Energy and Power Engineering, 2010, 2, 103-110
doi:10.4236/epe.2010.22015 Published Online May 2010 (http://www.SciRP.org/journal/epe)
103
An Improved Design of a Fully Automated Multiple
Output Micropotentiometer
Hala Mohamed Abdel Mageed1, Ahmed Faheem Zobaa2, Mohamed Helmy Abdel Raouf1,
Abla Hosni Abd El-Rahman1, Mohamed Mamdouh Abdel Aziz3
1National Institute for Standards, Giza, Egypt
2University of Exeter, Exeter, UK
3Cairo University, Giza, Egypt
Received December 1, 2009; revised January 7, 2010; accepted March 21, 2010
Abstract
This paper describes in details a new design of a fully automated multiple output micropotentiometer (pot).
A prototype has been built at the National Institute for Standards (NIS), Egypt to establish this highly im-
proved AC voltage source in the millivolt range. The new device offers three different outputs covering a
wide frequency range from only one outlet. This valuably supports the precise sourcing ranges of low AC
voltage at NIS. The design and the operation theory of this prototype have been discussed in details. An
automatic calibration technique has been introduced through specially designed software using the Lab-
VIEW program to enhance the calibration technique and to reduce the uncertainty contributions. Relative
small AC-DC differences of our prototype in the three output ranges are fairly verified. The expanded un-
certainties of the calibration results for the three output ranges have been faithfully estimated. However, fur-
ther work is needed to achieve the optimum performance of this new device.
Keywords: Multiple Output pot, AC-DC Transfer Standard, Single Junction Thermal Converter,
Calibration, Uncertainty
1. Introduction
In recent years new types of AC instruments and high
precision AC-DC voltage transfer devices for low volt-
ages and high frequencies have been widely adopted in
metrological and industrial laboratories. Therefore, new
activities for developing systems operating at these levels
have been undertaken [1,2].
The measurement of AC voltages and currents may
involve several methods [3]. However, thermal convert-
ers are mostly used in national measurement institutes
and other laboratories as the basis for deriving AC quan-
tities (voltage, current, and power) referring to known
DC quantities [4]. Normally, the measurement of AC
quantities with thermal converters implies the transfor-
mation of the electrical energy into thermal energy by
means of the Joule heat dissipated in the thermal con-
verter heater resistor [5,6]. Evidently, these thermal con-
verters are the most accurate AC-DC transfer standards
for the transfer of alternating voltage and current to the
equivalent DC quantities [7,8]. Commonly, there are two
types of thermal converters: single-junction thermal
converter (SJTC) and multi-junction thermal converter
(MJTC) [6]. Although MJTCs in planar technique are
used for voltages down to 100 mV [9], SJTC are mainly
used as thermal current converters because they are sim-
pler and easily available [8].
In national metrology institutes, the accurate AC in-
struments are mostly calibrated using micropotentiome-
ters (pots), which are basically voltage sources [10,11].
The µpots are developed for the generation of accurate
low AC voltages at wide range of frequencies. In actual
fact the SJTC µpots are still highly admitted specially
those designed with thin-film radial resistors, because
radial resistors so far ensure optimum frequency re-
sponse [12].
As accurate calibration of low AC voltages presents
several challenges, we implemented the design of a new
simple versatile accurate multiple output µpot in order to
enhance the capabilities of NIS by extending its precise
output sourcing ranges of low AC voltage.
The traceability for the precision sourcing of AC
voltages in the ranges of 10 mV, 25 mV, 50 mV, 100 mV,
200 mV, and 500 mV at frequencies from 10 Hz to 20

H. M. A. MAGEED ET AL. 104
KHz had been derived at NIS from a set of SJTC pots
which had been previously fabricated and calibrated [13].
Nevertheless, each μpot of that set produces only one
output AC voltage range.
In this work we aimed to build a fully automated pro-
totype of a multiple range μpot offering three different
output millivolt ranges from only one outlet through two
disc resistors. These three output ranges are 300 mV, 400
mV, and the parallel resistances combination equivalent
value (171.4 mV) at frequencies from 10 Hz to 10 KHz.
Furthermore, the automatic calibrations of the new de-
vice, the AC-DC differences (), and the expanded un-
certainties for the results had been fully investigated in
this work.
