Curve Fitting Analysis of Time-Current

Characteristic of Expulsion Fuse Links

G. B. Costa A. C. Marchesan; A. P. Morais; G. Cardoso Jr. and M. Gallas
Federal University of Santa Maria – UFSM UFSM – Santa Maria-RS, Brazil UFSM – Santa Maria-RS, Brazil
Santa Maria-RS, Brazil e-mails: [email protected] e-mails: [email protected]
e-mail: [email protected] [email protected] [email protected]

Abstract—This work aims to modeling the Time-Current Due to the popularization of expulsion fuse links, a great
Characteristics (TCC) of expulsion fuse links through number of studies focus on the distribution protection schemes
mathematics equations. To achieve this, the Least Square Method considering the presence of these devices. When research
has been used and the Residual Sum of Squares (RSS) as a involves computational simulations, it is necessary to represent
comparative parameter. From this parameter, it is possible to
the fuse element using digital models. Generally, the data used
identify models that have the better adjustment in global scope
considering the different nominal current ratings, speed ratios and in these models are obtained through Time-Current Curves
manufacturers. The coefficients from the equations with the minor (TCC).
RSS will be available, so researchers of distribution system More specifically, the TCC fuse is defined by two curves:
protection can use it in future works.
“Minimum Melting” (MM) and “Total Clearing” (TC). These
Keywords—Curve Fitting; Time-Current Characteristic; curves are obtained from standardized electrical tests [3]
Expulsion Fuse Links. performed by manufacturers and made available through graphs
in log-log scale, where current is the horizontal axis and time is
I. INTRODUCTION the vertical axis. The MM characteristic represents the
minimum time in which the fuse element melts at a given value
Expulsion fuse links are widely used in the protection of of fault current. The TC characteristic gives the maximum
overhead distribution systems, specifically in branches and melting time plus the arcing time at a given fault current,
equipment such as transformers and capacitor banks due to their
representing the time for the fault to be completely isolated [4].
simplicity and low cost in relation to other protection devices
[1]. Generally, in these applications, the fuse link is installed From the data obtained from the TCC curves, there are
inside the fuse holder associated with a switch. This set is called several approaches proposed in the literature for fuse links
the fuse cutout or cut-out fuse. An example of a fuse link (a), a modeling, such as, published in [5]. In this work, the aim is to
fuse holder (b) and a fuse cutout (c) commonly found in outdoor evaluate the methodologies that employ analytical equations to
distribution systems is shown in Figure 1. represent the mathematical behavior of these devices in
computational simulations.
II. EXPULSION FUSE LINK MODELING

To model the fuse links, [6] and [7], a set of points from the
MM and TC characteristics was used. When necessary,
intermediate values are obtained by linear interpolation between
two consecutive points. Another approach is proposed in [8]-
[10] based on linear behavior when considering small intervals.
As a result, they use a curve fitting method to approximate the
segment of interest by linear equation (1).According to the
authors; this procedure significantly reduces the calculation task.
log(t ) = a. log(I ) + b (1)
Figure 1 - Example of: fuse link (a), fuse holder (b) and fuse cutout (c).
where t and I for the fuse denote operating time and current
During the operation of a fuse link, the heating caused by the respectively, and coefficients a and b can be calculated by a
fuse element fusion causes the sudden expansion of air inside curve fitting method.
the tube, expelling vaporized metal and contributing to the
extinction of the arc [2]. This feature gives the name to this type Although the authors in [8]-[10] use a technique that allows
of fuse link. linear adjustments, they affirm that TCC curves in log-log scale
can be approximated in their entirety by a second-order
polynomial function.
The authors are thankful to Brazilian government through Capes
(Coordenação de Aperfeiçoamento de Pessoal de Nível Superior). This work
was supported by Capes.

