The Standard Fall 03

8/8/2019 The Standard Fall 03

1/32

The StandardFall 2003 Page 1The StandardVol. 17, Issue 3 The Newsletter of the Measurement Quality Division, American Society for Quality Fall 2003

Chai r s Co lum n

Message from the Chairman (2003-2004)

Dilip Shah

I returned from the NCSLI conference in Tampa inAugust and wanted to update you on the session thatMQD presented and other events that transpired there.

The Certified Calibration Technician session at theNCSLI conference in Tampa attracted approximately300 session participants, with Chris Grachanen, SallyHarthun (ASQ HQ) and I presenting and Jay Bucherhosting the session. Following the session presenta-tion, a panel discussion ensued where Hershal Brewer

joined the session presenters to answer questionsabout the CCT exam.

The Metrology Handbook group with the exceptionof Dave Brown was also able to meet in Tampa and talkface to face about the project.

During the Tampa meetings, we also started work

on getting the Joseph D. Simmons Memorial Scholar-ship Award off the ground with Norm Belecki leadingthat effort. The Joe Simmons Scholarship is a jointeffort between the MQD, NCSLI and the MeasurementScience Conference. After the conference, a telecon-ference meeting took place with representatives fromall three organizations in September, 2003. TerrelleWilson has agreed to be the MQD representative forthe Joe Simmons Scholarship committee.

The conferences give us a venue to meet in personwhere, normally, most of the time our business is

conducted via teleconference calls. It was good to

meet old friends and make new ones.Speaking of conferences, our last scheduled MQD

conference was to be on September 13-14, 2001.Unfortunately, the tragic events of September 11, 2001forced us to cancel that event. Plans are underway tohave an MQD conference in September, 2004. Watchthis space for further announcement on the time andvenue. We look forward to organizing a great programof speakers to enhance your knowledge and hope thatall of you can attend.

As we come to the end of the year, we embark onmany holidays observed by different faiths and tradi-

tions. Have a safe and happy holiday!

E = mc3 Solutions197 Great Oaks Trail #130Wadsworth, OH 44281-8215Voice: 330-328-4400Fax: 330-336-3974E-Mail: [email protected]

IN THIS ISSUEChairs Column................................................... 1Advertising & Article Info.................................. 2Chris Grachanen-Certification.......................... 3Division Membership Trend.............................. 5Phil Painchaud-The Learning Curve..................6Phil Stein-The Contrarian Metrologist...............8Editors Note..................................................... 10Division Conference News.............................. 10Simmons Scholarship ..................................... 10William McCullough-10012.............................. 11Division Finances ............................................ 13

Jay Bucher-Calibration.................................... 14Jack Gale-Repair .............................................. 26Upcoming Events............................................. 27MQD August Meeting Minutes..........................28MQD October Meeting Minutes .......................29MQD Officers List............................................. 30Regional Councilors List................................. 31

IN THE WINTER ISSUEMetrology Handbook Update


2/32

The StandardFall 2003 Page 2 The StandardVol. 17, Issue 3 The Newsletter of the Measurement Quality Division, American Society for Quality Fall 2003

PublicationStaff

Managing Editor

Mark SchoenleinP.O. Box 206Perrysburg, OH 43552Voice: 419-247-7285Fax: 419-247-8770E-Mail: mark.schoenlein@

us.o-i.com

Advertising2003 Rates for a single publication:

Business card size .....................$201/4 page .....................................$351/2 page .....................................$70Full page ..................................$120

A 15% discount will be applied for multi-edition ads.

Ads must be formatted in MS Word oras a TIF file.Advertising must be clearly identified asan ad and should relate to the field ofmeasurement quality.Ads must not imply endorsement by theMeasurement Quality Division or ASQ.

Letters to the Editor

The STANDARD welcomes lettersfrom members and subscribers. Lettersshould clearly state whether the authoris expressing opinion or presentingfacts with supporting information.Commendation, encouragement,constructive critique, suggestions, andalternative approaches are accepted.If the content is more than 200 words,we may delete portions to hold thatlimit. We reserve the right to edit lettersand papers.

Publication InformationThe STANDARD is published quarterlyby the Measurement Quality Division ofASQ; deadlines are March 15, June 15,September 15 and January 15.Text information intended forpublication can be sent via electronicmail or through postal mail on 3 1/2"diskette in Microsoft Word saved inRich Text Format (RTF). If it is notfeasible to send text in electronic form,clean printed text can be submitted.

Graphics or illustratios must be sent ina TIF file format.Photographs of MQD activities arealways appreciated.Publication of articles, product releases,advertisements or technical informationdoes not imply endorsement by TheMeasurement Quality Division of ASQ.While The STANDARD makes everyeffort to ensure the accuracy of articles,the publication disclaims responsibilityfor statements of fact or opinion madeby the authors or other contributors.Material from The STANDARD may notbe reproduced without permission ofASQ. Copyrights in the United Statesand all other countries are reserved.

2003 ASQ, MQD. All rights reserved.

Website Information

The Measurement Quality Divisionhomepage can be found on the internetat www.measurementquality.org. Pasteditions of the STANDARD in PDFformat are available there.


3/32


4/32

The StandardFall 2003 Page 4

Uniqueness within the quality profession Applicability within industry Public domain breadth of the body of knowledge Consistency with ASQs objectives, policies and procedures Training availability Support and commitment Market needs

Market justification was proven via an environmental market survey by a major calibration service provider andconservative inferences and from the U.S. Department of Labor, Bureau of Labor Statistics. In addition, ASQsMarket Research Department conducted a survey which yielded the following statistics when calibrationpractitioners were posed with the following questions:

Yes No Dont Know

I believe there is an industry-wide need to develop a CCT program 84.9% 6.2% 8.8%

I, or somebody in my organization/div./dept., would be interested in taking 82.1% 8.8% 9.1%

a CCT exam

The final proposal was submitted to ASQs certification board and a formal presentation given at ASQs 2002Quality Congress Conference in Denver, Colo. The presentation, given to the ASQ certification board, lasted alittle over 10 minutes. It was followed by a question-and-answer session lasting approximately 5-7 minutes.Immediately after the question-and-answer session a motion was made to give approval to the creation of the CCT

program. The motion passed unanimously! Smiles, hugs and congratulations were in abundance. It was rightabout this time that a member of ASQs board of directors seated at the back of the room related that the boardwas of the opinion that it should not be necessary to present the CCT program proposal to the board. Instead anagenda change motion was going to be presented to the board which would, if approved, forego the formal CCTpresentation and agree to the creation of the CCT program. The boards opinion was based on what was describedas the thorough, specific, and well-written CCT program proposal coupled with the approval of ASQscertification board and ASQs professional development council (PDC). About an hour later, ASQs board ofdirectors voted unanimously to accept the agenda change motion, thereby formally creating the CCT program.

The next big step in CCT program development was to conduct a formal job analysis survey. The main purposeof the survey is to help determine the body of knowledge (BOK) for use in the CCT program. The CCT BOKconsists of information deemed important by those actively practicing in the field. Survey topics are generallyweighted by level of importance and frequency of engagement. The survey itself is the aggregate compilation

of multiple telephone interviews, a two-day workshop, a practitioner beta test and countless review hours. Theseefforts were conducted under the auspices of a contracted psychometrician to ensure subject matter centeredon pertinent calibration / metrology information and that questions made sense from the perspective of the focusaudience. Participation for the written survey was the highest ever for a new ASQ certification, reflecting theenthusiasm of the target audience. The psychometrician compiled, analyzed and summarized all survey data intofinal written reports, identifying both topic areas and associated weighting in terms of importance and frequency.This report, coupled with survey participant demographic information in terms of location, industry involvement,experience, size of facility, etc., was the centerpiece of another workshop known as the Test Specification (TS)workshop. The TS workshop, attended by 14 metrology professionals, analyzed the survey knowledge areaspresented in a summary report and made recommendations for inclusion into the BOK. The results of the TSworkshop were compiled and submitted to ASQs certification board for final approval. The certification boardunanimously passed the CCT BOK. The CCT BOK is divided into the following six major topic areas:


5/32


I. General Metrology

II. Measurement Systems

III. Calibration Systems

IV.Applied Mathematics And Statistics

V. Quality Systems And Standards

VI.Uncertainty

The finalized BOK serves as the basis for developing items for the CCT exam. Item development for the CCTprogram involved an Item Writing workshop, an Item Review workshop and an Exam Review workshop. After theExam Review workshop, the CCT pilot exam was finalized and administered. After the CCT pilot exam was givenand the results determined, a Cut-Score Study was conducted to determine the threshold necessary for passingthe exam. The results of the Cut-Score Study gave the June, 2003, CCT candidates approximately a 70% passrate. You can get the information about the CCT program, exam dates, reference listing, etc., at http:// www.asq.org/cert/types/cct/ and at http://www.measurementquality.org/ .

