Académique Documents
Professionnel Documents
Culture Documents
Gain (sensitivity) Error 0.01% 0.5% 20% 0.01% 0.5% 3% 0.01% 0.05% 3% 0.01% 1%
Phase Error 0.2˚ 4˚ 30˚ 0.1˚ 0.5˚ 3˚ 0.01˚ 0.1˚ 1˚ 0.01˚ 1˚
†
Voltage Output options available in ±1V and ±10V
Gain / Phase
Gain (DS200, typical) Phase (DS200, typical)
2%
0
0%
-2% -5
Phase (Degrees)
-4% -10
Gain
-6%
-15
-8%
-10% -20
-12% -25
-14%
-30
-16% 10 100 1000 10000 100000 1000000
10 100 1000 10000 100000 1000000
Frequency (Hz)
Frequency (Hz)
GMW
MW Associates • www
www.gmw.com
Volume 25, Number 3
www.callabmag.com
FEATURES
DEPARTMENTS
2 Calendar
3 Editor’s Desk
12 Industry and Research News
19 Cal-Toons by Ted Green
32 New Products and Services
36 Automation Corner
ON THE COVER: Anna M. Escarcega and Brian Valenti test a four stage compressor at Escarcega Technology Services, Inc. in Garden
Grove, CA.
Jul • Aug • Sep 2018 1 Cal Lab: The International Journal of Metrology
CALENDAR
Nov 12-16, 2018 International Conference on Precision nouveaux produits et services à plusieurs centaines de visiteurs
Engineering (ICPE). Kamakura, Japan. The aim is to provide an internationaux. Cet événement est organisé par le Comité Africain
international forum for experts to promote, share, and discuss de Métrologie (CAFMET). http://www.forumesure.com/
various issues and developments in the field of the precision and
related engineering. http://www.scoop-japan.com/kaigi/icpe2018/ Apr 8-10, 2019 Annual Meeting of the Council of Ionizing
Radiation Measurements and Standards (CIRMS). Gaithersburg,
Nov 13-16, 2018 General Conference on Weights and Measures. MD. The 27th Annual Meeting of the Council on Ionizing Radiation
Paris, France. The kilogram, ampere, kelvin, and mole will be Measurements and Standards will focus on Strengthening the
redefined at this meeting. It is expected that this meeting will agree Economy and Homeland Security with Radiation Measurements
the redefinition of the SI. As part of this Conference, BIPM will be and Standards. http://cirms.org/
organizing a number of activities, including a press conference,
to mark the occasion and are looking to broadcast the event live. Apr 10-12, 2019 METROMEET – 15th International Conference
https://www.bipm.org/en/cgpm-2018/ on Industrial Metrololgy. Bilbao, Spain. We provide information
about the latest technological, the progress made in the sector and
Jan 20-23, 2019 ARFTG Conference. Orlando, FL. Co-located we constitute a forum for debate on metrology and its development
with Radio & Wireless Week The most important part of the in a fast changing industry. Also, the latest news about; new digital
ARFTG experience is the opportunity to interact one-on-one with & optical developments and the European & international norms.
colleagues, experts and vendors of the RF and microwave test and http://metromeet.org/
measurement community. http://arftg.org/
Apr 16-19, 2019 Measurement Science Conference (MSC)
Mar 25-26, 2019 FORUMESURE. Nantes, France. FORUMESURE Training Symposium. Anaheim, CA. The conference is offering
est un salon annuel, qui s’adresse aux sociétés et aussi aux a series of excellent technical programs covering the various
institutions qui souhaitent présenter leur savoir-faire, leurs disciplines of the measurement sciences. http://www.msc-conf.com/
CALIBRATION LAB
HV CALIBRATION LAB CAPABILITIES:
• UP TO 450kV PEAK 60Hz
• UP TO 400kV DC
• UP TO 400kV 1.2x50μs
ISO/IEC 17025:2017
LIGHTNING IMPULSE CALIBRATION CERT #2746.01
• HV VOLTAGE DIVIDERS
• HV PROBES
• HV RELAYS
• HV AC & DC HIPOTS
• HV DIGITAL VOLTMETERS
• HV CONTACTORS
• HV CIRCUIT BREAKERS
• HV RESISTIVE LOADS
• SPARK GAPS
• FIBER OPTIC SYSTEMS
R OSS
ANSI/NCSLI Z540-1-1994 ACCREDITED
ISO 9001:2015 QMS CERTIFIED ENGINEERING CORPORATION
N.I.S.T. TRACEABILITY 540 Westchester Drive, Campbell, CA 95008 USA | Ph: 408-377-4621
N.R.C. TRACEABILITY info@rossengineeringcorp.com | www.rossengineeringcorp.com
Cal Lab: The International Journal of Metrology 2 Jul • Aug • Sep 2018
EDITOR’S DESK
Community
PUBLISHER
MICHAEL L. SCHWARTZ We initially purchased the magazine back in January of 2011 because we
found it an effective means of advertising for our metrology software and
EDITOR consulting business. When we approached our accountant with the numbers,
SITA P. SCHWARTZ he recommended against it. But, the magazine meant more to us than just
advertising; like the previous publisher, we felt it filled a void in the industry.
CAL LAB
PO Box 111113 Each quarter, as I compiled event dates, artwork, news, and articles, I
Aurora, CO 80042 was reminded of the whole of the community of which we are a part. The
TEL 303-317-6670 • FAX 303-317-5295 measurement sciences are at such a critical moment with technician and engineer
office@callabmag.com
www.callabmag.com attrition coinciding with the rapid acceleration of technology! Now, I’ve come
to terms that our community is very fragile. I don’t see a whole lot of common
EDITORIAL ADVISORS sentiment around measurement sciences. At least here in the states, I see people
and businesses working against each other. The calibration community operates
CHRISTOPHER L. GRACHANEN in a fractured and competitive environment.
TRANSCAT
The challenge we anticipated down the road was not what we expected.
MIKE SURACI The challenge did not stem from the publication industry, but from within
SURACI CONSULTING SERVICES
LEAD ASSESSOR, ANAB
the calibration and metrology industry. I am expressing myself here because
it’s been wearing on me… and because I can. Maybe those reading this can
MARTIN DE GROOT
MARTINDEGROOT CONSULTANCY
commiserate and find a little relief that they are not alone in feeling the same.
On a brighter note, I got to take a drive out to the lovely green and blue world
JESSE MORSE of the high plains for picture taking and a tour of the Colorado Engineering
MORSE METROLOGY Experiment Station in the northern part of the state. I put together an article
JERRY ELDRED about CEESI for this issue, but first, we have a Metrology 101 article on
TESCOM “Calculating the Frequency Response of an Oscilloscope from a Transition
Duration (Risetime or Falltime) Measurement,” by Donald Larson of Entegra
Subscription fees for 1 year (4 issues) Corporation. Don is an interesting and smart guy who has helped us on an
$50 for USA, $55 Mexico/Canada, engineering project in the past. Sometimes an article we publish will pop up
$65 all other countries. just at the time the reader needed it; so hopefully, this too will be one of those
Visit www.callabmag.com
to subscribe. helpful, reference pieces for you.
We keep an article index of past articles online, with links out to the article
Printed in the USA. posts. From the article posts, readers can download a PDF version. Please take
© Copyright 2018 CAL LAB.
ISSN No. 1095-4791
advantage of past articles by visiting: https://www.callabmag.com/article-index/.
Happy Measuring,
Sita Schwartz
Jul • Aug • Sep 2018 3 Cal Lab: The International Journal of Metrology
CALENDAR
SEMINARS: Dimensional Jan 8-11, 2019 Gage Calibration. Aurora, IL. Mitutoyo Institute
of Metrology. Mitutoyo America’s Gage Calibration course is
Dec 3-5, 2018 Dimensional Measurement Training: Level 1. a unique, active, educational experience designed specifically
Telford, UK. Hexagon Metrology. A three day training course for those who plan and perform calibrations of dimensional
introducing measurement knowledge focusing upon dimensional measuring tools, gages, and instruments. https://www.mitutoyo.
techniques. Level 1 is applicable to all industrial sectors as a com/events/seminar-gage-calibration-3/
stand-alone qualification or as a building block for further NPL
Dimensional Measurement Training Levels - 2 & 3. http://www. Jan 9-10, 2019 Gage Calibration & Repair. Omaha, NE. IICT.
npl.co.uk/commercial-services/products-and-services/training/ This 2-day training offers specialized training in calibration
training-courses/ and repair for the individual who has some knowledge of
basic Metrology. Approximately 75% of the workshop involves
Dec 5-6, 2018 Hands-On Gage Calibration and Repair. Oshkosh, "Hands-on" calibration, repair and adjustments of micrometers,
WI. IICT. This 2-day hands-on workshop offers specialized calipers, indicators height gages, etc. Course also covers NIST
training in calibration and repair for the individual who has Traceability, Certificates of Conformance, Gage Management,
some knowledge of basic Metrology. Course includes hands on Standards, etc. http://www.iictenterprisesllc.com
calibration and repairs and adjustments of micrometers, calipers,
indicators height gages, etc. http://www.iictenterprisesllc.com Jan 15-17, 2019 Gage Calibration Methods. Cincinnati, OH.
