—

?

,..’.’---

NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS

ORIGINALLY ISSUED

February 1945 asAdvance Restricted Report ESBO1

MEASUREMENT OF OPERATING STRESSES IN AN AIRCRAFT

ENGINE CRANKSHAFT UNDER POWER

By Douglas P. Walstrom

Aircraft Engine Research LaboratoryCleveland, Ohio

NACA WARTIME REPORrS are reprints of papers originally issued tu provide rapid distribution ofadvance research results to an authorized group requiring them for the war effort. They were pre-viously held under a security status but are now unclassified, Some of these reports were not tech-nically cdite(i. All have been reproduced withcmt change in order to expedite general distribution.

L - ..--ih.-.::u...-::u

-...

NACA AKR NO. 3!5601

NATIONAL ADVISORY C~ FOR AERONAUTICS

ADVANCE RESI?RICTEDREPORT

~ OF OPERM!IEG SJIWSES IN AllAIRcRNm

ENG- CFWKHWT UNDER POWER

By l)OU@aS P. Wtistrom

SUMMARY

Strain-gage measuramnts were made on the crankshaft of aWright R-1820-73 test engine while under power. Bakelite-bondedwire strain gages were Installed on the front half of the crms.k-shaft and the connecting wires were brought out through the oilpassages and connected to slip rings at the rear of the engine.

Bending and tensile-ccmpreasive strains at various sectior.ak the web were measured by pairs of strair.gages and strains atseveral points cc th~ front half of the crankshaft were measuredby sing19 strain gagoe. Peak stresses were determined from oscll-lqy?aph recordg end the harmorilccomponents of strass were deter-mked hy a wave analyzer.

Strains were measured with the eriginebeing driven by thedynamometer at a range ~f speeds and elso with the emgine operatingunder p~wer at a range of speeds and powers. A harmonic analyelsof the gas pressures In the cylinder was made at constant speedand a range of powers.

Results shwed that the inertia-excited stresses varied almostdlr9ctly with speed and that the stresses cau8ed by gaa-pressureforces varied with the Indicated mean effective pressure, When theengine operated under severe knock, observations showed no change Incr~shaft etresses. No direct relation was found between stressesin the crankshaft and the torsional-vibrationamplitudes as deter-mined by a torslograph Installed at the rear of the crankshaft.StralL-gage measurements on an aircraft-qine crankshaft in flightor In an engine-propeller test stand are believed to be practicable.

IULOAARR NO. E5801

INTRODUCTION

The determination ~f operating stresses in an airoraft-enginecrankshaft under power would be of value in orenkshaft developmentbecause lt would supplement theoretloal calculations and laboratorytests. Ba.kelit9-bondedstrain gages applied to the acoesslble partsof the operating crankshaft, with connecting wires brought out throughthe oil passages, make possible the followlng:

1. Measurement of actual peak stresses

2. Separation of stresses into bending, torsional, end tensilestresses

3. Analysis of the forces applied to the crankshaft

Curtis (reference 1) applied strain gages ta the crankshaft of astationary Dies.>1engine. As the result of strain measurements Inthe orarikweb,he was able to explain a series of breakages and toshsw that Installation of torsional-vibration dampers reduced thestrama t~ a safe value.

The tests reported herein give the results of strain-gage meas-urements made at the NACA Cleveland laboratory during the summer of1944 on the crankshaft of a Wright R-1820-73 (CWO) test engine overa racge of 9peratl~q cmditlms. The technique of strain-gage meas-urement and the acceasmy r~pparatusused is described. Data aregiven on bending snd tmslle-compressive stresses at various sectionsof the web and m stresses at several p~ints on the front half of theorankeheft. Alternating stresses only were meaaured.

The two types of Gxciting force, inertia and gas pressure, wereseparattiiyanalyzed. Stressaa caused by inertia forces were deter-mined by driving the engine nth the dynamometer and the gas-pressureforces were qualitatlv91y determined wltn a pressure indicator. Theengine was cperatcd under power (the mndfticm where both inertia andgm-prmaum forces are present) at a ruqge of speeds and manifoldpressures end the stresess were measurud. In order to determinewhether kncck prmiuced any stresses in the orenkshaft, the enginewas op~rated Undbr severe lumck. Torsional-vibrationmmsurementswere made to supplemmt th9 istressm~asurements.

API?KRATIJS

Tests were conducted on a Wright R-1820-73 engina convertedfor test purposes. Cylinder 1 was operated under power; cylin-ders 4 and 7 were operated without firing for balance with pistons,

NACA ARR NO. E58(ll 3

ccmneoting rods, and valves in place but without push rods. Theother oyllnders were removed. The propeller reduction gears werereplaced with a “directoonzmctlon and the pendulum dampers werereplaoed tith rigidly attached oounterwelghts. The superchargerimpeller and gearing were removed. Auzlliary equipment used forccmtrolllng and measuring engine conditions is shown in figure 1.

