NAFL INST. OF STAND & TECH R.I.C.

N!ST
isurement Services: REFERENCE
AlllOS 3SMmE PUBLICATIONS;

NIST Multifunction
Calibration System

NIST
Special
Publication
250^6

Nile Oldham and Mark Parker

U.S.Department of Commerce
Technology Administration
U57 National Institute of Standards and Technology
0.250-46
I

'

998
 #he National Institute of Standards and Technology was established in 1988 by Congress to "assist industry in

the development of technology . . . needed to improve product quality, to modernize manufacturing processes, to

ensure product reliability . . . and to facilitate rapid commercialization ... of products based on new scientific

discoveries."
NIST, originally founded as the National Bureau of Standards in 1901, works to strengthen U.S. industry's
competitiveness; advance science and engineering; and improve public health, safety, and the environment. One
of the agency's basic functions is to develop, maintain, and retain custody of the national standards of

measurement, and provide the means and methods for comparing standards used in science, engineering,
manufacturing, commerce, industry, and education with the standards adopted or recognized by the Federal
Government.
As an agency of the U.S. Commerce Department's Technology NIST conducts basic and
Administration,
applied research in measurement techniques, test
the physical sciences and engineering, and develops
methods, standards, and related services. The Institute does generic and precompetitive work on new and
advanced technologies. NIST's research facilities are located at Gaithersburg, MD 20899, and at Boulder, CO 80303.
Major technical operating units and their principal activities are listed below. For more information contact the
Publications and Program Inquiries Desk, 301-975-3058.

Office of the Director Physics Laboratory

• National Quality Program • Electron and Optical Physics
• International and Academic Affairs • Atomic Physics
• Optical Technology
Technology Services • Ionizing Radiation
• standards Services • Time and Frequency'
• Technology Partnerships • Quantum Physics'
• Measurement Services
• Technology Innovation Materials Science and Engineering
• Information Services Laboratory
• Intelligent Processing of Materials
Advanced Technology Program • Ceramics
• Economic Assessment • Materials Reliability'
• Information Technology and Applications • Polymers
• Chemical and Biomedical Technology • Metallurgy
• Materials and Manufacturing Technology • NIST Center for Neutron Research
• Electronics and Photonics Technology
Manufacturing Engineering
Manufacturing Extension Partnership Laboratory
Program • Precision Engineering
• Regional Programs • Automated Production Technology
• National Programs • Intelligent Systems
• Program Development • Fabrication Technology
• Manufacturing Systems Integration
Electronics and Electrical Engineering
Laboratory Building and Fire Research
• Microelectronics Laboratory
• Law Enforcement Standards • structures
• Electricity • Building Materials
• Semiconductor Electronics • Building Environment
• Electromagnetic Fields' • Fire Safety Engineering
• Electromagnetic Technology' • Fire Science
• Optoelectronics'
Information Technology Laboratory
Chemical Science and Technology • Mathematical and Computational Sciences^
Laboratory • Advanced Network Technologies
• Biotechnology • Computer Security
• Physical and Chemical Properties^ • Information Access and User Interfaces
• Analytical Chemistry • High Performance Systems and Services
• Process Measurements • Distributed Computing and Information Services
• Surface and Microanalysis Science • Software Diagnostics and Conformance Testing

'At Boulder, CO 80303.

^Some elements at Boulder, CO.
NIST Special Publication 250-46

NIST MEASUREMENT SERVICES:

NIST Multifunction Calibration System

Nile Oldham and Mark Parker

Electronic Instrumentation and Metrology Group

Electricity Division
Electronics and Electrical Engineering Laboratory
National Institute of Standards and Technology
Gaithersburg, MD 20899-0001

February 1 998

U.S. Department of Commerce

William M. Daley, Secretary

Technology Administration
Gary R. Bachula, Acting Under Secretary for Technology

National Institute of Standards and Technology

Raymond G. Kammer, Director
 National Institute of Standards and Technology Special Publication 250-46
Natl. Inst. Stand. Technol. Spec. Publ. 250-46, 28 pages (Feb. 1998)
CODEN: NSPUE2

U.S. GOVERNMENT PRINTING OFFICE

WASHINGTON: 1998

For sale by the Superintendent of Documents, U.S. Government Printing Office, Waslnington, DC 20402-9325
 TABLE OF CONTENTS

1. INTRODUCTION 2

2. DESCRIPTION OF THE MCS 3

2.1 MCS Hardware 3
2.2 MCS Software 4

3. CHARACTERIZING THE MCS 6

3.1 DC Voltage (DC V) 6
3.2 DC Current (DCI) 8
3.3 DC Resistance (DCR) 10
3.4 AC Voltage (AC V) 12
3.4.1 1 mV to 1 kV from 10 Hz to 1 MHz
Voltages from 12
3.4.2 Voltagesfrom 1 mV to 250 mV from 10 Hz to 1 MHz 14
3.4.3 Voltage from 250 mV to 7 V from 10 Hz to 30 MHz 15
3.4.4 Voltages from 1 mV to 1 kV from 0.01 Hz to 1 kHz 16
3.5 AC Current (ACI) 17
3.5.1 Currents from 2 mA to 10 A from 10 Hz to 100 kHz 18
3.5.2 Currents from 2 mA to 10 A from 0.01 Hz to 1 kHz 20

4. CALIBRATIONS AND TESTS USEslG THE MCS 21

4.1 Calibration of DMMs Using the MCS Calibrator 21
MCS DMM
4.2 Calibration of Calibrators using the 23
4.3 High Accuracy Special Test of DMMs and Calibrators 24
4.4 Special Test of Low Voltage Thermal Voltage Converters 24
4.5 Special Test of Micropotentiometers 24

5. QUALITY CONTROL AND UNCERTAINTIES 24

7. ACKNOWLEDGMENTS 25

8. REFERENCES 26

1
 NIST MULTIFUNCTION CALIBRATION SYSTEM

Nile Oldham and Mark Parker

Electronic Instrumentation and Metrology Group
Electricity Division
National Institute of Standards and Technology
Gaithersburg, MD 20899

Abstract - The NIST automated Multifunction Calibration System (MCS) for voltage, current, and
resistance is described. Developed primarily to calibrate digital multimeters and calibrators, the
system can also be used to test thermal converters, and micropotentiometers. Methods for
characterizing the MCS over a wide range of amplitudes at frequencies from dc to 30 MHz are
described.

1. INTRODUCTION

Digital multimeters (DMMs) from a number of manufacturers claim uncertainties and stabilities that
approach those of the laboratory standards used to support them. This is possible because these
DMMs are essentially miniature standards labs with built-in zener references, stable resistors, and
ac-dc transfer standards, all controlled by a central processor that applies corrections and performs
data analysis. In addition to being easier to use than most standards, DMMs are also quite rugged,
able to withstand the mechanical and thermal shocks of transportation from one laboratory to another
with little degradation in performance. These properties make them ideal transport standards to
provide traceability for the five quantities that most DMMs measure (dc voltage, current, and
resistance, and ac voltage and current).

