C.P. No. 223 (17,993)

A.R.C. Technd Report

C.P. N;;) 223 A.R C. Technical Report

MINISTRY OF SUPPLY

AERONAUTICAL RESEARCH COUNCIL

CURRENT PAPERS

Report of the Third Year’s Flying on the Development

of Flight Testing Techniques for Finding and Measuring

Natural Icing Conditions

G C. Abel, &SC., D.I.C.

LONDON HER MAJESTY’S STATIONERY OFFICE

I956

THREE SHILLINGS NET

C.P. I'io. 223 neport NO. miz3,1ke~/205

18th January, 1955 Report of the third year's fly-in& on the developmnt

of flight testmg technrqws for Ym3.i.~.d m.aiu-i~ natural ~.cin,~ condltmns .

BY

G. C. Abel, &SC., 3.1.C.

This report covers the third year's search for naDaal icin:: cond~tlons to&her rrith some results for the vhih0l.e three year period. Vamoue form of unserv;ceability limtzd this year's flymg to approxinmtcly four morlths. Ear the tlree year penoil only half the icing forecasts for layer type cloud over U.K. yielded icing conditions suffm5ently continuous to allm an znti- icing system to be tested. i, number of thesc were too lk,;~tod to allow more than one test nln. By lurntlng testl;ng to the best synoptic candxtlons yet found, forocnsts w.11 probably be co:;nsletc:l.y correct &ltho,agh very llnrt~d m number. h good coetmg of we -as obtained from almost all forecasts for cLn11u1us clouds. Cccnslons on winch scins scvent y was o,rcr?sti.uatr.d durxng tne past two ye%?3 m-e d0ub-h those 12 tbo frrst year, ~resU;;,bly due to the smaller numbor 0.. upper air soun.?zmgs w~zdroie. Tht tech;llTde of photograph- ~-n< the leading edge of the tailplane has produced several 1ntcrLstug stutiwa of xx fomatlon. Ihc paHAy to the lzuutcd flyin: time no great advance in li~stnment-ntlon has bcon achievc3 thin year.

/L& of Contents.,..

-2-

List of Contents

1. 2.

3.

4.

5‘

6. 7. 8.

Introduction Details of test aircraft snd its instrmenta~on 2.1. hrcraf t 2.2. Instrumentation

2.2.1, Oiled slxde ssqlmg cqupucnt urd r.<cro cmera

2.2.2. Ted caxwra 2.2.3. Electrolytic 1-nndscrcen 2.2.4. Thermal ice detectors 2.2.5. Rotating discs

K&hod of test 3.1. PErLOd of &Its 3.2. NOma tests Results

5;:;: Success in finding Ice Icing conditions mssed &LVJ to unserticeabllity

2:: Tyye of 3.ce formation Instmmnt readmgs

4.5. Electrolytic vindscreen Iiiscuss1on of results 5.1. Conditions suitable for testrn: aucraft r?e-lci.n~:

or mti-icing system 5.2. Severity of acing conditions 5.3. Ice formation 5.4. Readings of uX&.vidual mstrucnts

5.4.:'. Rotating cylinders 5.4.2. Rotating dxscs 5.4.3. Oiled slide sapling e~Uiplltilt and micrc,

camera 5.4.4. iJ i L. L.C. - Smiths icmg mS.ioator

Conclusmns Xxrthcr dtvelop~~mts References

List of Tables

Ccmparrson of ileteorol gical forecasts and concktlons encountered Surirmxy of rneesurements made in icing condltlons Cmptison of severity of icing forecast with severity of icing

observed Values of' Rg calcuiated for thf rotating cylinder neasurments

compsred with observed icing severity

List of Illustrations

Ice formation on leading edge of tailplane

Ice formation ?n leading edge of taili>lme Ice L"orwtion on the t~lplane and other parts of the aircraft

(Flights 76, 65, 73, mtt 75) Diagrmrmtic layout of modified olled sli?& ssmpllng eqtipuent

and micro camera Record of height, speed and xing indzcatxons during severe

icing on flight 73. Compnson of port and starboard rotating disc recor5.s Comparxson of port and starboard rotatins tlrsc results CalxbratiLn of smiths ice detector agolnst rotat;% disc

/Introduction....

G.P. No. 223

ERRATA

Page 5. Last word "Includwi" should be"Including".

page 6. Line 4. "larger" should be'layer". Para. 5.2, line 9. the "i" from dying has been omitted.

Page II. Ref. 4, line 2. "ellipsord" should be "ellipsoid".

Table I. Columns 13, 14, 16 and 17 should contain the following additional figures.

Columns Nos.

13 24 16.53

0.02 -~- 16.55 16.55 16.58

17.0% 17.43 27.43 17.43 17.54 18.21

18.36 18.48 19.13 19.31 19.41 19.59 20.08 20.13

20.28 20.33

16 11

0.19 0.17 The date should be 8.7.53. Not 3.7.53.

0.17 0.11 0.21 0.06

0.06 0.04. 0.14 0.09 The date should be 2.2.54.

Not 2.2.59. 0. oy 0.06 0.11 0.04 0.22 0.10 0.15 0.06 0.14 0.07 0.30 0.12 0.27 0.16

Ykpressionw should be inserted in Col. (4).

3.26 0.23

In Table 2 flqht 72,line 3, colmn (6) "-775" should be tt-7.5tp.

-3-

1. Introduction

Work has non been proceedm~ at the i. $ L.::.C. for three yxars on a jomt programe with the 8.b.C. ar,d t!lc Xetcorolo@m.l Office designed to discover nom about zurcraft Icing.

The aspects of the problm on ntic!? work has been ooncentrated have been:

(I) Correlnti?q the occurrmc~ of icing cmtitions vlth the Net. Office forecasts.

(2) ikveloping uwtnmcnts to r~~aswi*e the paraixters on nlvch icing conditions dqmxd.

ti 3 Studjrrng ice Pormtion on azrcraft arid Its effects. 4 Developing water sprays to produce LOC in zUight in clear air.

The f'lrst +XTO years t;ork has been reported in reforonces I and 2. The present rqort deals ijlth Ltcms I - 3 for the period from June 1953 to NW 1954.