QC Training. This 3-day hands-on workshop offers specialized
Dec 13-14, 2018 Hands-On Gage Calibration and Repair. training in calibration and repair for the individual who has some
Bloomington, MN. IICT. This 2-day hands-on workshop offers knowledge of basic Metrology. Attendees will be equipped with
specialized training in calibration and repair for the individual the knowledge to meet current and future calibration needs, be
who has some knowledge of basic Metrology. Course prepared to save the company money on calibrations and grow
includes hands on calibration and repairs and adjustments of professionally. https://qctraininginc.com/courselist/
micrometers, calipers, indicators height gages, etc. http://www.
iictenterprisesllc.com
Cal Lab: The International Journal of Metrology 4 Jul • Aug • Sep 2018
CALENDAR
Jan 15-17, 2019 Dimensional Measurement Training: Level 1. Standards, etc. http://www.iictenterprisesllc.com
Bristol, UK. INSPHERE Ltd. A 3-day training course introducing
measurement knowledge focusing upon dimensional techniques. Feb 5-7, 2019 Dimensional Measurement Training: Level 1.
Level 1 is applicable to all industrial sectors as a stand-alone Bristol, UK. INSPHERE Ltd. A 3-day training course introducing
qualification or as a building block for further NPL Dimensional measurement knowledge focusing upon dimensional techniques.
Measurement Training Levels - 2 & 3. http://www.npl.co.uk/ Level 1 is applicable to all industrial sectors as a stand-alone
training qualification or as a building block for further NPL Dimensional
Measurement Training Levels - 2 & 3. http://www.npl.co.uk/
Jan 24-25, 2019 Gage Calibration & Repair. Minneapolis, MN. training
IICT. This 2-day training offers specialized training in calibration
and repair for the individual who has some knowledge of basic Feb 5-7, 2019 Coordinate Measuring Machine (CMM) - Basics.
Metrology. Approximately 75% of the workshop involves Minneapolis, MN. QC Training. This course is designed for
"Hands-on" calibration, repair and adjustments of micrometers, inspection personnel, engineering personnel, production
calipers, indicators height gages, etc. Course also covers NIST personnel, entry level inspectors, and those who would like
Traceability, Certificates of Conformance, Gage Management, to enter into the field of dimensional measurement and CMM
Standards, etc. http://www.iictenterprisesllc.com programming. https://qctraininginc.com/courselist/
Jan 30-31, 2019 Gage Calibration & Repair. Altoona, WI. IICT. Feb 6-7, 2019 Dimensional Measurement Tools - Basics.
This 2-day training offers specialized training in calibration Minneapolis, MN. QC Training Services’s Basic Dimensional
and repair for the individual who has some knowledge of Measurement Tools and Methods (BDMT) class is an introduction
basic Metrology. Approximately 75% of the workshop involves to some of the most common gages used on the shop floor.
"Hands-on" calibration, repair and adjustments of micrometers, The purpose is to provide practical instruction on the different
calipers, indicators height gages, etc. Course also covers NIST tools available and their proper use to build workers’ skill and
Traceability, Certificates of Conformance, Gage Management, knowledge. https://qctraininginc.com/courselist/
• Traceable to SI
• 0.5% of Reading RH Uncertainty
• High Flow Capability of 50 L/min
• HumiCalc® with Uncertainty Mathematical Engine
• ControLog® Automation Software Scan
• Generate: RH, DP, FP, PPM, Multi-point Profiles For
• Externally Driven Chamber Fan More
Info
• Multi-point Touch LCD
• Fluid Jacketed Chamber Door
Jul • Aug • Sep 2018 5 Cal Lab: The International Journal of Metrology
JofraCloud is a new way to perform remote reading
and setting of Jofra temperature calibrators over an un-
limited distance. The data is presented as a copy of the calibrator’s display, and can
be viewed at jofracloud.com. The webpage is accessible via a browser on any
device —PC, laptop, tablet, or smartphone.
JOFRA
RTC Series
JOFRA
PTC Series
Crystal
nVision
Crystal
XP2i
Feb 13-14, 2019 Electrical Measurement. Australian NMI. Feb 4-8, 2019 Fundamentals of Metrology. Gaithersburg, MD.
Lindfield, NSW. This two-day course covers essential knowledge NIST. The 5 day seminar is an intensive course that introduces
of the theory and practice of electrical measurement using participants to the concepts of measurement systems, units, good
digital multimeters and calibrators; special attention is given to laboratory practices, data integrity, measurement uncertainty,
important practical issues such as grounding, interference and measurement assurance, traceability, basic statistics and
thermal effects. https://www.measurement.gov.au/ how they fit into a laboratory Quality Management System.
https://www.nist.gov/pml/weights-and-measures/about-owm/
Feb 18-22, 2019 MET-101 Basic Hands-On Metrology. Everett, calendar-events
WA. Fluke Calibration. This course introduces the student
to basic measurement concepts, basic electronics related to Feb 11-15, 2019 Fundamentals of Metrology. Gaithersburg, MD.
measurement instruments and math used in calibration. https:// NIST. The 5 day seminar is an intensive course that introduces
us.flukecal.com/training participants to the concepts of measurement systems, units, good
laboratory practices, data integrity, measurement uncertainty,
Apr 22-25, 2019 MET-301 Advanced Hands-On Metrology. measurement assurance, traceability, basic statistics and
Everett, WA. This course introduces the student to advanced how they fit into a laboratory Quality Management System.
measurement concepts and math used in standards laboratories. https://www.nist.gov/pml/weights-and-measures/about-owm/
https://us.flukecal.com/training calendar-events
Jan 17, 2019 Basic Uncertainty Concepts. NIST Webinar. This 2 Dec 4, 2018 ISO/IEC 17025:2017 Bridging the Gap from 2005.
hour webinar provides a very basic introduction to uncertainty Frederick, MD. QC Training Services, Inc. In this course, the
calculations and reporting using the 8-step process published participant will become aware of the significant and subtle
in NIST SOP 29 (NISTIR 6969), beginning with some definitions changes to existing ISO/IEC 17025 laboratory system, as well as
and concepts from the Guide to the Expression of Uncertainty in the necessary steps to ensure conformity to the new Standard.
Measurement (GUM) with simple calculations. https://www.nist. https://qctraininginc.com/courselist/
Jul • Aug • Sep 2018 7 Cal Lab: The International Journal of Metrology
Trust in every measure
Your Single Source for Reference Standards Calibrations
Reference Mass Calibrations Reference Temperature Calibrations
NEW
RF Standards Calibration
(to 50GHz) 17025 Accredited
Also available:
• Humidity • Dew Point • Dimensional • Standards Rentals
SEMINARS: Industry Standards (continued) and apply the requirements of ISO/IEC 17025:2017. Attendees will
examine the origins of the standard and learn practical concepts
Dec 4-5, 2018 ISO/IEC 17025:2017 for Testing and Calibration Labs. such as document control, internal auditing, proficiency testing,
Chicago, IL. International Accreditation Service (IAS). Receive a traceability, measurement uncertainty, and method witnessing.
complete overview of ISO/IEC Standard 17025:2017 with valuable https://www.anab.org/training
insights and tools to help you implement the requirements of the
standard. https://www.iasonline.org/training/ias-training-schedule/ Jan 29-31, 2019 Calibration Certificate Evaluation. NIST Webinar.
This 4-hour webinar is conducted on two days, with a 2-hour session
Dec 5-6, 2018 Internal Auditing. Frederick, MD. A2LA. This on each day. The learning event will introduce concepts necessary
course introduces participants to the internationally-recognized to successfully implement ISO/IEC 17025:2017 compliant calibration
approaches of ISO 19011 Guidelines for Auditing Management certificates. https://www.nist.gov/pml/weights-and-measures/
Systems for conducting effective internal audits. The course includes about-owm/calendar-events
easy-to-implement methods for involvement of personnel, and
continual improvement of the audit process. https://www.a2la.org/ Feb 21, 2019 Internal Auditing Best Practices. NIST Webinar. This
events/internal-auditing 2 hour webinar will consider internal auditing techniques and best
practices that are used by a metrology laboratory to comply with
Dec 18-19, 2018 ISO/IEC 17025:2017 - The New Standard for ISO/IEC 17025:2017 criteria. https://www.nist.gov/pml/weights-
Laboratory Competence (MS 111). Frederick, MD. A2LA. and-measures/about-owm/calendar-events
This course is a comprehensive review of the philosophies and
requirements of this international Standard. The participant will Mar 11-13, 2019 Internal Auditing to ISO/IEC 17025. Indianapolis,
gain an understanding of conformity assessment using the risks IN. ANAB. The 2.5-day training course prepares the internal auditor
and opportunities-based approach. https://www.a2la.org/events to clearly understand technical issues relating to an audit. Attendees
of this course will learn how to coordinate a quality management
Jan 17-18, 2019. Introduction to ISO/IEC 17025. Las Vegas, NV. system audit to ISO/IEC 17025:2017 and collect audit evidence and
ANAB. The 1.5-day training course will help attendees understand document observations. https://www.anab.org/training
sales@IndySoft.com
Jul • Aug • Sep 2018 9 Cal Lab: The International Journal of Metrology
CALENDAR
SEMINARS: Management & Quality Feb 28, 2019 Introduction to Estimating Measurement Uncertainty.