The National Institute of Standards and Technology (NIST) offers calibration services^ for dc
voltage, dc resistance, and the ac-dc difference of thermal converters With these three services,
[ 1 -4] .

standards laboratories can provide direct NIST measured by most

traceability for the functions
DMMs. Based on customer demand, special tests^ for DMMs (all functions) were offered beginning
in 1990 [2-4]. This technical note describes the NIST Multifunction Calibration System (MCS) that
is used to provide special test services for DMMs, multifunction calibrators (the programmable

sources designed to calibrate DMMs), and special tests for low voltage thermal converters, and
micropotentiometers. The MCS is periodically calibrated using reference electrical standards
available at NIST. Results of these system calibrations and of frequently performed self-tests are
analyzed statistically to maintain quality control of the MCS.

NIST calibration services are well documented and analyzed measurements offered at a fixed cost.

NIST special tests are generally undocumented measurements that are charged "at cost."

2
 :

2. DESCRIPTION OF THE MCS

2.1 MCS Hardware'

A block diagram of the MCS instrumentation is shown in Fig. 1 . The basic system is quite simple,
consisting of a controller, a characterized calibrator (to test DMMs), and two characterized DM Ms
(to calibrate calibrators and serve as check standards for the system calibrator).

Fig. 1 . Simplified diagram of the MCS.

The NIST test facility consists of two such systems which utilize the instruments shown in Table 1

Table 1
MCS Instrumentation

System 1 System 2

Fluke 5720A Multifunction Calibrator Wavetek 4808 Multifunction Calibrator

Hewlett Packard 345 8 A Digital Multimeter Hewlett Packard 345 8 A Digital Multimeter

Wavetek 4950 Digital Multimeter Wavetek 4950 Digital Multimeter

PC Controller with GPIB Card PC Controller with GPIB Card

Keithley 7001 Switch System Keithley 7001 Switch System

Commercial instruments and software are identified in this report to describe the measurement system.
Such recommendation or endorsement by NIST, nor does it imply that the instruments
identification does not imply
or software are necessarily the best available for the purpose.

3
The system calibrator and DMMs are periodically characterized using the support equipment and
NIST reference standards shown in Table 2 and in Fig. 2.

Table 2
Support Equipment

Function Support Equipment and NIST Reference Standards

DC voltage NIST 1 0 V Josephson array

(DCV) Wavetek 491 1 10 V zener reference
Fluke 752 resistive voltage divider (0.1:1 to 100:1)

DC current NIST standard resistors (0.1 Q to 100 MQ)

(DCI) System DMM (DCV)
DC resistance NIST standard resistors
(DCR) System DMM (RES)
AC voltage System calibrator (DCV)
(ACV) NIST thermal voltage converters (0.1 V to 1 kV, 10 Hz to 100 MHz)
Ballantine 440 and Holt 12 micropotentiometers
(1 mV to 200 mV, 10 Hz to 30 MHz)
Wavetek 4920 ac digital voltmeter (1 mV to 200 mV, 1 Hz to 30 MHz)
NIST DSS-5 Digitally Synthesized Source (1 mV to 7 V, dc to 1 kHz)
Keithley 181 and 1 82 dc nano voltmeters
Hewlett Packard 34420A nanovoltmeters

AC current System calibrator (DCI)

(ACI) NIST thermal current converters (2 mA to 20 A, 10 Hz to 100 kHz)
NIST DSS-5
NIST Transconductance Amplifier ; i,

NIST ac resistors (0. 1 Q to 1 00 kQ) -

Instruments are controlled using a PC (486 or higher) and a National Instruments GPIB-PCIIA
general purpose interface bus (GPIB) interface card.

2.2 MCS Software ^^1^^ i-"' ^ --^

The systehi software, developed in National Instruments LabVlfeW™, allows the operator to perform
allof the tests needed to characterize the system calibrator and DMMs, store and edit correction files,
and set up and perform tests of the instrument under test (lUT). Since the instrumentation is

4
 10Hz - 30IVHZ 0 .1 Hz - 1 kHz 10Hz - 100kHz

200rTV-1 kV 1 - 200rrV 1 nV - 7V 2rM - 20A

"TT-ERMN.
MCRO- acrrALLY THERMflJ.
VOLTAGE SmTl-ES12ED CURRENT
COMzKlhRS POTHvmOMETERS
SOURCE OOTA/eTTERS
(TVC) (Mp)
(DSS) (TCQ

V V V V
CONTRCXIB?
AND
SOFTWARE
m MULTIFUNCTION
CALIBRATOR
AND
AtVPURStS <
AAA
aaTAL
MULTIWETH^
(DMA)
INSmsJtJEtfVS
(CMVl DETECTOR
SWITCHES. ETC.)

JOSB=HSON DC VOLTAGE RESISTIVE QUANTIZED

> > «
STANDARD €
ARRAY STANDARD VOLTAGE
RESISTORS RESISTANCE
DMDB%S
0-10V 10 V 1:1000 -lOOai 0.01 Q- 100 Mn 6.4 kO

Fig. 2. Block diagram of MCS support equipment.

controlled through theGPIB, most tests can be completely automated, requiring an operator only to
change connections between the calibrator and DMM
for certain functions. Test data are numerically
processed using commercial spreadsheet software. In addition to the software developed at NIST,
the MCS also makes use of commercial software to control the 4950 DMM.

Table 3
A list of the software routines available in the MCS

Title Description

DMM cal Used to calibrate any of the ftanctions on a test DMM using
the MCS calibrator.
Calibrator cal Used to calibrate any of the ftinctions on a test calibrator

using one of the MCS DMMs.

Thermal V/I cal Used to calibrate the ac voltage and current ftinctions of
DMMs and calibrators using thermal converters.
DC V/I cal Used to calibrate the dc voltage and current functions of the
MCS calibrator and DMMs.

5
 DCR Used to calibrate the resistance function of the MCS
caHbrator and DMMs.
AC-DC diff Used to determine the ac-dc difference of thermal
voltage/current converters and micropotentiometers.

4950MTS Software that uses a 4950 DMM to test a calibrator, and

store a new set of firmware corrections in Wavetek
calibrators.

3. CHARACTERIZING THE MCS

3.1 DC Voltage (DCV)

A Wavetek 4911 10 V zener reference is part of the MCS equipment used to support the DCV
fiinction of the system calibrator and DMMs. This instrument is periodically compared to the NIST
10 V Josephson array [5]. A control chart with the zener reference corrections is shown in Fig. 3.

3 1

^2

J
-3 '

0 50 100 150 200 250

Time (weeks)

o Measured -•- Regression

Fig. 3. Control chart with measured and predicted

corrections for the zener reference.

