.

I 9 MAR1948
NATIONAL ADVISORY COMMITTEE
a

‘
FOR AERONAUTICS

TECHNICAL NOTE

No. 1537

APPARENT EFFECT OF INLET TEMPERATURE ON ADIABATIC

EFFICIENCY OF CEN’HUFUGAL COMPRESSORS ‘

By Robert f: Anderson, William IC.*R;&r

and Shirley R. Parsons

Flight Propulsion Research Laboratory

Cleveland, Ohio

.
FM?RMww —.
NOT
TOBE TAKEN FROM Tt-liS ROOM

~g~

Washington
Februazy 1948
NACALIBRA.RY
&ANt3LEY MEMORIAL AERONAUTIUA.L

1 . LABORATORY
V*.
Ia5zleyJ.hel&
“.
 . .—. —

,, . . .. . . .... .- r

.+ a
FORAEROXWJTIH
NATIONAL JIDVH2RY CQMMI!IT!EE
.— —
fiV~CA~ N(j~ NO. 1537 .
...
——— -. I

a~=~ mm @’ IXG!3TTEMPERA= . ON .A))IXBATIX ‘-’”-.

. .,.
E3’FIC=NC; ,OF C$~ ccMFmms
.-
By Robert “J;Anderson} William K. Ritter
and Shirley B. Parsons
. . ..
., . s-y ..”---- .:. .

An investigation was made to determine whether the apparent

variation Of adiabattc”effieien’kywith inlet temperattirefotioen-
tz%’ugal compressors oould be kedkxed or eMminS%e@. “The effects .-
..of“heat.
transfer dnd Reynolds number on adiabatio efficiency were
considered. Three centrifugal compressors were teste~””iman’
insulated variable-oom~onent compreeeor at severa~’imlet tempera-
tures with oonstant discharge pressure and at two discharge Pres-
cures-with”oonstant irilettemperature”. Disoharge-tet@eratUrS
were made”at-the diffuser.dlsdarge and “at-theentrance
measuremerzts”
to the dMsharge pipe$ in’addition tithe measuremen%:at’tli6stan-
dard station in the discharge pipe, to detemnine’whethe~ a tempe%&ture-
measuting stationqould be -foundwhere the”’effectb,:bfheat ttiEn&fer
would be less than those at the standafd station. -’~ :
..
The’standard methods of determini~ oentrii%g&X-Compressor
efficiency by temperature-rise measuremxi% %bnd to give erron~
eously high values when ambient room-air inlek temperd%ures are
used, and the apparent variation of Miabatic dffi.ciencyoan be
eliminated.by basing efficiency on temperature measured at the dif-
fuser discharge. ‘.
..1-
-.

INTRODUCTION

- =vestigations of the performance of centrifugal compressors

at theNACA and other Iaboratortes have shown tkit the adiabatic
efficiency apparently increased with inlet temperature at:constant
equivalent tip speeds and equivalent volume flows, but that the
pressure ratio is unaffected. This apparent chimge in efficiency
has been shown to be related to heat transfer (ref%renoe 1} but no
adequate nor complete explanation has been avaibble to account for
the varlatioh of efficiency with inlet temperature.
.
 ,. -.. !,
.-.+- ,,-.
... . . d“.
. ... .,,... ,-—
2 tiACiEN No. 1S37
,..
*
The lack of agreement of effioienuies-determinedfor the same
compressor at the same equivalent tip speeds but at different inlet
temperatures renders the interpretation c$ corniressor perfonwxx
incomplete. Il!he.
unoertaltitiesthat arise from the “mriation in
temperature-rise-ratioefficiency are: ,,●....
,.t,.
(a) Inconsistencies in efficl&& in compressor investigations
conducted with different:inlet-atb-tetipemt~s but at corresponding
equivalent tip speeds and @th.the da.md!:jyxfdedures.

