What specifically must be calibrated per ISO 9001:2000?

[jrcook5]

I know that this topic has been touched on before but I haven't seen a difinitive answer on exactly what devices ISO 9001 and 13485 require to be controlled to the level of the standards. We are getting inundated with requests to calibrate equipment and I have been commisiond to define the criteria for determing which devices to include in the control system. Please comment on the following statements. Agree, disagree or am I all wet?

Control of "monitoring and measuring devices" is required by ISO standards and FDA regulations. ISO 9001:2000 and 13485:2003 require "devices needed to provide evidence of conformity of product to determined requirements" be identified and controlled to ensure valid measurement results. The CFR Title 21, Part 820, Subpart G, Section 820.72 stipulates "inspection, measuring, and test equipment" be "suitable for its intended purpose" and "capable of producing valid results." Both ISO Standards and FDA Regulations specify equipment calibration as a component of the control system and outline the necessary requirements to ensure the device measurement system achieves the intended purpose.

ISO, in previous releases of 9001, specified IMT&E "used by the supplier to demonstrate the conformance of product to the specified requirements" must be controlled, and periodic calibrations performed. ANSI/ASQC Q9001-1994 element 4.11 CONTROL OF INSPECTION, MEASURING, AND TEST EQUIPMENT, directs the supplier to refer to ISO 100012 QUALITY ASSURANCE REQUIREMENTS OF MEASURING EQUIPMENT for guidance implementing a control system. ISO 100012 is very specific in scope regarding measuring equipment. It is applicable to measuring equipment "used in the demonstration of compliance with a specification"… "it does not apply to other items of measuring equipment".

FDA has provided guidance for implementing a calibration program in Chapter 7 EQUIPMENT AND CALIBRATION, of HHS Publication FDA 97-4179 MEDICAL DEVICE QUALITY SYSTEMS MANUAL: A SMALL ENTITY COMPLIANCE GUIDE. The Guide indicates that "Some instruments need only to be checked to see that their performance is within specified limits, while others may require extensive calibration to a specification." It continues "Sometimes equipment used only for monitoring a parameter need not be calibrated but should be identified (e.g., for monitoring)."

These clarifying statements by FDA along with those by ISO, indicate that:
a. Only devices used to demonstrate product conformance must be controlled to the level outlined in the standard/regulations.
b.Other devices may be checked, verified, or tested based on utilization.

Your comments please,
Jim

 

[Al Rosen]

Calibration by any other name is still calibration.

calibration
n : the act of checking or adjusting (by comparison with a standard) the accuracy of a measuring instrument; "the thermometer needed calibration"

 

[qualitygoddess - 2010]

I hate to say it this way, but err on the side of caution. You have to respond to the requirement to calibrate based on your business' needs for control of measurement equipment, and with what standards your business needs to comply. You are certainly taking this assignment seriously, so you are the right person for the job. If you have to defend the calibration of every piece of equipment to some management group, then I suggest you choose the appropriate standard(s) for your business and get started.

 

[jrcook5]

Calibration by any other name is still calibration.

I don't get it. I know the definition but how does that help me to determine which devices get calibrated and controlled to the requirements of the standards.

 

[Rob Nix]

Anything you use for measuring anything has a certain level of "calibration" attached to it. You buy a bathroom scale and assume it's accurate at the time of purchase and give it no thought afterwards (until it starts indicating higher than usual readings - in that case it is in error ). The thermostat at work may never be checked - unless it is used in a temperature controlled lab.

So it is all a matter of degree. And you the business decide the level of control. Keep a list of the things you consider worthy of being calibrated and use it when being audited. Again it is a matter of risk vs. benefit.

True, normally ISO, et al tends to direct the requirement at the measuring and testing of product, but any variable check that affects the bottom line might be "calibrated" if needed.

Now, if the auditor finds some sort of device and asks about calibration, you either have to show him it is covered, or have a good explanation why it doesn't need to be calibrated.

It might help if you gave us a few examples of things you currently calibrate and some things you think do not need it. That oughta get us going again.

 

[Paul Simpson]

If you need it you need to calibrate it

I don't get it. I know the definition but how does that help me to determine which devices get calibrated and controlled to the requirements of the standards.