Periodically, special resistive dividers, a Fluke 752, and another similar divider designed at NIST,
are used to compare five calibrator voltages (100 mV, 1 V, 10 V, 100 V, and 1 kV) to the 10 V zener
reference. The dc voltage corrections to the calibrator C^^^^y are related to the calibrator setting S and
the actual calibrator output voltage F^^ck by:

Kvcv=S(l + C,,cv)- (1)

6
A control chart with selected calibrator corrections is shown in Fig. 4. Over long intervals, calibrator
corrections tend to drift linearly with time, so a least-squares fit to the correction time series is

performed and a linear equation is used to predict the calibrator correction between calibrations.

0.5

0
-0.5

-1 I

1.5
o
U -2 I

-2.5 + -f- -I- 4- +

12 16 20 24 30 34 38 42
Time (weeks)

O.I V (measured) O.I V (regression)

o 10 V (measured) -•- 10 V (regression)

Fig. 4. Control chart with measured and predicted

corrections for the MCS calibrator for DCV.

The system DMMs are calibrated at 10 V directly against the zener reference and at other voltages
using the characterized calibrator. The DMM corrections C^^cf related to the DMM reading R and
the true voltage by:

Vj,c = RO + C,^cy)- (2)

The system calibrator and DMMs rely on an internal zener voltage reference and timing to generate
and measure voltage. The calibrator uses a scaling technique known as time-division-multiplexing,
while the DMMs employ multislope integrating analog-to-digital converters. Both techniques are
capable of excellent linearity. The linearity of one of the system DMMs has been shown to be better
than 1 part in 10^ of full scale between 1 V to 10 V using the Josephson array [5]. The linearity of
other voltage ranges is measured using a calibrated DMM
and resistive divider.

For both the calibrator and DMMs, calibration at fiill scale and 10% of full scale in each range is

generally adequate to characterize the full range. The system software linearly interpolates between

calibration points to compute calibrator corrections over the full range (>1 0'' discrete voltages). These
software corrections are applied during the test of a customer's DMM. DMM corrections are
processed in a similar manner and applied during the test of a customer's calibrator.

The uncertainty of DCV measurements performed using the MCS depends on uncertainties computed
during the test (Type A) and uncertainties associated with the calibration of the MCS (Type B). A

7
description of these uncertainty types and methods of combining them is given in Section 5 and in
reference [6].

The Type B standard uncertainties of the MCS caHbrator for DCV are given in Table 4. These figures
include uncertainties associated with the DCV support equipment described in Table 2. The
uncertainties in Table 4 are for the set of points where the MCS is routinely calibrated and monitored
(Calibrated Points) and for all other points within the range (Full Range) where corrections are
interpolated estimates.

Table 4
MCS Type B Standard Uncertainties for DCV

Voltage Range Calibrated Points Full Range

1 mV to 100 mV 5 ^iVA^ + 1 ^V 10 |nVA^+ 1 |LiV

100 mVto 1 V 2 nVA^ + 1 [iV 4 \iV/W + 1 liY

1 V to 10 V 1 liW/V + 1 nV 2 [lY/W + 1 |iV

10 to 100 V 2 ^iVA^ 4 [lY/W

100 V to kV
1 3 ^iVA^ 6 |nVA^

3.2 DC Current (DCI)

The DCI function of the MCS calibrator and DMMs is periodically calibrated by applying various
of NIST reference, four-terminal resistors and measuring the voltage across
calibrator currents to a set
the resistors using a system DMM. The set of reference resistors consists of values between 0.1 Q
and 1 MQ that are maintained in oil and air baths with a temperature control of ±0.1 °C, and are
periodically calibrated against the NIST resistance standards [7].

To measure the influence of compliance voltage, calibrator currents are measured with the following
burdens:

a. reference resistor only,

b. reference resistor and each system DMM,
c. each system DMM only.
The compHance voltage is measured for each burden and the current differences vs compliance
voltage are plotted. If there is a significant difference, additional burden measurements are made to
cover the expected range of customer's DMM burdens. Linear interpolation between data points is

generally adequate to correct the calibrator current.

8
The dc current corrections to the cahbrator C^^ci ^re related to the caHbrator setting S and the actual
calibrator output current I^j^^., by:

IcDa=S(l + C,^„). (3)

The system DMMs are also calibrated at various currents by this procedure. The DMM corrections
CdDci related to the DMM reading R and the true current loc^y-
Ioc = RO +C,^a)- (4)

A control chart with selected DCI calibrator corrections is shown in Fig. 5. DCI corrections tend to
drift linearly with time, so a least squares fit to the correction time series is performed and a linear
equation used to predict the calibrator correction between calibrations. The system software also
is

linearly interpolates between calibration points to compute corrections over the fiill range. As with
DCV, software corrections are applied during DMM and calibrator tests.

40

35

'Sso
a
325
.2 20
o
g15
a 10
5

0 4- 4- + +
12 16 20 24 28 32 36 40
Time (weeks)

^ 1 0 mA (measured) 10 mA (regression)

o 1 A (measured) -•- 1 A (regression)

Fig. 5. Control chart with measured and predicted

corrections for the MCS calibrator for DCI.

Measurements are performed at cardinal currents between 1 )nA and 10 A. Measurements at other
DMMs, the current linearity is generally as good or better
currents indicate that, for the calibrator and
than the current stability. measurement at other than the calibrated points, the MCS
However, for
corrections are interpolated estimates and uncertainties are somewhat larger. As with DCV, the
uncertainty of DCI measurements performed using the MCS depend on Type A and Type B standard
uncertainties. The Type B standard uncertainties of the MCS calibrator for DCI (which include the
uncertainties of the DCI support equipment described in table 2) are given in Table 5.

9
 Table 5
MCS Type B Standard Uncertainties for DCI

Current Range Calibrated Points Full Range

1 |a,A to 1 mA 20 |aA/A+ lOnA 40 |iA/A+ lOnA

1 mA to 10 mA 15 |iA/A 30 liA/A

10 mA to 100 mA lO^A/A 20 liA/A

100 mA to A 1 20 [lA/A 40 [lAJA

1 A to 10 A 25 \iA/A 50 jiA/A

3.3 DC Resistance (DCR)

The DCR function of the MCS calibrator and DMMs is periodically calibrated against the set of NIST
reference resistors described earlier. Each reference resistor is measured by the system DMMs using
the 4-wire DCR function. The dc resistance corrections to the DMM Q^ic^ are related to the DMM
reading R and the true dc resistance R^^ by:

R^c=RO + C,^cr)-
~
(5)

Once the DMMs are calibrated, they are used to measure the set of resistors in the system calibrator.
The corrections to the calibrator resistors are based on the average of values assigned by the system
DMMs. The dc resistance corrections to the calibrator Q^i^;? are related to the cahbrator setting S and
the actual calibrator dc resistance R^^cr by:

Kdcr =S(1 + C,^cr)- (6)

A control chart with selected DCR calibrator corrections is shown in Fig. 6. DCR corrections also
tend to drift linearly with time, so a least squares fit to the correction time series is performed and a
linear equation is used to predict the calibrator correction between calibrations. The system software
also linearly interpolates between calibration points to compute corrections over the full range
(DMMs only). As with other ftmctions, software corrections are applied during DMM and calibrator
tests.

