Applied Physics Research; Vol. 6, No. 6; 2014
ISSN 1916-9639
E-ISSN 1916-9647
Published by Canadian Center of Science and Education
AC Breakdown Characteristics of Nomex and Kapton as Insulation
Material Used in Pancake Tape Coil Structure for Resistive-Type
Superconducting Fault Current Limiter
H. Sun1, F. Wang1, H. Gao1, X. Liu1, Z. Jin1 & Z. Hong1
1
Department of Electrical Engineering, Shanghai Jiao Tong University, Shanghai, China
Correspondence: Z. Hong, Department of Electrical Engineering, Shanghai Jiao Tong University, Shanghai,
China. E-mail: zhiyong.hong@sjtu.edu.cn
Received: October 10, 2014
doi:10.5539/apr.v6n6p96
Accepted: October 24, 2014
Online Published: November 26, 2014
URL: http://dx.doi.org/10.5539/apr.v6n6p96
Abstract
A resistive-type superconducting fault current limiter (SFCL) is one of the most promising superconducting devices
in an electrical power system. However, an SFCL has the very obvious disadvantage of a large space requirement
when applied in high-voltage-level power networks. Considering the space utilization efficiency, a pancake
structure in which superconducting tapes are wound into coils should be adopted. When certain faults occur in a
power grid, a relatively high voltage difference will be present in the tape coil. The inner and outer layers are in
direct contact; therefore, special measures have to be carried out to overcome high-voltage insulation safety issues.
Two commonly used materials are considered in this study: Nomex and Kapton. The basic AC breakdown
characteristics in air and liquid nitrogen are explored. A new breakdown-strength testing platform is manufactured
to obtain the breakdown voltage. Considering that the edges of the superconducting tape are very thin, the impact of
the superconducting-tape thickness on the breakdown strength is explored. Two methods, parallel and vertical
lapping, are used to lap the insulation layer onto the tape and are subsequently compared. Finally, a face-to-face
electrode structure breakdown test is conducted to simulate the actual working conditions in a tape coil. The results
obtained can be used to design and optimize the geometrical structure of the SFCL pancake component.
Keywords: SFCL, Pancake component, tape coil, breakdown characteristic, insulation material, lapping method
1. Introduction
The discovery of high-temperature superconductivity has raised a great deal of interest in the exploration of its
potential applications in a power system. A resistive-type superconducting fault current limiter (SFCL) is
regarded to be one of the most promising superconducting applications. Resistive-type SFCLs have attracted
significant attention, and many related studies have been carried out worldwide (Hassenzahl, 2000). In our
university, a 10-kV resistive-type SFCL has been built and tested (Hong et al., 2012). In 2014, a higher voltage
prototype will be designed and demonstrated. Different types of primal designs have been reported. Among them,
a design using a stack of pancake components (Hobl et al., 2013) has the best performance for space utilization
efficiency. In each component, superconducting tapes are wound into a coil. A relatively high voltage difference
will be present between two adjacent layers during faults in power system. Special measures have to be taken to
overcome high-voltage issues. Lapping an insulation layer around the tape is one possible solution. Therefore,
the selection of insulation materials, lapping methods, and other aspects should be studied. The breakdown
characteristics of different polymer materials has been tested (Ombello et al., 2002). Further, the insulation
properties have been explored at room temperature (Ul Haq, & Raju, 2006) and cryogenic temperatures (Zhang,
Zhang, Tan, Luo, & Tu, 2012). A comparison between the DC and AC breakdown characteristics has also been
carried out (Seong, Seo, Hwang, & Lee, 2012). The results obtained from all of these tests are valuable and
comprehensive. However, all of these former studies have used a standard electrode configuration. No research
has considered the actual electrode structure in a pancake component used in an SFCL.
In this paper, we consider Nomex (polyamide) and Kapton (polyimide), which are very commonly used
insulation materials and have very good mechanical and dielectric properties in liquid nitrogen (LN2). Tests are
conducted to measure the AC breakdown characteristics of these two materials in LN2 and to verify the
effectiveness of our newly created test platform.