(b) Inability to determine precisely the perfcn?nmnceof’a.

compressor from performance dat’ai>bf the same compressor obtained at
a different inlet-air tem~rature ..
..,, .. * ... .. - ...-
~ (c) InabdMty to determihe’the~true ‘effYcien& of a &kpres&or
from.temperatuzm’-r.l~se.m easu.twmbnts without a “determinationof’%he
effeot of-inlet+ir~”tim~nture and tlieprobability that ’eff~tiiencies
obtained by’standard m6thods”at-&5a-Ie?el atmospheric ten@eratur6)-
have.been higher than tktrue.va~~
,... .. ... ...$.
.:’ . .::.
. :-r! .,--..H,:..::
..- -.$.
‘.,.
. .. ..
,., ..J. .
.-
...
....,
.
-;\ i:: .—

In.an investigation of,the’effbbt”df blbde burvattire’o~06’fi*~fugal- —

Impel.lerperfczbance’(R#eretioe 2)~.eff$oienuiesbasdd “on’rne&@%ents
made according to ’standardprocedures W6PS fou)idtm’b’einwxmiste’nt
when.different Inlet tempkmatures were used. Thapressiirera+io was
pradtcally unaffected by’vuiatidns ’ininlet tetip&a*~~ WhCYeaS-
the tempemture-rise ratio was affbctdd. The%atic’pam.meiers” 0+’”
equivalent tip speed and equivalent volume flow were the same.at all .
points of compartsonj but the Reynolds numberflvaried
%e@ius&the
ca~essor-dts~rge’ptiss~ was’imirita3n6dtinstant tigardless of
inlet temperature. “Heat-transfereffects might have’been res@xMble
for the’laok”of”agreembnt in temperature-rim i’atiosat var$ous” ““
inlet temperatures, and ths variati.onof”Reynolds number wi$liinlet
temperature might account for a disagreement of pressui% ratioj
temperature-rise ratioJ or both,

c h. investigation was madeat the ’NACA Cleveland laboratory to

determine whether the apparent variation of adiabatio efficiency with
inlet temperature could be reduced op eliminated, “ In order;to red~ce
the effewt of heat transfem on the disoharge-atr temperature:,temper-
ature measurements.were made at the dfffuser’dischargb and at the
entrance to the dischargepipe in.addition to the standard measurement
In the di%oharge pipe”, Adiabatic effiolexxxb%’based on”hischarge ‘
tem~atures taken at the diffueerdischarge and at the standard
statton in the discharge pipe M! diameters downstream of the’oollictor
are presented for three.Impellers at sevexal equlvalbnt tip speeds
and inlet temperatures. The results of varying only Reynolds number
for several equivalent tip speeds are presented.and probable reasons
for the apparent variation of adtabatic efficionoy with inlet temper-
ature are discussed.
NACA TN NO.- 1537 3

