internal combustion engines - forgottenbooks.com · with the growth and development of the...
TRANSCRIPT
PREFACE.
TH AT this work is pla ced on the market a t all is due principal ly to the la ck of satisfactory , compact reference bookstreating on the subj ect in question .
There are many excellent books of reference which treatthe subj ect from a theoretica l standpoint and deal largelywith the growth and development of the internal-combustionengine. Many of these books
,however
,have not been
brought down to date and,while beyond reproach as expo
nents of theory, fall far short in the matter of present practi ceand modern design . I t would be well to supplement theuse of this book with any one of several works on the gasengine, in order tha t the mathematical side of the subj ectmay not be slighted . Works by Clerk
,H utton, and Donkin
are particularly available along these lines .A complete knowledge of thermodynamics is inva luable
for the perfect understanding of the theory of internal-combustion engines, one of the best text-books on this subj ectbeing “ Thermodynamics
,H eat Motors and Refrigera ting
Machines, by De Volson Wood.
H owever, it has been the a im of this work to eliminate, asfar as pra cti cable, the more involved mathematical formulasand to confine the matter conta ined to the more practicaland applied phase of the subj ect. I n the chapter on “ Com
pression several thermodynamic formulas have been used toprove the relation of the compression to the thermal efficiency ;these formulas, however, have no immediate bea ring, exceptin a general way
,on the problems of actual design and
operation,but the formula PV” C, by far the most impor
tant formula used in the actual designing, is found andderived in this chapter
,and its discussion is ta ken up in the
following chapter on“The I ndicator Card .
V
PREFACE
For practical information and data contained in this workthe author is indebted
,to a large extent, to different manu
fa cturers who have placed the result of practica l tests at hisdisposal . I t has been the intention to use only that information which appeared most reliable and in keeping withactual practice .
I n the matter of design,average practice has been con
sidered, and while the formulas given should not by any
means be taken to give results in keeping with each a nd
every engine on the market,their use will insure results
closely in keeping with the average .
The tables and formulas herein contained should fill allaverage requirements
,either for the designer or the operator
,
and while neither original nor compiled especially for gasengine practice (they may be found in any standard handbook), the fact that they may be found here assembledshould be of advantage .
I t is the desire of the author in issuing this book that itmay find a place for itself and fulfil l
,in part at least
,his
intentions .W I LLI AM M . HOGLE.
TOLEDO, OH I O, December, 1908 .
CONTENTS.
I NTRODUCTORY .
PAGEH istory of internal combustion a s a motive power .
Development of internal combustion a s a moti ve p ower .
CHAPTER I .
-" TH E BEAU DE ROCHAS CYCLE .
Sequence of cycle .
CHAPTER I I .
— TH E CLERK CYCLE .
Descript ion “
of Clerk principle
The Robson engine
The Day engine .
The Day cycle .
CHAPTER I I I .
— TH E DI ESEL MOTOR.
Compression spaceFuel econ omy .
Relation of card to cycle
CHAPTER I V.
— COMPARI SON OF THE CYCLES .
PAGEThe {Our-cycle principleAutomobile motors .
Marinem otors .
Comparative power development of two a nd four cycle engines .
The Diesel motor
CHAPTER V . PRACT I CAL OPERAT I ON .
Starting a stationa ry engine
Starting an automobile or ma rine engine
Cooling of engine cyl inder
The valves
Proper care Of governor .
Proper mixture for successful operation
Troubles a nd Rer‘
nedies .
Failure to start
Carburettor out of adjustment
Engine stops .
I gniti on tube cold .
M ixture too:rich Back firing in exhaustBack firing in compression stroke
CONTEN TS. ix
CHAPTER VI . START I NG DEVI CES.
PAGEMethods of startingStarting cams
CHAPTER VI I . CARBURETTORs, VAPORI ZERS , AND I NJECTORS .
Methods of securing explosive mixtureThe carburettorRequirements for successful carburetting of fuelCa rburetting alcoholCarburetting petroleums
The in jector principleThe H ornsby-Akroid method of fuel in jectionThe Meitz and Weiss method Of fuel in jectionThe Diesel valves a nd method of fuel in jectionThe Daimler carburettor
Surface carburettor .
Spray carburettorMixing valvesThe Schebler carburettorThe H olley carburettorAlcohol carburettors
CHAPTER VI I I . PRODUCERS .
Pressure producersFuels available for use in pressure producersDist illing producers .
Qua ntity and heating value Of gas from distilling producerCombustion producers
Quantity and heating value of gas from combustionThe suction producer .Fuels available for suction producers .
Operation of the suction producer .
Comparison of steam and ga s producer power plantsGas analysis from suction producer
X CONTENTS .
CHAPTER I X . FUELS AND COMBUSTI ON .
PAGEGaseous fuels s
Advantages of gaseous fuelsNatural ga s compared with producer ga sBlast furnace ga s .
H eating values of fuels (tabulated)Volumetric analysis of Pennsylvan ia gases (tabulated)Analysis of gases .
Liquid fuels .
Petroleum distillates .
Properties of petroleum disti llates (tabulated)Composition of crude oils (tabulated).
GasolineKeroseneH eat of combustionMeasurement of heatAir necessary for combustionAir required for combustion of different fuelsVaporizationRequirements for complete vaporizationLaws for perfect gases .
Vapor pressure of saturationAvogadro
’s law a s applied to vapor pressure .
Temperature necessary for a perfect m ixtureAcetylene .
AlcoholRelative heating values of gasoline a nd alcoholPower derived from alcohol a s compared with that derived fromgasoline
Cost of alcohol a s compared with gasoline
CHAPTER X . COMPRESSI ON .
Limits to which compression may be carriedCompression temperatures (tabulated)I gniti on Obtained by means of high compressionDerivation Of the formula for the ideal indicator ca rdThe theoretical card
CONTENTS X1
CHAPTER XI .
— TH E I NDI CATOR CARD.
PAGEWhat constitutes a perfect cycle in a ny given cylinder H ow com
The cams as related to the cardValues in general use for 7 .
Computation of values for the ideal cardDetermination of the constant for the expansion curveDesign of engine as related to ideal card .
Chart for determining compression pressure
CHAPTER X I I . GENERAL DI MENSI ONS.
Mean effect ive pressure .
Average values of mean effective pressuresDetermination of bore a nd strokeThe fuel factorMechan ical efficiency of multiple cylinder engine
CHAPTER XI I I .
— TH E CAM MECHAN I SM .
Location of the cams .
Transmission of cam motion to valvesCams with lever transmission .
Shifting of lever to bring starting cams into OperationCams classifiedMethod of laying out single cam
The double cam .
Application of double cam to vertical engineMethod of laying out double cam .
Ma terial necessary for cams
Types of gearing in useSpeed ratio in skew gearingAdjustable gear .
Fiber gearing
xl 1 CONTENTS
CHAPTER X I V.
— TH E VALVES AND PORTS .
Mushroom valves .
Effective valve opening .
Design of inlet a nd exhaust passagesDetermination of eff ective valve openingMinor valve dimensions .
Methods of setting valvesThe suction inlet Valve .
Ports in two—cycle design .
Design a nd location of two- cycle portsThe exhaust port lead .
CHAPTER XV . TH E CYL I NDER .
The air-cooled cylinder .
The water—cooled cylinderThickness of cylinder wallDepth Of water jacketThickness of outer water jacket wallCopper water jacketsLength of water jacket .
Design of cylinder to fac ilitate boringOpenings for in let a nd dischargeGrinding of cylinder .
Bolts .
Material for cylinder ca stings
CHAPTER XVI . TH E FLYW H EEL .
Function of flywheel .
Calculation of weight of WheelDesign of flywheelTable of keys .
CH APTER XV I I . TH E FRAME .
Purpose of frameAdvantage of heavy frameFrame for horizontal engiFrame for vertical engineThe crank- case engine .
The sub-base
CON TENTS
CHAPTER XVI I I . ENGI NE FOUNDATI ONS .
Drawings for foundationsAdvantage of good foundationMaterial for foundat ionDesign of foundationFoundation boltsLaying out foundation (the bolt template)
CHAPTER XI X .
— TH E CRANK SH AFT AND RECIPROCAT I NG PARTS .
Style of piston .
Strength Of crank shaftDesign of shaft a nd length of bearingThe balance weights .
Determinat ion of necessary weightsCrank shaft bearings a nd brasses .
Oil rings .
The connecting rod
The Piston , Wrist Pm a nd Piston Ri ngs.
The wrist-pin bearing .
The outer diameter of pistonDesign a nd construction of ring .
The two-cycle piston head
CHAPTER XX .
— GOVERN I NG DEV I CES .
Methods of governingThe governor controlling mechan ismDesign of centrifugal governorThe simple fly-ball govern orThe loaded governorDevices for throttlingThe inertia governor
CHAPTER XXI . I GNI T I ON .
Methods of ign iting chargeJump
-spark ign itionMake-a nd-break system of ignition .
xiv CONTENTS
PAGENon—inductive resistance a nd condenser .
Connections for single cylinder with Ruhmkorff coilWiring diagram for four-cylinder engineUse Of commutator .
Types of make-a nd-break I gn itersThe commutator how constructedTypes of commuta tors . .
The spark plug I nsulatio n,et c
Types of Spark plugs .
Dynamo ign ition .
The Apple ign iter. .
The Bosch type of dynamo .
The Remy magnetoFlame ignitersBarnett ignition cockThe hot -tube ign iterAuto-ign ition .
Time of ign itionFiring order for multiple cylinders
CHAPTER XX I I . ENGI NE TEsTI NG .
Methods of testingThe Prony bra keDerivation of brake formulaFactors for Prony brake (tabulated)The belt dynamometerTesting with Prony brakeLog of testTesting of gasoline, alcohol, and Oi l engines
CHAPTER XXI I I .
— REPORT OF TESTS .
Form of reportWeight a nd specific heat of gasesThe p lan imeter .
Determ ination of the mean effective pressureThe heat balanceDetermination of brake horsepower
CONTENTS
MI SCELLANEOUS
Defin ition of uni tsWire and sheet metal gauges (table).
Machine screw sizes (table)Wrought iron pipe sizes (table).
Circumferences a nd area of circles (table)Trigonometric functions (table)
I N TERNAL COM BUST I ON ENGI NES
the attention Of the public to the steam engine,the develop
men t along the lines of intern al combustion ceased , and itwas not un til about the year 179 1 that any suggestions weremade which were improvements on the engine Of Abbé deH autefeuille . I n this year an English inventor
,by name
John Ba rber,took out a patent on the use Of a mixture of
hydrocarbon ga s and air in an exploder . ”
I n 1794 this patent was followed by one covering theproduction of an explosive vapor by means of a liquid andair . This paten t wa s also taken out by an En glish inventornamed Robert Street .
I n the year 1799 Philip Lebon,of Brachay
,Fran ce ,
tookout a paten t on the prin ciple as well a s the constru ction ofa n engine using the explosion Of coal ga s a s in ot ive power .
This inven tor also took out paten ts on a pump for thecompression of the explosive mixture a nd a machine
,
operated by the engine,for the production of an electric
spark for igniting the charge .
The career Of this inventor term inating abruptly shortlya fter this time
,and before he had developed his inventions
,
closed what might have been an epoch-marking period ingas—en gine development .
From 1799 until 1860,in which year the first practically
successful engine was designed a nd built,several different
schemes were a dv a n ced . One brought ou t by W right in theyea r 1833 w a s very well developed from a theoreticalstandpoin t
,a governor bein g used in con n ection to vary the
mixture of ga s to make it proportion al to the work beingdone a nd to regula te the compression Of the charge .
A double- actin g engine produced by Johnston,and
devised for the use of hy drogen a nd oxygen,two parts of
the former to one part Of the latter,wa s somewhat unique
in its operation,and had it not been for the cost of the fuel
would doubtless have been used quite extensively . The
hydrogen bein g exploded , formed with oxygen a water vaporwhich on being cooled was precipitated and a partia l vacuumformed
,the unbalan ced force Of the atmospheric pressure
t hen acting during the return stroke of the piston .
I .VTI fODUCTOI I’ Y
I n 1838 Barnett took out a patent covering substantiallythe same ground as did that of Lebon
,two pumps being
used to compress separately the gas and air and then forcethem into the cylinder . The explosion was produced bymeans of the so- called Barnett ignition cock
,later described
in Chapter XX I I I on “I gnition .
”
The use of the magneto a s medium for producing thesparking curren t wa s suggested by Stepha rd in 1850 .
I n the year 1857 Barsanti and Matteucci devised a motorwith a very long cylinder fitted with a piston to which arack meshing with a spur gear on the fly-wheel shaft wasattached . On the explosion stroke a pawl allowed the rackto run freely
,but on the return stroke the pawl engaged
and the rack caused the spur gear and shaft to revolve .
The explosion of the charge drove the piston upwards in thecylinder
,and its inertia caused it to pass the point where the
internal pressure was equal to the atmosphere and in consequence a vacuum was formed . The cooling of theexploded charge increased this vacuum
,with the result that
the piston was forced down with considerable force .
I n 1858 an engine was devised by Degrand in which thegases were compressed in the cylinder , but because ofmechanical diffi culties it did not meet with any success
,
although the idea was a forerunner of the engine of thepresent day .
The appearance of the Lenoir motor in 1860 marked anepoch in gas- engine construction
, a s it was the first enginecapable of comparatively regular and effi cient work . The
machine was constructed along the lines of a double-actingsteam engine
,the ignition was obtained by means of a
primary battery and Ruhmkorff coil producing a j umpspark
,and altogether it wa s a very decided advance over all
existing forms of gas engines up to that time . But theLenoir engine was uneconomical , requiring about 100 cu . ft .of gas per hp .
-hr . and four times as much water for coolingas was used in a steam engine of l ike power . The great heatin the cylinder required that the piston be kept flooded withoil . I n V iew of these several difficulties the Lenoir engine
I NTERNAL COM BUSTI ON ENGI NES
disappeared in a very short time, but not before it hadstirred the minds of the inventors to renewed activity alongthe lines of the internal-combustion en gine .
I n the same year,1860 , H ugon introduced a motor in
which he attempted to keep down the temperature Of thecyl inder by means Of the , inj ection of a spray of water.This engine was more economical in the consumption Of gas
,
requiring a trifle more than 80 cu . ft . Of gas per hp .
-hr ., and
the temperature Of the exhaust gases was appreciablydiminished .
Several other ideas were advanced about this time,all of
them being either of minor importance or repetitions Of
previous attempts . I n the year 1862 M . Beau de Rochastook out a method patent setting forth
,theoretically ,
thebest working conditions for an internal- combustion engine .
H is cycle of operations wa s in all respects the same as thatin use at the present day in the so-called Otto-cycle engines .The following proposition s were embodied in his patent :
1 . The largest cylinder capacity with the smallestcircumferen tial surface .
2 . Max imum piston speed .
3 . Greatest possible expan sion .
4 . Greatest pressure at beginning of working stroke .
While the honor o f promulgating the theory belongs,
beyond a doubt,to M . de Rochas
,he did not in his patent
set forth a ny means for producing the theoretical propositionin practice
,and
,owin g to irregu larity in the proceedings
,
his patent became public property soon after the applicationwas filed
,but not until 1878 wa s atten tion again called to
it . I n that year the Otto ga s en gine , substantially a s itnow appears , was first placed on the market . Previous tothis time , about the year 1872 , Otto , in connection withLangen , placed on the market the so-called Otto and Langenengine
,of which
,due to its comparatively economical Opera
tion,they were enabled to sell quite a large number
,notw ith
standing the fact that it was of the free-piston type a nd
exceedingly noisy in its operation . I ts gas consumption
I NTRODUCTORY
was about 26 cu . ft . per hp .-hr .
,and the cost of energy
produced was somewhat less than with the existing steamengines .Continuing his experiment
,Otto
,in 1878
,produced and
placed on the market the first four- cycle engine Operatin gon the Beau de Rochas cycle but commonly known a s theOtto cycle . This engine was almost immediately adoptedas the standard type of in ternal-combustion motor
,the
perfection of which ha s been the problem of designers .I n 1879 a modificati on of this en gine wa s produced by
Dugald Clerk and formed the basis for present-day twocycle engin e practice . I n the Clerk engine the charge wascompressed and exploded once every revolution , as againstone explosion every two revolutions in the engines of theOtto type .
Since the year 1880 several motors of greater or less valuehave been placed on the market , but without exception theyhave disappeared
,and at the presen t time the engines of
the four-cycle and two- cycle types, with greater or lessmodifications
,hold the field.
CHAPTER I .
THE BEAU DE ROCHAS CYCLE.
REFERENCE to Fig. 1 will explain fully the Beau de Rochasor Otto cycle , the four strokes of which a re designated a s
follows :1 . Suction ; 2 . Compression ; 3 . Explosion ; 4. Exhaust .The four strokes as above enumerated form the basis for
the term “ four -cycle a s applied to gas-engine practice .
Fig . 1 . The Four-Cyc le Principle .
I n figure,A represents the cyl inder , B the piston
,C
the inlet valve,a nd D the exhaust valve . An ideal i ndicator
card has been placed directly above that part of the cylindercomprising the piston displacemen t , in order to facilitatethe explan ation Of the cycle of Operations. The proj ectedpath of the crank pin ha s been divided in to ten equal spaces
,
as has also the path of the piston . From this latter set ofpoints ordinates have been erected a nd the indicator carddrawn . On the indicator card referen ce points have been
TH E BEAU DE ROCH AS CYCLE
indicated by small letters ; corresponding points on thecrank circle are indicated by the same letters with the
subscript c . Outside the crank circle two larger circleshave been drawn on which the cycle of operations is shown .
The inner circle represents the suction and compressionstrokes and the outer one the explosion and exhaust .On the suction stroke the port
, C,opens as the piston
travels from a’ to b’
,a s shown on the indi cator diagram
from a to b. Unless valves C and D are suffi ciently large,
point,a, on the suction l ine would tend to be slightly above
atmosphere a nd point,b,would tend to be sl ightly below ,
due to the resistance of the ports to satisfactorily supply thecharge and remove the products of combustion . Especia llv
would this be true on high-speed engines. These results a regenerally present
,to a greater or less extent
,in most engines
placed on the market . I t is obvious that in the diff erentdesigns Of engines the proportion of the ports and valves willvary ,
and for this reason these defects,in the card
,are quite
pronounced in some,while in others the suction a nd exhaust
lines will be found to follow very closely the atmosphericline . The carburettor may also be a factor in determ iningthe suction line of the card , as it is apparent that one witha capacity too small for the cylinder displacement cannotsupply a full charge under all conditions .I t would then appear that a suction line whose in itial
point,a ,was above and whose terminal point, 6, was below
the atmosphere,would indicate that either the engine valves
were too small to do their work in a satisfactory manner orthat the carburettor was not sufficiently large .
On the compression stroke the piston moves from b’ back
toward a’
,as shown on the indicator diagram , from b back
to an indefinite point z'
on the compression curve , be, or to
to on the crank circle , at which point the compressed chargeis ignited
,the pressure rising rapid ly to the maximum ,
which should be just as the engine passes dead center at a ’ .
As the combustion of the gases is more or less slow , the ignition point is
,of necessity , given a positive lead in order to
obta in this result . As the veloci ty Of the piston increases
8 I N TERNAL COM BUSTI ON ENGI NES
it is Obvious that the igniter lead must be in creased in orderto give the charge suffi cient time to burn and reach itsmaximum pressure at dead center . I f a card shows a shapea s at d, Fig. 2
,it would indicate that the ignition was not
suffic1en t ly advan ced , in con sequence of which the maximumpressure is not reached un til the en gine has passed its dea dcen ter . The expan sion line bein g nearly an adiabatic, a lossof energy necessarily resul ts. The pressure at release iscorrespondin gly higher, as shown by comparison of thedotted expansion curve of Fig. 2 with that shown in fullline , mean ordinate 1s decreased
,owing to fact
Fig . 2 . Card showing La te I gn it ion .
that the maximum pressure is not developed until after thepiston is in the working stroke
,and a s the ordinates
,
dim in ished in length by this late ign ition,appear in that
portion Of the stroke where the pressure is highest,they
a ffect the value of the mean ordin ate most .On the expansion stroke the piston again moves from
a.’ toward b’ , or, as shown on the indicator diagram
,from
d to e,a nd on the crank circle the pin moves from 0 to co,
at which point the exhaust port Open s,allowin g the exp ansion
curve, which up to that poin t , accordin g to Thurston ,is
n early a n adiabatic , to drop Off quite sudden ly nearly toatmosphere . I n effect , then , the exhaust takes place froe to f . The l ocation of the release
,e,varies con siderably in
differen t makes of engines, the theoretically correct point
10 I NTERNAL COM BUSTI ON ENGI NES
indicated by the card is too high , the expansion line willsume a ragged outline, due to the harmonic motion Of thespring in overcoming the eff ect of the inertia. This raggedappearance may continue until well into the next compressionstroke.
CHAPTER I I .
TH E CLERK CYCLE .
TH E Clerk cycle engine , more common ly known a s thetwo-cycle engine
,as prev iously mentioned , wa s first intro
duced by Dugald Clerk about the year 1879 , a nd wa s the firstof the compression motors buil t
,receiving an impulse every
revolution . I n the Clerk motor its inventor in troducedthe charge into the cylinder under compression
,as is done
in the present types Of two—cycle engines . I nstead of obtaining his primary compression in the crank case
,as is the
modern practice,he used a n aux il iary pump . The exhaust
ports were arranged in the cyl inder wall , bein g uncoveredby the piston on its downward stroke, the inrush of thecompressed charge expelling the products of combustion inpracti cally the same manner as this is accomplished in thelater two-cycle engines . This type of motor was composedof two cylinders
,one the power cylinder and the other the
primary compression cylinder . The primary cylinder commun ica ted with the explosion cyl inder at the top of thecompression space
,and herein differed from the present type
,
in which the charge is introduced through a port very nearlyin line with the exhaust port ; it being uncovered , with theexhaust port
,by the downward stroke of the piston .
The theory of the Clerk engine is the same a s that of theDay two-cycle (i .e . introduction of the charge under compression), but the added number of parts with the consequen taddition of weight and complexity is a feature which madethis type of engine impracticable for general use
,more
especially where a light motor is required ; and as nothing isgained
,in its construction , over the Day type , the balance is
all against it . Nevertheless the honor of first describingthe two-cycle principle must be accorded to Clerk and thepractical improvemen ts a nd application to later inventors .
1 1
I 2 I N TERNAL COM BUSTION ENGI NES
The Robson Engine.
A forerunner of the Day type of motor was the Robsonengine
,manufactured by Messrs. Tangye under Robson
’spatent . I n this engine the cylinder wa s closed at both endsand used a piston rod . The forward end Of the cylinderwa s used for obtaining the primary compression
,the charge
being drawn in on the suction stroke and compressed duringthe greater part of the explosion stroke
,a nd the gas thus
compressed was forced into the power cylinder through anautomati c l ift valve
,which operated when the piston was
fully out and the exhaust valve wide open . This chargewa s then compressed by the return stroke of the piston andexploded a s in the ordinary two- cycle motors. Two valveswere necessary in this engine
,a n automatic valve for
admittin g the charge to the primary compression chambera nd one open ing from this space into the power cyl inder .
The en gine wa s of rather neat design and not nearly a s
cumbersome as the Clerk production .
The Stockport Engine .
This engine wa s exactly sim ilar in principle to the Robsonengine , but the forward end o f the cylinder was not ut ilizedto obtain the primary compression . The engine wa s
similar in design to the two—cylinder opposed motors of thepresen t day
,one cy linder being used for the primary com
pression , while the other w a s used for the power cylinder .Several othe r types of engines were evolved and placed on
the market about this time,but the most of them were
either too cumbersome or too complicated to meet with general use , a nd the Day cycle proper has almost, if not quite ,displaced these earlier types .
The Day Engine .
As the Day engine is the analogue , in two-cycle construetion , of the Otto engine in four—cycle design , a completedescription of it will be given . I n the description of theDay engine and cycle we are describing
,in prin c iple
,the
original Clerk idea .
TH E CLERK CYCLE
Referring to Fig. 4, A represents the cylinder, B thepiston,
C the inlet valve, a nd D the connection between the crankcase, in this instance the primary compression space
,and
the power cylinder . An idea l indicator card has been placeddirectly above that part of the cylinder comprising thepiston displacement, in order to facilitate the explanationof the cycle of operations . The proj ected path of the crank
Fig. 4 . The Two-Cycle Prin ciple.
pin has been divided in to equal spaces, a s has also the path
of the piston . From the latter set of poin ts fordin a teshave been erected and the indicator card drawn . On theindicator card reference points are indicated by smallletters ; corresponding points on the cran k circle are referredto by the same letter with the subscript c . Outside thecrank circle another larger circle is drawn
,on which the
cycle of operations in the cylinder is shown ; on a still largercircle the cyc le of operations in the crank case is shown .