96
www.ccsennet.org/apr
Applied Physics Researrch
Vol. 6, No. 6; 2014
The fact thhat a superconnducting tape is very thin m
may underminee the breakdow
wn strength. F
Furthermore, a very
sharp edgee will aggravaate a partial disscharge, and m
more initial eleectrons are injeected into the dielectric. The use
of copper (0.1-mm thicckness) and suuperconductinng (0.3-mm thhickness) tapes as a compaarison will pro
ovide
further dettails regarding this issue.
There are ttwo main methhods to lap thee insulation layyer onto the tappe: parallel lappping and vertiical lapping. In
n this
research, w
we will demonstrate how thhese two methhods work, annd the breakdoown characterristics of these
e two
methods w
will be comppared. The ressults show thhat vertical laapping can ennhance the brreakdown strength.
Furthermoore, a face-to-face electrodee structure breeakdown test is conducted to simulate tthe actual worrking
conditionss in a tape coil..
2. Experim
mental Setup
The pancaake-structure resistive-type
r
SFCL has beeen conceptuallly designed, aand a schemattic diagram off one
componennt is shown in Figure
F
1. A layyer of insulatioon material is llapped around the tape.
Figure 1. Schematic diagram of one pancake compoonent
The breakdown tests useed a 50-Hz AC
C power sourcce. Nomex andd Kapton sheetts were preparred and designed to
have dimeensions of 70 × 70 mm2. The sheets were cleaned with aalcohol and pllaced in air or LN2 environm
ments
for 24 h beefore each test.
A schemattic diagram of the experimenntal setup is shoown in Figure 2. This setup iis used for perfforming breakd
down
experimennts and is modiffied from the pprevious setup ((Li et al., 20144). The double-layered glass vvessel has a vac
cuum
between thhe two layers. In order to preevent the occurrrence of bubbbles in the LN22 from affectinng the experim
mental
results, theere is a 10-m
min interval tim
me between evvery two expeeriments. The supporting strructure is mad
de of
glass-fiberr-reinforced plaastic (G10) beccause of its perrfect electrical and mechaniccal robustness. In addition, an
n AC
voltage is applied, increeased by 200 V at a time, and maintaineed for 30 s too observe the occurrence off any
breakdownn. The initial vooltage is set to be 40% of thee expected breaakdown voltagge. If breakdow
wn occurs withiin six
increases iin voltage, thee initial voltagge is lowered, and the test iis performed aagain. For each sample, testts are
performed five times, andd the average vvalue is used too calculate the bbreakdown streength.
Fiigure 2. Experrimental setup for the dielecttric-sheet breakkdown test
97
www.ccsennet.org/apr
Applied Physics Researrch
Vol. 6, No. 6; 2014
2.1 Dielectric-Sheet Breakdown Test
A newly ffabricated electtrode configurration and the entire experim
mental structurre are shown inn Figure 3 (a). The
configurattion of electroode conforms to the IEC 600243 standardd, and all metaal parts are m
made of brass. The
high-voltaage and groundd electrodes arre 25 mm in diiameter and heeight and havee 3-mm roundeed edges. Diffferent
sample thiicknesses are prepared:
p
0.05, 0.08, and 0.113 mm for Nom
mex; and 0.033, 0.05, and 0.007 mm for Kapton.
Tests are cconducted in aiir and LN2 to make a compaarison.
2.2 Influennce of the Thicckness of the Ta
Tape on the Breeakdown Strenggth
Another seet of electrodees is fabricatedd according to the IEC602433 standard. Thee high-voltagee electrode is 6 mm
in diameteer and 25 mm in height and has 1-mm rouunded edges. S
Superconductinng (100 mm inn length, 12 mm
m in
width, andd 0.3 mm in thickness) and ccopper (100 mm
m in length, 122 mm in widthh, and 0.1 mm in thickness) tapes
are used aas the ground electrode.
e
The tapes are presssed beneath a metal slug coonnected to thee ground wire. The
entire expeerimental struccture is shownn in Figure 3 (bb). The paralleel lapping methhod is adoptedd, and the tape
es are
lapped witth Nomex (0.13 mm in thickn
kness) and Kappton (0.07 mm in thickness).
Figgure. 3 Electrodde configuratioon: (a) dielectrric-sheet breakkdown test andd (b) the use off copper and
superrconducting taapes as the grouund electrode
2.3 Compaarison Betweenn Parallel Lappping and Vertiical Lapping
The paralllel and vertical lapping methhods are show
wn in Figure 4. The results aacquired are coompared to explore
the differences between these two lappping methods.