2. Details cf test aircraft md Its instmwntation.

2.1. hrcra*t. is dewmbcd in reference 1 the aircrlft used was a Vlkang, a b&n en,$.ned nircrzft designed to cnrr~ 24 passengers.

2.2. Instrwientation. The najonty of the instrumentation vas tl-e sane as described in reference 2 but the mucro camera had been redesigned md repositLoned to bring ~11 i-he major instruam:ents, with the exoeplxon of the rotatm;: cylinders, under the control of one operator. This operator controlled the auto observer, t'ne two rotating discs, the fixed cylmder wtich was seldom used ralnly due to lack of tl.le because of other comtments rrhen lil ian;: conclitxons, thz bslenoed bqdge thermmeter, the HussenotL20 recorder anii the redesqasd macro cax~ra and the oiled sl.Ae saqLling W+likli;nt , 3etails of thh: ch3~zs 13~de to the lnstrwentatxon are given below:-

2.2.1. Oiled slule sax~lri? ., Lqv.xpmcnt and micro cmera. To enable szu.@es to be photographed at a tcqxrature below frcezlng the uicro cmxra had been nodiflLd to allcm the aled slCie to be brought straght into the positlan u which it was >hotographsd, uuedutely after sa@i.ing:. The ix~crosoopo stage was kclt cold as before by ear from outsid6 the arcraft. A di~&r<am.t~c sketch of km this ms achievcti 1s ,gxven in Figure 4. 6 tube nas fitted through the KLCTO camera >slth a gap at the v~~roscq~o stage for photography axd a. slide carrier was inounted on thu end oi a flemble cable running throurih thu tube. On the inboarj. sale of the macro camera there ms a gap in the tube to ~L~low a frerh sLdc to be fitteA to the slule carrier, bile beyond the gap the tube was slotted to control the angle at vhwh the slide vu, hell. On the outboard side of the 111cro ca~?';ra cno tube ran lnslde an outer tube vbch in turn was inslde a stroanCl.inod outer cover. This outer cover with tne two tubes In It pissed through the skin of the aircraft and projected sane +xo feet snto the axrstreau. Ln air intake on top of thx outer cover duected am air flow lilt0 the rxxo oaxerp. to ::~a.xrAim the twperature at the uuxosoope stage below freezing. As before the lens and the stage were sh.xelded from the dxrcct blast of the air to prevent water dro+ts from contaminating the sample when flying through cloud. The imar tube was fixed so that Its outboard end coinmded nlth the Lnd of tho outer cover but the scoond tujc could be pusned out Into the air stream amd vxthdrawn again by swans of a rod :rhxh also passed through the ixicro camera to the inboard side.

The nlethcd of sa~&u1?; xas to insert an oi.le.3 slide Into the slClc curler on the mnboai-d side of the 'inoro caa~rz and then to i:ush it outboard by fixeans of ?;he hmdlc -xojectux,, 7 throu,$ the slot in the mnboard tube. The slride passed through the uuoro caera stage ad through the xi.rcr&'t skin until it was near the outboard end of 1% i;"idxng tube. Lfter a. short uxtcr- ml to make sure that the 0x1 was below frouzuq tozriiil-raturc, the outer tube

/riolad. be.. *. .

-4-

would be pushed beyontl the end of the outer cover, the slide would be turned from horizontal to vertical by passmg the oontrollmg handle through the appropriate part of the slot in the lnboord tube and then r~ould be push& out to the end of its travel. The slide mould then be beyond the end. of the outer cover but was still shielded from the sustrem by the outer tube. To make the exposure the outer tube was retracted mto the outer cover and then pushed out to Cover the slrde again. The slldc nas j~~~mliately withdram into Yne photo- graphin& position and photographed as quickly as possible while the outer tube was retracted into the outer cover to prevent Ice from building up on It. After photography was oomplete the ~llde was brought to the inboard side of the micro cmera end removed. The whole scquenoe could then be repeated.

2.2.2. Tad oamcro.An F73 35 m camera wss mounted at the rear win&m for photograp'hmg the ice foru.tmq on the lead&G ed,rr,e of the tazlplane. A thin 2" graduated rod ms mounted f'rm the leading e&ge to show the thickness of the Ice formation, the outer half of the rod nas mrked m $,, but the inner half m~s marked in t". The camera was operated by a T35 control wLnch could be set to take photographs at intervals of f~rorr 2 - 55 seconds.

2.2.3. Xlectrolytiz windscreen. L sme.11 test electrically heated wandscreen approximate& 8" sw~~ro was mounted in front of the fliht engineer. Reat ~ras~$duoed~b~ pass& SAL alternating current through en- electrolyte ~&ile the ter~peratze was neaoureZ by themooou$es mounted between the two pairs o f &ss l&nations that an&e up the front and rear faces of the screen.

2.2.4. Themal ice detectors. Several ferns of thermal ice detectors were fitted for various periods throu,ghout the jzar but no final production model was made available for test,

2.2.5. Rotatinq discs. Two rotatin;: il~scs were fitted, one half 'AQIY back in the original position on th c starboard side while the other was Well forward on the ;,ort side.

3. Kethod of test

3.1. Period of tests. Ihs tircmft nas grounded for several long periods so that the tuli: avallablc fez test fljmnlz nss only about four months m the whole year. The first ti,?o pcrz~ods of unserviceability which were separated by a lieek were in June and early July anc Bert ciue to modifications to the riain spars l;'flat Were g:dnerd to xl1 Vilangs. In July ihe arcraft was unserllceable for one week but was then avaz1abl.e till the 10th August when sane modifications were made to the instrumentation, Very shortly afterwards it nas decided that the fin and taQlane roust be investigated by the Contractur for possible corrosion. Althovgh a replacenent fin was not available the fm and tailplane were sent to the firm and the arrrcraft d&I not f%y agzun till the 11th Deceriber after the original fin had been returnal. Various radio snags, bad neather and the Xmas break prevented any teat flying until the 3lst December. From then till the boe&~~~ng of April there was an accumulation of only about one neck's unserviceability and four days when the slrfield was unseticeeble because of snoi~, so the majority of the years' flying was concentrated In ths period. On the 5th April one engine failed and coul& not be replaced before a ujor inspection fell due. A C. of A. inspcotion was performed conourrently nlth the na~cr znspectlon and soile modifications were also made. The Inspections and modifications were not completed before the be&wing of June.

