...

,.;

‘,

“,.’

l?ECHITICAL NE?lOR~DUMS

IiATIOITAL ADVISORY COMMITTEE FOR AERONAUTICS.*-

I?L~GIIT-TEST

.——. —. —

NO, 7’08

DATA ON THE

OT VARIOUS’

STATIC I’OR.E-AND.-M?T

GERMAN AIRPLANES

By ?alter iJkbner

STABILITY

... ... -..,1 ,

,.’—..-. —----

TTashing*o,nMay, 1933

-Zeitschrift f;r I’lngtechn’ik und Mot brluftschiffahrt~01. 24, iq(i. 2, Janua~y 28, 1933

Verlag von R. Oldenbourg., Munchen und Berlin

https://ntrs.nasa.gov/search.jsp?R=19930094709 2020-06-17T00:09:49+00:00ZCORE Metadata, citation and similar papers at core.ac.uk

Provided by NASA Technical Reports Server

-.

L

~Illllllllllflwwlllllllllllll‘31176014373741

.____—.

NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS

TECHNICAII MEMORANDU14-.

FLIGHT-TEST DATA ON THE STATIC FORE-AND-AFT STABILITY

NO. 708

OF VARIOUS GERIdAN AIRPLANES

By Walter Hfibner

The static longitudinal stability of ad airplane withlocked elevator is usually determined by analysis; in spe-cific cases, by model tests. The extent of agreement be-tween analysis or model test and full-scale tests is notsufficiently known, since actual flight tests have beenvery meager. The present report purposes to supply theresults of such measurements.

We used the same method as before (reference 1), al-though the accuracy in these tests was enhanced and theinterpretation more complete. This method consists inrecording the dynamic pressure versus elevator displace-ment at different center-of-gravity positions in unaccel-erated flight. In ‘order to establish definitely the oper-ating attitude of the engine the records were made at fulland closed throttle. The interpretation is limited to theflight range between ca N 0.2 and ca x l~O; that is, thezone within which the lift coefficient is approximatelylinearly dependent on the elevator deflection.

The measurements reveal the relationship of the liftcoefficient to the elevator deflection: ca = f (~H) atdifferent c.g~ positione, so that the pitching moment co-efficient versu”d lift coefficient can be determined: cm =

f (Ca). The value ~cmH/~ca is a criterion of the staticstability and is shown in figures 1-7 for the differentairplanes versus the e.g. position.

As anticipated, the stab.i-.lttychanges linearly withthe c,’g. position, that is, in the same ratio in all air-

., planes when the e.g. is expressed iilpercent of the mesachord. The straight lines which re~reseilt the acm~/~ca

versus the e.g. position slope for all airplanes, at full-—————-- ——— — —-

*l13r.gebnisse von ijessungen der statfschen L~ngsstabilit&teiniger I’lugzeuge. “ Z.3’.X., January 28, 1933, PP.47’-52.

L,...

-,-—, --—,.. ... .-—--—-,. ,,..-. —..-.-.. .. . ... .. . ... .. . . .. .-—---

{.-:,

2 N. AqC. A. T6chni”cal Memorandum No. 708

throttle as well as by. closed throttle, at,

By virtue of the singular relationship with the c,g.position, the stability of an airplane can, by a certainoperating attitude of the engine, he numerically given forany trim; that is, for any load attitude, provided thee.g. position is known at w’hich the airplane is neutrallystable :

Figure 8 shows this neutrally stable position of theCeg., of the examined airplanes for no-load enginepower a~~ for a number of ~lodels according to wind-tunneltests vePsus tHFE/tm F.* It is readily seen that the meas-

ured values lie ia a ~one bounded hy the two lines:

r. IH FH and r. IH FHT; = 0.2,2 +-0.33 ~~~- 7; = 0“30 + 0“33 tml? “

For the first approximate.on of the neutrally stalle e.g.position of an airplane, tlie lower limit, that is,

is very expedient.

The measured ~~H/aca ,are illustrated in figures 1-7

versus the c.go position. The static elevator effect

—.———-.. ——---—-—-_..— . ——..— . .. .. ___ ._—--- -_—

‘*3y wing area is meant the total area of the wing project-ed in a horizontal plane, by horizontal position of thepropeller axis. The projection of the fuselage por}ion ly-ing between leading and trailing edge of the center sectionis included in the wing area.

