.-z .. .*@2&&i.#,.& ,.

“~eMR No. E5L19

NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS

wim’m’m Iuwoir”ORIGINALLY ISSUED

December 1945 asMemorandum Report E5L19

LABORATORY INVESTIGATION OF ICING IN THE CARBURETOR AND

SUPERCHARGER INLET ELBOW OF AN AIRCRAFT ENGINE

III - HEATED AIR AS A MEANS OF DE -ICING

THE CARBURETOR AND SUPERCHARGER INLET ELBOW

By Richard E. Lyons and Willard D. Coles

Aircraft Engine Research LaboratoryCleveland, Ohio

,., . .

WASHINGTON

NACA WARTIME REPORTS are reprintsofpapersoriginallyissuedtoproviderapiddistributionofadvanceresearchresultstoan authorizedgroup requiringthem forthewar effort.They were pre-viouslyheldunder a securitystatusbutare now unclassified.Some ofthesereportswere nottech-nicallyedited.All have been reproducedwithoutchangeinordertoexpeditegeneraldistribution.

E-172.

https://ntrs.nasa.gov/search.jsp?R=19930093069 2020-05-14T19:00:34+00:00Z

—— — . . . .— . . .. . . ._ ...—

.-

3117601363 m ,—

HACA MR Ho. E51J9

lwcAAm2Rmr l!3mrlERmE4RcHLABaRAmRY

. . . . . .. . . ..- . . MmoRmmMREFQRl!

for the

Alr Technlod Servloe Command, Amy Air l’oroem

LABORATORY INVESl?IGATII)lVOF ICRfG m TBE MRBUH3DR

suPERoEARmR mJm! ELE4)wOR’AN AzRcRAiTEmm

Irt-BEmEo An As AMEAmoFnE-IcmG THE

~R ~ ~E’&UXR m -

By Richard E. Quns and Willard D. Coles

A twin-%arrel InJecl.Ion carburetor and a supn?char~r inletelbow, forminq part of tw III15:vtlon eystem of an airorsd’ten@ne,were used in a lahoratozy Invwd igatIcn to o~tabliah quantItativelyths rdat Ion between wst- and dxy-buli.temperatures of the de-icingair c=d the t- required to ZWOOVOY A gi~en air-flow 10GS due toIcing. Tests were r.inat one ssvere icing condition during simulatedoperation at normal rated power and at two severe Icing conditionsat 60-peroent normal rated -r. The air-flow recovery t-s weredete:whed for dry and humidifled de-icing air at various tempera-tures fzmm 45° to 120° F.

The time required to restore 95 peroent of the maximum possibleair flow after the Inltlal air flow had been reduced by icing toappzwxizmtely 2000 pounds per hour was found to correlate with wet-bulb temperature of the de-icing alr and to de?rease only slightlyas the heated-air wet-bulb tempmature was ticrensed above 800 F.Uhder Impact Icing oondltlons, the meterl-~ chazzcterlet~cs of thecarburetor were seriously affeotedJ ~wsulthg in vary ir.%gularvalues of fuel-alr ratio during the de-icing period. U-r the sameInltId conditions of carburetor-alr tempemture and humidity, theIoe formed at 60-peroent power oonditlons mqulred mom time toremove under the same de -Ioing oonditIons than that formed at normalmted power conditions. . .

,

.—. _ .—

1 m m ,,, .,, - , .,-..,,., ,--, .-,-. ,. -. ,, , ,,, , ,,, ,,, , ,,, ,,mmm, mmmmn,, ,,,1111111111

I

2

nmoIxJcmm

At the request of the Air Technloal Service Comand, S AirForcee, an Inveatlgation of the Ioing oh=acteristloe and eliminationof ice fomation In the *tIon system of a fighter airplane wasconduoted during the fall of 1944 at the NAM Cleveland laboratory.The series of teats reported herein was made on a carburetor mountedon an engine-stage eupezwharger aeeembly and covers that part of theinvestIgatIon ooncernlng the determinantIon of the relation betweenthe wet- and dry-bulb.tenpmdaare of the de-ioing a- and the timerequired for removing Ice from the system.

The results of a previous Investigateion of InductIon-systemde-icing by heated air show that the time required to recover a givenair-flow loss due to icIng is dependent on the wet-bulb temperature ofthe de-icing alr (reference 1).

The Iriductionsystem for this airplane, which has no direct pro-vision for heatIng the carburetor air, Incorporates an alternate airIntalm to prevent free water from entering the system and offersseveral Itiirect methods of oontrol.lingcarburetor-air temperat-are. Ammually controlled valve allows the direct ram air to be shut offand eir from the nmln wheel wells to be drawn through the inductIonsystem. Some tempwature rise always oocurs through th~ turbosuper-charger and by regulatlng tti intercooler flaps the net temperatuzwrise can be controlled. Added heat ~ ulso be obtained by Increasingthe powwr output of the engine.

