intermediate Check-Ensuring the Confidence of Your Reference Standards During Field/On-Site Calibration

with 30 Comments


I received a comment in my other post asking me on what to do about the high readings of environmental conditions (temperature and humidity) that they are experiencing during their onsite or field calibration compared to their defined specs when inside the lab. He is worried that an auditor may find an issue once it is displayed in their calibration certificate.

I believe this is also a concern to some or most Calibration Techs that performs calibration outside their lab so it is worth a topic to discuss. This can become an audit issue if not properly addressed since it is required to be displayed in the calibration certificate.

Also, one of the questions I received during an audit is, how do you ensure that the reference standards you use outside the laboratory are still in confidence or in good condition once it left and/or returned from the fieldwork?

In this post, I will share with you a quality control procedure that we can easily execute to ensure confidence in your reference standard before or after it is being used outside the calibration laboratory. This is what we call a Functional Check or an Intermediate Check.

I will also share in this article the related clauses under ISO 17025:2017 and ISO 9001:2015 where intermediate checks are a requirement. Once we performed Intermediate checks, we will be confident that we implement these clauses.

This is a must in order to detect any possible issue that can be faced during or after an onsite calibration work. This is also applicable to those instruments or reference standards that are sent out for 3rd party calibration.

I will show to you the step by step procedure on how to implement this.

What is a Functionality Check or Intermediate Check?

An Intermediate Check is a quality check performed to ensure that the reference standard used is still in good condition as your “reference standard” where results are verified and analyzed that it is still within defined limits.

Me during my on-site job exposed to a 46.9 deg C environment
Me during my on-site job exposed to a 46.9 deg C environment

This is one type of calibration quality control that is implemented internally where the resulting data should be recorded and analyzed. This is a planned activity where the frequency of implementation should be laid out.

This is also termed as functionality check, because this is literally what you do, to check and verify the functionality of the reference standard before and after use specific to a fieldwork.

Since we will be focusing on reference standards that are used outside the lab, the frequency is every time it is being used outside the lab. This can fall under a work instruction for all reference standards that will be brought outside the facility.

This is applicable whether the standard is for 3rd party calibration or simply performing an onsite or field calibration.

This is a just a simple cross check so there is no need to perform a full range check. The important thing is to have a program or procedure that is implemented and documented.

It is just a simple way to detect any out of specs reading that may have resulted during the exposure to the different environment or strong vibrations or even mishandling outside the lab.

Why Do We Need to Implement Intermediate Check?

In addition to the reasons above on why we need to perform intermediate check, another reason is that, Intermediate checks are required by Standards, Specifically the ISO 17025 and ISO 9001 Standards.

Below are the related clauses for ISO 17025:2017:
  1. Clause 6.4.2:  When the laboratory uses equipment outside its permanent control, it shall ensure that the requirements for equipment of this document are met.

  2. under clause 6.4.10, which states that:  When intermediate checks are necessary to maintain confidence in the performance of the equipment, these checks shall be carried out according to a procedure.

  3. under Ensuring the Validity of Results, clause 7.7.1 The laboratory shall have a procedure for monitoring the validity of results. This monitoring shall be planned and reviewed and shall include, where appropriate, but not be limited to: (e). intermediate checks on measuring equipment;
Below are the related clauses for ISO 9001:2015: Under Clause 7.1.5: Monitoring and measuring resources

7.1.5.1 General
The organization shall ensure that the resources provided:
b) are maintained to ensure their continuing fitness for their purpose.
The organization shall retain appropriate documented information as evidence of fitness for purpose of the monitoring and measurement resources.

How to Verify the Environmental Specs of your Reference Standard?

UUT specs
Verify the environmental specifications of the reference standards in its manual to ensure it is suited to the environment that you will be exposed to.

Once this intermediate check has been performed, regardless of what temperature or humidity the reference standard has been exposed to you are confident that the calibration you performed outside of your lab is not compromised.

But before exposing your reference standard to a different, more harsh environment, make sure to check first its environmental specifications in its user manual to ensure that the reference standards are designed for that range of environmental condition to avoid damage.