3.2. Normal tests. The no-1 method of test was in Tenera the same as in previous years. The Met. forewst ;ias obtained each morning and when icing conditions were forecast a search was made in the area. After ioe nas found measurements 7-were made IcLth all the avarlable instruments and photogrzghs were taken of the ice formtion. Pkasurements -;ere ncrrrally continued until no further worthwh-lhile lclng oondittlon could be found, or tests had to be abandoned due to the accw&&icn of =ce on the alroraft. Orring to the very Lmitod peno& over rhlch the aIrcraft tics avazlable some flights were made when the forecast was less pro;llsin:; thz.n the forecasts used in previous years.

/I;. . . .

4. Results

4.1. Success in flndlng 1ce. 'kble I sham the conGitwns uniter vrhlch ice was found conoared to the conditions that WWF: forecast bv the N&Office. Thcra irere 19 cncountcrs witi ice in 17 flight s whde on 3 flights no ice or only very slight xcing ims found. One of the successful encounters was obtamcd by using informa'aon Q-cn~ other aLrc.r,aft that i~we uxpemwcin:: I.~UTT contitlonn :?hilo ,?11 the others were obt,uned by :'oll~~-&n~ the Met. forecast. Of the 3 failures one T-vs due to cmmlus cloud not hating bxili &XV: frecx.n~ level by the time the olrcraft vvos axrching the zrea. Lt is l>C,sSiblC tiiat it my have built up later on in the &y. The chances of fsn!lin~~ ice forocnst for the flight on which the second failure occurred vcrc not wry :;xxl but sew&i to justif'y a flight to test c? modi.fYcatio~~ to L new iw detuctor. !cno front in nhlch the loiq{ wa.s oxpectca was lane vherc a. search COT&~ not be rxdc.

echll$y%%o lymg a-Long cm -I?? 9x0 third f,ai.lure nna only partld but

only very slight ice wz.a found FS the r&win bo&y of cloud was boluor freezing leveltith only stall peaks above. In this case there was x broken layer of clmd. Fbave, 1.n which it was just posslblc icing eentitions found but, aiing to the incxpencwc of the observer in char::c 3s not tried.

4.2. Icing; concZjtxms xi osed ~duuc to unncrv~ccaM.lity. Durin? the ezght months of unscrmility thcrz -erc 40 ~'dd for,rccnsts of icing that

5. Discussion of results

5.1. ConCations suitAle f3r testing: aircraft de-iclw or anti-icing SJsfens. OS the 20 flights made 3 were mado pnr~nrlly to test

instrwmnts and 4 others sfcre not successful in fin<G+ a really useful icing codition. Of the renai~n~ 13 fE;hts 8 nch~cvci. therr success nainly from cu~ulw clouts dthouzh these were cm3e~~~Xi in n lzyer of strcto o~~~ulus in tvra cases. This leaves 5 flights in lzycr type cloudvvh~ch r.u,rrht wdl hnvc been satisfactory had 2n icill,y system been w&r test. f@inst these 5 flithts e further 4 were nttcmlpted on forecasts frcn mhoh It was hoped that an adequate quantity of ice noulc? bc feund. lWc of these 4 rvere rather lnconclusivc ~3 con5litims were sush that It is psmblt t‘nnt n r.lore cxpcrienccd observer night have been able to ~cnicve greeter success but at least 2 of tha 9 forecasts did not lead to suxtable cond3.tions for tcstiq h". icing systw:,

/Included.....

-6-

Including the fll+hts :&,c in tha previous pars there is a total of 20 flights on which a test on an icing system might well have been :&e, against a further 18 fcrecasts from which it was hoped to obtazn good test c~ndiitrons. This shows that, over the British Isles, the duration of ice in larger ty;,,e cloud throughout tne year is unlikely to be sufficient to alla? even one satisfactory test on an icing systw3 in nearly 50 per cent of the connc?xtions forecast. The conditions suitable for several test r&m. on a system rep-cscnt a still smaller perceni+Te and are almost invanabl; confined tc the winter season. The picture is net, however, q uzte so blazk as it ai>?efls as it represents layer type cloud from all types of synoptic conditxons throughout the whole year end then only when the aircraft was serviocable. If testird is confined to t

9 xinter period and to the most pomisiq conditions rihc.c;l

so far seem to h u&able mrstream with a very lm freesin-, level, it seems p&able that almost all forecasts nfll be correct even though their total number still not be very great, possibly as low as 5 per year. Further experience may also allow fliqhts in other synoptic conditions that should 5e nearly as suitable thereby increasing the number of possible forecasts that are almost certain to lead anaircraft to the required conditions.

The picture on cul~ulus cloud flyin; 1s very much better. For this year 8 out of 9 forecasts were successful vihile the zZtwre over 3 years IS 30 cut of 34. The duration of irnn? conditions in cumulus clouds is too short to be of much value for testin& anti-icing systems. A good coating of ice can, hwever, be built up and this might be used for testing some types of de-ioinb systems , particularly nech,uical ones. In ad&tion the behaviour of an air- craft with a coat of ice on is in itself iprtant to allow for the case of the protection system failin+ it the ssme time, unprotected items such as tank vents, for instance, say be checked for i>ossible failure in ice.

5.2. Seventy of ioir;: conditions. Table 3 shc&vs that the forecasts of the severity of icing condationo were net so :rood ES in nrevlous .vears. In only 12 out-of 20 forecasts ';'ere the correct icing seve&ti.es found. ThlS is ;?artly due to searches being made on forecasts that gave a smaller chance of finding ice than in previous years, In one such case a search could not be made in the condition associated vj,th the forecast as the only active cloud was aonfmed to an sir lane. In another case tbc, auxm& did not take off till so late that '$ $8 doubtful if the forecast severLty still aL?:;lred. In yet another case t, e9pao%vity was t "y'.% out more rapidly than hail been expected. SO that by the time the <aircraft reached tne con:2ition only light icing could be found. On one occa,ion the search had to be curtailed after a late start CLue to lack of daylight although it is believed that the full forecast severity was present in the condition, while once the sersch may have been made a l.a.ttle too early. On the remsirnng three forecasts on at least ~-TO occasions there did not al'pear to be any cloud above freezing level capable of sustaining more severe conditions than these found. It seems probable thatvath more recent upper air data, cloud conditions could have heen forecast more accurately. This theory is borne out by the fact that over the past ti-ro years the number of oocasions on which icing severity was underestimated is double that of tne frrst year when more soundings were available.