**According to Lapresle (reference 2), the model tests givefor the neutral siability - e.g. position:

r. z~ ~H— = o.225 + 0.37 q~.t*

N~A.’’A.A. Technical ilemorandum No. 708, 3

>—.

was determined for each airplane at every operating atti-tude of the engine. The obtained figures reveal that...

“nH = dcmH”F tm

m ~ F~H

for the different control surfaces. The dcnH/d~H values’for throttled flight together with various ~ind-tunneltests are shown in figtire 9 versus the aspect ratio of thetail surfaces.

The tail surfaces show a marked difference from eachother in shape and position of the elevator axis. The ef-fect of the different division in stabilizer and elevatorwas minimized by converting dcnH/d~H by multiplication with

to a ratio of elevator to control surface area: ~R = o-5.~

According to Toussaint (reference 3), we have~ -1

dcn~- [1.27fi @ - o.215q

0.095 ~H.-——- =d~H ‘H+l .73

for a control group of full contour with continuous eleva-tor. The flight tests with airplane of standard type re-veal, according to the curves in figure 9, figures of from30 to 40 percent lower than stipulated by this formula.The sole exception is the Focke-7hzlf W!ntell, ‘which withdcn~- = 0.050 ajjproaches that of the calculated valued@H ,,

dcn~~ = 0;055.

,.dCnH

“The marked. discrepancy of —dBH

of the other air-

plafies from the theoretical figure–is, in the first place,attributable to the blanketing of the control surfaces bythe propeller at no-load, body effect, effect of cut-outin elevator, effect of open gap between. elevator and sta-bilizer, and effect of form, especially where balanced el-evators are used. These effects do not exist on the !tEnte.~l

4 N. A. C,A. .Technical .Memorandum Ho 708

But. by the usual, arrangement and form of the horizon-tal control surfaces the loss on elevator effect must beincluded because of the above cited causes. This is all-owed for by assuming:

r -1

Owing to the Lack of wind-tunnel data on all but twotypes, the comparison between wind-tunnel data and flightrecords had to be confined to these two. (See tabl~ 11. )

TAdLE II. COMPARISON OF WIND-TUNNEL DATA TO FLIGHT RECORDS.—. — ! ——

‘MeutralX”stable po-

1L

dcnHsition of e.g. : r.Type_

.L

a~—.— —.. ——

Flightp tunnel‘ligh-d=el —

Junkers A 35 I 42.2$ tm!

44.04 t~I

0.030 I 0.0299I I

Xocke-wulf 1“ / I,.

ll~ntellA

[-~6&3$ tm ~ ,-~~.~$ tm ‘. ~.——.—.-.-.-— L.—...~... .— ——

The agreement is better with the llEktellthan with theJurdcers; probably because the slipstream effect in the for-mer is less owing to the locatton of the horizontal tailsurfaces and of the fuselage.

The approximate limits of the range of the e.g. posi-tions due to load chages in practical service are shownin fi~ures 1-7. These limits are very nearly the same asthose set up in the type tests as limits of unobjectionableservice qualities;* right-hand limit of the shown rangewhich gives t“he maximum permissible rearward position ofthe c.gti is of particular si.gliificaace, for, it correspondsto the cig. positiofi at which ad airplane with releasedelevator aud at cruising speed (i.e~, about 60 to 80 per-cent of the full horsepower) , is just sufficiently stableabout the lateral axis. It is seen from the figures thatthe limit of’ stability with elevator released is, in allairplanes. with the exception of the. “lInte,ltonly reachedby rear e.g. position, at which instability already pre-

I?.A. C.A.. Technical Metiorandum No. 708 5

vails with locked elevator and the same running attitude. of ,~~&”.engina: ...., .. .

The range of stability with released elevator in con-sequence extends up to greater e.g. positions than withlocked- elevatoT (reference 4).

Current practice demands that every airplane shouldbe stable with plevator released. For the stability withelevator released determin~s tlie direction of the eleva-tor forces, which is decisive from the point”of view offlight attitude and of landing the airplane in flightbThe question of whether an airplane should also be stablewith’ elevator locked, however, still remains to be an-swered.