Tests were performed at shzlated normal rated and cnislngengine power conditIons at carburetor-ati tempemtums of 25° F forboth powers and at 35° F for only the cruise pwer. These test con-d.ition8were selected beoause the determination of Icing character-istics reported In reference 2 shows that at below-freezing alr tem-peratures serious lclng oan occur on the air metering parts on top ofthe carburetor at any power condition; whereas, at inlet-air tempera-tures above fnez ing, the iohg is Imme serious below the throttlesat the low power conditions.

AITARATUS

The Induction system de-icing installation for these tests, whichconsists of a twin-barrel inJeotion-type carburetor mounted on anenghe -stage supercharger assembly, Is shown schematIoally In figurs 1and is described In reference 3. .

nAOABm Bo. E5L19 3

The iolng and de-icti alr duets leadlng to the carburetor areequlppd with tIght shut-off valves oonnectd by a linkage thatprovides for simultaneous opep~ of one valve and oloslng of theothr, givlbg a rapid means of ohanglng fmm Icing to de-iolng oon-ditlons. Alr bleed lines from eaoh duet, whloh are equipped withvahes, permit oonttiuous flm of de-ioing or iofng alr supply whennot flowing through the carburetor. Simult-us Operation of thefoub duet and bleed valves is aooomplished by the use of a singlesolenoid-operated valve to oontrol the alr supply to the two valve-aotuatIng oylinders.

The range of oondltlons under whioh the de-icing tests wereoonduoted is as follows:

Normal InltIal Icing condltions De-lolng-air=ted Air flow Fuel-alr Carburetor- Water temperatures

(~m~nt)(lb/hr) ratio air tem- Injeotion Dry bulb Wet bulb

perature(or)

(gr#M/mln) (%) (%)

Go 4G20 o.Oeo 35 250 50-120 35-120 .60 4620 .WO 25 250 70-120 59-120100 7700 .095 25 250 “4!5-110 45-110

A water InJectIon of 250 gnams per minute at thd carburetorcorresponds to a freewater content within the duct of 8.6 and5.2 grams per cubic meter for 60-peroent normal rated power and mted~r, respectively. These values were used to correlate the de-icingresults with the determination of Icing charaoteristlcsreported inreferenoe 3 and were selected without knowledge of the aotual valuesobtainable at ,thecarburetor in the airplane. Controlled humld+ifica-tion of the air, ioing and de-icing, wae aocomplfshed by inJeot@eteem Into the air duets at a point a sufficient distance from theoar-tor b insure oomplete mixing of the alr and 8team. Thecarburetor-deok pressure was maintained equivalent to an al.tItude ofapproxlmtely 2000 feet.

.

When icing of the Induction sytiem had oaused the air flow todrop to approximately 2000 pods per hour, the valves were shiftedto allow the do-lolng air to flow throu@ the cmburetor. Both airstipplieswere bypassed at the rate of approximately 2000 pxmds perhour when not being drawn through the induotion system to preventM* dw* - tier the shifting of the valves and to *lowoonstant molstum oontent and tempemture to be maintained. An

.._ ____ _.. -

...--.-.., -- .-,.., ... ,. -,-. ..,. ,,,. , ,,, .,,. . ,, . .. . ... . ...... . ,.- . . . ... . . . . . . .m..—. mmm -mm,,, M ,n m

Ioing-atr flow of 2000 pounds per hour was ohosen as the value atwhich *O begin de-iolng because the rate of Inorease of”alr flowfrom that point allowed euffiolent tim for recotiing the de-ioingdata. The de-Icing-air flow and the fuel“flowwere simultaneouslyreoorded every 0.1 minute for 1 minute, every 0.2 minute for 2 min-utes, and every O.5 minute thereafter up to 5 minutes. Eeoause theair flow varied during de-icing, the amount of steam inJeoted intothe air streem wae ad$zsted to rmlntain a oonstant wet-bulb temper-ature. If the alr flow showed no algae of rgGom~ dur~ tharbitrary 5-minute test period, the test was diecontInuecl.

The Inoomlng de-Iofng-air temperature was approximately -60° Fbefore It was heated and humldiffed. When no steam was added tothe de-icing alzz,the relatIve humidity was assumed to be O peroentat the temperature of the tests. No tests were made with free waterinJeoted during the de-icing mns beoause the tests reported inreferenoe 1 indioated that this ~t of free water has no effecton the de-it@ time within the llmlts of observational error.

RESULTS MD DJBCUSSION

The data obtained from the de-iolng tests with the Iabomtorysetup of the hduct ion system are eunmmrized In table I, whioh inoludesthe initlal lc~ oondltions, the de-ic~ oonditlons, and the recoverydata.