Brisbane, QLD. Australian NMI. This one-day course (9 am to 5 pm)
Dec 4-5, 2018 Measurement Systems Analysis Using Excel will give you a clear step-by-step approach to uncertainty estimation
Statistical Tools (EMU 110). Frederick, MD. A2LA. This course with practical examples; you will learn techniques covering the
provides the participant with essential tools to be efficient in whole process from identifying the sources of uncertainty in your
laboratory operations, including making continual measurement measurements right through to completing the uncertainty budget.
process improvements in the calibration and testing environment. https://www.measurement.gov.au/
Industry-proven Statistical tools such as Statistical Process
Control (SPC), False Accept Risk, Hypothesis Testing and others Mar 5-6, 2019 Understanding & Evaluating Measurement
are covered in this workshop. https://www.a2la.org/events Uncertainty. Teddington, UK. National Physical Laboratory
(NPL). The course includes an introduction to the philosophy
Feb 7, 2019 Conducting an Effective Management Review. behind the Guide to the expression of uncertainty in measurement
NIST Webinar. This 2-hour webinar introduces the Laboratory (GUM) and the first Supplement to the GUM concerned with
Management Review process, an important tool to foster using a Monte Carlo method for uncertainty evaluation, and
communication between top management and laboratory covers material contained within those documents. http://www.
personnel to improve laboratory operations to produce quality npl.co.uk/commercial-services/products-and-services/training/
calibrations and highly satisfied customers. https://www.nist. training-courses/
gov/pml/weights-and-measures/about-owm/calendar-events
Mar 14-15, 2019 Fundamentals of Measurement Uncertainty.
Apr 30, 2019 Implementing Metrology and SPC concepts with Indianapolis, IN. ANAB. Attendees of this 2-day training course will
MS Excel. York, PA. Morehouse Instrument Company. This one- learn a practical approach to measurement uncertainty applications,
day workshop prepares the metrology professional to apply the based on fundamental practices. https://www.anab.org/training
power of Microsoft Excel’s mathematical and statistical tools to
assist in managing the laboratory’s Quality Management System SEMINARS: Pressure
including Measurement Uncertainty. http://www.mhforce.com/
Training/TrainingCourses Mar 4-8, 2019 Principles of Pressure Calibration. Phoenix, AZ.
Fluke Calibration. A five-day training course on the principles and
SEMINARS: Mass practices of pressure calibration using digital pressure calibrators
and piston gauges (pressure balances).The class is designed to focus
Mar 11-22, 2019 Mass Metrology Seminar. Gaithersburg, MD. on the practical considerations of pressure calibrations. https://
NIST. The Mass Metrology Seminar is a 2 week, "hands-on" us.flukecal.com/training
seminar. It incorporates approximately 30 percent lectures and
70 percent demonstrations and laboratory work in which the SEMINARS: Software
trainee performs measurements by applying procedures and
equations discussed in the classroom. https://www.nist.gov/pml/ Dec 3-7, 2018 TWB 1051 MET/TEAM® Basic Web-Based Training.
weights-and-measures/about-owm/calendar-events Fluke Calibration. This web-based course presents an overview of
how to use MET/TEAM® Test Equipment and Asset Management
Apr 29-May 9, 2019 Advanced Mass Seminar. Gaithersburg, Software in an Internet browser to develop your asset management
MD. NIST. This 9 day, hands-on mass calibration seminar system. You will learn a systematic approach to recording the
focuses on the comprehension and application of the advanced information you need to manage your lab assets routinely,
mass dissemination procedures, the equations, and associated consistently and completely. https://us.flukecal.com/training
calculations. https://www.nist.gov/pml/weights-and-measures/
about-owm/calendar-events Dec 10-14, 2018 TWB 1031 MET/CAL® Procedure Development
Web-Based Training. Fluke Calibration. Learn to create procedures
SEMINARS: Measurement Uncertainty with the latest version of MET/CAL, without leaving your office.
This web seminar is offered to MET/CAL users who need assistance
Dec 4, 2018 Introduction to Estimating Measurement writing procedures but have a limited travel budget. https://
Uncertainty. Port Melbourne, VIC. Australian NMI. This one-day us.flukecal.com/training
course (9 am to 5 pm) will give you a clear step-by-step approach
to uncertainty estimation with practical examples; you will learn Feb 11-15, 2019 TWB 1051 MET/TEAM® Basic Web-Based Training.
techniques covering the whole process from identifying the Fluke Calibration. This web-based course presents an overview of
sources of uncertainty in your measurements right through to how to use MET/TEAM® Test Equipment and Asset Management
completing the uncertainty budget. https://www.measurement. Software in an Internet browser to develop your asset management
gov.au/ system. You will learn a systematic approach to recording the
information you need to manage your lab assets routinely,
Dec 20-21, 2018 Applied Measurement Uncertainty for consistently and completely. https://us.flukecal.com/training
Calibration Laboratories (EMU 202). Frederick, MD. A2LA.
During this workshop, the participant will be introduced to Mar 11-15, 2019 MC-206 Basic MET/CAL® Procedure Writing.
several tools and techniques that can be applied in the calibration Everett, WA. Fluke Calibration. In this five-day course, you will
laboratory environment to efficiently and effectively create learn to configure MET/CAL software to create, edit, and maintain
measurement uncertainty budgets which comply with ISO/IEC calibration solutions, projects and procedures. http://us.flukecal.
17025 requirements. https://www.a2la.org/events com/training
Cal Lab: The International Journal of Metrology 10 Jul • Aug • Sep 2018
CALENDAR
Mar 18-22, 2019 Metrology.NET ® Advanced User Training. examine the ISO/IEC 17025:2017 requirements related to software.
Aurora, CO (Denver). Cal Lab Solutions, Inc. This 5-day training https://www.nist.gov/pml/weights-and-measures/about-owm/
will provide an understanding of Metrology.NET and demonstrate calendar-events
how it can be optimized for your lab, covering: hands-on creation
& configuration of a test project, development of device drivers, SEMINARS: Temperature
debugging techniques, creating and using resources in Metrology.
NET, and more! http://www.metrology.net/5-day-advanced-user/ Mar 5-7, 2019 Practical Temperature Calibration Training.
American Fork, UT. Fluke Calibration. Three day course loaded
Apr 8-12, 2019 MC-205 MET/TEAM ® Asset Management. with valuable principles and hands-on training designed to help
Everett, WA. Fluke Calibration. This five-day course presents calibration technicians and engineers get a solid base of temperature
a comprehensive overview of how to use MET/TEAM® Test calibration fundamentals. http://us.flukecal.com/training
Equipment and Asset Management Software in an Internet browser
to develop your asset management system. http://us.flukecal.com/ SEMINARS: Vibration
training
Feb 25-27, 2019 Fundamentals of Random Vibration and Shock
Apr 11, 2019 Software Verification and Validation, Part 1. NIST Testing. Equipment Reliability Institute. After this short course,
Webinar. 2 hour session on Software Verification and Validation will you will be able to measure vibration and shock, calibrate vibration
focus on the use of Microsoft Excel in calibration laboratories and and shock measurement systems, convert field measured data into
examine the ISO/IEC 17025:2017 requirements related to software. a test program, interpret vibration and shock test requirements,
https://www.nist.gov/pml/weights-and-measures/about-owm/ supervise vibration and shock tests, specify and experimentally
calendar-events evaluate vibration and shock test fixtures, perform ESS, HALT
and HASS. https://equipment-reliability.com/upcoming-training/
Apr 25, 2019 Software Verification and Validation, Part II. NIST
Webinar. 2 hour session on Software Verification and Validation will Visit www.callabmag.com for upcoming metrology
focus on the use of Microsoft Excel in calibration laboratories and events & webinars!
