Corona From Floating Electrodes

Published on January 2017 | Categories: Documents | Downloads: 45 | Comments: 0 | Views: 219
of 12
Download PDF   Embed   Report

Comments

Content


Journal of
ELECTROSTATICS
ELSEVI ER Journal of Electrostatics 37 (1996) 67--78
Corona from floating el ectrodes
Francisco Rom~.n*, Vernon Cooray, Viktor Scuka
Institute of High Voltage Research, Uppsala UniversiO,, Husbyborg, S-752 28 Uppsala, Sweden
Received 17 October 1995; accepted after revision 9 January 1996
Abstract
It is not unusual to have insulated conducting objects located close to the conduct ors of
a Lightning Prot ect i on System. However, the separation of these objects from the Lightning
Prot ect i on System could vary from a few millimetres to some centimetres. When the system is
exposed to t hunderst orm electric fields, discharge could be initiated between the Lightning
Protection System and the floating conductive body. We have performed the study reported in
this paper t o gain knowledge concerning this discharge process. The critical field for the
initiation of cor ona currents from a floating electrode, when a negative DC voltage is applied to
the main gap electrodes, is detected by the discharge of the floating electrode to the earth
electrode. The electric field intensity at the high curvature (sharp) points on the floating
electrode surface was calculated numerically to determine this critical electric field. The
discharge repetition frequency was studied as a function of the applied background electric field
and a well-defined relationship was observed between these two parameters. The cor ona
currents were calculated and good agreement was found between calculated and measured
values.
Keywords: Corona; Critical field; Cor ona current; Floating electrode; High voltage; Lightning;
Breakdown voltage
1. I nt roduct i on
Several wor ks have e xa mi ne d t he i nf l uence of a f l oat i ng b o d y o n t he b r e a k d o wn
vol t age bet ween a hi gh vol t age el ect r ode a nd ear t h when t he b o d y is i nt r oduc e d
bet ween t hem. The r e duc t i on i n t he b r e a k d o wn vol t age a nd t he i ncr ement in t he t i me
t o b r e a k d o wn c a us e d by a f l oat i ng el ect r ode l ocat ed a ppr oxi ma t e l y 1 mm f r om t he
ear t hed el ect r ode have been i nves t i gat ed unde r t he appl i cat i on of a negat i ve i mpul s e
vol t age in pr e vi ous wor ks [ 1~, ] . Fu r t h e r mo r e , t he r e duc t i on in t he b r e a k d o wn
vol t age caus ed by a f l oat i ng b o d y s i t uat ed at a b o u t one - t hi r d of t he gap di s t ance f r om
* Corresponding author. Tel: 46 18 532703; Fax: 4618 502619. Permanent address: Universidad Nacional
de Colombia, A.A. 80789 Santa F6 de Bogota, Colombia.
0304-3886/96/$15.00 © 1996 Elsevier Science B.V. All rights reserved
PII S0 3 0 4 - 3 8 8 6 ( 9 6 ) 0 0 0 0 5 - 8
68 F. Roman et al./Journal o/Electrostatics 37 (1996) 67-78
t he ear t hed el ect rode was investigated by several aut hor s under the appl i cat i on of
a positive pol ar i t y impulse. The results were summari sed in the CI GRE r epor t of 1990
[5]. The influence of floating el ect rode cor ona sources on the observed reduct i on in
t he br eakdown vol t age of t he compl ex gap was studied by Phi l l pot t et al. [6]. The
mai n pur pose of t he present wor k is t o investigate t he cor ona phenomena associated
with a floating el ect rode l ocat ed 0.5, 1 or 2 mm from t he ear t hed el ect rode t hat can
cont r i but e to the observed br eakdown vol t age reduct i on ment i oned in our previ ous
wor k [ 1- 4] . This is achi eved by st udyi ng the onset of cor ona current s on t he 1.2 mm
radi us t op of a fl oat i ng el ect rode (see Figs. 1 and 2), when a negative DC field is
applied. The appl i ed fields are in the or der of magni t ude of nat ural electrical fields
present dur i ng t hunder st or m condi t i ons.
