Annual report 1952: Conference on Electrical Insulation (1952)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

- 7 Since SgF-Lo obaeTvei i t i s supposed that the activated complex dispioportionqtes aa follows; (SgFio)* S2F2 + 8F thus accoiuiting -^or the production of SgFg. Thermal decomposition of ^2^2 produces the SF2 observed. A l l other products may be accounted f o r by secondary reactions of these two gases. DIELECTRIC STRENGTH OF S F 5 IN A NON-UNIFORM FIELD C. N. Worlrs and T. W. Dakin Westinghouse Besearch Laboratories East Pittsburgh, Pa. This paper presents some of the ancmalles i n the d i e l e c t r i c strength of SF5 as a function of pressure and electrode separation i n a non-unifona f i e l d . Over certain pressure ranges the a.c. voltage break- down gradient may remain nearly constant or even Increase with increasing electrode spacing rather than decreasing as normally expected. This yields unexpectec^ly high breakdown strength, characterized by the presence of corona before sparkover. This was termed "Corona Stabilized" breakdown i n a recent paper by Cam i l l i , Gordon, and Plump.(1) This phenomenon w i l l be discussed i n view of the r e l a t i o n t o be shown t o exist between the d.o» and a.c. d i e l e c t r i c strength of SFg and the e f f e c t of electrode spacing on the p o s i t i o n of the maximum which occurs i n the ciurve of d i e l e c t r i c strength vs. gas pressure. The posi t i v e and negative d.c, 60 cycle, and impulse d i e l e c t r i c strengths as a function of pressure f o r a given elec- trode configuration i s also I l l u s t r a t e d and t h i s i n f o m a t i o n used t o i n d i - cate the time required f o r corona s t a b i l i z a t i o n t o take place. I n F i g , 1 i s shown the curve f o r positive d.c. d i e l e c t r i c strength of SFg as a function of pressure f o r a rod gap i n a grounded steel tank. I n the noble gases the d i e l e c t r i c strength w i l l usually increase l i n e a r l y w ith pressure f o r a while and then the curve w i l l f l a t t e n o f f . I n SFg the d i e l e c t r i c strength passes through a pronounced maximum. This maximum occurs i n a region where coiona piecedes sparkover. The curve of corona onset increases smoothly with pressure and at about 5 atmospheres sparkover 1) Gaseous Insu l a t i o n f o r High Voltage Transformers, G. Camil l i , G. S. Gordon, B. E. Plump, A.I.E.E. Paper 52-78, A.I.E.E. Transactions Vol. 71, 1952.

- 8 - occurs without the previous presence of corona. This i s not a new observa- t i o n as curves showing a maximum i n the positive d.c. d i e l e c t r i c strength of SF5 may be found i n the l i t e r a t u r e . ( 2 , 3 ) However i t i s believed that t h i s i s the f i r s t time the positive d.c. sparkover voltage on a gap as great as 2" has been reported. There I s considerable discussion of the mechanisms which may produce such a maximum i n the sparkover curve I n the l i t e r a t u r e cited. ( 2 , 3 ; Another anomaly i n the breakdown strength of SFg i s shown i n Pig. 2, where the 60 cycle d i e l e c t r i c strength a t atmospheric pressure I s pl o t t e d as a function of electrode separation. The capacitor plates pro- duce a condition which approaches a uniform f i e l d , a l l other electrode arrangements shown i l l u s t r a t e quite divergent f i e l d conditions. For such divergent f i e l d conditions, the d i e l e c t r i c strength may Increase more ra p i d l y than the spacing. Instead of less r a p i d l y as i p usually the case, producing an i n f l e c t i o n i n the curve. I n t h i s case breakdown i s preceded by corona which decreases the divergence of the f i e l d by increasing the e f f e c t i v e radius of the p o i n t . This unusual behavior of d i e l e c t r i c strength vs. gap spacing raised the question of whether t h i s characteristic i s r e l ated t o the occurence of a maximum i n the curve of pos i t i v e d.c. d i e l e c t r i c strength vs. pressure. To answer t h i s question, the positive d.c, negative d.c, and a.c crest d i e l e c t r i c strength vs. pressure of a 1 inch gap between a rod and plane were coiipared as shown i n F i g , 3o The curve f o r the a . c crest follows the shape and values of the p o s i t i v e d.c. curve rather closely. Having found t h a t the a.c. d i e l e c t r i c strength vs. pressure i s very close t o that f o r positive d.c, the next step was t o obtain the ef f e c t of changing the gap spacing on t h i s c h a r a c t e r i s t i c This i s shown i n Fig. k whldx gives the a.c. d i e l e c t r i c strength f o r a 1 inch and ̂ inch gap. I t i s evident that as the gap spacing i s increased the maximum l i i the curve occurs at lower pressures. Therefore, at a pressure equal t o or less than t h a t a t which the maximum f o r the largest gap considered occurs, i t follows that as the gap spacing i s Increased, at a given pres- sure, the d i e l e c t r i c strength w i l l increase r a p i d l y because we are ap- proaching a maximum I n the breakdown vs. pressure curve. Hence, curves f o r d i e l e c t r i c strength vs. gap spacing are expected t o be as shown i n Fig. 2 . With negative voltage on the sharp electrode, the d i e l e c t r i c strength increases continuously w i t h increasing pressure f o r a l l elec- > trode configurations and spacings so f a r tested. See Fig. 3 . I n t h i s case corona always appears t o precede breakdown. 2) The Effe c t of Pressure on the Positive Polnt-to-Plane Discharge.in N2, O2, SF6, etc., H. C, Pollock and F. S. Cooper, Physical Eeview 56, 170 (1939). — 3) Positive Point-to-Plane Spark Breakdown of Compressed Gases, T. B. Foord, Nature I66, 688 (1950).