The existence of stability with elevator released,even if very slight, is re~dily and accurately determina-ble by the pilot from the elevator forcos. TIIO decisionas to whether an airplane with locked elevator is stableor not, especially by small absolute stability figuresand pronounced elevator effect, demands pilots particu-larly trained for this work, unless instruments are used.The conclusion lies close that the stability with lockedelevator is not only of less significance for the airplanepilot than that with elevator released, but that it is al-together unnecessary.

In measurements such as these the airplane is flownunder varying degrees of stability. The changes in be-havior due to tile dagroe of stability are especially plain-ly visiblo when a stated dyilamic pressure must bo held.

By instability with elevator lo-cliedit is”vory diffi-cult to maintain even an approximately constant dynamicpressure for any l~ngth of .t,~me;even in calm weather itrequires continuous up and down movements o< tile elevat,or.By stability’with elevator locked it is only necessary tohold the elevator at its exact setting;” the airplane thenmaintains the dynamic pressure for this deflect.,ion hy it-self.* Stability with elevator locked therefore facili-tates in maintaining.a certain” flight attitude, but theso

*By high stability ‘with elevator locked tk.edynamic pres-sure recordor can be exchanged for au elevator d.isplaco-ment recordor, bocauso of t-he singular relationship botwoenelevator displacement and dyilamic pressure, for a certainengine load and 3. stated stabilizer setting.

6 NiA.C.A. Technical, ,l$emorandum No. 708

advantages become especially noticeable at landing, sinceit requires only a steady pull on the elevator without upand down movenent. Handling an airplane at landi”rig ismuch more simple by stalility with elevator locked thanby instability, provided, of courso, the elevator effectis sufficient.

For that reason, airplanes should be stable with el-evator released as well as with elevator locked, exceptthose used primarily for acrobatic purposos.

SUMMARY

~cmH1. Stability with elevator locked: ~z.- changes

linearly with the e.g. position in ~very airplano tostcd,that is,

.

2. The neutrally stable position of the e.g. forthrottled flight can be estimated conformable to

,.

ro= ~ 22 + ~ 33 ‘H ‘H ~~ndtm “ . tm J?

dcnH3. ~

(for the conventional tail surface designsand arrangements) according to:

,-

dCnH 0.095 AH fF;(

FR—= 0.6 X -——

)]1.27#~ 1 - 0.215~ .

d@H AH + 1.731

dcnH4* The agreement for r. —— between windand d~~

tum.nel and flight test is satisfactory.

5. Every one of the examined airplanes of standarddesign is still stable with elevator released at e.g. po-sitions at which it is already unstable when the elevatoris locked.

6. From the point of vim of landing in flight, ev-ery airplane, unless primarily intended for acrobatic pur-poses, should be stable with released as well as withlocked elevator,

N. A. C.A. Technical Memorandum No. 708 7

REl?ERENCES. ,,. .. ...=...— ... , ,’

1. V. Heidelberg: ~;leasur.ements of Rudder Moments on anAirplane in Flight.’ T.N. NO. 38, N.A.C.A., 1921.

H#bner, W.: Mes;ung der H~hensteuerkr~fte und derLangsste.lilitat eines Tlugs’euges Voxn Muster Junkers ‘F 13 ge. T.G.L. Jahrbuch 1929, pp. 158-164; andD.V.L. Jahrbuch 1930, pp. 638-644. -,

Blenk, H.: Flight Tests for the Determination of StaticLongitudinal Stability. Toll. 584, IT.A.C.A., 1930.

H~bner, W.: Ergebnisse von Messungen der Stabilit~t urndie Querachse. D,V.L. Jahrbuch 1931, pp. 684-690.

2. Layresle, A.: The Aerodynamic Wind V:~-neand the Inher-ent Stability of Airplanes. T.M~ Uo. 607, N.A.C.A.,1931.

3. Haus, Fr. Ch.: Stahilit~ et Maniabilitd des Avions.Paris, 1930, p. 61.

Toussaint: LIAviation Actuelle, Paris, 1928, p. 140.

4. 131enk, H.: Uber die ~~ngsstabilit~t eines Flugzeugesmit losgelassenem tiohensteuer. Z.F.M., Vol. 21, NO.8, 1.930, pp. 189-196; and D,V.L. Jahrbuch 1930,pp. 61-68.

