TECHNlcAL NOT ES

NATIONAL ADVISORY COMMI TTEE FOR AERONAUTICS.

No. 210

THE TESTI NG OF AVIA7ION ENGINES

UNDER APPROXIN~TE ALTITUDE CONDITIONS·

By R. N. DuBois, Bureau of Standards.

December, 1924.

NATIOJAL ADVISORY OOMMI TTEE FOR AERONAUTIOS .

TEOHNICAL NOTE NO. 210.

THE TESTING OF AVIATI01 ENGINES

UNDER APPROXIMATE ALT I TUDE CONDITIONS·

By R. N. DuBois.

The de.si rabi 1i ty of a test to det ermine the perforrr.ance of

aviation engines a t a ltitudes was early recognizedn With informa­

ti on obtained f rom such a test, the performaLce in flight could be

predict ed, possibl e engi ne failures at high al ti tudes foreseen,

and the danger to pilots t he refrom greatly miniMized . This led to

the const~lction , at the Bureau of Standards, of a laboratory in

qhich the conditions encounter ed at altitudes up to 30, 000 feet

carl be simulated . A c..escription of this laboratory, which has

been in successful operation fo r over six years, is contained in

Report No ~ 44 of the Jational Advi sory Committee for Aeronautics.

The co st of a laboratory of t ht s type has led to reany sug­

gestions for a modifi ed method of testing which ~ould not require

the eqll ~,pment necessary for a II true ll al ti tude te st , but which

mig'l!'u give approximately correct resul ts at less expense . In mak­

ing an al ti tude test in the laboratory at the Bureau of Standa~d.s ,

the pressures at the carbureter ent rance, at the exhaust ports and

i n the chamber sur~oundi ng the engine are reduced to a value corre­

sponding to the desired alt itude . Among the suggested approximat e

methods of test is one in wh j.ch the p r essures at the carbureter

N.A.C·A. Teohni cal Note No . 210 2

entrance and exhaust ports are reduoed as above described, but the

air surrounding the engine al lowed to remain at sea-level pressure.

;~sts oan be made in this way without the use of the strongly re­

inforoed ai r t i ght chamber e~8entiEtl to the "true H a l ti tude test.

This paper is based upon tests made in the altitude laboratory

of the Bureau of Standards to determine the value of thi s approxi­

mate me thod of t est.

The engine used in the tests -as a Curtiss, Model D-12 (1145),

having 12 cylinders of 4 . 5 i n . bore by 6 in. stroke, comp. ratiO 5.3.

A oomparison was made between runs under the approximate alti­

tude conditions, i.e ., with ai r surrounding t he engine at sea-level

pressure, and s i milar runs made under IItrue ll altitude conditions in

which the surround i ~g air is reduced to a pressure corresponding to

the a ltitude. Runs under both oonditions were made at air pres­

sures corresponding to sea level and a ltitudes of 5 , 000, 10,000,

and 15,000 feet . At eaoh a ltitude , the group of runs under approx­

imate conditions wa s i mmediately followed by a similar group under

t he 11 true" condi tions.

Any i mportant ohange in engine performanoe brought about by

the ohange in p res sure of the air surrounding the engine should

manifest it self by a ohange either in speoifio fuel consumption or

in brake horsepower developed . Figs. 1 and 3 are t herefore of pri­

mary interest.

From Fig . 3, it is seen that the relation betw'een specifio

f uel consumntion (lb . fuel per indioa ted horsepower-hour) and fuel-

N.A.C·A. Technical Not e No . 210 3

air ratio was the same for both types of test.

Fig. 1 shows that the maxi mum brake horsepower developed at

f'c;.ch al ti tude was the same under both condi tions. The value s gj.ven

in this CLlI've are taken from the curves of fuel-air ratio versus

brake horsepower shown in Fi g . 2.

Since the power developed in t he two types of tests was the

same, no difference in actual volume tr ic efficiency was to be ex­

pected. Past experience , however, has shown that the observed vol­

umetric efficiency is often less than the true value, due to unno­

ticed leaks in intake mani fo ld s and the air line to the carbureters.

In both methods of alt itude test, the reduced pressure at the car­

bureter entrance is obtained by throttling the air entering the car­

bureter. If leaks had existed in the carbureter air line, their

presence WOUld, of course j have caused a much greater error in the

air measurements when the surrounding air was at sea level pressure

than when it was reduced to approximately the same pressure as at

the carbureter entrance. Since no consistent difference was found

bet~een the volumetric efficiencies in the two types of test (Fig.

4), it is safe to assume that in these tests appreciable leaks were

not encountered.

Since the pressures at carbureter entrance a~d exhaust po rts

are the same under eit~1el' approximate or ntrLle" altitude conditions,

there should be no difference in the conditions in the combustion

chamber.. HOi7ever, the increased pressure in the crankcase during

an approximate test may have some slight influence on the amount of

·,A.C· A. Technical Note No . 210 4

oil passing by the piston. Another effect of increased crankcase

pressure is to increase the work of moving the piston on the down­

w~rd stroke. This, however, is compensated for by the decreased

~ork of the upward st roke and therefore does not affect the power

output.