Criterion of de-Icim effoctlveness. - Pbr a supercharger oper-ating at oonetant speed with constant inlet pressure, an increase inalr temperature deoreases tlw mass air-flow rate through the Induc-tIon system. Thus, with the de-icing ah, the maximxm lyxsible air-flow rate recovered was less than the Initial ioing air-flow rate.The varlation of air flow”with temperature was experlmntally deter-mined. Figure 2 shows the change in mass air flow tith dry-bulbtemperature, maintaining oonstant carburetor-deck pressure, enginewed, and throttle setting. The tIme talmn to zwcover 95 percentof the maximum possible warn-air flow was taken as the criterion ofeffeetiveness of the de-icing, because above 95 percent the slopeof the reoove~ ourve become quite flat, as shown by figure 3. Inall the tests conduoted on de-lclng, the de-icing was found to be acombinatIon of the- and meohanioal.prooesses. The ice surfacenext to the metal parts was melted and then whole sectIons of ioewen swpt Into the supercharger Impeller by the air stream, usuallyresultlng in a rather irregular recovery of air flow. M all cases,the fuel-ah ~tlo was In the operable range at the time of reoove~but, in several tests where the reoove~ was extrekLy =pid, thelag of the fuel rotemter flost was appreolable, indicating a leanermixture than was aotually being metered by the carburetor.

—...

rmcA Bm Ho. E5L19 5

The -lmum allowable de-iolng time of one engine of a two- “engined alro~ oannot be est~ted hoause it is governed by suohfaotore as the absolute altitude, the pwer rqserve 0? the otherenglnsj ~ wh@her engine stoppage or excmmive roughness omurs.As shown In table 1, when the heat content above @ F of the com-bustion air 1s omr 40 Btu per pound, tti de-Iolng or reoove~ timeis O.6 minute or less.

Results of 60-Pmoelit mted m “r teste. - Tb &sults of testsat tiltial carburetor-airtemperatureEIof S - 2* ~ Indioatethat the wet-bulb tem’pzvatureis the mst si@fioant faotor Indetermining the cte-ioingtims for almilar icing conditions.. Elg-ure 4(a) shows that after lolng at an Inltlel alr temperature of35° F, the de-icing tbe varies with &y-bulb tempemture andrelatIve humidity. When the same data are plotted agalnet wet-bulbtemperature (fig. 4(b)), the ourve obtained corresponds closely tothe mrve for saturated de-iolng air In figure 4 a). The results

&of the teats at an inltlal ah tenipemture of 2S F are presentedIn figure 5. As for the tests at an tiltial air tempemture of35° F (fig. 4(b)), a sti@.e ourve represents the relation of do-icingtime and wet-bulb temperature (fig. S(b)). When the wet-bulb tem-perature of the de-icing air is Inomased above @ F, no appreciablereduction in de-ioing time ocours.

Results of nomal rated power tests. - The de-iolng tim forsimulated normal =ted power tests with an Initial oaz-buretor-airtemperature of 25° F is elso depedmt on the wet-bulb temperaturebecause all the erperlmntal points, regardless of mlat ive humldlty,lie close to a single curve h figure 6(b). The minimum de-icingtime Is again obtained at a wet-bulb temperature of approxlmtel.y600 F. Tk experimentally detezmlned 100-peroent relative humidityourve of de-loing time as a fhnotion of dry-bulb temperature alongwith constant relative-humidity ltnes that were oaloulated from the100-peroent relative humidity ourveare shown In figure 6(a).

Eff13ctof LC* oondltions on de-Iolzu?tIme. - The heat require-ments for de -Iolng during the tests at simulated 60-peroent nomalrated -r, with an Inittal carburetor-air temperature of 35° F, “were less than those at 25° F with the same simulated pawer (fig. 7).These reduoe(lheat requlremnte are due to the lower engine andoarburetor-metal tempezwture at 25° F and the differeme In type ofioe pmduoed at the two idng oondftions. The Ichg conditions at350 F pmduoed a soft, spongy, fuel-evaporationtype of Iolng In theInlet elbow and on the under Surfaoes of the throttle plates. Thistype of ioe oontalned a large.percentage of fuel and was poorlybondsd to the metal. surfaoes. Hazd tightlyadhering impaot Ioe wasfo-d on the upper surfaoes of tb throttle plates during tk

.-. —

6

icing conditions atthe metal surfaces,

25° 1’. The lmpaot

was mre dlffioult

EACA IdRllo.E5L19

ice, being well bonded toto remove than the spo~

fuel-evapoxwhlon I&e.