As in previous years thevalues of Rg, the rate of build u;, on a standard ice collector, have been calculated from the rotating cylinder resuits and compared with the severities observed on the flight. This comparison made in table 4 fits in exactly vai.th premous years results that the values of Rg above between 4 and 6 were obtained for severe ioinz wrtile values of Rg belV;f about 2 - 3 indicnted lqht icin-i.

5.3. Ice formation. The catlera photograping ice formation on the tail- plane has produced several interesting sets of results, mcluoini; some unusual formations caused by runback. The first photograph in fi,pre 1 shaz.s the formation on the leadmnz edge of the tailplane on flight 69 when the tamgerature and limd water content were both fairly low and the droplet size was small (about10 - 12 rmorons). The photograph \Tas taken after the last reading for this flight given in Table 2. It will be seen that the nmn a.ce formation is confined to between the two T.X.S. strips, R &stance of no inore than 19 while there is subsidiary build uLl from the upper T.K.S. stri;>. The interestin,

/feature....

-7-

feature is the my u :-Jliich the see has bx~lt u;i in ~~11, alrrost unconnected, columns tach pointxz in a dxactxon that clearly shows the trnjzctory of the suyer coded water droplets that froze to form then. The trajectories smst have been almost prsllel to the inn; surface, except close to the sta.gnatmn point, showing that the aster droplets vtere very small. Indirect confoxuation of how s;lall the droljlets were, cm be obtaineil from the shape of the EE that has grmn from the end of the 2" rmasursng rod pr0JeCt.i~ from the leamilg ed.ge. l'hls is veil narrow by co!lparison wit'n the coax-e-- .,i,ondxg formation shown on the o&her phctographs in the saue fi,gure, Ths mrron type of fomation 1s due to tach dropiet fraeeinJ on mpact before the next drcplet is cw&t and withoui hanng tx.l= to flow any apweciablc distance. This onQ haL~pcns with 1~ tePrJerature, xn this case -12°C, and %zth lo:7 ;;-ater contelts x;uch are norz.jLly assocated xxth s;lell drcrglets. The larger 76d.5 forL7ation on tihc other photo- @a&z was caused by rwre and larger droplets caught at a hr::lcr tcngernture so thst his took an appreciable tire to freeze after iipaot a& durxn;, that t%le they flowed snd amalgamated wxth adJacent drops. In addition suhscqucnt LroL3s arrived beforzthe initial Lrop had completely frozen so that a well k&t Ice film vas fomled. This is the classic &f'ferencc between the method 3f formation of rime 2nd glaze Ice. The last four photo,-i,ra;zh s in the f'iFre >Tere taken on flizht 75 at Lntervals of 63, 10 and 5 runutes respectively. The first photo- gra:>h was taken I$- r.xnutes before the start of the results tiven for tins flleht in table 2, The last p!?oto,qaph was iakbn at 16.01 at the end of the period of icing; starting at 75.56. The air ttzz~perature was from? -7% to -5s°C dunqy tie sequence dale the boplet size reached 20 uxrons or over. Th-rc were ciao peak Values of liquid water content of over 1 ~1. per cubic net% but the overa~:cs over periods of ap?moti3atelJ 2, 2 and 5 tinutes respectlveljr were of the order of 0.4 zls. per cubic b:etre. Earller in the fllkht the aircraft had flown zhrou;h an udng cloud at approxhately freezinq level in whxh it had picked up a snLL1 quantLty of Ice, much of ;rrblch had been famed after runtuw; back 2 few inches behind the leading erige. This forme the foundation for the scattered nodules that can be seen In the first photograph an inch or so above the upper T.K.S. Stmp and also below the lower T.K.S. strip. Rivulets of --rater that froze y:hlle rursxL% back to these areas also fol-iled a foundation for SOLE of the ice foma- tlon that can be seen on and above the upper T.K.3. strxs. The ice thus bult up on the earlier degoslt over a wrder area than st would have xf this deposit had not been present. The nodules furiher back on the uptxr and lcxrer surfaces ~re;s on the founclatum of the earlier +osrt to such R $ze that, as CF~ be seen from the last photograph, they started to 1:reak off agzun. In the meantime tne sin&e ~WKJ of ice on the top *f tbe r~e2sux-m:~ rod buit Ui? tili, on the last ?hotograph, it ha& almost reached ztts rxz&xm sue before breakin,c-; o.?f. Even though the thickness o. tnu ~.ce on tne leading{ ed<e cf the tailplane did not exceed &" its rough, wi,de character affected the elevator controls although not qute far enough to cause overbalance, and increased the drag of the aircraft. Another undesirable featie of thu t-aje of ice formation was thst it affected the en+ne coolmg to such an extent that even with the ,&ls fully oyea the olrcrsft had to increase speed ;;y descending to Jrevent the en<clne tempera-cures from reaching lxatations.

k s~.milar effect mxs producecl on fl@d 723 by the ice formation shcwn in f&m3 2. Bere all the 1ce foxmcd very close to freearny: level. Tia distance to wtuch the water ran back before freezlng was considerable. In the first photograph it can be seen freezing as far back as the third black l;nc on the Upper surface. As the lrnes are an inch rylde this means at Least six inches. Later on the Ice eras observeJ to have formed as far hack as 12" f'ror.l the leading edge. It can be seen that during the first 6 photographs Ice is gradually rrets- in;? on the upper surface although not rizuch has fonned orl the T.K.S. strips. On tiie seventh photograph a 2" deep area has praotxcally cleared Itself although the amount of ice on the leading edge has increased. In the last tire photographs the clearance has pro,~ressed considerably. The d.e-icing system vas noi oWrat.ted during this time so clearsxce was caused by flying fxrst just helm then just above frcsziq level and finally descendi~ a little belw it c'unnv the last pm? of the sequence. Xedi~s from a thermocouple near the measurug rod on the ieadin[g e3ge shoTI that for the first three photographs ~ficl the last +xo v?e temperature on the leactinr: &e was +2'C. The:T;;iocoqale reahngs ?fere not t&m at ti,ws correspond&~ to the rcmimng photopraphs but over ths ;,er-ioa the air tem;7erature re&ced fro,0 a3out -1.5% to -4%. b.5 the te;il_)erotiire w.s so close

/to I'reezlnz...