Measurements are performed at nominal decade resistance values from 0.1 Q to 100 MQ using the
DMMs. Decade resistance values between 1 Q and 10 MQ are available on the system calibrator.

Other than decade values may be measured using the system DMMs, which demonstrate resistance
linearity as good or better than resistance stability. However, for measurement at other than the
cahbrated points, the MCS corrections are interpolated estimates and uncertainties are somewhat
larger. The Type B standard uncertainties of the MCS DCR (which include the uncertainties of the

10
DCR support equipment described in table 2) are given in Table 6.

4 T-

? 2-

-4 -M 1 1 1 1 1 1 1 1 1 1-
1 4 8 12 16 20 24 28 32 36 40
Time (weeks)

^ 10 ohms (measured) 10 ohms (regression)

o 100 k-ohms (measured) -•- 100 k-ohms (regression)

Fig. 6. Control chart with measured and predicted

corrections for the MCS calibrator for DCR.

Table 6
MCS Type B Standard Uncertainties for DCR

Resistance Range Calibrated Points Full Range

1 Q 20 [iQ/Q 60 \iQ/Q

lOQ 10 iLiQ/Q 30 |LiQ/Q

100 Q 6 [xQ/Q 18 [iQ/Q

1 kQ 3 [iQ/Q 9 ^Q/Q
10 kQ 3 |iQ/Q 9 [xQ/Q

100 kQ 4^Q/Q 12 |iQ/Q

1 MQ 5 [iQ/Q 15 ^iQ/Q

10 MQ 20 [iQ/Q 60 nQ/Q

100 MQ 30 [iQ/Q 90 [iQ/Q

11
 0

3.4 AC Voltage (ACV)

The low-frequency ACV function of the MCS calibrator and DMMs is periodically calibrated using
a multirange thermal voltage converter (TVC) that has been calibrated against a set of NIST reference
TVCs, micropotentiometers, and a calculable DSS [8] . The ac-dc differences of these support
instruments are ultimately referenced to a set of characterized TVCs that are the NIST standards for
ac-dc difference [9]. The test procedures for characterizing the system calibrator and DMMs for ACV

are more complex than the other functions and will be described in detail.

3.4.1 Voltages from 1 mV to 1 kV from 10 Hz to 1 MHz

The multirange TVC is connected to the MCS calibrator and to the DMM
(both set to the ACV mode)
as shown in Fig. 7. The nominal ac voltage Fat frequency // is programmed and the output emf of
the TVC, e„ is measured using the dc digital voltmeter (DVM). The calibrator function is switched
to +DCV and the amplitude is programmed to +V. The calibrator voltage is adjusted to V+^j to bring
the TVC emf e^^j to within ±t of . [The operator selectable tolerance t is typicallybetween 1

)iV/V and 50 |iV/V.] The calibrator is switched back to ACV and the TVC emf and the DMM
ACV reading R^, are recorded. The calibrator is then switched to -DCV and adjusted to V_^/ to bring
the TVC emf e_^j to within ±t of e^^ . The above procedure is repeated for frequencies through f-

The TVC is disconnected, the calibrator is switched to ACV, frequencies fj through / are set at the
same nominal voltage V, and the DMM
readings through i?^- are recorded.

Fig. 7. Setup for characterizing the ACV fiinction of the

MCS using a multirange TVC.

12
 /

The calibrator correction for each test voltage at frequency f- is described by:

CcAcvi = (V,i - V) / Ku + (e.^ - ej/(nej + D. + C.^^y, (7)

where V^- = (7^^^ - V_^i)/2 {ave. dc voltage}

^di
= (^^di + ^-diM^ {ave. TVC emf with ±DCV applied}
n = the TVC power coefficient at the nominal voltage
Di = the ac-dc difference of the TVC at the nominal voltage at

QocF ^ the average calibrator DCV correction.

The DMM correction for each test voltage at frequency / is given by:

where = the DMM reading of the test voltage at f^.

Notes: i. The above procedure is designed to allow the MCS to be calibrated using the
reference TVCs as well as the multirange TVC. By driving the dc voltage to
match the TVC output when ac voltage is applied, the importance
of an
accurate knowledge of the TVC power coefficient n is minimized by
minimizing the numerator in the second term of (7). For the reference TVCs,
this coefficient varies between 1.6 and 2 as a function of the test voltage and
must be measured at each voltage. The multirange TVC has a nominal power
coefficient of 1, which is relatively insensitive to the test voltage, so the
tolerance t may be increased to reduce test time.
a. The reference TVC time constants are typically 1 s to 2 s; therefore, a wait
period of up to 30 s is required before recording the TVC emf. To minimize
the effects of drift, important that all of the wait periods are the same for
it is

a particular set of measurements. The multirange TVC settles in about 10 s

and has a low drift rate so timing is not as critical.

Hi. It is important that the dead time in switching from the ACV to the DCV
functions is minimized. For the highest accuracy using reference TVCs, a
switch and auxiliary supply are used to keep the TVC energized while the
calibrator is switching between functions. The multirange TVC is less
sensitive to switching time.
iv. The dc DVM should be able to resolve the TVC emf to within ±1 ju V/V. This
resolution corresponds to several nV for the reference TVCs (which have
output emfs from 2 mV to
10 mV) and several /iVfor the multirange TVC
(which has a full scale output of about 2 V).
V. Depending on the random error of the measurements, an actual calibration
consists offrom 1-10 repeats of the described procedure.

As the test frequency increases the calibrator becomes more sensitive to the load imposed by the TVC
and the test DMM. Therefore, to obtain the lowest uncertainties, the calibrator must be characterized
for each DMM type using the procedure described above. The second set of readings i?^, is DMM
13
used to generate unique corrections C^^cvhx) for all DMMs of that type (type-X DMMs for this
example). These corrections apply only when a type-X DMM is connected to the calibrator using the
same connectors and cables that were used to obtain the set of R^j readings. The unique set of
calibrator corrections for each DMM type are given by:
^cACVi(X) ~ ^cACVi ^di(X)- (9)

where Q,^^^
= (R^kx) -
Rai(x)^^i represents the additional correction neededcompensate for the
to
difference between the calibrator output vohage (programmed to with the TVC and type-X DMM
connected and with only the type-X DMM connected.
3.4.2 Voltages from 1 mV to 250 mV from 10 Hz to 1 MHz
A reference micropotentiometer (i^pot) is connected to the system calibrator output and a DMM is
connected to its output as shown in Fig. 8. The calibrator is set to the ACV function at kHz and its 1

voltage is adjusted until the DMM reading R^g is within ±t of the test The
voltage V. DMM reading
R^Q and the |apot output emf e^g are recorded. The calibrator frequency is programmed to // (the first
test frequency), the calibrator voltage is adjusted until the |j.pot emf e^, is within ±t of e^g, and the
DMM reading R^, is recorded. The calibrator is programmed to 1 kHz, the calibrator voltage is

adjusted until the nominal test voltage V±t is read by the DMM, and its reading R^, and the )ipot emf
e^, are recorded. The procedure is repeated for frequencies /, through f- interleaving a 1 kHz
measurement between measurements at each test frequency.

