ptd-ht-23-1126-71 iß · 2018-11-09 · mtwmwiwwi'«'""1'"...
TRANSCRIPT
■—ra "r-w PTfli r-w-- "■"»■»■" :•■ '■'■^•; —^"^*,'—■ "v^w
:
PTD-HT-23-1126-71
iß
O FOREIGN TECHNOLOGY DIVISION
N
Q «33
STUDY OF THE TIME CHARACTERISTICS OP THE ELECTRICAL BREAKDOWN OF SHORT GAS GAPS IN THE
NANOSECOND TIME RANGE
by
Yu. I. Bychkov and G. S. Korshunov
Rtproduced .from K^
D D C r.iif in /.,■
best availabl« copy.
Rtproduccd by
NATIONAL TECHNICAL INFORMATION SERVICE
SprinsftoW, Vi. 22151
APR 26 1972
^ E
Approved for public release; distribution unlimited.
\
'—M*^^^»—--■'---
I
;
"1 UNCLASSIFIED
iii'Lii'iarnTiTni.
(i««wr<»r tUMIIlMllg »I MM«,
Foreign Technology Division Air Force Systems Command U. S. Air Force
DOCUMIHT CONTROL DATA. R ft 0
7 CTWCT ff?Cl
u. mveur iiewniTv cktuifICATIO^
UNCLASSIFIED i», tMeu»
STUDY OF THE TIME CHARACTERISTICS OF THE ELECTRICAL BREAKDOWN OF SHORT GAS GAPS IN THE NANOSECOND TIME RANGE
4. OltCNIPTIVI NeTllf|y**»fM»«MMtflMllMl**«WN)
h »W»H6 JMfl/UjffiiflP. naa: u
! Bychkov, Yu. I.j Korshunov, G. S.
rmCTTäJrr 1967
U. CONTDkCT OH «NANT NO.
k. »ROJtCT NO. 7JJ
t * DIA Task No. T66-01-8 I«. OKTmauTIOM «TATIMIMT
I*. TOT»t NO. Or »««It
M. omttNAToirt MireiiT NUM*(III«I
FTD-HT-25-1126-71 SR erHm SiTSar SoS! MRT «MM
Ml« IMMI) S55S5 *•• «My »• M «IfMrf
Approved for public release; distribution unlimited,
H. mwm
II. tHONMUlNtMIklTAMV ACTIVITV
Foreign Technology Division Wright-Patterson AFB, Ohio
Delay and switching times were studied In the static and pulse breakdown of the short (0.05 - 2.2 mm) gas gaps within the nanosecond range. Given statistic delay was measured at E equals 500 - 1400 kV/cm, when the efficient electrons are supplied by autoemisslon from the cathode. The average given statistic delay was found to decrease sharply wlti the decrease in pressure and to be 0.7 nsec. at 20 mm Hg; a given time at E equals 500 kV/cra decreased from 1400 to 1 nsec, with the in- crease in the length of discharge gap from 0,02 to 0.08 cm. , The 14-fold overvoltage across the gap caused a decrease of /a given time to 1 nsec. Irradiation of the 0.01 - 0.02 cm long gap with the spark light from the discharger caused the 100-fold decrease in a given time. Switching time was studied in break- down of the 0,05 - 2.2 mm gaps with various gases at 1 - 7 atm. pressures and in breakdown of the 0.1 and 2.0 mm long gaps in the air at atmospheric pressure. In the case of static break- down, the near-electrode phenomena and Irradiation of the gaps were shown to be without any significant effect on switching process. A significant decrease in switching time was observed, when the gap was filled with Ar rather than hydrogen, nitrogen,, or air at atmospheric pressure. [ARgoilO^]
DD ^..1473 UNCLASSIFIED Stcuntv Justification
aaaaaimBMäiflaMimd
mmmmrmmmmmmm^ ™ « n wn 111 i wiw m\u^*^^i^^^^^m^m^***mmmmt^^ivmiat i wmu^
UNCLASSIFIED •MWlly CICMUMWtlM
■ ■V MMt
Irradiation Time Switch Pulse Generator
bfM« *
«•hi ■?