Directly above the proj ected path of the crank pin thecrank-case indicator card has been constructed , the pointsof referen ce being indicated by the letters a: and y withsuitable subscripts.
I 4 I N TERNAL COM BUSTI ON ENGI NES
On the suction stroke the piston travels from b’ toward a’
,
or , a s shown on the crank- case indicator card , from 3: toward
y. Just as the forward end of the piston passes point P ,
the inlet valve, C. starts to Open . A vacuum ha s been
produced in the crank case up to this point , by the pistontraveling upwards in the cylinder , a nd in consequence whenthis port commences to open a charge of ga s rushes in fromthe carburettor
,F
,a nd continues to flow un til the vacuum
in the cran k case is entirely balan ced or until the piston onthe return stroke completely covers the port
,C. At 50 on
the crank—case indicator card the suction l ine would tend tobe sl ightly above atmosphere ,
due to the fact that theopen ing D between the cran k case a nd the cyl inder , whenun covered by the piston head on it s forward stroke
,will
not,in all probability ,
allow the passage of en ough of thecompressed charge in the cran k case to the cylinder to bringit s pressure down to atmosphere . Then on the back strokeof the piston
,the cran k- case space being en larged
,the
pressure falls,un til
,when the forward end of the piston
uncovers port, C,at y on the crank-case card
,a considerable
vacuum has been produced . The charge rushing in throughport
, C,causes the line of the card to rise a s the pressure
in the case is increased , a nd when ,on the forward stroke ,
the port,C
,is again completely covered , the card should
show a pressure of about atmosphere as at y, . I f port,
C , is too small , this will not be the case , but the point y, wil lstill show a slight vacuum
,which will n ecessarily aff ect
the maximum cran k-case compression at as,when the
piston head un covers port , D . From y2 to as,on the card
t he pressure should rise regularly until at x,the pressure
should be maximum ,which should be from 6 to 10 lb . per
sq. in .
,in no case less than 5 lb . per sq . in . At this poin t
,
port D being opened by the piston,the pressure line com
mences to fall , and continues to do so un til the pressure isequalized on both sides of the piston , or until the pistonon its return stroke again closes valve , D (shown at point ,r. on the indicator card).
Since the displacement of the piston is the same on the
TH E CLERK CYCLE
cylinder end a s on the crank-case end,the charge taken
into the crank case on the suction stroke should exactlyfill the cylinder space
,and the pressure at x
,should be equal
,
in pounds , to the vacuum at y, . But it is practically impossible to so proportion the ports that they will produce theseresults at all speeds
,unless they are made abnormally
large . This is especially true in the three—port type of enginehere described . The charge introduced in to the crankcase on the suction stroke does not vary with the speed atwhich the engine is running , a s the vacuum produced will notvary to any considerable exten t except a s the increasedspeed of the piston gives less time for leakage of air into thecrank case before the in let valve open s
,consequently on
slow speeds the port being Open longer allows a largercharge to rush in . This accoun ts for the inability of sometwo-cycle motors to run at high speeds and deliver power inproportion .
As noted,these conditions are more noticeable in the valve
less engine than in the en gines using a poppet valve for theinlet port
,a s indicated at G, Fig. 4
,in whi ch case inl et
port, C,
is not used . I n this construction it is alwayspossible to get a ful l cran k-case charge with a correspondingly higher pressure at a: but we are still confrontedwith the diffi culty of making port D large enough to give afull charge at all speeds , but , a s pressure a: is greater
,the
flow will be somewhat more rapid . I t is probable,however
,
that this advantage is suffi cient to make the balance favorthe engine with the valve over the more simple three-portengine ; in fact many manufacturers have tried and discarded the three-po rt type . The crank-case card for thetwo-port type is shown in Fig. 4 below the crank case .
H aving followed through the cycle of operations in the
crank case,let us l ook at the operations taking place in the
cylinder . On the cylinder indicator card let a representthe point of opening of the exhaust port , E,
b the openingpoint of inlet port
,D
,c the point of closing for inlet port, D ,
and d the point of closing for exhaust port , E.
On the forward stroke of the piston B ,the charge having
16 I NTERNAL COMBUSTI ON ENGI NES
been exploded,at point a on the card the piston head
commen ces to un cover exhaust port,E. At this point
,then
,
the pressure of the expansion line falls off rapidly,forming
some such a break in the card as is indicated . At poin t,
B, on the indicator card the port from the crank case intothe cyl inder commences to open ; at this point , b, it is Obviousthat the pressure of the exhaust gases in the cylindershould have fallen to such an extent a s to make the cylinderpressure less than that of the crank case
,otherwise the
cylinder will exhaust back into the cran k case and causewha t is known a s back firing
,with con sequen t loss of power .
For this very reason the exhaust port is given a lead overthe inlet port
,varying to some extent in different two-cycle
design s (see Chapter XVI ). From b to c on the indicatorcard the crank—case charge rushes through pOrt , D ,
to fil lthe cylinder space . A baffle plate
,G
,is placed on the piston
head to deflect the in coming charge to the top of the cyl inder,so that it may more effectively force out the burned gaseswithout bein g exhausted itself . At poin t
,d,the exhaust
port closes a nd the cyl inder compression commen ces . The
charge is compressed from point, d, up to the point of
ignition,which of course is varied for different speeds by
the spark- timin g device . From 3 to f the explosion takesplace a nd the expansion follows from f ba ck to a .
I n two- cycle work the following points should be strivenfor : 1 . Quick exhaust , with consequen t large exhaust port .2 . Unrestri cted exhaust port . 3 . Crank- case compressionhigh enough to make the primary pressure greater than thecylinder pressure when the inlet port Opens. 4 . As largein let ports a s possible . 5 . As long a compression andexpan sion line as possible . A l ittle study of the card wil lserve to show that some of these points must necessarily besacrificed
,to a greater or less exten t , in order to attain the
others . The expan sion line must be shortened in order togive suffi cien t exhaust period ,
a nd a shortening of the expa nsion line produces a corresponding shortening of the compression l ine . The most satisfactory arrangement of theseseveral poin ts will be taken up later in the design .
18 I NTERNAL COM BUST ION ENGI NES
mean effective pressure may be higher , while with a lightload the temperature and pressure will fall
,due to the
quan tity of fuel being less .
After the period of fuel inj ection,which comprises about
10 per cent of the working stroke,is completed, the fuel
valve closes and the ignited charge expands until 90 per cen tof the working stroke has been completed , at which pointthe exhaust valve opens in order to relieve the pressurebefore the expulsion stroke commences . The manufacturersclaim their pressure at exhaust to be about 35 lb . per sq . in .
for n ormal load,which pressure would necessarily be in
creased or diminished a s the engine was operated at overload or running light .The fourt h stroke in the cycle is the expulsion stroke
durin g which the piston,travelin g upwards with the exhaust
valve open,ej ects the burned charge .
I n the Diesel cycle there is no opportunity ‘whatsoeverfor premature explosion
,sin ce the fuel is not inj ected until
the beginning of the working stroke . The high compressionand correspondingly small compression space , about 7 percent of the cylinder volume
,make it possible to ej ect
nearly a ll of the burned g a ses and to secure a charge ofalmost pure air to support the combustion during theworking stroke . The fuel economy , with the theoreticalconditions attained
,would necessarily be high
,and a ctua l
results seem to prove this to be the case . There is somequestion , however , as to the wear and tear resulting from theheavy parts made necessary by the long-sustained highcompression . This high compression causes the temperaturein the cylinder to approximate for a much longer periodthe temperature of combustion
,but as this temperature of
combustion is much lower than in most internal-combust ionengines, it is probably true that the parts subj ected to thisheat are not damaged - to any appreciable extent , and it isdoubtful if the long-sustain ed compression would be moreharmful than the suddenly applied pressure induced inengines operating on the Otto prin ciple .
Fig. 5 will explain fully the cycle of operations taking
TH E DI ESEL MOTOR
place in the Diesel engine . The same arrangement ofdiagram has been followed in this as in the previouslydescribed cycles .
xa represents the clearance .
a b represen ts the stroke .
The stroke is divided in to ten parts by the ordinates 0— 10,
as shown . I n the engine diagram 1)f
og ,and T 3 are, respec
t ively the inlet, exhaust and fuel valves . On the first
Fig. 5 . The Diesel Pri nciple.
ward stroke the piston,from a to b, occurs the suction of
pure air, valve , bein g open . On the backward strokeof the piston the air is compressed
,following the curve
be on the card,all the valves being closed . On t he second
forward stroke of the piston,from cd on the card
,the fuel
valve,v3 opening, the fuel is inj ected . The curve ed
,as
previously described,is an approx imate isothermal ; at d a
break occurs,as the fuel valve is closed ; and from d to e, or
durin g about 80 per cent of the working stroke,expansion
takes place . At e, the exhaust valve , v2 , being opened , thepressure falls and reaches atmosphere at b; and from b to (1during the expulsion stroke
,on the second backward stroke
of the piston,the products of combustion are discharged .
CHAPTER I V
COMPARI SON OF TH E CYCLES.
TH E four- cycle engine has found most favor with thegeneral public
, a nd in consequence has been most widelymanufactured . I n stationary engine practice this design isin almost un iversal use
,although there a re some two- cycle
engines in use for this class of work, and it seems that theyare gain in g some headway .
There are several good reasons for the two—cycle engine notgaining general popularity as readily as the four— cycle .
For stationary en gines,the question of lightness of parts
,
or of the complete engine,plays no important part
,the
design tending more strongly toward stability and weight,within reasonable bounds . As the weight is no inducemen t,the en gine which can be most readily controlled , which ismost certain in operation
,and most economical in fuel
consumption,gains precedence . These three important
points a re found in the four—cycle type of engine with allvalves mechan i cally con trolled and all working parts reducedto a scien tific a nd mechanical basis
,when
,with the quality
of fuel known,the cycle of operations
,with the resulting
power del ivered,m ay be depended on a s unvaryin g, pro
v ided, of course , that ign ition is insured by means of a perfect
sparking device .
I n automobile engines the four- cycle type , While notun iversally adopted , is used in the very large maj ority ofcases, but in a somewha t modified form . H ere the diff erentmanufacturers have striven to create a m achine with the
greatest power a nd the least weight ; in other words, theytry to make the weight of the engine per horsepower a s
l ow a s possible . To accomplish this many manufacturersuse the suction inlet valve
,bu t not to as large an extent as20
COM PARI SON OF TH E CYCLES
formerly,the mechanically controlled inlet and exhaust
being looked upon with more favor at the present time,
due to the fact that the suction in let ha s not,in all cases
,
given universal satisfaction , and in fact wa s quite unsa t is
factory in many instances .I n marine-engine practice probably about an equal number
of two a nd four cycle engines are manufactured a nd sold .
The two—cycle type is more satis factory for marine use thanfor any other purpose
,as the necessary amount of cooling
water is more readily available . Sin ce in the two—cycle typeof motor
,an impulse or explosion occurs at every revolu
tion,it natura lly fo llows that the cyl inder heats up more
rapidly than in the four-cycle type , in which the impu lse ,occurrin g on ly once every other revolution , gives the cylindermore chance to cool .While the two-cycle engine grows hotter
,due to the more
frequen t explosions,it should produce more work and
steadier power for the same reason . This will be found tobe the case
,i f the ports are properly proportioned ; but no
two-cycle engine ever bu ilt cou ld produce twice as much workfor the same number of revolutions
,at all speeds
,stroke and
bore bein g the same,as a four- cycle engine . The reason for
this is found when one tries to proportion the ports for a
ga s speed of 100 ft . per see. for the in let and 90 ft . per sec .
for the exhaust and finds that the ports must necessarilybe quite large to admit a fu ll charge at 800 rev . per min .
a nd under the most favorable condition s,that is
,with the
cyl inder completely sc avenged of the previous charge andthe carburettor sufficiently large a nd properly adjusted toallow a fu ll charge to enter the crank case on t he suctionstroke . I n the maj ority of two- cycle engines the ports a re
made much smaller than they should be , even for theirnominal speed
,with the result that when the speed is
increased the charge is very greatly diminished . At speedswhere the four-cycle en gine would obtain a fu ll charge ofgas it is probable that a two-cycle motor does not obtainmuch more than half a charge , and as the speed is furtherincreased , even less than that amount . I t is doubtful
22 I N TERNAL COM BUSTI ON ENGI NES
if the average two-cycle motor does more than 30 per centmore work than a four—cycle motor of the same size , bothoperating under most favorable conditions .
The two—cycle motor is more wastefu l of fue l than the fourcycle type , and trouble with crank-case explosions is frequent ,due to the explosion following back into the crank case whenthe crank—case compression is lower than the cylinder pressure a s the in let port opens. This may be obvi ated , in alarge measure
,by placing a screen baffle plate in the in let
passage ; the screen ing acts on the prin ciple of a Davyminer ’s lamp . The baffle plate must be a close fit , however,to be effective
,and must be in clined in the passage
,so that
the meshes,through which the ga s passes, may be
‘equal inarea to that of the port i tself .The Diesel motor is gain ing some favor a s a prime mover
in power plan ts . I ts performances,a s far as known
,are
excellent,its manufacturers claim in g the exceptionally high
effi cien cy of 38 per cen t . The engines are accompan ied bya guarantee
,a s to fuel con sumption per horsepower-hour
,
good for one year from the da te of installation . The questionof first cost and the n ecessarily heavy parts possibly inducingan extraordinary amount of wear
,are
,as a general rule
,the
most serious obst a cles to prospective purchasers . H owever,
reports from plan ts in actual operation tend to show thatthe manufacturer’s guarantee is n on e too high . I n fact
,some
operators claim t heir fuel consumption to be below the manufa cturer
’
s guarantee .
The municipal lighting and waterworks plant of Bryan ,Ohio , report that their motors have given not the slightesttrouble from regulation or wear
,and tha t they have made
runs of a month at a time without stopping .
*
* The a uthor is indebted to M r . S . L . Folk of B rya n , Ohio ,. for the
in forma tion in rega rd to the pra ctica l operation of the Diesel motor.
CHAPTER V .
PRACTI CAL OPERATI ON.
S ta rting a Sta tionary Engine— Starting a gas engine is in
most cases a simple operation,if a few rules are remembered .
A gas engine will not start without it obtains enoughinitial power from some outside source to enable i t to commen ce its cycle of operations .
The mixture of fuel and air must be neither too rich nortoo poor in fuel
,for if either of these conditions obtain
,an
explosive mixture will not result .
I n starting a stationary engine by hand , or in fact a ny
other way,the ignition should be given a negative lead
,or
,
in other words,the sparking point should be past the upper
dead center in the direction in which the engine is runn in g .
Failure to note this importan t point will result in backfiring
,with more or less disastrous results .
Compression should be relieved and the load thrown off ,
un less a powerful sta rtin g device is used , as , for instan ce , thecompressed-air system .
Always be sure that the en gine is well oiled , and the oilcups are fil led .
See that the ignition apparatus is in good order and thesparking points clean , if electri cal ignition be u sed . I f thefuel used is clean and burn s without producing a largeamoun t
'
of soot and crust , the sparking points will remainclean much longer than if a dirty fuel is used . Use a cyl inderoil with a high flashing point , in order to obtain the bestresults, as an oil which flashes
”
at a low temperature willassist
,very materially
,in foul ing a cylinder. A foul spark
ing device cannot be made to yield good results .Be sure that all wiring connections are close and clean and
that the batteries and coils a re in good working order.23
24 I N TERNAL COM BUSTI ON ENGI NES
I f a hot tube ign iter is used , bring the tube to a cherryred heat and adj ust the flame to maintain this temperature .
I f the en gine has a starting cam (a double cam acting onthe exhaust valve and serving the same purpose as a reliefcock), it should be thrown in to startin g position . I f the
engine is provided with a relief cock, instead of the cam ,
open the cock .
Set the ign iter to the proper starting position a nd openthe valve in the ga s supply pipe about one—quarter full
,or
possibly a little more . Now give the en gin e an impulse,a nd
,a s soon as it begins its cycle of operations, commen ce
slowly to open the ga s valve and con tinue until the engineis getting i ts maximum supply and is running at its regularspeed . Do not open the valve too rapidly ,
or the enginewill get too much ga s, and in consequence too rich a mixture ;i t will soon slow down a nd stop . While the valve is beingopened
,the relief valve mechanism may be thrown out of
gear , or the relief cock closed , and the spark or ignitiondevice advanced to running position
,unless a governor
actin g on the ign iter mechan ism is used .
When the engine is well started , the load may be thrownon a nd the water turned into the water j acket until thedischarge water is at a temperature of from 160 deg. fahr .to 180 deg . fahr . , for stationary engines of low compression .
For high- compression en gines,a somewhat lower temperature
is n ecessary .
I f the en gine is provided with a starting device,allow it to
make several revolutions before openin g the gas valve .
(See l ater chapter on
As soon as the engine is running well,inspect all oil cups
a nd make sure that they a re feeding properly .
S toppi ng— To stop a ga s engine turn off the gas valve, and
if i t is desirable to stop the momentum of the flywheel , afriction brake
,in the shape of a plank
,m ay be made to
press again st its rim ,by placing the stick against the floor
or other available ful crum , and prying against the rim .
Turn off the oil a nd water supply and turn off the flame inthe hot tube igniter , if one be used , or if electri ca l ignition
26 I N TERNAL COM BUSTI ON ENGI NES
Open the oil supply valves .
Now give the engine a few turns with the starting crank,and
,if i t fails to start at once
,try priming the carburettor;
this wi ll usually solve the difficulty ; but if it sti ll fails tostart
,prime the cyl inder, through the priming cup usually
provided,using care not to use too much gasoline ; a few
drops is sufficient . Failure to start at once , after theseattempts
,shows that the cylinder either is getting too
rich ga s, or is flooded , or that some other part is out oforder . (T roubles a nd remedies a re more ful ly discussed laterin this chapter .)There are two ways of starting a two-cycle marine engine .
One way ,the engine is turned over the same a s a four- cycle
until it takes up i ts operation . The other way,a nd the one
most frequen tly employed in small engines,is to work the
flywheel back and forth to get a charge in to the cylinder,then w ith the spark retarded in the direction in whichthe engine is to run (a two-cycle engine is reversible),turn the flywheel sharply ba ck aga inst the compression
,
until it sparks,instead of turn ing it over dead cen ter .
The same di fficulties in starting are found in the two-cycleengine a s in the four-cycle type .
When the engine,either automobile or marine
,is well
started , gradually advance the spa rk , and open the throttle torunning position ; a s this is a va riable quan tity in thesetypes of engines, no fixed rule can be given ,
but do not openout too quickly ; give the en gine time to
“ catch up .
”
Always be sure that the engine circulation is good andthat the oil supply is working properly ; too much oil is bad ,but too little oil is worse .
S toppi ng.
— To stop an automobile engine,throw off the
spark or close the ga s supply,shift the transmission in to
n eutral and apply the tran smission brake . After stoppingthe engine , close the oil supply valves, i f a force lubricatoris not used , and , i f the machine is to be left any length oftime , remove the plug from the coil or throw off the
switch .
I n stopping a marine engine , the propeller acts as a brake ,
PRACTI CAL OPERATION
and it is only necessary to throw off the spark or close the
ga s supply . The same general rules apply to closing the oilvalves and leaving everything about the engine ready tostart again .
Ca re of Engine.
As a matter of fact , a gas-engine plant requires less attention , by far , than a steam plant of the same size . H owever,the ga s—engine owner or operator should not confuse thisstatement, or similar statements, into meaning that a gasengine requires no attention and will “ run itself afterstarting.
A stationary gas engine should have its regular attendant,
who,while he need not give his entire atten tion to the engine
,
should be depended on to see that it is always in good runningcondition . A gas engine should always be as clean and a s
well oiled a s a steam engine,and it should a lways have a
suffi cient supply of j acket water to maintain a uniformtemperature of from 1 70 deg. fahr . to 180 deg. fahr . I t is ofimportance that the temperature of the cylinder be keptuniform
,especially in the case of an engine running electrical
machinery,as variations in temperature may be readily
detected in the operation of the engine . I n order to maintain a uniform temperature
,the pressure of the water at
the j acket must be kept constant . One of the best waysto accomplish this is to depend on the water pressure of auniform head of water instead of direct pressure from thecirculating pump . This may be accomplished by pumpingthe water first to an elevated tank and allowing it to circulatefrom there to the engine and then to the pump , from whichi t is again pumped to the tank. Suitable means for coolingshould be provided
,either in the shape of a cooling tower
or other device by which the temperature of the water maybe lowered a s rapidly as possible . I t is always advisableto use the cooling water over and over , since , after two orthree circulations through the j acket , it will be
“ broken ,”
that is,the lime or other impurity contained will have been
precipitated ; frequent renewal of the j acket water will
28 I N TERNAL COM BUSTI ON ENGI NES
qu ickly cause a crust to form in the j acket or sediment tolodge at some poin t
,at which place a “ hot-spot ” will be
produced .
I f,a s is sometimes the case
,the exhaust valve is provided
with means for circulating the water through it,the water
passages should be drilled out a s often as once a week inorder to insure their remain ing open to circulation .
The exact point of ignition should be known,so that
,in
starting,the engineer may know when to expect the explo
sion ; also the sparking device may,for some reason
,become
out of adj ustment , or i t may be n ecessary to remove it forrepairs . W ith a make-and-break electrical ignition system ,
this poin t may be determ ined as follows : slowly turn theengine over un til the ign iter snaps
,at which poin t the spark
is produced . Now ,without moving a ny part , make corre
sponding points on the flywheel a nd frame or on the pistonand cyl inder, the latter way being most desirable if possible .
I t is obv ious,then
,that the engine m ay ,
at a ny time , beturned to its sparking poin t
,even though the igniter is
removed ; that is, i t may be turned to the exact distance fromdead cen ter where the ign ition occurs ; but here , in the fourcycle engine , a difficulty confronts us : we must be sure wea re in the explosion stroke and not in the suction stroke.
This may be most readily determined by inspecting the
cams . I f both cams are down,then both valves a re closed
,
a nd we a re all right ; but i f the in let cam is just commencing toraise the valve , we are in the suction stroke a nd must turnthe engine over one complete revolution un til the referencema rks again correspond , at which poin t the sparker maybe set to snap .
All first - class en gines , when they leave the shop,should
have their valve a nd spark positions marked ; and thesemarks , together with prin ted instructions , should enablea ny average mechanic to reset the valve or ign itermechan ism .
The care of the ign ition mechan ism is an all-importantpart in the operation of a ga s engine . Electri cal devices
,
if properly cared for , give excellent sa tisfaction , while if
PRACTI CAL OPERAT ION
allowed to become dirty or out of adj ustment they will givevery poor satisfaction . Be sure that the connections areclean and close a nd that the battery
,i f one be used , is not
allowed to run down . I t is good practice to have two sets ofcells connected up with a switch
,by means of which either
set may be thrown into circuit .I f a tube ign iter is used
,the best material that can be
purchased is none too good . Nickel alloy or porcelain withstands the action of the heat a nd gases best
,a nd
,with
ordinary care , a tube of either of these substan ces willlast a comparatively lon g time, while an iron tube needsto be replaced every few days . Several tubes should alwaysbe kept readily available so that
,in case of acciden t to one
,
another may be quickly substituted . The tube should bekept at the very lowest temperature at which the gas willignite and should never be hotter than a cherry-red . I n
practice,some gases will be found to inflame more readily
than others as the quality is richer or poorer .The bearings and runn ing parts of a gas engine should be
well lubri cated with a good grade of machine oil,but the
cylinder should be lubricated with a gas—engine cylinder oilof high flashing point
,or otherwise the carbon ized oil
produced will soon choke the passages, prevent the valvesfrom seating , a nd, becoming incandescen t , cause prematureignition and back firing . I n any event the exhaust passagesshould be cleaned occasionally to prevent a ny possibleaccumulation from reducing their effective area
,thus
producing a back pressure a nd reduction of power . The
valves shou ld be frequently examined and ground in wi thflour
,emery and oil
,if they leak ever so l ittle. The valve
stem springs should be stiff a nd strong ; i f they become
weakened it is not always necessary to replace them,but
they may be removed and stretched to increase theirstrength
. The few engines that use the suction inlet observethe reverse of this rule , and in their case the valve springshould only be strong enough to properly seat the valve
,
so that a small vacuum , in the cylinder , will open it quickly .