'GO freezing the ice accretion instruments gave unreliable results s0 the true Values of droplet size and liquid water content are u&now. The physical effeCts of the ice formation were similar to those experienced during flight 75.

Figure 3 shows several different points about ice formation. The first two photographs r;hich acre taken on flight 76 at an interval of l+ minutes shot" that the ice formation of an aircraft is not alweys confined to ice formed from supercooled droplets. In the first photographs a number of frozan particles have been caught in the film of ice formed from the supercooled droplets ;.ith uJhich they Caere mixed in the cloud. In the second photograph the same particles CM be identified well covered with a layer of glaze ice and i.ith further pertioles added. This type of formation is not uncommon in cumulus clouds, particularly if they sre wall developed , and it clearly demonstrates hoi. ice accretion instrwents for measuring liquid v,ater content may give inaccurate results. The second paw of photographs she% two view of the ice formation On the leading ddge of the tailplane Obtained On flight 65, taken after the aircraft had landed. Part of the ice had been broken off to measure the thbknzss ad. to make it easier to see the fornation. As in flight 75 later ice formation grew on this foundation to almost the same extent as that on the leading edge. One further point of interest is the thickness of the ice at the leading edge. Various estimates made from the photograph by casual observers fuere all considerably higher than the truth Qich was that the ice ivas only $' thick. This show hoi-J easily inexperienced observers may be in error so that normal pilot's reports of the thickness of ice formation On en aircraft must be treated -Lth reserve. The remaining two pairs of photographs are of the ice fOZmatiOn on flaghts 73 and 75 on parts otha- thtln the taQ&.ne. The first in each set shows the ice formation round the engine co1Jir.g and On the air intake. In each ease the formation w.s enough to upset the flo\l of cooling air 60 that even >"ith the gills fully open the engine temperature ws still rising towrds its limitation. This ws more the case on flight 75 %.here it can be seen from the spinner that there wx considerable runcack as the ice formed. Cn flight 73 the ice formation on the spinner, althOugh not visible from the photograph, covered only the tip snd the area of the first dark ring seen on the flight 75 photograph, shoring that although, there may have been a little run back, there nas not much. lhe smaller photographs show the ice on Rebecca aerials and the rou.hness of the formation on flight 73 shags up t-cl1 against the much smoother mushroom shape on flight 75. The drag caused by the roughness of the ice on flight 73 uas such that even using climbing pow%- it divas only possible to maintain level flight at an indicated speed of 125 knots, vehich is of the order of 50 knots lowr then would be expected for the clean aircraft. It was found neoessary to descend below freezing level to shed the ice. A record from this flight is given in figure 5. It show hou, the speed Zropped off, even though power uas increased, until it had fallen almost 50 knots. Unfortunately the heater on the rotating disc strut did not prevent ice building up until it covered the front of the disc end stopped it from recording after 6 minutes. The Smiths ice detector continued to operate and the times between its pulses sho\$ the continuing severity of the conditions.

5.4. Readings of individual instruments

5.4.1. Rotating cylinders. Readings from the rotating cylinders follow the same general tendency as last year in that all the values of liquid water content over 0.3. grsms per cubic metre come above the theoretical Ludlem limit from the largest cylinder and so must be subJact to some doubt. The amount of error is rather indeterminate but can be quite large at larger values of liquid water content 54th high sub freezing temperatures. For loner values of liquid w&z content end lower temperatures the error should not be excdssivd so the rotating cylinders still give a fair indication of the true liquid water content in this region.

/5.4.2. *. .

-9-

5.4.2. ~Rotatlnl~scs. The rotating discs were both of a produc- tion pattern which had not been= use far long. Lxring the yeear several sr.rd.l faults were found, most of which were flnally&ininated. Apart from the inadequacy of the heater under severe icing conditiorm mentioned above, the t-m main faults arose from the adjustment of the feeler and the scraper. The feeler tensioning spring had to be reduced in size to give more sdjusiment at low tensmn and the friction of the feeler bearings had to be improved. The scrqer was found to be ineffective on some types of ice and the tension of' Its sg~rrng had to be increased. Lastly the feeler and the disc being of the same metal, a groove was worn in the feeler which was not noticed at first but which made calibration difficult. The feelers are being replaced by new feelers of a harder steel. The other failures that were experienced inere of a minor form such as broken electrical connections and wre not attributable to the design of the disc. Now that the teething troubles have been overcome the discs are givingvery useful measurements of liquid nater content, Typical records of both discs are given in figure 6, for a small part of flight 71. This illustrates one of' the early failings in that the feeler on the port disc pushed off the first piece of ice formation and so did not start to measure. until some six seconds after the starboard disc. The starboard disc gives a lower liquid water content than the port of the order of 0.1 grams per oubic metre This hamened fairly generally as can be seen from figure 7 ~&ere the values of liquid water content obtained from each disc are plotted against one snotner. The port disc vms sited much further forward than the starboard disc and it is to be expected that it would give a higher reading due to the effect of the fuselage on the droplet trajectories. This point is clearly illustrated in reference 4 T&here the liquid w.tcr concentrations round an ellrpsoid are compared with the conditions in undisturbe3. air.

5.4.3. Ollcd slide samplmg equinmnt and micro annera. Troubles were experienced mtc& cooling system of the micro camera due to drops of water depositing on to the slrde both before and after sampling. The various apertures through &I& these drops penetrated were gradually traced and eliminated but, although under certain flight conditions this contamination was cured after a few flights , it ~3s not until the end of the period that the difficulty ~uas completely uvercfxne. This accounts for Yne very limited number of successful measurements made by this wthod this year.