As i ndi cat ed in Fig. 1, a floating el ect rode is an isolated metallic body l ocat ed in the
gap space at a cert ai n di st ance from the ear t hed electrode. In the present paper, the
convent i on used is t o define this metallic body as a floating el ect rode (FE) onl y when
it is isolated. Otherwise, when it is connect ed to t he ear t hed electrode, it is called
"FE- ear t hed". The symbol s in Fig. 1 describing t he distances in the gap are the same
as t hose used in papers [ 1- 4] , and in t he CI GRE r epor t [5]. Thr ee different electric
fields are also shown in Fig. 1: E~ and Eb are t he electric fields at t he t op of the floating
el ect rode and at the secondar y gap, respectively. Ec is a reference gap electric field
obt ai ned by dividing the appl i ed negative DC vol t age by t he distance (D + h).
The cor ona source on the FE, act i vat ed by the reference electric field Ec, will cause
chargi ng of the FE and consequent l y t he br eakdown of the secondar y gap, see Figs.
1 and 2. The cal cul at ed electric field on t he high cur vat ur e (sharp) poi nt s of t he FE
surface at the i nst ant of the br eakdown of the secondar y gap is consi dered t o be the
critical field for the i ni t i at i on of cor ona current s.
H V (-) plate
electrode
reference
electric I
field= No
d~
Za
m ~
ii
(D+h) =5, 9m
I
d. - - - - - - _
Fig. 1. Geomet ri c paramet ers of a complex gap arrangement cont ai ni ng a floating electrode and the
measuring system. (Legend: FE is the floating electrode; dl is the pri mary gap distance; d2 is the secondary
gap distance; D is the complex gap. Ea, Eb are the electric fields at the t op of the floating electrode and at the
secondary gap, respectively. Ec is the reference electric field. Measuri ng system: (1) frequency meter; (2)
analogue oscilloscope, rise time 0.9 ns; (3) Rogowsky coil, 2 ns rise time. See Fig. 2 for ot her dimensions.
F. Rom&n et aL /Journal of Electrostatics 37 (1996) 67-78 69
••
R 1 =1. 2
166
N3=11
h = 1582
2R= lO
Fig. 2. Di agr am showi ng t he fl oat i ng el ect r ode geomet r y and t he electric fields E~ at t he t op of t he FE and
E~ at t he secondar y gap. The di mensi ons in mm are given in t he figure; d2 = 0.5, 1.0, 2.0 mm.
In Sect i on 2, some concept s are discussed concer ni ng t he critical field for the
i ni t i at i on of cor ona current s in the presence of a FE, hencefort h simply referred t o as
t he critical field. The exper i ment al pr ocedur e and t he results obt ai ned are descri bed in
the Sect i on 3 and t he concl usi ons are present ed in Sect i on 4.
2. Theory
Fl oat i ng el ect rodes are metallic obj ect s t hat are di sconnect ed or i sol at ed from the
eart h pot ent i al , and which, under cert ai n well-defined condi t i ons, can acqui re a par-
t i cul ar electric pot ent i al , t hus accumul at i ng a par t i cul ar pot ent i al energy. The pot en-
tial level and t he amount of pot ent i al energy t hat the fl oat i ng el ect rode acquires is
a direct funct i on of its ar ea and its l ocat i on in t he gap. It is a funct i on of the FE
capaci t ance. In an ar r angement such as ours (see Fig. 1), the floating el ect rode can
acqui re a vol t age whi ch depends on t he vol t age di st ri but i on in the gap, t he fl oat i ng
el ect rode' s geomet r y and its l ocat i on in t he gap, and is st rongl y dependent on the
secondar y gap di st ance d2 t o t he ear t hed electrode.
When a metallic body is i nsert ed in t he gap of length D, see Fig. 1, and a negative
DC vol t age is appl i ed across t he gap, t he FE di st ort s t he el ect rost at i c field. The
electric field will be enhanced on t he high cur vat ur e poi nt s of t he FE and t he whol e
metallic body acqui res a cert ai n vol t age with respect t o eart h, al t hough its net charge
is zero. If t he acqui r ed vol t age of t he fl oat i ng el ect rode is l ower t han the vol t age t hat
can be wi t hst ood across t he secondar y gap, no br eakdown will occur, but if it is
70 F. Romhn et al./Journal of Electrostatics 37 (1996) 6~78
higher, the FE will discharge to earth. The discharge brings the floating electrode
to earth potential, and it becomes charged. Thus, the net charge on the floating
electrode becomes different from zero. In the present case it acquires a positive
charge. This charge reduces the electric field strength in the secondary gap but
increases the electric field at the t op of the floating electrode, see Figs. 1 and 2. If the
applied DC voltage across the gap remains constant, no further discharges will occur.