I wish to express my sincere appreciation to the many volunteers who have contributed to the CCT programdevelopment and to the many individuals who have developed or are currently developing materials for aidingtechnicians in preparing for the exam. Without your passion the CCT program would not have gotten beyond thewishful thinking stage. I especially want to thank the dedicated professional staff members at ASQ headquarterswho were our guides throughout the certification process (kudos to all of you).

For contact information see the officers list on page 30.

Monthly Measurement Quality Division Regular Membership Totals

N u m

b e r o

f M e m

b e r s

3500

3000

2500

2000

1500

1000

500

0

Jan-03 Feb-03 Mar-03 Apr-03 May-03 Jun-03 Jul-03 Aug-03 Sep-03 Oct-03 Nov-03 Dec-03

3037 3086 3095 3135 3163 3146 2452 2553 2689 2736 2753Members

Months


6/32


The Lea rn ing Curve

This is the thirty-fifth contiguous monologue on thegeneral subject of metrology education. Occasionallythis column deviates from its chartered subject andsoliloquizes on subjects in, around, over, under, beside,behind, in front of, and sometimes even near to itschartered objective. As usual, our format will be that of an open letter to the Editor-in-Chief of The Standard.

Phil Painchaud

Dear Boss:

Before we get underway, who among you cananswer this question that I got out of the Los AngelesTimes, Saturday, January 25, 2003, comic stripFRAZZ, by Jeff Mallet:

How come Canada uses a temperature scalewhere their zero is warmer than our zero when theirgeneral climate is colder than our climate?

See what you can do with that one. If your answeris enlightening enough, I might even try to get ourEditor to print it in a forthcoming issue.

Now the news about our four-year baccalaureateprogram in measurement science at the CaliforniaState University Dominguez Hills--its good. Our firstclass finished their first year in June, and the secondyear started in September. I unloaded about 300books and other documents and approximately thesame number of journals that have been taking upspace here in my cramped office. And we also had acouple of tons of lab equipment at Yuba College inMarysville, Calif., we physically transferred toDominguez Hills.

This educational resource has been a long timecoming; but it is now here, so I dont want ever again to

have someone come crying to me that there is noplace I can go to get an education in metrology. Theplace is the California State University at DominguezHills in the city of Carson, Calif. If you are unable totravel to California and live there for the four yearsrequired, you can get the same degree on-line; but itmay take a bit longer. Or, if you can get enoughinterested people together, the program is willing tocome to you for on-site sessions. For information onany of these options, contact Dr. Eugene Watson or hisboss Dean Margaret Gordon, California State Univer-sity-Dominguez Hills, 1000 Victoria Street, Carson, CA90747-0001, telephone: (310) 243-3737, e-mail:[email protected].

Now into my periodic homily. I am regularly chal-lenged by would-be metrologists who claim that theyare searching for an education in metrology but with a

Why do I have to take all of that cr when all I want todo is calibrate voltmeters (or gauge blocks, or micro-wave assemblies, or you name your own favoritedevices)? I answer, If your sole objective is tocalibrate something, you dont want to be a metrologist;you want to be a calibrator or maybe a metrologytechnician (there is a difference). If that is so, forgettrying to get educated as a metrologist and go some-where you can be trained as a calibrator (or metrologytechnician). There are several good institutions doing

just that. There is nothing wrong with or demeaningabout being a top notch calibrator or metrology techni-cian. A good one is worth their weight in gold. I wouldrather have a first-rate calibrator working with me anyday than a half-educated, self-styled metrologist.

Then I sometimes ask, What do you think that aneducation is supposed to do for you anyway?generally get an answer to the effect, It is supposed toteach me everything there is to know about somesubject I am interested in. Boy-oh-boy, are you goingto be disappointed! Nobody, but nobody, can knoweverything about anything; the human brain just isntbuilt that way. Theoretically, the ideal educationalcurriculum is supposed to be designed to teachyou to be able to recognize what you do not know;

to know how and where to find that information;and to know how to interpret and apply the informa-tion applicable to your problem. It is not supposedto just fill your head with facts and figures; that is whatreferences are for--to look up necessary data whenneeded. And dont forget that, despite the claims ofmany currently popular pundits, everything you mayneed is not on the Internet. You still need books andlibraries. May I suggest that a very valuable additionto your curriculum would be an elementary course inlibrary science.

It is pretty obvious that, if you are forced to look up

I


7/32


a great deal of material on a particular subject, andshould you retain the information you have looked up,you will eventually get to know quite a bit about thesubject, i.e., become an authority on the subject. In thecourse of looking up information and interpreting itcompatible with your problem of interest, you will needthe proper tools. I have preached many times in earliereditions of this column that you need as these tools asolid comprehension of the sciences of chemistry andphysics as well as a considerably more than just nod-ding acquaintance with most of the branches of math-ematics. Remember, metrology is not a technology tobe taught as a trade or a craft; it is a fundamental basicscience, one upon which all other science depends. Ifyou are a metrologist, a true metrologist, you aretherefore a scientist.

But why chemistry, why physics? I am often asked.

Elementary, and I reply with another question. Canyou measure anything without knowing something aboutthe nature of whatever it is you are measuring? Ofcourse you cannot. There are only two things that canbe measured, matter and energy. Chemistry is thescience of matter, and physics is the science of energy.And you need mathematics as the manipulartory toolfor modifying the measurement data. But also to beconsidered are the media and the environment in whichyou are attempting that measurement. Everything youmeasure is in some media and is in some environment.

For example, let us say that you want to preciselymeasure the voltage drop across a resistor of a lowohmic value with a relatively high current flowing throughit. First you need to determine if the current flowing istruly pure direct current, or does it contain constituentsfrom the frequency and/or time domains (AC or pulses).Here physics is necessary. Next you need to study thephysical shape of the resistor to determine the probableconfiguration and magnitudes of any fields that will begenerated around that device under test and the effectof these fields on the current paths through the device.Again physics--field theory is necessary. Now dont tryto stop there. We must also consider the matter that theresistor is composed of--probably some metallic alloy;

its specific resistivity; its temperature coefficient ofresistivity; its voltage and current coefficienties of resis-tance (Yes, those are rare but very real factors that maypeople overlook); and, if you are measuring a devicecomposed of a semi-conductor material, you may haveto consider its photo coefficient of resistivity. All of thesefactors reside in the domain of some branch of chemis-try, such as metallurgy or trace element chemistry.

And, we havent even gotten into the environmentalfactors as yet. How about the ambient temperature inyour laboratory (Physics)? Or the relative humidity(also Physics)? Or the material used in your tempera-

ture control bath (Chemistry)? How about the fieldstrengths, vectors, and spectral composition of theradiated fields in your laboratory (more Physics)? Doyou know the vectorial gravitational factors existing inyour laboratory (Geodesy)? I could go on ad nauseaum,but I think that you have gotten the point by now.

Then after you know all of these factors, and possi-bly many more, you need the appropriate mathematicsto develop the proper algorithms to combine them inproportion to their relative importance and to eventu-ally determine the Most Probable Value of yourmeasurement.

Can the typical trained calibrator do all of this? Mostprobably not. It takes a metrologist educated in thesciences to develop such a procedure. A well-educated metrology engineer should be able to de-velop the procedure and the necessary hardware fromthe criteria developed by the metrologist, and a well-trained calibrator (or metrology technician) should beable to follow those procedures using the specifiedhardware. If the individual is particularly well trainedand experienced, they might even be able to assist themetrologist or the metrology engineer in their tasks.

I know that this might raise the hackles of some ofyou, and you might want to argue. Thats what I amhere for--to carry on a dialog with our readers--argu-mentative or otherwise.

I am at the same old station where I have been formany years. In case you are a newcomer to thiscolumn, I am at:

Phil Painchaud1110 West Dorothy DriveBrea, CA. 92821-2017Voice: 714-529-6604Fax: 714-529-1109E-Mail: [email protected]

or [email protected]


8/32


The Cont ra r i anMet ro log i s t

Philip Stein

MEASUREMENT CAPABILITY

Measurements are made all the time-- in processes,of how well work is progressing, and in daily lives for

commerce and other purposes. Most of the time, thereis a trust that these measurements are adequate for thetask. Being adequate means having a small enoughuncertainty (enough accuracy, resolution, precision) tosupport that for which the measurement is being used.