I I
! I
UN« •
••kl »t hIMN «
Mb!
vmm MltM
■w-'^—""---"■"—"■ üajMBai ■ ■ :..~.:^....:.-:. i ...
m» '-1 wmm wmm > mmmmmmm ■WUpp— .umiinnimiiwiiiniwiiiiiniiii^
FTO-HT-23-1126-71
EDITED TRANSLATION STUDY OP THE TIME CHARACTERISTICS OP THE ELECTRICAL BREAKDOWN OF SHORT GAS GAPS IN THE NANOSECOND TIME RANGE
By: Yu. I. Byohkov and G. S. Korshunov
English pages: 8
Source: Tomskly Polltekhnicheskiy Institut, Izvestlya. (Tomsk Polytechnic Institute. News), No. 162, 1967, pp. 180-185.
Translated by: Mr. John Miller.
ÜR/0000-67-000-162 .pproved for public release;
distribution unlimited.
TNI TIANH.ATWN II A IINOmON OP TNI OllOt MAL PMIWM TIXT «ITNOUT ANY ANALYTICAL OR •MTOtlAL eOMMNT. STATIMINTI OR TNiORIII AtVOCATieeOIIIPLIIDARITNOIIOPTNl MURCI AMDM NOT NICIUAMLV RtPLICT THE POIITION 0« OMMON W TNI PORIMN TICNNOLOOT 01.
TRANSLATION OIVIWM PORIION TICNNOLOOT OIVHtON ffP'An, ONIO.
FTD-HT-23-1126-71 Date n ^n 19 72
J
Mtwmwiwwi'«'""1'" mmmmßwmmnm .■ i i/-^^™^w™wT^wnrWfWWWWi!WPPWW¥<Tm^W»W*wpfrfP^fiTf^^
ü. S. BOARD ON QEOORAPHIC NAMiSS TRAN£.LITERATION SYSTQf
31ock Italic Transliteration Block Italic Traneliteration A t A A, a P P F ^ Fi r 6 6 B B, b C e C < S, B B ■ B V, v T T T *f T, t r r r Q, g y ) y y U, u a A ß D» d » 4> <» 0 P, f E • B Ye, ye; E, e* X X X X Kh, kh M m M M Zh, zh u u LI M TB, tB 3 • 3 Z, z H H h V Ch, eh H « H I. i U 111 LI Ml Sh, Bh R • a Y, y W Ut H Hf Shch, Bhch K x K K« k 1> t z. » n n » n L, 1 y II hi M Y, y M N M M, ra b b I » t
H N H N, n 8 • 5 • E, e 0 o 0 0, o 10 B *> M Yu, yu n n n P, P fl II Jif f Ya, ya
♦ ye Initially, after vowels, and after % hi e «Is^where. vIKen written as 8 in RuBBlan, transliterate» as yS or 8, The use of diacritical marks is preferrwd, but such marks- may be omitted when expediency dictates.
FTD-HT-? 3-1126-7.1
-w.....^,.,.,.:-.^,,-.,,..—nj mmjfrn^ngj
uppp wp■ff—wpwwp'i "'' "■ ■■," —wMPr*'»» wwinIIIU'^I. ■-1 ■i^iiniimniuTr»"''iFW'»> w'ufiiFn Mmqpimi ww ■ 111" pffw ■■^w^TTfr1'^"'**""""''' WB]
STbDY UF THE TIME ChARACTtRISTICS OF THE ELECTRICAL BREAKDOWN OF SHORT GAS GAPS IN THE (lAiJOSECOuD TIHE RANGE
Yu. I. Bychkov and G. 3. Korshunov (Presented by the science seminar of the Scientific Research Insti- tute of Nuclear Physics)
Only a few works have been devoted to study of the time charac- teriaties of the electrical breakdown of short gas gaps in the nano- second time range, although study of these characteristics is impor- tant for explaining the physics of the process and for practical use in high-voltage nanosecond pulse engineering. It became necessary to investigate the time characteristics of short gas gaps in the nano- second time range because of the recent development, at the Tomsk Polytechnic Institute, of a great many high-voltage nanosecond pulse generators [1-3] which have been widely used in research on nuclear physics, quantum electronics, the physics of dielectrics, etc., and because of the need for constant improvement in the parameters of generators, I.e., increase of the steepness of the pulse rise time and the response time during triggering.
In this paper we give the results of a statistical study of the lag and switchinjj time during the static and pulse breakdown of gas gaps in the nanosecond time range. Here the region of investigated gap lengths was defined by the real values of the gaps used in high- voltage nanosecond pulse generators, 0.05-2.2 mm.