Also, in order to support the validity of the procedure even if the environmental conditions are different compared when inside the lab, below information should be reflected in your calibration record/report:

  1. The term “on-site” calibration – it is expected that a procedure with regards to onsite calibration is in place in your documentation. This is to differentiate it to the calibration performed inside the lab (bench calibration)
  2. Customer approval of the procedure performed– show evidence of customer or user approval.

As part of our quality control, we should consider or implement on how we ensure that our reference standards are still ok while it is being exposed in a harsh environment that can somehow affect its performance, like:

  • very high temperatures,
  • very high humidity
  • strong vibrations
  • contaminants

The procedure is to have another reference standard that we can use to compare and take a reading with. Both should have a valid calibration status. It is the same as calibrating a simple UUT (Unit Under Test) with a reference standard but it does not necessarily a more accurate standard.

As long as we can detect and verify the desired test point then it is ok. But of course, a higher level standard is better if we have it.

A simple Functionality Check for the RTD using a metrology well.
An example of Intermediate Check for the RTD using a metrology well.

One example is an RTD and a Metrology Well. The Metrology Well can be used as a reference standard in performing a functional check with the RTD. It does not necessarily a more accurate standard. I hope you got my point.

Another example of performing an intermediate/ functional check is by using the same reference standard to a previously calibrated UUT. A UUT that is designed for this purpose. See if they have the same or acceptable readings or results comparing it with the previous record.

7 Steps to Perform an Intermediate Check

In performing an intermediate check, below are the steps that I recommend and follow:

  1. The same as performing a calibration, all reference standards that will undergo an intermediate check should be left to stabilize in the lab for at least 1 hour.
  2. You should prepare a checklist or a form to document and list needed details about the intermediate check. Below are some details to include (and anything that is significant during checking):
    • The identification of the Reference standard to be used for cross-checking.
    • The name of the standard that will undergo intermediate check,
    • Brand
    • Model
    • Serial No.
    • Temperature and humidity
    • Date of functionality check when released
    • Date of functionality check when returned
    • Who performed the functionality Check
    • The physical conditions including accessories
    • The tolerance to determine pass and fail status.
  3. Choose a test point for the intermediate check as per available functions or nominal value.
  • For example: In an RTD, you can check only at 0 deg C and 100 deg C test point for a 0 to 200 range                                                                  For a Test Gauge, check only on its mid-range (50% of full scale)
  1. Perform at least 3 trials.
  2. Compute for the error and record. Based the tolerance in your procedure or with the manufacturer specs. It can be also on the calibration certificate. How to check?  See this link>> 3 ways to verify the measurement results in a calibration certificate
  3. Have it approved by an authorized person then it is ready for fieldwork.
  4. Then once the reference standard has returned to the lab, repeat steps 2 to 5.
  5. And you’re done

Conclusion

Intermediate checks are part of a laboratory quality control procedure. This can be used as an evidence of a good performance of a reference standard even if you use it outside the defined specs of the lab with respect to bench calibration.

This is how we can demonstrate the validity of results, if ever that you will be audited regarding the uncontrolled outside environment that we and the reference standard are exposed to.

I have discussed in this post the meaning of Intermediate check which is also called a functionality check,  the importance of an intermediate check as a part of a calibration program under quality control, and a simple procedure on how to implement this including the details to be displayed in your report.

Do you have other means of quality control that you perform during a field or onsite calibration for your reference standard? Please comment below.

If you liked my post, share and subscribe.

Your guide in calibration,

Edwin

30 Responses

  1. Jollin A Doshi
    | Reply

    Other way around is the master is calibrated so u know the uncertainty..

    Calibrate uuc of same accuracy and maximum range and calculate the En ratio if the same is less then 1 the master is perfect

    • edsponce
      | Reply

      Hi Jollin,
      Thank you for the comment. An additional method for quality control involving intermediate check, the same calculation using the interlab comparison method I believe. Very nice.

      Thanks and regards,
      Edwin

  2. Vinod
    | Reply

    Thanks for your post, our auditor also telling to calculate En value but how it possible to calculate En Value as we have only one cmc and one uncertainty for that standard?