2. Construction of the New Developed
Multiple Output Pot
Figure 1 demonstrates the block diagram of the multiple
output µpot which consists of a N type male coaxial in-
put connector, an ultra high frequency (UHF) type SJTC
with nominal rated current of 5 mA, and a microcontrol-
ler (AT89C2051). The microcontroller is used to control
two mechanical relays through four-push button
switches. The first switch actuates the PC to automati-
cally control all the circuit, while the other three can be
manually activated to change between two radial resis-
tors with nominal values of 60, and 80. Each one of
the two resistors consists of ten parallel equal resistors
and is securely soldered into the output female N-type
coaxial connector for producing the output voltage.
3
2
1
0
AT89C2051
Mic ro
Controller
P
C
Serial
Cable
Input
Current
Relay 1 Relay 2
P
us
h
-
button switches
R
el
.
2
R
el
.
1
SJTC
R1
(60 ohm)Output
Vol t age
R2
(80 ohm)
Type N
Male Connecto
r
Type N
Female
Connector
Figure 1. Block diagram of the multiple output µpot.
In this technique, each one of the two resistors can be
easily chosen when the corresponding push button switch
is pressed while; the third button introduces their parallel
combination (34.29) by activating the two relays at the
same time.
The AT89C2051 microcontroller is a low voltage,
high performance and powerful microcomputer which
provides a highly flexible and cost effective solution to
many embedded control applications. In addition, the
AT89C205 microcontroller is designed with static logic
(binary code) which is stored in the microcontroller
ROM through a C-language program [14].
The pot is normally designed to provide a precisely
determined voltage at its output terminal when it is ex-
cited with an external source. The input current flows
through the heater of the SJTC to the radial resistor and
the voltage drop across the radial resistor induces a low-
impedance source of AC voltage. The output voltage is
nominally the product of the heater current and the resis-
tance of the radial resistor [15].
The measurement principle of the SJTC is based on
converting the electrical signal to a heat power [16]. In
such a converter, energy dissipated by an AC current
flowing through a heater resistor, raises its temperature
above the ambient. It is then compared to an equal
energy dissipated by a DC current flowing through the
same heater. The increase in the temperature of the
heater resistor by the rms of the AC signal and the
equivalent DC signal is proportional to the dissipated
energy and it is then measured using the thermocouple.
A relative difference between the response of the
converter to AC and DC inputs, called AC-DC transfer
difference, is determined from these two measurements
[17]. The AC-DC transfer difference is the main objec-
tive of the metrological characterization of each AC-DC
standard [18].
In our adopted design, the normal radial resistors are
replaced by two radial resistors producing three output
values as explained before which gaining benefit of
getting three output voltages of 300 mV, 400 mV, and
171.4 mV by a very simple circuit. In this design, more
advanced variable resistors were used instead of the re-
sistors used in the single range µpot. This type of resis-
tors are called multi-turn presets resistor. It is mostly
used where very precise adjustments must be made
through its screw. Its screw is turned to give very fine
adjustment control of the resistor required value. The
multi-turn presets resistors are miniature versions of the
standard variable resistor. They are designed to be
mounted directly onto the circuit board and adjusted only
when the circuit is built.
Moreover a new software using LabVIEW program is
prepared to measure  of the pot and calculate the un-
certainties of the results.

H. M. A. MAGEED ET AL.105
3. Setup of the Whole Automated System
In metrological and industrial laboratories, programma-
ble instruments are now widely employed and increas-
ingly included in measurement systems, because they
can perform automatically the time consuming opera-
tions required in the calibration activity. Accordingly,
there is an effective request for software able to calibrate
such instruments [19]. However, building automatic sys-
tems for calibration is not straightforward, especially for
application in metrological laboratories that operate at
high level of accuracy. For this reason, a special pro-
grammable measurement system has been built.
As shown in Figure 2 the automated calibration sys-
tem of the multiple output pot consists of a highly ac-
curate (FLUKE 5720), programmable calibrator, used as
a precise source for both alternating and direct currents,
SJTC multiple output pot with three output ranges,
highly sensitive digital multimeter (DMM) with a very
high input resistance (10 G) and 10 nV DC resolution
(HP 3458A) to measure the output emf of the multiple
output pot at each range of the three ranges. Moreover,
the (HP 3458A) DMM implements a reasonable digital
method for the measurement of DC and AC voltages [20].
Finally, a personal computer (PC) is used to drive the
calibrator and to record the DMM readings by using the
specially designed LabVIEW program.