978-1-5386-3917-7/17/$31.00 ©2017 IEEE

 Thus, the work presented in [11] uses a curve fitting method implemented to obtain the coefficients and the RSS
to determine p0, p1 and p2 coefficients in (2) for each expulsion values.
fuse link.
4. A good function fit was considered to be less than 1 and
log(t ) = p0 + p1. log(I ) + p2 . log2 ( I ) + ... (2) did not produce an improvement of more than 10% in
RSS compared to the order of the next lower order.
On the other hand, the authors in [12] and [13] opted for the
Thus, this study evaluates different analytical functions
use of exponential equations (3) and (4), respectively, because
proposed in the literature to model the MM and TC
the function provides better adjustment results for both TCC
characteristics of expulsing fuse links, using the adjustment of
curves. Again, a curve fitting method is used to determine a, b,
Least Squares Method. As a criterion for evaluating the quality
c, e d coefficients in (3), and an coefficient in (4) (where n is an
of the analyzed adjustments, the RSS was adopted. The RSS can
integer between 0 and 4).
be calculated by (9). It was used to identify the models that
log(t ) = a. exp(b. log(I )) + c. exp(d . log(I )) provide the best adjustments in the global scope, considering the
(3) different nominal current ratings and manufacturers approached.
4
t = exp(  a n .(ln( I )) n ) (4)
RSS = (Estimated value – Original value)² (9)
n=0
Four mathematics functions were selected. The functions
In the studies [14] and [15], the adjustment of TCC curves of were presented in Section I.
fuse links was explored through equations developed for the
The most common used fuse in distribution system is “K”
computer representation of overcurrent electromechanical relay
type [4]. Only this type will be discussed. Three important fuse
characteristics. The authors describe the operation of an
manufacturers have been considered in this paper, i.e.: S&C
algorithm developed to evaluate and identify which of the
Positrol® fuse links [17], Cooper Power Systems Kearney™ Fit-
equations (5), (6), (7) or (8) give the smallest adjustment of
All™ fuse links [18] and Eletrofusi Power Systems fuse links
poerrors.
[19].
A In this way, one can expect that this study can contribute to
log(t ) = +B (5)
log(I ) n − 1 the development of further research in distribution system
protection, mainly in the academic scope, therefore facilitating
n
K  I  the modeling of expulsion fuse links.
t =C+ − b.  (6)
( I − h + ωI − 2 I ) q  50  IV. RESULTS
A1 A2 This section presents the results for modeling the MM and
log(t ) = A0 + + +
log( I ) log 2 ( I ) (7) TC characteristics of fuse links using (5), (8), Double
Exponential, and Polynomial function.
A1 A2
log(t ) = A0 + + + A. Reference [15]
(log( I ) − 1) (log( I ) − 1) 2 (8)
In [15], the authors described the operation of an algorithm
III. METHODOLOGY developed to evaluate and identify which of the equations (5),
(6), (7) or (8) give the smallest adjustment of errors. The present
Through this brief review, it is possible to observe that study selected (5) and (8) to evaluate the fuse link curve fitting.
numerous authors have proposed different ways to model the 1) Equation (5)
behavior of expulsion fuse links. The main reason for that is the
difficulty in finding a single function able to represent with great Figure 2Error! Reference source not found. shows the
precision the diversity of nominal current ratings and speed comparison between the manufacturers from the MM curve for
ratios of these devices. Still, the use of mathematical equations 6k, 15k, 40k and 100k using (5). Thus, from this figure it is
facilitates the TCC curves representation in computational possible to conclude that the RSS for all manufacturers are
simulations, favoring the performance of researches that employ greater than 1.
heuristic techniques and optimization methods for protective In general, the equation is not capable of producing good
device coordination in distribution systems. adjustments for the full range of fuse link nominal values.
The routine used in the present work was:
1. Obtaining the MM and TC curve from the
manufacturer;
2. Collecting at least 60 points from the MM and TC
curves using a curve fitting software;
3. Through Curve Fitting Toolbox™ [16], present in
MATLAB® functions, the selected functions was
 8
S&C Positrol®
B. Double Exponential Function [12]-[13]
Eletrofusi Power System
6,7848 Cooper Power Systems Kearney™ Results for Double Exponential Function, using (3), did not
fit considering the criteria used in this study. Figure 4 shows the
6
comparison among the manufacturers from the TC curve for
4,8209
25k, 40k and 65k. According to this figure, none of the RSS
values are in agreement with the work criteria, i.e. below 1.
3,9586
4
In general, only three RSS values of the eighteen studies are
RSS