FTD-HT-23-1126-71 l
Mi in M - -1" ■■■----'■■■■'■'-■ - ■■■'
iii>ii!iiWiii|,|iiiii.i»i'wm,iiiiMJ rpn—mwwi mm »mw 'WTUT'III'III I ii". "■" ." ""
Study of the Statistical Discharge Lag Time
As we know, the response time of a spark gap ocnsists of two
components:
'•-«n + V (1)
wnere ucT is tlie statistical lag time caused by the appearance of
an effective electron; t is the discharge shaping time. Since a.
Is associated with the expectation of an effective electron, It i^
a statistical value and has broad scatter. Ultraviolet irradiation
of the cathode (quartz lamp, spark discharge) creates a photocurrent
froj;. ihe surfaae of the cathode, by means of which the scatter of
OLT can be decreased. Fletcher [4] has shown that with Irradiation
of the spark gap with a spark of a nearby discharger o « 0.01 ns,
while T has no scatter and depends only on the applied field £.
Mesyats et ai. t5] have shown that the irradation effect is mani-
fested completely when the irradiation precedes the pulse by 70 ns.
The use of Irradiation is not always desirable, from a design
jtandpoint.
We studied the statistical discharge lag time for fields
E * 300-1400 kV/cm, i.e., when effective electrons are assured be-
cause of field enisslan from the cathode surface.
The experiment methodology is presented in [6J. The bandwidth a
of the registration channel is at least 3* 10 riz. We used an
ui-l4 Oodllograph. An automatic photodevice made it possible to
photograph the osclllograms of a great many breakdowns, up to 600
in each case. Such a large number of osclllograms give reliable
otatlstical distribution for the lag time.
Figure la shows the distribution of A-^ as a function of time,
where nt is the number of pulses with a given lag time ani n, is the
total number of pulse..;.
FTD-HT-J3-1126-71
liltuMi IBMII-'"-^-'-*"'■■'- If I'lllWllf "''"'''" fc**&n
inw,m„vm.'ww,™wv*w..ir7,m„r,vm,w„, ,,„,„ ,nm W^rr- ■"» ^m< jmi ,i.-r\mt"
a* | a Ts u u it * i / i '■ *" I . htgcm ! ,
Fig. 1. a - function A -'(«.». E ■
» l^JOO kV/om, 6 - 0.01 cm; b -
function /«I-(/,).I E ^ 1^00 kV/cm, I"» I
6 « 0.01 cm.
Designation: HCQH ■ na.
Thus, ^^L la tne relative number oi' breakdowns having a given «0
lag time. For example, 0.25 of tne breakdowns have a lag time with-
in 1.8-2 na.
figure lb showrf the same distribution on axes In — and t , where 3
n Is the number of pulses having given Jag time or greater and n^
lo tne total number of pulses. Tne term t shown in Fig. lb
is the mean-statistical lag time which oan characterize the degree
of scatter of the lag times. The greater the lag-time scatter, the
flatter will be the dependence of /,j ii: on t ana tau greater will
oe t With a decrease in scatter, t. decreases, 3 C;p. CT. ' ' 3 Cp. CT,
Figures la and lb show the distrlüutions of the lag time for a gap
of length 6 = o.Ol cm and with E = 1^00 kV^cm, with carefully polished
copper electrodes.
Figures 2a and 2b give the dependences of t on pres-
sure ana gap length, respectively, with constant field strength
E = 300 kV/cm. The electrodes are of carefully polished aluminum.
From the dependence in Fig. 2a we see that t. abruptly de- 3 Cp. CT.
creases with a drop In pressure, and is 0.7 ns with p = 20 mm Hg.
FTD-hT-23-1126-7]
^—■-■'■■ -WJ.........L-.. ■ ;tLu',i.ituA4^rluUi,.
Accorcilng to (1), o, '• 0.7 ns. The value of a in our case will t.e
determined by the fiel'"! emission current, a functioti only of field
strength E. dllth a rise in pressure, at t- ^ const a dnuulü rfciualn
lesd than 0.7 ns and, consequently, the time-lag scatter with a rise
In pressure which we jbserveci ax-e tlrne scatterü i
tue dependence of t, "I' Figure 2b shows
on the lengtn of the discharge gap for 3 Cp. CT.
E ■ 300 kV/cia and atmospheric pressure. In this case we have a de-
crease In t from 1400 ns to 28 ns with an increase In 6 from 3 Cp. CT.
