Metrology Monday! #60 Calibrating an Instrument to Worse than Manufacturer’s Specifications
Since last week I touched on calibrating a product to better than manufacturer’s specifications, I thought I should also spend time on calibrating an instrument to worse than manufacturer’s specifications.
First off, is this possible to do?? Yes, it is certainly possible to perform a calibration that may be less stringent than the published specifications for the product.? It is also possible to not test specific ranges or functions on a product.? Is it ever appropriate to do so?? Yes, the owners of test and measurement equipment may perform this process to optimize the time and effort that it takes for a calibration to be fit for purpose in their given application.? How do we distinguish this from a “standard” calibration?? Usually, calibration organizations identify instruments that are calibrated in this manner as a “limited calibration”, as shown in the picture below.
Here is what the Fluke Quality System document for Measurement Management states about limited calibrations.
“A ‘Limited Calibration’ status label shall be affixed to items of M&TE when the calibration includes limitations on use. Limitations for use includes degradation or enhancement of performance specifications, special handling, characterizations, system calibration, or any other deviation from a normal calibration in terms of regular use, functions, ranges and specifications. When an item of M&TE has been assigned a Limited Calibration status, a description of, or a reference to, the limitation shall be applied to the item of M&TE.”
While most organizations only define limited calibration as calibrating to less than or worse than manufacturer’s specifications Fluke decided to use the limited calibration label for either a degradation or an enhancement, meaning that this could be a special calibration or characterization as we discussed in post #59.
I think a more common definition for limited calibration is one that is described in the White paper, “Limited calibration…it’s not a dirty word!” by Philip Mistretta.
领英推荐
“A limited calibration is a unit that has NOT been fully calibrated to the OEM’s published specifications.”
His article is very good and is available at the Transcat website or in the 2008 NCSLI Conference Proceedings, and my thoughts align very well to his paper.
There are many valid reasons for performing a limited calibration on an instrument.? One reason could be that an instrument indicates that it has an operational failure for a function which requires repair, but the customer does not use that function and does not want or need to spend the money on the repair.
Another reason that I have observed is that someone in an organization has purchased an instrument that is much more accurate that the customer needs.? This could be because someone had extra budget when it was time to purchase instrument or many other reasons.? I recall working with an automotive manufacturer that needed to test the inductance of components in an anti-lock braking system.? The tolerance of the components being tested was about 5%.? The company purchased a very accurate LCR (Inductance, Capacitance, Resistance) meter that had a basic accuracy around 0.01%.? The cost of doing a calibration to manufacturer’s specifications for the product was well over $1,000, which was causing angst for the customer.? After some conversation with the customer, we were able to develop a limited calibration for the product that only calibrated the inductance range, and only to 1% (about 5 times better than the tolerance of the components they were testing) and the cost to do this was about $150.? The customer was happy as we were able to save both dollars of calibration cost and significant downtime required to remove the LCR meter from the customer site and perform the calibration in a proper laboratory environment with high quality reference standards.
Limited calibrations need to have good documentation of the justification and good visual management so that end-users can clearly see the calibration limitation.? It is very important that end-users can see the limitation so they don’t move the instrument to an area that would require a full calibration of the product.?
Oftentimes, in-house calibration laboratories need to justify their existence to senior management.? Effective use of limited calibrations can optimize the organization’s calibration costs by ensuring that calibrations are fit for purpose.? ?Showing senior leadership the planned use of limited calibrations demonstrates a responsible way of managing costs, which can also be quantified.? #MetrologyMonday #FlukeMetrology
?
Senior Principal Metrology/Quality Compliance Engineer at Northrop Grumman / NCSLI Baltimore Section Coordinator
8 个月Great article! It is certainly a very valuable process for internal calibration providers. Between reduced costs of service time and the standards needed for reduced calibration requirements, it can save an organization a lot of service costs. However, it's a little murkier when it's an external customer or when an organization's requirements differ for similar manufacturer models. One important piece to note is the documenting of the communication between the end user and the calibration provider. Limited calibration of a device may different from end user to end user so the process of contract review comes into play. Aside from the compliance piece, it's useful to have the request clearly documented in the equipment archives for the all-important lookback that eventually occurs. Whether it's internal or external, turnover occurs and knowledge transfer can often turn into a game of telephone when it does!
Paladin and Beard of Metrology Knowledge BOMK
8 个月Jeff, I liked it. Only not everyone may know of Fluke's policy. In my past experiences we had "Limited" and "Special" cal labels. We limited for exclusion of a function and to widen specs due to uncertainty. Special was specified requirements and enhanced capabilities. That might be a discussion, Use of Labels and Coefficients charts.
Let's Make a World where We Win Together
8 个月Hi Jeff. I have a question about interpreting linearity and absolute errors. Let's say we have 2 calibration power meters that have a consistent linearity error but a sizable absolute error (let's say -.2 dBm). Does that tell us that the problem is the calibration power meters being out of cal rather than the (in this case, the attenuated) source?
Manager, Calibration @ Daikin Malaysia
8 个月Thanks for sharing. Where a CAB, Conformity Assessment Body is responsible to ensure the UUT conforms with the specification (generally refer to manufacturer 's specification), what there need to do extra when limited calibration is conducted?
Founder @ METROTEKA calibration lab and LORISQ lab equipment management app. Current president of the Croatian Metrology Society. Generally, I like to improve things.
8 个月Jeff Gust you should publish this as a book. I want to use it as a crash course in sensible metrology for my new employees...