Thirteen strain gages were Installed on the front half of theorankehaft and mnneoting tires were brought out of the engine asshown in figure 2. The tires, No. 32 enameled double ootton oovered,were attached to the surfaoe of the crankshaft with Be&elite oement,BC-6035, up to where they passed through an oiltlgbt neoprene paokingInto the drilled oil passage. The wires were twisted together andcovered with a flexlble braided-glass sleevlng. Aholhw extensionshaft was attaohed to the starter-shaft coupling. The shaft ocm-tained a second neoprene oil seal that allowed the wires to passthrough to where they were oonnected to the slip rings. At the dis-assembly of the engine after 300 hours’ operation, this installationof connecting wires and oil seals was found to be in good condition,

The slip-ring devioe (ordinarily called a “pineapple”),whichwas attached to the extension shaft, consieted of a set of 12 monelslip rings mounted on a spindle rotating with the shaft and a sta-

. tionary outer case supported by bearings to allow relative rotation.This outer case supported the silver-graphitebrushes, 1/8 inch Indiameter, which were held against the slip rings by small helicalsprings. Two brushes, 180° apsrt, were used for each ring. Atimang devloe oontained in the pineapple generated a sharp electri-cal impulse at each revolution of the shaft. A view of the pineappleIlwtalled on the engine is shown in figure 3.

Individually shielded wires conneoted the pineapple to a ter-minal box containing load resistors, switches, and terminals. Con-nections could be made from this terminal box to the wave analyzer,the cathode-ray oscilloscope, or to the 12-channel amplifier andrecording osclllograph. Provision was made for applying a calibra-ting voltage of lmownmagnltude to each ohannel of the enplifler.

Amagnetostriction-type pressure indioator was installed in atapped hole in the cylinder and connected to the terminal box.

Construction and Application of Strain Gages

Strain gages suitable for airoraft-engtie application must beresistant to high temperature and engine oil. A bakelite-bonded

reslstanoe-td~e electric strain gage fulfilling these requirammtshas been devdoped by the NACA. (S00 referenoe 2.) This type ofstrain gage can be ussd at temperatures up to 500° F, Is unaffectedby engine Gil, and is ruggad and durable. Some of the strain gagesused in this invostigatlon were made ky the NACA, using Advance wire.The other gages were bakellte-boniledstrain gages coumerclallyman-ufactured,uskg Iso-l?lastlotire. (See table i~ fig. 4.)

The strain ~ages were attached with Bakolite cement, BC-6035,and the entire assembly was baked in an oven. The oannectlng wiresware attached to tho crankshaft with Balmllte cemmt at the smuetimf3,

Gage constant

When a resistance-type strain gage Is cemented to a test obJeot,the component of strain in the test material in the direction of thegage axis prcduc~s a oha.ngein the el~ctrlcal resistance of thestrain gage. This ch=~e in resistance hes Ixmn fcund to ba a linearfunctlcn of the strain w~thin the elastlc limit of thu tust material.Th~ ratio of unit change in strain-gagu rtmlstance to unit strain inthe member is defined as the gage Congt=Lt and may bo exprossad as

AR—.k)(~R

(1)

AR change in strain-gage r“.-sistance,ohms

R resistance of strain 6age, ohms

k $ag13constant

c strain in material at sw.’l%oe,mchus per inch

The value of k d~pends upon the material used for the gage-reeistanca wire.

Basic Strain-Gage Circuit for Single Strain Gage

The basic circuit for measurements #.th a single strain gageis shwn in figure 5(J). The strain gage is ~; Rd 1s a leadresistor that is approximately equal to R .

eUnder the conditions

of zero strain, the voltage drop o ucross the strain gago is

rmc~ Am Ire.E5Bol

where

.““~ Iia%iyvolkiige ‘ ‘ - ~ “ “ ~ --.... ...

5

Under strati the gage resistame ohanges by an amount AR andthe voltage aorose the strain gage beoomea

RR+tJ?# e’ = % Rg+AR+~

In the measurement of vibratory strains, this oircuit Is ooupledto the amplifyhg or measur~ oiroult wltb a Gondenser that blookaoff the direot mmponent of battery voltage. The alternating oompo-nent of battery voltage aoross the gage is

Eao =

u

where

cl-e

Ea. alternati~ oamponent of battery voltage, peak volts

Inasmuch as AR in practice is very small compared with Rg

=d %J the last term in the denominator may be dropped with neg-ligible error. When it Is assumed that ~ = a X Rg, the result Is

EBAREac=~x(l~a)2

Substituti~ frau equation (1) and rea&~ing

(2)

., . .