    • edsponce
      | Reply

      Hi Mr. Vinod,
      I Appreciate you liked my post. Did you perform an interlaboratory comparison? It is another method for quality control utilizing the process of interlab comparison by calculating the En value. A process of comparing the results of a sample/Device Under Test from a reference lab with your results based on measurement results and its uncertainty. In any case, below is the formula of En. En Formula
      The result should be less than 1 in order to be considered passed. You may check this standard for further reading: ISO/IEC 17043:2005 “Conformity assessment — General requirements for proficiency testing”. This standard will explain more how the formula will be used.

      For any other concern, please comment further.

      Thanks and regards,
      Edwin

  3. beverlyne Sandoval
    | Reply

    How can i performed intermediate check for the dimensional equipment such us gauge blaock set

    • edsponce
      | Reply

      Hi Ms. Beverlyne,

      First of all, thank you for visiting my site.

      We perform an intermediate check to ensure that the reference standard (gauge block) is in confidence while it is still within its calibration interval.

      Gauge blocks have no mechanical or moving parts that can drift or wear-off, therefore, what can we do as an intermediate check is a visual check only.
      During the visual check, this is what you can do:
      1. check if it has a scratch
      2. check if it is deformed in any way
      3. check if it is cleaned and with oil for corrosion resistant

      If a scratch or deformation is observed, that is the time to use a calibrated caliper to verify its reading and parallelism.

      Ensure that the gauge block is stored back to an acceptable and stable environmental condition while performing a verification.

      Hope this helps.

      Edwin

  4. IBRAHIM
    | Reply

    THANKS A LOT ,
    How can i performed intermediate check for Temperature ? I have 2 temp. sensor (FLUKE) with same model PT 100
    Also for St. weight M1 ?

    • edsponce
      | Reply

      Hi Ibrahim,
      You are welcome. I am glad you liked it.
      There are 2 ways in which you can perform an intermediate check for your temp sensor.

      FIRST is to compare them directly. If you do not have a standard like a metrology well or dry block, you can use the ambient temperature. Our objective here is to have a reference reading before and after the use of the PT100. Do not forget to document or record your readings.

      SECOND, if you have a calibrated dry well, you can use that as your reference for the comparison of temperature. Refer to the procedure I presented in my post above for the execution.

      For std weight, usually, this type of standard can be check through visual only. Just record everything that you observe. Does it have scratches, dents, or a sign of deterioration after it was returned to the lab. Record all observation. If you observe any irregularities like too many scratches, that is the time to perform verification by using a calibrated weighing scale. Or worst, have it recalibrated.

      But if your resources can support, in addition to the visual check, you can use a weighing scale to verify each M1 standard before and after use.

      Hope this helps.
      Edwin

  5. Ansar
    | Reply

    Hi,

    Thanks for your information about intermediate check.If you could help me to know about what criteria we can go for RTD thermometer which we are using as master thermometer.Please reply me

    Regards
    Ansar

    • edsponce
      | Reply

      Hi Ansar,
      Thank for reading my post.
      If you are referring to acceptance criteria, then below are my suggestions.
      1. You can check the manufacturer specifications, look for the accuracy part. You can use the accuracy as the tolerance for the acceptance criteria.

      2. Another way is to check the calibration certificate and look for the uncertainty results on the range you have chosen. The uncertainty results can also be used as the tolerance for the basis of acceptance criteria.

      3. The final method is to use the computed standard deviation at 95 % confidence level.
      If you are performing an intermediate check for example at 100 deg C using metrology well, measure at 100 degC 20 times.
      a. Record all the 20 results in an excel sheet and use the standard deviation formula.
      b. Compute for the standard deviation,
      c. Then the final result, multiply it by 2 to expand it to a 95% confidence level.

      This can now be used as your tolerance for acceptance criteria.

      There is no wrong choice here, it depends on what method you are most comfortable with. If you want more strict acceptance criteria, you can choose the 2nd option.

      There are other ways to select or determine acceptance criteria, the important thing to do is to document the method you choose.