Jul • Aug • Sep 2018 11 Cal Lab: The International Journal of Metrology
INDUSTRY AND RESEARCH NEWS
Helping the Microchip Industry Go The exact amount of gas injected into measurements over time as gas fills a
(Very Low) with the Flow the chamber is critically important to collection tank through the MFC.
these processes and is regulated by a “Concerns about the accuracy of
August 22, 2018, NIST News - A device called a mass flow controller that technique came to our attention
new study by scientists at the National (MFC). MFCs must be highly accurate recently when a major manufacturer of
Institute of Standards and Technology to ensure that the deposited layers chip-fabrication equipment found that
(NIST) has uncovered a source of error have the required dimensions. The they were getting inconsistent results
in an industry-standard calibration potential impact is large because chips for flow rate from their instruments
method that could lead microchip with incorrect layer depths must be when they were calibrated on different
manufacturers to lose a million dollars discarded. RoR systems,” said John Wright of
or more in a single fabrication run. “Flow inaccuracies cause NIST’s Fluid Metrology Group, whose
The problem is expected to become nonuniformities in critical features members conducted the error analysis.
progressively more acute as chipmakers in wafers, directly causing yield Wright was particularly interested
pack ever more features into ever reduction,” said Mohamed Saleem, because for many years he had seen that
smaller space. Chief Technology Officer at Brooks RoR readings didn’t agree with results
The error occurs when measuring Instrument, a U.S. company that obtained with NIST’s “gold standard”
very small flows of exotic gas manufactures MFCs among other pressure/volume/temperature/time
mixtures. Small gas flows occur precision measurement devices. system. He and colleagues developed
during chemical vapor deposition “Factoring in the cost of running a mathematical model of the RoR
(CVD), a process that occurs inside a cleanrooms, the loss on a batch of wafers process and conducted detailed
vacuum chamber when ultra-rarefied scrapped due to flow irregularities can experiments. [https://www.nist.gov/
gases flow across a silicon wafer to run around $500,000 to $1,000,000. Add publications/errors-rate-rise-gas-
deposit a solid film. CVD is widely to that cost the process tool downtime flow-measurements-flow-work] The
used to fabricate many kinds of high- required for troubleshooting, and it conclusion: conventional RoR flow
performance microchips containing as becomes prohibitively expensive.” measurements can have significant
many as several billion transistors. CVD Modern nanofabrication facilities errors because of erroneous temperature
builds up complex 3D structures by cost several billion dollars each, and values. “The gas is heated by flow work
depositing successive layers of atoms it is generally not cost-effective for as it is compressed in the collection
or molecules; some layers are only a a company to constantly fine tune tank, but that is not easily accounted
few atoms thick. A complementary CVD and plasma etching. Instead, for: it is difficult to measure the
process called plasma etching also the facilities rely on accurate gas temperature of nearly stationary gas.”
uses small flows of exotic gases to flows controlled by MFCs. Typically, Wright and colleagues found
produce tiny features on the surface of MFCs are calibrated using the “rate that without corrections for these
semiconducting materials by removing of rise” (RoR) method, which makes temperature errors, RoR readings can
small amounts of silicon. a series of pressure and temperature be off by as much as 1 percent, and
perhaps considerably more. That might
not seem like a lot, but low uncertainty
is critical to attaining uniformity and
quality in the chip manufacturing
process. And the challenge is growing.
Current low-end flow rates in the
semiconductor industry are in the
range of one standard cubic centimeter
(1 sccm)—about the volume of a sugar
cube—per minute, but they will soon
shrink by a factor of 10 to 0.1 sccm.
Precise flow measurement is a
particularly serious concern for
manufacturing processes that use
etching of deposited layers to form
trench-like features. In that case, the
MFC is often open for no more than a
few seconds.
“A tiny amount of variation in the
Process Engineer Richard Kasica of NIST’s Center for Nanoscale Science and Technology flow rate has a profound effect on the
holds a wafer of the type typically produced in the plasma-enhanced chemical vapor
deposition chamber at center. Credit: Curt Suplee/NIST etch rate and critical dimensions of the
Cal Lab: The International Journal of Metrology 12 Jul • Aug • Sep 2018
INDUSTRY AND RESEARCH NEWS
structures” in very large-scale integrated a model uncertainty analysis that role in industry. To provide the
circuits, said Iqbal Shareef of Lam others can use to know what level of metrological infrastructure required,
Research, a company headquartered agreement to expect between MFCs PTB has been pursuing an approach
in California that provides precision calibrated on different RoR systems. that describes the dynamic behavior
fabrication equipment to microchip “NIST serves as a reliable reference of the force transducer by means of a
manufacturers. for knowledge and measurement mathematical model. The transducer
“So, we are extremely concerned where industry can assess agreement and the calibration device are modeled
about flow rates being accurate and between their systems,” Wright said. as a series arrangement of mass-
consistent from chamber to chamber “As manufacturers’ measurement springdamper elements. The mass, the
and wafer to wafer,” Shareef said. “Our needs push to ever lower flows, so will stiffness and the damping parameters
industry is already headed toward very NIST calibration standards.” of the force transducer are identified
small flow rates.” Note: Any mention or depiction of applying the model equation to
“We are talking about wafer commercial products is for information dynamic measurement data. The goal
uniformity today on the nanometer only; it does not imply recommendation is the general characterization of the
and even subnanometer scale,” Shareef or endorsement by NIST. dynamic behavior, irrespective of the
said. Source: https://www.nist.gov/news- particular measuring application or the
That’s very small. But it’s what events/news/2018/08/helping-microchip- type of force excitation. Calibrations
the complexity of three-dimensional industry-go-very-low-flow with shock or sine excitations should
chip manufacturing increasingly provide consistent parameters.
demands. Not so long ago, “a 3D Dynamic Calibration of Earlier tests using a high-bandwidth
integrated circuit used to have four Force Transducers force transducer did not give consistent
layers of metals,” said William White, results. A new model offers a thorough
Director of Advanced Technology at PTB-News 2.2018 - Good agreement explanation. The force transducer used
HORIBA Instruments Incorporated, between the results of different as a case study (measuring range ±1 kN,
a global firm that provides analytical calibration methods (shock or sine) has force introduction via two threaded
and measurement systems. “Now been achieved for the first time. This rods) was subjected to shock or sine
companies are regularly going to 32 was made possible by a mathematical forces. Intended variations of the
layers and sometimes to 64. Just this model of a measuring setup with the dynamic behavior could be realized
year I heard about 128.” And some force transducers elastically coupled by using additional load masses. The
of those chips have as many as 3,000 on both sides. pulse duration ranged from 0.1 ms to 1
process steps. The reliable measurement of ms. The maximum excitation frequency
“Each 300 mm wafer can cost up to dynamic forces plays an important was 30 kHz.
$400, and contains 281 dies for a die
size of 250 to 300 mm2,” Brooks’ Saleem
said. “Each die in today’s high-end
integrated circuits consists of about
three to four billion transistors. Each
wafer goes through 1 or 2 months
of processing that includes multiple
runs of separate individual processes,”
including chemical vapor deposition,
etch, lithography and ion implantation.
All those processes use expensive
chemicals and gases.
Many companies are already re-
examining their practices in light of
the NIST publication, which provides
needed theoretical explanations for
the source of RoR flow measurement
errors. The theory guides designers of
RoR collection tanks and demonstrates Comparison of measured and modeled resonant frequencies of the force transducer
easy-to-apply correction methods. investigated as a case study for dynamic calibrations with different load masses.
RoR theory shows that different The diagram shows the modeled resonant frequencies and the resonant frequencies
temperature errors will occur for the determined experimentally with shock and sine excitation as a function of the load
mass. The measurement point for1 kg represents a typical sinusoidal calibration with
different gases used in CVD processes.
large masses and excitations up to a few kHz. In addition, the resonances determined
The NIST publication also provides experimentally at shock excitations are also plotted. Credit: PTB
Cal Lab: The International Journal of Metrology 14 Jul • Aug • Sep 2018
INDUSTRY NEWS
2019 TECH
The force transducer has two dominant resonances whose
characteristics depend on the size of the coupled mass. In
the case of a typical shock excitation without load masses,
the lowest resonant frequency is caused by the vibrating
transducer housing, whereas it is due to the elastically
FORUM
March 28-31 | Reston, Virginia
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
coupled mass itself in the case of a sinusoidal excitation
with high mass values. The new model of three elastically Be part of one of the accreditation
coupled masses considers a two-sided elastic coupling of industry’s largest annual events.
the transducer. The resonant frequencies measured with the . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
different measurement setups were compared with those of
the model, and the stiffness parameters of the transducer Join attendees from over
were thus identified. 12 different industries,
The improved model-based dynamic calibration now
such as automotive,
provides consistent parameters from measurement data
with sinusoidal and pulse-shaped force excitation. This environmental,
proves the suitability of this new calibration methodology. pharmaceutical,
Complementary investigations with finite-element methods calibration, and more,
have confirmed these results. The dynamic measurement for one of the largest,
behavior of the force transducer can therefore be transferred multidiscipline events in
to a specific measurement application by extending the
the accreditation industry.
model correspondingly.
Contact
The A2LA Technical
Michael Kobusch, Department 1.7, Acoustics and
Dynamics, Phone: +49 531 592-1107, michael.kobusch(at) Forum promotes
ptb.de discussion and training
Scientific publication on topics related to the
M. Kobusch, S. Eichstädt: A case study in model-based accreditation industry
dynamic calibration of small strain gauge force transducers.
through face-to-face
ACTA IMEKO 6, 3–12 (2017)
Source: https://www.ptb.de/cms/en/presseaktuelles/journals- interactions with staff,
magazines/ptb-news.html members, assessors,
and accredited organizations.