The seat for a suction—inlet valve must always be perfect,
30 I NTERNAL COM BUS TION ENGI NES
since the pressure of the spring is not sufficient to exert anyappreciable grinding effect .Never turn the circulating water in to a hot cylinde r too
rapidly,or the sudden cooling of the walls may cause them
to contract,while the piston is still hot and expanded , with
the result that the piston sticks and cuts the inn er surfaceof the cyl inder . A ring cut
,once started in the cylinder,
will grow until the compression of the engine is ruined andits power gone . When a cylinder is cut
“ badly it can berepaired only by reboring and providing a new piston andrings.
The governing device should receive frequen t attentionto prevent its becoming clogged or gummed up with greasea nd losing its sensitiveness. This is especially true in thecase of governors contained in the crank case
,where they
a re in a position to accumulate a great amount of dirty grease .
For this reason it would be much better design to place thegovernor in an apartment by itself
,or even to leave it
exposed where it may be easily attended to . The hit-andmiss governors act on the gas supply by opening andclosing a ga s valve ; a s the engine in creases speed beyond acertain limit , the governor catches and closes the valve , orreleases i t and allows it to close
,a nd i t will remain closed
until the engine slows down enough to allow the valvemechanism to connect aga in . I f the ga s supply valveis not open enough
,the engine will not get a charge and
impulse the first time the governor connects,and the engine
will slow down until the aperture opens wide enough orlong enough to allow a charge to en ter the cylinder . A
hit - and-m iss governor,properly adj usted , and with proper
opening of the ga s valve , should govern the engine verv
closely .
There are a number of the so called hit-and-miss governorsof different design on the market
,all acting on the same
general principle of closing off the gas supply .
With the hit -and-miss governor,the first impulse received
,
after the governor connects,is always stronger than the nor
mal,due to the fact that all the hot exhaust gases have
PRACTI CAL OPERATI ON
been expelled , and the cylinder, in consequence , gets a fullcharge and a cooler one than usual .The quality of the mixture should be watched
,and may
be determined by inspecting the exhaust,which should be
almost colorless , what color there is being imparted by thecylinder oil , which will always burn more or less . The
nature of the combustion may also be determined byopening the rel ief cock
,if one be provided ,
and watchingthe color of the flame
,which
,for perfect combustion
,should
be deep blue , bordering on a violet .
The cylinder and piston and the valve stems should becleaned occasionally with kerosene
, a nd no oil that will gumor carbonize should ever be used on the valve stems .The crank—pin bearings and the main bearin gs should be
inspected from time to time,and adjusted at the first sign
of wear or looseness . The method of impulse,in a gas
engine,will loosen and wear the bearings much more rapidly
than in a steam engine,and once they start to loosen and
the engine commences to pound,the trouble will grow very
rapidly .
Troubles and Remedies .
Trouble in the operation of a ga s engine is due morefrequently to ignorance in handling than from any faultof the engine itself . I gnorance in handlin g may also beunderstood to include careless handling and inattention tosmall details
,which
,if given their proper consideration
,will
assist very materially in the successful operation of anengine .
I n the enumeration of the troubles connected with gas orgasoline engine operation
,the subj ect will be treated a s a
whole,i t being understood that the carburettor difficulties
apply only to that class of engine in which liquid fuel isused in connection with a carburettor . Nearly all othersources of trouble are common to both engines using liquid
and gaseous fuel .
EngineFa i ls toSta rt— I f the engine wil l not start , examine
the gas valve to see if it has been open too long and allowed
32 I N TERNAL COM BUST I ON ENGI NES
too much ga s to leak in to the cylinder , or if it is open toowide
,allowing too ful l a charge to be taken . I f either of
these conditions is found to be true , close the valve entirelya nd turn the en gine over on ce or twice , to clear the cylinder,or un til an explosion occurs ; then open the valve to startingposition and try starting the engine . I f this fails, go over
the ignition system thoroughly , as described under heading“Spark Weak or Wanting,
” or,if the hot tube igniter is
used , see that the tube is hot enough to ignite the charge.
Cyli nder Flooded. Partially close the gasoline supply andturn the engin e over enough times to satisfy yourself thatall surplus gas ha s been worked out of the cylinder .
Ca rburettor out of Adjustmen t — As every carburettor isdifferent
,the engine operator must familiarize himself with
his spec ial one and find in what adjustmen t it produces, on
the average,the best results . I t is useless to attempt any
fine adjustmen t of the carburettor while the engine is notrunn ing , but it may be set to its approximate adjustment ,once that is known .
Spa rk Weak or Wa n ting.
— I f the spark grows weak, thebatteries are probably poor or old . This trouble may beremedied
,to a certain extent
,by adjusting the poin ts of the
coil ; for weak batteries the points shou ld be set much closerthan when the ba tteries a re stron g. The spark m ay betested by removin g the spark plug a nd holding it, by meansof the in sulated wire
,against the cylinder ; then turn the
engine over to see if there is a good fat spark between thepoin ts. I f the spark is weak and uncertain when exposedto the open air
,it wil l be very much weaker when under
cyl inder compression ,with the probability that there will
be no spark at all between the poin ts under these conditions. When the spark plug is taken out
,see that the points
a re set the proper distan ce apart ; the size of this spark gapwill vary
,to some extent
,with the a ge or stren gth of the
batteries,but in . is about r ight . Be sure that the poin ts
are clean and free from soot ; to insure this condition , theyshould be cleaned , from time to t ime , w ith gasoline .
I f an extra set of cells is carried,throw them into the
34 I NTERNAL COM BUSTI ON ENGI NES
the presence of a deposit ; and to improve the circulation , thisdeposit must be removed . Manyengines provide an opening in to the j acket at these points a nd fit the open in g witha pipe plug
,which may be removed , when n ecessary , to
flush out the j acket . I f the engine is so equipped , removethe plug
,a nd
,usin g a bent wire
,break up the deposit and
then,with the water outlet closed
,force water through the
j acket un til it is clear . I f no such plug is provided,and the
engine con tinues to give trouble from overheating , i t may befound necessary to drill a nd tap a hole for a 1 -ih . pipeplug .
I gn i tion Tube Cold. I f the ignition tube is too cold to fireevery charge
,then some unburned gas will be discharged
into the exhaust passages and explode there. I f the tubedoes not fire the charge frequently en ough to keep up thecycle of operations
,the engine will stop .
M ixture tooRi ch.— This condition usually results in explo
sions in the exhaust passages,or in stopping of the engine .
Ba ck Firing.
— When this condition exists,the charge fires
back,in the compression stroke
,again st the direction in
which the engine is running . Back firing may be due toa ny one of a number of causes ; the compression may be toohigh
,but this should not result in back firing
,except at low
speeds,as on high Speed the charge should be ignited con
siderably ahead of dead center in order to allow the gas toexpand to its max imum pressure by the time dead centeris reached ; it is doubtful i f, unless the pressure were abnorm ally high
,the charge would ignite before this critical point
was reached . There would be more likelihood of the condition being encountered in gases of low ignition temperature
,
a s gasol ine vapor . Back firing may also be caused by thecyl inder becoming overheated , or by proj ections , or fins , onthe inside of the cyl inder becoming in candescent and holdingtheir heat
,derived from one explosion
,long enough to ignite
the next partially compressed charge . A parti cle of carbon ized oil may become incandescent with the same result .The spark may be too far advanced for the speed at whi chthe engine is running .
PRACTI CAL OPERATION
I n the two-cycle type of motor,premature explosions may
occur in the crank case, due to its compression being poor,as ha s been previously explained .
Premature explosions are accompanied by pounding ofthe engine bearings , although a bearing pounding does notnecessarily indicate that premature explosions are takingplace
,as a bolt or nut may be loose and produce the same
result .Wa ter in the Cyli nder .
— This may result from water beingintroduced in the mixture
,or, as is sometimes the case when
the engine is made with a detachable cylinder head,from
the gasket blowing from the cylinder into the water space .
The condition is accompanied by loss of power , or, as isusually the case
,by stopping of the engine . The igniter
mechanism,if electrical , becomes grounded . The only
remedy is to repack the head , an operation often done thewrong way by those inexperienced . The packing, whichshould be a good grade of wire-woven asbestos, should becarefully cut and fitted to the cylinder head , being sure toprovide the Openings for the water spaces and any othersthat may occur. Carefully cut all bolt holes , making themlarge enough to permit the bolt to pass through freely
,
without drawing the gasket out of place when they arescrewed up . Cut all openings as nearly to the exact size aspossible . Now place the gasket carefully in place on thehead or cylinder
,as is most convenient ; place the head in
position and insert the screws or bolts . With the engine
cold, draw up the bolts as tight as they will go , of courseusing a reasonable amount of j udgment and not twistingthe heads off . Now,
with the j acket dry , run the enginefor 3 or 4 min .
,or until it is good and warm ; this will soften
the rubber,or other cementing material in the gasket, and
allow the bolts or screws to be tightened up to their fina lposition
. Failure to perform this last tightening operation wil l mean that the gasket will blow again, as, when theengine is hot, it sometimes takes three
-quarters of a turn onthe screw
,to ta ke up the gasket and squeeze the cement into
all the cra cks so as to produce an absolutely tight j oint .
36 I NTERNAL COM BUSTI ON ENGI NES
EngineSmakes— Smoke, issuin g from the exhaust, indicates
too rich a mixture or too much oil . Smoke , issuing fromthe front of the cylinder
,indicates that the piston is leaking,
due to the rings being worn or the cyl inder out of round , orthe engine may be running hot . The remedy for these condit ions has already been mentioned .
Va lves Leak.— See if the stem s are sticking, or if the seat
is crusted or cut,or if a sprin g is weak . The remedy ha s
been previously suggested .
Engine Ra ces. I f the en gine,running light
,races
,or runs
faster than it can be supplied with gas,it is an indication
that the spark is too far advanced for the amount of mixturebeing fed to the cylinder. The remedy is to retard the sparkor give the engine more mixture .
CHAPTER V I
STARTI NG DEV I CES .
TH ERE are a number of difieren t methods in use forstartin g gas engines, all of which are used
,more or less
extensively,a s requirements demand .
They may be enumerated about as follows : 1 . H andstarting , which is
‘
used most extensively in the startingof engines of moderate Size, and requires that the enginebe provided with compression relief cocks or startingcams . I n using this method of starting
,care must be taken
that the ignition is so set that the charge wil l not be prematurely exploded
,causing back firing
,with accompanying
dan gerous results to the operator . Engines which are to bestarted by this method are generally provided with anautomati c throw-out collar which enables the operator toclutch the shaft with the starting crank , but which , whenthe engine starts
,automatically throws the cran k out of
connection . Several devices,of greater or less effi cien cy ,
have been placed on the market, the obj ect of which is tocause the startin g cran k to disengage as the engine startsor back fires
,thus insuring immunity to the operator .
2 . I t is sometimes possible in mul ti-cylinder engines , andeven a t times in engines of but one cylinder , to start , aftera moderately short stop , by retarding the sparking apparatusand igniting a cylinder containing part of a charge drawn inbefore the engine was stopped . To do this successfullydemands that the engine be stopped with the spark andthat the piston rings be a tight fit , insuring a tight cyl inder .
3 . The engin e may be turned over un til it takes up itscycle of operations by some external source of energy .
Electric motors are often used to advantage for this purpose ,or the large engine may be provided with a starting engine
37
38 I N TERNAL COMBUSTION ENGI NES
small enough to be turned over by hand and of suffi cientpower
,when running
,to start the large engine .
4 . An explosive mixture of gas and air may be stored inan aux iliary air-tight chamber . This may be accomplishedby the en gine itself charging this receptacle before it isstopped . On starting the en gine
,it is turn ed over dead center
into the explosion stroke,and a charge of the explosive mix
ture is admitted to the cylinder by open in g a valve in thesupply pipe . The explosion of this charge will generally besu ffi cient to give the engine enough impulse to make it takeup its cycle of operations .
5 . The l ast—named method may be varied by using anair pump , operated by hand , to compress a charge of explosive mixture into the cylinder . I n either of these last two
methods named,the charge may
be exploded by a n electri cal spark,i f the electrical system of ignitionbe used
,or by means of a match
starter ; see Fig. 6 .
6 . The method of in serting anexplosive cartridge in a tube
,
opening in to the cyl inder , andexploding it bymechan i cal means
,
has been used to some extent .7 . A charge may be exploded
in an aux iliary chamber and theFig. 6 . M a tch I gn iter . resulting pressure conveyed to
the engine cyl inder . H utton,
in his treatise on The Gas Engine,illustrates such
a starter,the operation of which is shown in Fig. 7 .
Ga s enters the auxiliary chamber , A,through the supply
pipe,B
,and
,the poppet valve
,C
,on the engin e being
closed,passes out through the cock
,D
,where it is ignited .
As long a s the ga s valve in the supply pipe is kept open ,the pressure in the explosion chamber is maintained sufficien t ly high to prevent the flame at the j et , D ,
from runningback into it ; but as soon as the supply is cut off , the gas inthe chamber is gradually consumed at the j et until the
STARTI NG DEVI CES
mixture is such that the flame runs ba ck and ignites theentire charge, and the resulting pressure is admitted to thecylinder through the poppet valve
, 0 .
8. By far the most widely used method of starting is bymeans of compressed a ir, compressed and stored by the engineitself or by means of a smaller auxiliary air compressor .
Fig. 7 . Auxi li ary Chamber Starter.
I n order to opera te the compressed-air starter,it is neces
sary that the cam movement in a four-cycle engine be soarranged that one or more of the cylinders may be converted
,
for starting purposes, into a compressed-air engine . To
accomplish this it is necessary that the cylinder exhaust onceevery revolution and that the inlet valve remain continually closed while the starting operation is proceeding . The
a ir may then be turned into the cylinder,either by means
of an air cock actuated by hand or by automatic means .Where a single cylinder, in a multiple-cyl inder engine, isthus arranged , it is necessary that the engine be turned overby hand until the piston of the air cylinder is at the beginning of the working stroke, when the starting mechanismmay be thrown into gear and the air admitted .
As mentioned in Chapter X I V, i f the cams a re made to
40 I N TERNAL COM BUSTION ENGI NES
operate a n in termediate lever moun ted on a shaft, whichmechanism
,in turn
,opera tes the valve rod , it is an easy
matter to make the shaft, on which the levers are mounted ,
to shift in such a way as to bring the starting mechanisminto gear .Such an a rrangement is shown in Figs . 8 a nd which
figures show the arrangemen t of the cams for three
Fig. 8. Compressed Air Start ing Cams.
cylinder,four-cycle engine. I n the il lustrations A,
B and Ca re the three exhaust cams
,and D
,E a nd F are the in let
Cams, a ll mounted on the one cam shaft,G. The trans
mission levers,H
,are shown mounted on the shifter shaft , I .
On the transmission levers are the ha rdened steel contacts,
Sect ion X-X
Fig . 9 . Det a i l of Sta rting Cam shown in Fig . 8.
J, on which the valve stems impinge . Cams A and D are
the double starting cams. Cam A is provided with twoeccentric portion s a s shown
,so that the exhaust valve is
made to open once every revolution,while cam D is made
with one—half of it with the outline of the regular in let cam,
while the other half is concentric to the Shaft at all points,
CHAPTER V I I .
CARBURETTORS, VAPORI ZERS AND I NJECTORS.
TH E gas,gasoline
,al cohol
,and oil engines operate on the
same general prin ciple,as far as the generation of power is
concerned , but the methods pursued for obtaining therequisite fuel in gaseous form vary with the several differenttypes . Thus
,in the gas engine
,that is
,the engines which
operate on some form of ga s a s a fuel , no intermediate stepsare necessary for the transformation of the fuel from aliquid to a vaporous or gaseous form
,although
,in the
engines operating on producer ga s, a n apparatus, known a s
a producer,is necessary to distill from the fuel
,as it appears
in a solid state,a ga s, available for use in internal-combustion
engines .The different devi ces used for the production of the com
bust ible gasoline , alcohol , or oil m ixture w il l first be discussed ,
after which the operation of a suction gas producerwill be taken up.
There are three general methods in use for securing anexplosive mixture from liquid fuels :
(1) Carburettin g, (2) Vaporizing , (3) I nj ecting.
The carburettor , in a ny one of i ts many forms,is a device
by which the liquid fuel is transformed in to a vapor bypassing air either over , through ,
or across a portion of thesupply and takin g up particles of the l iquid in a va por form .
To facilitate the operation,when carburetting gasoline
,i t
is much better , al though not absolutelv necessary ,that the
air,as wel l a s the gasoline ,
be wa rm, especially in cold
weather ; and for this rea son we find the engine manufa cturers leading their suction from a hot-box
,
” locatedeither on the exhaust manifold or on the cylinder base
,or
CARBURETTORS, VAPORI ZE'
RS AND I NJECTORS 43
where the wa rm water, from the j a cket, may be made tocircula te around it .When carburetting al cohol it is necessary
,under all
circumstances , that the fuel be warmed ; the reason for thisis fully explained in Chapter I X .
I n carburetting petroleums it is necessary,especially with
the heavier grades, that the air under pressure he forcedthrough the liquid in order that it may break up or pulverizethe fuel and carry a portion of it, in suspension
,to the
engine cylinder . I n many engines,after the oil is thus
broken up, the mixture is carried to a heated chamber orthrough heated coils where it is vaporized and mixed withair to form the proper explosive mixture . I n other types
,
the vaporized fuel is carried direct to the cylinder,a nd the
residual heat of previous explosions produces the samegeneral results, although in a less satisfactory manner . Stillanother method consists in heating the fuel oil , by passingit through coils exposed to the action of the exhaust gases ,and thereby driving off an oily vapor which , due to its heat,has suffi cient pressure to carry it past an air nozzle wheresuffi cient air is mixed with it to produce the proper explosivemixture . The same result is also obtained by causing theoil to fall , a drop at a time, on a hot plate, thus causing it tovaporize .
The vaporizer (and i t may here be said that they are onlyapplicable
,as ordinarily designed
,to the use of gasoline
or naphtha) differs from the carburettor in that the latteralways has a supply of gas on hand
,while the vaporizer
,or
mixing valve,makes only enough gas for each revolution or
charge,as required . Many
,so- called , carburettors are , in
real ity , improved types of mixing valves, and in fact, it maybe said that the maj ority of them are .The vapori zer consists, essentially , of a gasoline valve, of
needle design , capable of being adjusted to deliver thenecessary amount of the fuel to produce the requisite vaporfor the mixture required , and an adj ustable a ir valve , bymeans of which the air supply may be regulated so as tovary the quality of the mixture, as requirements demand .
44 I NTERNAL COM BUSTI ON ENGI NES
The gasoline is dropped in the path of the entering air andcarried along in the form of a finely divided spray
,or it is
made to rise in a nozzle,placed in the path of the entering
air, which carries enough of the fuel with it to produce thenecessary mixture .
Vaporizers a re made supplied with a throttling deviceand in all respects Similar to the maj ority of the
,so-called
,
carburettors on the market with the ex ception that they arenot supplied with an automatic float feed device .
I n j ecting,a s the n ame implies
,consists in injecting into
the cylinder , or a chamber adj acen t thereto , a quan tity of thefuel m ixed with the requisite amount of air . This methodof introdu cing the fuel in to the cylinder is practi ced qu itelargely by the differen t oil-engine manufacturers. The
H orn sby-Akroid a nd the Meitz and Weiss,the in j ection a nd
ign ition of whose charges a re later described under “I gni
tion,
” make use of this principle,as does also the Diesel motor .
The Diesel method is as follows. Referrin g to Fig. 1 0,
the air valve,the exhaust valve
,a nd the fuel valve are
plainly marked . The air valve allows the air charge,a s
previously described , to enter the cyl inder on the suctionstroke . At the beginn ing of the working stroke the fuelvalve is open ed a nd the charge of oil is forced in to the cyl in
der by means of compressed air under a pressure of 800 lb .
per sq . in . The construction of the fuel valve is somewhatunique ; the fuel en ters the valve through the pipe, A,
andthe auxil iary compressed air , through the pipe, B . The
valve,proper
,consists of concentri c washers
, C,dril led with
small holes,a s shown , parallel to the spindle , D ,
which, by
mean s of the govern or acting through the bell cran k,E
,
opens and closes the valve at F. The capillary attractionof the oil
,a s it falls on the washers , causes it to fill the holes
above men tioned,and when
,on the workin g stroke
,the
valve,F
,is opened
,the oil is carried with the air in a finely
divided spray to the cylinder where,as described in the
chapter on I gn ition ,”the heated air contained in the
cyl inder completely vaporizes and ign ites it . The valvestem is made of nickel steel , as it has been found by experi
CARBURETTORS, VAPORI ZERS AND I NJECTORS 45
ence that where the packing abrades the spindle,as it is
moved back a nd forth,it soon becomes worn and requ ires
to be replaced .
This method , as used on the Diesel motors, is very econom
i cal of fuel and could be applied to engines of lower compression with probably as satisfactory results .
Fig. 10 . Diesel V a lves.
Returning to the subject of carburettors and referring toFig. 1 1
,we have a n example in which the carburetted air
is obtained by passing it through the fuel and thence to theengine . This method is sometimes spoken of as mechanicalebullition . I n the illustration in question , A represents thesuction pipe to the engine ; B ,
the screened openings, throughwhich the auxili ary air supply is drawn ; C,
the tube , terminating in the float
,E,throughwhich the carburetted air is drawn ;
46 I N TERNAL COM BUST I ON ENGI NES
D,a shield actin g on the principle of a separator and causing
the surplus particles of gasoline to separate from the vaporwhen they impinge against its surface ; F,
a n indicator andgauge by which the height of gasoline in the carburettor may
be determined . The end of tube ,C
,term in ates in such a position a s
always to be just below the surfaceof the l iquid ; the float , E, causingit to move up or down , as the
elevation of the surface of the liqu idis changed . On the suction strokeof the engin e
,air is drawn through
the tube, C,
and mixes with theauxiliary air supply drawn in at B ,
a nd the mixture thus obtained iscarried to the cyl inder of the engine .
This carburettor wa s used in theearlier Daimler en gin es and wasfi rst devised by Gottlieb Daimler .There were
,however
,two very
marked disadvantages in its use .
I n the vaporization of any liquid acertain amoun t of heat
,known as
the latent heat of vaporization,is
lost ; now unless heat be suppliedto the liquid
,from outside sources, a s this vaporization
continues,the temperature of the liquid con tinues to fall
until it may become so cold that it w il l have lost nearlyall of its volatility . (See Chapter I X for detailed descriptionof this condition in differen t fluids .) This was the greatdifli cul ty encountered in the operation of this type of carburet tor . Added to this
,trouble wa s also experienced from
fractional distill ation ; that is, the lighter portions of theliquid naturally rose to the top and , i n the process of vaporiza t ion
,passed off first , leavin g in the carburettor the heavier
part of the fuel . As a result of this condition the last partof the liquid to be vaporized
,or distilled
,was of poor qua l ity,
naturally aff ecting the operation of the engine.
Fig. 1 1 . Da imlerCarburettor.
CARBURETTORS , VAPORI ZE'
RS AND I NJECTORS 47
Several carburettors of this general type were designedand used to some extent
,but the same diffi culties were
experienced in a ll .
Fig. 1 2 represents one of the earl ier carburettors , in whichcarburetted charge was obtained by passing air across
1 2 . Sur fa ce Carburet tor .
a large surface on which the fuel to be vaporized wasmaintained .
I n the illustration , A is the in let valve, through which theair to be carburetted is drawn ; B is the suction pipe toengine ; C is the auxiliary air supply valve
,by means of
which the mixture may be regulated ; E is a light metalcasing
,containing the spiral E
,which is fastened . to the
top of the casing and allowed to proj ect nearly to the bottom ,
as Shown . On either side of the spiral partition is bastedlight flannel or felt .When the carburettor is partially filled with fuel , as shown
,
i t is a pparent that the air, entering at A, must pass com~
pletely through the spiral to reach the outlet at B . The
48 I NTERNAL COMBUSTI ON ENGI NES
flannel or fel t,reachin g down into the liquid
,serves in the
capacity of a wick , and , by capillary attraction , the partabove the liquid is kept saturated w ith the fuel . The
passage of the air currents across this large expanse of wicksurface evaporates the fuel and saturates the air .Experience ha s shown that
,in carburettors of this type,
the best results are obtained when the carburettor , i f about8 in . deep
,is half ful l of the liquid to be evaporated .
The principal obj ection to the wick carburettor is that thewick gradually becomes clogged with foreign matter
,taken
in with the air supply or con tained in the fuel , a nd ceases toperform its functions properly .