5.4.4, l?.A.B. - Srmths icing indicatar. Fqure 8 abows a calibm- tion cf the Smiths indicator results against the amount of ice as measured by the port rotating disc. It will be se& that the points do not agree with the line from last years results. This was not discovered until after the aoints had been plotted vjhen, on investigation, it was found that tine indicator head had been knocked and bent back through an angle of 4". This brought the intxzG heater much closer to the inside of the four pressure holes, partialarly the lonest one, so that the head was partially converted to a baffled type srdlor to the Canadian type described in reference 4, while the effective area of the holes was reduced. The neu~f&mula for liquid water content with the head.,.,, bentbackbecamcm = -- gms per cubic metre as oppose.: to m =

before the head was damaged. This i closer ta the corresnondi~s formula for the Canadian type which was m = 595 i IS the true a;ir si>e;c m knots

Tpqg( . T IS the off time between indicator signals in seconds and Q i.s the air temneratwe in OC). The main difference in these formulae is that mith the damagebixxd the constant that is divided by Q is srmller so that the time between signals ten be shorter before the denominatcr of the fraction becomes so small. that it makes the value of m ridiculously Urge. This allows more reasonable results to be obtained at tenrperatures only a few degrees below freezing.

6. Conclusions

6.1. Over the past 3 years the duration of consistent icing conditions found u layer type clouds over the E!ritish Isles has been too &or-t for even one satisfactory test on an aircraft protection systan on nearly 50 per cant of the layer type icing conditions forecast. The u5ng encounters found to be suitable for several tests represent a still smaller percentage and were almost

/invariably....

- IO -

invariably conf~-~ed tothe Winter months. Although some ice was usually fowd as a result of the other folprasts It 'CELS too scattered to be c12 value for tests on the aircraft systems. The most suitable type of synoptic condition for ice can be forecast with considerable accuracy but it ocours only very seldom, possibly of the order of five times per year. Good testing conditions can be obtained in other synoptic conditions but experience is still too Limited to give any special guidance on them.

Forecasts of ioing conditions in cumulus clouds have beenvery good and, although too short for testing most systems, a good casting of ice can be built up by fly- repeatedly through such clouds.

6.2. Occasions on which icing severity was overestimated during the. past 'mm years are double those in the first year, preswbly due to the smaller number of upper air soundings available.

6.3. The technique of photographing the leading edge of the tailplane has produced several interesting studies of ice formation including examples of "run back" ice . The danger of acing oonditions at temperatures near to the freezing point has been emphasised due to the greater area over which the ice may form thus giving a foundation to s&sequent formations.

6.4. Comparison of the records of the twa rotating discs shows that, allowjr~g for positioning on the aircraft, this instrument gives a cl.ear ard reliable indication of the variation of hquid water cartent throughout tho icing conditions, The existing heater on tho supporting strut is inadequate for prolonged severe iozng.

6.5. The modZ.fied oiled slide sampling equipment has now reached a stage where samples can be photographed almost zmmediately nfter being taken . but its full. possibil~taes have not yet beentested.

7. Further Developments

7.1. Further experience m.& be gained in the synoptic wndltions giving the most continuous dcing. These are mwe cornplacated than most of the titua- tions mentioned in reference 2 and the most suitable cabinations will be watched for during the coming winter.

7.2. The photogaphs of various types of ice formation will be used for comparison with the artificial ice obtained in flight by means of water sprws.

7.3. The oiled slide sampling technique needs further testing Thile the rotating dizc requires an improved heater.

7.4. Improved thermal ice detectors and rotating discs are still needed while an improved version of the Smiths type of indicator should also be inrestigated.

8. References

I. AAB/Res/272. Report of the first year's flying on the G.C. Abel development of flight testing techniques fcr finding an? measuring natial icing wxxbtions. G.P. 221.

2. AJ.?E/Res/278. Report d the second yea& flying on the G.C. Abel development of flight testing techniques for finding sndmeasuringnatwalicing conditions. C.P. 222.

/3 . . . .

- 11 -

3. N. k. c. A. T. N. 3153. Variation of local liquid water R. G. mrsch concentratmn about an ellmsord ii. s. Brun. of fmeness ratio 5 moving h a droplet field.

k. N. A. E. Conada m. -71.

Characteristics of an orifloe type D. Fmser. icing detectcr probe.

NO. I Of ; SynopcIc /

ICIII:: .Cloud Condition Sww1t.y

(I) ! (2) (3) I ii) (51

60 16.6.53 ’ CuriulusjUnstable I II+ I<odwate

1 /,irstrem /

I Swere /

61 /17.6.53 ,C”mlus,Unst~blc Moda’ote

/ SN’lY

; ;.irStrEal

7.w Cu.rlulUS lO,coo -5 !

i I

i .w / 149.15

) j i

/- ,-j

! ’ I I

,

6,500 1 Alto /Light j 12.m -8 1

63 i9.7.Y Cu?“lun ““stcblc / j

1,500 j Alto 4 izht Irith 1 ( h,500- ,-5--g ( GUI ulus jn little 8,000

1.45 I 1

154.10 / 0.35 / I

I - i j

- / 00 1 clouds / I t 7,o

WeP t$ inch build UP on WIL’

Plaw. But clcud not all COntimOUS.

i

Jcrluslon 4 I’

I I

I

67 ‘22.1.54 ‘Stratc CWllS’

! j

)GlUSlOI? Li*t -

j Moderate

Icontmucd.....

4 (33) i (4) (5) 1 (6) J (7) (8) ! I

i (s)

Licka 6,000 I’ Llfm

Light 7,OOO

Llsht

Lfcht 1’ 8,000

7,500

10,oo

4,50

/

-2- ---

(10) ___

-7

., 2 - -1’

-7

f,5- -I1

7.5 - -7

-11

-6

-9

f ! I

I

1

/ /

I

/

I I

1 /

!

I I

U)i (2)

0.27 /

I

60 25.1.54

67 I2.2.59

I

3.05 / 159.25 /

I 3,coo -j mrato- 4,000 ~ClrUlUS

162.25 j 0.15 I

1 ;

70 11.2.54

71, !lG 2 =J& . . &I:.

