08:26 May-20-1998 Rolf Diederichs Director, Editor, Publisher, Internet, PHP MySQL NDT.net, Germany, Joined Nov 1998 602
NDT Newsgroup topic: Manual UT instrument "calibration" frequency
BACKGROUND: In the currrent NDTnet issue May 1998 we summarize the new European Standard 'EN 126681-1 Part 1: Characterization and Verification of Ultrasonic Examination - Instruments.' http://www.ndt.net/article/0598/en126681/en126681.htm
This standard gives methods and acceptance criteria for verifying the electronic performance of ultrasonic instruments. with A-scan displays, employed for manual ultrasonic non-destructive examination. A specific group of these methods and acceptance criteria shall be used at least once every twelve months to verify the fitness for purpose of an ultrasonic instrument during it's lifetime. Other two parts of this standard consists of Probes (2) and Combined Equipment (3).
DISCUSSION: A discussion on this topic have been hold on the NDT newsgroup. Find below the intire message followed by its 3 replies.
One main question was: Is it really necessary to return the instruments for calibration every year? According to EN 126681-1 (PART 1 !) it is necessary to return the instrument to the manufactorer (or his agent). However, notice also the PART 3 'Combined Equipment'. The Part three of this standard gives methods and acceptance criteria for a complete system, with the probe, cable and ultrasonic instrument combined. Initially these shall be used to verify the compatibility of a probe, cable and ultrasonic instrument and to characterise the system's overall performance. On site these methods and acceptance criteria shall be used to routinely check the probe, cable and ultrasonic instrument during their lifetime.
Part 3 shall be performed at regular intervals on site!
You may continue the discussion on the NDT Newsgroup or on the UTonline Forum. I will try to keep both sites on prospective messages informed.
----------------------- Date: Tue, 12 May 1998 22:52:51 GMT From: email@example.com Subject: Manual UT instrument 'calibration' frequency
I am looking for some input on UT instrument 'calibration'. Most instruments come with a manufacturer's calibration that expires one year from the initial date. Is it really necessary to return the instruments for calibration every year? Please give me your opinions for the following 'types' of instruments: 1.Digital thickness gauges (with or without A-scan feature) 2.Digitally controlled CRT instruments (such as Staveley 136) 3.Digital flaw detectors (such as Panametrics Epoch series of K-B USN series)
All the instruments are set up and calibrated before and after each use. The flaw detectors have vertical, horizontal and gain control linearities performed before and after use.
Thanks for your input - you can respond to the address above or post to the news group.
----------------------- Date: Wed, 13 May 1998 13:37:54 BST From: John Alldred
It is worth noting that there are (at least) three different meanings/usages of the term 'calibration' (of any type of instrument):
(1) Adjustments to bring the indicated values to within the permitted tolerance of true values (this may be a 'factory only' adjustment);
(2) Verification (without adjustment) that the indicated values are within tolerance of the true values;
(3) Preparation of a table or graph to allow conversion from indicated values to true values.
Items (2) and (3) should be able to be performed by the user, although they may require reference 'standards' or special test equipment; item (2) is particularly relevant as it may be used to determine whether an instrument requires a 're-calibration' as per item (1).
There are situations where a user is /competent/ to perform calibration checks, but is not "accredited" to do so (or believe that to be the case); this sort of confusion is likely to arise when "QA schemes" are in operation which are more concerned with weight of paperwork than with actual practicalities ;-)
The one year expiration period for calibration certification indeed seems to be almost universal industry practice. We follow that practice here at Panametrics, as do most other manufacturers. People seem to expect it, if nothing else.
But to answer your question, it is entirely up to you how often you return instruments to the manufacturer for calibration recertification. This in turn will depend on how they're being used, and what code requirements you may have to follow in your job. The fact that you are verifying performance before use is an excellent practice -- I always recommend that to anyone using UT equipment in critical applications, since the fact that someone put a calibration sticker on a piece of equipment a few months ago doesn't actually prove that it's working properly (or more importantly, has been set up properly by the user) today.
Calibration certification really just shows that the instrument was working according to manufacturer's specifications on a particular day. If you're doing a critical test, it's up to you to make sure it's working properly *today*. If you're doing a non-critical test and you're satisfied that the instrument is working properly, you may not need up-to-date cal certification at all. We have some customers who return instruments for manufacturer's re-certification more than once a year because they are working to procedures that require it. We have other customers who re-certify equipment rarely if at all because their procedures do not require it.
Modern digital instruments tend to fail catastrophically if they go. Many have self-test protocols that will detect any problems on start-up or during operation. They are not subject to the sort of horizontal drift that could affect analog instruments, and vertical (gain) failures often tend to be obvious too. Most digital gages, scopes, and flaw detectors will essentially maintain calibration forever, barring a major (and usually obvious) failure.
Two potential problems that are *much* more common than calibration drift or failure in digital instruments are incorrect setup by the operator, or use of an inappropriate or failing transducer. These problems can be avoided through the implementation of daily (or at least regular) test procedures of the sort you're probably using.
--Tom Nelligan Senior Applications Engineer Panametrics, Inc.
-------------------------- Date: Wed, 13 May 1998 10:46:08 -0700 From: rocky
Calibration is a complex issue and in general each user must make up his own mind. In some cases calibration is regulated by standards of practice. The ASME Code for example requires calibration to ASTM E-317, every three months for Section III. You may wish to look at http://www.xdcr.com/calibration.html where I discuss calibration at greater length.
It is necessary to understand that the calibration you do before and after a shift is not calibration but an only a calibration check or reference standard calibration. The calibration furnished by the vendor is usually a certification with traceable standards meeting some standard like ASTM E-317 or the manufacturers own requirements. The reference standards run before a test do not meet these requirements and do not meet the requirements in standards for periodic calibration of measurement & test equipment.
The basic function of the more stringent requirements is to test for nonlinearity that might go undetected when running the reference standards. Such an event is unlikely, although I have seen it occasionally in 25 years in ultrasonics, so the interval to do the calibration is essentially a question of risk management. Just how much money or time are you willing to spend to guard against an unlikely event. Or how much risk your customers are willing to take if you want to meet particular standards that specify calibration.
My advise is to examine the standards you must meet and see what they say about calibration of measurement & test equipment. Even without such standards I would not ignore the manufacturers recommendation. If their cal expires in one year I would renew at that time and ask for another certification and a report on how much the instrument deviated from specifications. The latter is important because if there are no deviations from specifications that is reasonable justification to recalibrate on a longer interval, assuming no standards require an interval.
One caveat is necessary and that is what you will do if the instrument does deviate from specifications when recalibrated. That is when it does deviate what effect will that have on all the NDT done with that instrument since the last valid cal? Only you can answer that question and only you can determine if a deviation is significant. Because the instruments are very good compared to what is usually needed you may be able to easily justify not repeating all inspections done since the last calibration. The chances of an invalid cal of course increase as you lengthen the calibration interval. If you have to repeat two or three years of NDT done with that instrument it can be expensive.
Regards, Robert (Rocky) A. Day Second Sound Ultrasonic Transducers 904 Cortland Avenue San Francisco, CA 94110 - 5633 (415) 641-4947 Fax: (415) 641-5502