To overcome the difficulty connected with the use ofwickin g
,carburettors are designed in which the l iqu id fuel
made to drop on very fine wire gauze,and
,forming a thin
film over the wires and Open spaces of the fabri c,is easily
evaporated by the en tering air.Of the carburettors , in which the air is carburetted by
passing it over the gasoline,we have many examples ; in
fact,the large maj ority of the modern designs use this
method . These carburettors a re common ly spoken of a s the“spray type ,
” and the name is aptly chosen , the fuel bein gin jected into the en terin g air , through a nozzle , in the formof a finely divided spray or mist .The operation of the carburettor is as follows : a nozzle
,
fitted with a needle valve , is so situated in the air passagethat the enterin g air must pass over it on the way to theengin e . On the suction stroke
,a partial vacuum is formed
in the carburettor , a nd the air,rushing over the sprayin g
nozzle,is attempting to fill this vacuum . I n consequence
its pressure is less than that of the atmosphere,and the
fuel,supported in the n ozzle
,is thrown off in to it by the
unbalanced pressure .
As has been previously noted , these carburettors have acapacity of but little more than one complete charge for thecylinder
,and are really improved types of mixing valves ,
their one distinctive feature bein g the float feed , with whichthey are almost universally equipped . I f not equipped with
I N TERNAL COM BUSTI ON ENGI NES
in the valve seat at E. The gasoline supply is regula ted bymeans of the needle valve
,the handle of which is shown at
1 313 . M ix ing V a lve.
F, and the amount of opening of the needl e valve is indicatedby the pointer
, G,and the dial
,H . The gasoline connection
is shown at I and the engine connection at J.
Pla in Pattern Generator Va lve.
Figs. 14,15 and 16 represent three diff erent types of
mixing valves manufactured by the Lunkenheimer Company,
of Cincinn ati , Ohio . Fig. 14 is the plain pattern generator
CARBURETTORS , VAPORI ZERS AND I NJECTORS 51
valve , Fig. 15 shows a valve with sc rews for varying theOpening of the poppet and the tension of the spring
,and
Genera tor Va lve wi th Adjust ab le Air Poppet.
16 . Genera tor Va lve with Throt t le.
Fig. 16 is a valve having throttle connections, mean s forvarying the air and gasoline supply a nd for changing thetension Of the valve spring.
52 I N TERNAL COM BUSTI ON ENGI NES
These valves,especially the one with throttle connection ,
give excellent satisfaction,and the operator is able to con
trol the motor very closely .
With nearly all types of mixing valves the best resultsare obtained when they are operated on ordinary stovegasoline . This is due to the fact that the lack of the floatfeed allows the fuel to run into the valve more freely than isnecessary ,
and,in consequence
,the mixture is l iable to
become too rich with the more highly volatile gasoline .
The Schebler Carburettor .
Fig. 17 is a cut of the Schebler carburettor, manufacturedby Wheeler Schebler , I ndianapolis, I ndiana . I t is of thespray-floa t feed type , and the working parts are shownqu ite clearly in the cut . Reference figures and letters areas follows : 9 i s the constan t a ir opening
,through which
the air,to be carburetted
,passes . The a ir
,entering at 9
,
passes upwards past the spraying n ozzle P,to the mixing
chamber of the carburettor,where the auxiliary air supply
,
entering through the poppet valve , A,is mixed with it
,and
the explosive mixture, thus formed , is drawn into the
CARBURETTORS , VAPORI ZERS AND I NJECTORS 53
engine cylinder . The'
floa t feed mechanism is shown atF
, J ,H
, and consists of a float, F,surrounding the constant
air supply tube, as shown , which operates the needle valve ,H
, through the lever, J , maintaining the level of the fluidsuch that it will just overflow the nozzle
,P . Gasoline enters
the carburettor , from the supply tan k,through the supply
pipe , G.
The nozzle , P,is fitted with a needle valve
, E,which is
adjusted permanen tly for low throttle by means of theknurled button
,I then for open throttle
,the needle valve
mechanism is raised or lowered,by the Operator
,by means
Of the lever, P,which actuates the cam mechanism
,Y
,
and causes the lever, Q , to revolve about the point, T ,
thusraising or lowering the needle poin t . The auxiliary airsupply is provided with two valves
,a damper valve Z 1
,
by means of which the Opening may be increased or diminished ; and the poppet valve A,
held to its seat by thetension of spring 0 , which tension may be increased ordiminished by means of the knurled screw M . Push pin Uis used for priming the ca rburettor when starting the engine .
Pushing the pin down lowers the float and opens the needlevalve
,H
,causing the nozzle
,P
,to overflow and producing
,
momentarily,a very rich mixture
,suitable for starting .
The float chamber is usually provided with a pet-cock,at its
lowest point, for draining off the poorer grade of gasoline,
which usually accumulates there . I f the gasoline containswater it also accumulates in the bottom of the float chamber,and may be removed from time to time .
Fig. 18 Shows two sect ion a l‘
views of the H olley carburettor
,manufactured by the H olley Brothers Company
,Detroit
,
Mi chigan . I t is also of the spray type , and i ts operation is asfollows : the incoming air enters through port A, which is provided with a fiber valve BC ; situated around the valve are theconstant openings (1 , through which air is constantly passed .
The valve proper is held to its seat by means of the spring ,b,as shown
,the tension of the spring being capable of
adj ustment by means of the adj usting screw 0 . The
spraying nozzle,D
,situated in the path of the incoming air
,
54 I N TERNAL COM BUSTI ON ENGI NES
may be Opened or closed by means of the needle valve, d.
The mixture passes to the en gine past the butterfly throttle ,E
,which may be opened or closed
,to suit requiremen ts,
by means of the lever,F. Gasoline enters the carburettor
through the gasoline connection,H
,which is opened or
cl osed by means of the needle valve,1,being raised or
Fig. 18 . Old Type H ol ley Carburet tor .
lowered by the float , G,acting through the lever
,J . The
mechan ism is set to main tain the height Of the l iquid suchthat it will just overflow the spraying nozzle . On low
speeds the air enters entirely through the con stan t airopen ings
,a,but
,as the speed in creases
,these openings not
being large enough to supply the requisite amoun t of airto overcome the partial vacuum formed
,the auxiliary
valve BC raises and allows a n extra supply to en ter thecarburettor . A pet-cock at K is used to drain the carburettor
,a s already described .
Fig. 1 9 is a later design of the H olley carburettor in whichan entirely new principle has been made use of to produce thevarying m ixtures necessary for the changing speeds . Moreor less trouble is encountered with ca rburettors
,in which this
variation is accomplished by mean s of an aux iliary airsupply
,owing to derangements of the Spring devi ce
,which
closes the auxiliary valve . Frequent adjustments of this
CARBURETTORS , VAPORI ZERS AND I NJECTORS 55
Fig. 1 9. H olley Ca rburet tor.
56 I NTERNAL COM BUSTI ON ENGI NES
spring are necessary,and even then trouble is often en coun
tered at different speeds .The H olley claims to have overcome this difli culty by
varying the amount of evaporatin g surface of the gasoline,thus changing the quality of the mixture .
The method of operation is a s follows : in the illustration ,the air enters at A and passes downward and then up througha U-shaped tube . At the lowest point of this tube the areais gradually constricted and the gasoline orifi ce
,B
,is located
there , the si ze of this orifice being adjusted,as in most
carburettors , by means of the needle valve, E. The mixturepasses through the butterfly throttle and to the enginethrough the outlet
, C. The float chamber surrounds the U ,
and has an annular cork float,J
,which con trols the needle
valve,L
,through the lever
,N
,pivoted at K . The lower
constricted part Of the U is , in principle , a venturi tube , andmakes i t possible to main tain a very high air velocity overthe gasoline orifice B .
The gasoline level , in the float chamber , is maintained sothat it will overflow B about {I in .
,a nd when the engine is
not operatin g , this condition will maintain . When theengine is started
,the suction does not have to lift the gaso
line but merely evaporates it off the top of the puddle,and
is ca rburetted by surface evaporation . As the enginespeed increases and the throttle is opened
,the increased
velocity of the air sweeps the puddle away,and on high
speeds the mixture is carburetted by the spray from theorifice . Drain pipe D is provided to prevent the puddlefrom growing deeper than 3
1; in . The engine is throttled by
means of the lever,F
,operating the butterfly valve , as shown .
The principle of operation of this carburettor is good , andresults seem to j ustify it .
Al cohol Ca rburettors .
As already mentioned , the only requirement for the carburet t ing of al cohol in an ordinary carburettor is that theair shall be of suffi cient warmth to vaporize the alcohol insufficient quantities to produce a properly saturated mixture .
CHAPTER VI I I .
PRODUCERS.
PRODU CER gas fuel,for gas engines
,may be generated
with apparatus operatin g under pressure,or by suction .
The first producers to be made and marketed were introduced by a Londoner , named Dowson ,
and they were usedto such an extent that the name Dowson gas came to bealmost synonymous with producer gas . They were of thepressure type and required for their operation a hard gradeof anthracite. The produ cer plants were quite compli cated
,
due to the necessi ty of scrubbers, cooling apparatus , and aga someter , in which the gas, Since it wa s under somewhatvarying pressure, had to be stored before being fed to theen gine .
There a re many cases, however , in which it is necessarythat a pressure system be used ; in fact for a ny other purposethan for use in connection w ith a gas engin e, where thesuction of the piston produces the n ecessary flow of airthrough the producer , pressure of air is necessary in order tooperate the apparatus and to convey the gas to the requiredpoin t .Nearly all pressure generators are copies of the original
Dowson idea and include a generator or retort,in which the
ga s is driven off from the fuel ; a n air—compressor fan or otherapparatus for blowing a mixture Of steam and air throughthe generator ; a scrubber, a ga s purifier and a gasometer .Fig. 20 represents, diagrammatically ,
su ch a plan t .The retort or producer consists of a metal shell
,l ined with
fire brick or clay,vertically mounted . A charging hopper
,
so arranged that the producer is never opened to the air,is
provided at the top . The bottom of the producer rests ona grate through which the ashes fall , the air and steam being
58
PRODUCERS
passed through these ashes to the producer, or a water seal
is provided , in which the generator sits, and the mixture ofair and steam is introduced under a conical hood whichprotects the open end Of the pipe from becoming clogged
Fig. 20. Pressure Producer.
with ashes and coal . The mixture should be superheatedi f possible
,in order that no more heat energy than is
necessary be lost in heating the entering air.The entering fuel should be carefully distributed
,a nd
means provided for breaking the clinker formation on thewalls of the retort ; some manufacturers provide tuyereopenings to accomplish this ; the apparatus must be tighta nd
,if not provided with a water seal
,suitable means
must be had for cleaning the grate .Distribution of the fuel charge is accomplished by making
the drop grate Of the hopper conical in form,which spreads
the coal over a large surface . (See Fig.
The air blast may be supplied to the producer in any oneof several different ways . The pressure may be obtaineddirect from a steam boiler maintained
,as nearly as possible
,
at 80 lb . per sq . in . pressure ; a. blower, operating on theprinciple of a draft inducer or inj ector ; by means of amechanical fan or centrifugal blower ; or by the use ofcompressed air.
60 I NTERNAL COM BUSTI ON ENGI NES
When blowi ng with steam,trouble is more than likely to
be encountered from the varyin g steam pressure producing differen t quali ties of gas . Mechanical blowers may bedriven directly from the engine
,i f used in connection with
a producer for a gas engine . The use of compressed airresults in a good even quality ofgas, but the cost of productionis
,necessarily
,high .
The gas leaving the producerpasses to the scrubber
,whereit
is cleaned of any dust which itmay contain by passing throughsprays of water and being fil tered
Fig . 2 1 . Charging H opper through beds of coke,calcium
for Producer. hydrate,moss
,or sawdust
,placed
on removable trays so arrangedthat the filtering or purifying material may be cleansed orrenewed .
From the scrubber a nd purifier the gas passes on to thegasometer
,where i t is stored ready for use . The gasometer
acts in the capacity of a pressure regulator a nd Should havesufli cien t capacity to take care Of any possible stopping ofthe production of gas for a brief time .
I n producers Operating on the pressure system,any
combustible or volatile materi al may be used to produ cethe gaseous fuel , and in nearly all cases the economy overthe combustion of the same material under a steam boileris very marked . This gas may be produced from sawdust
,
sawmill refuse , street sweepings , garbage, lignite , peat , etc .
The methods used for producing the gas vary to someextent — the ga s bein g obtain ed either by distillation orcombustion . Since these wa ste products and the cheapergrades of coal or peat in burning form a rather closelycompacted mass , considerable pressure is required to drivethe air through the producer ; hence , without exception , thesematerials are not available for suction producers .I n the production of gas from wood or wood refuse
,by
the distillation process , the material is placed in a cast-iron
PRODUCERS
crucible which is subj ected to the heat of a furnace and thevolatile part of the fuel distilled off
,leaving charcoal in the
crucible . The wall s of the crucible should be heated to acherry-red heat, between 1600 degrees and 1700 degreesfahr .
, and its diameter'
should not exceed 12 in .
Fig. 22 is a cut Of the Riche distilling producer as illustra ted by Ma thot in his “
Modern Gas Engines and ProducerGa s Plants . The heatedgases from the furnace passthrough the flue opening
,A
,
into the flue space,B
,which
surrounds the retort,as shown ;
the gases then circulate aroundthe retort and pass up andout to the stack through port
,
C. The crucible is chargedwith the fuel to be distilled
,
i n this case pieces of wood,
and the top closed to make anair-tight j oint. From thebottom of the crucible thegas generated is led to thescrubber and purifier andthen to the gasometer . The
heat generated in the cruc ibleraises the pressure of the
prOdUCtS being diStmed and , Fig. 22 . The Riche Dist i ll ingSince there is no outlet at the Pl ‘Oducer
top,the gases must pass from
the cooler part past the hot part of the apparatus on theway to the scrubber. This has a tendency to burn out theimpurities contained .
Producers operating on the distillation principle burnabout 1 lb . of coal to every to 3 lb . of material distilled ,and produce from 28 to 35 cu. ft . of gas having a heatingvalue of about 340 B . t .u . per cu. ft . or 9860 B . t .u . per poundOf coal , whereas 1 lb . of good coal will produce of itself about
B. t .u .
62 I N TERNAL COM BUSTI ON ENGI NES
The gas produced from the wood,however
,is of permanent
composition and can be transported lon g distances. The
residual charcoal,i f wood be used
,also has some value .
The residual weight of charcoal is approximately one—fifththe original weight of the wood depending
,however
,to
quite an extent on the amount of water in the wood,a wood
like elm containing a very large percentage .
Combustion producers produce the ga s by the combustionof the fuel in the presen ce of water . The products ofcombustion a re then passed on to a reducer
,which disso
c iates the hydrogen a nd oxygen con tained in the steam,
reduces the carbonic acid gas to carbon monoxide,and
produces the hydrocarbons. The reducer contain s coke,
which,when incandescent
,produces the n ecessary reaction s .
One pound of wood waste,in a combustion reducer
,
will produce about 1 0 cu . ft . of gas having a heatin g valueof approx imately 1 15 B . t .u . per cu . ft .I nverted producers Operate by forcing the air down from
the top,through the fuel . The distilled volatile products
,
when they reach the incandescent part of the fuel , are reduceda nd a permanent ga s, free from tar, is obtained .
The Suction Producer
As ha s already been men tion ed , the suction producerdraws the air charge through the fuel by means of thesuction of the en gine piston a nd
,in consequen ce of this fact
,
only certain fuels are available for use in these plan ts .
The pressure type , it is readily seen , ha s greater elastici tyin meeting the differen t fuel conditions
,but the suction
plant takes up much less floor space and the cost of installation is much less than for a pressure plant of the same Size .
The suction producer can use only anthracite coal,or
carbonized fuels,as charcoal or coke . The anthracite must
not be too small — not less than pea size — a nd it must beclean a nd carry a s small a percentage of ash as possible
,not
more than 15 per cen t . Undue resistance in the producerwill produce a n over amount of back suction on the enginepiston
,with consequen t loss of power .
PRODUCERS
I f anthracite coal is used,one pound of good quality will
produce about 1 b . hp .-hr .
,and it will require about lb .
of coke to produ ce the same result . For small units theamount of fuel is somewhat increased and may be as highas lb . per b . hp .
-hr .
Fig . 23 . Suct ion Producer.
I n operating the suction producer , a fire is kindled on thegrate and a bed is buil t upon it in the ordinary manner , thenecessary air bein g supplied
,for the time being
,with a
hand or belt—driven fan . Beyond the vaporizer,the prod
ucts of combustion escape through a waste pipe . Whenthe test cock shows that a good quality of ga s is being produced
,the scrubber a nd purifier is thrown into the circuit
,
and when good gas appears at the engine cock,the engine
is started and the fan stopped .
I NTERNAL COMBUSTI ON ENGINES
Fig. 23 illustrates a suction producer manufactured byR. D . Wood Co.
,of Philadelphia . I t is very compact,
an area of 15 by 35 ft . being suffi cient for a plant of severalhundred horsepower
,a smaller plant
,Of course, requiring
less space .
Fig. 24 shows very clearly the comparative effic iencies ofa gas producer and steam plant .*
Fig. 24. Compara t ive Effi cienc ies of Steam and Producer Plants.
The first cost of a producer plant is approximately thesame as for first -class steam engines a nd boilers of the samehorsepower
,but the resultant economies in fuel and atten
tion are very marked,one man being able to care for a large
plant . The cost of attention is from 50 to 75 per cent thatof a steam plant . No time is required , a s with the steam
The author is indebted to R. D. Wood Co . for the cuts on
suction producers , a nd the results shown in thei r diagram a re veryclosely in keeping with the results of modern practice where the steamengine used is of non-expansion , non
-condensing type.
CHAPTER I X .
FUELS AND COMBUSTION .
FOR the motive power in an intern al - combustion enginea ny gaseous fuel is available , as well as any other fuelwhich may be vaporized or transformed in to a gas . By
vaporized fuel we mean any fuel,such a s gasoline
,petroleum ,
oil,or alcohol
,whi ch may be used in the cylinder of a ga s
engin e without the intermediate step of tran sform in g it in toa ga s. There is absolutely no combustible substance whi chmay not be tran sformed in to a gas
,or rather have it s gaseous
products driven off, by the action of heat . Any one Of these
gases may be used,w ith greater or less efficien cy
,a s their
calorific effi ciency is greater or less,in the cylinder of a ga s
engine . I t is furthermore true that in all cases the powerObtained from a ny fuel first converted in to a ga s and thenburned in the cylinder of an internal- combustion engine
,is
always greater than if the same amoun t of fuel were burnedunder a boiler a nd the steam used to drive a steam engine .
This is true because the heat effi cien cy of a ga s engin e isabout 25 per cent , while that of the steam en gin e is from 10
to 12 per cent . As nearly a ll of the combustible part of thefuel becomes ga s especially is this true in the case of vaporized fuels it is Obvious that the fuel so used must be mu chmore economical than when fired under a boiler . The
manufacturers’ guarantee a ccompanying st a t ion a rv gasengines usually insures that their en gine will produce1 hp .
-hr . on from 1 1 to 1 2 cu . ft . of natural gas . The heatun its con tained in natura l ga s range from 900 to 1 100 percu . ft .
,according to the nature of the gas , a nd in producer
gas there are a bout 160 heat units per cu . ft . ‘ Runningon producer gas
,then , the same engine would require
66
FUELS AND COM BUSTI ON
1 100
160
of bituminous coal will produ ce 75 cu . ft . of ga s a nd, inconsequence
,to produce 1 hp.
—hr . would require that 1 lb .
Of coal be used in the producer . The very best steamengines yield but 1 hp .
-hr . on lb . of coal .I f we could use the relative heat effi ciencies of a gas and
a steam en gine as a basis of comparison,we would expect to
find that where 1 lb . of coal , as gas , in a ga s-engine cyl inder ,would produce 1 hp.
-hr .,i t would require 2 lb . of coal ,
or a trifle more,when used under a steam boiler . This
comparison may not be used in the comparison of producergas
,as a certain amount of the avail able heat is used in
the producer in distill ing off the ga s and in vaporizing thewater con tain ed in the fuel . As regards plant constructionand operation
,the gas producer is much cheaper and simpler
an apparatus than the steam boiler a nd requires less attention . A surplus of gas may always be kept on hand , in agasometer or storage tank
,and the en gine started on a
moment’s notice,whil e with a steam boiler time must be
consumed in getting up a head of steam .
Ga s too lean to be used under a boiler is found to igniterapidly when under compression in a gas-engine cylinder .
Blast-furnace gas is an example of the above ,large two
cycle engines operating on this gas being in use at the shopsOf the Lackawanna Steel Company , in Buffalo , New York.
The value of a gas as used in a gas-engine cylinder islargely dependent on the number of British thermal unitsi t contains
,although the richness of the gas must also be
considered . A lean gas may be burned completely in acyl inder with less air , and a consequen t larger amount of gas ,than a gas of high thermal va lue . The following tablegives the heat units per pound and per cubic foot for thedifferent fuels . Natural gas ‘ is seen to have the greatestheat value
,but
,notwithstanding this
,gasoline vapor
,with
its lower heat value , is credited with about 1 1 per cent morepower . This fact is due to the rate of flame propagationbeing more rapid in the gasoline vapor than in the use of
— 75 cu . ft . approximately . Now one pound
138 I N TERNAL COM BUSTION ENGI NES
natural gas , with the result that the combustion assumesmore the aspect of an explosion . A corresponding increasein the mea n effect i ve pressure results .
TABLE I .
The con stituen t parts of natural gases vary in differentlocalities . Table I I gives some of the different volumetricanalyses of Pennsylvania gases .
TABLE I I .
FUELS AND COM BUSTI ON
Many natural gases, especially those from western fields,
have much lower heat values than those here tabulated .
Oi l Ga s. Oil ga s, a s manufa ctured for municipal l ightingplants , is of the following approximate composition :
H ydrogen , H
Ma rsh-ga s , OH 4
Nitrogen , N
H ydroca rbonsCa rbon ic oxide, COOxygen , 0
W ater vapor,H 2O
The heat value of this ga s is slightly more than 600 heatunits per cu . ft .Producer Ga s . The so—called producer gases are of differ
en t volum etric composition s , a s their methods of production vary . The following may be taken a s a verage values :
The presence of the hydrocarbon element to a greaterextent in the bituminous producer gas makes its heat valuecorrespondingly greater .Wa ter Ga s .
— The production of this gas entails the lossof a large amount of energy and for this reason it does notplay a very impor tant part in the gas-engine field , althoughit is a gas very rich in energy itself . I t is made by theaction of a j et Of steam on incandescent fuel , the hydrogenin the steam being dissociated and taken up by the gas .R . D . Wood Co. give the following data on the produc
tion of water gas , the theoretical composition of which is
70 I N TERNAL COM BUSTI ON ENGI NES
equa l volumes of hydrogen and carbon monoxide,although
the gas generally contains some nitrogen and carbon i c acid
500 cu . ft . of H weigh500 cu . ft . of COweigh
Total weight of cu .
Now,as carbon mon oxide is composed of 12 parts
carbon to of Oxygen,the weight of carbon in 36 89 lb .
of ga s is lb . and of oxygen lb . When thi s oxygenis derived from water (steam) it liberates , as above , lb .
of hydrogen . The heat developed and absorbed in thesereaction s (disregarding the en ergy required to elevate thecoal from the temperatur e of the atmosphere to say 1 800degrees) is a s follows :
lb . of H a bsorb ,in dissoc ia tion from water, H eat un its .
X
lb . C bu rned to CO develop X
Excess of heat absorption over heat develop
The loss due to this absorption must be made up in someway or other .
lb . of carbon burned to carbon dioxide would supplythis heat , theoretically , but in practice , owing to the imperfeet and indirect combustion and radiat ion ,
more thandouble t his
,amoun t is requ i red . Besides this i t is not often
that the sum of the carbon monoxide and hydrogen exceeds90 per cent , the remainder bein g carbon diox ide and n itrogen .
The following is the average volumetric analysis ofwater gas :
Ca rbon ic oxide , CO
H ydrogen , H
Ca rbon ic a cid , CO2N itrogen , N
H ea t va lue , cu . ft
Suction producer plants are quite largely in use in connection with the Operation of gas engines . They are applicable
FUELS AND COM BUSTION
to the use of anthracite only,depending for their operation
on the suction of the engine instead of pressure on theproducer . The gas thus produced is of good quality
,with a
heat value higher than pressure producer gas with anthracitefuel . Following is the analysis
Carbon dioxide,CO2
Carbon mon ox ide , COH ydrogen , H
Marsh gas, CH 4
N itrogen , N
H eat va lue, cu . ft
This ga s is sometimes called semi—water gas,owin g to the
fact that steam is usually jetted into the suction pipe .