- 9 - The positive impulse strength, shown i n Fig. 1, shows a maximum at about 1 atmosphere pressure. This maximum i s only a l i t t l e more than ^ the value of the maximum of the p o s i t i v e d.c. curve. When a.c. voltage was applied t o t h i s rod gap by a h i g ^ voltage switch the breakdown strength was not decreased. These breakdowns usually occurred i n one cycle or more. Therefore I t appears that the time required f o r "corona s t a b i l i z a t i o n " i s much less than one cycle (of 60 cycle voltage) but greater than several microseconds as indicated by the Impulse curve. More work should be done on the e f f e c t of time on the breakdown voltage. A f t e r reaching a maximum the Impulse strength decreases but s l i g h t l y , and then Increases w i t h i n - creasing pressure g i v i n g higher values than po s i t i v e d.c. This indicates that more than a few microseconds are required f o r the positive streamers t o complete breakdown. For t h i s electrode configuration the corona onset point was determined w i t h an oscillograph and an a n i p l i f l e r . The negative d.c. corona onset curve i s lower than that for.the p o s i t i v e . This i s i n agree- ment wi t h the data of Pollock and C o o p e r , T h e a.c. corona onset curve appears t o l i e between the p o s i t i v e and x i e ^ t i v e d.c, curves, although recent work on short point t o plane gaps has shown that the corona onset f o r d.c, and a.c, i s the same f o r the same p o l a r i t y respectively. 2) The Effect of Pressure on the Positive Point-to-Plane Discharge i n 112, 02, SF6, etc.. H. C. Pollock and F, S. Cooper, Physical Eeview 56, I70 (1939). ^ * v

FIG I NEG D C POS IMPULSE P O S D C POS D C CORONA:; 1' P O S D C NEG D C > 8 0 60 CYCLE A C »-POS IMP POS D C CORONA - A C CORONA NEG DC CORONA . - N E G D C CORONA ATM I 2 ATM I 3 ATM CCORONA 1 2 0 1 6 0 2 0 0 P R E S S U R E - C M HG D I E L E C T R I C - S T R E N G T H OF SFg FOR 2 " G A P B E T W E E N Vz'SQ R O D S IN 12" DIA S T E E L TANK A S A F U N C T I O N O F P R E S S U R E 3 2 0 ^UNIFORM FIELD \ ( E S T I M A T E D ) CAPACITOR PLATES \ ROD TO BOD ROD TO 6 PLANE 10 80 •y SPHERE TO PLANE ^ 4 0 " F R O M PAPER BY CAMILLI GORDON AND PLUMP AIEE7I — F R O M WESTINGHOUSE RESEARCH LABORATORIES 0 50 100 150 2 00 2 50 3 0 0 ELECTRODE SPACING-INCHES 3 50 FIG 2 D I E L E C T R I C S T R E N G T H O F S F e A S A F U N C T I O N O F E L E C T R O D E SPACING F O R A NUMBER O F E L E C T R O D E C O N F I G U R A T I O N S AT I ATM P R E S S U R E AND 2 5 ° C 2 0 0 1 6 0 bJ t 8 0 4 0 • ^ 0 POS D C NEG D C 60CYCL 0 • • A C 0 0 1 A 1 2;5iT M 3ATI A 4 ATM FIG ? 2 0 0 1 6 0 1 2 0 (li (E O > 8 0 4 0 4 0 8 0 1 2 0 1 6 0 2 0 0 ^ P R E S S U R E - C M H G A B S 2 4 0 2 8 0 3 2 0 FIG 4 \P * lA 1 T M 2 A T M 3 A T I 1 II i_ 4 A T M 1 4 0 8 0 1 2 0 ISO 2 0 0 P R E S S U R E - C M H 6 . 2 4 0 2 8 0 3 2 0 D I E L E C T R I C - S T R E N G T H O F S F g FOR A T G A P B E T W E E N l^'SQ ROD AND A 6 * DIA P L A N E IN A 12" DIA S T E E L TANK A S A F U N C T I O N O F P R E S S U R E A C D I E L E C T R I C - S T R E N G T H O F S F , F O R Mt'AND I ' G A P S B E T W E E N V£SQ R O D A N D 6 ' D I A . P L A N E I N 1 2 * D I A . S T E E L T A N K A S A F U N C T I O N O F P R E S S U R E