Measurements are not always adequate, though.Some scientific and industrial processes can push thelimits of what can be done at all. Other times, theequipment or algorithms being used are not up to the

job. Often, the practitioners have simply not givenenough thought--or any thought--to the measurementrequirements.

Evaluating the quality of a measurement must al-ways be done in the context of the task--what are thecharacteristics of the quantity being measured, andwhat is required of the measurement in order for it to beadequate?

QUANTIFYING THE QUALITY OF A MEASUREMENT

This property is known as measurement capability.Just as the quality profession has defined measures ofprocess capability--indices such as Cp and Cpk--so

can we define capability indices for measurements.

A typical process capability measure will comparethe actual spread of the process (often expressed as astandard deviation) with the allowed spread of the

process (usually calculated as the total range of thespecification--the voice of the customer). If, for ex-ample, the allowed spread is plus-or-minus six timesthe actual spread, this represents the familiar sixsigma situation.

In order to demonstrate that a process is capable,the actual process values must be measured; and thatmeasurement must in turn be narrower (more ca-pable), otherwise it wont be possible to see the spreadof the process--all you will see is the spread of themeasurement. Sometimes the measurement rangecant be reduced any further, and that limits the abilityto report high-sigma values for the process. The spreadof the process might be very narrow; but, if it cant bemeasured, it cant be reported.

If the process spread is divided by the measure-ment spread, it will yield a measure that could be calledthe measurement capability index. The intent is thesame as Cpk. If the measurement is narrower than theprocess (five or six times narrower would be nice),most of the reported variation will be due to the pro-cess; and thats a desirable situation. Another way oflooking at this is to consider how many differentiablemeasured values can fit into the spread of the process.

During a consulting assignment 10 years ago, Ievaluated a sterilizer and its temperature controls. Thenominal temperature was 35C, and the process limitswere 33C to 37C. The temperature controls and mea-surement system had a resolution (minimum report-able change) of one degree. Thus, there were only fivemeasurement values available between the limits (notfive values within the process spread, which wouldhave been better). When attempting to analyze thebehavior of the sterilizer, histograms and control chartsshowed extensive stratification of the data due to thisvery limited measurement capability. Analysis wasdifficult, and the answers that resulted were weak.

EXTRACTING THE DETAILS

The measurement capability index shows the over-all performance of the measurement system withoutany reference to separating the individual contributorsand quantifying their separate influences on that per-formance. If the capability is adequate or better, some-times its not necessary to know those details. Whenthe capability is limited or if improvement is desired,some effort is needed to decide where the excess


9/32


variation is coming from and to set priorities for dealingwith it.

The process for accomplishing this separation isoften called a measurement capability study. Repeatedmeasurements of the same item or of a group of itemsare performed under deliberately varying conditions(operator, gage or tool used, measurement environ-ment). While there are many schools of thought as toexactly how to carry out these studies, most are varia-tions on two well-known approaches: the range method(Gage R&R) and the ANOVA (Analysis of Variance)method.

Both approaches are based on the same principle.Repeat measurements are expected to generate simi-lar results (this doesnt work for destructive tests).When two identical measurements of the same itemget different results, we look at what was different in themeasurement conditions. By analyzing variations gen-erated by different setups, we can assign a separatefraction of the total variation to each influence.

The Gage R&R range method is most commonlyfound being used in the automotive industrys Mea-surement Systems Analysis Manual, part of the QS-9000 series of standards. Its a pencil-and-paper tech-nique that has been deliberately limited so that it can becompleted and analyzed without the help of a profes-sional statistician or a computer. This is philosophicallyvery similar to the X-bar R control chart for smallsubgroups. A simple form leads the practitioner through

the calculations to a final answer.Lately, though, the limitations of the method have

come under increased scrutiny and criticism. At theOctober Fall Technical Conference co-sponsored byASQs Statistics Division, I attended a session spon-sored by the Journal of Quality Technology and waspleased to hear Connie Borrer deliver the paper, AReview of Methods for Measurement Systems Ca-pability Analysis that she co-authored with Rick Burdickand Doug Montgomery. The full paper can be read inthe October, 2003, issue of the Journal.

The analysis in Burdick et al focuses on the math-

ematical limitations of the range method, but metrolo-gists have other concerns. Gage R&R is based on afixed model wherein any variation in the measurementresults are assigned either to the gage (measuring toolof any kind) or to the operator. Statistically this is knownas a two-parameter fixed-effects model. Thinking aboutthe realities of measurement, there are many othermodels that might better reflect the underlying metrol-ogy; and using the R&R model may lead to confusingor even completely wrong results.

For example, if two operators were part of the study,but for convenience one day shift and one night shift

operator were used, and if the lab temperature wascooler at night, the analysis couldnt tell the differencebetween a variation due to the operator and one due totemperature. Operators might be retrained or evendisciplined when the fault lay with poor laboratorytemperature control.

The ANOVA method allows for any model. Anycombination of variables can be chosen as candidatesto see to what extent they influence the measurementresults, and the analysis will be more likely to properlyseparate the total variation and assign the correctfraction to each influence. A computer is likely to makethe calculation a lot easier; but, once the method hasbeen learned, automated experimental design soft-ware can do a good job of imitating a consultingstatistician for this purpose. Gage R&R simply givesmisleading or confusing answers too much of the time

to be considered a good tool for this purpose.

P.G. Stein Consultants

400 Oak Street

Pennington, NJ 08534-3316

Voice: 609-737-9144

Cell: 908-672-1480

Fax: 609-737-9411

E-Mail: [email protected]

www.measurement.com

Sign up fo r t he CCT!

The registration deadline for the next ASQCCT (Certified Calibration Technician) examis April 9, 2004. The exam is scheduled forJune 5, 2004. For more info visit www.asq.organd click on the Certification tab or call ASQat 1-800-248-1946 and request item B1331.


10/32


Manag ing Ed i t o r s No t e

Mark Schoenlein

I would like to thank Bill, Jay, and Jack for theirresponse to our request for more papers. You will alsonote that we are now able to publish more of thedivision stats-this is a direct result of Dilips leadership.

I hope you will find this edition of the newslettersatisfying. Let me know what you think. My e-mailaddress is [email protected]

Div i s ion Confe ren c eN e w s

We pleased to announce that a division conferencefor the fall of 2004 is now in the planning stages. Thetentative dates are September 23-24, 2004. It will behosted at the Air Force METCAL facilities in Heath,Ohio (20 miles from Columbus).

Further information can be obtaining by contactingeither Dilip Shah or Hershal Brewer. Their contactinformation can be found on the officers list on page30.

Notice!

Sim m ons Sc ho l a r sh ipt o be Off e r ed

The Joseph D. Simmons Memorial ScholarshipAward will resume offering an annual scholarship to ahighly qualified student of metrology beginning in the20042005 school year. The scholarship has beenrevitalized under its constitution, with the first twomeetings of its Executive Board being held in Septem-ber and October. The Board began to formalize itsoperating procedures and is developing a process forassessing applicants. Norm Belecki, a NIST retiree,was elected Chair; and Mark Kaufman of the Measure-ment Science Conference and an employee of theNaval Surface Weapons Center, Corona, was pickedto head the Publicity Committee.

Formal announcements about the scholarship, in-cluding application instructions, will be sent to relevanteducational institutions early next year.

The scholarship is a joint project of the Measure-ment Quality Division, the Measurement Science Con-ference, the National Conference of Standards Labo-ratories, and friends and colleagues of the late JoeSimmons in recognition of his vision and leadership inthe field of metrology and his many contributions to theefforts of the scholarships sponsors. Your Divisionsustains the scholarship by providing for conference-call meetings and Web-based operational support, inaddition to its financial contributions.

Individual contributions to and commercial spon-sorship of the scholarship are welcome; please contactNorm Belecki for details.

Contact information:

Norm [email protected]

Mark [email protected]


11/32


INTEGRATION OF ISO 10012 INTO U.S. MEASUREMENT PROCESSES

By William McCullough

ISO standards are generic to a purpose and written to encompass a wide range of uses and complexity. Theintroduction clause of ISO 10012:2003, Measurement Management Systems--Requirements for MeasurementProcesses and Measuring Equipment describes measurement systems with great latitude. It suggests that theuse of the standard is from basic equipment verification to application of statistical techniques. That means thatthis standard is applicable for a simple shop floor dimensional reading with a caliper to the pre-flight parametertest on the space shuttle. As with any standard, the application of 10012 should be scaled to fit the process.

