Jump to content

  • Quick Navigation
Photo
- - - - -

Criteria for calibrating scales, thermometers, rulers, etc.?


  • You cannot start a new topic
  • Please log in to reply
8 replies to this topic

Maryl

    Grade - Active

  • IFSQN Active
  • 2 posts
  • 0 thanks
0
Neutral

  • United States
    United States

Posted 21 October 2020 - 05:06 PM

Hello everyone, I am working on a cosmetics company and starting to work on a project with our scales, thermometer, rulers, etc, and I have a few doubts. How do you guys decide on what mass to use in the routine verification? Is it based on the use of the scale or its capacity? And how should I decide on tolerances limits, to determine if a equipment passes or fails?



Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 19,097 posts
  • 5313 thanks
1,281
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 22 October 2020 - 08:55 AM

Hello everyone, I am working on a cosmetics company and starting to work on a project with our scales, thermometer, rulers, etc, and I have a few doubts.

How do you guys decide on what mass to use in the routine verification? Is it based on the use of the scale or its capacity?

And how should I decide on tolerances limits, to determine if a equipment passes or fails?

Hi Maryl,

 

^^(red) - scales are typically calibrated with appropriate, traceable, certified, standard weights.

 

Values/Tolerances will depend on the "specifics".


Kind Regards,

 

Charles.C


beautiophile

    Grade - MIFSQN

  • IFSQN Member
  • 198 posts
  • 70 thanks
19
Good

  • Vietnam
    Vietnam
  • Gender:Male

Posted 22 October 2020 - 09:59 AM

Hello everyone, I am working on a cosmetics company and starting to work on a project with our scales, thermometer, rulers, etc, and I have a few doubts. How do you guys decide on what mass to use in the routine verification? Is it based on the use of the scale or its capacity? And how should I decide on tolerances limits, to determine if a equipment passes or fails?

At least:

A standard weight must be lighter than the lightest value you've ever measured.

A standard weight must be heavier than the heaviest value you expect to measure.

A standard weight has roughly the median value of those two.



Ryan M.

    Grade - FIFSQN

  • IFSQN Fellow
  • 1,261 posts
  • 456 thanks
258
Excellent

  • United States
    United States
  • Gender:Male
  • Location:Birmingham, AL
  • Interests:Reading, crosswords, passionate discussions, laughing at US politics.

Posted 22 October 2020 - 06:59 PM

If you have a scale contractor or service provider get their input.  Each scale / balance will have it's own limits and metrology verification is a lengthy discussion.  Best to leave it to the experts.

Typical, you will routinely verify a scale at the normal use weight / mass versus the capacity of the scale.  This would be your weekly, daily, hourly check (whatever frequency you determine to conduct internally).

 

A scale calibration verification is typically done by an outside contractor on a routine basis.  The frequency should be determined by the scale use, potential of falling outside of calibrations limits, and potential scale abuse.  For example, when I initially started at my current position we had an outside company come in quarterly to verify the scale calibrations were within tolerance.  We found that we could change it to two times per year because we did not experience calibration issues and our scales were not abused.



Charles.C

    Grade - FIFSQN

  • IFSQN Moderator
  • 19,097 posts
  • 5313 thanks
1,281
Excellent

  • Earth
    Earth
  • Gender:Male
  • Interests:SF
    TV
    Movies

Posted 23 October 2020 - 03:17 AM

If you have a scale contractor or service provider get their input.  Each scale / balance will have it's own limits and metrology verification is a lengthy discussion.  Best to leave it to the experts.

Typical, you will routinely verify a scale at the normal use weight / mass versus the capacity of the scale.  This would be your weekly, daily, hourly check (whatever frequency you determine to conduct internally).

 

A scale calibration verification is typically done by an outside contractor on a routine basis.  The frequency should be determined by the scale use, potential of falling outside of calibrations limits, and potential scale abuse.  For example, when I initially started at my current position we had an outside company come in quarterly to verify the scale calibrations were within tolerance.  We found that we could change it to two times per year because we did not experience calibration issues and our scales were not abused.

 

Contracting is of course convenient .

It may depend on the context, eg size/type/application of the scales and available technical resources, but in Principle (and IMEX) calibration can be done internally with the exception of annual revalidation of the standard weights in use.


Kind Regards,

 

Charles.C


Maryl

    Grade - Active

  • IFSQN Active
  • 2 posts
  • 0 thanks
0
Neutral

  • United States
    United States

Posted 23 October 2020 - 11:36 AM

Hi everyone, thank you so much for your inputs.