      Hope this helps, do not hesitate to message me if you still have a question.
      Edwin

  6. Izzah
    | Reply

    Thank you for your useful information, ive learned a lot

    How can i performed intermediate check for the load measuring machine – compress (ie cube machine) with my reference standard only load cell 50Tonne

    • edsponce
      | Reply

      Hi Izzah,

      You are welcome. I am glad that you liked my post.
      During the intermediate check, you only need to verify one or two test points depending on your requirements. My recommendation as per my post above is 50 percent of the range you are using. If you are using until 80 tons, you can measure 40 tons

      Get the readings and determine if it is a passed or failed based on your tolerance.

      Do not forget to record your results and include all other important details I mentioned in my article above.

      Our purpose is to verify that the load measuring machine is within acceptable conditions before and after use. I believe the load measuring machine is exposed to different conditions outside its normal use that is why you need to make an intermediate check.

      I hope I answered your question.
      Edwin

  7. koti penumala
    | Reply

    in intermediate checks how can i take present value and previous value and how can i give satisfactory results

    • edsponce
      | Reply

      Hi Koti,

      Thank you for visiting my site.
      An intermediate check should be done in a simple way. Do not be confused about performing a calibration during the intermediate check.

      Intermediate check is more on verification; we will verify the reading of our reference standard on a specific range only, to determine that it is still in-tolerance before and after it is used outside the lab.

      And remember that during the intermediate check, it is not necessary to use a higher accuracy check standard. Any previously calibrated instruments are ok to be used as long as it can display the range we needed.

      In order to perform verification, you should know your tolerance or the acceptable specifications of the reference standard that will undergo an intermediate check.

      To explain better, see below example.
      The reference standard for the intermediate check is a Test Gauge.
      1. First, determine the range. Test gauge range is 0 to 1500 psi.
      2. Then, determine the tolerance, As per manufacturer specs, the accuracy is 0.25% of the full-scale reading. Therefore, tolerance is (0.0025×1500) = +/-3.75 psi
      3. Next, based on our procedure, we will perform an intermediate check on the 50% of the full-scale range which is equal to 750 psi. We can use an ordinary calibrated pressure gauge to verify the output pressure of the Test Gauge, our only goal is to get an acceptable reading, (within the +/- 3.75 limit). You can add more test points if you want to.
      4. If during the verification and the results are within the tolerance limits, then it is a satisfactory result or PASSED.
      5. Record your result, name this result as ‘ Before Check’ or ‘Before Value’.
      6. Then once the Test gauge has arrived after the onsite activity, repeat the same procedure.
      7. Verify if it is still within the tolerance limits.
      8. Record again the result and name this as ‘After Check’ or After Value or Present Value.
      9. If all results are satisfactory, then you are sure that your reference standard is in good condition and the calibration you performed is not compromised.
      10. Make sure that you compile and keep all the intermediate check records.

      If you need more clarifications, please do not hesitate to comment further.

      Thanks and regards,
      Edwin

  8. LIM
    | Reply

    Hi Edwin,
    Would you give me some advice on extending the calibration intervals for calibration standards? We are using performance check, intermediate check and functional check as support documents to extend our calibration intervals. How do we use all the past records to determine the equipment stability and suitability to go for longer calibration intervals?

    • edsponce
      | Reply

      Hi Lim,
      Thank you for reading my post.

      Since you already have the past records or history, you only need to show the performance of the reference standards by using the control chart.

      Stability is determined by collecting data on a fixed interval. In can be daily, monthly or every 3 months.

      Our main objective is to determine if the encountered drift is acceptable. Drift is the variation of the performance over time which can be observed as an error. Drift can be detected through the calculated error which can be related to stability that can be observed by plotting the errors in a control chart.

      What I always use for analysis and presentation is the control chart. You can do this easily in excel. See below example:

      I have collected data through an intermediate check performed every three months on a Test Gauge.

      Below are the data.

      Intermediate Check Results

      Determine the error, calculate the mean, and the standard deviation

      The formula from excel:

      Excel formula for Control Chart

      Plot in the control chart.

      Control Chart fro Stability

      As long as the error is within the control limit, we can be sure that the reference standard is very stable and in control.