APMRA!KJSAND PROCEDURE ,
,,
~:,
Setup

Three centrifugal impellers, designated A, B, anilC, ‘amd a

variable-component compressor set’up in accordance with-the rec-
ommendation of reference 3 were.used for this investigation. A
schenmtic cross eection.of the variable-component compressor is: “
shown in figure 1. A typical installation of one of the impellers,
the rear half of the vaneless diffuser;’the collector, and other
component parts is shown in figure 2. w air discharged from
the vaneless diffuser .fldws into the cbllectir cham$r before
entering two tangential disoharge~pipe6 located 180 apart. The
variable-component cmu@reBMr was moun~e’d” separate”from the ‘“
driving gearbox to elitinate direct heat conduction between the
compressor and the gearbox,; Thecompressor was compldtel.yenclosed
in a“insulation box. “-The~wallsof the insulation box wei’ecom-
posed of 1/4 inch of hard asbestos} 1 incl$of Insulating board,
and 1/2 inch of plywood. The inlet and discharge pipes were
insulated with 3 tnches.of.@.geml.wool. ... ... ..
... ..
-- -“”Instrmie@dt
---- ion —
. .
The comPres&or was instrumented inaccordanoe with the recom-
mendations of.reference 3.w;th stattc-pzws~ure taps,.total-pressure
tubes, and exposed-bulb thermocouples. This instrumentation pro-
vides an inlet.measuring station 2..pipediameters upstream of the
impeller and a dfscharge measuring station in each discharge pipe
12 pipe diameters downstream of ti,eoolleotor. Pressure measure-
ments yere also nmde”at’tl% diffuser dlsbliargetk determine * .
pressure =tiohacross the coiupresstitip”ta-this point. ...
.
In additionto the stgmdqrd thdrmocotipies,”:t% follok;~ “ ~ ‘-
thermocouples”wei’einstalled on
.. the compressor:.-, - c_- “;--..
., -. :-. ..
(1) TWO high-recovery-factor thermocouples were placed “diame%~
??icall#opposite each other and midway b@ween the front and rear
diffuser walls at the diffuser discharge for air-temperature meas-
urements. These thernmcouples were shielded @ were similar to
the Pratt & Whitney probe described in.@~eienoe “4. The diffuser
thermocouple
,, locations’tire“tio~ in figure?2. _-
..- .-
(Z) One expoe6d-bulb thermocoup~e, the &me as those used
at the standard measuring””
station} was placed in the entrance to
. . .. . ..
one of the dl.
s6harge pi~s: ‘“ “ - ~
., . ..,., ,,,, ..i. . ----- : J.- ---

. ,/.
..- . ‘.
. . . . .
,“. .
4 .
NACA,TN No. I-537

1
T@ impeller speed was measured.with an automatically timed
eleuttic revolution counter. The precision of measurements is
estimated to be within the follcwtng liml.ts:
.,
Temperature$ % . .... . . . . *,8.”*.*.” . ’..,....,.. ., =ko.5
Pressuretin. Eg . . . .. . . . ..... . . . . ..i. ,... ..+ OeOZ~
Impeller tip speed, percent . . . . ....,. . ,..J. . . . . ~ . +OP5
Volume flowJ percent . . . .. .“. . . . . . . .... ... ..t., .&3,Q
,. ,,
.- ,4
. - Procedure .. ,.

Impellers A, Bj and C were run at a number of inlet tempera-

tures and,equivalent.(corrected) tip.speeds (reference 5) with the
discharge temperature mea8u~# at the s~wd dzscharge station and
at the diffuser discharge foz?i all ,threeimpellers. Air temperature
,was also measured at the entrance to’the discarge’ pipe for impellers
B and C, I%e folluwing range of equivalent tip speeds and inlet
tempemdmres wee investigated: .
‘? . .. ... . .. ...

Equivalent
tip speed . wet temperature) ‘R
(ft/see) Ihpeller A\ Impeller B ~Impeller C
900 400, 535 463, $30 “452,534
1100 ‘ ‘403, !545 399, 452, 526 431, 531 .:
1300‘‘ “400, 548 402, 529 . 457, 531
... . .
,-

Impeller A was aleo run,at equivalen% tfp epeeds of 879, 1170,

and 1350 feet per second at discharge pressures of 40 apd 20 inches
of nmrcury absolute to obtain an estimate of the pos”sibleeffect of
Reynolds-number variation on adialmtic efficiency. The inlet tem-
perature was nearly the same for both discharge pressures, and the
discharge temperatures were taken at only the standard station in
the dls~ge pipe. , . .
,.
‘.
.,’ ,. CaJculatlons ,“

,. The adiabatic efficiency, based on the pressure ratio across

the compressor up to the diffuser discharge and the temperatures
at the.standard discharge-pipe measuring station, was calculated
for all,runs. Adiabatic efficiency was also,calculated from the
same pm.~sure rati,ofbut for the tempe~tures at the diffuser dip-
charge and at the discharge-pipe entran~,. The difference ‘between
the adiabatic effieienoies for any run is therefore due to the
difference in temperature-rise ratios.
 NACA TN No.”1537 5

.
.. The’Reynolds numb@r’R 18 dafitid”as ‘“j’.’” ‘ ,. ,----
,.
:,...,..,.:” .,.:.
. . ~zu. :: ., —
,. .-. ;
.. “R= ~1”~,: ---- ‘ . ., ‘j..;.
, ...
1. .... ... ...
.. ,.
f:.
?..“.
-..
.
....... ...... . ,. ,. ..,..: -.-—.
.-
where ,, ....
.!..
. . . . . .. . . . . . . .!. . . -.——
. . .