ISO 10012s Measurement Management System (MMS) approach sounds like a new concept with added work,but you might be surprised to find that it is a process that most of us in metrology have been using day in and dayout for years. We all have been well trained on the following key elements that need to be in place in order to callsomething calibrated whether we do it ourselves or purchase it:

The organization needs to be legally described and identified. Each of the processes need to be identified and documented. Calibration must be a customer-focused service. Metrological functions should be closely aligned with the quality function, have clear quality objectives and

independence from production pressures. The quality, objectives and performance must be audited for conformance. Management and staff must be competent and trained in their defined functions. Calibration is a legal process requiring documented procedures and records proving traceability. Records include means of identifying all measurement equipment and their calibration status. Calibrations are sensitive to and require control and monitoring of the environment. Change is a reality that requires the maintenance of calibration intervals. Calibration failure is a customer risk requiring possible corrective action. The value of a measurement is arbitrary without a calculated uncertainty.

Calibration standards such as ISO 17025, General requirements for the Competence of Testing and CalibrationLaboratories, and ANSI/NCSL Z540-1, American National Standard for Calibration--Calibration Laboratories andMeasuring and Test Equipment--General Requirements provide guidance for the calibration and test laboratoriesand could have been used to create the above list. To make a point, the list was not created from either standard.The list was actually gleaned from 10012 demonstrating that, although 10012 may seem just a bit different, its corepremises are still basic, fundamental metrology.

When we decide to purchase a new measurement device, without thinking about it, the MMS process is ofteninvoked. The general requirements for the instrument are first determined. Because many instruments havemultiple uses, specifications that will meet general requirements are stipulated. The developed specifications arethen used to evaluate candidate instruments. The instruments that align with the general specification arevalidated to the general requirement and are candidates for purchase. Only then should the best bang-for-the-buck selection be made.

Factual (sales information) is not always actual (product realization), so the validation and selection processshould be sufficiently rigorous to include some form of testing to ensure that the instruments specifications arealigned with its actual performance. In another form of validation, the supplier is selected based on their abilityto meet the documented requirements. (6.4) This is important because in the next step the instrumentsspecifications will be subjected to the calibration process.


12/32


In the calibration process, a standard with uncertainties suitably low enough not to affect the calibration resultstests the entire instruments functions, ranges, and scales. If the validation and selection was not done withsufficient rigor, it is possible for the instrument to fail its calibration. Failure of initial calibration should beconsidered evidence of a suppliers ineffective quality program and should trigger the re-validation of thatsuppliers instrument(s). Successful calibration assures that the validated instrument is calibrated and ready forconfirmation for specific use.

Confirmation deals with the specific use that the instrument will undertake after calibration. Each quantity tobe measured will have a specified or inferred limit. That limit is compared with the specifications or uncertaintyof the instrument. When the instrument is determined to be sufficiently accurate as to minimize its effect on themeasurement, the measurement is confirmed and you have a MMS. This is a standard process for metrologistbut not always the case outside of the calibration lab. Consistent with 10012s concept of confirmation, Z540-1and 17025 require us to do an uncertainty analysis on every measurement to determine that the measurementmeets its requirement. Again we are dealing with new terminology, not concept.

The validation/calibrations/confirmation process used for simple low-end meters and hand tools can be lookedat as Cal-Lite or Cal-X (eXtreme) for major systems or measurement standards requiring in-depth parameterevaluations. Regardless of the level of confirmation, and by any other name, confirmation has always been partof calibration. ISO 9001-1994 asked that measurement systems be adequate for their intended use. ISO 9001-2000 has a new expression for the same idea: meets the determined requirement for the product. Requirementfor the product is a variable that ranges from Cal-Lite to Cal-X.

The first issue to come to grips with is that 10012 is a measurement standard; and, although within it calibrationis encapsulated, it is not a calibration standard. The MMS requirement correctly places the burden of evaluation,selection, and use of a valid measurement system on the customer. Stay tuned for a discussion as to who thecustomer is. As a minimum with Cal-Lite, the customer makes themselves aware of the measurementequipments specifications and applies them to the measurement. In the most demanding Cal-X measurementsfull uncertainty evaluations, type A and B, are understood and documented. In both cases the results of theunderstood uncertainty are used to ensure that the measurement system is of sufficient accuracy to producemeaningful measurements with low risk of error.

As metrologists our responsibility is to make sure that our internal MMS is an example which the rest of thecompany or our customers can follow. Unfortunately, we may have no control over our customers. Lets look atwho the customer is.

The Customer can be the end user but is not necessarily a person. It could be a contract, quality manual,standard operating procedures, or predefined requirements. The customer can be considered to be any mutuallyunderstood requirements. ISO 9001 as well as 10012 make a very strong commitment to fulfilling the customersrequirements. To make customer requirements less nebulous, we need to focus on why measurementequipment is being calibrated. The instruments metrological characteristics must conform to the requirement forits intended use or in terms that we are used to; it must meet its published specifications. Most of the timecalibration of an instrument to its specification meets the customers requirements, so the only thing new here isthe terminology.

One more bit of terminology that should be dealt with is product. ISO 9001 describes product as the outputof a process. Some are struggling with the term process because they are making it more difficult than it reallyis. ISO 9001 further defines process as: set of interrelated or interacting activities, which transforms inputs intooutputs. The output of a process is a product, and calibration is a process with the product being a calibratedinstrument. There are processes within or feeding other processes such as ordering parts, writing procedures,or recalling instruments. Long before 9001 came about, good management practices suggested, in similar terms,that processes should be identified, documented, and controlled.

An internal lab calibrating instruments for the changing faces of production workers that know little about theirtarget measurements seems to be a process with an ambiguous customer and ambiguous requirements. Weneed to understand that there can be more that one customer and product for any process. The productionworkers are customers that have a requirement of getting their instrument back so they can get their work done,but they are not the customers that set the calibration requirement. That customer is the entity responsible forestablishing the use specifications.


13/32


Most of the world has come together on one quality management system (QMS), the ISO 9000 family ofstandards. ISO 10012s MMS is aligned with ISO 9001s QMS to embrace a customer-focused process approachas applied to measurements. It also focuses on continual improvement and error prevention.

When customers ask that companies and their products conform to ISO 9001, they have set their primaryrequirements. ISO 9001 clause 7.6 changed its measurement terminology to more precisely define therequirement of conforming product. It now uses conform to determined requirements, which in the U.S. isanalogous to referring to published specifications. ISO 9001 clause 7.2 relates the customers requirements tothe product design. So it can be said that, in the customers default view, products are expected to meet theirdesigned and advertised specifications.

To make sure that the output of a process relating to measurements is effective, under control, and minimizesthe customers risk of receiving a nonconforming product, ISO 9001 clause 7.6 suggests the use of 10012 for theorganizations measurement processes. Calibration is a subset of all measurements so it is included. Foraccreditation of calibration laboratories, 10012 defers the reader to 17025. Those standards define the customerdefault requirements. To realize those requirements, products must be designed, produced, and tested to the9000 family of standards or U.S. equivalents. To finalize the process and assure customers satisfaction, theproduct must be tested rigorously enough to reduce the customer risk of receiving non-conforming product usingthe guidance in 10012 and 17025.

The products final testing is one of the calibration labs strongest links to the internal customer, the design, andtest engineers. The final tests were specifically designed by test engineers to meet the products designedspecifications. Proper execution of design and test will delight the customer by providing what was ordered. Thosetests were done with instruments that the calibration lab made sure met their specification through calibration, andthe final test made sure that their test instruments were confirmed to determined requirement for the product.

So, in conclusion, those that have an operational calibration laboratory or process that already meets the needsof our customers by conforming to current standards have few changes to make. When a new standard isintroduced, we are not going to scrap what we have and rebuild the laboratory. No, we will do a gap analysis todetermine what exactly we will need to do to align our current processes to the new standard. If the new standardincludes something new, we will integrate it into the current processes. If the standards indexing or somethingwithin our labs current scope has changed, we will adjust our process and re-index our calibration manual.

The use or the addition of 10012 to Z540-1 doesnt require much change to calibration laboratories. It does,however, present an opportunity for improvement outside the calibration function by extending the principles ofmetrology to others that make measurements. Most of us in quality and metrology have, at one time, suggestedthe need for just that. This is a good thing.

I f y ou d on t h av e e no ug h i nfo rm at io n t o a rg ue b ot h s id es o f t he i ss ue , y ou d on t h av e e no ug hi nfo rm ati on to m ake a n i nfo rm ed d ec isi on . S ch izo ph re ni a c an b e a n a sse t.

Div i s ion F inance s

As of October 31, 2003

Cash: Liabilities:Checking 18,284.68 Owe HQ 170.00Money Market 72,369.48

Receivables: Total Assets: 91,832.26Advertising 105.00 Total Liabilities: (170.00)HQ owes 1,073.10

Total Assets: 91,832.26 Fund Balance: $ 91,662.26


14/32


YOUVE FOUND THE CURE FOR CANCER. . .