D.O'J ::>. to U.ü8 cm. In small gaps a decrease In time-lag scatter
can bt- achieve! by means of nuperhigh surges. From Fig. 1 It Is evi-
uent that only with a l^-fold surge is It posslble to reduce
CT. to 1 1,3.
0, MM pmcm
dependence t3 CptCT(p). Fig. 2. a
E = 300 kV/cm, 5 ■ O.O'p cm; b ,(<5), E » 30
3 Cp.CT '
kV/cm, p ■ 1 atm(tech).
dependence t, "» up • i- i
atm(tech).
Designation: MM pr.cT. »mm Hg.
It should be noted that Irradiation of the gap by the spark of
a nearby discharger sharply reduces t , so that for gaps of 3 Cp■ C T•
0.01-0.02 cm, with irradiation, t nn oT decreases by a fact r of
more than 100.
component a ,
cp. CT. Consequently, gap irradiation decreases not only the
but also T..
FT D-HT-.3-1126-71
^MMIMHHtakaii lüMügiia^MMliM« ^gäMmaoM i
Investigation of tue Switcnlng Time
During tue iparK--;ap switching period the Initial voltage on the
eiectrodes U0 is reduced to the value f^ f^
i^e know [1] that with an increase in field strength E In the gap,
wnicn cdn it- acnieved by raising the pressure P or creating surge
ß, the swliciiln^ time uecreases. However, there is very little exper-
imental material on the study of switching in the nanosecond time
ratine during the static and surge oreakdown of narrow gaps. In this
regard we conducted a study of the switching time during the static
oreakdown of a gap with lengtn 6 = 0.0W.2 mm In various gases at
pressures P = i-7 atn ana surge gaps 0.1 and 2 mm long in air at at-
laospherlc pressure. Here the experiments were conducted with and
without illumination of the gaps by a PKK-5 lamp.
Tue experiment methodology is shown in [?]. Electrode diameters
d .vere selected so as to exclude the influence of interelectrode
capacitance on the switching time, described in [8j, and were pre-
pared for 6 = 0.0^-0.2 mm, d = 1,2 mm; for 6 ■ 0.^-1 cm, d ■ 6 mm;
and for 6 » 2,2 mm, d « 20 am.
Tne switching-time characteristic is the time t [i].
/„ f ,. j... ... 9£:^ (2)
waere 10 is the current amplitude, (di/dt) ^ is tne maximum steepness
of tne current time rise, and a is a constant which depends on the
type of gas. Fluctuations were detected during measurement of t ^
Tnerefore each value of t was selected as the arithmetic mean of 20 rt
or more measurements.
The static gap breaivdown wa- investigated line by line during
discharge (wave resistance z^ •> 75 ohms). Figure 3a shows the de-
penaence of t on pressure P and gap length S. With a decrease In
6 tne time t decreases, Just as with a rise in pressure, since tne
FTD-HT-23-U26-71 5
maHUAi^L ...v..-..:.1...'-J^.tv.i..A~.. ..y...^,..^,.........
w*m ■ •l"ll•, •"■■■" wmmmm J^^MIIIIIMIIIillllllllllMIJIMIIIIJII I
field .Hrerifctn Ui-Ti-asua In both caae..;. When E > 1^0 kV/cm the time
t_ is, for ali intents 3nd purposes, no longer* a function of P and f>.
We estimated the coefficient a using Formula (2) for the data in Fig.
3a. We obtained satisfactory agreement with the r.ürr.pe-Welzel theory
[9.1. The maximum of the dependences for 6 » 0.05, 0.085, 0.13 mm
also agrees with this theory.
S
it
2 n
Fig. 3. a - dependence
t(v,(p): 1 - 4 ■ 2.2 mm,
2 - 6 - 0.98 mm, 3 - 6 ■ » 0.7 mm, ^ - 6 « 0.1 mm,
t3-6»0.2min, 6-Ö"
= 0.05 mm, 7 - 6 = 0.085
mm, 8-6» 0.13 itun; b -
dependence t^E) for
various gaoes: for air
N'?, H2, U0 « 15 kV; for
Ar with E - 9.3 kV/cm,
bfl « 7-2 kV, for remalr-
i'ig points U0 » 11.5 kV.
KLY: (I) Air.
Designations: HCBH ■ ns;
ar = at; HB = kV.