5,300 4 St-ato- 7,CoO lcrllulus

4,500 1 Strata-

i cumlus 4,5x -j Alto- .3,030 1 cvwulus

/ cunu1us

/ Llint 72 : 16.2.54 1 PD. I

01 0’

73

7G

i Mw!wat;

! I

0

~ ‘0

Light v.ith/ 7,00C 0 little / ixderate

2,Goo ctlxxhs

/

MOdEPate

I

+

j,

1,000

1,5co

2,500

/contmuod.,....

-

65

G8

' 8.7 53

I

j 9.7.53

I

/ I 6.1.54

I

i 25.1.54 I

.2.54

11.45 11.55

11.59 12.00 12.07 12.14 9,~~ i 136 ! -3.5 12.20 7.4co , 130 I -4 12.25 , 9,5m : 138 -4

-10.5 / -8 / -8 1 -7

12.07 4 t,m j 157 12.14 / 7,000 14: 12.15 , 7,350 ; 153 12.21 / 6,CO'l ; 133

14.50 j 7,600 t 161 14.51 j 7,600 ! 180 15.17 / &WC 174 15.23 )

( 6,coL 1 165

IL.24 &f-W 163 :

-6 -8 -1 -7 -7

14.52 j 4,ow / 126 14.55 15.01 15.02

14.47 14.51 14.55

-12 -11 -12.5 -12.5

15.09 I / 4,3x 15.12 15.i4

j 4,"" / ‘,,iOD

15.17 4 6CC 15.22 I L'RW 1:/a? 9 l&f&

~(GrznslCubic i!etre; 1 Med. drop dia.

1. Rot. Disc. i TOE over ufiich a%. t&en (SEC.?) (MiLxmL9)

wt.cy1s.j Port Disc.!Sfbd. Disc. I Q.ot.CylS. IOiled shde.

0.30 ; 0.20 '_ 1

0 25 0.15 0.50 0.40 / - 0.53 0.35 300 :

z

i4j /

1,2

::;Ir: c.25 0.55 I - it2j 320 ,

/Continued.. . ,

0.20 i

1’ ) (2) -is (iI) (5) ' (6) /

-----i-

:hl.2.54 11.27 , li.2Y

11.30

; 7,700 : 170 -8

i ; 11.31 11.46

/ 11.51

?E.2.54 IO.24 lC.25

i 11.07

11.14 ll.r6

( 11.17

7,700 : 163 7,7CC I ;aJ 7,7G@ ' 102 7 ,oco ; 763

I 7.200 159

, 11.21

,'8.2.54 1

15.rxJ 15.01

' 15.02 : 15.04

15.15 , 15.24 / 15.26 I 15.27 ' 15.a

15.31

! 15.32 15.33 / 15.47

I ' 15.51 , i 15.54

bs.2.54 ' 15.oG 15.06

-6 -0 -8 -7 -8

-9 -9 -10.5 -11 -11.5 -11.5 -3.5

-7 -7.5 -775 -7 5 -1 -7.5

‘3

15.13 16.07

t IL37

0.25 II.20 C.10 0 I

1 - i - / -

1:

I - / -

555 247 193

\ 45 I 15 I 523 / 247 I157

I

I* i 1 21

139 1 14 1 loo 1 65 i 66 I

j 43 I h /la, ' 91

ID -

30 39 75 -

i-

ll

31 Gl

P-

: -6.5 -7

/ 0.15 0.45

(0.45 0.35

-

0.55 0.25

/ 0.15 0.35 0.50

j 0.35 0.50

1 11

! 8 , /

/ '4

'17

1 9

‘0 /:

I

0.30

0.05

0.15

0.35

3.05 I( 0.10

0 15 0.10 0.10 0.20 0.15 0.20 a.15

0.45 0.45 3.55 0.55 0.10

G.15 0.10 0.25

0.10 0

0.2c G-30

I / -

I

0.75 j 0.60

1.10 1.15 0.20

1.05 1.05 9.10 3.10 C.x) 0.15 3.15 3.35 3.25 3.20 3.35

I - i 0.35 i 0.35 1 0.10

0.05 0 10 0.10 0.10 0.10 0.05 0.15 5.25

20

00 1

20

31 361

32

G.20 0.20 0.50

-

0.15

0.35 C.P

/ 1: -7

j -7 I -7 -7

j -7

/ 7 j -5.5

1 z.5

3.20 ; 0.15 1.10 i 0.05

3.15 8 0.05 , 0 I

3.25 0.15 1.65 0.25 3.60 I 0.25

i ’ 262

I

i 5,4CO j IG5 C,!XC 15

:5 ,OCCl ' IO / 5,200 14G 6.5

‘continued.....

-3-

(5 ( (IO) .-

I -r t ;I’! j (5) i (6)

i

,,DX 7,x0 7s.m 7.3011 7,7^O 7,m 7,2LO G.&CO

-9

-9 i -f&

/ -9 -10

/ -9

I -65

-7

5,4oc c,4co 6,IKl 6,105

5,m 5,3co 5,500

6,510 7,400 7,600 ",500 8,5CO

d,593

125 I40 145 133 I31 136 140 134

142 142 150 154 144 I39 I46

I73 145 148 138 131 130 140 140 I35

I37 130 I33 131 I43 143 141

1 9.5 -6

j -a ! -7.5 / -8 1 -8.5

-7.5

i am5oo 7,530 7.5m

IO,cm 10,oce 5,930

1 '&OS

>,5cc 5,7':e 1?,2CO

, -1c / I -10 i -12 /

/ -12 -12 , -12

j -10 -10

, -12.5

, -12.5 4 -12.5

-13 / -12 , -12 / -13.5

-13.5

(71 ' (8) i?) i !3) I 4 (1% I 12) (i2) iI3 / (14)

I

.15 .25 2.5 .35 .20 .25 .30

40

i - I 196

102

a 69 27

; 112 I 77 1 64 i i

69 27

, i I23 I 1 I34

:15!

10

253

0 16

13

11

16 1s 11

(17)

16.2.54 i

'1

, I

;4 /

1

I

I

! I

j.

j-

j

1

.3.54

..3.54

31.3.54

0.40

0.20 0.30

0.9

0.20

0.55 0.7" 0.35

1.0

w 0.55 0.50

-

0.25 0.30 0.40

0.40 0.55 0.35 0.55 0.45 0.50 0.7:

1.40 O.ffi 0.65

0.10 ).60 :.20

-

.35

I.95

1

i( j

-

.30 .40 ,.05 ,.35 '.I5

0 0

.