      My suggestion is that in addition to the intermediate check for calibration interval analysis, it is best to include the results in a calibration certificate every time the reference standards are brought back from calibration.

      Gather the data and use the same principle that I have presented. This will show you that for example, during the 3 years period of calibration performed, where you will observe based on the chart that the instruments are still stable.

      Therefore, you can extend the interval for 3 years since you have the evidence of its performance within a 3 year period.

      You can also include in your analysis its accuracy performance through the use of manufacturer specifications as the control limits or tolerance.

      This simple method is acceptable as per my experience, but if you want to have a deep or more detailed analysis, there is a statistical tool called Minitab, I did not try it yet but I suggest you check it.

      This can automate the process of data analysis which includes the control chart, repeatability, reproducibility, trend analysis and a lot more.

      I hope this helps,
      Thanks and regards,
      Edwin

      • PEIK SEE LIM
        | Reply

        Thanks Edwin for your guidance, it would definitely helps me on this.

        • edsponce
          | Reply

          Hi Lim,
          You are welcome.

          Thanks and regards,
          Edwin

  9. Charlie
    | Reply

    sir, We’re not a calibration laboratory but a manufacturing company. In our lab i perform inhouse calib based on the seminars that i attended. Am i qualified to do this? Our customers audited us and required us to make a calibration procedure for in house calib but i am skeptical about it bcoz the some method that i used was a combination of my experience and based on instruments manual only. Is it acceptable when i include this in my documentation of method used? Thank you.

    • edsponce
      | Reply

      Hi Charlie,
      If you are already trained by a qualified training provider with a training certificate, I believe you are qualified plus you already have experience.

      Below are my inputs on where we should reference a calibration method:
      1. A calibration procedure should be referred (or totally use) from an internationally recognized organization, it means that it is from a well-known organization like ASTM, EURAMET, NIST, Local Government acquired methods or other related organization.

      2. Reference from the Original Equipment Manufacturer OEM, like Fluke or Mitutoyo.

      3. It is from the internally produced method but it is properly validated before use.

      The first 2 items where we reference our methods are called standard methods. These are methods that are published and internationally accepted.
      The last one is called a non-standard method.

      As per your concern about the calibration procedure or methods that you are using, if you use the manufacturer recommended method as a reference, then it is ok and acceptable.

      But if you want to take your procedure to the next level, I will advise using the methods from a well-accepted organization like EURAMET if they have an existing method that you can use. They offer free downloadable guides.

      In any case where there are no manufacturer recommendations or other International reference methods (standard methods) that you can get, then you need to use a non-standard method but need to perform Method validation. This is to ensure that before you use your non-standard method, you have validated it that it is suited for your customer or internal calibration needs.

      Keep all the documentation when performing method validations.

      You can check this guide from EURACHEM for method validation>> The Fitness for Purpose of Analytical Methods

      Thank you for visiting my site. Appreciate your comments.

      Edwin

      • Charlie
        | Reply

        Sir thank you so much for the information. In addtition sir, i would like to ask if i’m doing this right. We have a 2000 g capacity top loading balance. During verification i tested a 100 g std mass based on the mass of the sample that i will be weighing and i also applied correction based on the calib cert. But this will be dilluted to 600 g. I have to tare the mass of the container which is 400 g. Is the mass of the container has an effect to the desired mass that i will be weighing which is 600 g or i have to make a new correction? Thanks in advance.

        • edsponce
          | Reply

          Hi Charlie,

          You are welcome.

          The mass of the container has no effect if you have tared it properly. Meaning the container is clean and in the correct position (Middle).

          If my understanding is correct, you have a standard mass with a known value of 100g. And you will mix this 100g to a 600 g samples. Where in the total mass will be equal to 700 g.

          What I can suggest is:
          1. First, zero the balance,
          2. Second, put the container then tare.
          3. Once tared, then put the 600 g sample, record the reading.( Note that this is a sample, not a standard mass, therefore, it is not expected to be exact)
          4. Then add the 100 g standard mass, record the total reading. Your expected value should be Sample mass + 100g. (approximately equal to 700 g)
          5. Now, this is the time for you to check for corrections from the calibration certificate.
          6. Then record the final answer or reading once corrections are made.
          7. Finally, compare with your tolerance or specifications if it is a pass or a fail.