Wi t h further increases in the applied DC voltage across the gap, the electric field in the
secondary gap increases again, and when the breakdown electric field of the secondary
gap is reached anot her discharge to eart h takes place. Again, if the voltage remains
const ant , no further discharge takes place. Wi t h increasing DC voltage across the
mai n gap, this process will lead to a gradual accumul at i on of positive charges on
the floating body and to a gradual increase in the electric field at the top of the
floating electrode until the critical electric field for the initiation of corona currents is
reached.
2.1. The cri t i cal el ect ri c f i e M f o r t he onset o f corona current s
When the critical electric field is reached for the initiation of corona discharges on
the high curvat ure prot rusi ons of the FE, a corona discharge starts from the floating
electrode. The floating electrode acquires a charge of the opposite pol ari t y to t hat of
the cor ona source. That is, in our present experiment, positive corona will charge the
floating electrode negatively.
The new acquisition of negative charges neutralises the previously accumul at ed
positive charges on the floating electrode increasing the electric field at the secondary
gap and decreasing it at the t op of the floating electrode. If the electric field at the
secondary gap exceeded the electric field st rengt h there, a breakdown of the secondary
gap will soon occur. If not, it is probabl e t hat the accumul at i on of negative charges on
the floating electrode can decrease the electric field E, at the t op of the floating
electrode and the corona discharge will be extinguished. This is really a t hreshol d
si t uat i on where a small increase of the applied DC voltage to the gap can lead to
a cont i nuous repetitive charge-di scharge process. This t hreshol d or critical voltage
V c r i t is a funct i on of the secondary gap distance: higher secondary gap distances imply
t hat the applied DC voltage must be higher. In this situation, the critical electric field
for the initiation of cor ona currents will be slightly lower t han the field at the t op of
the FE necessary to initiate the repetitive charge and discharge process of the floating
electrode.
It can t hen be concl uded t hat if the electric field E~ is equal to the critical electric
field for the onset of cor ona currents, the FE will be negatively charged, and if
Eb is equal to the secondary gap br eakdown electric field, a sudden discharge will
transfer this negative charge to earth. Once the discharge occurs, the FE acquires
a positive charge and a voltage t hat depends on the charge accumul at ed on
the floating electrode. The corona sources on the FE can lead to a further in-
crease of the secondary gap electric field Eb, and a charge-di scharge process is
initiated.
F. Romhn et al./Journal of Electrostatics 37 (1996) 6 ~7 8 71
2.2. The secondary gap discharge repetition frequency (DRF)
The dur at i on of t he chargi ng process depends on t he electric field i nt ensi t y at the
curved poi nt s of t he FE surface. The dur at i on can t ake several seconds for an electric
field close t o t he cor ona onset, but it can be faster if t he electric field is higher. The
reason for this r educt i on in the chargi ng time of the FE is t hat the cor ona current s are
enhanced by a hi gher electric field and t he fl oat i ng el ect rode capaci t ance will be
char ged faster. However , it must be not i ced t hat t he charge accumul at i on reduces the
electric field on t he high cur vat ur e poi nt s l ocat ed at t he t op of t he FE. If t he field
remai ni ng after chargi ng is hi gher or equal t o the critical electric field, per manent
secondar y gap sparki ng occurs. The difference bet ween t wo subsequent discharges is
t hen a funct i on of this remai ni ng electric field. It can be concl uded t hat this di scharge
repet i t i on frequency ( DRF) increases wi t h appl i ed electric field.
The upper limit for t he DRF is when br eakdown activity ceases and a per manent
connect i ng channel bridges t he secondar y gap. Under this condi t i on, t he cor ona
sources are capabl e of mai nt ai ni ng t he t hermal i zed secondar y gap channel and
a per manent connect i on bet ween t he FE and the eart h el ect rode is established.