Liqu id Fuels.
The l iquid fuels,derived from crude petroleum by succes
sive distillation,are gasoline
,naphtha
,kerosene
,gas-oil
,and
fuel Oil s . Alcohol,derived from the fermentation and
distillation Of organic matter , is being largely experimentedwith as a probable substitute for the petroleum products
,
which,owing to the decrease of the visible supply and to
trade condition s,seems destined to increase in price
,while
the probable tendency of alcohol will be to decrease in price .
While the rate of flame propagation in alcohol vapor is slowerthan in that of gasoline or benzene
,it is claimed
,and the
claim is substantiated by government tests,that the mean
effective pressure obtained by the use of alcohol equalsor exceeds that of gasol ine under the same conditions
,
providing that the apparatus for vaporizing the fluid is
suited to the requirements . As we shall subsequently Showthat the temperature required to secure a mixture mostsuitable for complete combustion is considerably higher foralcohol than for the lighter petroleum products , i t is necessary that more heat be supplied in the vaporizer or ca rburet tor in order to Obta in these results .Table I I I gives the approximate percentage of the suc
cessive distillates of petroleum ,with their specific gravities
,
72 I NTERNAL COMBUSTI ON ENGI NES
boiling points , and flashing points . The values are subjectto considerable variation and should not be taken as morethan avera ges .
TABLE I I I .
PROPERT I ES OF PETROLEUM DI ST I LLATES .
TABLE I V .
COMPOS I T I ON OF CRUDE OI LS .
Engler 1’ Mabery , Noble County .
Table from Oil Engines, Coldingham .
The heating value of this Oil , per pound , is about
74 I N TERNAL COM BUSTI ON ENGI NES
TABLE V .
FRACT I ONAL DI STI LLATI O‘
N OF GASOLI NE .
At the final temperature there were still 5 cu . cm . notvaporized .
A l ighting gas composed of about equal parts of airgasoline vapor
,about 320 cu . ft . being Obtained from a
gallon of fuel,is not explosive and is used quite extensively
a s a n illum in an t . The gasoline is stored in an outdoor tankor carburettor
,and the pressure
,equivalent to that due to
from 1 in . to in . of water,is Obtained by an air pump
actuated by mea ns of a weight or su itable motor . The
operatin g prin ciple of the pump is a n inversion of the wet
ga s meter principle . A m ixing valve automatically con
trols the mixture by means of a balan ce Obtained betweenthe atmosphere a nd the m ixture of gasoline vapor and air.This il luminating ga s, when properly mixed with air, becomesa good fuel for gas engines.
Seventy- one test gasoline weighs lb . per gal . ; about
27 cu . ft . of it s vapor is obtain able per pound a nd 703
heat un its per cu . ft .
Kerosene .
Kerosene oil has a specific gravity of from to andweighs
,for test
,lb . per gal . I t ha s a heat value a
trifle higher than that of gasoline,about B . t .u . per
lb . The oil ignites when heated to from 130° to 1 40° fahr .
and flashes a t about 120° fahr .
FUELS AND COM BUSTI ON
H ea t of Combustion .
The heat of combustion of different fuels is obtained bymeans of a calorimeter
,in which the heat generated by a
known amount of fuel is absorbed by a given weight ofwater . The standard of measurement in this country isthe which is the heat required to raise the temperatureof one pound of water from 39° to 40° fahr . I n Europeancoun tries
,where the centigrade thermometer is used
,h eat
of combustion is measured in calories per kilogram . H eatof combustion in calories is five-ninths that of the heat ofcombustion in B . t .u .
The heat of combustion of the petroleum oils varies fromto B . t .u . per lb . and is an average
value . The value for pure ethyl alcohol,as obtain ed from
experimen t,is about or approximately two-thirds
that of the petroleum oils ; as experiments have shown thatethyl al cohol as a fuel has about two-thirds the value ofgasoline for the same weight, these figures are significant a sserving to show that the relative thermal efficien cies areabout equal .All l iquid fuels contain hydrogen in greater or less
amount,which when burned is converted into water vapor.
When the fuel is tested in a calorimeter, this steam,coming
in contact with the walls of the calorimeter , is condensed andcontributes quite an appreciable amount of heat to the waterof the calorimeter . Since the products of combustion ina gas-engine cylinder always leave the cylinder at a tem
pera ture considerably above that of boiling water , the
engine does make use of the latent heat of condensation,
which the calorimeter measures,and , in consequence , the
effective heat value of the fuel , when burned in the cylinderof an engin e
,is of a value lower than the theoretical as
obtained by calorimeter test . For this reason it is customa ryto compute
,and deduct
,this latent heat of condensation
when comparing engine fuels . The resulting value Obtainedis known a s the low heat value. Supposing the heat valueof gasoline, as obtained by calorimeter test, was B. t .u.
76 I NTERNAL COM BUSTI ON ENGI NES
Taking the chemical symbol to be CGH H , the percentage of
H,by weight
,would be
6
iw —
4
X
X
1
" and since hydrogen
unites with oxygen to produce water,in the ratio of l to 8 ,
by weight, then the weight of water (vapor) produced in thecombustion of 1 lb . of gasoline would be X 8)
and the latent heat of water at atmosphere beingfound
,from the steam tables
,to be then
,for
1h. the laten t heat would be x a nd
1680 the low heat value forgasoline of the above chemical composition . Since thechemical composition of gasoline i s subject to such wideranges
,the actual low heat value can on ly be Obtained by
securing its volumetric analysis,to determine the actual
amount of hydrogen in its composition,and then proceeding
a s above .
Air Necessa ry for Combustion .
The air required for the complete combustion of a fuel ofdefinite known composition ca n be a ccurately determined .
This cal cul ation may be made for a fuel such as alcohol , butcan only be approximately determined for the petroleumfuels.
For ethyl alcohol we have the formul a CZH SOH . I t s
molecul ar weight is :
Ca rbonH ydrogenOxygen
For the complete combustion of one molecule of a l cohol,
the two atoms of carbon require four atoms of oxygen toproduce CO2 (carbon diox ide) a nd the Six a toms of hydrogenrequire three atoms of oxygen to produce H 20 (watervapor). There is one atom of oxygen in the molecule ofalcohol and hence the oxygen which must be derived fromoutside sources is (3 4) 1 6 atoms . The weight Ofthe 6 atoms is 6 X 16 96 . H ence complete combustion
of 1 lb . of requires" lb . of oxygen,and
FUELS AND COM BUSTION
Since in 1 lb . Of dry air there i s lb. Of oxygen,to obtain
lb . would require 3223 lb . of air . I f the air is
quite damp, one pound wou ld contain less free oxygen andhence more air would be requ ired .
Assuming the composition of gasoline to be CsH u ,the
amount of air required for complete combustion wou ld bedetermined in like manner . The molecular weight is :
Ca rbon 6 X 12 72
H ydrogen 1 4 x 1 14
The 6 atoms of carbon require 1 2 atoms of oxygen and the14 atoms of hydrogen require 7 atoms of oxygen . Now
since all must be derived from outside sources the amountof oxygen requi red is 1 2 7 19 . The molecular weight
304of the oxygen would be 19 x 16 304, and 86lb .
of oxygen per pound Of gasoline, and 3233 lb . of air
required .
Table VI gives the amount of a ir required for the combust ion of one pound of different fuels
, assuming theirchemical formulas to be correct.
TABLE VI .
AIR REQUI RED FOR COMBUST I ON OF DI FFERENT FUELS .
78 I N TERNAL COM BUSTI ON ENGI NES
The mixture of water with the al cohol reduces the amountof air necessary for complete combustion . The ordinary commercia l denatured al cohol generally carries 10per cent of water .
Vaporiza tion .
I n order that a liquid fuel mav be used in an internalcombustion engine
,it must first be vaporized
,and the vapor
thus Obtained mixed with air in the proper proportion .
For their complete vaporization , fuels , such as kerosene ,with a high boiling point
,require much greater heating in
the vaporizer,than fuels with a comparatively low boilin g
point . Al cohol,kerosene
,and the crude oils require greater
heat than the lighter hydrocarbons . Gasoline,being ex
t remely volatile , is easily vaporized at ordinary temperatures.
All substan ces which liquefy at ordinary temperatureshave a definite limit to the amount of their vapor whichmay exist in a ny given space at a given temperature ,
Accordin g to the laws for perfect gases,which we assume
to hold true in the case of these vapors,at a given constant
temperature the weight of a ny vapor presen t in a cubicfoot of space is proportional to its vapor pressure a nd may
be measured by it . I llustratin g this,imagine a cylinder
containing any vapor,at a pressure corresponding to 20 mm .
of mercury,and maintained at a constant temperature
of 60° fahr .,during the experiment
,to be fitted with a
perfectly tight piston . I f the vapor is now compressed,by
means of the piston,to half its original volume
,the pressure
will rise to 40 mm . of mercury,and if this volume is again
halved,its pressure will rise to 80. But for any vapor there
is a saturation pressure corresponding to every degree oftemperature
,and if the compression is carried beyond this
point the pressure will not rise but will remain constantand
,as the volume is further decreased
,a part of the vapor
will be condensed into a liquid,but the amount of vapor
per cubic foot of space wil l remain constant . Suppose,
for illustration,that the vapor being compressed was that
of gasoline,and that we carried the pressure to that of
160 mm . of mercury . We would then find that a part of
FUELS AND COM BUSTI ON
the vapor had begun to condense , a s the vapor pressureof saturation for gasoline vapor at 60° fahr . is about 1 57 .
Table VI I gives the vapor pressure of saturation forseveral substances as found in the Bu l letin 191
, UnitedStates Department of Agriculture .
TABLE VI I .
VAPOR PRES SURE or SATURAT I ON FOR VARI OUS SUBSTANCES .
Mercury
W a ter .
Gasoline , being an extremely volatile substance, ha s acomparatively high saturation pressure
,while for the alcohols
it is comparatively low .
Sin ce the vapor pressure determines the amount of vaporwhich may be contained in any given volume for a giventemperature and pressure , if we know the amount ofvaporous fuel necessary to produce the best explosive mixture
,when mixed with air
,we may determine the lowest
temperature at which this may exist at atmospheric pressure . As the vapor pressure of alcohol is low, we mayexpect to find the necessary temperature correspondinglyhigh . The existence of different gases or vapors in thesame receptacle
,if they have no chemical aff inity for each
80 I N TERNAL COM BUSTI ON ENGI NES
other,does not affect their action . Then a mixture of dry
air and al cohol vapor,or vapor of a hydrocarbon product,
in the proper proportions,a s previously determined , for
complete combustion,will
,when under pressure
,derive
a portion of its vapor pressure from the air a nd a portionfrom the vapor of the fuel in the mixture . By means ofAvogadro ’s law we are able to compute the vapor pressuredue to the vaporized fuel . Pressure a nd tempera ture re
ma in ing con sta n t,the densi ty of a ny ga s i s proportiona l to
i ts molecula r weight. H ydrogen ga s is taken as the basis Ofcomparison . Since the molecular weight of hydrogen is 2and that of ethyl alcohol is 46 , then the relative density
23 . That is to say, 1 lb . of ethyl alcohol vapor
confined in a given space has a vapor pressure equal toone-twen ty-third that of hydrogen confined in the samespace a nd at the same constant temperature . The densityof air bein g on the same basis of comparison
,then
lb . of air,the amoun t
,as previously determined
,
necessary for complete combustion,would have part1
of of the vapor pressure of 1 lb . of hydrogen . Then therelative vapor pressures of the alcohol vapor and air are as1 or a s The total vapor pressure may23
then be considered as equal to of
which 00 435 per cen t is due to the alcohol vapor,
and per cent is due to the dry air . Under ordinaryatmospheric pressure of 760 mm . of mercury
,that is, lb .
per sq . in .
, the vapor pressure o f the alcohol would be760 X 49 . mm .
,which we find from the table cor
responds to a temperature of a trifle more than 7 1° fahr .H ence , a perfectly combustible mix ture of air and alcoholcannot exist at a temperature of less than 71 ° fahr. , althougha mixture carrying an excess of a ir may exist at a muchlower temperature . I f the air is damp and the alcohol
,
as is nearly always the case,carries a certain per cent
8" I N TERNAL COM BUSTI ON ENGI NES
Acetylene liquefies, at atmospheric pressure, at a com
para t ively high temperature, about 1 16° fahr .,while under
a compression of 600 to 700 lb . it l iquefies at temperatureshigher than the ordinary temperatures of air . I t is suggested by some that liquid acetylene would be a veryavailable fuel for internal—combustion engines .Acetylene has a heatin g value of about B . t .u . per
lb .,its formula being C2H 2 ,
and it requires lb . of airfor the complete combustion of one pound Of gas .
One pound of cal cium carbide will yield about lb . ofthe gas and will use about lb . of water in the process .
One pound of liquid acetylene , evaporated , will producelb . Of gas at atmospheric pressure .
Acetylene gas ignites spontaneously at lower pressuresthan gasoline vapor or n atural gas
,a nd
,in consequence
,is
not available for high—compression engines .
Results of foreign experiments seem to Show that itrequires from 5 to 7 cu . ft . of acetylene
,mixed with air in
the proper proportion,to produce 1 hp .
-hr .
The rate of flame propagation is high,the temperature of
combustion is high,and the energy derived is high
, but the
cost of production of the gas is more than twice that ofgasoline vapor
,for the same amoun t of work . The l iberation
of the gas,in a closed vessel
,may also produce a dangerously
high pressure . I t is doubtful if the use of acetylene as afuel will become very general unless the cost of produ ctionis considerably lessen ed or the cost of the other availablefuels increased to such an extent a s to make its usecomparatively economical .
Alcohol .
We have already shown the action of alcohol,in regard to
its vaporization and combustion , to be considerably differentfrom that of gasoline . The low heat value of denatureda l cohol is about while that of commercialgasoline is slightly more than it is thus apparent thattheir heating values are approximately as 2 3 , or, in other
FUELS AND COM BUSTI ON
words, the thermal value of alcohol is a trifle less thantwo-thirds that of gasoline . This fact is significant when weconsider that the average of the government tests on thecomparative values of gasol ine and al cohol shows that theratio of their con sumption per b . hp . in the same engine andunder the same conditions was approximately asI t seems to Show that the operation of al cohol and ofgasoline in an internal-combustion engine is practicallythe same a s regards the thermal effi ciency . I t is possible,however
,with alcohol
,to use a higher compression than
with gasoline,and earlier ignition may be used and still
produce a smooth-running engine .
A gasoline en gine of ordin ary design may burn al coholwith more or less success, but to secure the best resul ts thecarburettor must be adapted to the requirements
'
of alcoholvaporization . When properly arranged an engine willdeliver slightly more power with alcohol than with gasoline .Al cohol being of known chemical composition , unmixed withimpurities other than water
,has no inherent tendency to
foul the in terior of the cylinder .The exhaust from a gasoline engine is much hotter than
that of an engine Operating on alcohol,and
,in consequence
,
there is less danger from fire in the latter .Al cohol may be used and stored with much less danger
than gasoline,and its use
,both in the matter of the exhaust
products and in the handling,is much more pleasant .
Al cohol can be produced anywhere from the distillation oforganic waste products .The cost of alcohol is still much more than that of gasoline,
and unless the price of the latter increases considerably orthe price of al cohol is decreased
,it is doubtful if it will come
very rapidly into general use.
CHAPTER X .
COMPRESSI ON .
TH E compression pressure of an internal-combustionengine varies from an extreme high limit of 500 lb . persq . in . or a trifle more to 60 lb . per sq. in .
,both these pres
sures being above atmosphere . As the pressure is increased ,the temperature Of the compressed gases increases un tilwith a pressure of 500 lb . the gas attains a temperature ofnearly 1000° fahr. and is incandescent.The compression in any type of en gine is governed entirely
by the conditions . I f a fuel with a high ignition poin t beused , the compression may be high , as in the use of producer
ga s, in which engines the compression is carried as high a s
from 150 to 200 lb . per sq . in . I n the Diesel type of motorit is possible to use the extremely high compression of 500lb . per sq . in .
,owing to the fact that during the compression
stroke the gas in the cylinder is pure air,and
,as previously
described,ignition is Obtain ed by the inj ection of the fuel
into this incandescent cylinder . The compression of a lowspeed gasoline engine should not exceed 85 lb . per sq. in .
,as
this produces a temperature in the cylinder sufficient insome cases to ignite the charge . As the engine speedin creases
,the compression may be increased to such a poin t
that ignition,due to the compression
,does not take place
previous to the proper lead of ignition,in the compression
stroke,for the speed in question . Many automobile engin es
are bu il t with a compression a s high as 95 lb . gage,and
give excellent satisfaction when operating at an averagespeed of from 800 to 900 rev . per min . ,
whil e a two—cylinderengine which operated at a speed of from 300 to 400 rev . permin . and with the same compression gave constant troublefrom premature ignition .
86 I N TERNAL COM BUSTI ON ENGI NES
the H ornsby-Akroid, Obtain their ignition by means of highcompression and residual heat in the vaporizer . Theseengines compress their charge to about 200 to 2 10 lb . persq . in .
The efficiency of an internal-combustion engine increaseswith its compression
,as may be shown .
We have the two equations for the heat absorbed at consta nt volume and constant pressure .
dH Cvdt pdv
dH d l vdp
Supposing the gas,to expand without the transm ission of
heat
Dividing (b) by (a ) gives
d_p Q9 fl fi v
(r being the specificP
hea t at constan t pressure to the spe cific constantvolume).
I ntegrating we have
P:
I n which pm and vI a re the in itial lim its a nd py and vyare
the general limits.
* Then
Pg); In} ;
Now a perfect gas is one in which the rate of expansionequals the rate of temperature increase if under constan tpressure
,or the volume bein g constant , the temperature
will in crease a s the pressure . Then
pyvy cc tu and v ,rO: t
I n a n in dica tor diagram the lim its px a nd vx correspond to thepressures a nd volumes at the beginn ing of the compression or the
end of the exhaust stroke .
COMPRESS ION
We may now el imina te pa a nd write equation- 1
Now, supposing v to be a constant, we would have from
equation 1 , since the derivative of a constan t is 0 ,H Cv (tr t
u) (4)
(The heat absorbed at constant volume is equal tothe specific heat at constant volume times the range oftemperature .)Drawing the theoretical card 1
,2, 3 , 4 (Fig. in which
heat is considered as being absorbed from 4 to 1 a nd rej ected
Fig. 25. Theoret ica l Card .
from 2 the constant volume lines, 4— 1 and beingthe theoretically perfect explosion and release lines in aninternal-combustion engine, the effi ciency would be
[ge m = 1(to
.
ta)
Hl — 4
t". t4 (t 1 I ‘)
But v1 v4 and v2 v then if T be constan t, we have ,from equation 3 ,
— L1 = 1 _
88 I N TERNAL COM BUST I ON ENGI NES
as the compression is increased,the value of
fraction,23 will decrease
,since the initial pressure p3 is that
of the atmosphere,and p4 in creases . The value of -E will
then approach 1,as a limit . I f it
’ were possible for E toequal 1
,then the efficien cy would be the maximum theoretical
value,which value is n ever even approached in a heat engine .
Equation 2 is that of the ideal indicator card and itsapplication will be fully discussed in the following chapter.
90 I N TERNAL COM BUS TI ON ENGI NES
ordinates . The general form of equation for the compression curve cd and also for the expansion curve ef is
p i)" k (5)
I n which
p pressure above a vacuum .
volume at pressure p .
k consta nt depending upon conditions but whi ch wi ll
rema in the same for a ll poin ts of the same curve. For conven ience we will denote the constant for the compressioncurve by Ice and that for the expan sion curve by he . Con
sidering the volume of the cylinder as unity , the va lue of theconstant kc will be the pressure of the atmosphere , whichwe assume to be the pressure in the cylinder at the beginningof the compression stroke and which is taken to be lb .
per sq . in . The determination for the value of he takes intocon sideration the fact that the pressure of the explosion isapproximately four times the compression
,and
,having
this value determined,we m ay solve the formula for k
which will remain constan t throughout the expansionstroke and is the value of p if the gases were to release atthe end of the stroke
, 9 .
I f a volume of air is con tained in the perfectly tightcyl inder fitted with a perfectly tight pist on and compressedat constant temperature , then the value of 7’ would become 1 .
I f the volume of air is rapidly compressed and all the heatproduced by this compression is retained
,then the value
of 7” will increase . The value of 7 is the ratio between thespec ific heat of air
,or whatever ga s is under compression , at
constan t pressure and its specific heat at constan t volume .
According to Ran kine,
r
A
")
for air .
For the gases in common use in the cyl inder of the ga sengine the value of T for the compression and expansioncurve is considerably lower , due to the ratio of the specificheats being different from those for air and to the losses dueto leakage past the piston and va lves and the loss of hea tthrough the cy l inder walls . For genera l practice the v a lues
TH E I NDI CATOR CARD
of r may be taken as for the compression curve andfor the expansion curve
,and these values will be found to
produce results very closely in keeping with the actualworking pressures obtained by indicator tests on efficienttypes of internal-combustion engines .The value is very nearly the average of results obtainedcomputation of different samples of natural gas ; so
Fig . 26 . I dea l I nd ica tor Card .
it may be said that the average internal-combustion engineis designed on the basis of natural gas as a fuel and actingunder ideal conditions .The value r is taken for the expansion curve in
view of the fact that the higher temperature of the gasesduring the workin g stroke results in a greater difference intemperature between the contents of the cylinder a nd thej acket water
,with a consequent greater loss of heat .
I n order to show just how the ideal indicator card may bebuilt up by means of the formulas , we will compute thevalues for several points of the compression and expansioncurves in Fig. 26 .
Before laying out the diagram or computing any values,it is necessary to decide upon some pressure for the compression or to assume some ratio for the volume of thecompression space to that displaced by the piston duringone stroke
,and to solve for the compression pressure . I n
92 I N TERNAL COM BUSTI ON ENGI NES
this model card we will assume the ratio to be aswhich will result in a rather high compression pressure .
The total volume of the cylinder at the end of the suctionstroke then will be 1 times the pistondisplacemen t . For convenience in the computations we
will assume this volume as un ity . The percentage of thecompression space to the total volume of the cylinder is
thenfi
Which represents the value of
Now using the formula pv“ 3 = k
,we have p X
or p
I n order to solve the above equation it is necessary to uselogarithms.
Log
Log is obtained as follows : Log as foundin the table of logarithms (that is the tabular log), is
Since the tabula r log ha s a n egative characterist ic
,
* we must find the true log by subtractin g the man
tissa (number to right of decimal poin t) from 1,as follows.
may be written — 10 .
Then we may subtract
10 . 000000 — 10— 10
T rue logMultiplying by to raise to that power
20 173 1 1
672437
0 874 168 True log1 125832 T abular logNow log log log p.
— 10 log— 10 log
Tabular log p .
The cha ra cteristic is the number to the left of the decima l pointa nd is 1 for a ny number in tens pla ce
,a s 2 for a ny number in
hundreds pla ce . as 3 for a ny number in thousa nds pla ce , aset c . For dec ima ls , the cha ra cteristic is : 1 for a ny number
in tenths , as — 2 for a ny number in hundredths, as 3
for a ny number in thousandths , as etc .
94 I N TERNAL COM BUSTI ON ENGI NES
H ence the equation for any point of the expansion curvein question is po
l '3s The intermediate workingpressures, corresponding to the computed compressionpressures , at and of the working stroke , arefound by subtracting the logs of the different values of v, a spreviously determined
,raised to the 1 .35th power from the
log of They a re a nd
Suff icient points now hav in g been determined,the curves
are drawn a s shown in Fig. 26 . The scale of indicator springin this diagram is taken to be 140 lb . per in . and the lengthof the diagram from a , to c is 4 in . Then every inch in heighton the diagram represents 140 lb . pressure
,a nd un it volume
is represen ted by the length 4 in .
We now have completed the theoreti cal diagram cdeg,
which would represent the action in the cylinder if theexhaust valve opened a t g a nd the pressure dropped on avertical l in e to c . Plain ly this would be purely an idealcondition . To complete the actual card it becomes neces
sary to pass a curve from the intersection, f , of ordinate 1 and
the expansion curve,at which point the exhaust is supposed
to open,to point c. This curve should be the arc of a circle
,
havin g its center on ordinate 1 and intersecting pointsf and 0 .
