Document gDNK8XDbz1BE5Nmd1VzV8z4zq

* IEEE 'Conference Record of 1976 IEEE International Synposiu on Electrical Insulation, *].pp. 170-172, Montreal, Quebec, Canada, June 14-16, 1976. A FLAMMABILITY TEST FOR DIELECTRIC tt-UlDS L. Mandelcorn and R. L. H i l l e r West ln*houe R e s e a r c h L a b o r a t o r i e s P ittsb u rg h , Pennsylvania 13235 Suarary A B o d lilc J tlc n o f th e C leveland Open Cup T e s t, ASTM D92. v ss developed (o r d l e l e e e r l c f l u i d s In non( l r e hazard power c a p a c ito rs and tr a n t f o m a r s . This atas on the b a sis of c o n sid e rin g f i r e s i t u a t i o n s In a h e s e a p p a ra tu s . Two c r i t e r i a fo r n o n - f l a m a b l l l t ? were deduced and a p p lied in t h is t e a t : ' th a t burning xay only be I n itia te d a t th e b o ilin g p o in t, ond th a t burning cea se cn c o o lin g whan s e p a ra te d f r e e th e h e a t a o u re e . A number of f l u id s were t o t t e d . E ffe c ts of soot ere noted. Introduction '.'" v ia undertaking to dovelop a lte rn a te d le le e tr le flu id # (o r t r i - and paota- eh lo ro b lp b an y is, (or non* flan o a b le or n o n -fire hasard power c a p a c ito ra and d istrib u tio n transferable, I t v u re e llte d that there lo e need to d eteraln e the e s s e n tia l flu id (non-) flasm ab illty properties for these tp p llc ttio n s . Possi ble problecs of ecological pollution by chlorinated b ip h e n y ls, tl>at nay h a le t h e ir continued u ea, have necessitated searches (or a lte rn a te s . It Is the fluid component In impregnated power equipment th a t causes p a rtic u la r consideration for firm -hazard. A ttention waa d ir e c te d to u n d erstan d why f ewer equipm ent has been s a tis fa c to rily non-fla--sbla w ith these c h lo ri n a te d b ip h e n y ls - a s shown by e x p e rie n c e o f AO y e a rs so th a t the sane could be achurrad w ith nan flu id s . f"Tjj. 'JBile-was deenad that -th e levaland ~ I t vsa realised 2, which is Only when th e l a t e e r o ccu rs can th e r e be e x te n s iv e burning of the flu id , w ith oxygen av ailab le fron the am bient, s in c e th e re ia no gas space In a power capaci to r and the aoount of gas above the flu id in a tra n s former is vary H a lte d . Then, the f ir e could spread. Slnee the flu id la ig n ite d by an a rc , which dus to lta very high tem perature could Ig n ite alm ost any flu id , the magnitude of lta ig n itio n tem perature hat l i t t l e e f f e c t a t t h is p o in t. The s i t u a t i o n is h e lp le s s w ith in the u n it w hile a rc in g p o rs io to . C apacitors Impregnated w ith tr lc h lo r o b lp h e n y l have been consumed a lm o st com p l e t e l y when th e f a u l t c u r re n t s u s ta in in g th e a re waa not in te rru p te d fo r o long tim e, which could occur oaly ra re ly when th e c i r c u i t b re ak e r or fuse f a ile d to o p e ra te . However, burning w ith in a u n it should be ex tinguished very soon a fte r the cu rren t is In terru p ted . Hut m oreover. I t la re aso n a b le to i n s i s t th a t th e b u rp --.- . ln g f l u i d be e x tin g u is h e d im m ediately when such f lu id escapes from th e u n i t, and hence from the source of h e a t. D lls would prevent sp read in g e f the f i r e . i t should bo noted t h a t, under tbeaa circumstances, che tem perature in the mesa o f a burning f lu id cannot exceed i t s atm ospheric p re ssu re b o ilin g p o in t. This la b ecause tb a f l u i d I s wm conflaed, due to th e ru p tu re at i t s c o n ta in.e r . j \ ;. These c o n s id e ra tio n s lo a d to two o s a e n tla l flanmab i l l t y c h a r a c te r is tic s o f a s u ita b le f lu id fo r non flammable or m o s-flra hasard power equipm ent. I t should mot burn f re e ly la a i r below I t s b o ilin g p o in t. And, tt i t dons burn a t th a t tem perature, the flam e i s a x tlp g u ie h e d ^ s ao o n , a s i t . s t a r t s to c o o l wtnpi removed from th e -'s o u rc e o f h a s t i.