V/////////C3PIB // / /////////

MULTIFUNCTION
CALIBRATOR
mPOT
DMM

DVM

CONTROLLER

Fig. 8. Setup for characterizing the ACV function of

the MCS using a |Lipot.

The jipot is removed, the DMM is connected directly to the calibrator output, frequencies // through
are set at the test voltage V, and the DMM readings i?^/ through i?^, are recorded.

14
The DMM corrections for each test voltage at frequency/ is described by the following relationship:

Q= (Rr, - RJ/Rr, + (e„, - e,,)/(nej + D, + (10)

where Rj^i ^ (Rro-i) + Rn)/^ {ave. 1-kHz reading}

^Ri
^ (^r(i-i) + ^ri)/^ {^yQ. i^pot emf with 1 kHz applied}
n = the )ipot power coefficient at output voltage V
D- ^ the ac 1-kHz difference of the jipot at output V -dXf -

C^n = the DMM 1-kHz correction.

The calibrator correction is given by:

c.iA = (R,i(i + cj-v)/v (11)

where R^^ = the DMM reading of the test voltage at

3.4.3 Voltage from 250 mV to 7 V from 10 Hz to 30 MHz
Measurements are performed using the calibrator 50-Q wideband output port. Special 50 Q TVCs,
calibrated out to 30 MHz, are used to measure the amplitude flatness of the calibrator using a
technique similar to that described in 3 .4. 1 except the calibrator is left on the ACV function and each
,

test frequency is referred back to 1 kHz.

A 50 Q TVC is connected to the wideband output of the MCS calibrator and to a wideband DMM
manner as shown in Fig. 7. The nominal voltage Fat frequency // is programmed and the
in a similar
TVC emf is measured using the dc DVM. The calibrator firequency is set to 1 kHz and its
amplitude is adjusted to V^g to bring the TVC emf to within ±t of e^. The calibrator frequency is
set to fj and the TVC emf e^, and the reading R^, are recorded.DMM The calibrator is
reprogrammed to 1 kHz and adjusted to V^, to bring the TVC emf e^, to within ±t of e^,. The above
procedure is repeated for fi"equencies f through f .

The TVC is disconnected, frequencies f through f are set at the same nominal voltage V, and the
DMM readings i?^/ through i?^, are recorded.

The calibrator correction for each test voltage at fi-equency f is described by:

Cacvi = (VRi - + (e^, - ej/(nej + D, + Q„ (12)

where V,^- = (V^^,,,^- V_^-)/2 {ave. 1-kHz voltage}

^di
= (e^di + e,^i)/2 {ave. 1-kHz TVC emf}
n = the TVC power coefficient at V
Z). = the ac-dc difference of the TVC aXVatf

C^n = the calibrator 1 kHz correction at V.

15
The DMM correction for each test voltage at frequency / is then given by:

C,,cvi = (V(l + C,,cvi)-RJ/K. (13)

where = the DMM reading of the test voltage at f-.

3.4.4 Voltages from 1 mV to 1 kV from 0.01 Hz to 1 kHz

A programmable digitally synthesized source (DSS) developed at NIST is employed in place of the
system calibrator to perform high accuracy, low- frequency tests of DMMs. The DSS synthesizes a
staircase approximation of a sine wave using a sine look-up table and a precision digital-to-analog
converter (DAC). Output voltages are programmable between 1 mV and 7 V (for higher voltages an
external dc-coupled amplifier is used). A DMM (on its DCV frmction) is used to measure the dc
voltage of each of the steps in the approximation. The rms value of the output voltage is computed
by this step calibration:

V, = (Tv,'/N)''\ (14)

where v,. = the dc voltages of the steps

A'^ = the number of step per period.

The frequency response rms error of the synthesized signal measured between 20 Hz and 1 kHz
is

using a calibrated multirange TVC. Below 20 Hz, it is assumed that the rms value remains constant
to within ±2 |iV/V. At 7 V, the frequency response out to 1 kHz is flat to within ± 5 |nV/V. This
performance degrades at lower voltages; however, even at 1 the DSS may be treated as an ac mV,
calibrator with zero correction below 20 Hz. DMMs calibrated using the DSS are assigned
corrections Q^ck/ by:

C,,cvi-(K(l + C,ss^-RJ/R,, (15)

where R^^ = the DMM reading at frequency f-

Cjjssi
= the DSS correction to at frequency .

Control charts for the ACV function corrections of the system calibrator and DMMs are quite
complex covering voltage, frequency, and DMM type. A ACV calibrator
sample chart for the
corrections at 10 V at 10 kHz is shown in Fig. 9. ACV corrections are less predictable than the DCV
corrections; however, until the drift patterns are better understood, a least squares linear fit to the
correction time series performed and a linear equation is used to predict the calibrator correction
is

between calibrations. The system software also linearly interpolates between calibration points to
compute corrections over the full range. Specific software corrections for the type being tested DMM
are applied where possible and all interconnection parameters are specified. In a similar manner,
when a calibrator is tested using a system DMM, the DMM type and all interconnection parameters
are specified. Measurements are performed at a specific set of voltages and frequencies from 1 mV
to 1 kV and 0.1Hz to 30 MHz. Measurements can be made at other than the calibrated points;

16
however, because the corrections at these points are interpolated estimates, the MCS uncertainties are
somewhat larger. The Type B standard uncertainties of the MCS ACV function are given in Table 7.

60 r
55
?50

25
J
20
0 10 20 30 40 50 60 70 80 90 100
Time (weeks)

o 10 V @ 10 kHz (measured) 10 V @ 10 kHz (regression)

Fig. 9. Control chart with measured and predicted

corrections for the MCS V and 10 kHz.
calibrator at 10

Table 7
MCS Type B Standard Uncertainties for ACV
Voltage Range Calibrated Points Full Range

1 mV to 10 mV (10/v)(70 + f)|iVA^ (10/v)(200 + f)|aV/V

lOmVto mV100 (70 + f72)^VA^ (200 +f72 )^VA^

100 mV to 500 mV (20 + f /5)^VA^ (60 + f/5)[iY/V

500 mV to 100 V (10 + f710)^V/V (30 + f/10)|iVA^

100 V to kV
1 (20 + f/5)[iV/V (60 + f75)^VA^
Where v = test voltage in mV and f = test frequency in kilohertz.
3.5 AC Current (ACI)
The ACI function of the MCS calibrator and DMMs is periodically calibrated against a set of
reference thermal current converters (TCCs), ac shunts, and a calculable DSS. The ac-dc differences
of these support instruments are ultimately referenced to a set of characterized TCCs that are the
NIST standards for ac-dc difference [10]. The test procedures for characterizing the system calibrator
and DMMs for ACI are also described in detail.