At (1 a sl ight curve,representing the increase of pressure
due to the advance of the ignition,is drawn . A straight
vertical line drawn from the in tersection of this curve andordin ate 10 to point e completes the diagram .
The diagram having been drawn,the mean effective pres
sure may be computed as described in Chapter XX I I I . The
mean effective pressure hav ing been found,the designer
may compute the horsepower of any proposed en gineof l ikecompression
,or he may determine the dimension of bore
and stroke to use for any proposed horsepower .Fig. 27
,represen ting the compression curve pv k
,may
be used to determ ine the compression space a nd to locatea ny point on the compression curve from 20 up to 500 lb .
By the use of this curve the computation of the values forpoints on the compression curve is rendered unnecessary .
Fig. 27 a lso shows the terminal pressures of the expansion
TH E I NDI CATOR CARD
curve for compressions from 60 to 120 lb. ,all these values
being computed on the a ssumption that the absolute explosion pressure is four times the absolute compression pressure .
To use the curve proceed as follows : I f the compressionhas been determined and the per cent of compression spaceis required , find, on ordinate P or P ,,
the pressure of com
Compression Curve. P V l -3=K.
9 101 48 623 1 18
65 78 80 102 48 678 1 14
66 78 91 108 1 1 5 49 108
67 80 92 104 1 1 6 49 187
68 81 93 106 48 783 1 1 7
69 82 94 106 48 80 ) 1 18
70 46 65 1 83 47 708 85 48 .864 10 7 119
71 84 47 770 96 48.416 108 120
Fig. 27. Compression Chart.
pression and trace the horizontal ordinate passing throughthis point to its intersection with the curve ; from the pointof intersection trace the vertical ordinate to the horizontalordinate V or V,, where the per cent of compression spacerequired for the compression in question may be read .
I nverting the order of procedure will give the pressure whenthe per cent of compression space is known . I n the columnsto the right are given the terminal pressures for the expansioncurve. With the curve and the values of the terminalpressures given
,in order to construct a complete ideal card
it is only necessary to compute the intermediate points forthe expansion curve .*
I n using the curves to obta in values from 500 down to 125 lb . ,
use the reference ordinates P and V in connection with the curve A.
For values from 125 down to 20 lb ., use the reference ordinates P, and
V1 in connection with curve B.
96 I N TERNAL COM BUSTI ON ENGI NES
As stated in the chapter on Engine Testing,the
mechanical efficiency of any engine is the ratio of the b . hp .
to the i . hp . I t is the general supposition that the i . hp . a s
obtained from an actual card is correct within reasonablelimits . Un fortunately experien ce has proven this to beerroneous in tests where proper care wa s not exercised . The
cards taken from a steam engine running under a steadyload vary much less than those taken from a ga s enginesupposedly under as favorable conditions
,and in couse
quence, it is apparent that the mean i . hp . as obtain ed fromthe cards taken in a test may vary considerably from theactual mean power developed in the cyl inder .
I n the tests of the committee of the I nstitution of CivilEngineers on the efficiency of intern al- combustion enginesit wa s brought out that a small 5—hp . engin e showed amechanical efficiency of 90 per cen t , while a 20—hp . engineof the same type showed an effi ciency of only 80 per cent .I t was man ifest to the comm ittee that these results werein error
,a s the effi cien cy of a ny engine tends to increase as
its size increases . Continuing their tests , however , theydemon strated that very accurate results could be obtainedif a properly designed indicator wa s used a nd care taken tomain tain a un iform gas pressure . A paper by BertramH opki nson
,read before the I n sti tution of Mechan ical
En gin eers and excerpts of which may be found in The
Engineering M aga zi ne for January 1908, gives the resul tsof his exten sive experimen ts alon g these l ines a nd shows thathe was able to obtain diagrams correct to within 1 or 2 percent . I t may be added tha t the pressure of the coolingwater should also be m ain tained as n early uniform as possible
,in order to secure the best results .
98 I NTERNAL COM BUSTI ON ENGI NES
delivered by a ny cyl inder , the mixture is assumed to betheoretically correct
,th is factor need not be considered
in the design . The highest mean eff ective pressure isobtainable by the use of gasoline
,with natural ga s a close
second ; then come the illuminating gases in the order of theircandlepowers
,then water gas
,producer gas
,and the poorer
quality of gas derived from the blast furnaces.
The average mean effective pressures range from 40 lb .
per sq . in . to 97 or 98 lb . per sq . in .
,the latter values being
found in the Diesel oil engines . An average value fornatural gas or gasoline is from 65 to 70 lb . per sq. in . The
above values a re all gage pressure readings.
The cyl inder compression having been determ ined upona nd the theoretical card con structed , as described , the
mean effective pressure obtained from this card m ay nowbe used to determine the bore and stroke necessary toproduce a given b . hp .
Let ( I diameter of cylin der in inches .
l len gth of stroke in in ches.
n revolutions per m inute.
7) mean effective pressure .
Pb brake horsepower .
I n this formula the fuel used is supposed to be equal inheat va lue to n atural ga s or gasoline and the pressure ofthe explosion to be four times the compression pressure .
For poorer gases a fuel fa ctor must be introduced in theformul a . Then for a four-cycle en gine
dzpzn
And for a two- cycle engine
For an en gine operatin g on producer ga s a nd with thesame compression the fuel factor may be taken a s approxi
mately Producer ga s engines,designed parti cularly
GENERAL DI M ENSI ONS
for that purpose,generally carry their compressions much
higher . (See chapter on For enginesoperatin g on blast-furnace ga s a fuel factor of to isnecessary .
The above formulas may now be used for determiningthe dimensions of a ny en gine of l ike compression to bedesigned . For example
,supposing it is desirable to build
a single-cylinder,four-cycle engine capable of delivering
40 hp . and that the mean effective pressure ha s been determined a t 70 lb . per sq . in . I t is necessary
,in order to solve
the equation,that but one value be unknown . I n order to
accomplish this result we must assume the value for twoOf the unknown quantities and solve the equation to determine if the remaining unknown value is in the properproportion . As the speed at which the engine is expectedto run is generally known
,we assume this value and the
length of stroke,leaving the formula to be solved for d.
I n this case we will assume the engine speed to be 200rev . per min . and the length of stroke to be 18 in . whichwould result in a piston speed of 600 ft . per min .
,approxi
mately the correct value for a stationary,low-speed
,hori
zon t a l engine .
Tbe“
and
The percentage of compression space of the entire cylindervolume being known
,it becomes a n easy matter to deter
mine the length of the space,having determ ined the value
of d. For example , supposing that the per cent of compression space necessary to obtain a mean effective pressure of65 lb . was found to be 23 , or the compression space wa sof the total cvlinder volume .
Then V= v V.
I n which V the total cylinder volume andv piston displacement .
I NTERNAL COM BUS TI ON ENGI NES
Solving we obtain the value of V,and , the area of the
compression space being known,we are able to determine
its length .
The proper ratio of bore to length of stroke is,for stationary
four-cycle engines,a s or For two- cycle en gines
this ratio should be decreased,some engines even being made
square,that is
,with the stroke equal to the bore . Auto
mobile and marine engines,owing to the requiremen t that
they set low ,are made with their stroke from 1 to times
the bore,and often run at a piston speed a s high as 800 ft .
per m in .
I n the design of an en gine it i s often found expedient tofollow as closely a s possible the gen eral ratio of dim ension sof some. engine of the same type , the operation of which ha sbeen found by experien ce to be uniform and satisfactory .
As an example take an en gine with a cylinder diameter of12 ln .
,a stroke of 18 in .
,and a mean effective pressure as
shown by the card of 60 lb . per sq. ih . ,which ha s been found
to give good satisfaction at 1 90 rev . per min . a nd to del iver30 b . hp . when operating on natural ga s a s a fuel .Then applying our formula for four- cycle engines we have ,
a s an unknown value , the value of the denom in ator of thefraction . Calling this value :v the formula may be writtenas follows :
60 x 18 X 190
a:
Solvin g for so we find its value to beThen the equation for a simi lar en gine of any desired
horsepower would be
I n a similar way a formul a may be found for a ny enginethe operation of which is known .
Formul as 6 and 7 may be simplified a nd made to applyto special cases only . That is
,when the piston speed
is determined as 600 or 700 ft . per min .,the diameter of
CHAPTER X I I I .
TH E CAM MECHANI SM .
PROBABLY the most important mechanical feature in thedesign of a n intern al-combustion engine is the cam mechanism
,as nearly every engine with mechan ically controlled
valves uses this mechanism a s a basis for transforming therotary motion of the crank or cam shaft into the straightline motion n ecessary for the operation of the mushroomvalve .
The c ams in a four-cycle en gine are commonly located on
the cam or half-time shaft,so called because it is designed
to revolve at a speed equal to half that of the crank sha ft .
The reason for this is apparen t when we con sider that in acomplete Otto cy cle the crank shaft makes two revolutions
,
a nd since the exhaust and in let valves open but once duringthis cycle
,their operating mechanism must work half a s
fast a s the cran k shaft.Occasionally engines are manufactured which use some
mechan ical principle other than 2 1 reduction gearing a nd
cam shaft to attain this end . An eccen tric mechan ism may
be devised,or other suitable methods may present themselves
to the designer . Fig. 28 represents a half- time mechanismoperated by mean s of a n eccentric on the crank shaft. I n
many small en gines , as has been prev iously mention ed , onlythe exhaust valve is mechanically Operated
,the inlet being
Of the suction type .
I t is good practice , and one adhered to in the manufactureof most of the large engines, to tran smit the cam motion tothe valve stem by mean s of an intermediate member . The .
maj ority Of the smaller engines of the automobile and marinetypes interpose a push rod between the valve stem andthe cam it is not good design to allow the valve stem even
102
TH E CAM MECHAN I SM
when carrying a good steel roller to take its motion directlyfrom the cam
,as the side thrust will tend to wear the rod
bushing . I f for the purpose of reducing cost this method ispursued
,the stem must be given a good liberal bearing to
minimize this tenden cy as much a s possible . The camshould also be set out Of l ine in thedirection toward wh ich the eccentricportion is rotating as it approachesthe push-rod roller ; this will reduce thecomponen t of the cam movement acting at right an gles to the axis of thevalve stem . I n Fig. 29 a b is the centerline of the valve stem and roller ; cd isthe center l ine of the cam ; A is the
guide block, of squared stock , forkedat the lower end to carry the hardenedsteel roller
,a nd having the push rod
tapped into its upper end a nd lockedwith a thin lock nut
, C,a s shown
,thus
providing adjustment of the mechanism so that the cam may commen ceto lift the valve at the proper insta n t ;B is the guide
,usually of brass
,
although it may be integral with thecase
,D . Even when an intermediate
lever or push rod is interposed,the
Eccen tric H a lfoffset tmg of the cam is good practice . time Mechan ism .
A common form of cam mechanismwith lever transmission device is shown in Fig. 30 ; a
being the hardened steel roller,on which the cam operates
,
and b a hardened steel pin , or other suitable device,
which impinges on the push rod 0 . Bearing d carriesthe transmission lever c . This mechanism is used quiteextensively where the engine is started by compressedair
,and at least one exhaust cam is made double . I n
order to bring this double-acting cam into startingposition
,it becomes necessary to shift either the cam or
the cam roller , the latter method being the more con
I NTERNAL COM BUSTI ON ENGI NES
veulent, owing to the fact that to shift the cam shaft wouldentail throwing the ha lf-time gears partly out of mesh ,or making the crank-sha ft pinion with wide enough face to
Rela t ion of Cam to Push Rod .
allow the cam-shaft gear it s necessary movement . I t is
compara tively easy to make shaft d the shi fting medium ,
cause the roller a to move into the proper position for
Fig. 30. Cam w ith Lever Transmission .
starting ; pin b must be replaced either by two pins or ahardened steel block of suffi cient len gth to allow the lever eto move into position for starting and yet maintain a contact
I N TERNAL COM BUST ION ENGI NES
valve required . The period of opening of the exhaustvalve is in terms of the crank motion
, X 180°
2 16°
(since the valve Opens before the end of the explosionstroke and closes after the beginning of the suctionstroke). On the cam shaft
,then
,the period would be equal
2 16
2
H aving determined the period of Opening on the cam shaft,
proceed as follows : the size of the cam shaft being known ,determine the diameter to be used for a
,and make the
diameter b of the con cen tric portion of the cam,or “ dwell
,
”
enough larger to al low for a good finish on the cam proper .H av ing determined the diameter b of the “ dwell
,
” lay out acircle of diameter c
, 113. in . greater in diameter than b. The
eccen tric por tion of the ca m. i s to be la id out on this la tter ci rcle,
a nd not on the ci rcle of diameter b. The reason for this willbe explained later . The angle Of the cam having been determined as lay Off 54° on either side of the center l ine xx
,
a s shown . Through the poin ts e and e, where the 54° l ines
intersect the circle c,pass tangents to the circle b of
the cam . I n order to draw these l ines through e and e
tan gent to b,the angle a must be kn own
,and may be deter
mined by means of the following formula :
0angle of action 1
2
Substituting : a 2 cosl—g
2 (0 4° 38°
Angle a being known , the tan gent surfaces of the cam maybe readily drawn through poin ts e. From the formula for
-Xa = 2
rI n the above trigonometrica l formu la , the term, cos
— l
rmea ns
I O r 0 l
the angle whose cosme i s;l I n the cam i n question r
, fl a nd r=
ta. I n decimals, r, r and From the
table of functions we find tha t the angle whose cosine is is
about ( I t is really 16°
THE CAM M ECHAN I SM
eff ective valve a reas we have determined on the lift necessaryfor the valve which this cam is to actuate . Supposingthe lift to have been determined as in . ; l ay OR a con
centric circle Of in . greater diameter than circle c and ,when this circle intersects lines Oe
,the cam outline will be
completed . Corners g should be slightly round in themachining
,so as to make the movement of roller from the
tangent rise to the “ dwell ” 99 as gradual as possible ,without materially affecting the valve opening .
The reason for laying out the eccentric portion of the camon a circle 1
13 in . larger in diameter than the cam itself is
that there may be no lost motion to take up during theangle of action . As in most cam mechanisms there is aslight clearance allowed between the valve stem and themember interposed between it and the cam
,this clearance
must be taken up before the valve can possibly commenceto open . Added to this clearance allowed there are almostsure to be some other slight losses , due to looseness of bearingpins or to wear Of parts . (Especially is this true in olden gines .) Now when the roller traveling on the camcircle b encounters the tangent portion of the cam itcommences to rise
,and when the roller passes the point of
intersection of this tangent and the circle 0, on which the
an gle of action is laid out,i f our assumption as to the amount
of lost motion is correct, the valve should be j ust commencingto open
,and should remain open until the roller passes the
intersection of the other tangent surface and circle e. As amatter of fact 31, in . is probably a little too much to allowfor play
,and may be reduced , i f the designer sees fit, to a
smaller amount ; but it is doubtful if, after the engine ha sbeen run a few months , 31; in . would be at all excessive .
H aving laid out the exhaust cam, the inlet cam may bedrawn in the same manner, the angle of operation of thecam being determined in the same way as for the exhaust.Some manufacturers obtain the opening of the exhaust
and inlet valves by means of a double cam in connection witha rocker arm. Fig. 32 gives the general idea of this camand va lve mechanism,
Rocker arm a , pivoted at b, is
I NTERNAL COM BUST I ON ENGI NES
provided with two bearing points, 6 and d, which operate ,
respectively,the exhaust and inlet valves, e and f. The
push rod is connected at g, as shown, and its movement isrestricted by the coil Spring h. The push-rod mechanismmay be identi cal with that used for an ordinary simple cam ,
Doub le Cam.
and need not be described . The j oint at 9 must be madecompensating, to preven t binding, as its center follows thecurve xx. Fig. 33 , showing a sectional V iew of a vertica lengine
,illustrates the method of double cam valve mechanism
as it has been applied to automobile engines . The generaloutline of cam is shown at A,
Fig. 32 , but in practice the camOutline , instead of dropping suddenly into the depressiona s shown
,is made to cut across on a chord of the cam circle
,
as shown in Fig. 33 . While this reduces the noise of operation
,it plainly reduces the mean eff ective opening
,a s will
be later described . The Operation of the double cam is a s
follows : in the position shown, the inlet valve f is wideopen
,the spring h having caused the lever a to move about
I N TERNAL COM BUSTI ON ENGI NES
contact with the stem of f . When the roller reaches thetangent portion of the cam ,
a t 2 , the push rod rotates lever aabout b, causing the exhaust valve e to open a nd rem ainopen until poin t 3 is reached
,at which poin t the roller starts
in to the depression 3 to 1,a nd allows spring h to again open
the inlet valve f . Fig. 34 shows the method of laying out
Fig . 34 . Double Cam Lay-out .
this That part of the cam from e to e laidcircle 0 a s ha s been already described . The methodlaying out the in let portion is a s follows : draw circlediameter d 1
11; in . less in diameter than the cam circle b, the
reason for this being the same a s for the constru ction ofcircle 0 . The angle of operation of the inlet being determ ined
,a s well as its location in the cycle
’
(the inlet valveis supposed to open immediately after the exhaust closesand remain open during a period of or a trifle more
,
on the c rank circle . I n this particu lar cam we will keep thein let valve open through exac tly a nd cause it to openimmediately after the exhaust closes), l ay off angle 7, asshown , equal to I t is obviously impossible to lay off
the inlet portion from Ce without cutting down the eff ective
TH E CAM M ECHANI SM
openin g of the exhaust valve in order to obtain the curvefrom c to f. I n order to overcome this diffi culty
,l ay off Cf
four or five degrees from Ce,and lay Off Cg 90
° from Cf .
Draw the curves ef and gi , as shown ,tangent , respectively ,
to the tangent rise and the “ dwell ” of the cam,and connect
the points f and g. A little study of the cam outline willshow that the period of maximum openin g Of the inlet valveis but an instant
,the valve Opening to its maximum at h
,
and immediately commencing to close again . Due allowan ce must be made for this feature
,and the valve made large
enough to produce an average open in g of sufficient area toprevent throttling the charge . While the double camreduces the number of parts necessary
,it is claimed
,and
with good reason,to be more n oisy than two separate cams .
The cam is usually made of machine steel,case hardened
,
a s are also the rollers . They should be located on the camshaft and secured in place with taper pins . The angularlocation of the exhaust and inlet cam
,if on the same shaft
,
is an important operation,a nd
,un less great care is used
,the
timing of the valves wil l be wrong. Either cam may bereadily located alone , but to secure their relative locationsbeing correct requires extreme precision . The best methodto use
,i f manufacturing in quantities
,is to provide a drilling
a nd reaming fixture for the shaft and cams in which therelative position of the two cams has been accurately located
,
a nd in which position they are held while the pin holes arebeing made . The gear may then be keyed to the shaft andthe cams set to their proper location in the cycle .
The Reduction Gea ring.
The gears con necting the cam shaft to the crank shaftare of the ordinary spur variety if the cam shaft and crankshaft are parallel ; if the cam shaft is at an angle to the crankshaft
,but in the same plane
,bevel gears are used ; if at an
angle,and not in the same plane
,skew gears must be used .
The spur gear consists, essentially, of a disk, the teeth of thegear being cut on its periphery ; the bevel gear has its teethcut on the surface of a cone . The skew gear has its teeth
I NTERNAL COMBUSTI ON ENGI NES
cut on the periphery of a disk, but at such an angle with
the shaft that they will mesh with the teeth of a gear theshaft of which is neither parallel nor in the same plane .
Fig. 35 shows these three types of gears ; at a is shown a pairof spur gears
,the gear marked 1 bein g the crank shaft pinion
and the one marked 2 the cam shaft gear . The size of pinion 1is half that of gear 2
,in order to secure the two to one speed
redu ction required . At b is shown a pair of bevel gears for
Fig . 35. Types of Gea rs .
changing the direction Of motion and at the same time
makin g the two to one reduction . The center line of thebevel pin ion 3 , which is carried on the crank shaft
,a nd that
of the bevel gear 4,which is the cam shaft gear
,l ie in the
same plan e and,in this particu lar example
,at right an gles .
As in the use Of the spur gears,the size of the cran k shaft
pin ion is made half that of the cam shaft gear,in order to
secure the proper speed reduction . At 0 is shown a pair ofskew gears
,by the use of which the speed reduction may be
secured and at the same time maintain the crank shaft geara s large or
,as most frequently is the case
,larger than the
cam shaft gear. The skew aff ords a quiet method of drivebut one in which the gears, due to their action , are subj ectto an abn ormal amoun t of wear . As the design Of gearingis a subject by itself, and one on which several comprehensiveworks have been written
,the author will not consume
valuable space in it details. Cut gears a re usually boughtdirect from manufacturers or
,if the output of a concern
I N TERNAL COM BUSTI ON ENGI NES
quiet running , the cam shaft gear is often made Of fiber, orrawhide , and brass. The fiber or rawhide is held bet ween
brass plates and the three are bolted or riveted together
Fig . 36 . Adjustab le Spur Gear .
and then cut a s one piece . The gea r shown iscon structed in this way
,parts of the rim m arked 1
,2,and 3
being brass,fiber
,a nd brass
,respectively.
CHAPTER X I V .
TH E VALVES AND PORTS.
TH E valves of the standard gas engine are Of the mushroomtype ,
and may be flat or angle seated most valves being Ofthe latter variety . See Fig. 37 for illustra tion s of thesetwo kinds of mushroom valves .The flat seated valve is most available for use as the suction
valve in engines using this type of in let, as the comparatively
Fig. 37. Types of M ushroom V a lves.
l ight weight,compared to an angle sea ted valve
,is desirable
,
sin ce lightness of the moving parts is a n important featurein order that the inertia may be reduced to a m in imumamount .The angle seated valve for all purposes other than the one
mentioned is far superior . The valve , meeting its seatat an angle
,tends to wedge itself in to place a nd seat more
firmly ; there is less tendency for carbon to collect a nd agreater tendency for the valve to grind out what maycollect . The flat seated valve
,being lighter, has greater
tendency to warp and leak
I N TERNAL COMBUSTI ON ENGI NES
Valves should be drop-forged or cast in one piece, andthen machined to finish size
,as any two-piece valve tends
to pull apart,whether the stem be tapped or riveted into
the head . I f tapped or riveted,the head must, of necessity ,
be made heavier and clumsier in order to give suffi cientbody for the stem . Fig. 38 illustrates a two-piece valvetapped and riveted .
The material of which the valve is to be made should beconsidered ; exhaust valve in particular should be made
of a metal not easily affectedby the heat to which it willbe continually subjected .
The best metal for the exhaust valve is nickel alloy .
*
This alloy is not easilyburned and will producesatisfactory results and insure a long life to the valve .
The inlet valve,while not
subj ected,except on its
n piece Va lve.
surface,to the hot exhaust
gases,should
,n evertheless,
be made of a tough steel . Many manufacturers use nickelsteel for the inlet
,with excellent results.
The best of material is none too good for the valves of agas engine ; un fortunately , however, this does not prevent themanufacturers
,in many instan ces
,from using inferior
material,with more or less successful results . Many
engines have both the inlet and exhaust valves made ofordinary drop—forged steel stock
,but while these valves may
,
in some cases,give excellent results for a time
,their use in
the manufacture of strictly high-class engines should bediscouraged .
The size of the eff ective valve Openings and of the inlet
SO-cal led n ickel a l loy is not the German si lver of commerce.
Al loy su itable for valves may be manufa ctured,or purchased di rect
from the manufa cturers, who use their own formu la . Amanganese,copper alloy, contain ing per cent of a luminum ,
is also a tough non
corroding material .
I N TERNAL COMBUSTION ENGI NES
I n ca se the inlet and exhaust passages are circular insection
,we may simplify the formulas 1 11 order to obtain
d and dl direct ly.
Substituting
(13)
d DVET. (14)
I f the engine being designed is of the stati onary type,for
which 600 ft. per min . would be a good value of S ,then we
may use the following values
a A,
A,
d D,
d,= D.
These last named values are clearly special cases,covering
but one class of design, and should not be ta ken as generalvalues .Table I X gives: briefly the values of a
, an d and d1 forpiston speeds of from500 to 1200 ft . per min .
TH E VALVES AND PORTS
TABLE I X .
eff ective valve opening must next be considered .