!llBL3I. AIEWYUMIC GIA2AOTERISTICSAMDT22TIWM Or TIE~TIQATEDAISP~

.,.4

*Upperwing.●*withspecialcontrolsurfaces.

WiqgarmSpm

Meanchord(chordat2b/31ffrm wingcenter)

Positioaof the leading edge of themeanwingchordbadof theleadingadgeof the center sectionof thewingAreaof horizontaltailsurfaceeSpanof horizontaltailsurfacesWidthof f’uselego,at leadingedgeof etabiliserFreespanof IwrisontaltailsurfacesAspect ratiOof horismtaltailsurfacesw # Ii N II II

Areaof stabiliserArw of elevatorBelativ.elevatorareaDistamcaof elwator-is frome.g.of airplme

Deadwoi@tTotalweightMgino t~e

HorsepowerOperation-cog.positionrang.

lkmtrallystabl.positionof e.g.by filllthrottleby closedthrottl,

Staticcontrollabilityat f’illlthrottle

at clos.dthrottla

Rig.ofnomal force coefficientof longitudinaltailsurfaceswithel.vatordisplacmmnth fullthrottleflight..........................

-1.

,-,

\XL Io

Heinkel metSD32 u12i

(82)18.3 I 24.8 I25.1: (m) 7099 10.45 1000

h (m) 1.34 I 1.35I1.36

t

(m) 0.163*-&— (m) 2.3 -h (m) 2.8AH (m) 0.38

2.423.42.s61.11.20.523.761.34

Soo

-t

- --SiwmllflSE12.

H hp 112$tm - 33.1

...39.5

t

0.3W 0.23’3.4 3.33.6 3.20.4 0.323.23.83.01,42.00.594.452.21580850+

2.88.3.072.422.81.30.44.s51.36

mo

t

SiemensSimm.SH12 8H11112 66

- 27.4- 27.4...2&.4...35.!

Unkers‘13ge

44.417.75—..

2.62

0.1587.0 -5.60.4-5.2_4.53.93.66.3.30.4%6.72.481440

mkersL5.300

rov $%s 30.8 28,% 33*35 36.s‘OL %& 34.2 8s.9 37.3 39.0

mlC$)1. v 0.044 0.0755

I0.035 0.050

d%inclosedthrottleflight ....................... — 0.028 0.0335 0.02450.035

.

ankm135

30.115.94

2.02

0.1114.94.210.423.93.652.962.82.060.425.422Jl11001600

mikwL6-200- m!..35.{

33.(42.2

DJJUN

D.013

0.036

0.029!

=1==UbatrosPocks-mm

30.1

1

2s.512.5 14.0—_.— . .-—

1.732 2*5

- 0.18*4.4-4.0(0.42)

-(::~}

(;::)

0.086,25.2

5,24*3540364.B

- 2;6+

1970.58 0.2756.08 4.710= 2.51300 12551635 1650IUU 2s8imms

--L_Va & 14

320200- 16.9z -28.5...23.0...-23.2

-16.344.1 -

-, 0.0155

0.018 -

0.039

0,036 -

IT.A.C.A.Technical MemorandumNo.708 Fig.1

Feet

Meters

.08

.04a, N1 throttle

~ dc)o 0

b, Approximate.mm operating

-.04 range.

-.(36 C,+o+nmzi ~J”= j ‘fiflc~:$’”

18 22 26 30 34 38 42 46Per cent tm

C.s. position r

4

0CcLtdAC)* -4

-818 22 26 30 34 38 42 45

Per cent ~c.~. position r

a. Full throttle

b+ No ~oad

Figure l.-l?wo-viewd.rawiilgsof Raa-o-iktzeilsteinKL.Ic “swallow” and testdata on static stability with elevator locked.

I

3T.A.C.A. Technical Memorandum No.’7O8

G&3.i:~jl~

f—---+.E-,?5&__+__ ‘--~..-.--——

_dd: ,,.

0 4 8 1216 20 Feett4-’l+’*024 6 Meters

Per cent tmc.p. position r

a, Full throttle

b, Approximateoperatingrange.

c, No load

12a, Full throttle

8

4b, IToload

G

-418 22 26 S0 34 38 42 46

Pel-cent ~e.g. position r

Iig.2

Figure 2.-Two-view drawings of HeirikelED 32 and test data on staticstability with elevator locked.

,.

