Full resolution (JPEG) - On this page / på denna sida - 1958, H. 2 - Effect of Non-slandard Surge Voltages on Insulation, by Sune Rusck
<< prev. page << föreg. sida << >> nästa sida >> next page >>
Below is the raw OCR text
from the above scanned image.
Do you see an error? Proofread the page now!
Här nedan syns maskintolkade texten från faksimilbilden ovan.
Ser du något fel? Korrekturläs sidan nu!
This page has never been proofread. / Denna sida har aldrig korrekturlästs.
Initiation mechanism of a sparkover
The initiation mechanism of a sparkover in a gap
with a non-uniform field can he summarized in the
following way1,2. Initially a corona discharge forms
in the highly stressed regions round the electrodes
at a voltage which is lower than the sparkover
voltage of the gap. If the voltage is higher than the
sparkover voltage the corona streamers bridge the
whole gap with a velocity which is so high that the
necessary time for their formation may be neglected.
The next stage of the initiation mechanism is the
development of a leader stroke from the electrodes
of the gap. This leader stroke propagates across the
gap with a velocity depending on the voltage of the
surge and the distance of the gap not yet
overbrid-ged by the leader stroke. When the leader has
crossed the whole gap the last stage — the main
stroke — occurs and follows the ionized path
established by the leader stroke. If the voltage is
chopped to a certain value before the leader has
bridged the whole gap the propagation of the leader
can be stopped and the main stroke avoided. It is
evident that if we know the law governing the
velocity of the leader stroke it will be possible to
pre-calculate the impulse strength of a gap subjected to
an irregular surge.
The velocity of the leader stroke has been studied
by many investigators. Unfortunately, however, it
seems as if most results published either refer to
cases where information of the applied voltage is
lacking or incomplete or where the velocity of the
leader stroke has been controlled by the constants of
the circuit i.e. voltage drop across the series
resistance of the surge generator caused by the current
surge accompanying the leader stroke. In a Cigré
report from 19543 it was, however, shown that the
velocity of the leader stroke is a function of the
instantaneous voltage across the gap and the
sparkover voltage of that part of the gap which is not
overbridged by the leader. This, however, is no
complete solution of the problem since tests carried out
by Asea show that the formula given in the
above-mentioned paper can not be utilized on other gaps.
Method of investigation
A direct measurement of the velocity of the leader
stroke involves the use of either high speed
photography or photo-multipliers. If we, however, accept
as a hypothesis the idea that the velocity of the
leader stroke is a function of the instantaneous
voltage and that distance of the gap which the leader
has not over-bridged a much simpler method of
investigation can be chosen, which only involves the
use of ordinary high-voltage measurements.
A reasonable approximation in this connection is
to assume in agreement with measurements reported1
that the voltage drop along the leader channel is
negligible. With this assumption the leader stroke
acts as an extension of the electrodes. If the applied
voltage U can be kept constant during the
development of the leader stroke it is possible to determine
by experiment the time lag r as a function af the
voltage U and the distance s of the gap as
t = f(U,s) (1)
By differentiating the equation (1) the initial
velocity Vi of the leader stroke is obtained as
Vi =
(U,s)
(2)
If the assumption above regarding the leader
channel acting as an extension of the electrodes is true
the instantaneous velocity may be written
dx _
1
dt
3f-{V,x)
(3)
in the case of a constant voltage U.
If the voltage U is a function of the time i.e. U —U(t)
the propagation of the leader stroke is obtained as
the solution of the differential equation
(4)
By measuring the time lag of the gap when subjected
to a voltage varying with the time the formula (4)
can be used to check the truth of the assumptions
made.
Fig. 1. Circuit used in the measurements with a constant voltage.
ELTEKNIK 1958 1 18
<< prev. page << föreg. sida << >> nästa sida >> next page >>