17
 /

3.5.1 Currents from 2 mA to 10 A from 10 Hz to 100 kHz

The reference TCC is connected to the system cahbrator and DMM
(both set to the ACI function) as
shown in Fig. 1 0. The cahbrator is programmed for the nominal current / at frequency// and the TCC
output emf e„ is measured by the dc DVM. The calibrator function is switched to +DCI and
programmed to +/. The calibrator current is adjusted to I^^, to bring the TCC emf e^^, to within
±t |iV/V of e„. The calibrator is switched back to ACI, the TCC emf e^, and the ACI reading DMM
R^, are recorded. The calibrator is then switched to -DCI and the negative polarity dc current is
adjusted
to to bring the TCC emf to within ±t p.V/V of e^, . The above procedure is repeated for
frequencies through f^.

GPIB

MULTIFUNCTION
CALIBRATOR

DMM /
B
SHUNT

TCC DVM

CONTROLLER

Fig. 10. Setup for characterizing ACI function of MCS

calibrator using a TCC.

The TCC is removed, the calibrator is switched to ACI, frequencies /; through / are set at the same
nominal current / and the
, DMM
readings R^, through i?^, are recorded. The calibrator correction at
each test current at frequency / is described by:

C.c/,- = r4 - /;// + (e^, - ej/(ne,) + A + C^oc. (16)

where 7^- = (7+^,. - {ave. dc current}

e^- = (e+^i + e_^i )/2 {ave. TCC emf with ±DCI applied}

n = the TCC power coefficient at the nominal voltage

Z), = the ac-dc difference of the TCC at the nominal voltage at

^cDci ^ the average calibrator DCI correction.

18
The DMM correction at each frequency Q^c// is given by:

dACIi (i(i + c,,„i)-RJ/i- (17)

As the frequency increases the cahbrator becomes more sensitive to the load imposed by the TCC and
the test DMM. DMM type,
Therefore, for the highest accuracy test of a particular the above
procedure must be performed using that DMM type. The second set of DMM readings 7?^, is used to
generate unique corrections C^^cii(X) for all DMMs of that type (type-X DMMs for this example).
These corrections apply only when a type-X DMM is connected to the calibrator using the same

connectors and cables that were used to obtain the set of i?^,- readings. These corrections are given by:

C
^cACIi(X)
= C
^cACIi + C
^di(X)' (18)

where Q,y;j.^
= (Ri,i(X)
-
Rai(X))^^ represents the additional correction needed to compensate for the
difference between the calibrator output current (programmed to /,) with the TCC and type-X DMM
connected, and with only the type-X DMM connected.
As with DCI, ACI measurements are made at various compliance voltages and a correction factor for
the calibrator compliance voltage is applied when testing a DMM type that has not been characterized
with the system calibrator.

Control charts for the ACI function corrections of the system calibrator and DMMs are also quite
complex covering current, fi^equency, and DMM type (or compliance voltage). A sample chart for
the ACI calibrator corrections at 1 A, at 1 kHz for one of the DMM types shown in Fig. ACI is 1 1 .

corrections are less predictable than the DCI corrections; however, until the drift patterns are better

40
35

10 15 20 25 30
Time (weeks)

o 1 A @ 1 kHz (measured) -•- 1 A @ 1 kHz (regression)

Fig. 1 1 . Control chart with measured and predicted

corrections for the MCS calibrator for ACL

19
understood, a least squares linear fit to the correction time series is performed and a linear equation
is used to predict the calibrator correction between calibrations. The system software also linearly
interpolates between calibration points to compute corrections over the fiall range. Specific software
corrections for the DMM type being tested are applied where possible and all interconnection
parameters are specified. In a similar manner, when a calibrator is tested using a system DMM, the
DMM type and all interconnection parameters are specified.

3.5.2 Currents from 2 mA to 10 A from 0.01 Hz to 1 kHz

The NIST DSS and Transconductance Amplifier are employed to perform high accuracy, low
fi-equency current tests of DMMs. A DMM (on its DCI function) is used to measure the dc current
of each of the steps in the waveform approximation. The rms value of the output current is

computed by this step calibration:

i = (i:if/Ny^\ (19)

-
where /• = the dc currents of the steps
= the number of steps per period.

The frequency response error of the synthesized signal is measured between 20 Hz and 1 kHz using
a calibrated TCC. Below 20 Hz, it is assumed that the rms value remains constant to within ±2 |liA/A.
At currents between 100 mA and 10 A, the fi"equency response out to 1 kHz is flat to within
±100 |iA/A. This performance degrades at lower currents; however, even at 1 mA, the DSS may be
treated as an ac calibrator with zero correction below 20 Hz. DMMs calibrated using the DSS are
assigned corrections Q/jc// ^y:

C,ncn = asa + C,ssJ-RJ/R,. (20)

where i?^, = the DMM reading at frequency f-

^Dssi
^ the DSS correction to at frequency f- .

Again, a least squares linear fit to the correction time series is performed and a linear equation is used
to predict the calibrator correction between calibrations. The system software also linearly
interpolates betiveen calibration points to compute corrections over the full range. Specific software
corrections for the DMM type being tested are applied where possible and all interconnection
parameters are specified. In a similar manner, when a calibrator is tested using a system DMM, the
DMM type and all interconnection parameters are specified.

Measurements are performed at a specific set of currents and frequencies from 1 mA to 1 A and
0.1 Hz to 1 kHz. Measurements can be made at other than the calibrated points; however, because
the corrections at these points are interpolated estimates, the MCS uncertainties are somewhat larger.
The Type B standard uncertainties of the MCS ACV function are given in Table 8.

20
 Table 8
MCS Type B Standard Uncertainties for ACI

Current Range Calibrated Points Full Range

1 mA to 10 A (50 + 2fmA/A (150 + 3f)>iA/A

Where f = test frequency in kilohertz.

4. CALIBRATIONS AND TESTS USING THE MCS

Once the system calibrator and DMMs have been characterized, the MCS is used to test customer's
DMMs, calibrators, and related instruments.

Prior to testing, the test instrument is energized in the MCS lab overnight, to allow and it to stabilize
to satisfy any warmup requirements. Then, self-calibration or auto-zeroing routines are performed
to bring the instrument to its rated uncertainty.