BUT YOU CANT REPRODUCE YOUR RESULTS!

COULD YOUR TEST EQUIPMENT BE THE WEAKEST LINK?

By: Jay L. Bucher

To paraphrase from a compliance guide covering Good Manufacturing Practices (GMP): Manufacturers areresponsible for ensuring the establishment of routine calibration on their test equipment so it will be suitable forits intended use. The core requirements for establishing and maintaining an in-house metrology department canbe found in various regulations and recommended practice guidelines. This paper explains how the MetrologyDepartment at Promega Corporation meets all these requirements. Topics covered include identifying testequipment, writing procedures, training technicians, recording data, traceability of measurement and standards,uncertainty budgets, and scheduling practices. Also covered are some valuable lessons learned that might helpyou establish, improve, or maintain a metrology program that instills confidence in your measurements and test

equipment.

1 INTRODUCTION

1.1 Most in the Metrology community are familiar with calibration laboratories and how they function andoperate. What about companies that want to meet Good Manufacturing Practices (GMP) or the InternationalOrganization for Standardization (ISO) requirements, manufacture products worldwide, or are trying to find thecure for cancer? How can they guarantee their products and/or results are reproducible, at home or abroad,without sending all their test equipment off-site for calibration? Not only is this true for GMP but also ISO and anymanufacturer or research facility requiring reproducible results or products.

1.2 Lets assume your R&D section thinks they have discovered the cure for cancer but can replicate their

findings only if they use one special set of pipettes. Or a research scientist believes she has found the answerto Alzheimers disease but gets various results depending on which spectrophotometer she uses. How about aproblem your production facility has with cross contamination, unless they use the same autoclave each time forsterilization, holding up the production line, and leaving three other autoclaves idol in the process. Or even thepossibility that foreign-made parts wont align with American-made parts on the assembly line, even though all aresupposed to be made to the same tight specifications. Could the lack of calibrated test equipment possibly be themissing or weakest link in each scenario? Or worse yet, the use of improperly calibrated test equipment might bemagnifying the problem or masking the solution. A reputable metrology program that meets established guidelinescould be the answer to some or all of these problems.

1.3 By incorporating a metrology program that encompasses a quality system used by trained technicians andsupervisors and which follows established metrology principles, you will greatly reduce your risk, lower your overallcost in the calibration and maintenance of your test equipment, and be able to get accurate, repeatable, and

traceable measurements anywhere in the world. It takes dedicated people, with the proper training, usingtraceable equipment, and a quality system that pays more than just lip service to its guidelines. All of theseprinciples are tried and true, have been in use for decades, and allow companies, both large and small, tomanufacture products all over the world, meeting the same specifications and achieving high quality, no mattertheir location, climate, or the nationality of the workforce.

1.4 Allow me to explain what I feel are the main differences between a calibration laboratory (commercial orprivate) and a metrology department. Most calibration laboratories are self-contained, located in their own facility,areas or set of rooms. The customer usually delivers their test equipment to the scheduler or the cal labs contactpoint. And the average calibration technician never sees or interacts with the equipment user/owner. In themetrology departments with which I am familiar, the vast majority of calibrations occur in the room, lab, orproduction area where the test equipment is located. Going on-site could mean stepping next door or having totravel to another facility, taking your standards, calibration procedures, forms, and labels with you. Here at


15/32


Promega, we are on a first-name basis with each of our customers and have a good idea how each piece ofequipment is used on a daily basis. Each calibration technician is trained to do their own scheduling, using oursoftware program, and is responsible for all the test equipment located in the building where they are assigned.This not only gives them a well-rounded background in the calibration and repair of a wide variety of equipmentbut also builds self-confidence, initiative, and a pro-active attitude in addressing customers problems andpresenting solutions.1.5 The basic premise and foundation of a good quality system is to Say what you do, do what you say, recordwhat you did, check the results, and act on the difference. Simply stated: Say what you do: write down in detailhow to perform every job (have calibration procedures); Do what you say: follow those procedures every timeyou calibrate your test equipment; Record what you did: accurately record the results of your measurementsand adjustments; Check the results: ensure the equipment meets all your specified tolerances; and finally, Acton the difference: if the instrument is out of tolerance, you must inform the user because they may have to re-evaluate manufactured goods, change a process, or recall a product. This concept is not new. It has been aroundas long as companies have wanted to produce a quality product, no matter the location of the manufacturing facilityor skill level of the workersfollow established instructions, record the findings, and act on the results.

1.6 You should establish a quality management system to ensure that all operations throughout themetrology department occur in a stable manner. The effective operation of such a system will result in stableprocesses and, therefore, in a consistent output from those processes. Once stability and consistency areachieved, it is possible to initiate improvements. Technicians must follow the calibration procedures, collect thedata as it is found, and document the results accurately. Only then can trends be evaluated, intervals increasedor decreased, and improvements to your processes and/or procedures implemented.

1.7 According to the small entity compliance guide, under calibration requirements: The Quality Systemregulation requires in section 820.72(b) that equipment be calibrated according to written procedures that includespecific directions and limits for accuracy and precision. GMP calibration requirements are: Routine calibration according to written procedures; Documentation of the calibration of each piece of equipment requiring calibration; Specification of accuracy and precision limits; Training of calibration personnel; Use of standards traceable to the National Institute of Standards and Technology (NIST), other recognizable

standards, or when necessary, in-house standards; and Provisions for remedial action to evaluate whether there was any adverse effect on the devices quality. [1]

1.8 Each of the areas identified above will be discussed in detail in section 2 through section 8. Section 9contains some lessons learned from an ISO 9001 compliant metrology program that has been in place for fiveyears. A short summary can be found in section 10.

1.9 Promega Corporation is a worldwide leader in applying biochemistry and molecular biology to thedevelopment of innovative, high-value products for the life sciences. The metrology department at Promega hasa manager, an electronics engineer/repair technician, and two calibration technicians. We support 6500+ items,with more than 3700 items requiring calibration and/or preventive maintenance inspections. We have had noitems overdue calibration since February, 1999. The mention of any particular companys product or service in

this paper does not imply an endorsement by Promega Corporation or any of its employees.

2 METROLOGY MANAGEMENT

2.1 Managers and administrators should understand the scope, significance, and complexity of a metrologyprogram in order to effectively administer it. The selection and training of competent calibration personnel is animportant consideration in establishing an effective metrology program. Personnel involved in calibration shouldideally possess the following qualities: Technical education and experience in the area of job assignment; Basic knowledge of metrology and calibration concepts;


16/32


An understanding of basic principles of measurement disciplines, data processing steps, and acceptancerequirements;

Knowledge of the overall calibration program; Ability to follow instructions regarding the maintenance and use of measurement equipment and standards;

and Mental attitude which results in safe, careful, and exacting execution of their duties. [1]

2.2 Training new calibration and/or repair technicians can be a daunting task. Hiring qualified, trained, andexperienced personnel can be difficult, depending on the skill levels needed, geographical location of your facility,or economic status of your location and company. This should not keep you from having a training program inplace. At Promega, we list all the required training and update according to the changing environment of testequipment we support. Integrity, traceability, documentation, computer skills, and a sense of humor are thefoundation for our training program. No matter what new equipment is purchased, or old equipment repaired andplaced back in service, the above disciplines have to be understood and met by each employee in the department.

2.3 I wrote the following in a paper published in Cal Lab: Allow me to paraphrase RP-6 [2], paragraph 5.12.:

The selection and training of competent calibration personnel is an important consideration in establishing andmaintaining an effective metrology program. Calibrations should be performed by personnel having thenecessary education, training, and experience. They should have a basic knowledge of metrology and calibrationconcepts and the ability to follow instructions. There are different ways to get proper training and experience:attendance at vocation or trade schools, through the militarys Precision Measurement Equipment Laboratory(PMEL) school, specialized classes given over a few days such as the Lighthouse Training Group [3] offers, oron-the-job-training under an experienced Metrologist. NCSLI can assist in locating various sources as your needsdictate. The training and knowledge gained by an experienced PMEL technician can be useful in filling theserequirements. Most PMEL technicians understand traceability, authoring of calibration procedures, and the set-up and maintenance of an effective scheduling program. This has allowed them to develop the skills and patienceto produce a quality product as a matter of course. You could consider this paper a blueprint for setting up ametrology program. Like any well-designed program, you need the right person to read and understand thoseblueprints. Youll need to modify the design to fit the requirements and circumstances of your organization. A good

place to start your program might be with an experienced metrologist who has a background in PMELmanagement, scheduling practices, writing procedures, and extensive experience calibrating a variety of testequipment. Id like to stress one point hereRegardless of the quality of your standards or the best intentions inthe worldif your technicians do not understand metrology concepts and disciplines and do not have the highestintegrity, your program is doomed to fail! Integrity is the glue that will hold your program together. [4] This is astrue today as it was when the Royal Egyptian Cubit was originally used more than 5000 years ago.