3,7011
3,3859
3,2484
3,0773
2,9336
3,0668 below 1 for Cooper Power Systems Kearney™. For Eletrofusi
2,7248 2,5986
Power Systems, none of the RSS values of the eighteen studies
2
2,0005
is in agreement with the work criteria. For S&C Positrol® just
one the RSS value are under 1. Thus, the Exponential Function
could not be considered a good approach for fuse link curve
fitting.
0
6k 15k 40k 100k 8
S&C Positrol®
Fuse Link Eletrofusi Power System
Figure 2 - Fuse Link x RSS from the MM curve for (5). Cooper Power Systems Kearney™ 6,7207

2) Equation (8) 6

By applying (8), an estimation from the first to the fifteenth

order was made using the RSS as a comparative. Seeking for 4,1301
balance between quality of results and complexity of 4

RSS
3,5655
implementation, a good function fit was considered to be less 3,1917

than 1 and did not produce an improvement of more than 10% 2,769

in RSS compared to the order of the next lower order. 2

Figure 3 shows the comparison between the manufacturers 1,0865 1,0549 1,172
1,3162

from the MM curve for 6k, 15k, 40k and 100k using (5).
0,8 0
S&C Positrol® 10k 25k 65k
Eletrofusi Power System
Cooper Power Systems Kearney™ Fuse Link
Figure 4 - Fuse Link x the RSS from TC curve for Double Exponential
0,6 0,5876
0,5648 Function.

C. Polynomial Function [11]

0,4 0,3799
Special attention was given to the polynomial equations,
RSS

because they are the most used in the literature. Using (2), an
evaluation of the adjustments from the first to the fifteenth order
0,2416
was made using the RSS as a comparative parameter. Seeking
0,2
0,1489 0,1585 0,1675 for balance between quality of results and complexity of
0,1385
0,0882 0,0991 implementation, a good polynomial fit was considered to be less
0,0476 0,0443 than 1 and did not produce an improvement of more than 10%
0,0 in the RSS compared to the order of the next lower order. Figure
6k 15k 40k 100k 5 shows a comparison between the RSS and the polynomial
Fuse Link order from MM curve. The comparison used 6k, 10k, 15k, 25k,
40K and 100K Positrol® fuses. From this figure we can noticed
Figure 3 - Fuse Link x RSS from the MM curve for (8).
that it is not possible use the same polynomial order for all fuse
Table I shows the minimum, average and maximum value links. In other words, for the 100k fuse link, a polynomial
from the MM and TC curve for all the manufactures. function with order 5 has a good response; however, for fuse link
TABLE I 10k, a polynomial function with order 12 is required.
MINIMUM, AVERAGE AND MAXIMUM RSS VALUES OF MM AND TC CURVE
This behavior also occurs for the MM and TC curves from
FROM THE MANUFACTURERS.
the same fuse link and manufacture. For example, the
Eletrofusi polynomial order for Cooper Power Systems Kearney™ fuse 6k
Cooper S&C
Curve Power
Kearney™
Systems
Positrol® is 7th for the MM curve and 10th for the TC curve.
Minimum 2.13E-02 3.68E-03 1.03E-03 Despite the fact that it is not possible represent all fuse link
MM Average 8.85E14 1.57E12 3.19E08
Maximum 1.18E17 1.65E14 3.25E10
with the same polynomial order, this kind of equation results in
Minimum 2.41E-02 0.0038 2.50E-03 lower RSS values compared to the others equations. Hence,
TC Average 8.42E04 1.86E04 6.20E01 Table II, III, IV, V, VI and VII in Annex A present the
Maximum 1.14E07 1.46E06 5.39E03 coefficients of the MM and TC curves of Cooper Power
Systems Kearney™, Eletrofusi Power Systems and S&C REFERENCES
Positrol®, respectively. The polynomial equations are ready for
[1] C. R. S. Pierre and T. E. Wolny, “Standardization of benchmarks for
use in future protection studies and researches. protective device time-current curves,” IEEE Trans. Ind. Appl., vol. IA-
12 22, no. 4, pp. 623–633, July 1986.
6k [2] J. Slepian and C. L. Denault, “The expulsion fuse,” Electrical
10k
Engineering, vol. 50, no. 11, pp. 903–903, Nov 1931.
10 15k
25k [3] IEEE Standard Design Tests for High-Voltage (>1000 V) Fuses, Fuse and
40k Disconnecting Cutouts, Distribution Enclosed Single-Pole Air Switches,
100k
8 Fuse Disconnecting Switches, and Fuse Links and Accessories Used with
These Devices, IEEE Std. C37.41-2008 (Revision of IEEE Std C37.41-
2000), March 2009.
6
RSS