4.1 Calibration of DMMs Using the MCS Calibrator

Unless otherwise requested, the DMM is calibrated at its input terminals. Using an adjustable
platform, the DMM terminals are brought "nose-to-nose" with the MCS calibrator terminals so that
only a few centimeters of connectors are required to interconnect them as shown in Fig. 12. The
calibrators have six output terminals including outputs (hi and lo), sense (hi and lo), guard and
ground. Test DMMs have a similar set of input terminals. Tables 9-11 list the interconnections made
for tj/pical DMM calibrations.
Table 9
Connections and configuration for a DCV or ACV calibration

From To

Calibrator voltage output hi DMM input hi

Calibrator voltage output lo DMM input lo
Calibrator guard and ground

The calibrator is programmed to local sense and local guard, and the DMM is programmed or
switched to local guard.

21
 Table 10
Connections and configuration for a DCI or ACI calibration

From To

Calibrator current output hi DMM current input hi

Calibrator current output lo DMM current input lo
Calibrator guard and ground

The calibrator is programmed to local guard, and the DMM is programmed or switched to local
guard.

Table 11
Connections for a DCR (4-wire) calibration

From To

Calibrator resistance output hi DMM resistance input hi

Calibrator resistance output lo DMM resistance input lo
Calibrator resistance sense hi DMM resistance sense hi
Calibrator resistance sense hi DMM resistance sense hi
Calibrator guard DMM guard
Calibrator ground

The calibrator is programmed to remote sense and local guard, and the DMM is programmed or
switched to local guard.

Calibrations are performed by programming MCS calibrator to the desired function and setting,
the
triggering the DMM (the first reading is ignored), and recording the next N readings.
The average reading of the DMM under test R, its associated correction Cj^^^, the system calibrator
setting S, and its correction C^cal ^^e related by the following equation:

R(1+C,^mm)=S(1 + Cscal)- (21)

22
Certain DMMs have software correction tables that are used to keep the instrument witliin a particular
specification. Correction tables are updated by applying a reference signal from the MCS calibrator
and then initiating a GPIB "correction" command that changes the appropriate correction value until
the DMM displays the proper value to within a selectable tolerance. Both the MCS software and the
Wavetek 4950 software have this capability. Repeat tests are performed to verify that the DMM
correction table has been accurately modified. This service represents a change in the long-standing
NIST policy of not adjusting instruments submitted for test. However, instead of making an analog
adjustment which can not generally be undone, in this case the adjustment is digital. The initial
corrections are stored in the instrument and the residual corrections are given in the test report. The
uncertainties of the residual corrections are based on the standard uncertainty of the calibrator and the
standard deviation of the repeat tests.

4.2 Calibration of Calibrators using the MCS DMM

Unless otherwise requested, calibrators are tested at their output terminals using the setup shown in
Fig. 12. The interconnections are the same as those given in Tables 9-11. The test calibrator setting

S is related to the system DMM reading R:

S(l+Cj^^i) -R(l + CsdmJ, (22)

NOSE TO NOSE FAST DMM OR FAST CALIBRATOR TEST SETUP

TO COMPUTER CALIBRATOR
^
TO COMPUTER

ADJUSTABLE
PLATFORM

Fig. 12. Setup to calibrate a test DMM using the

characterized MCS calibrator.

where Cj-^^^ is the test calibrator correction and C^^mm is the system DMM correction.
As with DMMs, certain calibrators have software correction tables that are used to keep the
instrument within a particular specification. Correction tables are updated by applying a nominal
signal fi-om the test calibrator to the MCS DMM, and then initiating a GPIB "correction" command,

23
which changes the appropriate correction value until the DMM displays the proper value to within
a selectable tolerance.

4.3 High Accuracy Special Test of DMMs and Calibrators

The highest accuracy test of a DMM or calibrator involves direct comparisons against the standards
that are used to characterize the MCS DMMs and calibrator using the techniques described in 3.4.
These are inherently more time consuming and expensive than the automated tests described in 4.2.

4.4 Special Test of Low Voltage Thermal Voltage Converters

Low voltage thermal voltage converters are tested using the MCS multirange thermal voltage
converter. Tests are performed by placing the two instruments nose-to-nose, and then applying ac
and dc voltages, while monitoring the output of each device. The ac-dc difference corrections for the
TVC under test are calculated by using the techniques described in 3.4.

4.5 Special Test of Micropotentiometers

Micropotentiometers are tested using the MCS multirange thermal voltage converter. Tests are
performed by connecting the )ipot input to the calibrator output and the |apot output to the thermal
converter input and then applying ac and dc voltages, while monitoring the output of each device.
The ac-dc difference corrections for the i^pot under test are calculated using the techniques described
in 3.4.

5. QUALITY CONTROL AND UNCERTAINTIES

The MCS on standards that are calibrated in other laboratories in the NIST Electricity Division
relies
and it is important to maintain traceability back to these standards. Each of the support artifacts or
instrument standards is calibrated on a regular interval (normally once a year) and a control chart
describing its value and uncertainty at any time is maintained. The MCS calibrator and DMMs are
normally characterized four times a year using the support standards. The calibrator and DMMs are
also intercompared several times per month to track long-term stability. The standard deviation of
this intercomparison represents a composite stability figure for the calibrator and two DMM
types.

The standard uncertainty of each function in the MCS calibrators and DMMs is based on the standard
uncertainties of the associated support instrument, the repeatability of the transfer process, and the
long-term stability of the calibrator. The figures given in the tables in section 3 represent one-
standard deviation (1-sigma) estimates of the system standard uncertainty. Note: These are
composite figures that serve for both the calibrator and the DMMs. Over a large parameter space
the figures apply equally to both; however, at high frequencies, the stability of the calibrator may be
significantly better than theDMMs. Under these conditions, a different set of uncertainty figures is
used. The total uncertainty of a test performed using the MCS includes the short-term stability of the
instrument under test. For example, the uncertainty of a test DMM
calibrated by the MCS calibrator
on the ACV function depends on the uncertainties of the following parameters:

24
1. The ac-dc difference of the muhirange TVC, fZ/y^ (see 3.4)
2. The correction of the dc voltage function of the cahbrator, U^f-. (see 3.1).
3. The standard deviation of the process in which the calibrator ac voltage is compared to its dc
voltage, (see 3.4).
4. The long-term stability of the calibrator/DMM at the test point, Ui^j^ (the standard deviation
of periodic comparisons with a system DMM).
5. The short-term stability of the test, U^ts (the standard deviation of the measurement including
the MCS calibrator and the instrument under test).
These values (all 1 -sigma estimates) are combined according to the procedures recommended in [6]
to give the expanded total uncertainty, Uj

U, = 2(Uso + U,ri + Usri + U^c' + UrycT (23)

where the first three terms are considered to be type A standard uncertainties and the last two are type
B standard uncertainties.