^'TWe two o re l a t e r - *' ' tartY cliftula* r l - eod'*`"" ~ ' , ' ! ^ 1t i ^ 4 W w w b i d r t a M . ilatlfla4 as '" ^^UiiAabldJbdirhW '-xtuoys' g il y iftliib lc. - * *** M , u / t stiv e significane, ^ R a th e r Implying d l f f l c u i e i / ' `i t probably a lm o st a n y th in g w i l l b urn i t h i f f k l m t h a s t mod oxygen are-made av ailab le to i t . ' F risc o and 1. S. Matheel caused polytecraflu ro eth y lan a (Teflon) to burn, in A e la p le to o t where th e oxygen c o n c e n tra tio n and /''T em perature were hig h . 'T a sto 'foC 'daion-) fla s m a b illty . 'o f m 'rllgdtd generally g ive/eg -in te n sity fa c to r th a t w ill cause the liq u id to. ig n ite o r bum contlouously, " ' r ` '$ Jtestpe r a to r * t n th m * ln v p lR a d 'Open Cup T s s t and f g o i j j f t h t cori oMn t6k :((IU^ %C fO3 tlcetsll.O/ x y ti lndex '.jr i^ w i^ lfc ft^ x h e 1 f Inambili^7 d r f l r n -hasard probleo due 1` -, - '/ / t p ^ ild C d 'in a power ca p e c ifo r o r trans former is -. - t b M M 'by^An arc^ vhlcb^caa cause chlorinated biphenyls " t o ' burn i n 'I t s v i c i n i t y ,. The f a c t chat non-flaomahLllty l a achieved With .them,-'in Chose a p p lica tio n s, 1 there- .Jfe tA m o c u s e q u iv ic s lly -fn so c ia^ d w i t h . t h e i r h ig h f i s s e p o is t s ' i n t h e Cleve lam i _ 6 p d t f ^ T e * t :' i f A p n lrn d a n W -rh o :m n tirm p , |-- -- --f r .n___ X I :.*:**t * ? '*&:)? ,2... . *... An a d d it i o n a l problem i s p re s e n te d whoa f l u i d Is heated to th e atm ospheric p re ssu re b o ilin g p o in t and h ig h e r te m p e ra tu re s . ' Some o f I t d l l bn v a p o ris e d l a t e th e amb ie n t , which c o u ld c o n s t i t u t e on e x p lo s iv e m ixtwra w ith the oxygon. ,,Thin b atard can be am plified 'nr aven" dom inated by Che p re se n c e o f flam m able v o le tlltr ia s m d *' ouch aa h y d rogen, g e n e ra te d by a r c in g l a th e f lu tjd ^ _____ Such a s itu a tio n I s analagous to re le a s in g compressed : f l ammab l e g a e , from A t a n k , 'i n t o a n 'a r e s w here t h e r e ***- !r- e ra flam es n r sp a rk s. Thin m a tte r e f a p o te n tia l : . 7 . explosion a r is in g from a f lu id exposed to a f a ile r e arc should be d e a lt w ith s e p a ra te ly , and no t from the point of view o f f l a s s M b i l lt y , a lth o u g h th e r e I s soma r e l a t i o n s h i p . ' nrfci*tr V. *.*.,& - - .. . '- 7 . - . background - t l i m i t t l% ~ C o o a l d a r a t Iona : - y ,-tt p i e l e t t t k f l u id s tx p erim eiltai and R esults A flasm a b illty t e i t was -accordingly d ev laed ;to ~ determ ine che ig n itio n tomporacurc J o r continuous burn condition of a ing of a flu id , w ht)ib(^it was b o ilin g t h e n / " or- x ra n s f o r n a r, m h e c h a t;lt v i d tU M seftp .b u t"m a'i t e r - i t v o s -a q p. ok to 4 b 'ic9 T l l t y proper*- " from th e so u rc e .The events le e d lo breakdown (due tkon high current T est c e l l ; ist 'em tr^ii *Mgfc t e n p e r s t n fe h o c > la tb * fc a * j r ... C of the container. (Corning T t - J J ) ,. U aluminum b lo ck 'd in w i u w ^ b d . l h i ' - A. r*-A*.V * - .