Addi t i onal l y, under this condi t i on, the FE is "ear t hed" by the di scharge channel and
no mor e secondar y gap discharges are expected. Thi s limit is obvi ousl y a funct i on of
t he secondar y gap di st ance and decreases with it. However, the discharge repetition
frequency of the secondar y gap is an i mpor t ant par amet er , and is essential for
under st andi ng t he obser ved r educt i on in the br eakdown vol t age of the mai n gap
caused by t he FE i nt er act i on [ 1 4 ] .
2.3. The energy stored in the FE
The pot ent i al acqui r ed by t he FE under t he act i vat i on of cor ona sources is a direct
funct i on of t he FE capaci t ance and t he maxi mum electric field t hat t he secondar y gap
can sustain. Thus, a well-defined amount of energy is st ored in t he FE. The maxi mum
potential energy W st ored in the gap can be cal cul at ed using
w = ½ c v ~, ( 1 )
where C is the FE capaci t ance and V is the br eakdown vol t age of the secondar y gap.
Consi deri ng t he secondar y gap to be homogeneous, the vol t age t hat can be sust ai ned
across it is a funct i on of the br eakdown electric field Eb and is given by:
V = Ebd2. (2)
Combi ni ng Eqs. (1) and (2), t he energy can be cal cul at ed from
W I g " A 2 1~'2
= 2 ~'~ ~ 2 " t ~ b • ( 3 )
As t he capaci t ance C is inversely pr opor t i onal t o the secondar y gap distance, it can be
concl uded t hat t he pot ent i al energy st ored in t he gap is a di rect funct i on of t he FE
area A, t he permi t t i vi t y e of t he i nsul at i ng medi um, the square of the br eakdown
72 F. Romdn et al./Journal of Electrostatics 37 (1996) 6 ~7 8
electric field intensity of the secondary gap, i.e., Eb 2, and the secondary gap distance, i.e.
W =f ( A, e, E g, d2). (4)
In terms of the charge Q measured at the eart hed electrode, the energy W is given by
W = ½Qd2Eb. (5)
2.4. The rate of change of charge during discharge
The concepts of FE capacitance, cor ona effect, secondary gap electric field, dis-
charge repetition frequency (DRF) of the secondary gap and energy accumul at ed on
the FE have been discussed. Anot her factor t hat must be t aken into consi derat i on is
the rate of change of the charge in each discharge in the secondary gap. This is related
to the secondary gap distance and the rise time of the discharge current for this gap.
For a secondary gap of 1 ram, the rise time is of the order of nanoseconds. The fast
vari at i on of the charge in a time of the order of nanoseconds can produce i mpor t ant
effects in the complex gap electric field t hat will be anal ysed in future work. Wi t h
reference to these effects, it can be added t hat the rate of change of the charge is also
a rate of change of the energy stored in the FE. If a certain amount of potential energy
is stored in the FE, the discharge process is onl y the t ransformat i on of this pot ent i al
energy i nt o a spark. The consequence is a rapid change of the charge accumul at ed on
the surface of the FE. If this rate of change is high, a high current will be generated.
The mean current t~ which is obtained, can be estimated:
T- AQ (6)
At '
where At is directly related to the length of the discharge process, and AQ with the
charge stored in the FE capacitance. For a small gap of 1 mm this is of the order of
some nanoseconds and with a charge of the order of a few nanoCoul ombs, the rate of
change of charge is of the order of amperes. The increment of this current increase is
a compromi se between the capaci t ance of the FE and the secondary gap electric field:
for small secondary gap distances, At is very small but the amount of charge AQ t hat
can be stored is low because of the low breakdown voltage of small secondary gap
distances.
The current t hat is drai ned to the eart h in each discharge must be directly related to
the cor ona current t hat charges the FE between two subsequent discharges, as shown
later.
If the floating electrode has a low radius of curvat ure (highly curved surface, e.g.
a sharp point), the corona onset electric field will be reached with relatively low
voltages being applied across the mai n gap. The corona onset electric field strength
has been calculated using Eq. (7) derived by Hi l gart h [7]:
Ea = 6K, (1 + K2/x//~k), (7)
where Ea is the maxi mum field on the sphere, c5 is the relative air density, K1 = 30 kV/
cm for air, K2 = 0.47 c m- 1/2 and rk is the radius of curvat ure of the sphere (in cm).