Figure 4. Dem
monstration off the lapping m
method: (a) paraallel lapping annd (b) verticall lapping
2.4 Face-tto-Face Electroode Breakdow
wn Test
To simulatte the actual working
w
condittions in the tappe coil, supercconducting tappe is used as bboth a high-vo
oltage
electrode aand a ground electrode. Thhe electrode coonfiguration iss shown in Fiigure 5. The eentire experim
mental
setup is im
mmersed in LN
N2 in a foam pllastic tank. The high-voltagee and ground electrodes are llapped with No
omex
98
www.ccsennet.org/apr
Applied Physics Researrch
Vol. 6, No. 6; 2014
and Kaptoon insulation laayers, respectivvely, using thee parallel lappiing method. Itt should be pointed out that there
are two layyers of insulatiion material beetween the higgh-voltage and ground electroodes.
Figuure 5. Face-to-fface electrode configuration
3. Results and Discussion
3.1 Dielecctric-Sheet Breakdown Test
Figure 6 annd Figure 7 shhows the relationship betweeen the breakdoown strength annd the thickness of the Nomex or
Kapton shheets in air andd LN2. The brreakdown streength is calcullated as the brreakdown volttage divided by
y the
thickness. In air, the breeakdown strenngth of Nomexx increases as the sheet thicckness increasses. In contrastt, the
breakdownn strength of Kapton
K
decreaases as the sheeet thickness inncreases, and the breakdow
wn strength tends to
saturate att 91.4 kV/mm. The breakdoown strength oof Kapton is aalmost three tim
mes that of Nomex. In LN2
2, the
breakdownn strength of both Nomex and Kapton decrease as thhe sheet thickkness increasees. The breakd
down
strengths ttend to saturatee at 56 and 1166 kV/mm for N
Nomex and Kaapton, respectivvely, and the bbreakdown stre
ength
of Kapton is only two tiimes that of Nomex. Furtherrmore, Kaptonn has better dieelectric propertties than Nomex in
both air annd LN2. Compparing with thhe values in airr, the breakdown strength of Nomex is inncreased by at least
two times but only 27%
% for Kapton. This result maay mean that L
LN2 has perm
meated into thee Nomex sheet and
significanttly enhanced its
i dielectric sttrength. This pphenomenon iis called the coomplex insulaation effect (Zh
hang,
Zhang, Taan, Luo, & Tuu, 2012) and iss used in transsformer insulaation by a com
mbination of soolid sheets and
d oil;
however, iit has not beenn demonstratedd for LN2.
Figure 6. Reelationship bettween the breakkdown strengtth and
Figure 7. Relatiionship betweeen the breakdo
own
the thicknesss of the Nomexx sheet
sttrength and thee thickness of the Kapton sheet
99
www.ccsennet.org/apr
Applied Physics Researrch
Vol. 6, No. 6; 2014
3.2 Influennce of the Thicckness of the Ta
Tape on the Breeakdown Strenggth
To study the influencee of the thickkness of the tape on the breakdown sttrength, the uuse of copperr and
supercondducting tapes is compared. The superconducting tapees used in thee experimentss are provided by
American Superconducttor with a thicckness of 0.3 mm, and the thickness of the copper tappe is 0.1 mm. The
parallel lappping method is adopted. Figgure 8 shows tthe breakdownn voltages of thhe Nomex andd Kapton insulation
layers. Thhe breakdown strengths are calculated, annd Table 1 lists the breakdow
wn strengths ffrom the sheet test
with parallel lapping. Thhe breakdownn strengths of N
Nomex and K
Kapton are 50.66 and 93.4 kV
V/mm, respectiively,
when coppper tape is useed as the grounnd electrode. IIn comparison with the breakkdown strengtths of the diele
ectric
sheet in LN
N2, the breakdown strengthh decreases byy 9.6% and 19.5% for Nomeex and Kaptonn, respectively.. The
copper tappe has very shharp edges; thuus, the electricc field around these edges iss highly non-uuniform. The initial
electrons aare more likelyy to be producced here, and the partial disscharge could be very intensse. The breakd
down
strengths oof Nomex and Kapton are 566.8 and 101.1 kkV/mm when the supercondducting tape is used as the grround
electrode. In comparisonn with the breaakdown strenggth of the dieleectric sheet in LN2, the breaakdown streng
gth of
Nomex is almost the sam
me, yet the brreakdown strenngth of Kaptoon decreases byy 12.8%. From
m the experim
mental
results, wee can concludde that a veryy low thicknesss will lead too a decrease iin the breakdoown strength. This
influence iis more obvious for Kaptonn. For Nomex, the influence is minimal w
when the thicknness is greater than
0.3 mm.