7. ACKNOWLEDGMENTS
The authors wish to acknowledge those people who helped in the development of the MCS: Barry
group leader of the Electronic Instrumentation and Metrology Group, for his strong support;
Bell, the
Ron Dzuiba, and Dean Jarrett for sharing their expertise in resistance measurements; Dick Steiner and
June Sims for assisting in dc voltage support; Joe Kinard and Tom Lipe for providing ac-dc transfer
calibrations and overall guidance in ac-dc voltage measurement techniques; Robert Palm for
prototyping and drafting support; and Kathy Magruder for running tests, and performing much of the

data reduction necessary for the MCS.

25
 8. REFERENCES

N. M. Oldham, M. E. Parker, A. M. Young, and A. G. Smith, "A High- Accuracy 10 Hz -

1 MHz Automatic AC Voltage Calibration System," IEEE Trans. Instrum. Meas., IM-36,

pp.883-887, Dec. 1987.

N. M. Oldham, "New Low-Voltage Standards in the DC to 1-MHz Frequency Range," IEEE

Trans. Instrum. Meas., Vol. 40, No. 2, pp. 368-372, Apr. 1991.

N. M. Oldham, P. S. Hetrick, "A Calculable, Transportable Audio-Frequency AC Reference

Standard," IEEE Trans. Instrum. Meas., Vol. 38, No. 2, Apr. 1989.

N. M. Oldham, "A 50-ppm AC Reference Standard which Spans 1 Hz to 50 kHz," IEEE

Trans. Instrum. Meas., IM-32, No. 1, Mar. 1983.

C.A. Hamilton et al., "The NBS Josephson Array Voltage Standard," IEEE Trans. Instrum.
Meas., IM-36, No. 2, pp 258-261, June 1987.

B. N. Taylor, C. E. Kuyatt, "Guidelines for Evaluating and Expressing the Uncertainty of

NIST Measurement Results," NIST Technical Note 1297, 1994 Ed.

R.F. Dziuba et al, "NIST Measurement Service for DC Standard Resistors," NIST Technical
Note 1298, 1992.

N.M. Oldham, P.S. Hetrick, and X. Zeng, "A Calculable, Transportable Audio-Frequency AC
Reference Standard," IEEE Trans. Instrum. Meas., Vol. 38, No. 2, pp. 368-371, April 1989.

J.R. Kinard et al, "NIST Measurement Services: AC-DC Difference Calibrations," NIST
Special Publication 250-27, May 1989.

26
 THE SP 250 SERIES ON NIST MEASUREMENT SERVICES*

SP 250-1 Spectral Radiance Calibrations SP 250-21 Calibration of Beta-Particle

PB87 179883 Radiation Instrumentation
PB88201579
SP 250-2 Far Ultraviolet Detector Standards
PB87227609 SP 250-22 Platinum Resistance Thermometer
Calibrations
SP 250-3 Radiometric Standards in the Vacuum
PB881 38367
Ultraviolet
PB87227625 SP 250-23 Liquid-in-Glass Thermometer
Calibration Service
SP 250-4 Fricke Dosimetry in High-Energy Electron Beams
PB89 128888
PB88 110374
SP 250-24 Standard Cell Calibrations
SP 250-5 Alpha-Particle Calibrations
PB881 23690
PB881 68620
SP 250-25 Calibration Service for Inductive
SP 250-6 Regular Spectral Transmittance
Voltage Dividers
PB881 08550
SP 250-26 NBS Phase Angle Calibration Services
SP 250-7 Radiance Temperature Calibrations
PB88225636
PB881 23674
SP 250-27 AC-DC Difference Calibrations
SP 250-8 Spectral Reflectance
PB892222616
PB881 09905
SP 250-28 Solid-state DC Voltage Standard Calibrations
SP 250-9 Calibration of Beta-Particle-Emitting
PB881 68703
Ophthalmic Applicators
PB881 08535 SP 250-29 Traceable Frequency Calibrations
PB881 68364
SP 250-10 Radioactivity Calibrations with the "4tt"
Gamma Ionization Chamber and Other SP 250-30 GOES Satellite Time Code Dissemination:
Radioactivity Calibration Capabilities Description and Operation
PB881 23708 PB881 68760

SP 250-1 1 Dosimetry for High Dose Applications SP 250-31 Mass Calibrations

PB88201587 PB89 153894

SP 250-12 Neutron Personnel Dosimetry SP 250-32 A Calibration Service for 30 MHz

PB87227617 Attenuation and Phase Shift
PB88238324
SP 250-13 Activation Foil Irradiation with Californium
Fission Sources SP 250-33 A Calibration Service for Voltage Transformers
PB882 17443 and High-Voltage Capacitors
PB882252903
SP 250-14 Activation Foil Irradiation by Reactor Cavity
Fission Sources SP 250-34 High Vacuum Standard and Its Use
PB88217435 PB89 193841

SP 250-1 5 Photometric Calibrations SP 250-35 The Calibration of Thermocouples

PB881 53747 and Thermocouple Materials
PB89209340
SP 250-16 Calibration of X-Ray and Gamma-Ray
Measuring Instruments SP 250-36 A Calibration Service for
PB882 11826 Current Transformers
PB91216770
SP 250-17 The NBS Photodetector Spectral Response
Calibration Transfer Program SP 250-37 Photometric Calibrations
PB88201595
SP 250-38 NIST Leak Calibration Service
SP 250-18 Neutron Source Strength Calibrations PB921 49772
PB88211818
SP 250-39 NIST Pressure Calibration Service
SP 250-19 Gamma-Ray-Emitting
Calibration of
Brachytherapy Sources
SP 250-40 Absorbed-Dose Calibration of
Ionization Chambers in a
PB891 93858
^°Co Gamma-Ray Beam
SP 250-20 Spectral Irradiance Calibrations SN003-003-03034-1 S2.00
PB881 23781

*
number (SN003-003-) and price can be purchased from the Superintendent of Documents, U.S. Government
Entries containing a stock
Printing Office, Washington, DC 20402. GPO will accept checks, money orders, VISA, and MasterCard. For more infomation, or to place
an order, call (202) 512-1800. Be sure to cite the stock number on all orders.
Entries containing PB numbers can be purchased from the National Technical Information Service, Springfield, VA 221 61 NTIS will accept .

American Express in addition to the payment methods listed for GPO. For more information call (703)487-4650; to place an order call
(800) 553-6487. Fax: (703) 321-8547. Be sure to cite the PB number on all orders.
Entries without stock or PB numbers are in preparation.
33
B- s
^H
(-0 rt
9
3
'

I—)
-1 O P
3- —
'ii*'
rti
5 w
era
~— —
O 1
S 3
I 3
ft n,
O
to
o 00 o
w
o 00
o
o
o
I

-
3
n
-t
n
n