D .:““~@pe&~~~~t~r, f>e~ ~ -“” “,, “- “ “ .’- : “ j’-“T:.:i~ ; . . . : ;F_

.-. . . . .
.,,

.,

Pi, inlet
. s$a@.ation den~ity,,s~~s per’cubio.foot ‘ “““
,,;. . . . .. ,. ..-
u .S’
--imraller’ti$ epeeji, “.. , . .. . -
.. : .< +t” per second ,&
..
.. r.- .— —.
-:.., ...
.
V1 ‘ absolute vi~cQ~i~ tit,. ~lqt” “sta.@@ion.oonditions$ slu@ per
foot-second ‘. .~ . .; ,..
... .-l.. .,. .!Z.,-,
=.... ,:--
,,, ....’..J -<-....,-. .. . -.
oThis Re~old6 -nhber’ ‘d~te.~iqqtionh&-..yu@u@ Onlj- in; compaI?h&’ “
po~nts at ,’theaa@ equivalent” volw” f.io~. ~ eq,utvalmst=ii~“=peed
for different inlet.of ‘.d~sohewge.co~~t@Q8 ~f t@praturo ror’~.“ .-
,--- . ;..
pressure. . ..- “, .
,..
.. .--..
. . .
4-..
--.:-----
.,.

RESULTS ‘ANDDISCUSSION . “’: ““”--”‘

Apparent 3Mfect of.InWt Tm@erdture on Ad~abat~b ~ficlbnc~ ,

,, . . -. i-””
::, ..- ..-..,:..
The apparent variatton of adiabat’id”efflciinby
~witli.
i+i+ %q-
peratu~ was ‘examinedby oomparing effioienoi8& based on temperattim
measunmnts at the standard discharge-pi@. st&tiofi; ‘,,
at the e-ntiadQe
to the dischar~e pipe, and at the diffuser dischatige
.’ ‘-” - --- -

Adlabattc efficiency based ‘on dtsc@a.@@-piTe teraperatiures.- .

The adiabatio efficienoy ~ad i6 presented fOr th?X30OqUi=lent t~p
speeds U/fi( ej ratio of inlet temperature to NM2$ standard sea-
. .
‘.
0 R) of fmpellers”A, B, and C in figurqs 3}
level temperature. 518.’4
4, and 52.respectively. The effictencies based on.the temperatures .
. ,. measured at.the standard.“ineasw”i~stationb in,the dtsc~gp pipe’s
are shown.fob impellers As Bj ahd C in figures 3~a), 4(a)l @ s(a))
respectively.. These figur’esslibw’thata: a“@veq-eqUiy&lent tip.
speed and equivalent voWm6’ fl,ow”~ #19 (where Q~ ~ “is inlet-vplume
flow at stagnatioripz$essure);tti a~iabarticeffici~ncy apparently’
inoreased with inlet temp6~ture’.- The effteI~ncy of @p@ller B ~
(fig. 4(a)) at an .equlvalenttip -sp%d, of’1100 fdet’@r* spooridwas
reduced about 0.02 when the inlet tempe~ture was reduced from 526°
to 452° R> and aboutfi04 for”tin.i
nls%-temperature reduction from 526°
to 399° R- The same order of ,magnitude of eff,iolencyvariation for
similar ;changesin inlet tenqy5datui%a“re’shdwq for .imp611erA-(fig.
3(a)) =nd”impeller C [fig.”:5(d)). we tqmpahatu$~s:~a~~ed.at the
.. .. .. ,..
->... -: ..: ,.
,- -.
 ●

6 NACA TN-No, 1537

1
entranoe of the discharge pipe were the =~ as those at the -standard
station; any heat loss from the disoharge pipes was therefore too
small to make a measurable temperature drop between these thermocouple
locations, and the adiabatic e~ficlency wa~ the same as at the stan-
dard station in the discharge pips.