2.4 To help get started on a training program, here are some of the common areas that should be explainedto new and, in some cases, experienced calibration technicians. Traceability and documentation are discussedfrom the start and repeated often. Since our department is paperless in data collection and storage, competentuse of computers, both PC and laptops, is essential. Once the ability to properly document, save and storeinformation is shown, we can move on to other topics. The proper use of our management software is critical intheir ability to schedule their own workload and see what is coming up, overdue, or needs to be repaired. We usean in-house program that has been tailored for our specific needs. There are numerous off-the-shelf programsthat can be adapted to fit your needs and requirements. As a starting point, LABMATE by Norfox [5] andCALIBRATION MANAGER by Blue Mountain Quality Resources [6] are used throughout the industry and providedemo software for trial periods. We also explain how and why we sometimes limit the calibration of someequipment. If a particular function on an instrument is not working but not used by the customer, and that functionhas no effect on other functions of the test instrument, it might be cost effective to delay or cancel repairing theunit and give it a limited calibration.

2.5 This is when you would use a limited calibration sticker to inform the customer that all functions of thetest equipment were not calibrated. If a test instrument requires calibration at specific settings due to the needsof the customer or process, a limited calibration would also be appropriate. There are many other areas that wecover, but most are specific to the type of test equipment used throughout a biotechnology manufacturingcompany.


17/32


2.6 One area that is of concern to most, no matter what you manufacture or produce, would be in determiningthe accuracy of each type of equipment. The main concern here is with instruments that have digital readouts.When setting tolerances for these types of items, keep in mind that the least significant digit (LSD) can impact yourreadings and accuracy limits. Also, explaining the difference between tolerances that specify percent of readingversus percent full scale can be difficult for new technicians to understand. Using examples of specific testequipment where they can see the difference it makes can help get them off on the right foot.

3 PROCEDURES

3.1 Allow me to give a partial quote from 21CFR820.72 Inspection, measuring and test equipment: (a) Eachmanufacturer shall establish and maintain procedures to ensure that equipment is routinely calibrated, inspected,checked, and maintained. These activities shall be documented. (b) Calibration procedures shall include specificdirections and limits for accuracy and precision. These activities shall be documented. (1) Calibration standardsused for inspection, measuring, and test equipment shall be traceable to national or international standards. (2)The equipment identification, calibration dates, the individual performing each calibration, and the next calibrationdate shall be documented. [7] Can you see a trend developing here? Proper documentation is critical, not only

to the process, but also for future statistical analysis, review of calibration intervals, and production trends and rootcause analysis when problems are identified. By having the proper documentation in place, you do not have tore-invent the wheel every time a person in a critical position moves on, retires, or gets hit by a train. Continuityis an important ingredient of everybodys production and manufacturing processes and procedures.

3.2 A typical equipment calibration procedure includes : Purpose and scope; Frequency of calibration; Equipment and standards required; Limits for accuracy and precision; Preliminary examinations and operations;

Calibration process description; Remedial action for product; and Documentation requirements . [1]

3.3 Please refer to Example 1 and Example 2 on the following pages. At Promega, we specify the calibrationinterval for each particular instrument on the matching calibration worksheet (shown in Example 3). This is wherethe calibration dates are also recorded and used in helping the calibration technician fill out the calibration labels.


18/32


Title: Balance and Scale Calibration Procedure Procedure No. SOP169 Rev. No. 03

Submitted by: Anna Terese Public Date: 2/29/00 Approved by: Ayumi Jane Deaux

READ THE ENTIRE PROCEDURE BEFORE BEGINNING.

1 PURPOSE

This Standard Operating Procedure (SOP) describes the responsibilities of the Metrology Department as theyrelate to the calibration of all balances and scales. The intent of this SOP is to give the reader an example ofhow to format and structure a calibration procedure.

2 SCOPE

This SOP applies to all balances and scales that impact the quality of the goods supplied by Acme WidgetCorporation, Eatmorecheese, Wisc.

3 RESPONSIBILITIES3.1 It is the responsibility of all metrology technicians who calibrate balances and scales to comply withthis SOP.

3.2 The person responsible for the repair and calibration of the balance or scale will wear rubbergloves and eye protection. The balance or scale must be cleaned and/or decontaminated by the user beforework can be accomplished.

3.3 A calibration interval of 12 months will be used for all balances and scales, unless otherwise stated.

4 DEFINITIONS

4.1 NIST National Institute of Standards and Technology

4.2 TI Test Instrument.

TEST INSTRUMENT SPECIFICATIONS

Manuf. P/N Usable Range Accuracy

Allied 7206A 500 mg ~ 500 g 30 mg (500 mg ~ 30 g) > 30 g 0.1% of Rdg

Denver 400 500 mg ~ 400 g 20 mg (500 mg ~ 20 g) > 20 g 0.1% of Rdg

Ohaus V02130 50 mg ~ 210 g 3 mg (50 mg ~ 3 g) > 3 g 0.1% of Rdg

EQUIPMENT REQUIREMENTS (STANDARDS)

Weight Size Accuracy (mg) Class

25 Kilograms 2.5 grams F

5 Kilograms 12.0 1

50 milligrams 0.01 1

1 milligrams 0.01 1

Example 1. Calibration Procedure Requirements.


19/32


5 PROCEDURE

5.1 General inspection

5.1.1 This gives the reader an idea of numbering and formatting of these procedures.

5.2 Leveling the TI

5.2.1 Be as specific as possible in your instructions.

5.3 Calibrating the edges of the weighing pan

5.3.1 Following the example in Figure 1, place a weight equal to approximately 1/2 the capacity ofthe TI, on edge 1 (place the single weight 1/2 the distance between the pan center and the usable edge).Record the reading on the calibration worksheet.

1 1

4 2 4 2

3 3

Figure 1

5.4 Calibrating the TI

5.4.1 Tare or zero the TI.

6 RELATED PROCEDURES

7 FORMS AND RECORDS

8 DOCUMENT HISTORYRev. # Change Summary

00 New document

01 Added edge calibration.

02 Changed accuracies to be 0.1% of Reading, and instituted usable range at low and high endof all balances.

Disconnect And Secure All Test Equipment

Test Instrument Calibration Points

Manufacturer Part Number Calibration Test Points in Grams

Allied 7206A .5, 1, 50, 100, 200, 500

Denver/Fisher 400 .5, 10, 50, 100, 200, 400

Ohaus V02130 .05, .5, 5, 10, 100, 210

Example 2. Calibration Procedure Requirements.


20/32


3.4 Weve developed an Alert/Action process for equipment that is found out of tolerance. Under this system,whenever test equipment is found out of tolerance by up to twice its specification, we call it an Alert. If it is outof tolerance by more than twice its specification, it falls into the Action category. Your needs and processes willdetermine where you draw the line between Alert and Action. We inform the equipment user of all the datacollected and, if applicable, any history we have on the item. It is their responsibility to determine if the equipmenthad an affect on product or if a recall is needed. We aggressively track all our Alert/Action items. They have provento be a good source of information for early detection of items that need their intervals changed, specificationsre-examined, or combinations of both; but this must be done judiciously.

4 CALIBRATION RECORDS

4.1 Calibration of each piece of equipment shall be documented to include: Equipment identification; The calibration date; The calibrator; and

The date the next calibration is due. [1] 4.2 An important part of your system will be your records. Whether you generate hard copy or electronicrecords, they need to be accurate, controlled and uniform. Youll also need to include the range and tolerancesof the instrument being calibrated, identification of the standards used during the calibration, identification of theprocedure used to perform the calibration, what your standard(s) read, and the As Found data of the testinstrument being calibrated and As Left if the unit was adjusted and/or repaired. We place all of the requiredinformation on the calibration label, in the calibration worksheet that is saved and stored in a secure location, andalso in our Metrology Automated Management System (MAMS). When we are audited, we can show our papertrail of traceability through the calibration record and MAMS. Within MAMS, the date due calibration of thestandard(s) used at the time the test instrument was calibrated is automatically recorded for future reference. Itis also a way for the technician and the person co-signing their paperwork to check that the standards were notoverdue at the time they were used to calibrate the test instrument (it is assumed that each and every time a

standard is used, the calibration label is checked to ensure the standard is not overdue calibration; but the realityis most technicians take it for granted their standards are not overdue--management takes care of that, right?).Example 3 on the next page shows a calibration record used at Promega.