Figure 8. Breakdownn strength whenn copper and ssuperconductinng tapes are used as the grouund electrode fo
or
Nomeex and Kaptonn
Table 1. Breakdown streengths of Nomeex and Kaptonn using parallel lapping
Material
Sheet Test
(kV/mm))
Copper
Tape Useed
(kV/mm))
Superconnducting
Tape Useed
(kV/mm))
Nomex
56
50.6
56.8
Kapton
116
93.4
101.1
3.3 Compaarison Betweenn Parallel Lappping and Vertiical Lapping
Figure 9 sshows the breaakdown voltagges of the insuulation layer w
when it is lappped onto a coppper tape using the
parallel annd vertical meethods. Both N
Nomex and K
Kapton are testted, and a coppper tape is uused as the ground
electrode. For parallel lapping,
l
the bbreakdown streengths are 500.6 and 93.4 kkV/mm for Noomex and Kapton,
respectivelly. For verticaal lapping, the breakdown sttrengths are 600.6 and 104.9 kV/mm for N
Nomex and Kapton,
respectivelly. Compared with the expeerimental data of parallel lappping method, the vertical laapping method
d can
increase breakdown streength by 19.8%
% and 12.3% for Nomex annd Kapton, resspectively. Partts of the insulation
layer are ddouble-layeredd when verticaal lapping is uused. This is m
may be the reeason why verrtical lapping has
h a
100
www.ccsennet.org/apr
Applied Physics Researrch
Vol. 6, No. 6; 2014
higher breakdown strenggth. However, this enhancem
ment is very lim
mited because vertical lappinng is more like
ely to
introduce air bubbles annd other impuurities. In addiition, the breaakdown voltagges are more ddispersed when
n the
vertical lappping method is adopted. W
When the breakkdown point iss in a one-layeer area, the breeakdown volta
age is
relatively llow; however,, the breakdow
wn voltage is reelatively high when the breaakdown point iis in a double-layer
area.
3.4 Face-tto-Face Electroode Breakdow
wn Test
To simulaate the actual working condditions in a taape coil, facee-to-face breakkdown tests aare conducted. The
high-voltaage and groundd electrodes aare made of suuperconductingg tape lapped in the insulattion layer with
h the
parallel lappping methodd. The high-voltage and grouund electrodess are pressed ttogether face-tto-face with a G10
slug and T
Teflon screws. Two layers oof insulation m
material exist between the electrodes in this structure. The
breakdownn voltages aree presented in Figure 10. It is shown thatt the breakdow
wn voltages arre only 1.3 and
d 1.2
times of thhe value of onne layer of insuulation for Noomex and Kaptton, respectiveely. The breakdown strength
hs are
only 38.1 aand 62.9 kV/m
mm for Nomexx and Kapton, respectively, inn this insulatioon structure annd decrease by 32.9%
and 37.8%
%, respectively.. The results shhow that the fa
face-to-face inssulation structuure will lead too a very signifficant
decrease inn the breakdow
wn strength and will affect K
Kapton more thhan Nomex.
Figure 99. Breakdown strength
s
of parrallel and vertical
lapping for Nomex and K
Kapton
Figgure 10. Breakkdown strengthh of a face-to-fface
insulation struucture for Nom
mex and Kapton
4. Conclussion
The basic insulation prooperties of Nom
mex and Kaptoon sheets are ttested in air annd LN2 using a newly fabric
cated
test platforrm. The breakddown strengthhs of both insullation materialls will greatly increase in LN
N2. The breakd
down
strength off Kapton is higgher than that oof Nomex, andd the complex insulation effeect is observedd for Nomex.