F. Romhn et al. /Journal of Electrostatics 37 (1996) 67-78 73
Fr om Eq. (7), t he expect ed cor ona onset electric field st rengt h for a sphere of radius
1.2 mm is 7.3 MV/ m with 6 = 1.04.
3. Experimental procedure
A homogeneous negative DC electric field was generat ed in the ar r angement shown
in Fig. 1. The HV el ect rode was a 100 m z el ect rode l ocat ed 5.9 m above the eart h,
out si de the hi gh-vol t age l abor at or y. A 300 kV DC source was used t o generat e the
el ect rost at i c field across t he gap. Wi t h increasing vol t age the gap goes i nt o break-
down. However , this br eakdown process is not regular, but by furt her increasing t he
vol t age the br eakdown of t he secondar y gap becomes regular. The vol t age reached in
this stage was t er med the critical vol t age (V~rjt). The cur r ent gener at ed dur i ng the
br eakdown process of the secondar y gap was measur ed by a Rogowsky coil ( Pear son
El ect roni cs Inc. model 2877 with 2 ns rise time) connect ed to the anal ogue st orage
Tekt r oni x oscilloscope (model 7834, amplifier 7A19, with a fast measuri ng unit of rise
time 0.9 ns). The di scharge repet i t i on frequency ( DRF) of t he br eakdown impulses was
measur ed wi t h t he frequency met er Dat a Preci si on model 5845, which is capabl e of
measuri ng t he average of a number of fast pulses pr oduced in a t i me span of 10 s.
Dur i ng the exper i ment t he at mospher i c condi t i ons were: pressure 995 HPa, t emper-
at ure 4 °C and relative humi di t y of 30%. The current , and hence the charge released to
gr ound in each discharge, was also measured.
As ment i oned before, t he rise t i me for a secondar y gap of 1 mm is of the or der of
a few nanoseconds. Wi t h t he present measuri ng system it was not possible t o measure
the act ual rise time of the di scharge cur r ent impulses for gaps of the or der of
millimetres or fract i ons of millimetres. As a consequence, for small gaps, the cur r ent
measur ed coul d be smal l er t han the cur r ent flowing dur i ng the secondar y gap
br eakdown. A typical measur ed cur r ent waveform is shown in Fig. 3. The charge was
obt ai ned by i nt egrat i ng t he measur ed current .
Fig. 3. Photograph showing the current measured by the Rogowsky coil (31 and the oscilloscope (2) in
Fig. 1. Current scale: 1 A/div; time 1 ns/div; the charge can be calculated as: Q = 1 nC/square. The FE was
located at a secondary gap distance of 1 ram.
74 F. Roman et al./Journal of Electrostatics 37 (1996) 67-78
4. Results and Discussion
The mai n results of the present exper i ment are the cal cul at ed and meas ur ed cor ona
onset electric field, t he meas ur ed cur r ent and char ge st ored in the FE, and the
s econdar y gap di scharge r epet i t i on f r equency ( DRF) as a funct i on of the appl i ed
vol t age or reference electric field.
4.1. The c or ona ons et el ect ri c f i e l d
When the critical vol t age (Vcrit) was det er mi ned, its val ue was conver t ed by di vi di ng
by t he di st ance (D + h) t o obt ai n t he equi val ent electric field. The resul t is the critical
reference electric field for t he c or ona i ncept i on E¢~. By assumi ng t hat t he spar ki ng of
the s econdar y gap bri ngs the pot ent i al of the fl oat i ng el ect rode to gr ound pot ent i al
and by usi ng t he a mount of char ge rel eased t o gr ound in each di scharge, t he electric
field at t he t op of the fl oat i ng el ect rode, Eeartl a (see Figs. 2 and 3), and its pot ent i al j ust
before t he spar ki ng of t he s econdar y gap were cal cul at ed. Tabl e 1 summar i ses the
cal cul at ed and meas ur ed par amet er s at the critical condi t i on.
First, not e t hat the critical vol t age necessary for t he cont i nuous di schargi ng of the
s econdar y gap i ncreases wi t h i ncreasi ng s econdar y gap length. Thi s is r easonabl e
because a l onger secondar y gap requi res a l arger vol t age VFE across it to cause
br eakdown. Thi s in t ur n requi res a l arger negat i ve char ge on the fl oat i ng electrode.