[4] T. A. Short, Electric Power Distribution Handbook. Boca Raton, USA:

CRC Press, 2003.
4 [5] W. Tian, C. Lei, Y. Zhang, D. Li, R. Fu and R. Winter, “Data Analysis
and Optimal Specification of Fuse Model for Fault Study in Power
Systems,’’ 2016 IEEE PES General Meeting, Boston, Massachusetts, July
2 2016.
[6] J. J. G. Ledesma, L. R. d. Araujo, D. R. R. Penido, and J. L. R. Pereira,
0 “Algorithm for the analysis of coordination of protection for distribution
0 2 4 6 8 10 12 14 16 systems,” IEEE Trans. Latin America, vol. 13, no. 7, pp. 2168–2175, July
Order 2015.
Figure 5 - Comparison between the RSS and the polynomial order from MM [7] E. M. Campitelli, L. G. W. D. Silva, R. A. F. Pereira, and J. R. S.
Mantovani, “Interactive system for placement and coordination of
curve for Positrol® fuses 6k, 10k, 15k, 25k, 40k and 100k.
overcurrent protective devices,” in Proc. IEEE/PES Transmiss. Distrib.
The average RSS value for MM and TC was 0.2725 and Conf. Expo.: Latin America, Aug 2006, pp. 1–6.
0.3036 for Cooper Power Systems Kearney™; 0.2172 and [8] S. Chaitusaney and A. Yokoyama, “Prevention of reliability degradation
from recloser-fuse miscoordination due to distributed generation,” IEEE
0.2220 for Eletrofusi Power Systems; 0.3984 and 0.3114 for Trans. Power Del., vol. 23, no. 4, pp. 2545–2554, Oct 2008.
S&C Positrol®, respectively.
[9] A. Fazanehrafat, S. A. M. Javadian, S. M. T. Bathaee, and M. R.
In order to show a consequence of using a fuse link modeled Haghifamt, “Maintaining the recloser-fuse coordination in distribution
systems in presence of dg by determining dg’s size,” in Proc. 9th IET Int.
by an equation with high RSS, Table VIII shows the coefficients Conf. Develop. Power Syst. Protect. (DPSP 2008), March 2008, pp. 132–
for all functions and the RSS value from the TC curve for 137.
Cooper Power Systems Kearney™ fuse 65k when current is [10] A. F. Naiem, Y. Hegazy, A. Y. Abdelaziz, and M. A. Elsharkawy, “A
137.26A. One can see that the polynomial function present the classification technique for recloser-fuse coordination in distribution
smallest RSS value and the smallest value of the difference systems with distributed generation,” IEEE Trans. Power Del., vol. 27,
between the estimated time and the real time. no. 1, pp. 176–185, Jan 2012.
[11] Y. Tang and R. Ayyanar, “Verification of protective device coordination
V. CONCLUSION in distribution systems with photovoltaic generation,” in Proc. 40th IEEE
Photovoltaic Specialist Conference (PVSC), June 2014, pp. 2100–2105.
Expulsion fuse links are widely used in the protection of [12] [11] H. A. Abdel-Ghany, A. M. Azmy, N. I. Elkalashy, and E. M. Rashad,
