Quality Assurance Calibration Certifications

Let's answer the easy aspect of the question, "What's the difference?..."

Essentially, certification is the process of comparing a measurement device of unknown uncertainty (known as the 'device under test' or DUT) against another item of known uncertainty (the 'reference'). A calibration event takes place if an adjustment is made as a result of this comparison to bring the DUT closer to that of the reference device or standard being used. The procedures used to perform the work is laid down by the standards of the laboratory's accreditation body. The procedure is roughly as follows:

The laboratory technician takes measurements at several points over the desired range. If the display of the DUT is the same or within a certain accepted range of the reference it is being tested against, then the DUT is considered to be within specification. An 'As Found' test report (certificate) is produced and the device is considered to have been certified to reference using the standard used by the laboratory. By contrast, the device (DUT) is found to be outside the manufacturer's accepted range (known as 'out of tolerance') the device is adjusted or calibrated to the reference, and a separate set of 'As Left' readings are taken after the adjustment to document the state of the device both before and after the calibration event.

Which brings us to the second aspect of the question:
"Why should I bother?..."

The whole point of routinely calibrating a device is to prove its capability to maintain a known 'uncertainty of measurement', its ability to maintain this performance over a given time period, and therefore to be able to rely on the information it provides. OK - it's expensive and a hassle, but there are several reasons why calibrating your measurement instruments is a good idea.

In answering this, let's pose the question in reverse - what is the likely outcome if you don't? Some examples of the issues that might adversely affect you as a result of a lack of calibration include:

Getting unfairly booked because the local law enforcement didn't calibrate their radar detector.

 

Paying an ever increasing gas bill because the utility company didn't calibrate your meter

The regular calibration of a gas meter ensures you only pay for what you use. You wouldn't want to cop a fine if you weren't actually speeding. If we think about it, we all appreciate the confidence that comes with certainty in measurements. However in the same way that many things in life can't be 100% certain, neither can instruments. So the entire rationale behind calibration is to minimise the uncertainty of measurement, allowing buyers and sellers to have confidence that they are getting the weight, size, volume, etc. that they expect.

In order to achieve standards relating to quality, various bodies exist in every country of the world, aligned to the national standards of that country, which in turn are aligned to accepted international standards of metrology. The international standard that governs the competence of calibration and testing laboratories is ISO/IEC 17025General requirements for the competence of testing and calibration laboratories.

Setting up a calibration laboratory is a major investment in many respects, in equipment, manpower, training and accreditation just to name a few. Australian commercial laboratories have to be viable based on demand for re-calibration of equipment, whereas the overseas manufacturer has the uncertainty of the actual equipment as their benchmark. In addition, responsible manufacturers have an increased tolerance in order to deliver on their warranties regarding the uncertainty of the device. For these reasons alone, a manufacturer's certificate represents extremely good value, and often these manufacturer certificates are offered free, or at far lower cost and much faster than having the item certified after it arrives in Australia. But how do you know if it is acceptable for use in Australia?

"I MUST have a NATA certificate..."is a phrase we often hear when it comes to people purchasing equipment. In Australia, the organisation responsible to the government for the assessment and application of the relevant standards is the National Association of Testing Authorities (NATA). However NATA is not the standard... they don't issue test reports, and anyone insisting that a report must be "to NATA standard" is well off the mark! Instead, NATA is an accreditation body - they assess the laboratories to ensure the equipment, procedures & personnel meet the required standard (known as a conformity assessment).

On the world stage, an approximation would be that NATA is to laboratory facilities what the Australian dollar is to money. Try to purchase a hotdog in New York with an Aussie $10 note, and you won't get very far, yet there's nothing wrong with the money or its value. However if you apply the international standards and principles of foreign exchange to your currency, you can buy what you want internationally. In the same way, there are standards and principals that govern international laboratory competence, and the good news here, is that you don't even have to exchange the certificate. So how do you tell if the certificate you've got is acceptable in Australia?

• Does the test report state that it has been issued to the international standard, ISO/IEC 17025? If it does, you are pretty much good to go, however if you want to validate this further...
• Who does the laboratory rely on for their accreditation? Just as laboratories in Australia are accredited to NATA and bear this endorsement, a logo should be prominently displayed to indicate the body that accredited the laboratory.

With this information, head to the website of the International Laboratory Accreditation Cooperation (www.ilac.org) of which NATA is an accredited body and MRA partner. Provided the manufacturer's laboratory is currently accredited by one of the listed ILAC partners, the certificate you have is fully acceptable in the same way that one produced by a local NATA accredited facility is. SeeNATA's websitefor further details on their MRA program and accreditation to ILAC and APLAC.

What we provide:Transtek offer advice and supply manufacturer certificates for items where the laboratory is suitably accredited. If not, then thanks to the support our partner laboratories, we are also able to provide NATA certificates for any items you purchase from us at no extra cost than what you would have paid to the same laboratory. Of course, if you'd rather organise this yourself, we're happy to advise you on where to send your equipment (see Recertification of Equipment below)

Good equipment is expensive - there's no getting away from that. In the world of measurement devices, the higher the accuracy, the higher the price. However what you are really buying is a lower uncertainty (or conversely greater certainty) of measurement.