All the necessary precautions were accomplished in
order to attain the optimum performance of the measur-
ing circuit. Technical considerations were also given to
fulfill correct grounding connections in this type of
measurement and to avoid any interference from high
field strength [15,16].
4. Experimental Work
Throughout the experimental work, the DC and AC cur-
rents had been applied to the µpot. The DMM then pre-
cisely recorded the output emf of the µpot through a spe-
cific LabVIEW command. The software was controlled
to send the gathering results to the computer to be saved
in a prepared excel worksheet. Indeed, the selected cur-
rent combined with the appropriate frequency for each
range was software processed.
Figure 2. Automated system of the multiple output µpot.
The nominal current (5 mA) was applied to each range
of the multiple output µpot in the sequence (AC, DC+,
DC-, AC) at approximately equal time intervals to readily
eliminate the DC reversal error [21].
Sufficient time had been allowed after each current
change for the SJTC to reach its final emf value [22].
After completing these four steps, the AC-DC difference
(δ) was evaluated at each frequency from the following
relation [15]:
AC DC
DC
EE
nE


 (1)
where, EAC is the average output emf due to the AC cur-
rent. While, EDC is the mean emf value due to the for-
ward (DC+) and the reverse (DC-) currents and n is a
dimensionless characteristics [21]. In fact, n is a neces-
sary factor in the equation due to the square law response
of the thermal converters. It approximately equals “2” (≈
1.6 to 2) at rated heater current [22].
At the beginning of the test the value of n must firstly
be determined. It was measured using the change in the
output emf, E, when the nominal input current is varied
by I according to the following relation [23]:
/
/
EE
n
I
I

 (2)
where,
I, is the nominal rated DC current (5 mA).
E, is the output emf corresponding to the nominal
rated DC current.
E, is the change in the output emf due to small
changes in the applied current, I.
Noting that, I, had been programmed to be ± 0.5
percent of the nominal rated current, and the final
AC-DC difference resulted at each frequency was the
average of 30 determinations of AC-DC difference under
the same measurement conditions.
The determined values of δ at different frequencies
were then added to the calibrated value of the applied
DC current with 5 mA nominal value to calculate the
actual values of the input AC current by using the fol-
lowing equation:
(1 )
ac dc
II


 (3)
Afterwards, by applying the actual values of the AC
current the actual values of the AC output voltage of
each range of the multiple output µpot at each frequency
could be obtained.
Furthermore, to get convincing evidence that the mul-
tiple output µpot performs as expected, the uncertainty
budget was thoroughly estimated. The uncertainty is de-
fined as the range of error of a measurement within
which the true value of the measurand is estimated to lie
within a stated level of confidence [16]. Type A and
Type B evaluations are the two approaches to estimate

H. M. A. MAGEED ET AL. 106
the uncertainty sources. It is cleared that, the AC-DC differences for the three
output ranges of the multiple output µpot are fairly small
at frequencies from 10 Hz to 10 KHz, as it is known that
µpots can be used to generate AC voltage signals in mil-
livolt ranges from 10 Hz to 1 MHz with AC-DC differ-
ences ranging from 20 ppm to 1000 ppm [25].
For AC-DC measurements, the Type B uncertainties
are generally dominating [23]. Type A evaluations of
standard uncertainty components are founded on normal
distributions, while type B evaluations are founded on a
suitable chosen distributions.
The combined standard uncertainty equals to the Root
Sum Square (RSS), of all the uncertainty contributions
[16,24]. All components of the combined standard un-
certainty (Type A, Type B) were taken into consideration.
The expanded uncertainties of the multiple output µpot
were calculated with confidence level of 95% (coverage
factor K = 2).
The gained results show that from 20 Hz to 1 KHz the
multiple output µpot has a very stable output where the
AC-DC differences are much smaller than the lower
limit of the admitted AC-DC differences range.
Although, at 10 Hz and 10 KHz the AC-DC differ-
ences are relatively high, they are still very near to the
lower limit. However, rather different but not highly ef-
fective behavior appeared at the 10 KHz for the 400 mV
output where it shows a negative sign with its . This is
most probably due to some dielectric losses inherent in
the resistors.
5. Results and Discussion
The results obtained at the rated current for n of each
voltage range were 1.69, 1.72, and 1.73 for the 300 mV,
the 400 mV and the 171.4 mV output AC voltage respec-
tively. Also, the AC-DC differences for the three ranges
of the µpot had been fully investigated. They are pre-
sented in Table 1 and Figure 3. The plots illustrate the
AC-DC differences of the 300 mV, 400 mV, and 171.4
mV ranges at frequencies from 10 Hz to 10 KHz in part
per million (ppm).