However , wi t h i ncreasi ng char ge t he electric field at the t op of the el ect rode
Ea decreases. Therefore, in or der to mai nt ai n t he electric field at the t op of the
el ect rode so t hat a cont i nuous c or ona cur r ent is sust ai ned, the vol t age of the pr i mar y
gap shoul d increase.
Table 1
Summarised results at the critical condition
Parameters,[, Secondary gap distances d2
0.5 1.0 2
Vcrit (kV) 91 100 103
E,,~, (MV/m) 7.9 7.8 8.4
Ebk (MV/m) 4.7 4.8 3.4
Eck (kV/m) 15.4 17.0 17.5
Ee,rth (MV/m) 8.7 9.1 11
VvE (kV) 2.4 4.8 5.2
Note: dz, secondary gap distance in mm; V c r i t , measured critical voltage of the main gap; Eak , calculated
critical electric field at the top of the FE just before sparking; Ebk, calculated electric field at the secondary
gap of the FE just before sparking; Eck, calculated critical reference electric field obtained by dividing the
applied critical voltage Vcri, by the distance (D + h); E e a r t h , calculated electric field at the top of the FE for
the critical voltage condition but with "FE-earthed" and VF~, calculated voltage of the FE corresponding to
the critical voltage of the main gap.
F. Romhn et al. /Journal of Electrostatics 37 (1996) 67-78 75
Second, not e t hat t he critical field Eak at the t op of t he fl oat i ng el ect rode does not
change significantly wi t h t he secondar y gap length. Thi s is t he case because t he electric
field at the t op of t he fl oat i ng el ect rode shoul d be mai nt ai ned at this critical electric
field, whi ch is i ndependent of t he secondar y gap length, t o mai nt ai n a cont i nuous
c or ona current . The electric field at t he t op of t he fl oat i ng el ect rode, Eearth, is al so
gi ven in Tabl e 1 when it is gr ounded ( FE- ear t hed) and when t he appl i ed vol t age is
equal t o Vcrit. Not e t hat this electric field i ncreases wi t h i ncreasi ng secondar y gap
length. Since the critical electric field at t he fl oat i ng el ect rode for cont i nuous cor ona is
equal to E,k and as this is i ndependent of t he act ual pot ent i al of the fl oat i ng el ect rode,
t he above results show t hat the vol t age across the compl ex gap necessary for cor ona
i ni t i at i on is less when t he fl oat i ng el ect rode is gr ounded.
The ampl i t ude of t he measur ed cur r ent was a di rect funct i on of the secondar y gap
di st ance as i ndi cat ed in Tabl e 2. Usi ng Eqs. (1)-(6), the energy st ored in the FE was
cal cul at ed. Ot her par amet er s i nvol ved in t he di scharge are al so i ncl uded in Tabl e 2,
e.g., the FE capaci t ance, the cal cul at ed vol t age of the FE cor r es pondi ng to the critical
vol t age of t he mai n gap, and the secondar y gap electric field.
Not e t hat t he peak cur r ent i, t he char ge Q and the ener gy st ored in the FE i ncrease
wi t h the secondar y gap di st ance. Thi s is expl ai ned by the fact t hat the fl oat i ng
el ect rode pot ent i al difference VFE across t he secondar y gap i ncreases fast er t han the
r educt i on in t he capaci t ance. The char ge Q is pl ot t ed in Fig. 4 as a funct i on of the
secondar y gap di st ance. I n t he same figure t he val ues of t he FE vol t age VFE and the
electric field in t he cal cul at ed secondar y gap Eb are al so included.
4.2. The secondary gap discharge repetition f requencv (DRF)
The secondar y gap di scharge repet i t i on frequency ( DRF) as a funct i on of the
reference electric field Ec was meas ur ed using the syst em descri bed in Fig. 2. The
results of these meas ur ement s are pl ot t ed in Fig. 5. The behavi our of a FE i mmer s ed in
an homogeneous electric field can be deri ved f r om the curves in Fig. 5.