,,

N,A.C.A. Technical Memorandum No.708

,.

“?4

I,.c$j~~---.“-

.,, ,., ,

7‘3‘IiII

(s=3’==4’,,,~>_-,..A------ -

.— .-.?””’”—

7

0 48121620i’,”,,11 I!I I 10246

‘L-a-

4“”5.12.------L. .k.—b—qq. -’ --

K-P“’--t-”” ~..... —-.

.o~ ~- ..J

*

. _.1.-..iJ.Jl..S......1 . _ \

-t/

IIEid .04 - -~-.fiw-22 --------.

o-++ ~Q

“-p:x:$.? -#-

-.c4-” “-i ‘H-”” ~’- “3

Feet

Meters

a, Full throttle

b, Approximateoperatin~range.

c, >~o1oad

’18 22 26 30 34 38 42 45Per cent till

e.g. position r

12

8 a, Full throttle

~ %, No load

c

L‘i8 22 26 30 34 38 42 46

Per cent ~cog. position r

Fig.3

Figure 3.-Two-view drawings of Udet U 12a “Flamingo” and test data onstatic stability with ele-vatorlocked.

— —

—.

“7-) ~-.

-1,.—

~! —..-

/:

/ I

IJlml“,’f~@===r---’-”~’!#.q ‘J------ +3 -.——- ... .

‘+*I\$j

— .>---

0 4 8 1216 0 FeetF+~+~+’- 70. 6 Meters

J:Kqi:;:~#’t*F;~fi$r;i.08 -. -+:&-.i ~~, .. .+t-.~-+-

[. I.&.~- .+ i..?~y~ -q--r--+-() _ _..t _ . _.1 ._T_. -- yj:.+:..~ -

...-— .

t_~””,””-,:~:,: :~:x:j‘- -.

L—L i--- -J18 22 26 20 34 38 42 46

Per cent ~

e.g. position r

8

4

0

-418 22 26 30 34 59 42 46

Per cent ~c.z. position r

a, Full throttle

b, Approximateoperatingrange.

c, No load

a, Full throttle

b, Ho load

l?ig.4

Figure 4.-Two-vie::{drawings of Junkers F 13ge and test data on staticstability with elevator locked.

1

.>, ,.

04 8121620F4* ‘-TW-?’0246

e.g. position r

18 22 26 30 34 3E 42 46Per cent tin

e.g. position r

Feet

Meters

a, ??d~ til~rottle

b, approximateoperatingrange

c, No ],oad

a, Full throttle

~, Xo load

Fig. 5

Figure 5.-Two-vie~-;drawincs of Jurkers A 35 and test data on.staticstability with elevator locked.

N.A.C.A. Technical Memorandum No.708

Fj@jo c)mm

18 22 26 30 34 38 42 46Per cent tm

e.g. position r

12

8

4

0

-418 22 26 30 34 38 42 46

l?ig.6

I’ect

Meters

a, HO load

a, No load

Figure 6.-AlbatrosL 75 “Ass’!,with special l~orizontaltail surfaces,andtest data on static stability with elevator locked.

—

,,, .,,

.

IT.A.C.A.Technical

●

✎

$$”m a .

-.

iiemorandumNo.70E!

i

Feet

Lieters

a,~ull throttle

b,Approxirnateoperatingrange.

Fig.7

. .

Per cent t~fle.g. position r

, ~,,J7-*~~,1l-- i

12-- - ‘. . :1 !+.. .

El~-4---- ,,,,--4-‘J--- -- -~-. .J–} -A!l-- + “--;&-‘:-J”+-:

w.. ~ 4 .--+- -J-*Z + - . “[:----b-

() ““ ~.-,,~,-4...__j...------:_ - ~ \_ _-p ‘<-“tj

,--40 -36 -32 -28 -24 -20-16-12”

Per cent tille.g. position r

Figure 7.-Two-vie~~drawings of l?ockeWulfI’19a “Ente” and test data onstatic stabilitywith elevator locked.

I

N.A.C.A. Technical Memorandum ITo. 708 8

FIGURE 8.- Neutrally stable position of the e.g., ac-cording to flight and model tests versus tH FH/tm ~.

.!.