4.3 At Promega, we use a five-digit bar code label attached to every piece of test equipment. It is the soleidentifier we use on the calibration worksheet and in MAMS. The part number and serial number were long usedas an accurate system for identifying test instruments. We found this to have a couple of problems. The first wasmost instrument data plates were located on the back or bottom of the equipment. This made it difficult to easilyidentify the item without moving, lifting, or sliding it around bench tops. Also, many items do not have serialnumbers assigned by their manufacturers. Scribing numbers and making sure you dont duplicate serial numberscan be time consuming and tedious. If we found an item without a serial number, we assigned it the same five-digit number we use as the ID number.


21/32

______________________________________________________________________________________

______________________________________________________________________________________


Title: Balance and Scale Calibration Worksheet Procedure No . MCW169 Rev No. 13

Submitted by: Anna Terese Public Date: 2/29/00 Approved by: Ayumi Jane Deaux

ID #: ___________________ P/N: _________________________

Range/Capacity: ____________________ Accuracy: _______________________________

Last Cal: _____________ Todays Date: _______________ Date Due Cal.: _______________

Room No: _________________ User Department: ______________________

Interval:o 1 month o 6 months o 12 months o 18 months

Edge Std Weight TI As Found Reading TI As Left Reading

12

3

4

ID No. Std Weight TI As Found Reading TI As Left Reading

This TI was tested against standard weights traceable to NIST (See the reverse side of this sheet for their accuracyand calibration due dates).

This TI was calibrated in accordance with SOP11C169.

This TI falls within specifications: o Yes

o No (If no, an Alert/Action Procedures will be initiated)

Comments: ____________________________________________________________________________

METROLOGIST: ____________________________________ Date: _______________________________

MAMS Updated: o Yes o No o N/A Cosigned/Approved By: ______________________

Example 3. Calibration Record Requirements.


22/32


4.4 We subscribe to the concept of using an actual day of the month when an item goes overdue calibration(as opposed to any time during a specific month). If you use a monthly schedule, an item can be calibrated on 1February, and be due in March (if it has a one-month calibration interval). If it is calibrated on 31 March, it has gone58 days between calibrations. If it is calibrated on 27 February, and again on 3 March, it has met the systemscriteria but gone only four days between calibrations. This could be a waste of valuable resources (standards, thetechnicians time, and data collected on the test instrument). But the biggest drawback under GMP or ISOrequirements would be: when does the unit actually go out of calibration or, more importantly, when does the userhave to stop using it? By having a specific date and time (midnight on the due date), there is never a question inanyones mind.

5 SCHEDULING EQUIPMENT FOR CALIBRATION

5.1 Measuring instruments should be calibrated at periodic intervals established on the basis of stability,purpose, and degree of usage of the equipment. A manufacturer should use a suitable method to remindemployees that re-calibration is due. [1]

5.2 The use of metrology management software cannot be over emphasized. Without it, you probably will notbe able to see what needs to be calibrated and what has gone overdue calibration and not have the ability to collectdata on trends, repairs, or calibration interval adjustments. As mentioned in paragraph 2.4, there are numerousoff-the-shelf software packages available. Tailoring them for your individual needs will allow you to gather therequired data with just a click of your mouse or the press of a button. What you do with that data could be a differentstory.

5.3 When we generate a 30-day schedule, it is sorted by where it is located (facility and room) and type ofequipment. Giving calibration technicians a look into the future and the ability to understand what needs to beaccomplished over a given period of time only enhances their productivity. By calibrating like items instead ofeach test instrument as their due date approaches, we can make better use of our resources. With limitedstandards available, the time to set up and tear down equipment is critical in the overall picture. By looking at allthe items coming due in a specific facility, conscientious time management, coupled with resource allocation, andreduced travel time have kept our overdues at zero and eliminated the need for additional manpower during times

of fiscal restraint and budget crunches. The majority of this came about because we use the tools (managementsoftware) that are at our disposal.

5.4 Most manufacturers recommend having their equipment calibrated at one (1) year intervals. In somecases, there are reasonable, historical data to support this time frame. However, in todays high-tech world, wheresafe journeys into space and under the sea happen as everyday occurrences, reliable test equipment has becomethe norm instead of the exception. By collecting data on how often items are found out of tolerance, and by howmuch, and how often they require unscheduled repair, calibration intervals can be realistically lengthened orshortened and both time and money saved in the number of calibrations performed in a given year. Theres a fineline between being on the cutting edge and being on the ragged edge when it comes to calibration intervals.Give any changes, especially lengthening of intervals, enough time to collect data to see if youve been a geniusor an idiot.

6 TRACEABILITY OF STANDARDS

6.1 The QS regulation requires that standards used to calibrate equipment be traceable to the NationalInstitute of Standards and Technology (NIST) or other recognized national or international standards. Traceabilityalso can be achieved through a contract calibration laboratory which in turn uses NIST services. [1]

6.2 If a company follows the guidance in ANSI/NCSL Z540-1-1994, paragraph 10.2(b) The laboratory shallensure that the calibration uncertainties are sufficiently small so that the adequacy of the measurement is notaffected. Well-defined and documented measurement assurance techniques or uncertainty analyses may beused to verify the adequacy of a measurement process. If such techniques or analyses are not used, then thecollective uncertainty of the measurement standards shall not exceed 25% of the acceptable tolerance (e.g.,manufacturers specification) for each characteristic of the measuring and test equipment being calibrated or


23/32


verified [8]; they will have traceable measurement. This 25% of the acceptable tolerance is where the 4:1 ratiois derived.

6.3 Heres an example: your standard has an accuracy of 0.25C; the most accurate item you can calibratecould be no better than 1.0C, or four times less accurate. You must ensure the combined uncertainty of thestandard(s) being used is at least four (4) times more accurate than what is being calibrated. By calibrating testequipment in the environment where it is normally used, you can determine how well it works and the repeatabilityof that instrument. This is also where traceability is derived. Traceability is defined as an unbroken chain ofcomparisons back to an international/national standard. During each calibration along that chain, a 4:1 ratio oruncertainty analysis must be documented. Only then can you show your equipment is traceable back to a specificstandard or level of accuracy.

6.4 NIST asserts that providing support for a claim of traceability of the result of a measurement or value ofa standard is the responsibility of the def08 provider of that result or value, whether that provider is NIST oranother organization, and that assessing the validity of such a claim is the responsibility of the def09 user of thatresult or value. NIST also communicates, especially where claims expressing or implying the contrary are made,that NIST does not def11 define, def12 specify, def13 assure, or def14 certify traceability of the results ofmeasurements or values of standards except those that NIST itself provides, either directly or through an officialNIST program or collaboration. Service providers are responsible for showing their traceability, no matter to whomit is traceable.6.5 Your documentation, whether it is your own calibration record or a calibration certificate from an outsideagency, needs to contain certain information. The items unique identification number and owner, when it was lastcalibrated, and its next due date are a given. There should be a stated uncertainty budget or ratio and, whenapplicable, a limit on the tolerances and specifications of the instrument that was calibrated. What standards wereused to calibrate the instrument should be shown, as well as when those standards were calibrated or are nextdue calibration. There should be a signature of the person performing the calibration and the date the calibrationwas completed. As-found and as-left data should be included or statements to the effect that adjustments werenot made and that the instrument met tolerances as received.

7 CALIBRATION ENVIRONMENT

7.1 As appropriate, environmental controls should be established and monitored to assure that measuringinstruments are calibrated and used in an environment that will not adversely effect the accuracy required.Consideration should be given to the effects of temperature, humidity, vibration, and cleanliness when purchasing,using, calibrating, and storing instruments. [1]

7.2 When a manufacturer produces test equipment, they specify certain parameters that must be met (e.g.,an operating range of -10C ~ 50C). Why then does the unit need to be calibrated in an environmentally controlledlaboratory when it is not used there? In most cases it doesnt! An environment that controls temperature, humidity,vibration, dust, etc., reduces the uncertainty of measurement and allows for the calibration of state-of-the-art testequipment to high accuracies that could not be ascertained in an uncontrolled environment. Im referring totolerances of three parts per million, or ten thousands of an inch, as an example.

7.3 When calibrating in an area other than an environmentally controlled and monitored calibrationlaboratory, the calibration environment need be controlled only to the extent required by the most environmentallysensitive measurement performed in the area. You need to consider not only the instrument being calibrated butalso the standards used to perform the calibration. Not all standards are created equal, so it is prudent to checkthe manufacturers stated specifications for environmental conditions that the standard can be used in, bothtemperature and humidity, when applicable.