The maintenance of reduced pressure at the exhaust ports dur­

ing an altitude test necessitates the use of exhauster pumps of

large capacity. The po ssibility of emergency tests without the use

of such equipment , or ith equ i pment of less capacity, made it ad­

visable to invest i gate the effect of changes in exhaust pressure on

the performance of the engine duri ng an approximate altitude test.

Runs were accord ingly made under approximate conditions, at carbu~

reter entrance and exhaust p ressures corresponding to sea level and

altitudes of 5, 000, 10J OOO and 15,000 feet. At carbureter en­

trance pressures cor~esponding to each altitude, runs were also

made iith exhaust pressures both greater and less than the carburet­

er entrance p ressu res, one group at each altitude being made with

sea-level exhaust pressure . The results of these runs are shovm

in Fig. 5.

Exhaust back-. ressure causes a reduction in the power devel­

oped 7hich may be considered as divided into two parts. The first

is the loss occasioned by the decrease in volumetric efficiency due

to the greater p ressure of the gases in the clearance space at the

beginning of the intake stroke. A second loss is caused by the in­

creased work done by the piston on the exhaust stroke in expelling

N.A.C·A . Technical Note Noo 210 5

the spent gases against a hi gher exhaust pressure.

The following computation may serve to illustrate this change

in power with change in exhaust p ressure:

Let P = absolute pressure at exhaust port. 1

"D .L2 = absolute pressure at carbureter entrance.

P3 = absolute pressure at end of suction stroke .

VI = clear ance vo lume .

V2 = volume occupied by clearance gases at end of suct ion stroke.

D = p iston displac ement.

n = exponent of adiabatic expansion, taken as 1.3.

To sirJpli fy the ca lcula tions, a number of assumptions are made:

First, that with or without exhaust back-pressure, the pressure in

the cylinder reaches the same value P3

at the end of the suction

stroke. (The values of P3 \7ill be slightly different due to

change in condit ions of flow into the cylinder with change in pres­

sure of the gase s in the clearance volume.) Second, that when

there is no back p re ssure, P3

equals the carbureter entrance pres­

sur8 P2 and therefore for that condi tion V2 equals VJ.' Third,

that the pressure in the cylinder a t the end of the exhaust stroke

is equal to Pl

, t he pressure at the exhaust port . Fourth, that

the indicated mean effective pressure is directly proportional to

the weight of char ge d r awn into the cylinder and hence to the volu­

metr ic efficiency . This is borne out by the results of many tests,

both in the alti tude laboratory and elsewhere.

N. A.C· A· Technical Note No . 210 6

For an example , t ake P1

= 14.7 lb. per sq.in.) v = 10 cu. in. , 1 .

D = 40 cU.in ., I. L" . E . P . deve10ped \"Jith no back-pressure = 100 lb .

D8r sq . in. It is desired to find the decrease in I._~.E.P. to be

expected v-nth an increase in exhaust pressure of 4 lb. per sq. in.

In this case the g ses in the clearance volume will have at the end .

of the exha'.J.st st ro ke a pre ssure P1 (14 .7 + 4 = 18.7 lb. per sq . in .),

and when expanded during the suction stroke to the pressure

P3

= P2

= 1 4 .7 lb. per sq .in., will occupy a volume V2 v7hich may

be determined by the use of the expression

then In ~ l~ 18~7

V2 = V1 P2

=:: 10 14 . 7 = 12.03 cu.in.

With no back- pre s sure , the clearance sases occupied the same

volume at the end of the suction stroke as at its be ginning ,

(V2 = V) = 10 cu,in.), whereas with 4 lb. per sq .i n. back-pressure

V2 = 12.03 cu .in. = V1 + 3 . 03 cu . in .

The space availabl e for the fresh charGe at the end of the

suction stroke is therefore 2 . 03 cu . in. less than under the first

cO~l.ii tiona This causes a reduction in volumetric efficiency of

2" 03 -, ~ 40 or 5'.;.../0' The reduction in I . i . E.P . due to the decreased vol-

un:etric efficiency is then 100 x .051 = 5.1 lb. per sq.in. To

this must be added the loss due to the work of expelling the gases

a painst a hiGher exhaust ~recsure. This is equal to the difference

P P = 4 I b 1 - :2 pe r sq . in. The total decrease in I.M.E.P. due to

N.A · C~A: Technical Note No . 210

an increase in exhaust pressur e of 4 lb. per sq. in. is then

5.1 + 4 = 9.1 lb . per sq . in.

7

The dotted lines of Fig . 5 are values calculated by this method ,

based upon the indi cated mean effective pressure developed at each

altitude with exhaust pressur e equal to carbureter entrance pres-

sure .

There are many assumpti ons ~ade in these calculations, but the

values obtained are near enough to the actual test values to make

the method of cal cu lation useful where it is desired to predict,

approximately, a l t i tude ge r formance without using the equipment

necessary to reduce the exhaust port pressure to the true value, or

where unintended fluctuations of this pre ssure have occurred.