In the ease WhOrO ~ = ~g~ then a = 1 and the expressictubeccmes

(3)

II II II IIHII ■ -l -1 mm ml ■ ■■■ m-mm. mm . ..- —--- . . .

1

6

The stress (lb/sq In.) Isvalue of modulus of elasticity

NACA

fmmd by substitutingE (lb/sq In.) for the

4E Eacstr9ss = —

~k

ARR No. E5601

the properteat material,

(4)

Measurement of Bending Streins

For the measurement of binding strains, a pair of similarstrain gagas is applied to the tsst memt9r back to back as shown infigure 5(d) and connect9d as shown In figure 5(l?), No load resistorio required. When a tmdir~ lead Is applied to the test member, thetop strain gage Is in tension and Increases in resistance, whereasthe bottom strain gag9 Is in compression and decreasss In reslstanca.If ancther load, such as a tensile lead, is applied in addition tothe bending, each strain gage will Increase in resistance an equalamount with a cancellationof eff9ct. The circuit is therefore sen-sitive only to hendlng strains. ‘Whenthe same kind of derivation isused fcr this circuit as is used for the single atraln gage, theexpressim for bending strain can be shown to be

2Eacstrain = —

~k(5)

It should be noted that the sensitivity obtained Is twice thatof the siqgle strain-gage caee ~f equation (3); that is, with equalbatt9ry voltage, gage constemt, and strain, twice the output voltageIs produced.

~Measurem9ntof Tensile-Ccmpresslve Strains

For the measureme~t of tensile-compressive strains, a pair ofsimilar strain g~es is applied to the test member back to back asshown in figure 5(e) and connected as In figure 5(0), The loadresistor Rd is usuelly made equal to the sum of th~ two strain8We9 Rgl and Rg28 When a pure tensile or compressive load is

applied to the test-member) each strain gage ~qually Increaaes crd~creases ia resistance. If an additional load, euoh as a bendingload, is applied to the tqst member, cne strain gage till increasein resistmce and the other will decrease in reslstanoe by the sameamount, with a canceletlcn of effect. The arrangement thereforeIs aensitlve only to t9nsile-compressive strains. If ~~ and

Rg are similar and If it is assumed that ~ =2

expreasim for strain can be shmn to b9

a(Rgl + ~g,), the

W!.OAARR NO. E5601

-.. ... - - strain . %0(1 + 8)2““l?gka

and.when p., + %2) = ~, the expression

strain = %~k

The sensitivity of this ocaubinationisstrain gage.

The fsrce

Foroes and Moments on

exerted on the orankplncomponents in the crenkweb as shown-in

7

(6)-.,---

is

(7)

the same as for a single

Crankshaft

may be resolved into fourfimares 6(b) to 6(e). The

rea~tion of the ccunterwsi~hteproduces ti-dltio& fcrce;. Pairs ofstrain gages were located to measure separately strains resultingfrom components (b) and (dj of figure 6, &a shown by figure 7. Com-pcne~ts (c) ard (e) of fl~ra 6 can also be measured by strain g“gesif dealred. %ge 11 waa epplied to measure straina in the fiilet!?Xdgage 12 wascounterweight.shown In figure

The use of

applied to meaaure 9tr&ins in the che9k near th~The combined circuit for all the strain gages ia4.

Determhatian of Stresses

a 12-channel amplifier and recordlmg oscillographmakes possible the reoording of several strain reoordsj a cylinder-pressure record, and a tlmi~ trace. These simultaneous reoordafacilitate the interpretation of forces. The 12 amplifier operateat cmnstsnt amplification and eaoh includes an input attenuatorthat can be set to give a known ratio between input voltage to theattenuator and the voltage applied to the amplifier.

In order to calibrate the system, an alternati~ voltage ofInmwn amount is applied to the input of all channels at a certainattenuator setting and an osoillograph record Is made. Records ofthe strain signals under test conditions are then made, the atten-uator setting being carefully noted. Inasmubh as the”height oftraoe produced by the lumwn voltage at a lumwn atteniiationand theheight of traoe of the unknown strain signal at a known attenuationis established, the alternating strain voltage Eac can be deter-mined, When alternating strain voltage, strain-gage constant,

battery volta$e, and modulue of elaatlolty are substituted In theproper formula, the stress Is obtained. An aoouracy of M perosntin the detamil.nationof stress is believed possible.