Effect of en@ne pxrer conditions on de-icing time. - Pbr slm-Ilar iolng conditions, the de-ioing tbe was greater for the 60-peroentrated power runs than-for the nozm&l rated po& runs at wet-bulbtemperature lees than 7@ F (fig. 8). At normal mted power thethrottle was open about 20° mre than at cruise mndlt ions and theimpact Iolng on the throttle plates had relatively little effect, theIce in the Inlet eltiw being mainly responsible for the deozwase Inalr flow. Ebr tests at s-ted cruising power under these same IcingoondltIons, the impaot Ice that fomed on the upper surfaces of thethrottle plates was mainly responsible for reducing the air flow. Theheat input and air velocity for the de-it* process are less at60-peroent norml rated power oondltlons and therefore, the de-icingthe for the sam wet-bulb tempezzttureswas longer for crulslng powerconditions than for norrml mated power oondltIons. The comparison oftests at simulated normal rated power and those at 60-gmxxnt normalrated power thenfore Indloates that the position of the throttle maydetermine which t~ of Ioe most affects the air-flow recovery tim.

Correlation between air-flow recovery and fuel-air ratio. -Typical dmngea in fuel-air mt lo with increased air flow during thede-icing prooess are shown In figure 9(a). At the beginning of thede-ioing period (air-flow rates, ~ to 2500 lb/hr), the curves showquite a wide range of fuel-air ratios but at an air-flow rate ofapproximately 3200 Punds per hour, the cmrves converge to nearly theoriginal fuel-air ratio of 0.080.

The flow-llmit curves for the twin-barrel InJection carburetorsuperimposed upon a band that has within its boundaries 85 peroent ofthe points of the experimental data under Icing conditions in whichthe air-metering parts of the carburetor remain Ioe free are shown Infigure 9(b). The flow-limit curves, as obtained from tho Alr TechnicalServloe Comand, Amy Alr lbroes, showed the fuel-alr ~tlo llmits for amixture-control settlng of automt ic rloh. TIM fuel-air ratio for theInitial air flow used in these tests was 0.080, which is higher thanthat provided by the carburetor at automatlo rich. When the mlfiuzwoontrol is manually set to obtain an Initial fuel-air mt io of 0.080,an effect equivalent to US* larger fuel Jets In the mrbumtorresulted without ohanging the metering characteristics. It was thuspossible to raise the ltilt curves for automatic rich to encompassthe Initial fuel-air ratio and air-flow conditions.

From these two sets of ourves (fig. 9(b)), It oan be seen that,as the alr flow approaches oomplete reoove~, the carburetor meters

.. —--- —.-...-—. ..

- ..

lUOA m No. E5L19

more necu?ly within the nomal-flow llmit range.mixture at the beginning of the de+lolng periodlag In-the tempemture compensator. ThuEJ,whensuddenly t-d on for de-loing, the carburetorfor a short txlnwas though the denser Iolng airand, henoe, gave a rloher mixture. Compensator

7

The overly riohwasduein parttoheated air wascontinued to meterwas Stiu f-~ iS undoubtedly

pzwaent wh&%er the air tempera~um is suddenly increased but theeffeots are most ~tlceable for Iolng at air temperatures above320 1’;because at carburetor-airte~ratures below 320 ~, th Iolngof the alr-metering parts produces ermt 10 nwterlng that outwelghathe lag. ~ compensator is expeoted to more olosely follaw theslowar temperature rise available In the airplane. Some doubtexists as to whether the temperature compensators - affeotedsimilarly In the airplane and In the labomtory setup beoause com-parative temperatuxw surveys were not made. It is believed, how-ever, that this error would be small and nearly oonstant.

The calculated fuel-air ratio durl~ the first part of therecovery was subJeot to error due to inertia of the rotameter floatand to the displaoemmnt of fluid by the float. Consideration ofthese errors 5ndicates that, because the error was large only duringthe first few seconds of the de-iolng period, a correction was notwarranted for the data of figure 9. The flow rate cormoted for thedisplacement ormr is the indicated flow rate plus the fluid displacedby the moving fkat In unit tinw. In run 7 at normal rated power(table I), the possible error In flow reading was approx=tely63.5 percent In the first 0.10 minute that tho fuel resumed flow. Theerror was reduced to approximately 14 percent In the next 0.10 minute.

The values of fuel-air ratio during the first part of thede-Icing porlod were Influenced by the Inaccuracy In obtalnlng oor-reot air-flow readings at the lower air-flow rates when using afixed orifice for the entire range. This effect arlees beoause fora given error In differentIal pressure reading aoross tho orifIoe,the air-flow rate determination is more In error at the lower rates.A stillar tnacouracy is to be expected In the metering charmteristosof

asIn

the carburetor at very low air-flow rates.

The deviations fk%m the allowable flow llmits of the carburetorthe air flow approached complete reoovery are due to variationsthe initial fuel-air ratio settings.