          Hope I explained it clearly. This is based on my understanding of your concern. You may have your own procedure for verification based on your actual process.

          But the procedure I explained above is the best way to do if you do not have the exact value of standard mass for the verification of a balance.

          Thanks and regards,
          Edwin

  10. Rye
    | Reply

    Wow very informative post sir!! Can i request a calibration procedure for laboratory oven and incubators. Thank you very much!!!

    • edsponce
      | Reply

      Hi Rye,
      Thank you for reading my post.

      Please check DKD-R 5-7: Calibration of Climatic Chambers for the reference procedure. I believe this will help you with the procedure you need. For a simple procedure, you may visit my other post in this link >> electric oven calibration

      Thanks and regards,
      Edwin

  11. Charlie
    | Reply

    Thank you so much sir. This will help me a lot.

    • edsponce
      | Reply

      You are welcome

      Best regards,
      Edwin

  12. koti penumala
    | Reply

    Dear sir how can i find out weighing balance accuracy

    • edsponce
      | Reply

      Hi Koti,
      Have you already check its manufacturer manual?

  13. Utpal Maji
    | Reply

    Sir please explain what is the difference between Onsite Calibration Procedure and Intermediate Check, because we do same process in onsite calibration Procedure. one more question sir can explain how we calculate Long Term Stability of a Instrument with control Chart.. please help me.

    • edsponce
      | Reply

      Hi Utpal,

      Thank you for reading my post.

      Onsite calibration is the calibration job we perform in the site or location of the customer. We perform the calibration outside the laboratory.

      Since we perform the calibration outside the lab, we carry our reference standard with us, when this happens, the reference standards will be exposed to different environmental conditions such as high temperature and humidity, strong vibrations and dusts.

      These environmental conditions may have an effect on the accuracy of our reference standards. In order to detect it and to ensure that there is no significant effect on the accuracy of our standards, we perform Intermediate check.

      Intermediate check is a simple verification and functional check that we perform inside the lab as per the procedure I present in the post above.

      We will perform the intermediate check before we send the standard outside then repeat the same verification after we return back the reference standards.

      For the calculation of results for long term stability, there are many ways to do it but I will share the most basic and simple to do. Just remember that the goal for stability calculation is to determine that the instrument or standard is functioning within specifications on a defined period of time.

      Below are the things that you can do to monitor stability of your reference standard.

      1. Collect all the past calibration certificates of your reference standards to be evaluated, the more the better. Check the data results, or..

      2. Stability is also determined by collecting data on a fixed interval. This could a data from your intermediate check. In can be daily, monthly or every 3 months.

      3. Organize your data on a table, see below example: (this example is the same above-see other comments)

      I have collected data through an intermediate check performed every three months on a Test Gauge.
      Below are the data.

      Intermediate Check Results
      Intermediate Check Results

      Our main objective is to determine if the encountered drift is acceptable. Drift is the variation of the performance over time which can be observed as an error.

      Drift can be also detected through the calculated error based on the difference of past performance with the present performance (past minus present) .

      Stability can be observed by plotting the errors in a control chart.

      What I always use for analysis and presentation is the control chart. You can do this easily in excel. See below example:

      Determine the error, calculate the mean, and the standard deviation

      Excel formula for Control Chart

      Plot in the control chart.

      Control Chart fro Stability

      Control Chart for Stability

      As long as the error is within the control limit, we can be sure that the reference standard is very stable and in control.

      You can also include in your analysis its accuracy performance through the use of manufacturer specifications as the control limits or tolerance.

      This simple method is acceptable as per my experience, but if you want to have a deep or more detailed analysis, there is a statistical tool called Minitab, I did not try it yet but I suggest you check it.

      This can automate the process of data analysis which includes the control chart, repeatability, reproducibility, trend analysis and a lot more.

      I hope this helps,

      Thanks and regards,
      Edwin

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.