Mark Type

d Raka 1:1 Ic

& Heinkel HD 32

V Uclet U 12 a

~ Juilkers F 13 gc

* Junkers A 35

9 Albatros L 75

6 Junkers A 35

+ lfVampyrll(glider)

Y. llGreifll

(glider)A Rohrbach

v I?ocke-1’iulfA 1((landplane)

ODFWCV

c1 Moranon

11

II

n

4 RomanoII

nII

b A;711

II

II

II

o 425

● 482

Test

Flight

II

II

It

II

II

model

II

IIII

IIII

IIIIIIIIIIIIIIIIII

II

II

II

II

II

n

II

II

1 l?~Z--/m.—0.35 I

0.45

0:44

0.40

0.43

0.48

0.43

0.265

0.250.47

0.440.27

0,2050.3050.4420e670.000-2050.3050-4420.670.000.2050.3050.4420.67Coo0,00

0.00

:0 fn $tm

..—-34.2

36.9

37.3

39.1

42.2

44.1

44.0

39.0

36.545.0

39.535.0

29.033.037.048.022.531.036s541.548.523.529-532.538e045.025.024.0

28e0

Reference——..— .

y

i

i\~This report

1

DVL Jahrb. 1930,i?ere 155

Sgttinger Ber. III

%~ttinger ~er. IIIGottinger Ber. III

G~ttinger Bor. IIIT.3. III

Bull. Tech. No. 66uIIIIIInIIIIIIII11IIIIII

n

Ygttinger 3er. II

G~ttinger Ber. III

I

—.-———.— ~–— — ..

~

—-——-—----–-y ———------

I..—.—.-—

f—...-—.. ———-—— .—j-_ ..-–

i

i

——.—. .—.—.————-—

.—.-.-——-— .—1- —~+--”---~ .——.—.,__i—-l

fhlX (’J i. ~~

~-l +-/ -a----”~–—- &.___~- ——

--———

1 I

!I~

...—-—

— r.—.%I c .1 .2 .3 .4. .5 .6 .7

Figure 8.-Neutrally

-h FH-tm 1?

stableposition of the e.g. according

—

—.

1~ FHto flight and model tests versus _ .

tm Fwwm

●

a)

—

.

N. A. C.A. Technical Memorandum ITo. 708 ‘9

I’IGURE 9a. - dcn/d~ versus aspect ratio of tail unit.,..

l?IGURE 9b. - dcn/d# fOr FR/FH = 0.5 Ver”sus aspect

ratio of tail g’roupor

dCnHI for ---—— = 0.7

0.095 h~l Ir(

FR FR

‘~H AH + 1.’73 11’27 ~ 1 - 0.215 —’11)]

dcnH 0.095 AHII for -— = 0.6 -“———

d~H AH + 1.73[1.27 &-(1 -0.215 *) 1

.—

Type

-—,Ra.ka 1:1 Ic

HeinkelHD 32

Udet U 12a

JunkersF 13 ge

JunkersA 35

Junker s

A 35

JunkersA 35

AlbatrosL ?5

Focke-WulfllEntell

G~tt. 409

G~tt. 646

--4--flight 2.55

II 3.0

II 2.45

II i3.9

I

II !2.95

II 2.95

Model 2.95

flight13.65

II

model

II

4.35

4*3

3.0

F~F~

——2.52

2.58

0.40

0.47:

0.42

0.42

0.42

0.59

0.27!

0.45

G.25

1._4d-Cn

dcn Tfi ‘orTfi ~~=o ~ Reference

FH ●

—. ——— ——-————0.028 10.0275 “-

0.034 ,0.032

0.0245 0.0265

0.035 0.0355

0.030 ~G.032 iI 1-

This report

DVI, Jahrb. 19300.027 0.029 Ber. 174

1’tDVL Jaiw~l. 1930

0.02990.032 Ber. 155II

0.038 0.0361

J

This report

G.039 0.050

0.037 !0.0385 IG~tt. 3er. III,p. 102

000~8 00(3375 T.B. I, NO. 6

●

.01 !—

I.~,J —-——

____ + ~---- —. fJ: -— —.r–——.— . ——.. . . . . . ___

.02 / ——— —-——

0 1 2 3 4 5 6

.

9b.

4

m:Aspect ratio of tailunit, AH

..C&“.

3

v

.!

,.’”

I ‘-l... .

. .,.. .

,’

,.

,,..’.,

,,,,

>,

,.

...,,