Uhder lclng conditions at carburetor-alrtemperatures below320 F, especially at s-ted nomal rated power conditions, themetering dmracterist ics of the carbtuwtor were seriously affectedand the fuel-air ratios were ve~ erratlo for approximately thefirst 80 percent of the reoovery. Ice that formed on the Impact

.

0 w MR lb. E5L19

tubes, the boost venturi, or th nozzle tended to ohange the meteringcharaoterlstios of the carburetor and random removal of these depositsacoounted for some of the apparent errat10 changes in fuel-air zatios.In sevez=l instances, the fuel nozzle and metering seotione were BOblocked by ioe that the fuel flow completely oeased. Variations infuel-air mxtio from O to 0.600 were observed during the first part ofthe de-ioing period of some runs.

SUMMARYOF RESUU2S

F&m teats of a twin-barrel in~ection carburetor and engine-stagesupercharger assembly, the following resulte were obtained:

1. W-icing time was a function of the nature of the ice formationand of the wet-bulb temperature of the de-ioing air.

2. De-Icing time decreased very rapidly with increase in wet-bulbtempemture to 80° F, above which the time required to de-ice was notmaterially reduoed as the wet-bulb temperature Inoreased.

3. Uhder the same initial conditions of carburetor-air temperatureand humidity, the ice fomtlon produced at simulated 60-peroent normalrated pwer conditions genemlly required more”time to remove under theeam de-Icing condltlone than that formal under simulated normal ratedynmr oonditIons.

4. Lag in the tempendan’e compensator invariably resulted in overlyrioh fuel-air mixtures at the beginning of de-icing periods followingIcing at carburetor-air temperatures above 32° F.

5. Ioe that fo?mwd on the carburetor-air-meteringparts atcarburetor-air temperatures below 320 1’ cawed more erratio fluctua-tions of fuel-air ratio during de-icing and required more time tode-ice at a given wet-bulb temperature than fuel-evaporation ice thatfozmed below the throttles.

Airoraft E@ne Researoh Laboratory,National Advisory-Comittee for Ae?xmautics,

Cleveland, Ohio.

—

mcA Im Ho. E5L19 9

.-. .‘.. 1. IWex,-Henry A., ti (kLbti, Ee?nmn B.: A Iabomtoq Investi6a-

tlon of Io- and Heated-Air De-Ioing of a Chedler-Evana 19CCCPB-3 Carburetor Mounted on a Pratt & Whitney R-183C-C4 lhter-mediate Rear Engine 8eotion. NAM ARR No ● E4JC3, 1944.

2. EB8eX, Henry A., lhlholland, Mnal.d R., and lblth, Wayne C.:Labo=tory Inve@..i@ion of Ioing in the C~buretor and Super-charger Inlet Elbow of an Alroraft lhgine. II - Deteminatlonof the Umitin&Io~ Cond$tlona. EACA ~ Ho. E5L18a, 194S.

3. MIlholland,Ikmal.dR., Rollin, Vem G., - Calvin, Herman B.:Laboratory Inve6ti@ion of Ioing in th Carburetor and Super--r -t Elbow of an Airoi’aft Engine. I - Deeoriptionof setup and Testing w)hnique. w NR r?o, E51J3, 1245.

.

1’

TABLE I

RESULTS OF HEATED-AIR DE-ICING TESTS OF A TWIN-BARREL INJECTION CARBURETOR

Run

AND ENGINE-STAGE SUPERCHARGER ASSEMBLY

CARBURETORMOUNTEDON ALLISONV-171O-89ACCESSORY HOUSING ASSEM8LY

!360 pert?entnormalratedpower- Initialicingco,ndit~ons:air flow at5° F. 4620 lb/hr:fuel-airratio.0.080;relativehumidity.100 Percent:water-injection,250 grams/ml~ - -

:nitiallirflowlb~ )

46754635464546654660462546354640466046454650464046304620462046404650467046554650464046554670464046504610467046454620

De-—-fet-bull;emper-kture,(°F)

3237.54$24347

::5051.555

2596062.56669757583838386

1%100100110120

>ingair)ry-bulbtemper-ature(°F)

5060

;:806060

%100’70

1::

1%806990

1%10083

l%120100100110120

at carbuRelativehumidity(percent

o0

52

:50

l%o

5:50

10:0

1%51515151100925151.5100100100100

torHeatconten8from O F(Btu/lb)

12.014.4516.616.019.120.820.s20.821.624.026.026.026.527●329.031.734.440.040.049.249.249.252●961.275.375.375.397*5126.4

Maximumair flowPecoveret(lb~r )

45404500405545004455452044804575445044404455451544354470422544204380431044304485436044704340430042104270438541954090

. .