overhead distribution systems as a result of their simplicity and “Optimizing DG penetration in distribution networks concerning
protection schemes and technical impact,” Electric Power Systems
low cost in relation to other devices. Due to the popularization Research, vol. 128, pp. 113 – 122, 2015.
of expulsion fuse links, several studies on distribution protection [13] S. Santoso and T. A. Short, “Identification of fuse and recloser operations
schemes consider the presence of these devices. When research in a radial distribution system,” IEEE Trans. Power Del., vol. 22, no. 4,
involves computational simulations, it is necessary to represent pp. 2370–2377, Oct 2007.
the fuse element using digital models. [14] A. C. Enríquez, E. V. Martínez, and H. J. Altuve-Ferrer, “Time
overcurrent adaptive relay,” International Journal of Electrical Power &
Hence, this study evaluated four different analytical Energy Systems, vol. 25, no. 10, pp. 841 – 847, 2003.
functions proposed in the literature to model the MM and TC [15] A. C. Enríquez and E. V. Martínez, “Enhanced time overcurrent
characteristics of expulsion fuse links, using the adjustment of coordination,” Electric Power Systems Research, vol. 76, no. 6–7, pp. 457
the Least Squares Method. Three major manufacturers were – 465, 2006.
considered: Positrol® fuse links of the S&C Electric Company, [16] MATLAB® and Curve Fitting Toolbox™ Release 2014b, The
KearneyTM fuse links of the Eaton Cooper PowerTM and fuse MathWorks, Inc., Natick, Massachusetts, United States
links of the Eletrofusi Power Systems. As a criterion for [17] S&C ELECTRIC COMPANY. (2017, feb.) Positrol® fuse links. [Online].
evaluating the quality of the analyzed adjustments, the Residual Available: http://www.sandc.com/en/products--
services/products/positrol-fuse-links/.
Sum of Squares (RSS) was adopted.
[18] EATON COOPER POWER. (2017, feb.) KearneyTM fuse inks. [Online].
The Polynomial function presented the best results among Available: http://www.cooperindustries.com/content/public/en/power_
systems/products/fusing/fuse_links/kearney.html.
the selected functions. The average RSS values were the lowest
considering the four functions. The disadvantage of the method, [19] ELETROFUSI POWER SYSTEMS. (2017, feb.) Fuse links. [Online].
according to the study criteria selected by the authors, is that Available: http://www.eletrofusi.com.br/eng/fuse_eng.htm#linkb.
there is no single polynomial order that fits for all the full range
of fuse links nominal values.
 ANNEX A
TABLE II
POLYNOMIAL EQUATION COEFFICIENTS FOR ADJUSTING THE MM CURVE OF COOPER POWER SYSTEMS KEARNEY™ FUSES.
Fuse p0 p1 p2 p3 p4 p5 p6 p7 RSS