Also, the  of the 171.4 mV output at 10 KHz is rela-
tively high due to factors contribute to frequency and
voltage dependant errors. In fact this needs more careful
investigation to reach the optimum design performance.
Nevertheless, the AC-DC differences of our developed
multiple output µpot faithfully prove to be highly suc-
cessful.
Tables 2-4 illustrate all the components of the uncer-
tainty budget for the multiple output µpot three ranges
respectively.
-4 0
-2 0
0
20
40
60
80
100
101001000 10000
Frequency (Hz)
AC-DC Difference (ppm)
AC-DC Difference of 171.4 mV
AC-DC Difference of 300 mV
AC-DC Difference of 400 mV
The AC input currents, and AC output voltages for the
three voltage ranges at 10 Hz to 10 KHz combined with
the corresponding expanded uncertainties are listed in
Tables 5-7.
It is clearly shown from the tables that, the expanded
uncertainties of the multiple output µpot proved to be
less than 5 ppm for all values. This presents that the out-
put voltages of the multiple output µpot are very satis-
factory.
6. Conclusions
Figure 3. AC-DC difference of the 300 mV, 400 mV, 171.4
mV ranges (ppm). It is fairly demonstrated that our new design provides a
Table 1. Results of the AC-DC differences of the 300 mV, 400 mV, and 171.4 mV ranges at different frequencies.
Frequency (Hz)  (300 mV Range) (ppm)  (400 mV Range) (ppm)  (171.4 mV Range) (ppm)
10 27 24 25
20 14.2 14 14.6
40 10.2 11 11
50 11.3 10 9.7
100 6.6 11 10
200 9.3 7 8.8
400 6.8 4 7.8
1000 2.2 2 4.4
10000 25.6 –31.5 78.7

H. M. A. MAGEED ET AL.107
Table 2. Uncertainty budget of the 300 mV range calibration in (ppm).
Sources of Uncertainty Type of Uncertainty Uncertainty Value (ppm)
Repeatability of o/p DC current (Type A) 0.2
U-Calibrator calibration certificate (Type B) 1.3
U-DMM calibration certificate (Type B) 3.0
U-Cables thermal emf (Type B) 2.9
Freq. (Hz) Value (ppm)
10 0.4
20 0.5
40 0.5
50 0.6
100 0.6
200 0.6
400 0.6
1000 0.6
Repeatability of δ (Type A)
10000 0.5
Freq. (Hz) Value (ppm)
10 2
20 2.2
40 1.2
50 1.4
100 1.3
200 1.1
400 1.2
1000 1.4
Repeatability of AC Output Voltage (Type A)
10000 0.8
Table 3. Uncertainty budget of the 400 mV range calibration in (ppm).
Sources of Uncertainty Type of Uncertainty Uncertainty Value (ppm)
Repeatability of o/p DC current (Type A) 0.2
U-Calibrator calibration certificate (Type B) 1.3
U-DMM calibration certificate (Type B) 3.0
U-Cables thermal emf (Type B) 2.9
Freq. (Hz) Value (ppm)
10 0.8
20 0.4
40 0.4
50 0.5
100 0.4
200 0.3
400 0.4
1000 0.6
Repeatability of δ (Type A)
10000 0.5
Freq. (Hz) Value (ppm)
10 1.7
20 1.4
40 1.2
50 1.6
100 1.1
200 0.6
400 1.3
1000 1.6
Repeatability of AC Output Voltage (Type A)
10000 0.9

H. M. A. MAGEED ET AL. 108
Table 4. Uncertainty budget of the 171.4 mV range calibration in (ppm).
Sources of Uncertainty Type of Uncertainty Uncertainty Value (ppm)
Repeatability of o/p DC current (Type A) 0.2
U-Calibrator calibration certificate (Type B) 1.3
U-DMM calibration certificate (Type B) 3.0
U-Cables thermal emf (Type B) 2.9
Freq. (Hz) Value (ppm)
10 0.8
20 0.4
40 0.4
50 0.4
100 0.4
200 0.3
400 0.4
1000 0.4
Repeatability of δ (Type A)
10000 0.7
Freq. (Hz) Value (ppm)
10 0.9
20 0.6
40 0.6
50 2.1
100 0.6
200 0.8
400 0.9
1000 0.5
Repeatability of AC Output Voltage (Type A)
10000 0.4
Table 5. AC input currents and AC output voltages combined with the expanded uncertainties of the 300 mV µpot.