EURAMET Publishes New Guidelines for
Vector Network Analyzers After you attend this
event, you will be
EURAMET News, April 5, 2018 – As the speed of digital informed and equipped
communications increases, radiofrequency and microwave
with practical, ready-to-
components must perform at ever higher data transmission
rates. Achieving increased speeds relies on the reliable implement solutions to
characterization of new component configurations during solve your organization’s
design development and prototype testing. Vector Network greatest accreditation and
Analyzers are used for this purpose. quality-related challenges.
The newly revised EURAMET calibration guideline no. 12
on the ‘Evaluation of Vector Network Analysers (VNA)’ gives
advice on how to measure the reflection and transmission
of guided electromagnetic waves with a vector network Purchase your tickets online at
analyzer in accordance with best measurement practices.
“Advances in VNA measurements over the last years www.A2LA.org/tech-forum
prompted the need for a revision of this important
calibration guide,” says Luca Callegaro (INRIM, Italy), chair or call 301-644-3248.
of EURAMET’s Technical Committee for Electricity and
Magnetism. “This guide is intended to become a reference for
all users aiming at traceable and accurate measurements with
vector network analyzers: National Metrology Institutes,
calibration laboratories and scientific and industrial research
centers.”
Jul • Aug • Sep 2018 15 Cal Lab: The International Journal of Metrology
INDUSTRY AND RESEARCH NEWS
Cal Lab: The International Journal of Metrology 16 Jul • Aug • Sep 2018
www.tegam.com
AUTOMATED RF POWER
SENSOR CALIBRATION
PS-Cal™ by Cal Lab Solutions
the newly developed DCC is based data and software and would thus A2LA Announces Acquisition of
on the internationally accepted and enable the measurement process to be WorkPlace Training
approved exchange format XML simulated. Physical weights already
(Extensible Markup Language), it have such digital twins which have August 27, 2018, Frederick, MD
is machine-readable; moreover all been successfully tested. This “digital and Wayzata, MN – A2LA and
indications, including the numerical weight” contains both information WorkPlace Training (WPT) today
calibration curves, can be directly from calibrations and estimations announced A2LA’s acquisition of
and automatically transferred into all concerning the expected behaviour WorkPlace Training. The addition of
digitally supported processes. At the of the weight under certain ambient WPT’s robust e-learning modules will
same time, cryptographic signatures conditions. complement A2LA’s wide-reaching
are used as security procedures to Contact instructor-led classroom training
guarantee that the integrity and the Siegfried Hackel, Project 1.01, offerings. E-learning modules offered
authenticity of a calibration certificate Digitalization in the Mechanics and by WPT include tutorials, practice,
is still ensured. The cryptographic Acoustics Division, Phone: +49 531 592- assessments and record keeping
procedures used for DCCs have 1017, siegfried.hackel(at)ptb.de on topics such as measurement,
proven their worth in other areas Scientific publication calibration and ISO/IEC 17025.
such as the civil register office S. Hackel, F. Härtig, J. Hornig, T. “Partnering with A2LA gives us
(“Standesamt”), waste management Wiedenhöfer: The digital calibration a unique opportunity to bring our
or in the purchasing department of certificate. PTB-Mitteilungen 127 (4), calibration and metrology training
the German federal administration. 75–81 (2017) to an even broader audience as a
The digital calibration certificate Source: https://www.ptb.de/cms/en/ part of their world-class accreditation
is already being further developed: a presseaktuelles/journals-magazines/ptb- and training portfolio,” said Paul
digital twin could contain even more news.html Hanssen, Founder of WorkPlace
www.entegra-corp.com
240-672-7645
Cal Lab: The International Journal of Metrology 18 Jul • Aug • Sep 2018
INDUSTRY AND RESEARCH NEWS
CAL-TOONS by
e x p aTed
n s i o n Green
of our training a n d comprehensive, relevant and current
accreditation services offerings,” m e a s u r e m e n t s c i e n c e t r a i n i n g
NOT SURE WHAT s t a t eTHE
d L o nAUDITORS
n i e S p i r e s , A 2WILL THINKSince their founding,
L A curriculum.
President/CEO. “Our customers WPT has provided training for
ABOUT A HEART-SHAPED have increasingly requested CAL STICKER…
A2LA to thousands of learners around the
provide its world-leading training in world.
e-learning format and we are making About A2LA
that a reality with the acquisition.” As the largest U.S.-based multi-
“ T h e a d d i t i o n o f W o r k Pl a c e discipline accreditation body, A2LA
Training. “We are excited to fortify Training completes a full spectrum offers an established infrastructure
our already long-term relationship of training offerings and delivery, engineered with over 35 years of
with A2LA and deliver our expertise from international standards to ISO standards development and use.
to reach even more technicians and measurements and from instructor- A2LA accreditation programs are
engineers.” led to e-learning, to bolster laboratory based on internationally-accepted
“A2LA is dedicated to improving c o n f o r m a n c e , c o m p e t e n c e , a n d criteria for competence. For over two
measurement results that impact confidence,” expressed Tim Osborne, decades, A2LA has also maintained
safety around the world, which Senior Director of Training Services. a world-class training program,
is supported by effective training “The values, vision and mission that providing a large spectrum of courses
f o r m e a s u r e m e n t a n d q u a l i t y WPT embraced for 30 years parallels designed to help organizations
assurance professionals. WorkPlace those of A2LA which results in an understand how to conform to
Training’s reputation for providing immediate win for the customer.” international standards. Find more
measurement-focused e-learning W P T w a s f o u n d e d i n 1 9 8 9 , information at www.A2LA.org.
t r a i n i n g m a k e s t h i s a n a t u r a l with the purpose of delivering a
Jul • Aug • Sep 2018 19 Cal Lab: The International Journal of Metrology
METROLOGY 101
Figure 1. Tektronix CSA 8000 with 80E02 sampler and Entegra Corporation Model 1150.
Cal Lab: The International Journal of Metrology 20 Jul • Aug • Sep 2018
METROLOGY 101
Frequency Response and Bandwidth While this procedure does work for many tests, it may
not always be correct. What if the oscilloscope’s frequency
The frequency response of an oscilloscope is the response has developed a dip at a frequency that is less
response of the oscilloscope to sinusoidal inputs, as a than f-3dB? This condition would not be discovered using
function of frequency. The 3 dB attenuation bandwidth, the method just described. Even if multiple measurements
f-3dB, or just bandwidth, is a key performance specification were made at different input frequencies (the “swept
of an oscilloscope, so much so that the bandwidth is frequency” method), as is done if the frequency response
printed on the front panel of every modern oscilloscope. is being measured, unless the input frequency is increased
The bandwidth is defined as the lowest signal frequency in small enough increments, any dips in the frequency
whose waveform amplitude is 3 dB less than the actual response may not be found. For example, if the frequency
input signal amplitude. response of a 500 MHz bandwidth oscilloscope contains a
dip at 435 MHz, this may not be discovered if the frequency
Bandwidth from Transition Duration - Rule increments are 25 MHz or even 10 MHz.
As will be explained in the next section, time domain
of Thumb Method measurements may provide the oscilloscope’s frequency
response with a fine frequency resolution from a single
A simple “rule of thumb” calculation has been used for
measurement.
many years to estimate f-3dB from the transition duration.
The equation is: The Fourier Transform – From Time to
ƒ −3dB= 0.35/τ Eq. 1
Frequency and Back
where τ is the 10 % to 90 % transition duration (the Some digital oscilloscopes include embedded software,
reference levels are given as a percentage of amplitude). the Discrete Fourier Transform (DFT), that will calculate
This equation is derived by assuming a frequency the frequency spectrum of the acquired waveform.
response waveform with a Gaussian roll-off. Although LabView, Matlab, and other software packages also
a Gaussian waveform is noncausal, this relationship include the DFT. The Fourier Transform describes the
gave a reasonable estimate until the advent of digital relationship between a time-domain representation
oscilloscopes, which have different frequency response of a signal and its frequency-domain representation
roll-offs. The major oscilloscope manufacturers each (or spectrum). If you are not familiar with the Fourier
advocate using a different number in place of the “0.35” Transform, there are several good texts on the subject [3].
[2]. For example, Tektronix advocates using “0.45” and Data that is a function of time (i.e. voltage vs. time or
Keysight Technology “0.50.” current vs. time) is called time-domain data. Similarly,
Though useful as a crude “Rule of Thumb” estimator, data that is a function of frequency is termed frequency-
Equation 1 is too inaccurate to be used in metrology, so, domain data. The Fourier Transform converts time-
a more rigorous method is the subject of this article. domain data into frequency-domain data and the Inverse
Fourier Transform converts frequency-domain data into
Common Bandwidth Measurement time-domain data. The Fourier Transform assumes that
Method the duration of the time-domain data, the waveform
epoch, is an integer number of cycles of a repetitive signal.
Some metrologists calibrate oscilloscopes using two This means that the first sample and the last sample of the
single frequencies to verify the oscilloscope’s bandwidth. waveform must have the same amplitude.