`dfj_fjae>.V*a ...^ux^r..- : , *.V * * NPC00026373 753710 : .1 tb e rro c o u p lc anti b rid g e , and a bonsen b u rn e r. The th rrrc c o u p lr lead v ire s were fasten ed to the cup H andle, and it* sensing bead was lo c a te d in sid e the cu p . about 1 /J J In Ire n Che b o tto e . A s n a i l p ie c e of T eflo n was a tta c h e d to th e bead to w eigh I t down and to keep i t iro n touching th e boated o c t a l . The hoc p l a t e and the a lin ln u a block were next to each o th er w ith th e ir aurtacea a t about the ease le v e l, to th a t the cup could be e a s ily noved Iro n the hot p la te onto the a lu n in u a b lo ck . They w are lo c a te d i n a d r a f t - f r e e e n c lo s u re . For any t e a t , th e h o t p l a t e waa alw ays a t i t s h ig h e s t te n p e ra tu re . and th e aluailnua block waa I n itia lly a t re taa p era tu ra. In a ty p ic a l te a t, a sa a p la of flu id In the cup was len t ted w ith the burner to continuous burning w hile iu ia tai on the h o t p la te , end th an noved onto th e s u r(Odd Si the a lu a in o n b lo ck end c o o le d u n t i l th e fla n * extinguished, tceparatura being recorded throughout, and a ls o tin e dyeing th e c o o lin g . AC th e Cine o f ig n i tio n th e f l u id use checked v i s u a l l y I f I t wee b a l l i n g . The cup vaa cleaned c a re fu lly b e fo re each t e a t , renew ing any s o o t, and wed f i l l e d w ith f l u i d to th e d e s ig n a te d l e v e l . A a p c c ln en wee esauned t o be b u rn in g c o n tin u o u sly when th e f l e a s p e r s i s t e d f o r 10 se c o n d s. When a f lu id waa te s te d f o r Che f i r s t t i n e , th e b u rn e r fla re v is applied for about one socoad every 5-10 C r i s e . I n a subsequent te n t the f l a n s n ig h t bo kept on th e su rfa ce fo r sev eral aoconde a t a lower bulk flu id tenperature to a ffe c t forced Ig n itio n . Or, tha saapla a t Ignited a t a higher, ta a p e ra tu ra than la* th e f i r s t t e a t . T his was tha n o et n a a a ln g fu l t e s t , p a r tic u la r ly whan th e a p ec ia e n uua ig n it e d a t o r M a r. b o i li n g . . -. . .... . ^ _ t ;n ->*> " ;* The follow ing flt&ds rare te ste d : pentaehloroblphenyl (Aroclor 1251), trlchtareblphenyl (MCS 1016), eln a ra l o i l , various n p iu tloar-of erlehlorablpbaay.l and a in arsl o i l , trierosylead trib u tyl phosphate, d lls o nonyl ptathalatn. and pelyd iaeth yl and polvnethylphenyl s ilic o n e (Dow Corning lOOjfJH^xar And/DC 350). T h * spectIvn'-narkar** ' M aarn^tdhfcsdhrdOd* A n tr im bastiM h m m i about IS C. both s ilic o n e o il req u ired re la tiv e );high terp e ratu re a to be ignited to continuous burnint. F ig s. 2 and 3. but once t h is occu rred , they continue: to burn during co oling. In fa c t, polydirethyl tilie e r e con tin u ed to b u m iro n about 290 C to le a s than 10 C w ithout i t s flo s e e x tin g u ish in g . I t should be noted th a t Che I g n i ti o n te n p e r a tu r e s of th e s ilic o n e o i l s end t r ic r e s y l phosphate a re about as high or higher then th at of trlc h lo ro b lp h e n y l, but they ere acre t li v e ab le by the c r i t e r i a p resen ted h e re. I t was obaervad Chat th e c h lo rin a te d biphenyl flu id s produced co n sid era b le soot and flesh es of fla re a t considerable d istan ces fron the specineo in th is t e s t , a s w e ll a s in th a C lav alan d Open Cup l a s t , when exposed to a f l o a t a t o r n a a r b e ll i n g . Tha so o t and fla s h e s t f f la n e s e e s to