Altering a tool after Calibration?

Hello all,

I am new here so I hope I am posting in the correct forum! Also, I am not sure if this has been answered before as I cannot find anything relating fully to my situation.

I am developing a Quality Management System for a small company working towards getting the ISO 9001 certification. For background they have never documented anything, so I am starting from scratch. Regarding calibration, I have come across an issue here that I have never seen before and as I am not that well versed in calibration I am hopeful you can help.

The company I work for sends out load cells for calibration to get “the piece of paper” but when they arrive back, they adjust them to get the readings they want (the companies’ tolerances are tighter than the standard). My questions are 1) Does this not invalidate the calibration? 2) if it is allowed to be done, how do I document this? The manager here doesn’t believe/want this to be documented so now we are “debating” what is or is not considered calibration, verification, adjustments etc. and how it fits into ISO Requirements.

I have asked why we don't request the tolerances we require and have been told many different reasons for why the calibration companies won't or can't meet them.

Please any advice would be greatly appreciated!!

I don't think we have enough information to say definitively about your specific situation, but we can certainly try to clarify things.
First, calibration is comparing the load cells measurement results against a standard. If it is within the load cell's tolerance, you document that. If it is outside the tolerance, you may make adjustments (if feasible) to bring it back into tolerance and document the as found and final results. Changing the load cell again after the outside source calibrated and documented the results invalidates that calibration.

Now, you may be dealing with a different situation, which must be handled differently from what you described. Calibration needs to be performed on the entire measurement system, not just an individual component of that system. I have seen several egregious examples of this on leak test and functional test equipment where an important component is removed from the entire system and calibrated separately then reinstalled without considering the rest of the system's influence.

A simple exaggerated example:
The measurement device consists of a Fluke multi-meter in a housing. The company added 20-foot leads to the meter to reach the test points on their product. They remove the meter on schedule and send it out for calibration. The resistance of the 20-foot leads is not considered during that calibration, so the entire measurement system is always biased by the effect of that resistance. Calibration should include the impact of the 20-foot leads.

Is your load cell part of a larger measurement system that could have an impact of the force measurements? If so, it needs to be calibrated in its entirety. Consider how a factory weight scale is calibrated. They don't remove and calibrate the load cell. They calibrate the entire scale to capture the influence of mechanical friction.

Thank you for advice! This helps with talking with the Manager. And yes the Load Cell is part of a measurement system so I will look into that too.
Thank you again @Miner!!

My experience with load cell calibrations is that they come back with output data, readings typically in mV / V with a calibrated load applied. Load cells are not "adjusted" - you use the calibration data to optimize the system measurement accuracy when you use them. If your people are applying the new calibration data to their measurement system they are doing the right thing, getting the most accuracy they can from their equipment. They are not adjusting the load cells, that would be a violation of any of the ISO 9000 family of standards ( 7.1.5.2 c) ) that require they are "safeguarded from adjustments, damage or deterioration that would invalidate the calibration status and subsequent measurement results"

Thank you @dwperron!!

For load cells generally, and calibration/acceptance/surrounding processes Morehouse Instruments has some great whitepapers and resources on their website.

They probably have the lowest uncertainties for force calibration outside NIST, and have the lowest torque uncertainties of any US cal lab.

If you can provide a little more about what exactly they are adjusting (just the signal conditioner scale and offsets? Something on the load cell itself? a digital interface?) that might help.

It is also interesting how the company tolerances are tighter than the standard...I'm probably just mis-understanding. Are you referring to a standard like an ASTM or IEC standard? or a standard like a reference standard load cell?

My wife works at transcat so I asked her. You should be able to state either oem calibration or a custom calibration if you ask them.

If you are adjusting a load cell after calibration I’m not sure why you don’t request them to just adjust to your requirements. Seems like double work.