8 AUDITING OF THE CALIBRATION SYSTEM

8.1 The calibration program shall be included in the quality system audits required by the QS regulation. [1]

8.2 Are annual internal audits conducted? Are they documented and distributed to the appropriate personnel


24/32


and managers? As a minimum, if there is no internal audit function, a self-inspection program could go a long wayin preparing you for audits and inspections. By setting up a self-inspection program, youre showing an effort tofind problems and see where youre not meeting the quality system you have in place. Demonstrating a desire tocontinuously improve your program through self initiative, and finding opportunities before they are found byothers, makes you proactive instead of reactive to problems and solutions.

8.3 One way to audit yourself is to follow the Say what you do, do what you say, record what you did, checkthe results, and act on the difference theme. Check if you are actually following the procedures, or are they simplythere for fluff? Do all your records contain the required information, and do they show a paper trail for traceabilitypurposes? If an item was found to be out of tolerance during calibration, was action taken? Was the customerinformed, and do you have data to show that it occurred? The more specific you are, the easier it is to answer yourquestions. Self-inspections can be an important continuous process improvement, but it takes time, effort, andhonesty at all levels.

9 LESSONS LEARNED

9.1 Of all the things youmustremember : identifying your test equipment, writing procedures, recording data,ensuring traceability of measurements and standards, uncertainty budgets and scheduling practices, heres theone thing you should never forget continually train, train, train! You cannot provide enough training in todaysever-evolving world of metrology. While everyone around you is updating their computer systems, goingpaperless, using automated data processors and purchasing the latest and greatest in new technology, dontforget to provide the critical training on each and every one of these new systems or ideas. Having the latesttechnology does not mean everyone knows how to use it. Continuous training is the only way to ensure all yourtechnicians are correctly using the standards, systems, and data collection devices you have. And documentingthat training is the most reliable form of ensuring everyone is singing from the same page at the same time.

9.2 Telling your customers what is expected of them can solve problems before they get started. During ournewcomer orientation briefing that is given to all new scientists, we not only get them started on the right foot withtheir test equipment but we start a partnership between them and us in support of Promegas goals through theuse of calibrated test equipment. We let them know what they are responsible for:

Informing Metrology of their requirements and needs; Getting the proper training in the correct and safe usage of test equipment; Maintaining their test equipment without abusing, contaminating or damaging it under normal operating

conditions; and Using Metrologys work order system for requesting service when equipment is broken, malfunctioning, or in

need of calibration.

9.3 The same theory also applies to our Metrology Department staff. TEAM METROLOGY is responsible for: Listening to our customers to understand their requirements and needs; Translating those requirements to the accuracy and specifications of the test equipment and support services

that meet or exceed their quality expectations;

Delivering test equipment that consistently meets requirements for reliable performance; Providing knowledgeable and comprehensive test equipment support; and Continuously reviewing and improving our services and processes.

9.4 There are many times when you can use your test equipment in a limited capacity. If a particular functionon an instrument is not working but not used by the customer, and that function has no effect on other functionsof the test instrument, it might be cost effective to delay or cancel repairing the unit and give it a limited calibration.

9.5 It is important in any quality system to have change control procedures in place, but it is critical in ametrology program. Whenever you make improvements or changes to your calibration procedures and/orcalibration records, you must have a system in place to ensure old copies are removed and destroyed, as well asmaintaining history in case you need to revert back to the original. You must also make sure everyone in thedepartment is trained on those changes.


25/32


9.6 A comprehensive training record will help make sure this occurs. We all know how repetitive and boringit can be to calibrate the same widget 69 times in the same week, while following the same old boring procedure.When someones life, or the lives of many, or the cure for deadly diseases is on the line, youd better be doing itright the first time; and following your procedures is the only way to ensure that is going to happen. And the onlyway to keep track of your changes and improvements is to document them within your quality system.

10 SUMMARY

10.1 Have you figured out the answers to paragraph 1.2? At Promega, we calibrate all pipettes againsttraceable standards after performing preventative maintenance inspections and any needed repairs. The result:all of our pipettes are special! We have more than 30 spectrophotometers, made by seven different manufacturers;each one receives a preventive maintenance inspection, alignment, and calibration at a regular interval. How dowe solve the autoclave problem? Simplewe calibrate every one of them with traceable standards to ensure theymeet specified tolerances. At Promega we follow an established program, using calibrated test equipmenttraceable to NIST. By doing so, test equipment will not be the weakest link in quality production.

References:

1. U.S. Food and Drug Administration Center for Devices and Radiological Health. Medical Device QualitySystems Manual: A Small Entity Compliance Guide, First Edition (Supersedes the Medical Device GoodManufacturing Practices [GMP] Manual). http://www.fda.gov/cdrh/ dsma/gmp_man.html

2. Calibration Control Systems for the Biomedical and Pharmaceutical Industry, Recommended Practice, RP-6,May, 1999

3. Lighthouse Training Group, 117 S. Albert Street, Mt. Prospect, IL 60056; Phone/Fax (847) 392-9796 e-mail:

[email protected]; www.lighthousetraining.com

4. J.L. Bucher, When Your Company Needs a Metrology Program, But Cant Afford To Build a CalibrationLaboratory, Cal Lab, San Diego, CA, October/November/December 2001, pp. 34-43

5. LABMATE Calibration Management Software, NORFOX Software, Inc., 3400 188 th Street SW, Suite 285,Lynnwood, WA 98037, 1-800-505-9292; www.norfox.com

6. CALIBRATION MANAGER, Blue Mountain Quality Resources, 208 W. Hamilton Ave.,

State College, PA 16801, 1-800-982-2388; www.coolblue.com

7. Code of Federal Regulations, Title 21, Volume 8, Part 820 Quality System Regulation, Sec. 820.72 Inspection,measuring, and test equipment, Revised as of April 1, 2001, pp 145-146

8. ANSI/NCSL Z540-1-1994, American National Standard for Calibration - Calibration Laboratories and Measur-ing and Test Equipment - General Requirements; Approved July 27, 1994

For contact information see the officers list on page 30.


26/32


OH WHERE, OH WHERE CAN I GET A REPAIR!

THE NEW AGE OF METROLOGY REPAIRS

By: Jack Gale

A vital piece of test equipment goes down. The you-know-what hits the fan and another piece of equipment isfound. After the sky stops falling, somebody has to see that the errant equipment gets fixed. Getting inspection,measurement and test equipment (IM&TE) fixed may be slowly going the way of the traveling knife-sharpener andiceman. Many technological and market forces have drastically re-shaped this field.

These changes have occurred over the past 20 years in metrology equipment and the industries that supportthe equipment that have impacted the owners ability to repair their test equipment. As an IM&TE owner, youshould be aware of these forces when making equipment service decisions.

Overall, the evolution has made it more difficult to get repairs on test equipment. These changes include:

Technology: Increasingly, equipment has used more advanced technology. Multi-layer circuit boards andmicroprocessors have replaced simpler components. While these components have superior reliability andfunctionality, they are easier replaced than repaired. Economically, the replacement of these expensivecomponents is not wise, and a replacement of the whole instrument is warranted in many cases for all but the morecostly equipment. This is reflected in the fewer number of service centers run by some of the major IM&TEmanufacturers. Maintaining parts capabilities with antiquated technologies is not the business they support withthe same enthusiasm as new equipment sales. Outsourced support: Many companies have outsourced the support functions for their older equipment. Theyno longer hold the parts, equipment, or procedures to support product lines they no longer sell. They typically willgrant the service support to a third party or, in a worst case, they make no provisions for support for olderequipment.

Equipment cost: A larger portion of IM&TE requiring service is inexpensive test equipment used in productionoperations by production personnel. This was an inevitable result of process controls and quality programsinstituted over the past twenty years (SPC, ISO 9001, etc.). Beyond fuses or batteries, the cost of repairing theseitems can exceed the cost of replacing the unit. Immediacy: Most companies do not maintain float equipment (equipment that can be shared within thecompany when equipment is out for service). Just In Time, FIFO, lean operations and all the other forces that drivea company to ship yesterday means you cannot wait for equipment to return. Vendors and resellers are willingto ship test equipment overnight to meet your more urgent schedule. Knowledge gap: A knowledge gap exists for accomplishing repairs. While skilled technicians do remain, aflood tide of forces is making it difficult to obtain the knowledge to accomplish t

The Standard Fall 03

Documents

Transcript of The Standard Fall 03