6k 3.80E03 -1.57E04 2.77E04 -2.69E04 1.55E04 -5.32E03 1.01E03 -80.9359 0.1586
10k 3.21E02 -7.86E02 7.71E02 -3.77E02 9.17E01 -8.86E00 0.1229
15k 1.57E02 -2.72E02 1.78E02 -5.16E01 5.57E00 0.2346
25k 2.14E02 -3.40E02 2.03E02 -5.38E01 5.33E00 0.1046
40k 2.45E02 -3.48E02 1.86E02 -4.42E01 3.92E00 0.7506
65k 4.08E02 -5.56E02 2.84E02 -6.44E01 5.45E00 0.1567
100k -9.41E04 2.21E05 -2.21E05 1.22E05 -4.05E04 7.99E03 -8.73E02 40.7057 0.4919
140k -1.26E05 2.77E05 -2.59E05 1.34E05 -4.15E04 7.68E03 -7.86E02 34.3248 0.2094
200k 1.26E02 -1.01E02 2.74E01 -2.52E00 0.2231
TABLE III
POLYNOMIAL EQUATION COEFFICIENTS FOR ADJUSTING THE TC CURVE OF COOPER POWER SYSTEMS KEARNEY™ FUSES.
Fuse p0 p1 p2 p3 p4 p5 p6 p7 p8 p9 p10 RSS
6k 6.79E04 -3.50E05 8.04E05 -1.08E06 9.47E05 -5.62E05 2.29E05 -6.35E04 1.14E04 -1.21E03 57.0299 0.1302
10k 9.30E02 -2.77E03 3.53E03 -2.49E03 1.05E03 -2.64E02 3.67E01 -2.18E00 0.0540
15k 2.34E03 -6.56E03 7.86E03 -5.21E03 2.06E03 -4.85E02 6.32E01 -3.51E00 0.0634
25k 4.28E02 -7.66E02 5.47E02 -1.94E02 3.42E01 -2.39E00 0.2357
40k 5.60E02 -9.21E02 6.05E02 -1.98E02 3.23E01 -2.08E00 0.5436
65k 7.17E02 -1.11E03 6.94E02 -2.16E02 3.33E01 -2.05E00 0.2063
100k -4.18E05 1.06E06 -1.17E06 7.32E05 -2.85E05 7.09E04 -1.10E04 9.66E02 -3.70E01 0.2938
140k -2.54E05 5.47E05 -5.01E05 2.54E05 -7.71E04 1.40E04 -1.40E03 6.00E01 0.3814
200k 1.73E02 -1.39E02 3.76E01 -3.43E00 0.8244
TABLE IV
POLYNOMIAL EQUATION COEFFICIENTS FOR ADJUSTING THE MM CURVE OF ELETROFUSI POWER SYSTEMS FUSES.
Fuse p0 p1 p2 p3 p4 p5 p6 p7 p8 p9 RSS
6k 1.66E04 -7.97E04 1.66E05 -1.97E05 1.44E05 -6.68E04 1.92E04 -3.13E03 2.21E02 0.1708
10k 2.12E05 -1.01E06 2.13E06 -2.59E06 2.02E06 -1.04E06 3.54E05 -7.72E04 9.74E03 -5.43E02 0.4030
15k 7.24E04 -2.72E05 4.46E05 -4.15E05 2.40E05 -8.81E04 2.01E04 -2.61E03 1.47E02 0.2212
25k 1.75E05 -5.98E05 8.91E05 -7.55E05 3.97E05 -1.33E05 2.78E04 -3.29E03 1.70E02 0.2460
40k 4.63E04 -1.26E05 1.47E05 -9.44E04 3.62E04 -8.30E03 1.05E03 -5.69E01 0.1900
65k 6.71E04 -1.69E05 1.82E05 -1.08E05 3.84E04 -8.16E03 9.58E02 -4.80E01 0.1183
100k 1.05E05 -2.46E05 2.45E05 -1.35E05 4.46E04 -8.80E03 9.61E02 -4.47E01 0.1711
TABLE V
POLYNOMIAL EQUATION COEFFICIENTS FOR ADJUSTING THE TC CURVE OF ELETROFUSI POWER SYSTEMS FUSES.
Fuse p0 p1 p2 p3 p4 p5 p6 p7 p8 p9 p10 RSS
6k 3.44E+04 -1.67E05 3.56E05 -4.40E05 3.45E05 -1.79E05 6.11E04 -1.33E04 1.68E03 -9.29E01 0.1740
10k 4.63E+05 -2.23E06 4.80E06 -6.08E06 5.01E06 -2.81E06 1.09E06 -2.86E05 4.90E04 -4.95E03 2.23E02 0.3829
15k 1.89E05 -7.38E05 1.27E06 -1.26E06 8.03E05 -3.38E05 9.43E04 -1.68E04 1.73E03 -7.88E01 0.2455
25k 3.87E05 -1.37E06 2.14E06 -1.94E06 1.13E06 -4.32E05 1.10E05 -1.79E04 1.69E03 -7.03E01 0.2283