The difficu l ties of the approximate type of altitude test are

chiefly those due to the ciifference bet~een the pressure of the air

surrounding the engine and that existing in the air horn, carbureter,

and exhaust pipes. Great care must be taken to prevent leaks into

the air horn and carbureter , as these will render the air measure­

ments inaccurate . Air bleeds to the carbureter must be enclosed

and connected to the air horn . Leaks in the exhaust p i pes will, of

course, greatly increase the diff i culty of keeping the exhaust port

·pressure at the .:)Toper value . It is believed that with the above­

mentioned precautions, this type of test may be found useful in

obtaining approxi mate a ltitude erformance data where facilities

are insufficient to r ep r oduce the true altitude conditions-

N.A.O.A. Te chnic a l Not e No . 2l0

x = Approximate al ti tude t est

o = True a ltitude test

Fig.l.

600 ~--~----~--~----~--~1 ----~I ----~--_r--_.

I 500 ~---1---~----~----+----+----+----+----r--~

! i I 1 I 1---1 I i ~ 400 1-----+-1,-~':--- · ~ I

j I ~~ I ~ 300 ~--~I----~--~----~-~"~~--_--~I----rl ----r---~

~ 1--_+ I: i ~, k~---+----l 200 I--------t-! --+--i I ~~,

I I I ~ i I I ~ -- -t-------+--. --H----+----+------l

I I I I

100 -----t- I +-t--L --,._--

1----~-+ I j---;.-- -+----+---+------f

I ! I O~--~----~--~----~--~----~--~----~--~

. 08 . 07 . 06 .05 . 04 Air densit y in lb . per Qu.ft.

Fig.l .

N.A.C.A. Technical Note No.210 Fig.2.

600

50 0

I( f-i 40 (])

~ o A (]) {f)

f-i o

.£l

{t-----(J) 30 ~ cd f-i

CO

20 i(

Ie iu

c .04

Fig.2 .

I

I

x :::: Approximate altitude test

o :::: True altitude test

I r)(-x t:>m. :::: ~3.7 c /~l ' Bar rn Hg

I !

V x~ p<o~ f3arom .j :::: 63 . f5 cm H~

IA-~ ~ f3arom. :::: 52. 7 cm H~

~-

_u ~) r--.. Barom . :::: 43

~.)(,.) .1 cm Hg I'

.06 . 08 .10 .12 Lb. of fuel per lb. air

N.A.C.A. ~eehnieal Note No.2l0 Fig.3.

Baromete r em Hg 73.9 63 .3 52.5 43 .1

Approximate altitude test x 0 + 0

True al t i tude test x 0 e ()

----

.9

.8

· f-l ..c: 1 .7 ·

H .6

f-l Q) p..

rl

~.5 ct-t

• ..0 ....:1.4

.3 .04

Fig.3.

I

L () V ./

/--£

J

() +~ ve-t-

~ -tff

-~~f I

I .06 .08 .. 10 .12

Lb. of fuel per lb. air

N.A.C.A. Technical Note No.ZID

nO p::

R 0

t--. LD to

II

• @ i4 I--cd

CO

--

c

x

0 X

x x

f---- )

o l U o ()) ()) rl

x = Approximat e altitude test

o = True alti tude test

~

I E 0 I rl . I K> x ~ b

" I . s · P P )( cd ~

X

~ f------ J:ill 0 P::)

El '.:1

Ox 0

1 t-- L[) ')( . .

N to 0 LD cDx

\I \I

x . . ~ 0 5 H H

~ x P5 _r

~

0

:x >x

c

(

(r--- -- r--- ---

I LD LD

Fig.4.

)<

X

b( p::

8

° t--. to C'-

II . ~ H

~

I

L[) o 0 o ()) ()) rl

Fig . 4 .

o N C\l

o o N

- .0 H

o cD r-I

o N rl

N.A.C.A. Technical Note No . ZlO Fig.5.

170

160

. .~150 . a' co

~140 P-t

.0 01130 ~ ·n <i)

~120 co co <i) f..j

P-tllO <i)

l> ·n +.:> 0100 <i)

'H 'H Q)

§ 90 <i)

S

'd (I) 80 .:>

cC o ·n 'd ~ "'0

H

60

50

x = Calculated values

o = Actual t est results

I

R. I) '.l'1 , lS~C 2000 C~rbure ~er ai r '~er":~fTa~ur~ =

J

~ r -+- I

~ . I

j Baron • = n1. 9 cr: JIg I I I I

! I I I -

~ :> r-~_

-~

~ I cm f. Barom. = 63 . 3 g

~ ~ r-----r----- Ba]:'om . = 52 ri5 cm 19

~~ r- I ~~ ~ ......... ~-1---_ i-----: --~r----Bare m. = ~ 3 . 1 cz. Hg x

'r--.." ---I ~

~ \

I o 2 4 6

Exhaust back pres sur e in lb. per sq. in.

Fig.5.