95 pposs

kirflow[lb/hr)

43604280

--------42604210427042804310417041104250424040504240400041904210418041604110410041204100406040203980403039503870

NAT 10NALCWMITTEE FOI

?centmaximum)lerecotecovery:ime(rein)

5.75.4

,-------4.03.01.41●2.8

1.3.7.5.75.35.43.4.4.8

::.2●3.12.1.3.2.2

::5.1

ery

Fuel-airratioatrecovery

0.078.078

--------.079.080.079●ml.079.081●083.079.079.080.079.082.076.081.080.078.073.078

----------------

60s1.074.078.080.074.073

~DVISORYAERONAUTICS

Run

303132333435363738394041424344454647

::

TABLE I - Continued

RESULTS OF HEATED-AIR DE-ICING TESTS QF A TWIN-BARREL INJECTION CARBURETORAND ENGINE-STAGE SUPERCHARGER ASSEMBLY - Continued.

L60-percent normal rated power - Initial icing conditions: afr flow at

5° F, 4620 lb/hr; fuel-alr. ra~io~ oeWo; relative hum~dity~ 100 percent;water injection, 250 grams/min]

[nitial~ir flow~lb/hr )

46204630463546404625464046504615463545904630461046354635461046004635460046204625

De-Ket-bulbtemper-ature

(OF)

515967697070707075808082909092

100100100110120

cing air at carbur

Dry-bulb Relativetemper-ature

(°F)

60’70808070

;:709080801009090110120100100110120

Wmidity(percent ]

555452

1To100100100

l%100

1$01005251

100100100100

borieat?ontentFrom 0° F(Btu/lb )

21.426*532.634.435.435*435.435.440.045,545.547.958.658.6610275.375.375.397.5

126●4

Wximumair flowrecovered(lb/hr)

40704405457045504415448546304345447043254640434543005010423541504655425541704390

95 percent maximumpossible recovery

\ir flow~lb/hr )

.------”

4200416041604240425042604230411041504180405041~o4120400039504060403039803920

?ecoverytime(mln)

--------6.31.4.45.4

::1.3

::.3.4.3.2.2,2

::.2.3

Fuel-airratio atrecovery

--------

0 ● 083●079,076.070.080● 077● 081● 080● 080.079,082,076● 068● 083.079● 086● 057● 076● 084

NATIONAL ADVISORYCOMMITTEE FOR AERONAUTICS

Run

TABLE I - Concluded

RESULTS OF HEATED-AIR DE-ICING TESTS OF A TWIN-BARREL INJECTION CARBURETORAND ENGINE-STAGE SUPERCHARGER ASSEMBLY - Concluded.

[Normalrated power - Initial icing conditions: air flow at 25° F~7700 lb/hr; fuel-air ratio, 0.095; relative humidity, 100 percent;water injection, 250 grams/mi~

Initialair flOM(lb/hr )

766076907’700‘77007740776077307725774077007720772577007770775077307720774577207750

DeWet-bultemper-ature

(°F)

40434550556060657070757580808590

1%105110

,cing air at carburetorDrv-bulblRelatlve lHeattemper-ature

(°F)

humidity content(percent) from 00

(Btu/lb)

100100100100100

1::10061.510079.5100

1%100100100100100100

15.617.118.120.823.927.427,4310035.335.340.040.045.545,551.558.666.075.385.697.5

Maximumair flowrecovere(lb/hr)

52807010750075657400795075607440739075007335728071857250724073257400732072557815

95 percent maximumpossible recovery

=T%TE&

-------e

--------

732072657240717071907130699070606970703068407015694568806810678567106685

-------- .--0----

-------- --------

8.9 0;.0902.16 +0911.75 .090

..0881:17 .092

●090:: .089.54 .069

.085:: .090.38 .078●55 ●091

.088::4 .081.25 .079.2 ●074.27 1.078.2 ;.076

co

NATIONAL ADVISORYCOMITTEE FOR AERONAUTICS

1-

I

NACA MR Ho. E5L19

w“ Ill,

k,Air supply pressure tap6-in. automatic pressure-control valve2-in. automatic pressure-control valveAir-supply temperatureHeaters, icing ductHeaters, de-icing duct$team bleed, icing ductSteam bleed, de-icing ductSteam injection, icing duct$team injection, de-icing ductHeater air temperature, icing ductHeater air temperature, de-icing ductAir-teperature control danpers, icing ductAir-temperature control danpers; de-i; ing duct 29Orifice, icing duct aOrifice, de-icing duct 31Air-temperature control element, icing duct 32Air-temperature control element, de-icing duct33l/8-in. steam velveWater separator ;:l-in. steam valve 36Steam bleeds 37Observation window, icing duct 38Observation window, de-icing duct 39Steam injection, ●lternate, icing duct No$team injection, alternate, de-icing duc$

Air bleed, icing ductAir bleed, de-icing ductWater drain, icing ductwater drain, de-icing ductWet-bulb and dry-bulb temperature unitFlexible duct supportIcing air damperWater-injection nozzlesWater coolerDe-icing ●ir damperAlternate water-injection nozzleRemovable ductTransparent removable ductOeck temperatureCarburetorAccessory housing asaemblyManifold temperatureStatic electricity jumperManifold-pressure control damperOilution air supplyDilution air control damperOiluting air pressure tapExhaust stack and air dilution ejectorCareeraDynamometer

NATIONAL AOVISORYCOWMITTEE FOR AERONAUTICS

Figure i. - Induction system de-icing installation.