Yes, we do have a company that comes once a year and calibrate them. We perform routine checks only, in house, but what I've observed is that we decided on check with 3 standard weights (3 points) and this results are been filed but never looked into, so I raised the question about how do we know if its good or not, or what is the criteria do say this is  its good to use, or its a warning or even need to call the calibration company. I've notice that is basically just being recorded as a mere procedure but no actions neither documents/methods are being taken.  On our checks we just perform sensitivity test, so I was wondering if it would be good to start performing eccentricity and repeatablity as well, less often though. Also the standard weights we have apparently have not been calibrated after we bought them, which is another concern that I had when I started looking at it. So those are the main points I was thinking about looking and propose something different. Regarding to the evaluation, do you guys perform any statistic test with the routine tests data to determine what actions to take or its just based on the values you see and how its compared to the tolerances you pre set before?

 

Regards,

Mariana



Ryan M.

    Grade - FIFSQN

  • IFSQN Fellow
  • 1,261 posts
  • 456 thanks
258
Excellent

  • United States
    United States
  • Gender:Male
  • Location:Birmingham, AL
  • Interests:Reading, crosswords, passionate discussions, laughing at US politics.

Posted 23 October 2020 - 06:21 PM

True, good point Charles.

 

Contracting is of course convenient .

It may depend on the context, eg size/type/application of the scales and available technical resources, but in Principle (and IMEX) calibration can be done internally with the exception of annual revalidation of the standard weights in use.



Ryan M.

    Grade - FIFSQN

  • IFSQN Fellow
  • 1,261 posts
  • 456 thanks
258
Excellent

  • United States
    United States
  • Gender:Male
  • Location:Birmingham, AL
  • Interests:Reading, crosswords, passionate discussions, laughing at US politics.

Posted 23 October 2020 - 06:25 PM

I've never done any statistical analysis on scale check data.  It really depends on how you are using the scales and what kind of tolerances and repeatability you wish to see with the scales.  I'm not metrologist so I think it is a conversation for someone at the 3rd party scale company who comes in once a year for you.

 

It is a good idea to have the outside company check your standard weights with your scales to verify them.

 

 

Hi everyone, thank you so much for your inputs.

Yes, we do have a company that comes once a year and calibrate them. We perform routine checks only, in house, but what I've observed is that we decided on check with 3 standard weights (3 points) and this results are been filed but never looked into, so I raised the question about how do we know if its good or not, or what is the criteria do say this is  its good to use, or its a warning or even need to call the calibration company. I've notice that is basically just being recorded as a mere procedure but no actions neither documents/methods are being taken.  On our checks we just perform sensitivity test, so I was wondering if it would be good to start performing eccentricity and repeatablity as well, less often though. Also the standard weights we have apparently have not been calibrated after we bought them, which is another concern that I had when I started looking at it. So those are the main points I was thinking about looking and propose something different. Regarding to the evaluation, do you guys perform any statistic test with the routine tests data to determine what actions to take or its just based on the values you see and how its compared to the tolerances you pre set before?

 

Regards,

Mariana



beautiophile

    Grade - MIFSQN

  • IFSQN Member
  • 198 posts
  • 70 thanks
19
Good

  • Vietnam
    Vietnam
  • Gender:Male

Posted 24 October 2020 - 02:18 AM

I've never done any statistical analysis on scale check data.  It really depends on how you are using the scales and what kind of tolerances and repeatability you wish to see with the scales.  I'm not metrologist so I think it is a conversation for someone at the 3rd party scale company who comes in once a year for you.

 

It is a good idea to have the outside company check your standard weights with your scales to verify them.

I was a labo-chemist and I have seen a lot misconceptions among terms like calibration, validation and adjustment, in industrial plants.

That bold text is called validation and the 3rd party has to be either accredited or linked to an accredited institute. The purpose is to prove the traceability to national standards. During such "validation", technicians will do the calibration and adjustment on instruments, if necessary. In case of the weights, they just validate by comparing with "national weight prototypes".

The task OP's doing (using 3 internal standard weights) is truely a simple calibration for a scale. Only the sensibility check belongs to something else.

BTW, I have also noticed that most auditors (2nd or 3rd party) rarely understand well laboratory worlds to give NCs on the methodologies of calibration, although such methods aren't perfect. However, applying labo-grade implementations to normal industrial QC tests is just an overkill and a waste in cost (workhours,budget)/efficiency, in my experience. As long as a company has got controls on these concerns and its customers feel satisfied of its product quality; its practices are fine.






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users