Terminology
On this page we offer explanations for some of
the arcane language that is often used in 'Calibration'. You may
wish to remind yourself of the reasons that calibration is important.
Traceability
The simple concept behind calibration is that
measuring equipment should be tested against a standard of higher
accuracy. For any parameter/range we should be able to illustrate
this type of hierarchical relationship:-
National Standard.......................... Accurate to 0.002%
Calibration Laboratory........................................0.01
%
Company "Master" Item......................................0.07
%
Company Production Equipment........................1.0 %
Produced Product.............................................10.0
%
Of course, these calibrations need to be done on a planned, periodic
basis with evidence of the comparison results maintained. This record
must include identification of the specific standards used (which
must be within their assigned calibration interval) and some means
of knowing the method used and other test conditions. By examining
these records, it should be possible to demonstrate an unbroken chain
of comparisons that ends at the agency responsible for maintaining
and developing a country's measurement standards (now generically
known as a national metrology institute). This demonstrable
linkage to national standards, with known accuracy, represents 'traceability'.
In fact, it doesn't stop there because these laboratories routinely
undertake international comparisons which help to establish worldwide
consensus on the accepted value of the fundamental measurement units
- without which, there would be little confidence in, for instance,
successfully mating a 10 mm screw manufactured in one country with
a 10 mm nut produced in another!
Periodicity
This refers to the calibration interval
assigned to an item of equipment -- examples could be 3 months or
perhaps 2 years. An alternative way of expressing this is the calibration
cycle, usually how many calibrations are required per year.
Equipment used in any metrological situation must have known accuracy;
that is, a specification assigned by the manufacturer or by the
user. Since the performance of pretty much everything on Earth degrades
with time, or use (or potential abuse), the expected accuracy must
relate to a given period.
Characterization
Calibration can be just determination of attributes or
characteristics or include realignment to nominal (within
certain limits, or specs.). Characterization is most often associated
with parameters which are not specified or for those which, by the
use of measurement data, enable the user to enhance his own measurement
accuracy. Such data must be accompanied with uncertainty information
to be valuable to the user wishing to qualify his own test accuracy.
Uncertainty
Despite what some suppliers may often claim, no measurement can
be guaranteed to be perfect! An uncertainty is a figure of merit
associated with the actual measured value; the boundary limits within
which the 'true' value lies. Contributors to this "potential
for inaccuracy" include the performance of the equipment used
to make the measurement, the test process or technique itself and
environmental effects. Additional imprecision may result from behavior
of the phenomenon or item being measured. A skilled metrologist
will assess and combine these various components in an uncertainty
budget. To prove that a product complies with specification (or
doesn't), the uncertainty must be less than the unknown's specification.
Test Accuracy Ratio
It's generally considered good practice to use test equipment and
techniques whose combined uncertainty is 3 to 10 times smaller than
the specification of the unit under test -- see concept of traceability
-- which represents the test accuracy ratio: i.e. TAR = Spec/Unc.
Of course, the higher the TAR the better, but higher performance
test gear or extended test times for averaging, for instance, costs
more and the pursuit of an excessively high TAR is cautioned against.
Guardband
Particularly employed when TARs are low, this is a safety margin
having the purpose of tightening an acceptance (pass) limit when
testing a specified product. The guardband limit might, simplistically,
be set at a point equal to the specification minus the uncertainty
but is often 'tuned' to recreate the same confidence that would
result from using a 4:1 TAR with the acceptance point set at the
spec. limits. Guardbands are also employed in manufacturing where
routine testing may only be a subset of the product's full (customer)
spec. in extent or environment, yet assurance of compliant shipments
is desired.
Associated Topics
Certification, registration, approval, accreditation, assessment,
compliance, conformance -- they all mean the same thing, right?
Wrong! There are some particular words that should be used
in relation to conformity assessment.
Thumb through any instrument specification and you are presented
with a whole host of technical terms describing the product's capability.
There are some basic ones which are often misunderstood, though
-- accuracy, precision, resolution and sensitivity
spring to mind -- so we've devoted a separate page to explain specifications.
After working so hard on your study of the "Language of Measurement",
you deserve a period of relaxation while you consider the language
itself !
|
Search | Site Map | Contact Us | Login | Home 