U :o^ _j. i- iiV "* C,'"kM
The results give'i above for the time-switching study were ob-
tained with no lllurunation of the ^aps. With illumination of gaps
PTD-HT--?3-U26*71
nam tulrtiiiiHiiiiHiiiiii ti«iifm» ir rri liinlrn m I
WW——iaga
6 ■ 0.1-0.5 mm long using a PRK-5 leurp, no noticeable differences In time t. were noted.
It must b« mentioned that with 6 ■ 0.05; 0.2; 0.5 mm, the elec- trode material (Cu, Al, W, steel) and the number of Impacts ('vlOOO) had no Influence on t and the nature of the osclllograms UB(t). This Indicates that near-electrode effects have no noticeable In- fluence on the switching process during the static breakdown of narrow gaps.
Figure 3b gives the dependence of t on E for various gases with line discharge. Here, for air, N2, and H2, UQ - 15 kV; for
Ar, when E ■ 9.3 kV/cm, U0 • 7-2 kV, while for the remaining points U0 ■ 11.5 kV. For air, nitrogen, and hydrogen there is a tendency
to approach time t with increasing E. It is interesting to note
that in Ar at atmospheric pressure (which corresponds, for our case, to E ■ 9.3 kV/om, 6 ■ 7.8 nun), time t^ is much leas than for other gases. This fact has previously not been noted.
Pulse breakdown of the gaps is accomplished by feeding, to the studied gap, pulses with a steep front of varying amplitudes. The pulse front time was selected such that gap breakdown occurred on the flat portion of the pulse. We studied two gaps. With 6 * 0.1 mm the gap was Illuminated by the PRK-5 lamp; with 5 » 2 mm there was no illumination. Data on time t and surge J«^ where Unp Is the
''CT
voltage at which breakdown occurs and U„ is the static breakdown
voltage of the gap, are given in Table 1.
Prom Table 1 we see that identical values of t for gap 6 » 2 mm
are obtained with a much lower surge than for 6 ■ 0.1 mm. It is in-
teresting to note that for gap 6 ^ «i mm with ß ■ 1.25 the time t is
already several times lower than the time t which is obtained with
static breakdown of a gap with 6 ■ 2.0 mm, and p ■ 1 atm(tech).
FrD-HT-23-1126-71
- ■
»"ia-ÜiiWi^^.i^i.i.,.;^^;,,;
Table 1.
ImO.l.VM 5 1 2 3 4 5 6 7
Ua^\K» HCtK
0,65 0,8 0,75 0.7 0,6' 0,45 0,3
Km2 MM '» I.2S 1.1 1.7 1.95 2.15 3.4 2.«
üCT^S.ISW nee*
3.2 2,4 1.8 1.35 1,1 , 0,81 0;'/
[Designations: HCEH = n3, HB = kV],
The autiiora wouiü like to tnank G. A. Mesyats for posing the prob lern.
BIBLIüÜhAPHY
I. I'. A. Iliiiiiini.cu, f. A. Mtcif«. Tcxiiiii.n ()iopMirpooaiiii)i BUCOKODO.IUTIIUS naiio- .1*511.1111« iiMiiy.u.ooii. I'lKiirrtMiia't.ii, i'.HiX
■J. I". A. II o p o ii i. e H, P. A. M v c H ii. P. C K o i» in y n o n. nT3. M 2, 98. 10C3. 1. P A. Mocini, r. A. UojiofiLCB. 10. II. UUIKUD. PanmmmiM it MCKTpo-
urn». Ni 4, 10. I'.KI5. •t. K. C. rU'tchi'/. Phys. Rev., 76, IMI, 1919. 5. !'. A. Me CM it. K). II. Von u, r. C. Kopiu yiion. PaAHOTeitiiiiKii M MCXTPOIIIIK«,
M 5, «i'.'. IlKil. ii. K). 11. Ii u >i K o II, P. A, M r i mi /KT*. n ncia-, u. 7.1'. A. M >• c «ii, I". C. K ü p in y n o D. JKT<1>, D nciaTK. «. B. B. Kl'« Mil fin. P. A. Mccmi IIT3, Mi I, 175. 1956. 9. R. Roinpc, W. Weiirl. Zeit. Physik. 122, Uli, 1944.
?TD-HT-2--1126-71
UHm IM ■^■^■■-iiHi'niiiiiirirYnrrr'r i^^atunaämiiäiiä -,