The wave ~lyzer was used to analyze the hmnonic omuponentsof the mmplex-wave fozm of the strain signals and gave the valueof Eao for each ocxaponantin rms volts.

Slip-Ring Interference

A serious faotor in wire-strati-gagemeasurements on rotatingmembers 1s the varlatlm of resistance oaused by slip rings In theciroult In series with the strain gage. If the strains are small,this variation appeerl~ as an IR drop might be of the same orderas the strain-gage signal and w1ll cause error In the measurements.It Is possible, however, to obtain satisfactory results if brushesand slip rings are operating correctly. The data in thle reportwere taken using the series type of olroult shown In figure 4,

Some of the measurements were repeated with the strain gagesconneot9d In a Wheatstone bridge arrangement. The results were inegreement and it was found that the bridge oonneotlon praetioallyeliminated the sllp-ring Interference. Figure 5(f) shows the baslobridge circuit of a pair of gages amd figure 5(g) shows a combina-tion of oirouits on a rotating menker. The oirouits are suppliedby a ooxmnonbattery. Resistors Rl, R2, and ~ are mounted on

the rotating shaft. The condaneers serve to bypass the hi@-frequenoy components of the interfere~oe,of the bridge cirouit are a disadvantage.

TIWI PROCEDURE

In the tests reported herein, inertiaby driving the engine with the dynemometer

The added complioatlons

forces were determinedat a range of speeds,

&king os~illogra~ records, and-taking wave-analyz& read~ngs ~tseveral points. Peak stressee were determined from osclllographreoords as previously described and the hazmmnlo components ofstress (expressed as multiples of crankshaft speed) were calculatedfrom wave-atvalyzerdata.

In order to determine the exoitlng foroes oaused by gas pres-sure, the pressure-lndioator output was analyzed for the variousharmonic mmp~ta of cylind9r pressure with the engine underpower.

HACA ARR ~0. E!i601 9

Strains ~oduced by the combination of inertia and gas-pressureforces.were .detemned with the engine operating at mnstent power(Imep, 227 lb/sq In.) and aremge of speeds. Osclllo~aph and wave-analyzer data were taken of the strain ~lgnals.

The mgine was then opsrated at constemt speed and a range ofmanifold pressures. The variation of the harmonio oompcments ofcylinder pressure w5th -fold pressure was “determinedby waveanalysis of the pressure-lndloatoroutput and the variation incrankshs,ftetresses were detemnlned by oscillograph and wave-analyzer data.

The engine was operated under severe knooklng conditions at anengine speed of 2400 rpm to Investigate the possible effect of knookon crankshaft stresses. The strain-gage signala were oarefullyobserved, one at a time, on a oathode-ray oscilloscope soreen andosclllograph reoords were made for further study.

A torsiograph was installed on the rear extension shaft inplace of the pineapple ond osolllograph and wave-analyzer data weretaken at normal rated power (tiep, 227 lb/sq in.) aver a range ofspeeds.

All tests were cond’~ctedunder the follcJw.n&wine conditions:

011-in temperature, ‘%. . . . . . . . . . . . . . . . . . . . 160Temperature cf rear spark-plug bushi~ (rnnxlmum),% . . . . . 450Combustion-air temperature (except for knock test), ‘F . . . . 95Fuel-airrutia. . . . . . . . . . . . . . . . . . . . . . .0.095

The fuel used was 100 octane.

RESULTS AND DISCUSSION

Typical osclllograph records taken during the various tests areshown in figures 8 to 13. The reoords have been retraced by hand inthe reproduutlon prooess.

Some representative ourves shmdng the bending stresses causedby inertia foroes er9 shown In figure 14. An arbitrary straight linehas been drawn through the peak-stress points to represent theirIncrease with speed. It oan be noted thet a large third-order stressis present. Other harmonics were found In addition to those plotted.A value of E = 30 X 106 pounds per square inch was assumed in thestress calculations. Two typical osoillogr~ph records taken withthe engine being motored are shown in figures 8 and 9.

I —. —

. . -.

10 XVACAARR ~0. E5B01 ‘“

Results of the harmonic nnalysis of gas pressure in the cyl-inder are plotted in fi~re 15. It can be seen that there Is aharmonic component of gas pressure at each order and half order,

successively decreasing In amplitude, Half orders higher than ~were not determined. Orders and half ordars higher than 9 werefcmnd but were too small for dependable measurement, The samerGsults were obtained at different speede. ‘lThisfigure 1s shilarto that gimn by L&unbaum (reference 3). The dashad lines showcmapGncnts of cylinder pressures while the engine was being motored;the inertia forces therafcme Includo the forces caused by thepumping action. The actual cylinder pressures were not determined.