The comment we often hear is, "..but I don't need that." While that may be true some of the time, due to the relatively low demands of the application, more often it is a comment borne out of misunderstanding.

Firstly, we often find that users aren't thinking about the cumulative nature of uncertainty. For example, if you have more than one instrument in a process loop whose measurements are inter-related, it's the cumulative effect of the uncertainty of all the instrument readings taken together that affect the overall result.

Secondly, buying calibration equipment is all about investment in quality and your time, and not just the cost of the asset. Consider the following:

Uncertainty or accuracy statements made by manufacturers are an expression of the worst performance you can expect from the instrument. This means that the actual instrument uncertainty must be better by some degree of tolerance as in making the statement (or claim) the manufacturer must allow for a large number of possible variables over the lifetime of the device, even given the ability of a technician to adjust the device to try and compensate for these changes. If the manufacturers statement is too close to the best possible performance of the instrument, it'll soon be headed either for service or the bin (at your cost) or their warranty department!

The lower you can drive the uncertainty of your measurements, the longer the interval between the time you should have to make them. Part of the rationale behind routine calibration has to be the benefit of compiling data on the devices you are testing, and then making rational assessments regarding the need for re-calibration events to be shorter or longer given past performance. Doing calibrations costs time - you might as well do the best job you can with a decent calibrator.

A good rule of thumb is to select a calibrator that is a minimum of 3 to 5 times better than the lowest uncertainty of the devices you are going to test. In selecting your device, watch out for manufacturer 'specmanship' - percent of full scale is not the same as percent of reading, uncertainties based on 30 days are not worth the paper they are printed on, and total uncertainty includes all of the relevant factors: hysteresis, linearity, repeatability and resolution, not just the latter.

What we provide: Transtek provide unbiased assistance to our customers on the pros and cons of all the major brands and types of equipment. We're very selective in who we represent and while we back our suppliers, we'll always give you an honest answer if there's something better to get the job done!

The questions we get asked by those inquiring about recalibration are either "How often does it need to be recalibrated...", or"How often should I get my device recalibrated...". Both questions imply that it is up to someone else to dictate the answer, when in actual fact it should really be determined by:

• How often you use it?
• What are the conditions in which you use it?
• How much do you rely upon its performance?
• What are the manufacturer's recommended guidelines?

Obviously, it goes without saying, that if your company has a specific mandate about re-certification intervals, then those are the minimum standards you should adopt. However mandates were never designed as a replacement for critical thought, and should always be subject to review, given logical, evidence-based assessment. Clearly a device that is kept in a laboratory isn't subjected to the same wear-and-tear as a field device. Conversely, a device that is used to perform a high number of critical measurements should be subjected to a much shorter period of review than one that is only used sporadically. Time is money, and you don't ever want to be in the situation of having re-check work that you did with a calibrator that was found to be widely out of tolerance on review.

For this reason we highly recommend instituting traceable site standards for clients with a large inventory of measurement devices. These local standards allow periodic assessments of routine test equipment without them leaving site. Not only does this substantially lower the risk of out of tolerance equipment being used, but it also drives a better quality system through those engaged with the site recertification equipment, and dramatically reduces the cost and downtime associated with equipment being sent off site for calibration.

Finally, while on the subject of calibration laboratories, you should be aware of the following:

• There is no requirement for a NATA accredited laboratory to inform you if you send them equipment that is of a higher uncertainty of measurement (accuracy) than their laboratory accreditation. It is up to you to ensure that the lab you select can certify your equipment to the manufacturers stated uncertainty for each device. While this behaviour would no doubt be frowned upon by an assessor, there is nothing to stop a laboratory issuing a certificate for a 0.005% reading dead weight tester using a 0.01% full scale standard test gauge, even though the resulting certificate would be worthless in terms of uncertainty.
• Many laboratories also sell equipment. We've experienced the pitfall's here first-hand. While equipment certainly has a life, you need to be wary when a laboratory fails a piece of equipment and then wants to sell you a replacement, especially when the item you sent in isn't the brand they sell!
• There is no such thing as a 'NATA traceable' certificate. A NATA endorsed certificate can only be issued by an accredited laboratory, and indicates that the device(s) used to complete the test, and the person responsible for carrying out the procedures, are traceable to and accredited to NATA, and comply with the standards of ISO/IEC 17025. A traceable certificate is issued by a non-accredited person using a piece of equipment that has been certified by a NATA laboratory.

What we provide: For any item we sell, Transtek offer the same service to re-certify the device on either the manfacturers' recommended certification interval, or another of your choosing. We provide email reminders, which are sent to the original purchaser of the item or the person who last arranged the certification service at least one month prior to the item requiring re-certification. We'll help you take the guess work out of certifying the device, either by recommending the most appropriate calibration facility to ensure that all certificates are issued in line with the manufacturer's accuracy specifications for your piece of equipment, or by taking care of the legwork for you - your choice!

Hopefully the above has been helpful in answering some of your questions in this area. If you'd like further assistance, please don't hesitate to contact us.