40k 1.16E05 -3.37E05 4.23E05 -3.02E05 1.34E05 -3.79E04 6.66E03 -6.64E02 2.88E01 0.1833
65k 2.73E04 -6.38E04 6.36E04 -3.50E04 1.15E04 -2.26E03 2.45E02 -1.13E01 0.2174
100k 4.02E05 -1.02E06 1.13E06 -7.16E05 2.81E05 -7.03E04 1.10E04 -9.73E02 3.76E01 0.1227
TABLE VI
POLYNOMIAL EQUATION COEFFICIENTS FOR ADJUSTING THE MM CURVE OF S&C Positrol® FUSES.
Fuse p0 p1 p2 p3 p4 p5 p6 p7 p8 p9 p10 p11 p12 RSS
6k 7.38E05 -4.48E06 1.22E07 -1.94E07 2.02E07 -1.43E07 6.98E06 -2.32E06 5.02E05 -6.40E04 3.65E03 1.0493
10k 5.19E07 -3.32E08 9.66E08 -1.70E09 2.00E09 -1.67E09 1.01E09 -4.47E08 1.43E08 -3.26E07 4.97E06 -4.57E05 1.92E04 0.8498
15k 1.23E05 -4.60E05 7.52E05 -6.97E05 4.01E05 -1.47E05 3.35E04 -4.33E03 2.44E02 0.3124
25k 9.65E06 -4.14E07 7.95E07 -9.02E07 6.68E07 -3.38E07 1.18E07 -2.83E06 4.42E05 -4.07E04 1.68E03 0.2781
40k 5.96E04 -1.63E05 1.89E05 -1.22E05 4.67E04 -1.07E04 1.36E03 -7.33E01 0.1216
65k 1.18E05 -2.98E05 3.19E05 -1.90E05 6.73E04 -1.43E04 1.67E03 -8.39E01 0.2711
100k 3.18E03 -5.20E03 3.39E03 -1.10E03 1.78E02 -1.15E01 0.1280
140k -2.14E04 4.12E04 -3.27E04 1.38E04 -3.25E03 4.06E02 -2.11E01 0.2464
200k -4.47E-05 3.20E01 -2.39E01 5.97E00 -5.03E-01 0.3289
TABLEVII
POLYNOMIAL EQUATION COEFFICIENTS FOR ADJUSTING THE TC CURVE OF S&C Positrol® FUSES.
Fuse p0 p1 p2 p3 p4 p5 p6 p7 p8 p9 p10 p11 p12 RSS
6k 3.44E04 -1.67E05 3.56E05 -4.40E05 3.45E05 -1.79E05 6.11E04 -1.33E04 1.68E03 -9.29E01 0.1740
10k 8.88E05 -4.70E06 1.13E07 -1.62E07 1.56E07 -1.06E07 5.15E06 -1.83E06 4.69E05 -8.46E04 1.02E04 -7.39E02 2.43E01 0.6420
15k 1.58E05 -6.31E05 1.12E06 -1.18E06 7.98E05 -3.68E05 1.17E05 -2.52E04 3.53E03 -2.91E02 1.07E01 0.2604
25k 1.92E06 -7.88E06 1.46E07 -1.61E07 1.17E07 -5.95E06 2.14E06 -5.46E05 9.68E04 -1.14E04 7.95E02 -2.51E01 0.2346
40k 1.98E05 -6.19E05 8.54E05 -6.83E05 3.49E05 -1.18E05 2.64E04 -3.78E03 3.13E02 -1.15E01 0.0955
65k 7.94E05 -2.37E06 3.12E06 -2.39E06 1.17E06 -3.81E05 8.22E04 -1.13E04 9.08E02 -3.21E01 0.2038
100k 8.57E03 -1.60E04 1.24E04 -5.12E03 1.18E03 -1.45E02 7.34E00 0.0826
140k 1.11E02 -9.09E01 2.49E01 -2.30E00 0.7453
200k 3.89E01 -1.99E01 2.46E00 0.3644
 TABLE VIII
COMPARISON AMONG THE FUNCTIONS FROM THE MM CURVE OF COOPER POWER SYSTEMS KEARNEY™ FUSE 65K WHEN CURRENT = 137.26A
Absolute
Estimated Real
Function Order Coefficients Time (s) Time (s)
(Estimated– RSS
Real)
Equation A B n
1st 115.0873 149.4898 34.4025 3.1969
(5) 133.6209 -1.1262 4.5586
Equation A0 A1 A2 A3
3rd 154.36267 149.4898 4.8729 0.2332
(8) -1.56E01 6.35E01 -9.52E01 5.23E01
Double a b c d
1st 17.3918 149.4898 132.0980 31.6417
Exponential -0.0027 1.9815 292.6497 -2.4914
p0 p1 p2 p3 p4
Polynomial 4th 147.8671 149.4898 1.6227 0.1567
4.08E02 -5.56E02 2.84E02 -6.44E01 5.45E00