With the e~~ine operatlri under power, the alternating strainsresult frcm the combinad inertia and gas-pressure fcrces. Curvesof the peak stress and of the tiportunt harmonic components areplott~d In figures 16 and 17 for bending and tenslla-compressivestresses, ruspoctlvely. A comparison of figure 16 with figure 14shows that the half- and first-ord6r harmonics have increased inmagnitude but that th~ third-order hmcnic has decreased somewhat.This decrGase ~~y be explained by the fact that the gas-pressure-excitGd third-brder ccmponent appsr9ntly acts in phase opposition tothe linger Ir.ertin-excitedthird order. It can be noted from fig-ure 17 that the third-order harmonic of tensile-compressive stressIs the lnrgcmt of ths harmcmlcs. I&cm figures 16 and 17 it can benGted that the inertia-axcited stresses fcrm a large part of thetotel strossos, The largo third-ord~r component Is undoubtedlychnraeteristic of the three-cylinder test engine and wculd not beexpected in tho rsgclar njna-cylinder engine.

Figures 10, 11, cmd 12 shcw oscillograph records of strainGlgnals tnkon whllo the engine was under power. It can be seenthat the peak stress9s increase with both speed and manifoldpressure.

A simultaneous record of wveral strain signals is presentedin figure 18. A careful study and comparison of the various tracesreveals Ird?ormatlonon the nature of the forces applied to thecrankshaft, such as:

1. Th9 peak in the bending stress from strain gages 2 and 3cozncides with the maximum cyllndcr preseure.

2, The tensile-ocmpressive stresses flmm strain gages 4 and 5resemble those from strain gages 6 and 7 but are opposite in directlrn.

m..

NAM

gage

Am No ● E5Bol

3. Stresses of appreciable12 located near cne of the

11

_tude are shwn by strainoountmwelghts. In the regular

englne%zlth-thependulum dampers, in?ormat~on oh the toroe~ exertedon–the crankshaft by the dam~ers”may be gained by similar strainmeasurements near the damper holes.

Strain gages 8 and 9 ad strain gage 11 produced signals too smallfor dependable measurement and were disconnected.

A plot of relative pressures against mean effective pressure(flg. 19) indicates that eaoh harmonic compunent of the cyllnderpressure increases with manifold pressure. These results aresimilar to those shown in reference 4, The Increase In gas-pressureexciting forces results in increased bending and tensile-omnpressivsstresses as shown ~n figure 20. The exception hers again is thestrcq, inertia-excitedthird order.

Cargful observation fallgd to show any effect of hock in thestrmn-gage slgmls won under severe krackir~ conditic~~. ml ssbservwhcn is ~n agreemer.twitk the thearetlcal enalysls of ‘&lger(ref9renc9 5).

Curves of the peak torsional-vll,rationamplitudes ~d of themain harmcnlc compuner.tsarn shcm in figure 21 and rItypicalosclllogra~h raccmd Is shcwn ic figure 13. Th.2hump m tho first-order harmonic In figure 21 at an cr@n9 speed of 2100 rpm lGbelicwad t,obe cauoed by ~1rusmanc~ in the crunkshaft-fl~heul-dynamometer syqtem ~f the test c~.me. A omcpari:~oncf th(>torsional-vilratlon ampli~udes with the stresses shown in fjgures 16and 17 shows no direct relation. !lhafirst-order pGak :.snot ljre~~ntin the stress curves and the prcmnent third-order stress Is notapparent in the torsional vibrations.

SUMWRY OF RESULTS

From an investigation of the measurementin an airoraft-engine crenkahaft under power,were obtained:

of operating stressesthe following results .

1. Inertia-excitedbending and tensile-compressive stressesincrease almost directly with speed.

bothIs a

2. A large inertia-excitedthird-order stress was found inthe bending and tensile-compressive cases. This excitationoharacteristlc of the three-cylinder test enginG.

12

3. Harmonics of cylinderwere fmmd far each order and

NACA Al& NO. E501

gas pressura of appreciable ma@tudehalf order up to 9.

4. The peek in the bending stress in the web ooincides withthe maximum cylinder pressurg.

5. Each harmonic of cylindar gas pressure was f&nd to increasealmost directh~ with manifold pressure. The measur’edstressesincremed In similar fashi~n.

6. No effect of hock on the cranhhaft stresses was notad.

7. The torsional-vibration amplitudes at the rear of the eru?lnewere not directly related to the s&esses In the crankshaft.