They do this because they can use a multi-product An important property of the Fourier Transform is that
calibrator as follows. A reference, low frequency sinewave the duration of the waveform (its epoch) will determine
is applied to the oscilloscope input channel (a 50 kHz the frequency resolution, Δf, of the frequency-domain
sinewave for example). The amplitude of the acquired data (the spectrum). The longer the waveform epoch,
waveform is assumed to be the same as the calibrated the smaller will be Δf. The Fourier Transform provides
amplitude of the input signal and is recorded. A second a complex (real and imaginary) spectrum that can be
sinewave with the same calibrated amplitude as the converted to a magnitude and phase spectra. We will only
reference frequency but at the same frequency as the consider the magnitude here.
oscilloscope bandwidth is next applied to the input of the
oscilloscope. The amplitude displayed by the oscilloscope Step-like Waveform vs. Impulse-like
is again measured. If the measured amplitude of this Waveform
second, high frequency sinewave is equal to or greater
than 70.7 % of the reference frequency amplitude, the A step-like waveform has an upper and lower state
oscilloscope bandwidth specification is considered met. whereas an impulse-like waveform has only one state (and
Jul • Aug • Sep 2018 21 Cal Lab: The International Journal of Metrology
METROLOGY 101
is the derivative of a corresponding step-like waveform). resolution are limited by the oscilloscope hardware,
There are various advantages and disadvantages to using obtaining a full cycle of the step-like waveform while
either an impulse-like waveform or step-like waveform having a sufficient number of samples in the transition
when determining the impulse response or frequency region may not be possible. Over the years, many
response of an oscilloscope. different ways of dealing with the discontinuity in
For many years, metrologists in the national metrology amplitude between the first and last waveform samples
institutes preferred to use step-like waveforms from have been proposed [5]. An intuitively simple method
calibrated pulse generators, this was primarily for two is the Nahman-Gans Extension. The acquired step-like
reasons. The first is that waveform averaging can be used waveform is extended by inverting a replica of the step-
to reduce signal noise without impacting the amplitude, like waveform and offsetting its level by adding to this
even in the presence of jitter. After averaging, the replica the value of the sum of the amplitudes of the first
amplitude of a step-like waveform can be more accurately and last samples, and then appending this inverted and
computed, this is done using a histogram or other method offset waveform to the original waveform. The resulting
[4]. Once the amplitude is determined, the 10 % and waveform has twice as many samples as the original but
90 % reference levels can be found, the 10 % and 90 % the first and last samples have identical amplitudes. Since
reference level instants determined, and the transition only the measured data is used, no new waveform features
duration calculated. or additional frequency content are introduced. However,
This option is difficult for an impulse-like waveform, a step-like waveform with a duration of N/2, where N is
signal averaging in the presence of jitter will reduce the the total number of samples, is created because of the noise
amplitude and as a consequence, the transition duration in the first and last samples, and this will introduce an
will increase. Therefore, other methods such as a peak error in the spectrum that is proportional to the waveform
detection method must be employed. One method is to noise. The primary draw-back to this method is that N
fit a parabola to the peak of the impulse [4]. is doubled so all the numerical processing takes twice as
The second reason that a step-like waveform is long to compute. Also, because of this unique process,
preferred over an impulse-like waveform is the impact of all even harmonics in the spectrum are zero.
oscilloscopes with different bandwidths on the measured Another effective method of dealing with this problem
waveform amplitude. For an impulse-like waveform, both is to convert the step-like waveform to an impulse-like
the amplitude and transition duration are dependent waveform by taking the first difference of the waveform
on the oscilloscope bandwidth. In fact, an impulse data. No new frequency structure results from this
can be too fast with too little energy in the frequencies calculation. However, because of noise in the waveform,
within the oscilloscope bandwidth. If the oscilloscope the values of the first and last waveform samples in the
bandwidth is too low, the impulse amplitude may be so difference waveform may not be equal. This discontinuity
small as to be undetectable. In this case, another longer, will still introduce a small error in the frequency response
calibrated impulse generator must be used. This does not spectrum because of assumed cyclic convolution in the
happen with a calibrated step-like pulse generator. The DFT.
amplitude of the step-like waveform remains constant for We recommend using a step-like waveform, determine
oscilloscopes with different bandwidths. the pulse parameters of amplitude and transition
Of course, the measured transition duration from duration, form an impulse-like waveform using the first
either a calibrated impulse-like or calibrated step-like difference method, and calculate the Fourier Transform
pulse generator will be longer when measured with to determine the frequency response and f -3dB of the
an oscilloscope with lower bandwidth, but only one oscilloscope. These computations presume that the
calibrated step-like pulse generator is required for a oscilloscope bandwidth to be 10 times less than that of
wide range of oscilloscope bandwidths. It is important the pulses input into the oscilloscope. If this is not the
to note that the impulse response is the derivative of the case, additional considerations are necessary, which is
step response function. It is also important to note that discussed in the next section.
all transition duration values mentioned herein use 10 %
to 90 % amplitude values. Convolution Effects
Fourier Transform of Step Response There is another effect that must be considered to obtain
accurate measurement results and that is convolution.
As previously mentioned, the Fourier Transform When a signal generator output is measured using any
requires that the amplitudes of the first and last samples waveform recorder like an oscilloscope, the result is a
be identical. A step-like waveform does not meet this convolution of the finite impulse responses of the two
requirement unless an integer multiple of complete cycles instruments. Suppose that the output of a step-like pulse
are acquired. Since the number of samples and the time generator with a short transition duration and wide
Cal Lab: The International Journal of Metrology 22 Jul • Aug • Sep 2018
METROLOGY 101
bandwidth, fg, is used to determine the step response “Computation of waveform parameter uncertainties” [6]
and 3 dB analog bandwidth, fs, of a fast-response (high- and in “Pulse parameter uncertainty analysis” [7]. One
speed) oscilloscope. The acquired waveform is described component of the uncertainty in the bandwidth may be
as a mathematical convolution of the pulse output determined by using the expanded uncertainty, uτ, of τ to
from the pulse generator and the impulse response of represent a set of step-like waveforms that have a range
the oscilloscope. The -3 dB analog bandwidth of the of transition durations equal to τ - uτ to τ + uτ, where
acquired waveform, fsg, may be approximated by using these step-like waveforms are generated numerically.
a Gaussian approximation of fg, fs, and fsg and the “root The DFT of each waveform in this set is then calculated, a
sum of squares” (RSS) method to relate the oscilloscope bandwidth for each spectrum obtained, and the standard
bandwidth fs and the pulse generator bandwidth fg: deviation of the resultant set of bandwidths is used as the
primary contributor to the uncertainty.
(√ 1 ) ( 1 )
___________
Four different pulse generators and four different
1
___ 2 2
+ ___
= __ Eq. 2 sampling oscilloscopes and samplers were used to
ƒsg ƒs ƒg
examine the variation in fsg as the transition durations
A numerical example is helpful to illustrate the were varied. The initial waveforms were acquired and
magnitude of the effect of convolution. Suppose that then a set of waveforms were created from each of them
by adjusting the sampling intervals to obtain the desired
fs = 2 GHz and that fg = 2 GHz, the result of Equation 2
variation in τ. The waveform epoch was altered by this
is fsg = 1.414 GHz. That is, the acquired waveform has process as well, which, as mentioned previously, affects
less bandwidth than 2 GHz even though the bandwidth the frequency interval or resolution, Δf.
of both the oscilloscope and pulse generator is 2 GHz. Figure 2 (on the following page) depicts the five initial
Deconvolution of the pulse waveform from the acquired waveforms and the results are presented in Table 1. The
waveform must be used in this case to get an accurate
percent variation in fsg very closely followed the percent
estimate of the bandwidth of the oscilloscope or a
variation in τ. All acquired waveforms used a 2 ns epoch.
calibrated pulse generator producing pulses with faster
The number of samples in the waveform for the Tektronix
transition duration must be found. For example, if fg = 12 067-1338-00 #2 was 1000, for the Tektronix 067-1338-00
GHz (like an Entegra Model 1012 High Speed Reference #1 and Fluke 9500B-9550 was 1024, and for the Entegra
Pulse Generator), the effects of convolution, estimated by 1150 was 4096. The same Entegra Corporation Model
Equation 2 would be greatly reduced; instead of fsg = 1.414 1150 Reference Pulse Generator was used to generate
GHz, it would be 1.973 GHz or less than 1.4 % error (Eq. 3). both waveforms depicted in Figure 2, only the sampling
___________ oscilloscopes acquiring the waveforms were different.
1
√ ( ) ( )
1 2 1 2
___
+ ___
= __ Eq. 3 The Agilent 86118A has a fs ≈ 70 GHz and the Agilent
ƒsg 2 12 83484A has a fs ≈ 50 GHz. In Table 1, the differences in
transition duration and bandwidth are noted.