From Fig. 37, showing the ang le and flat seated valves, it isapparent
,after a little consideration
,that in the former the
effective opening of the valve is not the lift proper,but some
Fig. 39 . Diagram of Effective V alve Open ing.
function of it,depending upon the angle of the seat . Fig. 39
,
as a diagrammat ic illustration,will make this fact clear. I n
the figure,the lift may be represented by a ,
but the effectiveopening is a smaller quantity, b. The angle of the seat maybe made which is good practice although the value issometimes varied . Constructing the triangle as shown , wehave
b a cos a ( 15)
I n this case,the value of a being the formula becomes
b a cos
120 I N TERNAL COM BUSTION ENGI NES
The effective open ing i s, of course, the a nnular spa ce
be“
found fol lows
Or substitu ting
— b sin a
I NTERNAL COM BUSTI ON ENGI NES
points on its interior surface being equal . Such an explosionchamber is impossible in a gas en gin e because the compressionspace would be altogether too large and the radiating surfaceof such extent a s to render the idea impracticable . Amodified form of such an explosion chamber has been tried byseveral manufacturers of small engines
,the piston head
bein g in the shape of a re-entrant cone and the cylinder headarched .
The setting of the valves in the cylinder head is,however
,
a common practice,but there are still many design ers who
adhere to the practice of setting the valves in pockets . Figs.
40 and 41 illustrate these two methods.
Fig. 40 shows the inlet and exhaust valve let into thecylinder head at an angle and designed to be operated by
Fig. 40 . M ethod of Sett ing Va lves in H ead .
mean s of a rocker arm and double-actin g cam,as previously
described in Chapter X I V. I n this design no valve cageshave been used
,the cylinder head being cast separate and
the valve seats ground into it as shown .
I n the Rathbun gas en gines both valves are let into thecylinder head in cages and operated by means of twoseparate cam mechanisms a nd levers.
I n Fig. 4 1 the inlet valve is let in,in a cage
,a s shown .
As there must be an opening at,greater in diameter than the
TH E VALVES AND PORTS
exhaust valve head , in order that the valve may be put inplace, the cage forms an easy and convenient method ofbushing this opening and making both valves readily accessible . Details of the cage and cap are shown and need nofurther explanation . The cage and cap may both be ofmalleable iron . The cap yoke should be ribbed and strength
ln ler Va lve Ca p
Fig. 41 . Method of Set t ing V a lves in Pocket .
ened as much as possible, as experien ce has shown that inhigh-compression engines a drop-forged yoke Of ordinarydimen sions will not stand the strain to which it is subj ected .
The author has in mind an instance where three or four dropforged yokes
,of differen t sections
,failed
,and it was found
in this case that a yoke of manganese bronze gave the bestsatisfaction .
The method of inserting the inlet valve in a cage may be
I NTERNAL COMBUSTI ON ENGI NES
used to good advantage in engines Operating with thesuction inlet . The only difference in design would be anincrease in height of the cap in order to allow the valve springto be placed in the cage . Fig. 42 illustrates such a valveassembled .
The spring for a suction inlet valve should be as light aspossible
,so that the slightest vacuum in the cylinder will
cause it to open . A spring of
Ilg
' in . spring wire and six coilsto the inch
,loose
,is approxi
mately correct for an outsidediameter of in . Some manufa cturers prefer a rectangularsectioned spring wire in .
by in .,winding the spring
with the long dimension parallelto the axis.
The method of placing the
Fig . 42 . Suct ion I n let valves i n separate pockets,on
V a lve in Case opposite sides of the cylinder,
ha s been used to some extent .The valves in this design are both set in a manner sim ilar tothe exhaust valve in Fig. 4 1 . An en gine with a double camshaft and two exhaust pockets has all the disadvan tages ofplacing the valves in a sin gle pocket and more
,inasmuch a s
it is impossible, owing to the in creased compression space,
to obtain a high compression without runn ing the pistonexceedingly close to the cyl inder head . Added to this
,the
increased expense of the extra cam shaft makes this methodof design undesirable .
Amethod of placin g the valves side by side,and in a single
pocket,is shown in Fig. 43 . This construction may be used
as shown,making the va lve box detached from the cylinder
,
or it may be cast integral , the last method , however , makinga much more complicated casting . The gas enters the valvebox and cylinder through the inlet valve
, a s shown by thedirection of the arrows a
, and passes out of the cylinderthrough the exhaust as indicated by the arrows b.
I N TERNAL COM BUS TI ON ENGI NES
charge,or a complete discharge Of the exhaust products, the
ports of a two— cycle engine must be proportionately muchlarger than the effective valve opening in a four-cycle engineof same size and designed to operate at the same speed .
Considering a two-cycle engine of 5- in . stroke,it is common ,
although not universal,practice so to design the piston and
cylinder that the exhaust port will be uncovered by thepiston when it ha s completed about 4 in . of the workin gstroke . The exhaust port
,then
,would be Open
,i n part,
while the piston was traveling the remaining 1 in . of theworking stroke and the first in ch of the expulsion stroke , or ,in other words
,the port would be partially open durin g
2 in . of the piston travel ; but , since the port is not whollyuncovered durin g all this time
,the average len gth of open
ing of the en tire port would be while the piston was traveling but 1 in . of it s stroke . I f the speed of the piston wereuniform durin g its complete stroke
,then the period of total
openin g of the port would be approx imately one—fifth of thetotal time required to complete one stroke
,but the piston
,
during the period of open in g of the port,is traveling at its
minimum speed and in con sequence the port is completelyopen during about one-fourth of the time required for asingle stroke of the piston .
Since the port remains open but one-fourth as long as thevalve of a four- cycle engine
,it naturally follows that its
effective area should be four times a s large . I t is a safe ruleto follow to make the ports
,both the exhaust a nd in let
,of a
two- cycle engine four times the effective area of the v alvesof a four—cycle engine having the same bore
,stroke
,and
piston speed .
The above rule is,of course
,subject to some modification
,
and it is well to bear in mind that it is much better that theports should be too large than too small
,provided that
,in
order to increase the effective area,the exhaust port does
not open too early in the working stroke,allowin g the pres
sure in the cylinder to fall and materially weakenin g thepower of the engine . The average two- cycle engine utilizesthe expansive power of the gases during about four-fifths
TH E VALVES AND PORTS
of the working stroke,the remaining one-fifth of the work
ing stroke and the first one-fifth of the compression strokebeing used to expel the products of combustion and tointroduce the fresh charge .
The exhaust port is given a positive lead over the inletport
,in gorder that the pressure in the cylin der may fall and
the products of combustion be partially expelled before theOpen ing of the inlet port allows the rush of gas into thecylinder. I f this were not done , unless the crank-case
Fig. 44 . Methods of Obta in ing ExhaustPort Lead in Two-CycleEngines.
compression were unusually high,the hot products of com
bust ion would continually fire back into the case,due to
their pressure being greater than the pressure in the cran kcase . The necessary lead is a poin t on which designersdiff er . H owever
,if the exhaust port is made to open one
third to one-fourth of its width ahead of the inlet port,it
will not be far wrong .
The lead of the exhaust is accomplished either by placingthe port higher in the cylinder than the inlet or
,a s is becom
i ng more and more the custom,by making the piston with
the edge which uncovers the exhaust lower than the edgewhich uncovers the inlet
,but making the ports even . The
l atter method is the better,since the inclined surface of the
piston allows the hot gases to discharge out Of the exhaustmore readily
,there being no sharp angle past which they must
pass as is the case in the engines using a piston with a flat
I NTERNAL COMBUSTI ON ENGI NES
head . Fig. 44 illustrates these two methods clearly,the
small arrows showing the paths Of the exhaust gases .As the ports continue for quite a distance around the
cylinder,in order to prevent the piston rings slipping into
them it is necessary that bars be cast across the opening atintervals
,thus forming
,instead of one large port
,several
smaller ports aggregating in area the required effectiveopening
,the bars
,forming a bearing surface
,preventin g the
rings slipping into the openings and catching there .
I f a third port is used for introducin g the gas into thecrank case
,its area should be equal to
,or slightly larger
than,the cylinder inlet port and its period of opening the
same . More difficulty is encountered in in troducin g a fullcharge into the crank case through a third port than inforcing the same charge into the cylinder
,because there is a
very great possibility that the case will leak and allow theva cuum to fall before the inlet port opens .
I NTERNAL COM BUST I ON ENGI NES
water off at a , as shown , the valve box is thrown into thecirculating system and the j acket water must all passthrough its j acket to reach the outlet
,whereas if the outlet a
were placed on the top of the cylinder, the valve box would
Fig. 45 . Type of Air Fig. 46 . Type of W a ter-cooledcooled Cylinder. Cy linder.
be sidetracked , so to speak, and obtain but a very limitedcircul ation .
The thickness of the cylinder walls to resist the in ternalpressure to which it is subj ected may be found by theformula
adt2
I n which t thickness in in .
, p pressure in lb . per sq.
in . (four times the compression approximately), d diameter in in .
, f the allowable tensile strength,which may
be con sidered lb . per sq. in .,and is a constant .
See Unwin’s
“ Ma ch ine Design .
TH E CYLI NDER
I n actual practice,however
,the cylinder wall is made
somewhat heavier than the above formula would seemto indicate
,in order to produce a good casting, to guard
aga inst the j acket core falling and making the wall too thinon one side
,and to allow for reboring . I f the cylinder wa l l
is cast too thin,water from the j acket is liable to pass through
into the cylinder . I n actual practice the cylinder wall,
under the water j acket,varies in thickness from t D
to t D,in which D is the diameter of the cylinder
bore . The cylinder wall t from the water j acket to theflan ge
,may be made t. Automobile manufacturers use
the lightest cylinder permissible and it is common to find acylinder of 5-in . bore with a —in . wall . Some foreignconcerns manufacture a cylinder of steel casting somewhatlighter than the above dimension
,and claim that they pro
duce uniform results in casting . Stationary en gines , ofcourse
,tend toward the use of heavier cylinders , weight
being no especial drawback but on the other hand anadvantage up to a certain limit .The depth of the water j acket j is a dimension over which
there seems to be some question , and a s a matter of fact,
unless the capacity of the circulation pump be considered,
it would be impossible to arrive at.
even an approximatevalue
,and
,after having derived a somewhat complex
formula,taking into con sideration the conductivity of the
cylinder walls,the probable heat of combustion , and the
temperature of the inlet water, we would find that it wouldnot meet all condition s . The best value for the depth ofwater j acket would be an average of the practice of differentmanufacturers . I f then a circulation pump with a regula t
ing valve in the supply pipe be used,the amount of entering
water may be nicely regulated until j ust enough waterpasses through the j acket to make the temperature of thedischarge constant at from 160 to 170 deg . fahr .
I t is good practice to make the depth of the water j acketD to D (23)
We find that the automobile engines use the larger valueof D
,due to the limited amount of circulating water carried
I N TERNAL COM BUSTION ENGI NES
and the fact of its being but partially cooled after passingthrough the j acket . The capacity of the c irculation pumpis also made small for the same reason
,in order that the
water in the radiator may have as much time a s possible inwhich to become cooled .
The outer wall of the j acket should be made no thickerthan the designer may consider n ecessary to produce a goodcasting ; in small cylinders this thickness would be relativelygreater than in engines of larger bore .
I t is probably safe to make the thickness t , D
for engines up to 5 or 6'
—ih . cyl inder bore ,and D
,or even
less,for larger cylinders . I n engines using the copper water
j acket the thickness is gen erally about
317 in . Provision must be made in thisca se for the unequal expansion of copperand iron . This may be accomplished bymean s of corrugation s . Fig. 47
,or by the
use of a n expansion j oint . The first
men tioned method is effective and probably the least expen sive to use. The
copper j acket must be calked into placewith lead and oakum
,and the head
fastened securely in place with proper
Fig . 47 . Copper gaskets. W ith the copper j acket the
W a ter Ja cket . cost of the cylinder casting - is reduced,
but the savin g in cost is more than offsetby the additional cost of the j acket and the necessary laborrequired to place it on the cylinder .
The in let water should enter the j acket at its lowest point,
and the outlet should be placed at the highest point and onthe side opposite the inlet if the engine be of the vertica ltype . I f the water leaves the j acket at a poin t lower thanthe height of water in the j acket
,a pocket is formed in
which water stagnates a nd produces steam a hot spot inthe cyl inder is the result . As the pocket so formed is usuallyon the top of the cylinder
,which point is heated the most ,
i t is liable to produce premature explosions and rapidly todeteriorate the cylinder itself .
I N TERNAL COM BUSTI ON ENGI NES
deposit in the j acket,i t is necessary that the openings be
made somewhat l arger than if the engine were cooled withwater at hydrant pressure. For the ordinary stationaryengine a water inlet equal in area to a fi-in . wrought-iron pipeis sufficient, while the outlet should equal in area a 1 -in .
pipe . The outlet pipe should,in all cases
,be the larger in
order to allow for the expansion of the water due to the heatabsorbed . Automobile and marine engines use somewhatsmaller openings, it being good practice to make the inletin . and the outlet in .
I n case there is a tendency for the water to pass directlythrough the j acket without coolin g all parts of the cylinder
Fig. 48. Type of Copper J a cketed Cylinder.
wall,baffle plates
,for deflecting the water into its proper
channel,should be cast in .
The boring of the cylinder is an operation in which extremeaccuracy is necessary
, a nd the machines must be heavyenough to prevent all chattering or vibration . The bestengines have their cylinders bored a nd ground
,as when this
is done “ it takes considerably less time to “ work in the
cylinder.” When an engine is marketed with the cylinder
finished by boring alone,it is always found to be true that,
after using the engine for a number of months , i ts powerincreases as the cylinder walls and piston rings are worn to
TH E CYLI NDER
a glassy smooth sur face,a condition productive of least
fri ction and the highest compression . Bore the cylinderto an even decimal and make the piston smaller
,as will be
described later .The cylinder illustrated in Fig. 46 is designed to be fas
tened to the frame or crank case by means of four cornerbolts , or cap screws , passing through holes d. The size ofthese bolts may be found from the formula
19db D
f nI n which
db diameter of bolt at root of threads .D diameter of cylinder.
p maximum pressure .
f allowable un it stress. (About lb . per sq . in .)number of bolts .
This same formula should be usedin determinin g the size of studs orbolts to be used in the cylinder headif it is cast separate from the cylinder proper
,as shown in Fig. 40 .
For the cylinder head,however
, fshould not be taken at more than
lb . per sq . in .,due to the fact
that in itial stress is probably in troduced in tightening the bolts hot
,
as must be done to insure a tightj oint .I f the j ump-spark ignition sys
tem be used,the spark—plug hole 6
should be tapped to receive l the
plug ; - in . pipe tap is standardfor most spark plugs .I n the cylinders shown in Figs . Fig
Cy lil
n
g
derxit ‘
gégggfig48
, and 50 the same general H ead .
dimensions may be adhered to andfor that reason they are not discussed . They are insertedin this work in order to give the designer some idea of the
I N TERNAL COM BUSTI ON ENGI NES
different forms of cylinders,and while not
,by any means ,
fully covering all conditions,they should offer suffi cient
suggestions to enable the average draftsman to design a
good working cylinder .Cylinders with the water j acket and the head cast integral
form , at best , an expensive casting,usually ranging in cost
from eight to fourteen cents a pound . I t is imperative for
[Exh a us f
Fig. 50 . Two—Cyc le Cyl inder.
this reason that suitable means be provided for properlysupporting the j acket core and thereby making the problemof castin g as Simple a s possible . Cylinders
,especially for
a utomobile engines, are often cast in pairs or even three orfour together , the patterns , in consequence , being verycomplicated .
Cylinder castings should be of the best gray iron in orderthat they may be a s tough as possible
,a nd each casting
should be tested under water pressure before leaving thefoundry and again after ma chining . One of the leadingmanufacturers of marine engines ha s made the statementtha t 20 per cent of the finished cvl inders in their shop arefound to be so faulty that they are
“scrapped .
”
I N TERNAL COM BUSTI ON ENGI NES
The steadiness of speed obtainable in an engine is notdependent on the flywheel alone . I n a direct- connectedelectric-generating un it the armature is a potent factor inSpeed regulation
, a s are also the pulleys , shafting , and othermachinery
,when belt connected . Some direct-connected
units utilize an auxil iary flywheel on the tail shaft of thedynamo . As the balancing effect of the flywheel is determined
,most largely
,from it s rim
,which is the heaviest portion
as well a s farthest from its ax is of revolution,it is common
practice to disregard the weight of the hub and arms , or web ,a s a balan cing factor
,although they exert their weight but
in lesser extent . For this reason the web or arms should bemade as light as possible to resist the strain to which theyare subj ected . I t is obvious that the proportion of idlestrokes to the actual strokes is so large
,in a gas or gasoline
en gine,even a two— cycle engine under most favorable con
dit ions gettin g on ly one impulse per revolution where asteam engine receives two
,that the weight of the flywheel
must be comparatively large .
The only method of actually calculatin g the weight of theflywheel required is to make a graphical diagram showin gthe resultan t forces of the cylinder a s acting on the crank
pin ,together with the effect of the reciprocatin g parts
,a nd
from the force diagram thus derived to determ ine the weightof the wheel which will equalize the unbalanced portions .
This method is tedious and complicated and,while serving
to show clearly the theory of the flywheel a nd acting as abasis for the empirical formulas in common use
,it ha s no
very great value to the practical designer . For a detaileddescription of this method the reader may refer to a smallpamphlet on
“ Dynam ics of Reciprocating En gines,
”a s
revised from the M ichigan Un iversi ty Techni c of 1888 . The
empirical formu las used in general practice give results thata re accurate to within a few per cen t of those which wouldbe obta ined from an actual diagram . The America n
M a chin ist gives the following formula :
TH E FLYWH EEL
I n whi ch W = weight of rim ,a = for engines
firing charge every revolution and for enginesof the four-cycle type
,
A area of piston in sq . in .
S stroke in ft.R rev . per min .
D outside diameter of wheel in ft .
The revolution s per minute or speed for which engines a redesigned being known
,the above formula
,with the constant
a as given,is good for average purposes
,but , as we have
observed,the value of
27may vary, for diff erent purposes,
0
from per cent to 3 per cent ; hen ce it is obvious that forvarying conditions we should introduce a factor to take care ofth is allowable variation . As a matter of fact the constantsa = and a = are average figures
,and
the value of a should be taken from toand from to We are t hen able tomodify the formu la to suit different conditions as follows :
aAS
s z
I n which we con sider a a s in all cases for atwo—cycle engine
,a nd a s in a ll cases for a four
cycle engine , a nd a: is a constant , for d ifferent condit ions, a sshown in the followin g table :
Portab le enginesPumping a nd Ordina ry use
Driving ma chine tools .
Driving looms or texti le ma ch inery , et c . .
Dr iving electric ma ch inery , et c
Driving cotton spinn ing, et c
Automobile and motor-boat en gines permit, and theirusage demands, a lighter flywheel than for stationary purposes
,and in these. engines we may safely use a value of a:
a s low a s for automobiles and for motor boats .
H aving determined the weight of rim necessary , its con tentin cubi c inches is determined by dividing the weight by
I N TERNAL COM BUSTION ENGI NES
which is the approximate weight, per cubic inch , for
cast iron . Now assume some thickness for the rim —~thethickne ss should , wherever possible ,
be its minor dimension,
so that its center of gravity may be a s far from its center ofrotation as possible . The mean diameter of the rim will
,
then, be equal to its outside diameter
plus its in side diameter,divided by
two . The mean diameter and circumferen ce a nd thickness of rim beingknown
,it s width must be Such that
it m ay con tain the number of cubicinches
,above determ ined
,necessa ry
to produce the required weight . Use
the nearest lg in . above the calculatedwidth .
Most flywheels used on stationaryFig. 5 1 . Flywheel . engines are made with six spokes
,
while the maj ority of automobile and
marine engines use the webbed pattern . Fig. 51 is a. spokedwheel showing the average dimen sion s in good use . I n the
equation s, 3 cran k-shaft diameter .
Then
d 3 (approximately).
b 3 (approximately).
The dimen sion d may be computed from Unwin’
s
formul a,in which
D diameter of pulley in in .
B breadth of rim in in .
n n umber of arms,* thus :
d=o.6337 (Single belt)
d=0 .798 (Double belt)
For a flywheel transmit t ing no power a sma l ler va lue may be used .
I N TERNAL COM BUSTI ON ENGI NES
Table X Should be used for all fixed work wherein the keynot on ly drives
,but also holds the parts again st endwise
motion . These keys are tapered and bear all over . ”
A webbed flywheel,Fig. 53 , should
use a comparatively small thickness fori ts web . This thickness varies
,in auto
mobile and marine engines,from
to in .,and need not be computed
,
as,in small wheels
,a web as thin a s
could safely be poured would probablybe as strong as six spokes . I n mostcases the thi ckness of the web will befound to be approx imately one- fourth
Fig. 53 . Flywheel . the shaft diameter . I n a Spoked wheel,
always fillet the Spokes well where theyj oin the rim
,a s the large mass of metal may
,otherwise
,
cause a crack to form at the j oin t.The flywheel must be accurately turned and balanced .
I f,after finishing
,one side is found to be heavier tha n the
other,holes a re drilled in the heavy side in order to overcome
this difficulty . Some manufacturers go so fa r a s to fin ishthe face of the web in a webbed wheel
,but except for finest
machines this will be found unnecessary .
John Richards, in Ca ssi er’s M aga zi ne.
CHAPTER XVI I .
TH E FRAME.
TH E gas-engine frame serves a twofold purpose ; it rigidlysupports the cran k shaft and cylinder and resists theirtendency to separate from the force of the explosion
,and
,by
its inertia,absorbs a part of the unbalanced forces in the
engine and transmits to the foundation proper the remainderof these forces .The advantage of having a comparatively heavy foundation
on which to rest the engine frame is due to the fact that itis undesirable a s well as impractical to make the frameheavy enough in itself to effectively absorb the vibrations,as such a large mass of metal would be exceedingly cumbersome to handle . I n view of this fact it is customary andgood practice to make the engine frame of cast iron a nd a s
heavy a s may be,at the same time preserving its mechanical
appearance a nd l imiting its weight to such a degree as topermit of its being handled in the machining process . For
these reasons it is not customary for the engine designer tocompute any part of the frame
,with the exception of the
bolts,as it is assumed that it will be amply strong to resist
a ny shock to which it may be subj ected .
The formula to be used for computing the bolts that fastenthe cylinder to the frame is given in Chapter XV, page 135,and needs no further discussion .
H orizon ta l Engines. Fig. 54 will serve to Show thegeneral character of design of frame as used in gas engines.Nearly every manufacturer will be found to modify his frameto some extent
,and it would be useless to go more fully into
their special characteristi cs .I n the frame shown in Fig. 54
,the bearings are set at a
45° angle , in order that most of the thrust of the crank143
I N TERNAL COM BUSTI ON ENGI NES
shaft may be against the frame instead of against the bearingstuds . While an angle of 45° will not be such that themax imum forward thrust will be received by the frame , agreater angle would not be desirable
,as it would bring the
j oint in the brasses too near the bottom of the bearing. I n
Fig. 54 . Frame for H or izon ta l Engine .
the design shown brass liners,a,a re used to take up the play
in the Shaft . Four or five of these liners,made of light-gage
brass,Should be used . They are cut to conform to the
shape of the cap and to clear the shaft 315 in . Two or threemay be made of 30-gage B . S. and a like number of about40-gage .
A good rule to follow for the size of the bearing studs isto make their diameter
,at the base of thread
,the
diameter of the shaft .Some manufacturers carry a proj ection of the frame out
under the cylinder, on which lugs are cast, and bolts are
passed through corresponding lugs on the cylinder . I t isdoubtful i f this is good design
,a s the unequal expansion of
the cylinder and frame w i ll either loosen these bolts orcause a distortion of the cyl inder alignment .The air supply for this type of en gin e is generally drawn
from the hollow base,in order to make the operation as
quiet as possible and to secure a warm dry suction .
The Ver tica l Engine— I n this type of engine , the frame
usually becomes a crank case , completely housing the shaftand cam gearing . There are exceptions
,of course
,in
cheaply constructed single- cylinder engines for generalpurposes
,but the modern high-class vertical engine as a
146 I N TERNAL COM BUST I ON ENGI NES
Access is had to the interior of the case by means of platese and f. Plate f, a s shown ,
covers the cam and valve , aswell as the starting mechanism
,if used ; the push-rod bushing
being inserted at 9, above the cam shaft, Shown at h. The
bearing studs i,as Shown
,require that the side plate f be
removed in order to reach them . A better design wouldbe to use a long cap screw C,
with a shoulder to act on thebearing cap
,and continue it through the top of the crank
case,where it wou ld be easily accessible .
Manufacturers of both vertical and horizontal enginesoften make a sub-base or bed plate . Especially is this truein the manufacture of direct-connected un its
,the sub-base
forming an accurately machined bed,on to which the engine
and dynamo are fastened,and their perfect alignment assured .