Adiaba%ic efficiency based on diffuser-disoharge temperature, -

The a~iabatlc effi.clencybased on the temperature measured at the
diffuser disohaxge is presented for inpeller A,,,Bja@.C in figures 3(b),
4(b)J and !5(b),respectively, For each of @e tlu%e impellers, the
efficienq, based on diffuser-discharge temperature, was practically
unaffected by variation of the’inlet-air temperature? Impeller B
(fig. 4{b)) showEIcomplete agreement of efficiency and little variation
in volum flow. For impellers A and C, there are did’ferenoesin
efficiency and volume f).owthat are within the range of preoision
of the measurements; therefore, it is indefinite whether or not small
actual differences do exist. ,The adiabatic efficiency keed on
diffuser-dtsoharge@mp6ratu% remains mabstantial~ constant with
_ in fn~et-air ~emw~~u~ fdrthe t- @@hers u-d intti
variable-component compressor,

Effect of Reynolds Nu&ber on”Adiabatic Efficiency

The adiabatio efficiency’of tmpeller A Is shown ’iafigure””6for

disoharge pressures of 20 and 40 inches of mercury absolute a,tthree
equivaleti tip speeds and nearly constant Mlet’:temperature. At a
given equiva3.entvolume flow, the Reynolds nmdber for the 20-inch
dtscharge.pressure is approximately one-half the value for *6 40-fnch
.,
discharge pressure.

At equivalent tip speeds of 879 and 1178 feet per seoopdj’somp

small.disagreements existed between the adiabattc-efflcioncyctirvesat
the two discharge pressures (figs. 6’(a)and 6(b)”,respectively) but:
almost perfeat ’agreement,waa ob-ined at 1350 feet per second (fig. 6(c)).
The variation of the disohargepre’ssurefrom 20 to 40 inches of meroury
absolute at conetant inlet temperature oaused a ohange in Reynolds num-
ber one and one-hallTtimes that obtained by the greatest variation of
inlet.,
temperature at constaht$ischarge ~essure. Although ohanges in
the shape of the adiabatic-efficiet@y curvbs and inmaximm equivalent
volme flow resulted from a ohange lh”’Reynoldsnumber),these ohanges were
too small ati too inconsistent to aocount for the variation of adiabatic
effiolency with inlet temperature at constant discharge p?+ssure.
.. ..
,.
Discussion of Inlet+emperatwe Effect
:. ,
~ elimination of.the annarent effect of inlet-air temtirature on
adiabatic efficiency.~hen dfs~~ge temperature $s measured =’Ethe diffuser
disoharge and the l.wk of effect of Inlet-tempa%ture variation on the d
 —

“NACATN NO r‘1537 7

press&e relat”fon:&qggest that the tife@” is due to heat transfer

from the compressor to the atmosphere between the diffuser die- ,
charge endthe discharge pipes. The difference between the
dtifuser-dischar~ -temperature”and the t~perature in%he discharge
pipe bore a subqtantiqlly linear relation tb:the temperature dif-
ference be~ween the di$fuser discharge ~d. ambient room air. The
he-atlosses fra the veriable-cornponent .compressorthatWoUld be.-
required to account for the dlf’fer~ces in effiof.enciesbased on
the temperatures at the:diffuser’discharge.andin the discharge
pipe are shown in the following table. The values shown applyvtio

,..
data points frm figure 3 at the tip speed of 1100 feet per sec-
ond; the heat loss would occur through the collector (figs. 1
and 2), which is approxhuately 41 inches in diameter by 14 inches
maximum depth.
-.. ,. —.
,,’, .—
.. .“..
.. . “.- .-”... .7, . “.’ . . .
.’