-

,.55

,.50

120 94

3I7 136

I4 21 16 60 79 99

26 27 34 36 I4 29 39

15.22 15.26

'15.3c / 15.37 115.40

15.42

/15.41

/ ;;:;

115.h7

i :,f:;

' 16.27

'14.30 ;14.43 '14.47 15.06

!l5.C7 115.OP

I::.::

/ : 15 II

i 14.57 /15.ofj '%5.10 115.19 1'5.23

0

0

0

0

0 0

0

0

0

0

0 0

0

0

0

-

I.35 0.45 0.25 / - 1.05 IO.50

' I 0 : lb0

0.60 lo.35 0.40 / 0.20

I / 0 1 0

I 24

24

I i

94

/ 49

317 j 136

1

1 1

I -

-

I 35

16

-

I

I I

/ / / I I

I

0.90 iO.L5 c.35 IO.10 0.60 0.30 I.$5 C.‘P

I.50 jO.85 1.55 /O.P

I I -

1 .‘A 0.9 I.40 0.75 1 .I5 0.60

1 .a5 0.85

j 0.55 0.60

I.90 I.39 1 .&O '1.15 ll,wJ i 13v I i i

/Table 3.....

Table 2

Cowarison of severity of icing forecast tith severity of icing observe3

Forecast 1.1 Light Noderate g Severe Nil

Light-moderate

Severe Severe

Totals Totals i 4 i 4 /4 14 111611’ /4 14 111611’

w This flight was ma& prinarily to test instrurunts so the forecast was not very promising. The main cloud fornation lay 3.n e0 azlane and may well have contslnd the severity forecest.

e A very late start was m&e on one of theee flights a-d. it is doubtful that the forecast severity applied..

3 Table

comparison of severity of' icirq forecast mth seventy of icing observed

-', Observed , Overestimated forecasts

\

Correct forecasts ; Dogree,of overestimtion 2

Forecast \ : Severe I

. Light Moderate Nil Light Nil

Light-r,cLierate

xoaerate

Uderate-severe

Severe

Totals

* Thus flight was rmde primarily to test instrumnts so the forecast was not very promsi~. The main cloud fornation lay in m iurlane end may well have contained the severity forecast.

0 A very late &&was mde on one of theee flights and it is doubtful that the forecast severity applied.

T I

/

i

-

Table 4

values of , calculated frm the rotatin:! cylinder measureinents compared 16th observed acing severity

Liquid water content

gm. metre 3)

1.0

0.7 1.2 0.45 0.3

0.05

0.3

0.3

0.05 0.15

0.35

0.05 0.1

0.45

0.4

0.2 0.3 0.9 0.2 0.55

:::5 1.0

-r t i _I

i

Medz.an Lrop1et tianeter ‘tiCrOIlS

17

1: 20

7

17

12

11

a $4

17

19 20

10

20

1: 13 11 16 18 11 q7

Q* (g-s/cm2/hours)

12

T;' 6.5 0.35

0.5

2.0

1.7

:::

4.0

0.8 I.4

2.0

6.0

;:: 7.0 0.8

9"::

:i"

Observed tin&

sevulty

Xoderate and severe

Moderate and severe tight

Light

tight

U.ght

tight

Light

IXght >tith a little moderate Light with sane noderat Moderate

Sevxe

Type of Cloud

:unulus

strata CUmlllUS

Strato CumulUS

Strata CunulUS

strata Cumulus

Strato CIJI!Julus

Strato CunulUS

cui?ulus

GUIillilUS

Cumulus

CumulUS

3E Rg is the rate of ice accretion on a 3 " diameter cylinder flying at 200 m.p.h. at a height of 3 kilometers (nearly 10,000 feet) zn the international standard aixosphere.

FIG.1.

FLIG 69,

FLIGHT 75

ON LEADI G EDGE OF TAILPL

FIG 3.

FLIGHT 76

FLIGHT 65

FLIGHT 73 FLIGHT 75

THE TAILPLANE AND OTHER PARTS F THE AIRCRAFT.

FtG.4.

OUTBOARD

SIDE VIEW

INBOAf2D.

PLAN VIEW

DIAGRAMMATIC LAYOUT OF MODIFIED OILED SLIDE SAMPLING EQUIPMENT AND MICRO CAMERA.

FIG. 5.

RECORD OF HEIGHT, SPEED 8 ICING INDICATIONS DURING SEVERE ICING ON FLIGHT 73.

FIG. 6.

< c ----

------

-a--

2

-’

‘. c=‘,

, 7:

/

COMPARISON OF PORT 8 STARBOARD ROTATING DISC RECORDS.

LIQUID WATEQ

WNlEuT

9mlm3

l-5 +

/’

4

/’ I-0 ,++

+/

0’ “+ A

/ A +

0.5 I 4’

/ +i

d

A,

0,’ 0 O-5 L-O

UQUlD WATER CONTENT

9mlma

/ I 1 l-6

PEAK VALUES

LlQlVO WATER

CONTENT

9mP’

AVERAGE VALUES

+ \ \ \

\ \

\+ \

\ \

\

1 I I I 2 2 e 8

,\ \ \ t \ A \ ’ ‘\ t \ \i +

++\ : \ \+ + \\

i s mo 8 0

CALIBRATION OF SMITHS ICE DETECTOR AGAINST ROTATlNG DISC.

C.P. No. 223 (17,993)

A.R C. Technical Report

( ‘rmw cop?, rrfh/ resrrved

Printed ,iqd pubbsbed by Hen MAJESTY’S STATIONERY OFFICE

I-o be purchased from Eork House. K~ngsway, London w c .z

qz; Oxford Street, London W.I P 0. Box 5%. London s B.I

IJA Castle Street, Edmburgh z 109 St Mary Street, Cardiff ~9 Ktng Strwt, Manchester s

‘Tower Lane, Bmitol I 2 Edmund Street, Bxmmgham 3

Xo Chl‘hrster Street, Belfast or through any bookseller

SO Code No. 23-91309-23

C.P. No. 223