CONCLUDING RlMARK3

1 The results presented demonstrate the practicability-. ofmaki~~ direct strain measurements on a crankshaft under operatingconditions●

2. Strain me$mmements on an erqginecrankshaft, either inflight or in a prqmller test stand, maybe made. The applicationof strain gages roqulrea ordy minor alterations and the mounting oft.hc@nsapple In place of the starter is not a serious problam.Amplifying E@ rec~rdirg indawments for flight use are available.

3. !4cstof the rudial aircraft engines now in use have a hollowshtit couplgd to the cmnkshaft acceQsltle at tha rear of the engineand are well suited tc a strain-gagelatlcn a~ ~m in-llrieengine would be

hrcraft Engine Rmearch Laboratory,National Adviacry Commttae for

Clevgland, Ohio.

~nstall%tlon. Such an lnstal-moru difficult.

Aeronautic,

REFERENCES

1. Curtis, W. F.: Dynamic Strain Measurements m the Crankshaft ofa Dios31 l!hginc. Rap. R-154, Eur. ,Ships,Navy Dept., Oct. 1943.

2. Nattles, J. Cory, and Tucker, Maurice: Construction of WtreStrain GagoQ for I@na Application. NACA ARR No. 3L03, 1943.

NACA ARR NO. E5B01 15

3. L~renbaum, Karl: Vibration of Oranksheft-PropellerSystem.SAE Jour. (Trane.), vol. 39, no. 6, Dec. 1936, pp. 469-478..

4. Wilson, W. Ker: Practical Solution of Torsional VibrationProblems. Vol. 1. John Wiley & Sone, Inc., 2d cd., 1940,pp. 452-456.

5. Gei8er, J.: Engine Stresses Induced byRapld Pressure Rlee andFuel Knock. (lEromAuto, tech. Zeitaohr., Jahrg. 44, Nr. 13,July 10, 1941, pp. 327-335.) R.T.P. Trane. No. 1433, Mlnhtryof Aircraft Prod. (British).

4I m

AB

f-

D

E 1

Hanoaeter for coollae-a~r pressure drop :Haflomet*r for cooling-air pressure (upstream) zCoo//nkwlr re~ulator/ianom#;cr for ;an/fold pressure mCombust/on-atr surge tan& a

Fuel pumpm

Combustion-afr-temperature thermocouple =Cooling-olr-temperature thermocouple (upstream) ~Combustion-air flltcrCool/ng-oir-temperature thermocouple (dounstreanlmDynamometer scales cmFlyuheel aFuu/-neusuring rotamctcr

o

EngineFlexlble couplingDynumomturExhaust mffierOil heatur and coolerCombustiofi-air-tcmPeroturc thermocouple011-circuioting pumpCombust/on-a/r heater ,.

Combustion-air-prcssur# regulatorHunoueter for codwstion-oir-orifice differ#ntia/pressureitanomter for combustion-alr Pressure (uPstrcam)

un T \

El

!

k-l r’- A

IJiuu x

~ Figure 1. - D!agramatic sketch of test engine and auxiliary equipment.

NATIONAL AOVISORY

COM41TTEE FOR AERONAUTICS

nm●

.

I

011 seal

Alteration fdirect drive L

x

r Oil seal :

neapple

ADVISORY

COWITTEE FOR AERONAUTICS

n

Flgun? 2. - ~~;:~!fied cross section of test enfine showing method of bringing out ~m

N

.#

HACA ARR HO. E500i Ffg. 3

Figure 3. - Pineapple instollot}on at rear of engine.

NACA ARR NO. E5BOJ FiPressure Iftdicdtor

cage

12 and 34 and 5

6 and 78 and 9

101112

I 13

Funct/on

Single

Bend / ng

Tensf/e-compress Iue

Tensile-compress iue

Tensile-compress Iue

Single

Sfng/8

Singie

Singie

Resistance(ohms)

425498505

503408

392210

400

210

Strain gageconstunt. K

2.003.433.433*432.002.002.00

2.002.00

Str::r@gage

Aduonceiso-Elustic

iso-Eiostic

iso-Eiastic

Aduance

Advance

Advance

Aduance

Advance

.

Iing

1/2

1/4

1/4

1/4

1/2

1/2

1/4

1/2

1/4

NATIONAL ADVISORY

CWITTEE FOR AERONAUTICS

Figure 4. - Combined circuit for all strain gages.