I

.

IIACA14R HOC E5L191 I 1

I NATIONAL AOVISORYCOkWITTfE FOR AERONAUTICS I I

., —-- ,

4700

4500,

&~ bII

3“o

\A&hd \,.aW \m#’

430C.

0.

‘K

4100.10 30 50 70 90 110

Dry-bulb temperature, ‘FFigure 2. - Effect of carburetor-air temperature on mass air flow.

C!arburetor~deck pressure, 28.86 inches mercury absolute; enginespeed, 2200 rpm; throttle setting, constant.

.

I

500C

4000

1000

0

NATIONAL ADVISORYCOIMITTEE FOR AERONAUTICS

90-

80-0

70-

60.

I 504Wet.-bulbtempf‘ratulme

140Yl I 1, I (°F)

o_ 60w t30~ @ a 50

0 3720.

*U -

0 2 3 4 5 6Recovery time, min

temperature, 35°F-;relative

~igure3.- Typicaltimehistoryof heated-airde-icingforsimulated60-percentnormalrated~wer at600F d~-bulbtemperature.Initialicingconditions:airflow,~620poundsperhour;dry-bulb

humidity,100percent.

.

I

NACA MR #0. E5119

(a)

NATIONAL ADVISORYCOWITTEE FOR AERONAUTICS

. ... .. ,. . .

?

6 s\

Dry bulb.

Figurel+.- Effect of de-icing-airtemperatureon air-flowrecoverytimeatsimulated6&percentnormal rated power. Initial conditions: ~r flow,4620pounds per hour; carbureto+dr temperature, 35° F; fuel-air ratio,0.080. k-icing started after sir flow dro~ped due to icingto approximately2000 pounds per hour.

.

.

.

f!

\

20 , 40 60 80 100 120

.

MR No. E5L19

NATIONAL AOVISORYCOMMITTEE FOR AERONAUTICS’

.

7

6\

xx

5

4 d

3Re:ative humidi ty

(pertlent)

$$)2-: .0049-: 2

2 k

●

1c) o

+w“. -+

- ‘h ‘< ‘c B< J- ~o

Temperature, ‘F(b) Wet bulb.

Figure 4. - Concluded. Effect of de-icing-air temperature on air-flowrecovery time at simulated GO-percent normal rated power. Initialconditions: air flow, 4620 pounds per hour; carburetor-air tempera-ture, 35° l?; fuel-air ratio, 0.080. De-icing started after air flowdropped due to icing to approximately 2000 pounds per hour.

I

NACA

—,

MR No. E5LI$I

-1

NATIONAL ADVISORYCOMMITTEE FOR AERONAUTICS

,. .._. .,.– . . . ,. ,.. -,

I

. .

36I

~a*

i5

s0

d

;a~4~G~

iRela tive h dmidit y

$3(psrcelt)

k :1 309-59

ib

c}1’

0 m

0240 60 80 100 120

Temperature, ‘F(a) Dry bulb.

Figure 5. - Effect of de-icing-air temperature on air-flow recoverytime at simulated 60-percent normal rated power. Initial conditions:air flow, 1+620pounds per hour; carbureto+air temperature, 250 F;fuel-air ratio, 0.080. De-icing started after air flow dropped dueto icing to approximately 2000 pounds per hour.

II.

I

NACA

-..

MR No. E5LIg—1

NATIONAL AOVISORYCOW ITTEE FOR AERCWAUTICS

7 ‘

6 ‘

:

~al> Io15 ‘

33

~

m

E4

~4c@

: Rela tive h uuhdit y

x (percent)13m 01 30Jij +4 B -59

i!2

1

0

Temperature, ‘F(b) Wet bulb.

Figure 5. - Concluded. Effect of de-icing-air temperature on air-flow recovery time at simulated 60-percent normal rzted pow-er.Initial conditions: air flow, 4620 pounds per hour; carburetor-air temperature, 25° F; fuel-air ratio, ().0S0. De-icing startedafter air flow dropped due to icing to approximately 2000 poundsper hour.

No. E5L19I(ACA MR

,- ,.,—.