On the bassis of the concceptual design of a pancake--structure resisstive-type SFC
CL, superconduucting tapes sh
hould
be lapped with an insulaation layer. Thhe breakdown sstrength of thee insulation maaterial will siggnificantly decrease
when lappped onto the taape. This is moost likely becaause the tape iis very thin annd has very sharp edges at which
w
partial disccharge is moree likely to occuur. It is also shhown that Kapton is far moree sensitive to tthe thickness of
o the
tape than N
Nomex. Whenn the tape thickkness is greaterr than 0.3 mm, the breakdow
wn strength of N
Nomex remain
ns the
same. Thee influence of the
t thickness oof the tape on tthe breakdownn strength is veerified. Two laapping method
ds are
used in thiis study: paralllel and verticaal lapping. Thee comparison bbetween these two methods shows that vertical
lapping haas better insulation perform
mance, but thee increase is very limited owing to the greater chanc
ce of
introducinng bubbles andd impurities. F
Further, face-too-face breakdoown tests are cconducted to simulate the actual
a
working coonditions in a tape coil. The coil structure will greatly w
weaken the breaakdown strenggth of the insulation
layer. The results obtaineed can be usedd to design andd optimize the geometrical sttructure of panncake resistive-type
SFCL com
mponents.
101
www.ccsenet.org/apr
Applied Physics Research
Vol. 6, No. 6; 2014
Acknowledgments
The authors are with the Department of Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240,
China. This work was sponsored by Science and Technology Commission of Shanghai Municipality of China
(Project 14DZ1100200). The authors would like to thank the Key Laboratory of Control of Power Transmission
and Conversion (Ministry of Education), The Shanghai Engineering Center for Superconducting Materials and
System, and the State Energy Smart Grid R&D Center (Shanghai) for the support on assistance in experimental
affairs.
References
Hassenzahl, W. V. (2000). Applications of superconductivity to electric power systems. Power Engineering
Review, IEEE, 20, 4-7. http://dx.doi.org/10.1109/39.841342
Hobl, A., Goldacker, W., Dutoit, B., Martini, L., Petermann, A., & Tixador, P. (2013). Design and production of
the ECCOFLOW resistive fault current limiter. IEEE transactions on applied superconductivity, 23(3),
5601804-5601804. http://dx.doi.org/10.1109/TASC.2013.2238288
Hong, Z., Sheng, J., Zhang, J., Lin, B., Ying, L., Li, Y., & Jin, Z. (2012). The development and performance test
of a 10 kV resistive type superconducting fault current limiter. Applied Superconductivity, IEEE
Transactions on, 22(3), 5600504-5600504. http://dx.doi.org/10.1109/TASC.2011.2180278
Li, S., Sun, H., Chen, Y., Wang, F., Jin, Z., Yin, Y., ... & Hong, Z. (2014). A Study of Surface Flashover and
Breakdown Characteristics in Liquid Nitrogen for SFCL Application. http://dx.doi.org/10.1109/TASC.
2013.2289927
Ombello, F., Attolini, G., Caracino, P., Nassi, M., Spreafico, S., & Montanari, G. C. (2002). Insulating materials
evaluation for cold dielectric superconducting cables. Dielectrics and Electrical Insulation, IEEE
Transactions on, 9(6), 958-963. http://dx.doi.org/10.1109/TDEI.2002.1115490
Seong, J. K., Seo, I. J., Hwang, J. S., & Lee, B. W. (2012). Comparative evaluation between DC and AC
breakdown characteristic of dielectric insulating materials in liquid nitrogen. Applied Superconductivity,
IEEE Transactions on, 22(3), 7701504-7701504. http://dx.doi.org/10.1109/TASC.2011.2178222
Ul Haq, S., & Raju, G. G. (2006). DC breakdown characteristics of high temperature polymer films. Dielectrics
and Electrical Insulation, IEEE Transactions on, 13(4), 917-926. http://dx.doi.org/10.1109/TDEI.2006.
1667753
Zhang, H., Zhang, G., Tan, R., Luo, M., & Tu, Y. (2012, March). Study on DC Breakdown Characteristics of
Polyimide at Cryogenic Temperature. In Power and Energy Engineering Conference (APPEEC), 2012
Asia-Pacific (pp. 1-4). IEEE. http://dx.doi.org/10.1109/APPEEC.2012.6307392
Copyrights
Copyright for this article is retained by the author(s), with first publication rights granted to the journal.
This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution
license (http://creativecommons.org/licenses/by/3.0/).
102