Freq. (Hz) IAC (mA) VAC (mV) ± Expanded Uncertainty (ppm)
10 5.0006822 299.85727 4.8
20 5.0003622 299.64441 4.9
40 5.0002622 299.85719 4.6
50 5.0002897 299.82553 4.6
100 5.0001722 299.85764 4.6
200 5.0002397 299.82329 4.5
400 5.0001772 299.81784 4.6
1000 5.0000622 299.81959 4.6
10000 5.0006472 299.87278 4.5
Table 6. AC input currents and AC output voltages combined with the expanded uncertainties of the 400 mV µpot.
Freq. (Hz) IAC (mA) VAC (mV) ± Expanded Uncertainty (ppm)
10 5.0006072 399.63049 4.8
20 5.0003572 399.79723 4.6
40 5.0002822 399.78593 4.6
50 5.0002572 399.78323 4.7
100 5.0002822 399.82343 4.5
200 5.0001822 399.78184 4.4
400 5.0001072 399.78409 4.6
1000 5.0000572 399.63214 4.7
10000 4.9992197 399.53989 4.5

H. M. A. MAGEED ET AL. 109
Table 7. AC input currents and AC output voltages combined with the expanded uncertainties of the 171.4 mV µpot.
Freq. (Hz) IAC (mA) VAC (mV) ± Expanded Uncertainty (ppm)
10 5.0006322 171.38202 4.8
20 5.0003722 171.51913 4.6
40 5.0002822 171.58246 4.6
50 5.0002497 171.56305 4.7
100 5.0002572 171.58934 4.5
200 5.0002272 171.59201 4.4
400 5.0002022 171.56805 4.6
1000 5.0001172 171.56316 4.7
10000 5.0019747 171.61028 4.5
significantly stable AC voltage source in millivolt ranges.
The combination of the microcontroller and the radial
resistors with the SJTC offers three highly stable output
AC voltages, 300 mV, 400 mV, and their parallel com-
bination, 171.4 mV, from the same outlet. Experiments
with the prototype indicate that its AC-DC differences
for the three output AC voltage ranges are faithfully
small.
In actual fact, it is expected that the adopted circuit
will be a very good supplement or even may replace
other commercial standards AC voltage sources due to
its extreme simplicity, its very low cost and its excellent
portability besides its admirable accuracy. In addition,
this device is readily suitable for many other applications.
However, further improvement can be achieved in the
final product in the near future.
7
. References
[1] U. Pogliano, G. C. Bosco and M. Lanzillotti, “Precision
Measurement of AC Voltages in the Millivolt Ranges,”
Proceedings of XVII IMEKO World Congress, 22-27
June 2003, Dubrovnik, pp. 625-629.
[2] J. Zinkernagel, “A Low-Voltage and High-Frequency
ac-dc Transfer System Using Inductive Dividers,” IEEE
Transactions on Instrumentation and Measurement, Vol.
43, No. 4, August 1994, pp. 544-548.
[3] U. Pogliano, B. Trinchera and F. Francone, “Reconfigur-
able Unit for Precise RMS Measurements,” IEEE Trans-
actions on Instrumentation and Measurement, Vol. 58,
No. 4, April 2009, pp. 827-831.
[4] I. Budovsky and B. D. Inglis, “High-Frequency AC-DC
Differences of NML Single-Junction Thermal Voltage
Converters,” Institute of Measurement, IEEE Transac-
tions on Instrumentation and Measurement, Vol. 50, No 1,
February 2001, pp. 101-105.
[5] H. Laiz and M. Klonz, “A Simulation Tool for the
AC-DC Transfer Difference of Thermal Converters at
Low Frequencies,” IEEE Transactions on Instrumenta-
tion and Measurement, Vol. 48, No. 6, December 1999,
pp. 1155-1160.
[6] H. Fujiki, “New Thin-Film Multijunction Thermal Con-
verter Design for Improved High-Frequency Perform-
ance,” IEEE Transactions on Instrumentation and Meas-
urement, Vol. 7, No. 9, September 2007, pp. 1243-1247.
[7] M. Klonz, Laiz and E. Kessler, “Development of Thin-
Film Multijunction Thermal Converters at PTB/IPHT,”
IEEE Transactions on Instrumentation and Measurement,
Vol. 50, No. 6, December 2001, pp. 1490-1498.