9’

8

7

54s

i!!2

1

uNATIONAL ADVISORY

CHITTEE FOR AERONAUTICS

,, .-. ,.

lCo. ,. ,,.%

..50

25

10 A

!“

I I

I

I 1II

I \I Relat Ive htitity,

(pertert)9

I !! +

\ \

I\ \ i,

Test data— GiLcu .ated

\. \ ~m “ 1,

\\ ,

f

o

1 f$o 60 80 100 120(a) Dry bulb. Temperature,‘F

P’Sgure 6. - Effectof de-icing-airtemperatureon air-flowrscoveqtimeat simu-latednormalratedpower. Inlttal conditions: air-flow, 7’?00 pounds per hour;carburetor-air -per~ture, 25° ~; fuel-air ratio, 0.095. De-icing started●fter air flow dropped due to icing to approximate~ 2000 pounds per hour.

NACA MR NO. E5L19

9● o‘NATIONAL ADYISORY

COMAITTEE FOR AERONAUTICS—, ..—_.

8

7

6 -

4

~

~2es

soq

2

E4~ percex It) ,.@ii o 100

ii$ w

~3m 65h

i61.5

2

3*2 c

‘o

)1 (.

● *or \~o I-y-o+ ).+

o - “’20 @ 60 eo 100 120

Temper~ture,‘F(b) Net bulb.

Figure 6. - Concluded. Effeat of de-icing-air temperature on air-flow recoverytime at simulated normal rated power. Initial conditions: air-flow, 7709 poundsper hour;carburetor-airtemper~ture,25° F; fuel-air ratio, 0.095. De-ioingstarted after air flow dropped due to icing to approximately 2000 pounds per hour.

II

HACA MR Ho. E5L19

NATIONAL AOVISORYCOMAITTEE FOR AERONAUTICS a

.

7

6 - \

<I

;.

gI ‘

~5 “

%

3 Iy

3

54I

:I

4c1

E Ik

:3 ‘ I

mI

Icin ~; tem~ ratur (!

& .

8 I/o”\

—- -.

.3 1=2

\I

III

1‘ s\

20 40 60 80 100 1.20Wet-bulb temperature, ‘F

Figure7.- Effectof icingconditionson recoverytimeat 60-percentnormalratedpower. Initialconditions:airflow,L620poundsperhour;fuel-airratio,0.080.De-icingstartedafter air flow had droppeddueto icingto approximately2000 poundsper hour.

. . . . . .... ----- .. .. . . . .. . . . . . .. . _.—.. ... .. . ,- ... . . . —-----,- ‘,,.

.

NACA hfR NO. E5L19

.

NATIONAL ADVISORYC@WITTEE FOR AERONAUTICS

-7 L

6 . I

:

sI

2g I

g5 r1

g

:

:I

.

~4 ‘

2I FueT.- 1titial

:c1 -1:3

-— 60-pe?centa

a t*

II

Norma. rate1 .~*5 ‘7700&2 18 I~2

I

I.

1 I,’

,.

020 40

Wet-bulbtemperature,‘F

Figure8. . Effectof powerconditionson”recwery time at icingtemperatureof 25° F. De-icingstartedafterair flowdroppeddue to icingtoapproximately 2000 pounds per hour.

1111

.130 - n

Dry-)ulbt,lmpera;Ure,‘‘FL De-i:inea,?

o 60

J20 100A ~

\.110 L I

NATIONAL ADVISORYC-IT?EE FOR AERONAUTICS

~

~ .100

$~

2●W

\

.080

.Wo ,5m lMO 1500 2000 2500 3W0 3500 @oo

Air flow, lblhr,

(a) Typical runs showing carburetor metering characteriatica during de-icing of fuel-evaporation ice

4500

Figure 9. - Comparison of observed metering characteristics of twin-barrel injection carburetor during de-icing ~period with allowable flow limits. Initial conditions: simulated 60-peroent normal rated power; air flow,462) pounda per hour; carburetor-air temperature, 35° F; fuel-air ratio, 0.080. De-icing started after airflow dropped due to icing to approximately 2000 pounds per hour..L

.(..,.,

(b)

.130NATIONAL AOVISORY

COWITTEE FOR AERONAUTICS

.120

.110

centDf the experhnenta1 data for de-icing- fuel-evapr tion ice li~s wit Mn thi g bar d

.100

.090

.OaoAdjus Led flm lim.ts ofcarburetor :.n autXsaticrich

.070500 1000 1500 25CX) 3500 f@O 459

Air flow,lblhrSummation of test data superimposed upon adjusted flow limits of carburetor.

xo.

Figure 9. - Concluded. Comparison of observed metering characteristics of twin-barrel injection carburetor duringde-icing period with allowable flow limits. Initial conditions: simulated 63-percent normal rated power, airflow, 4620 poundsper hour; carburetor-airtemperature,35° F; fuel-airratio,9.08!3.De-icingstartedafter airflow diF5ppeddue to icing to approximately 2000 pounds per hour.