Table 2
Calculated and measured current related parameters
Parameters+ Secondary gap distances d2
0.5 1.0 2
i (A) 3 3.2 4
Q (nC) 9 15 21
C (pF) 4.0 3.8 3.7
VFE (kV) 2.4 4.8 5.2
W(I) (taJ) 11.5 44 50
W(5) (laJ) 10.6 36 71.4
Note: d2, secondary gap distance in mm; i, measured current on the earthed electrode, Q, measured
charge obtained by integrating the measured current; C, calculated capacitance of the FE; VFE, calculated
voltage of the FE corresponding to the critical voltage of the main gap; W(1), calculated energy stored in
the FE using Eq. (1) and W(2): calculated energy stored in the FE using Eq. (5).
76 F. Roman et al./Journal of Electrostatics 37 (1996) 67-78
25 12
20
~ e o ~o
5
- 1 0 ~
8 ~
6
4 ~
2
! !
0,5 1 1,5 ~ 2,5
s e c o n d a r y ga p di s t ance d 2 ( mm )
---q:3----- Q( n C) . . . . ¢, . . . . Eb ( MV/ m) . . . . O . . . . VFE( k V) ]
Fi g. 4. Ch a r g e Q, cal cul at ed el ect ri c field at t he s e c onda r y ga p of t he F E j us t bef or e s par ki ng, cal cul at ed
vol t age of t he F E c o r r e s p o n d i n g t o t he cri t i cal vol t age of t he ma i n gap, as a f unct i on of t he s e c onda r y ga p
di s t ance de. No t e t he i nc r e me nt s of t he F E c ha r ge a nd VvE as a f unct i on of t he s e c onda r y ga p di s t ance a nd
t he i nver t ed b e h a v i o u r of Eb.
g .
. 2
3000
2 5 0 0
2 0 0 0
[ ] d 2 = 0. 5 ..... - .................. ~ ...................
@ d 2 = l [ i
.... i ............. J---i ...................
t
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
' . " D
1 5 0 0 .................................................... ~,---~ .................. [ ..................
........................................ - I ......... i .................. i ................
1 0 0 0 t" ::
. . . . . . . . . . . . . . . . . i . . . . . . . . z ~ . . . . . . . ~ . L , , ; ~ . 7
i t ,,-* s" ! s . . " v
5 0 0 ~ la " ~ , o ' ~
.......... ~ .......................... ~,~,-,O ......... ~ ...................
. . . . . ~ . ~ - . . . . ¢ - ~ - . J . . ~ , , , t . . . . . . . . . i . . . . . . . . . . . . . . ~ . . . . . . . . . . . . . .
.............. -'~" -" ~ i
o d ~ i ~ ' i i
10 15 20 25 30 35
Ec Reference electric fieM (kV/m)
Fig. 5, Se c onda r y ga p b r e a k d o wn r epet i t i on f r equency as a f unct i on of t he r ef er ence el ect ri c field. No t e t he
i ncr eas e in t he r epet i t i on f r e que nc y wi t h t he r ef er ence el ect ri c field. The c ha nge in t he s l ope wi t h t he
s e c onda r y g a p di s t ance is r e ma r ka bl e . The best f i t t i ng cur ves ar e i ncl uded.
Fi rst , Fi g. 5 s hows how, f or a c ons t a nt s e c onda r y ga p di s t ance a n d f or c ons t a nt
i ncr ement s i n t he r ef er ence el ect ri c field (i.e., t he vol t age appl i ed acr os s t he ma i n gap),
t he di s char ge r epet i t i on f r equency ( DRF) i ncr eases a l mos t l i nearl y. Thi s is r eas onabl e,
becaus e of t he e n h a n c e d b a c k g r o u n d el ect ri c field, t he c o r o n a effect on t he sur f ace of
F. Romhn et al./Journal of Electrostatics 37 (1996) 67 78 77
the FE becomes enhanced and the duration of the charging process decreases. This
shows that the corona current increases almost linearly with the background electric
field. The mean value of the corona current Tc required to charge the FE capacitance,
can be obtained from
~c = Q x DRF, (8)
where Q is the measured charge from each FE discharge (see Table 2); DRF is the
discharge repetition frequency in Fig. 5. The current for the different secondary gap
distances was calculated and plotted in Fig. 6. The slopes of the diagrams showing the
relationships between the corona currents and the reference electric field E,: for the
three different secondary gap distances are remarkably similar. The measured values
of the critical reference electric field are in the order of magnitude of electrostatic fields
present during thunderstorm conditions. For that reason it can be concluded that the
described sparking process on floating electrodes with similar shapes as indicated in
Figs. 1 and 2 can be activated under natural atmospheric conditions.