If you make any judgement on the suitability of the product to meet requirements (customer or internal) then the equipment needs to be calibrated.

If you use a piece of equipment but there is another test / inspection / measurement further down the line that says whether the item is ok then you probably don't need to calibrate (and should mark the equipment as to its status - e.g. indication only).

 

[jrcook5]

cal criteria

Let me approach this from another angle. Initially I want to determine what devices I have to calibrate based on:
a: is the device is used to accept/reject product
b. does ISO 9001 or 13485 require it (applicable standards +FDA)
c. satisfy other safety,environmental, or internal requirement.

a. is easy, I just get a sales order and look at the product specs and work backwards to see what devices are used to measure that charecteristic of the product.

b. is where the trouble begins. My understanding is that ISO 9001:2001 and CFR Title 21 Part 820.72 indicate the same criteria as I listed in item a., if the device is used to demonstrate conformance to a specification, then it must be controlled which includes calibrating the device.

It makes sense to me to limit the devices that must be controlled to the level required by the standard to just the devices demonstrating product conformance. The key is not substituting calibration for control here. Calibration is a subset of control and they are not synonomous. Calibration is not the objective, control is.

However, I am constantly approached by employees who want a device calibrated and the basis for their decison is "they feel it affects quality."
Six hundred devices later it's time to define a more coherent criteria.

What I would like to establish as the criteria for what devices we enter into the control system would simply be, show me the product spec and we will control the device to the level dictated by the standard.
Other devices may get calibrated or verified based on utilization but this could be accompplised via another process such as the p.m. system.
I'm not sugesting that there may not be some benifit in calibrating other devices for a variety of reasons, only that they do not require to be controlled to the level of the applicable standard.

 

[jrcook5]

If you make any judgement on the suitability of the product to meet requirements (customer or internal) then the equipment needs to be calibrated.

If you use a piece of equipment but there is another test / inspection / measurement further down the line that says whether the item is ok then you probably don't need to calibrate (and should mark the equipment as to its status - e.g. indication only).

Paul, I like your answer. But let me make sure I understand. Are you saying that your answer would pass they smell test by a nit-picky ISO auditor.

 

[Jim Wynne]

The key is not substituting calibration for control here. Calibration is a subset of control and they are not synonomous. Calibration is not the objective, control is.

You seem sure about what the intention of the standard is, so I'm not sure what you're trying to find out. Just out of curiousity with regard to the calibration/control thing, can you give an example of gage control that doesn't involve calibration, but protects against erroneous measurement?

However, I am constantly approached by employees who want a device calibrated and the basis for their decison is "they feel it affects quality."
Six hundred devices later it's time to define a more coherent criteria.
What I would like to establish as the criteria for what devices we enter into the control system would simply be, show me the product spec and we will control the device to the level dictated by the standard.

This isn't an unreasonable approach, although I would be careful in discouraging people from being concerned about their measurements.
What Rob said earlier is good advice:

if the auditor finds some sort of device and asks about calibration, you either have to show him it is covered, or have a good explanation why it doesn't need to be calibrated.

I'm not sugesting that there may not be some benifit in calibrating other devices for a variety of reasons, only that they do not require to be controlled to the level of the applicable standard.

Forget about the standard(s), which prescribe only a minimum, common sense level of control. If it makes sense to calibrate (remember, "calibration" doesn't necessarily mean "adjustment") then do it because it's the right thing to do, if it is.

 

[Hershal]

First, ignore the ISO auditor.....s/he may or may NOT have a really good idea of what needs cal or doesn't, based on the process involved.

Now, look at the measurement.....if there is no need to calibrate something then what happens if you take away the measuring equipment? If a set of calipers or a thermometer or an oscilloscope "do not require calibration" then why have the items there? Remove them and solve the question.....BUT if manufacturing (or whoever) complains that they need them to "make sure the quality is good" then calibrate the items.

Of course, I always recommend using an accredited provider and getting the accredited cal.....then you know what you actually have.

Hope this helps. I imagine I will make some folks uncomfortable with my comments above.

Hershal