If pulses with even shorter transition duration were We also examined the change in fsg as the transition
used, like those provided by an Entegra Model 1120, with occurrence instant changed position in the waveform
fg =18 GHz, the result is fsg = 1.988 GHz or about 0.6 % error. epoch. In Figure 2, all of the transition occurrence
Figure 1 (page 20) depicts a Tektronix CSA8000 with a instants are at 400 ps or 20 % of the epoch. There was
80E02 sampling head (12.5 GHz bandwidth) measuring only a very small change (less than 0.06 ps) when the
the output of an Entegra Corporation Model 1150 (50 transition occurrence instant was delayed or advanced
GHz bandwidth). There is a 1.85 mm to 3.5 mm adapter up to 15 % (100 ps to 550 ps from the first sample
between the pulse generator output and the sampling of the waveform). Consequently, this effect can be
head input. If a coaxial cable, connector adapters, or neglected. Once uτ is determined, its contribution to the
terminations are used to connect the pulse generator to the
measurement uncertainty, fsg, expressed as a percentage,
oscilloscope, the effects of their bandwidths must also be
can be determined. These percentages were observed to
included by inserting additional terms inside the square
be approximately the same.
root in Equation 2. Please note that Equation 2 may also
As previously mentioned, the duration of the time
be used with transition durations by replacing 1/f terms
domain waveform, i.e. the waveform epoch, determines
with τ (transition duration) terms.
the frequency resolution, Δf of the frequency response
calculated using the Fourier Transform. This discrete
Uncertainty Analysis Δf is another contributor to the uncertainty in the
frequency response and bandwidth. The effect of Δf on
The uncertainty analysis for step-like and impulse- the uncertainty is handled similarly to how the time
like waveform parameters is covered by IEC 62754:2017 resolution affects the uncertainty in the transition duration
Jul • Aug • Sep 2018 23 Cal Lab: The International Journal of Metrology
METROLOGY 101
Figure 2. Waveforms acquired using four different pulse generators and four samplers (volts vs. seconds).
[7]. Since the probability of finding the -3 dB point on the affect the equality. The situation when the values of the
frequency response is uniformly distributed across the first and last samples are unequal is treated as an error and
interval f-3dB - Δf to f-3dB + Δf, the measurement uncertainty the waveform is corrected for this error before calculating
is 0.577 Δf. the Fourier Transform. Due to noise, measurement
Another contribution to the uncertainty is the waveform uncertainties are associated with this error correction.
epoch not being an integer number of signal cycles. Using There may be other measurement uncertainty
the Nahman-Gans, First Difference, or other method to contributors to the transition duration and thus the
address this situation may still result in a small difference frequency response that you may decide to include. For
in the amplitude between the first and last samples in the example, if the temperature in your lab is not very stable,
waveform. This will effectively introduce an additional both the oscilloscope step response and pulse generator
ideal step-like (square-wave) waveform to the measured output may vary [8].
signal and, thus, to its spectrum. This additional square- In keeping with NIST and the GUM recommendations
wave spectrum is more noticeable in the lower-frequency [9], the various uncertainty components examined above
parts of the spectrum as this is where it has the greatest are combined using the RSS formula similar to Equation
power. This effect is particularly noticeable on the acquired 2 above to yield the combined uncertainty. The expanded
waveform when using an impulse-like signal as the input uncertainty is then calculated by multiplying the combined
to the oscilloscope. Rarely are the values of the first and uncertainty by a coverage factor, usually 2 for a 95 % level
last samples equal for an impulse-like signal unless a full of confidence. The expanded uncertainty is the value
cycle is acquired. However, even in this case, noise will usually reported to your customer.
Cal Lab: The International Journal of Metrology 24 Jul • Aug • Sep 2018
METROLOGY 101
Transition
Bandwidth % Change
Pulse Generator Sampler (ƒs) Duration
(ƒsg) in τ % Change in ƒsg
(τ )
Agilent 86118A
Tektronix 067-1338-00 #1 12.56 ps 23.05 GHz ±2.5% +2.43 % to -2.56%
(70 GHz)
Tektronix 80E01
Tektronix 067-1338-00 #2 15.06 ps 21.46 GHz ±2.5% +2.42 % to -2.56%
(50 GHz)
Agilent 86118A
Entegra 1150 #1 9.50 ps 49.42 GHz ±2.5% +2.43 % to -2.57%
(70 GHz)
Agilent 83484A
Entegra 1150 #1 11.33 ps 48.83 GHz ±2.5% +2.44 % to -2.56%
(50 GHz)
Tektronix SD32
Fluke 9500B- 9550 28.36 ps 10.60 GHz ±2.5% +2.45 % to -2.55%
(50 GHz)
Table 1. Variation in Transition Duration and Resulting Variation in Bandwidth
Jul • Aug • Sep 2018 25 Cal Lab: The International Journal of Metrology
In Focus: Colorado Engineering
Experiment Station (CEESI)
Sita Schwartz
Editor
The Colorado Engineering Experiment Station, Inc. (CEESI), a subsidiary of Western Energy Support & Technology, Inc.
(WEST), provides accredited NIST traceable calibration for numerous types of flow meters for customers around the world
in the energy and aerospace industries. With roots going back to the University of Colorado in Boulder, they also engineer
flow testing systems, participate in inter-laboratory comparisons, and sponsor training and flow measurement events.
Cal Lab: The International Journal of Metrology 26 Jul • Aug • Sep 2018
In Focus: Colorado Engineering Experiment Station (CEESI)
Sita Schwartz
History
The Engineering Experiment Station began as a
program, run by the College of Engineering at the
University of Colorado (CU) in Boulder from 1951
to 1965 “for research, development, and testing of
small rocket nozzles. In addition, the Station became
the sole location for testing gas turbine meters as the
military began to use these meters [1].” The program
produced flow measurement systems and allowed
researchers to participate in the development of high
pressure gas flow standards and critical flow venturis.
The site in Nunn was originally one of 27 testing
sites for the Atlas E nuclear missiles. It was assigned
to the 566th Strategic Missile Squadron of F. E.
Warren AFB in Cheyenne, Wyoming [2]. The facility
served as a surplus missile silo for two short years,
becoming obsolete when the Air Force shifted from
Figure 1. Outside the missile bay, originally built for an Atlas E Missile, liquid to solid fueled propulsion. The military was
now used for flow meter testing. motivated to unload the sites for repurposing and
the underground facilities have since been used for
Jul • Aug • Sep 2018 27 Cal Lab: The International Journal of Metrology
In Focus: Colorado Engineering Experiment Station (CEESI)
Sita Schwartz
homes, document storage, machine shops, etc. But the engineer tests that utilize a wide variety of gases and fluids,
underground tunnels and staging areas also had great including replicating multiphase conditions found in oil
potential for flow testing. So, in 1966, Dr. Tom Arnberg and gas production. Most of the meters are calibrated in a
helped to separate the Engineering Experiment Station from test area known as the “Mechanical and Electrical” room
CU and move it to the decommissioned missile site. The by the Air Force crew. Located nearby are the primary flow
non-profit also went through a name change to Colorado standards and metrology lab for pressure, temperature,
Engineering Experiment Station, Inc. (CEESI). viscosity and density calibrations. The original “Missile
“During the late 1960’s and early 1970’s, the facility Bay” is now used to calibrate larger meters operating at
became home to two new high pressure primary flow higher flowrates, while the largest meters are calibrated in
standards, identified as ‘Primary A’ [volumetric] and an outdoor test area.
‘Primary B’ [gravimetric], which are still in use today. [1]” The largest test area is the Wet Gas/Multiphase facility.
In 1986, CEESI changed hands when it was purchased by From the second floor of the Wet Gas Building, an oversized
Walt Seidl and Steve Caldwell, both of whom had been picture window looks out over an expansive, rectangular
quite involved with the company already. Incorporated shaped gas loop. Rich Schoonover, who was recently
as a for-profit, commercial calibration facility, CEESI appointed President of Corporate Operations, explained the
has since expanded and grown. Where there was just an operations cost of keeping the flow of gas running through
underground missile bunker and tunnels, there are now the loop runs at $10,000 a month.
numerous test areas, storage tanks, and buildings spread Together with the Northern Plains Natural Gas Company
out over the facility. and several flow meter vendors, a high flow natural gas
calibration facility was completed in Garner, Iowa in 1999.
Capabilities This facility fulfills a need for ultrasonic meter calibrations
traceable to standards in the United States. For those not
CEESI is a NIST traceable, primary and secondary flow familiar with pressure and flow measurement (like me),
meter calibration facility and one of the largest in North ultrasonic meters are vital to the natural gas industry by
America, serving customers around the world. They can being able to measure the velocity of flow through a pipe.
Cal Lab: The International Journal of Metrology 28 Jul • Aug • Sep 2018
In Focus: Colorado Engineering Experiment Station (CEESI)
Sita Schwartz
Jul • Aug • Sep 2018 29 Cal Lab: The International Journal of Metrology
In Focus: Colorado Engineering Experiment Station (CEESI)
Sita Schwartz
“is a pumped continuous system allowing for thermal company with over 50 years of leadership within the flow
equilibrium and continuous operation without liquid calibration industry. At its locations in Nunn, Colorado,
[nitrogen] loss [3].” CEESI worked closely with NIST in and Garner, Iowa, CEESI performs NIST-traceable primary
order to meet operational criteria and match data from the and secondary calibrations on numerous types of meters
rebuilt system to historical data [4]. CEESI.com has a lot of and valves, using a wide range of flow rates, pressures,
great photos of the cryogenic lab in action, so I encourage temperatures, and test gases.
anyone interested in this type of
extreme temperature measurement
to check out their website. In fact,
CEESI’s website is a goldmine of
information—the more you look, the
more you find about their facility and
flow calibration.