Second, for the same applied reference electric field, the DRF increases with
a reduction in the secondary gap distance. This can be expected since the breakdown
voltage reduces with the secondary gap distance d2 for the range of secondary gap
distances studied. If the breakdown voltage reduces, the amount of charge that can be
stored without causing sparking in the FE reduces, and hence the sparking repetition
frequency increases for short secondary gap distances.
Finally, it can be concluded from these experiments that Eq. (7) gives the corona
onset voltage with high accuracy. Additionally, it can be mentioned that we have
presented mainly the experimental observations. The exact physical processes that
lead to these results are still under consideration. More detailed measurements with
other electrode geometries will be performed so that particular physical processes
shall be identified and quantified.
30
25
20
15
10-
5 -
0
10 15 20 25 30 35
Ec Re f e r e n c e e l e c t r i c f i e l d ( kV/ m~
....... O ........ 1 mm Ii ......... i ~ . . . . . .
4O
Fi g. 6. Th e me a n v a l u e of t he c o r o n a c u r r e n t r e q u i r e d t o c h a r g e t he c a p a c i t a n c e of t he F E i n Fi g. 2
o b t a i n e d u s i n g Eq. (8), as a f u n c t i o n of t he r ef er ence e l e c t r i c f i el d E~.
78 F. Romdn et al./Journal of Electrostatics 37 (1996) 67-78
Acknowledgements
The authors thank Ulf Ring for the help with the laboratory measurements.
Francisco Rom/m thanks the High Voltage Research Institute of Uppsala University,
COLCIENCIAS and Universidad Nacional de Colombia for their support as well as
L. Lozano for her time.
References
[1] F. Romfin, E. L6tberg, R. H6gberg and V. Scuka, Electrical characteristics of insulated metallic bodies
in a lightning breakdown field, Proc. 22nd Int. Conf. on Li ght ni ng Protection, Budapest, Hungary,
Paper No. R 6b-03 (1994).
[2] F. Romfin, The influence of a floating electrode on the breakdown voltage of a complex gap, Licentiate
Thesis, Inst i t ut e of High Voltage Research, Uppsal a University, Sweden, 1995.
[3] F. Romfin and V. Scouka, The influence of a series micro-gap on the breakdown reduction of a complex
gap arrangement cont ai ni ng floating electrodes, Paper 2171 presented at the Ni nt h Int ernat i onal
Symposium on High Voltage Engineering, Graz, Austria, 28 August-1 September 1995.
[4] F. Rom/m, V. Cooray and V. Scouka, The corona onset voltage as a function of the radius of curvature
of floating electrodes, Paper I 192 presented at the l 1 th Int ernat i onal Conference On Gas Discharges
and their Applications, Chuo University, Tokyo, Japan, 11 15 September 1995, pp. 192-195.
[5] CIGRE, Study Commi t t ee 33, Strength of external insulation during live line mai nt enance and repair
work with special reference to t ransi ent overvoltages, Task Force 07.02 of Study Committee 33: Britten,
Fonseca, Garbagnat i , Hutzler, Zheng-Jianchao. Electra N. 129, March 1990, pp. 108 131.
[6] J. Phillpott, P. Little, E.L. White, H.M. Ryan et al., Li ght ni ng strike point location studies on scale
models, Culham, 1975, Li ght ni ng and static electricity, Conference Paper, Royal Aeronautical Society.
[7] G. Hilgarth, Hochspannungst echni k, B.G. Teubner, Stuttgart, 1981, p. 72.
[8] ACE Program, developed by ABB Corporat i on under the name: The ABB Common Platform for the
2D Field Analysis and Simulation, Vaster~s, Sweden, 1993.

Sponsor Documents

Or use your account on DocShare.tips

Hide

Forgot your password?

Or register your new account on DocShare.tips

Hide

Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in

Close