7
Equiva- JMfiaiency I&et Roam Dtifuser’- T~pera- equired
lent at dif-, temperat-tempera= dzsoh~ge ture @p, heat loss
volume fuser dfs= ure ‘ t~e “ ,tiemp6ra- diffuser (Btu/see)
flow @itige “(’%) ““(~) ‘tureminus tO dlq-
~~’” ,roglntem- Char@:
. .. .

.
1
perature
(%)
PiPE .:
.(%).

6690 0;648 a5, 545 175:.8 . ;.’,:6:6 9,8

. 5817 ;780 178,7 6,3 7.0
4400 ,732”’” 188,3 7.7 6*5
--- ——
“...
6583 .0.641 “
4,m” ,.;!545 -6.0 0.4 “0.7
5563 ‘g
772 ,, ... -11.5 -v8 -1,0
4672 .753 -6,9 -e3 -*3

These Indicated heat=transfer’quantities havecnot been satis-

factorily accounted for by computations using coefficients for
uniform flows.

Although the explanation of the nature of the effect of inlet

temperature on adiabatic efficiency is incomplete, it was i%und
that if the efficiency is based on a temperature measured where
external heat transfer has had negligible opportunity to occur, —
the apparent variation in adiabatic efficiency Is eliminated. It
is also apparent that the efficiencies which vary most from the
correlation values are for the highest inlet temperature and that
these efficiencies are higher than the correlation valuee.
8 NO. 1537 .NACA ~
,.
Centrifugal-compressorefficiency ratings obtained by standard pro-
cedurq~,.yith,
ambient room:.qirtnlpt,.
t~rnperat~rp,thus.,tqndto be . ,
. . ....’.
.-t.. .:....
higher thqn,t,he!trueeffi~i,ency.,,... . . .i. ~,., -—
.. . . . .-
A method’of ‘o.>%ain.in$
conel~tentbf’ficie~$les,for”c~rnparatitq
per.fqqpa.nce,
in ,avariablp-com~nent” ce,ntrifuglil”cohpress@ has :, ,
been obta~ned;’”,however,these qfficienc,iqs’are,notne”~ssarily” ‘.
true effici.enci’es
:,’free”fr@,extei’nal”effdcts.,Obviously, the true.
values of .adiabatiqefficj.ency.&n be found”on~ by .o~tain$nga , —
complete heat balan-ce
. .f’oti
each comp”r:ssor’o<(by,ise of’axiac&zrate”
. . ,. .. .
d~amometer. .,, .,.,. :.
-.
.. ..
. .,,, :’coN@usIoqs:..,’:”:
.,, . . . .
“:,:,,
I?Yoman investigation of the apparent effect of inlet-air tem-
perature on adiabatio efficiency conducted in a variable-component
comp~ssor, the followlng conclusions,vere drawn: ..
,.., s.” , ,.- .- ,,..
-:.”.”..’-.. ,,
“~”:~eap~rext:.;a;t~t;;~of ,adiabatic:ef$iciencywith kle.t”
Ltemperhture of “Gbntrlftigalcon)pres~orsin a v~ia~le-componentibom-
pressor of %h& type used tit~is’inves$iga%ion can be.eliminated by
basing the efficiency”on discharge temperature meas~ed at~the
diffuser disc~rgeb ‘!: I ‘
,. :!
.
2. The standa methods of determining..centrifugal-c~pressor
efficiency by,temperature-rise measurement tend to g$ve erroneously
h@. values wpen ambient . room-air inlet temperature8~cabe ubed.
*,
.. “. , ,..

3. A reliably ac;u~ate determination d the eff$~,tency,,of a.centri-

fugal compre,sqorcan bQ made by temperature measure~nts o@y if a com-
ple$e,heat ba,knce of t~ particul~ compressor.is obtained. ‘“. ~
. *. . ..,.
.-
# ..’. ,,
.... ... ,,”.... . ..
Flight Propul.sioaResearch Laboratory,
Nationq.:~~vJ.soryCommittee for Aeronautics,
Cle.yeland,Ohio; ~ne’271 1947;”, ‘ : . .