[8] M. Klonz, H. Laiz, T. Spiegel and P. Bittel, “AC-DC
Current Transfer Step-Up and Step-Down Calibration and
Uncertainty Calculation,” IEEE Transactions on Instru-
mentation and Measurement, Vol. 51, No. 5, October
2002, pp. 1027-1034.
[9] H. Dintners, M. Klonz, A. Lerm, F. VBlklein and T.
Weimann, “AC-DC-mV-Transfer with Highly Sensitive
Thin-Film Multijunction Thermal Converters,” IEEE
Transactions on Instrumentation and Measurement, Vol.
IM-42, April 1993, pp. 612-614.
[10] I. Budovsky, “A Micropotentiometer-Based System for
Low-Voltage Calibration of Alternating Voltage Meas-
urement Standards,” IEEE Transactions on Instrumenta-
tion and Measurement, Vol. 46, No. 2, April 1997, pp.
356-360.
[11] T. Funck, M. Stojanovi´c and M. Klonz, “Improved High-
Input-Impedance mV-Amplifiers with Gain Factors from
10 to 900,” IEEE Transactions on Instrumentation and
Measurement, Vol. 58, No. 4, April 2009, pp. 859-862.
[12] J. R. Kinard, D. X. Huang and D. B. Novotny, “Integrated
Thin-Film Micropotentiometers,” IEEE Transactions on
Instrumentation and Measurement, Vol. IM-44, April
1995, pp. 374-376.
[13] H. A. Mageed, M. Halawa, A. F. Zobaa and M. M. Abdel
Aziz, “Establishment and Characterization of a Traceable
AC Voltage Source at NIS, Egypt,” Universities Power
Engineering Conference, Padova, September 2008.
[14] www.atmel.com/atmel/acrobat/doc0368.pdf
[15] D. X. Huang, G. Rebuldela and J. D. Harper, “RF-DC
Differences of Micropotentiometers,” IEEE Transactions
on Instrumentation and Measurement, Vol. 46, No. 2,
April 1997, pp. 337-341.
[16] Fluke Corporation, “Calibration: Philosophy in Practice,”

H. M. A. MAGEED ET AL. 110
2nd Edition, Legal Department, Everett, WA 98206-9090,
May 1994.
[17] P. S. Filipski and M. Boecker, “AC-DC Current Transfer
Standards and Calibrations at NRC,” Symposium of Me-
trology, Mexico, October 2006.
[18] U. Pogliano and G. C. Bosco, “Determination of the Fre-
quency Dependence of the AC-DC Voltage Standard at
IEN,” IEEE Transactions on Instrumentation and Meas-
urement, Vol. 46, No. 2, April 1997, pp. 373-376.
[19] U. Pogliano, G. C. Bosco and M. Lanzillotti, “General-
ized Automatic System for AC/DC Transfer, AC Voltage,
and AC Current Measurements,” IEEE Transactions on
Instrumentation and Measurement, Vol. 55, No. 5, Octo-
ber 2006, pp. 1747-1751.
[20] C. D. Capua, D. Grillo and E. Romeo, “The Implementa-
tion of an Automatic Measurement Station for the Deter-
mination of the Calibration Intervals for a DMM,” IEEE
International Conference on Virtual Environments, Hu-
man Computer Interfaces and Measurement Systems, La
Coruna-Spain, 10-12 July 2006, pp. 58-62.
[21] B. Stojanovic, M. Klonz, H. Laiz and S. Kraicanic,
“AC-DC Voltage Transfer Module with Thin-Film Mul-
tijunction Thermal Converter,” IEEE Transactions on In-
strumentation and Measurement, Vol. 52, No. 2, April
2003, pp. 355-358.
[22] T. E. Lipe, “Operation and Reference Manual for the
NIST Automated AC-DC Calibration Systems and Soft-
ware,” NIST Special Publication, May 2004.
[23] “The Expression of Uncertainty and Confidence in Meas-
urement,” United Kingdom Accreditation Service, NA-
MAS, M 3003, 1st Edition, January 2007.
[24] “Guide to the Expression of Uncertainty in Measure-
ment,” Saudi Arabian Standards Organization, 2006.
[25] D. X. Huang, J. Harper and D. Mcadoo, “Measurement
Assurance of RF-DC Differences of Thermal Standards,”
NCSL Workshop & Symposium, 1996.