be a sso c lsc a d t- Son* f lu id ray be to rb ad an th a t o o t , e n d , p erhaps by c a t a l y s i s , nay b u rs even a t a r e l a t i v e l y low t a a p e r a t u r a . The p o s s ib ility of a f i r e epreeding this*way should be con s id e r e d . b u t i t i s n o t a s dangerous a s when a flu id Ilk a a in a r a l a l l la ig n ite d , where the bulk flu id bums continuously, w hile here only very sn a il nounti ere involved. D iscussion The r e s u l t s of t h is t e s t e re s u ite reaso n ab le. It Is sig n ifican t that positive fIn sta b ility resu lts, in d ic a tin g f i r e b a sa rd , were obtained w ith polyaethyl phenyl s i l i c o n e (DC 5 0 0 ), p o ly d ln e th y l s i l ic o n e and t r i c r e s y l p h o sp h a te wheat C lev elan d Open Cup l a s t f i r e points a r t e ith e r about the etna a s ,o r higher than t r l chloroblphenyl. These f lu id s burn fre e ly , a f te r being ignited a t ataospberie co n d itio n s, because af th eir re la tiv e ly high hydrogen c o n te n ts. However, th e a a g n lto d e o f th e C leveland Open Cup t e a t f i r e p a i n t . Of. th a i g n i t i o n te n p e r a tu r e i n i t s m o d ific atio n k a t a / ' l a .s i g n if i c a n t when o s t i U e f i i l a - "" conjunction . p ? . acturlAg/Vtc4 , adjfrjfba .nixed. ituutTertnntlfcr' t - ..'7.'trioeasTrt;pbaapheta'ft a g ^ ^ ^WBlF |`:etg :V S fy n S ^ a a td ' a ster were af pdytlculjfr^.lfvtaxaxt nncauan.-they'aru s c - >.; r ` ' e a a io a a lly costaIdarad *" * p v c isliy aneS^TRet~ire _ f ir e - ha Sard jposetg>&i|olgiiLir^J 'eat; lik e ain a ra l o i l , wham a f ir e through ashop aiehap or sute-ceabuatloo ban a f i n i t e , hopefully very anel l , p ro b a b ility . Acknowledg ..SVtf-V f o o e ttp n o f 'klrff fuC % f^.-i#*lV iotk'a A a o o tla u eu a * 10 " second,*!lena to extinction i t tha flane w hile cooling, ^ Tha e t i t t r l f r r i ( U K 'W ^ f ir * hazardous l a - V .____ ___________________J __ odly.. pontachlarobiaheo- Tha au th o rs w ish to .th a a k lngbeuea ftm aaidCX sW yst work _____ r _____ , _______ X' m ineral S l l ^ FiUfr phenyl d ll M t m t bum : i 'T it b o ilin g f s l s t V A *' ; * K M i *richlorobi phdntf c 4 4 I htid it* , so lu tio n * of up to f a c t ? p a in ta c h lo ro b i- . encenda whan ig n ite d a t elnaral o il is ' '*v ` ------ ^r b o il-.U , to follow- th? tni^riM brta^at^aS u turning f lui i n v e s t i g a t i o n . ."r*'if:' .* -- " trvparyi- 4 V." tin w e ^ J tV b u m , f s r . s e n t o I f oo-^ '* *ncU*ch> ^ cioa^D t.tkip K nll U jacocm f t trin t,,IM S ~ f ' tv - q>rf?s`iAf r * c ~*3*>'v ' r W 1 F i f ; * Thi d - t t f a 2 i b - t i > b a * tv e d ln ' mom*' JfJe e te .h a ta fo 'T tlN fF iip ^ jrtta te ^ ftfc 2, 2* IM Jyipdiiltti F lt e u b l lj lt y aid Saak* Tooting T e c h n iq u e * , U nriea*<igU i g i * *si ^ te iu o t^ itfc tu r a tU im -a . g3ki6 r .. ' ....... f if ' *V5-" !H>> H .^st#r>; Vi;V" i itfK1:<*- k^J3V'tf'( 1. v ' T- 753711 zu zsi S9Z000DdN ns k w & v W iN li * m tf lis ta m i q A um M R U i qp r* p|OU MI M M IMI 0| i>iU * M q n f TM(*3 u * $ ptMtMQ w x p w * uf spK>0 SMUlf *IU fiumana t t t '* US C SfunM u Mmoui i wo I U0(u6| pVXMiX ptitinfoip) Mvt(j i art * Mr) iwo p u ijm o m tpM i * VM1 jfl|M|Mfuof||WWOSiM|isww#U-l *ffj * . . -c >. . :'.- /V - X v;' 0 pmjiw n m iiio3 t I I I i 1*l -pCuydfqw <W W W f - S O 1UMIW M02 .f t ii M on " u o ItMMfW VM - >(uM4tiJOM3fJ* IKHM autW 4C *( -- SO |tJ u |W V 2 4 ...IkWMWrttlMl S* " (iugdqajo|tp|Ji ] 1 ntz - HUp jg y n u iin M 1 ~ 6um | q - ___1___ i f 111 03 sd i 0 fV I - ___1___ 00