NACA ARR Ho. E5B01 F-ig. uPressure Indicator

.

fd. couplingdensers-

rGage

12 and 3

4 and 5

6 and 7

8 and 9

1011

12

13

JTimer inpineapple

Funct/on

SingleBend i ng

Tensfle-compressive

Tensile-compress iueTensile-com~ress iue

Sfng/e

Single

Sfngle

Singie

Resistance(ohms)

425

498

505

503408

392210

400

210

Strain gageconstant, K

2.003.433.433.432.002.002.00

2.00

2.00

Str::r@gage

Aduance

lso-Elostic

/so-E/ustic

lso-Elostic

Aduance

Aduance

Aduance

Aduaace

Aduance

Strain gageIenfth (in.i

1/2

1/4

1/4

1/4

1/2

1/2

1/4

1/2

1/4

NATIONAL ADVISORY

CtiMMITTEE FOR AERONAUTICS

figure 4. - Combined circuit for all strain gages.

fzEI(a) Basic circuit for

single strain#age.

(((j(d) Pair of strain

gages appliedfor bendingmeasurearentsc

tJE3lb) Circuit using

pair of stra;ntQU@S for bend-ing 9eosur8-wents.

, —- -— .

(e) Pair of straingages appliedfor tensile-com-pressiue measure-ments.

m(c) Circuit using pair

of strafn gagesfor tansi li-con-presslue measure-ments.

‘“m(f)

!r’ I v N?’” I v“ \ “2 T 20 WJ.50 fid* I

r 1Uheatstone bridgeconnection forbending measure-ments.

.

I I ~ac(single)

NATIONAL ADVISORY

COMMITTEE FOR AERONAUTICS

I

I

(h?)

1 Figure 5. - Strain

fac (badlng)

LOc (tensile-compressive)

Combination of bridge circuits’ on rotatingmember.

tage applications.

.

NACA ARR No. E5BOI

.,.

Fig.

(b)

/d)

o

$>Bending momenml to web.

(a) Three-cylinder test e~gine.

t nor- (c)

NATIONAL ADVIS~Y

C~lTTEE f~ AERONAUrlCS

Tensile-compressive (e)force on web.

%

?

.

8ending momentplane of ueb.

g

~

TNisting momentweb.

6

in

on

Figure 6. - Forces and moments on cran&shaft.

I

.

fiotatiofl

-a- EE1-17z3)t-

— —

KI L———-l---i-u l-l / I \

—

xo●

NATIONAL ADVISORY 7-.

COMMITTEE FOR AERONAUTICS La

Figure 7. - Location of strain gages on crankshaft. ●

u

HACA ARR HO. E5BOI Figs. 8,9,10,11,12,13

rims

(0)

- .

(b) 8,200 Ib/sq !n.

r “

*One

●lop deOO

rwolutfon center● b

(o)

(b)

v

rl~ —e

5,YCJ0 Ib/gq in.y

-’vV w- >

One TOp QCodrfwolut 10*

●center

Figure 8. - Record taken v[th mginebeint motored. Engine spi?ed,2B0 rpn; (a) gages 2 and 3, bend-ing; (b) gages 4 and5, ttmil+ccwpressive.

Figure 9. - Record taken uith enginebeing motored. Engine speed,2700 rpn; (al gages 2 and 3, bend-ing: (b) gages 4 and 5, tensii+ccwpressive.

Figure 10. -Record taken with ”engineunder pmer. Engine speed, 2B0 rpm:mw?ifold pressure, 30 inches mercuryOtrsoiute; (a) gages 2 and 3, bend-ing; (b) gagas 4 and5, tensile-cm-pressiue.

L 4Figure 12. - Recwd taken uith engine

under pm?r. Engine, speed, 2700 rm;manifold pressure, 40 inches mercuryabsoiute: (a) gages 2 and 3, b~d-ing; (b) gages 4 and5, tensi/e-compressive.

Pressure

Tim b

(0) . ) . 600 lb/sq in.

bh “ 4

Al

(b) -14,300 lb sq in.

Figure U. - Record taken uith engineunder pmz?r. Engine speed, 2300 rpm:manifold pressure, 49 inches nercuryabsolute; (a) gafies 2 and 3, bend-ing: (b) gages 4 and 5, tensile-cm-pressiw.

Figure 13. - Rectml of torsionalvibrations taken with engine underpouer. Engine sped, 2300 rm:mnifoid pressure, 40 inches mercurYabsolute.

NATIONAL ADVISORY

COW ITTEE FOR AERONAUTICS

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

NACA

. ..

ARR NO. E5B01 Fig. 14

MAT IOUAL ADVISORYCOWITTEE FOR AERONMICS

o Pe”kkStreasn Half-order harfnonla

. e * ~ eA Third-or er rmonlc

. .

6 .

—

Figure 14. - Hending stresses caused by inertia forces withengine be~ng motored. Gages 2.and 3.

3.00

2.75

2.50

2.25

2.00

1.75

1.50

1.25

1.00

●75

●50

[.25

t--OL-

I

II

I

II

1

Figure 15.- Harmonicindicator. Engine

NATIOMAL AOVISORYCOMMITTEE FOR AERONAUTICS

@’ D wer haaQ.z :01 d

pre5sure, 40 h. abs●;imep, 227 Ib/sq in.)