Corporate Info
CEESI has undergone further
changes in recent years. Besides
the addition of the Cryogenic and
Liquid Hydrocarbon Flow calibration
facilities, CEESI has seen a change
of hands at the management level.
Rich Schoonover was appointed
President of Corporate Operations
in November of 2017. Rich joins a Figure 7. Shot of the flow software interface from the control room.
Cal Lab: The International Journal of Metrology 30 Jul • Aug • Sep 2018
In Focus: Colorado Engineering Experiment Station (CEESI)
Sita Schwartz
Works Cited
CEESI Snapshot
[1] “Our History,” Colorado Engineering Experiment
Station, Inc., [Online]. Available: http://ceesi.com/our- • Locations: Nunn, Colorado and Garner, Iowa, USA
history. [Accessed 16 May 2018]. • 1951 Established as the Engineering Experiment
[2] “Atlas Missile Silo Home Page,” [Online]. Available: Station by the University of Colorado
http://www.atlasmissilesilo.com/. [Accessed 16 May
• 1986 Incorporated as an independent, commercial
2018].
facility
[3] CEESI, “Cryogenic Flow Calibration Facillity,” [On-
line]. Available: http://ceesi.com/cryogenic-services. • Founding members: Walt Seidl and Steve Caldwell
[Accessed 15 June 2018]. • Number of employees: 40+
[4] T. Kegel, “The NIST/CEESI Liquid Nitrogen Flow Fa- • Parent Company: Western Energy Support &
cility,” in European Flow Measurement Workshop, Technology, Inc. (WEST)
Metrology for LNG Workshop, Noordwijk, Nether- • Calibration services: Flow meters, including:
lands, April 2017. critical flow venturis/sonic nozzles, ultrasonic
flowmeters, orifice meters, differential meters, and
Acknowledgments turbine meters.
• Accreditation: NVLAP
Many thanks to Jason Tator for his time and patience,
• Sponsored events and training: International
to Tom Kegel for his help editing, and Rich Schooner for
Symposium on Fluid Flow Measurement (ISFFM),
his hospitality.
North American Fluid Flow Measurement Council,
Ultrasonic Meter Seminar
Jul • Aug • Sep 2018 31 Cal Lab: The International Journal of Metrology
INDUSTRY
NEW PRODUCTS
AND RESEARCH
AND SERVICES
NEWS
Keysight Technologies Infiniium UXR-Series • sampling rates of 256 GSa/s per channel on 40 to 110 GHz
Oscilloscopes models and 128 GSa/s per channel on 13 to 33 GHz models,
creating accurate reconstruction of high-speed signals;
SANTA ROSA, Calif., September 18, 2018 – Keysight • up to four full-bandwidth channels with less than 35 fs (rms)
Technologies, Inc. (NYSE: KEYS), a leading technology of inter-channel intrinsic jitter providing accurate timing and
company that helps enterprises, service providers, and skew measurements;
governments accelerate innovation to connect and secure
the world, announced the new Infiniium UXR-Series of • an optional full factory grade self-calibration module which
oscilloscopes, with models ranging from 13 to 110 GHz of true ensures ongoing measurement accuracy while reducing the
analog bandwidth, offering industry leading signal integrity, need to take the unit out of service,
and investment protection that meets the needs of technology • a chipset based on a Keysight-proprietary Indium Phosphide
advancements today and tomorrow. (InP) process that enables exceptionally wide bandwidth and
Keysight’s new Infiniium UXR-Series enables high-speed extremely low noise floor.
serial and optical designers, focused on current and next Keysight also announced two additional solutions that, when
generation technologies, to quickly create comprehensive paired with the company’s Infiniium UXR-Series oscilloscopes,
designs with higher margins, thereby reducing time to offer a full end-to-end solution from stimulus to analysis for
market of their innovation. Superior performance levels, PAM4 and 400G, 600G, as well as terabit coherent optical
combined with a wide selection of bandwidth choices, make interconnect designs. These solutions include:
the Infiniium UXR-series the ideal solution for engineers and
• N4391B Optical Modulation Analyzer (OMA) –a compact, real-
designers working with any generation of DDR, USB, PCIe or
time oscilloscope-based OMA designed specifically for complex
other serial technologies, as well as PAM4, 5G, radar, satellite
optical data transmission and terabit measurement challenges.
communications, and optical designs.
“Today’s electronic and communications systems have • M8194A 120 GSa/s arbitrary waveform generator (AWG)
become more complex requiring next generation test and –Keysight’s fastest AWG delivers a new level of stimulus
measurement,” said Jessy Cavazos, Industry Director, Test performance for generating challenging formats such as 64
& Measurement, Frost & Sullivan. “Keysight’s deep heritage GBaud 64QAM (quadrature amplitude modulation) and other
and extensive expertise in test and measurement is clearly wideband modulation schemes.
reflected in the new UXR series of oscilloscopes, which deliver “High-speed serial and optical technologies are continually
next generation innovations to address the rapidly evolving pushing speed and performance boundaries as the development
measurement and compliance needs in the high-speed digital, of terabit and 5th generation serial technologies accelerate,” said
terabit and optical research, 5G, RF, and aerospace and defense Dave Cipriani, Keysight Vice President, Digital and Photonics
markets.” Center of Excellence. “The expanded UXR oscilloscope series,
Keysight’s new Infiniium UXR-Series delivers a low noise the UXR-based OMA, and Keysight’s fastest AWG, collectively
floor and high vertical resolution to ensure measurements are offer high-speed designers an unprecedented end-to-end
not impacted by oscilloscope noise and signals are represented solution to rapidly advance their technology development
accurately. As a result, eye diagrams are significantly more today, while providing investment protection into the future.”
open, and true margins and performance can be determined. Additional Information
Keysight Infiniium UXR-Series 13 to 110 GHz oscilloscope
• More information about the Keysight Infiniium UXR Series real-
models enable designers to accelerate time to market of their
time oscilloscopes is available at www.keysight.com/find/UXR.
innovation, with:
• 10-bit vertical resolution and industry leading signal • More information about the Keysight N4391B OMA is available
integrity for superior effective number of bits (ENOB), at www.keysight.com/find/oma_backgrounder.
leading to faster compliance testing with higher margins; • More information about the Keysight M8194A AWG is available
at www.keysight.com/find/awg_backgrounder.
Cal Lab: The International Journal of Metrology 32 Jul • Aug • Sep 2018
INDUSTRY
NEW PRODUCTS
AND RESEARCH
AND SERVICES
NEWS
Jul • Aug • Sep 2018 33 Cal Lab: The International Journal of Metrology
NEW PRODUCTS AND SERVICES
Cal Lab: The International Journal of Metrology 34 Jul • Aug • Sep 2018
NEW PRODUCTS AND SERVICES
Jul • Aug • Sep 2018 35 Cal Lab: The International Journal of Metrology
AUTOMATION CORNER
No Faking Automation
Michael Schwartz
Cal Lab Solutions, Inc.
Automation has its own quality process, like the input impedance on theory. Even if you have detailed step
system automatically built in. When the 3458A can cause the calibrator to by step instructions, it’s difficult to
software performs the same test, the trip. Setting up a calibrator for say code anything without understanding
same way over and over again, it can’t 1,000 V, then resetting the 3458A, and the measurement process and the
be wrong. There is no fake it till you putting it into auto range; that trips the metrology behind the test.
make it. You can’t just put a sticker calibrator. That breaks the automation! The hardest part is when there is no
on it and hope nobody checks. The It is helpful when the manufacturer written test procedure (our company
10 * Log 2
THD
P12
P32
P2 ...Px2
life of automation is longer than a has a well-defined RESET state of the charges more for this because we have
V1
I2 IR
20 * Log 2
THD 2 P P I
V
V 3 ...V x
2 2 2 E P
single calibration. Users can always hardware with all of the default settings. to first write some sort of test process).
2
I
E R E PR
look at the measurements made by The worst is when a manufacturer If there is no known procedure, or
R I P EI
P
E
R R I R
automation and question the results. allows the end user to configure a something similar, then the automation
P 2
E2 E
Especially if something looks wrong or custom default state because now the needs to be written based on the
y yocto 1e^-24 kilo k 1e^3
z zepto 1e^-21 mega M 1e^6
a atto 1e^-18 giga G 1e^9
they are getting contradicting manual state is largely an unknown. And when calibration results from the factory or
f femto 1e^-15 tera T 1e^12
p pico 1e^-12 peta P 1e^15
n nano 1e^-9 exa E 1e^18
measurements. the equipment state is unknown, the from the equipment specifications. In
µ micro 1e^-6 zetta Z 1e^21
m milli 1e^-3 yotta Y 1e^24
Cal Lab: The International Journal of Metrology 36 Jul • Aug • Sep 2018
Better Software
Better Metrology
www.metrology.net
www.additel.com