., .. . . ...
,, .. ,, ,. .
,, ,,
,. .,.
.:!
 I?ACATN NO. 1537 9
.

REFERENCES
.
1. Caponj R. S., and Brooke, G. V.: The Application of Dimensibnd
Relationships to Air Compressors, with Special Reference to the
Variation of Performance with Inlet Conditions. R.&M. No. 1336,
A. R. C., 1930.

2. Anderson, Robert J.j RitterJ William K., and DildineJ DeanM.:

An Investigation of the Effect of Blade Curvature on Centrifugal-
Impeller Performance. NACATN No. 1313, 1947.
.
3. Ellerbrock, Herman H., Jr., and Goldstein) Arthur W.: Principles
and Methods of Rating end Testing Centrifugal Superchargers.
NACA ARRj Feb. 1942.

4. Eottel, E. C.j and Kalitins~z A.: Temperature Measurements in

Htgh-Velocity Air Streams. Jour. Appl. Mech.j vol 12, no. 1,
March 1945, pp. A25-A32.

5. NACA Subcommittee on Supercharger Compressors: Standard Method

of Graphical Presentation of Centrifugal Compressor Perfomanoe.
NACA ARR No. E5F13a} 1945.

.-.

.:
-.

——-
 .

,,

8
..

“..

:, .:. .
. .

.
..’. --- .- -—
,. . . . . .

,.

.
 NACA TN No. [537 II

//////////////////////

Insulaticm
kad-ai r
i . . ..-. \

VA ‘“- Wikiw ‘d-air”’””u

\
v/
7

Drive-shaft housing

--- -
NATIONAL ADVISORY
COMMITTEE FOR AERONAUTICS

Figure 1. - Cross section of variable-component compressor. ( Scale

3/8 in. = 1.0 in. )

.
. . . . .
,-
. .

Figure 2. - Front view of variable-component compressor.

\

. .
. . .
1.
NACA TN No. 1537 15

NAT IONAL ADV; SORY

C-i TTEE FOR AERONAUTICS

.W

.60

.40

.20

.00

w
c=
3%
“
d \a
f‘
.60 tom

.40
!cj;8
.am. mm m m lwx170Lnlmwaloaala 40&J !!4cae4m70wlKntY
EvJ1valant VOtlN@ flw, Q, ,Mc Cu nlain
,

Figure 3. - Adiabatic efficiency of impeller A.

16 NACA TN No. 1537

NATIONAL ADVISORY

.eo

.60

.40

.20

(al kmd on standard disdmrw-pipe tamwrsture. (b) Eased on dlftusw-dlschwoe ttimratur..

Figure 4. - Adiabatic efficiency of impeller B.

1
NACA TN No. 1537 17

NATIONAL ADVISORY
COMMITTEE FOR AERONAUTICS

.
Equlvalmt VOIU8flo
1s) @aaed on stmdard dlschtrge-pipe tesmratur~. (b) &imcd cm dlffusardlschargc t-pmatura.

Figure 5. - Adiabatic efficiency of impeller C.

18 NACA TN No. 1537

●W)

,70

.50

.30
(a) Equivalent tip speed U/@, 879 feet per second,
.90’
u
m
s
.
3
% ,70
rm ~
.-
0
G
w
v .
u
“; “
~ .50
m
.-
V
4
I I I i

(b) Equivalent tip speed U/w i178 feet per second. ,

,90~ —
NATIONAL ADVISORY
COMMITTEE FOR AERWAUTICS

.70 ‘m
[3== -

I
.5Q I
82? 463

. WI
--
2000 3000 4000 !XX)O 6000 7000 8000
Equivalent voiume flow, Qt ,/~, cu ft/mi n
,

(c) Equivalent tip speed Wm 1350 feet per second.

Figure 6. - Effect. of Reynoids number on adiabatic efficiency of impel-

ier A.