--- Eng1Re be~ng m >tore3

h

1z 3 4 5 6 7 a 9

Order number of components

components of gas pressure in cylinder as measured by pressurespeed, 2300 rpm.

.cm

NACA ARR HO. E5B01 Fig.

NATIONAL AOVISORYCOWITTEE FOR AERONMTICS

T---- Peek atre6s,With en 3 he ~elng

mot!>red (from f18. 14)o P

mA Thir(l-oralm harmonl:

12m 5

11~ (

10‘,\

c)

9() / (>

()

8

7 < r

6 ‘/“

.5

//“

/ e

4/ /

/(

/.-

//3 . /0 4>/

,.~

2<r

<*---? —4h--

4b~~ b

1. I ,1 \c bJ— —

0$ 1600 1=0 2000 2200 2400 2600Engine speed, rpn

F@me 16. - Bending stresses with engine under power.Manifold pressure, 40 Inahes mercury absolute; gages2aw13.

~ —.-––...–.– —...—

.

NACA ARR No. E5BOI Fig. 17

0--0 Peak10a ~ - ~ u ~ -—

o15.

14 /( )

NATIONAL ADVISORV (

CulTTEE FOR AERONAUTICS13

/

f -/t) () /

12 r / //

/11 .

s~ ~f )= /’

10->

//

9 .()

//

8 / ) ()/

() /)

7./

() //

6 /()

/

5. \i

4Ai

/

3’ Ak=AL

211— J

~— —$ =f ) -[

Y* <

1

{ >

& 1600 MOO 2000 2200 2400 2600Engine speed, rps

Figure 17.- Tensile-compressive stresses with engine being~tored and under Power. H@ifold pressure, 40 inchesmerc~ absolute; gages 4 and 5.

. ...-. . ,.,.- .. . --—————-...————.

NACA ARR NO. E5BOI Fig, 18

\

19,000 lb/sq In.

I _9;ZO0 Ib/sq In.1, I I

I-r’i’oo ‘b’sq‘“”1, I I A

15 ,600 Iblsq In.

\

I r14. 400 Iblsq In. I I I

IT- ”--

~ ‘0””’ ““’’’””’’’””’sO”center -1 CWIT7EE FOR AERONAUTICS

Cyl Inder

pressure

Gage 1(single)

Gages 2 and 3(bendlnf] ~

Gates 4 and 5(tensi/e-

compress iue)

Gages 6 and 7(tens he-conpress i ue)

Ho signal

Gag. 10(sinrlel

No signal

Gage 12(single)

igure 18. - Sin?ulrafieous record of several strain signals.Engine speed, 2400 rpm: manifold pressure,40 inches mercury absolute.

NACA ARR No. E5BOI Fig. i9

NATIONAL ADVISORYCOMITTEE FOR AERONAUTICS

A Twoe

2.0 ,

1.E?

1.4 .❑

1.2 /r

l.O ~ / /f

/. /

.2T IMan:fold pres~lure, in.

30 3: , +0 45100 100 200

Indicated mean effective presuure, lb\sq in.Figure 19.- Harmoni’ccomponents of cyllnder pressure atmanifold pressures. Engine speed, 2300 mm; fuel-air

variousratio,

0.095.

I(ACA ARR Ho. E5BOI

—

Fig. 20

9.

.—.

15. IC--J43

14

13

12

11

10

9

8

7

6

5

4

3

2

1

0

1

/

/u /

/ ~D

(>

>

/

NATIONAL ADVISORYCM ITTEE FOR AERONAUTICS

>---~

L h— . . — . — - , )

L*

*n-

“n---

n- ~ 3~

- ManM:ld pre~ m bE4: i9[r

L0 100 200 300

Indicated mean effective pressure, lb\sq in.Figure 20. - Vnriation of stresses with manifold pressme.Engine speed, 2300 rpm; fuel-air ratlo~ 0.095.

-.

NACA ARR No. E5BOi Fig. 21

+1

.5

.4

.3

●2

.1

0

NATl~AL h ISORYCOMMITTEE FOR AERONAUTICS

o

cA

~ Thixd-ord.er ha,rmon~c“

t,~(

..

c— [1

</Lh

iL—4L

AH Y<)~< >b

1600 2000 2200 2400 2600Engine speed, rpu

Figure 21.- Variation of torsional vibration with speed as measuredwith a torsiograph Installed at the rear end of the crankshaft.Manifold pressure, 40 inches mercury absolute.

I I1, .— —

.-’