CHAPTER XV I I I .
ENGI NE FOUNDATIONS.
TH E engine foundations are almost always bui lt by theowner or contractor from drawings furnished by the enginebuilder . The drawings consist of accurately dimensionedplates Showing location of the foundation bolts
,and also a
drawin g of a template to be made for locating these boltsin the foundation .
Without a good foundation an engine is bound to givetrouble
,sooner or later
,from settling or from the engine
workin g loose from the foundation and getting out of line .
The foundation of the engine may be considered as apart of the bed plate or frame
,in that it Should be given
suffi cient mass to absorb,by its inertia
,the eff ect of the
suddenly applied cylinder or crank-pin forces not absorbedby the bed plate and frame . I t should get a good bearingon solid ground
,the quality of the soi l governing
,to a large
extent,the depth of foundation necessary . Under no condi
tions Should an engine be fastened directly to the floor of theengine room
,except for a temporary j ob
,and furthermore
,
the floor of the building should be absolutely independentof the foundation
,in order to prevent the transmission of
vibrations to other parts of the bu ildin g,ex cept a s may be
tran smitted by the concussion of the air or the vibrationof the ground . When it becomes necessary to reduce thevibration still further
,a layer of some insulating material
,
2 or 3 in . thick,should be placed between the foundation
and the surroundin g soil . A substan ce such as deadeningfelt
,horsehair felt
,or cork may be used and a layer 8 or 10 in .
thick placed next to the foundation . Cork, however ,Should not be used below the foundation owing to its tendency to absorb water and swell . A bed of dry sand as abottom has been used with good results .
147
I NTERNAL COM BUSTI ON ENGI NES
I f it becomes necessary to install an engine on an upperfloor where a bearing on the ground cannot be obtained , acrib of heavy timbers should be built below the floor tosupport a foundation of con crete . This foundation willabsorb
,by its inertia
,a great deal of the vibration .
An engine foundation may be made of concrete , brick laidin cement
,or stone . I f of concrete
,the mixture should be
one part good Portland cement,two parts sharp clean sand ,
and four parts broken stone small enough to pass through a2-in . ring . The concrete Should be l aid in layers of not morethan 6 in .
,each layer bein g thoroughly tamped before the
next layer is put in . Foundations made of bri ck should belaid in cement mortar composed of one part Portland cementto two parts clean sharp sand . The brick used should be hardburned foundation brick . A very good foundation, up tothe floor line
,may be made of brickbats laid in the above
men tioned mortar,care being taken to fill all voids . This
foundation may then be built up of concrete or brick,in the
ordinary way,to receive the en gine . A brick foundation
should have a cap of l imestone or granite,or a cement cap
1 ft . in thickness may be put on . Stone foundation s shouldbe laid up in cement mortar , and in regular steps , being surethat the stones have a good level bearing
,so as to prevent
any tendency to slide . A concrete foundation should be
given a batter of about 2 or 3 in . to the foot,and a brick or
stone foundation should follow,in general
,the same outline .
A concrete foundation of standard design is shown in Fig. 56 .
I f,as frequently is the case
,the en gine is a direct- con
nected unit , that is, an engine directly connected to a dynamo ,then the foundation should be so designed a s to take in thedynamo and outboard bearings .
The weight of an engine foundation is seldom computed ,a s differently designed engines are subj ect to diff erentdegrees of Shock
,from their unbalanced forces
,and the
weight of their frame and bed plate may vary to such anextent that any formula would be applicable to but fewcases were the surrounding conditions the same . With thecondition of the soil varying, it would be impossible to
I N TERNAL COM BUST I ON ENGI NES
tion bolts Should be ample in size and of suffi cient numberto accomplish this result .I n layin g out a foundation proceed as follows : Dig down
to good firm ground,or hard pan
,a nd level this bottom off
carefully ; in case the ground is marshy it may be n ecessaryto use piling
,but this is unusual except in some localities .
I f it is found to be necessary,piles may be driven on 30- in .
centers and cut 0 5 at ground-water level . The concreteshould then be filled in between the piles a nd the foundationproper started up from that point . H aving secured a goodbed , locate accurately the position of the engine , and placethe bolt template in such a position over the excavationthat the foundation bolts may be suspended
,by their nuts
,
through the holes in the template,in the position which they
will assume when the en gine is set . The bolts,before being
suspended from the template,should have a piece of iron
pipe slipped over them,Fig. 56
,A. This pipe should be
large enough to allow a slight movemen t of the bolt in settin gthe engine . The pipe is supported at its lower end by the
bolt an chor , and its upper end Should come just below thetop of the finished foundation . The foundation may nowbe built up around these pipes . When the top course
,either
brick or concrete,is la id
,be sure that the surface whi ch is to
receive the bed plate is perfectly level . This levelingshould be done with a Y level and must be accura te . Afterthe engine ha s been set the pipes should be filled with cementmortar . A good size of pipe to use is a 2- in . standard ga spipe.
Foundations after being built should be allowed to set
for at least a month,unless circumstances deman d them to
be used sooner,I n which case the concrete should be made
as dry a s possible to secure a good mixture .
I t Should not be assumed that it is common practice toinsulate the foundation with a shock absorbent
,as above
described . This is done on ly when the location of the en gin eis such as to make it desirable that every precaution be usedto prevent a nuisan ce .
CHAPTER XI X .
TH E CRANK SHAFT AND RECI PROCAT I NG PARTS.
TH E crank shaft, connecting rod , and piston , together withtheir respective parts
,constitute the rotative and recipro
cating parts of a gas engine . The crank shaft, as previouslydescribed, carries the flywheel
,and balance weights
,if
used .
W ith but few exceptions,gas-engine pistons are of the
trunk variety . The reason for this very general use of thetrunk piston is that if the engines were provided with a pistonrod and made double-acting
,as is the practice in steam
engines,the rod end of the piston would be subj ected to as
much heat as the head end,and the rod , becoming heated ,
would expand and cut into the stuffing-box. Except inthe case of two-cycle engines
,in which the crank case is
used as a primary compression chamber,there is no necessity
of closing the forward end of the cylinder,the piston itself
being made long enough to act as a guide without the useof a piston rod and cross-head . A few engines have beendesigned which were double-acting
,the rod and packing box
being cooled by means of water made to circu late aboutthem
,but as these types may be considered in the “ freak ”
class no special discussion will be given them .
The crank shaft of a gas engine , while it serves in thesame capacity as the shaft of a steam engine, must be madevery much stronger
,Since it is subj ected not only to the
strain due to the conversion and transm ission of the powerreceived but
,at the beginning of the working stroke, it is
subj ected to six times the average power of the engine .
I n determining the size of crank shaft necessary , thediameter of the piston and the maximum cylinder pressureare taken as the basic conditions for the computation .
1 51
I NTERNAL COMBUSTI ON ENGI NES
Let D, the diameter of the shaft in in .
D; the diameter of the cylinder in in .
m the maximum cylinder pressure .
For a drop-forged steel shaft the average practice is tomake
D3 Dc VP",
For malleable or wrought iron,
D. D. f/Pm (32h)
These values will be found to produce results closely inkeeping with average practice when the stroke is equal to
Dc . E. W . Roberts gives the following formulas forcrank Shafts where the ratios of the stroke L to the cylinderbore is other than 1 .
For wrought iron ,D. \
3/PmLDc2
For steel ,
(33b)
As he remarks, the formul as wil l be found to give resultslarger than the average practice
,but for stationary engines
this is a good fault . For marin e and automobile enginesthe values will be found to be much larger than averagepractice requires .The length of the cra nk- Shaft bearing necessary is readi ly
determined,once the diameter is known
,from the following
formula :Dc
2 P
10 18 D,
which I length of the bearing .
De cyl inder diameter , as above .
D8 Shaft diameter, as determined .
P Mean effective pressure .
I n practice , however , the crank- Shaft bearings are usuallymade much longer than the value which would be obtainedby the use of this formula . A good rule to follow is to makeI : B a .
I N TERNAL COM BUSTI ON ENGI NES
and hence to be added to the weight of the pin when computing the balance weights necessary . The usual locationfor the balance weights is to place them one on either crankthrow and on the side of the shaft opposite to the pin . Whenlocated in this manner they should be securely fastened inplace” with pins
,and a perfect bearing secured by babbitting
the j oint or by machining both the weights and cran k armsto a perfect and tight fit. The slightest play in theseweights will increase very rapidly as the engin e is run a nd
will cause no end of trouble . When the weights are babbitteda groove in the weight and crank arm
,as indicated in Fig. 58
,
retains the babbitt and secures the weight firmly, and whenthe rivet and cap screw are in place
,as Shown
,a good j ob
is secured . When the j oint is machined the rivet and capscrew are used in the same manner . Some engines carry thebalan ce weights on the flywheel and on the side oppositethe crank pin . When located in this latter position theymay be much smaller
,owin g to their increased radius of
rotation , but their in creased distance from the center offorce introduces a greater amount of wear on the enginebearings.
I n balan cing a single-throw crank the first thing to determ ine is the weight
,con sidered centered in the crank pin
,to
be balanced by the counter weight . As previously men
t ioned,two- thirds of the connecting rod is considered a s
rotative . Then the weight to be balanced would be equalto Wr+Wp at .
I n which
W . Weight of connecting rod,brasses
,and bolts
,etc .
W,, Weight of cran k pin a nd that part of the arms con cen
tric to the pin .
Weight of the remainder of the crank arms con sideredcen tered in the cran k pin .
The values W, and Wp are easily determined , but thevalue of £23, accurately computed , would possibly require arather complicated equation . H ence it is customary todetermine the moment of W, and Wp and then to balance
TH E CRANK SH AFT AND RECI PROCATI NG PARTS 1 55
them with a weight 10 per cent in excess of the calculatedvalue .
Then (W, l moment of balance weightsnecessary (I being the distance from center of Shaft to centerof pin).Now taking the distance of the center of gravity from
the crank- shaft center to be I the weight necessary tobalan ce the crank and rod would be
H alf of this amount in each weight with their centers ofgravity l
ldistant from the shaft would nearly produce a
balance . To determine the value of I I it is customary tocut and try . Lay out the crank and balance weights , asshown in Fig. 58 . Then cut out a template of cardboard
a s l fi d l ‘d fed
Fig. 58 . Ba la nced Crank .
the exact shape of the weight as drawn ; find the center ofgravity of this template by ba lancing it on a pencil point andthen determine the distance of this point from the center ofthe shaft when the weight is in place . Usin g this value inequation (36) gives the total amoun t of balance necessary ,a nd half of this amount is to be placed in each weight . I f
the value of l, ,as determined
,is found to be so small that
,
I n order to make the two weights equal to W,it is necessary
to make them thicker than the crank arms,then a larger
template should be laid out and a larger value for ll deter
I NTERNAL COMBUSTI ON ENGI NES
mined,from which a smaller value for W will be found .
Crank shafts with more than one throw are not balanced,
since the different rods and arms balance each other ; neitheris it customary to balance the crank of the cheaper industrialengines .The more expensive engines have their shafts turned and
then ground in order to insure a smooth bearin g .
The shaft and connecting—rod bushings are made of somegood bearing metal
,turned and fin ished to the proper size
,
after which they should be “ scraped in and the oil groovescut to carry the lubrican t . Fig. 59 illustrates such a bushin g
Fig. 59. Bear ing B ron ze.
with the oil grooves cut in the customary manner . Thereare many good bearin g metals on the market ; some manufa cturers preferring a good grade of babbitt metal
,while
others use a bearing bronze one of the best being phosphor-bronze. There is very little choice between the two ,although the babbitt metal
,a s a rule
,requires less lubrican t .
One of the very best bearin gs is made of bron ze withbabbitt poured in
,in order to facilitate the lubrication .
on Rings .
I n order to secure good lubrication of the crank pin in
closed crank-case en gines when splash lubrication is used ,
means must be provided for conducting the oil contained inthe crank case to the pin bushing. This is often a ccom
plished by millin g a slot in the end of the rod cap a nd throughthe bushing ; the rod end passin g through the oil during it srevolution
,enough is picked up and works into t he pin
through the slot to keep the bearing fairly well lubricated .
1 58 I N TERNAL COM BUSTI ON ENGI NES
held in place by studs and castellated lock nuts . As thebearing becomes worn , the cap is removed and a pair ofliners taken out . Fig. 6 1 illustrates a rod commonly in use
on engines of the marine and automobile type . I t is knownas the “
I” section rod and is made of drop- forged steel
, the
web and flanges being formed in the dies . The cap and
Rod a n d and fo be forgedm one p i ec e
Figs. 60 and 61 . Connecting Rods.
l iners are shown in place . I n both rods illustrated the
general dimen sion s — ba sed on good average practicea re given in terms of the crank—pin diameter . Ken t givesthe following Simple formula for rectangular section con
nect ing rods :
t D V" + 0 6
I n which t Thickness of rod in in .
D Diameter of cyl inder as given above .
P Maximum pressure in cylinder.
H aving determined the value for t,the depth of the rod
at the crank end is made equal to t and the depth of thewrist pin end t . The length of connecting rod betweencenters varies between the ratios 2 1 and 3 I
,in which
.1_i_s equal to the length of stroke . The above formula is
based on the required section at the surge point and , inconsequence, surplus metal is found at the crank end .
TH E CRANK SH AFT AND RECI PROCATI NG PARTS 1 59
The Piston , W rist Pin, and Piston Rings.
The piston of a gas engine is connected to the rod bymeans of the “ wrist pin ” bearing . There are two methodsemployed for forming this bearing . I n the first and theone most commonly in use — the wrist pin is locked fast inthe piston , the rod working on it a s a bearing. I n the secondmethod the pin is locked firmly in the rod
,the bearing for
the pin being in the piston itself . Fig. illustrates the
In .9 9 9n, -1
Fig. 62. Piston and W rist Pin .
general form of piston in use for four-cycle engines . The
weight of the piston,as well as the rod
,Should be as
light as consistent with the strength required . The headand that part of the shell under the rings are made heavierthan the remainder of the piston
,as indicated in the drawing .
The sizes,as indicated
,are in terms of the crank—pin diameter
and length of stroke . The connecting rod and bushing areshown in dotted lines
,the wrist-pin bushing being pressed
into the rod end and allowed to proj ect,as shown
,to secure
the end bearing necessary . Finish marks indicate that partof the piston to be machined . The set screw and lock nut
I NTERNAL COMBUSTI ON ENGI NES
are shown in place . This screw should be made as tight afit as possible in order to prevent its working loose . The
wrist pin is shown beside the piston drawing, but while itsdiameter is given the same as the rod bushing
,it should have
a running fit of from to in .,depending on its size .
I t shou ld fit in the piston as tightly as may be withoutdanger of cracking the shell .The outer diameter of the piston should taper Slightly
from the head toward the crank end . The reason for thisis very apparent from the fact that the head end , beingsubj ected to the greatest heat and containing the greatest
. 0 0 5 0g
Fig. 63 . P iston Ring.
amount of metal, expands much more rapidly than the crankend and is liable to stick and cut the cylinder . Obviouslythis taper should increase with the piston diameter. A
good average rule is to make the head diameter equal toD, and the crank end equal to Dc . The length
of the piston should approximate the stroke with thewrist pin located practically at its center .The best pistons are usually provided with four eccentri c
rings,although some of the cheaper ones are made with three
or even two rings . The ring at the crank end serves in thecapacity of an oil ring and also provides a bearing for thatend of the piston . The piston ring is shown in Fig. 63 andits dimensions are given in terms of the cylinder diameter .The rings should be made of the best grade of gray cast iron
CHAPTER XX .
GOVERN ING DEVI CES.
TH ERE are four principal methods used for governing thespeed of a n internal-combustion engin e , V iz . : (1)By throttlin gthe charge ; (2) By cuttin g off the supply during one or morecomplete cycles
,commonly known a s hit or miss ; (3) By
keeping the exhaust valve open or closed during one or morestrokes ; (4) By interrupting the spark if electrical ignitionis used .
Greatest regularity of operation but poorest economy isobtained by throttlin g the charge
, but it is undoubtedly themost satisfactory when
,owing to the nature of work per
formed by the engine,the speed variation must be slight as
in the driving of electrical machinery . A combination ofthe two methods ha s been used in which the charge isthrottled up to certain limits a nd then cut Off . See Fig. 65
,
the Otto electric— l ight govern or . The properties of explosivemixtures of gas and air will Show the reason for the throttlingmethod bein g uneconomical . “
The l imits of change allowable in the proportion s of gaseous explosive mixtures arevery narrow
,the ga s present ranging from i, to 1
15 of the
total volume . A mixture containing 4 of its volume of coalgas in air has just sufficien t oxygen to burn it and no more ;a ny further increase of gas will pass away unburned
,there
bein g insuffi cient oxygen present for its combustion .
CLERK,
“The Ga s and Oil Engine
,page 226 .
A mixture of air and gas contain in g less than Ilg
' of itsvolume of gas loses its inflamm ability altogether . I t followsthat governors acting on the ga s supply should be set togovern between these limits only .
The mixture itself may be throttled without altering the1 62
GOVERN I NG DEVI CES
ratio of gas to air,thus producing instead of an inferior
mixture a greater or less amount of the same quality ofmixture . This method reduces the force of explosion andat the same time maintains a uniformly explosive charge .
Keeping the exhaust valve open is of course wasteful offuel and less economical
,as is also stopping the spark . Keep
ing the exhaust valve closed produces a mixture of unburnedproducts in the cylinder which must be completely exhaustedbefore the cycle of operation s can again be taken up .
The“ hit or miss ” type in which the fuel supply is entirely
cut off for one or more complete cycles is of course the mo st
Fig. 65. The Ot to Electric L ight Governor.
economical method in use, as the charge is never varied andthere is either an explosion of full force or none at all . I t
is apparent,however
,that “ hit or miss ” govern in g admits of
a greater range of speed variation ,although it is possible
with a properly adjusted governor to obtain regulationwithin four or five per cent as against two per cent with thethrottling governor .The mechanisms used to control the different governing
devices are of two general designs,viz (1) The centrifugal
I N TERNAL COM BUSTI ON ENGI NES
governor,whi ch may be used for any one of the above-men
t ioned methods of governing; (2) The inertia governorwhich is applicable only to hit or miss ” governing. A
modification of one of these two methods , more or lessinvolved
,is the principle of every governing device .
The centrifugal governor is essentially a revolving pendulum with inclined arms, advantage being taken of thetendency of these arms to revolve in the same plane as theirpoin ts of support . Figs. 66
,A and B
,illustrate the principle
of the cen tri fugal governor . At A the two fly balls a and bare connected
,a s shown
,to collars c a nd d. Collar c is
fastened to Shaft e,while collar d
,carrying the swivel f to
which bell crank h is fastened,is free to move . As shaft e is
revolved the ba lls a a nd b separate as indicated by thearrows and in doing so l ift the collar d. This movement isresisted by spring 3 , the tension of which may be adj usted bycollar i . When the speed reaches a certain l imit collar dwill be raised until bell crank h throws the push rod j awayfrom the valve stem as shown . Valve h then remainsclosed until the Speed decreases and j returns to its placebelow the valve stem . This device is of the “ hit or miss ”
variety .
To determine the lift of the governor balls for any givenspeed their weight and the resistan ce of the spring 3 beingknown we have , for a simple pendulum ,
the ratio
centrifugal force
Substituting for 19 its value 2 am ,
I N TERNAL COM BUSTI ON ENGI NES
I n whichh distance from plane of center line of ball s to the
plane of their points of suspension .
r radius of circle described by balls in ft .
9
E velocity of center of balls in ft . per sec . 2 am .
n number of rev . per sec .
N number of rev . per min .
The simple fly-ball governor is not isochronous ; that is itdoes not revolve at a uniform speed
,since the speed changes
with the angle of the arms. To remedy this defect thegovernor is loaded by means of a spring or weight .
For the loaded governor we have the value of the cen t rifugal force
,due to the weight of the balls
,unchanged
, but
the value of the weight now becomes equal to the combinedweight of the balls and the value of the spring load considered directly below the center of gravity of the balls.
Let I : len gth of arm 7; from the poin t of suspension tothe center of gravity of the ball , and let the length of thesuspending link I 1 x be the length of the arm y from itspoint of suspension to its poin t of attachmen t to the ball ;11) weight of one ball ; half the value of the sprin gload in pounds ; h= the height from plane of revolution ofba lls to point of suspension of y; then
N 2
the ra tio of half the Spring load to the weight of one ball
bein g 3 " and the relation of its suspension point to the center11)
(39)
of gravity of the ball bein g as If . Unless the l inks y and a:
a re equal in length,this relation will not hold true. The lift
being determined,the bell cran k or other mechanism con
nect ing the governor to the valves is ea sily laid out .
Sin ce a governor is design ed and set for a certain speedit must ' be run at this n ormal Speed regardless of that ofthe engine . I f the engine speed is increased or diminished
,
the governor must be geared down or up as the case may be .
GOVERN I NG DEVI CES
At B is shown another application of the centrifugal balltype of governor . The l ift of the balls
,a and b, may be
determined by means of another applica tion of the ratio
h Weightr Cent . Force
to becombined with the weight of the balls .)
Simplifying,
h
I n which n The rev . per sec .W. H alf the spring load .
W The weight of one ball .a The angle of the arm I .
I f we allow N to represent the rev . per min .,then the
formula will be h 2
39
324
This formula for small movements of the balls is sufli
cien t ly accurate for all purposes. I n this style of governor,
referring to the figure,we have two levers acting on f a s a
fulcrum and exerting their lifting force at c and d. The
bell crank g rides the cam roller h off or onto the cam i asthe speed varies
,and valve j, actuated by lever It , opens
or remains closed as the case may be . I t would be possible,
by making the lift of the cam variable as shown at I,to
produce a throttling eff ect at the valve . Throttl ing with apoppet valve, however, produces an abnormal amount ofwire drawing of the charge . A better arrangement is to usean independent throttle valve in the gas ma in . Fig. 67
represents such a throttling valve with governor connection .
AS the balls a and b move outward the bell crank c moveswith the collar d and the valve V is shoved in the directionof the arrow e. The openings f in the valve graduallymove across the openings g in the shell h, reducing the
I NTERNAL COM BUSTI ON ENGI NES
effective port openings into the annular space i,and throt
tling the mixture .
The principle of the inertia governor is illustra ted inFig. 68 . The weight a is connected to bell crank b andpivoted to the push rod c . The tendency of the weightto turn the bell crank in the direction of the arrow is
Fig. 67 . Cen tri fuga l Governor and Throt tle.
resisted by the spring d, but a s the push rod is movedrapidly the inertia of the weight partially overcomes thisresistan ce and the engaging parts of the valve gear areseparated when the speed reaches a certain limit . The
strength of the spring requ ired is determined by means ofthe following formula :
Wa l
I n which
W Capacity of the spring in lb .
Wa Weight of a in lb .
I1
Lever arm of a about c .
V Velocity attained by push rod b in ft . per sec .
9 ft . per see .
I n the above formula the weight of the rod carrying thependulum is neglected . To be accurate the resultantmoment of it and the weight should be used , but by threading the weight onto the rod , as shown , adjustment may be
CHAPTER XXI .
I GNI TI ON.
TH ERE are three general methods employed for securingthe ignition of the compressed charge in a gas-engine cylinder :
(1) By means of an electrical spark ; (2) By means of amechanically operated flame or heated surface ; (3) Byauto-ignition , in which heat suffi cient to ignite the charge isproduced either by means of the compression alone or bymeans of the combined eff ect of compression and residualheat .Electrical ignition devices are most extensively used , and
these may be subdivided into two classes : (1) The j umpspark system ; (2) The make-and-break system .
Jump-Spark ignition , as the name implies, consists incausing an induced current of high potential to spark betweentwo metalli c points conven iently placed in the compressionspace of the cylinder , the spark thus produced igniting thecompressed charge . To produce the spark at the properinstant in the cylinder a make—and—break contact must beplaced somewhere in the electrical circu it
,this contact being
so operated by mean s of the cran k or cam shaft that atsome point in every operating cycle the circuit will be closedand an e .m . f . generated suffi cient in value to cause a sparkto j ump across the gap . This make-a nd-break device isgenerally known as the commutator or spark advancer
,the
latter name having its origin from the fact that the commutator is so arranged that the en gine Operator, by meansof a suitable lever, is able to change the point at which thecharge is ignited so that it may correspond to the speed atwhich the engine is run n in g. This statement may be somewhat confusing inasmuch a s the speed regulation is obtained
,
to a very large extent, by advan cing or retarding the time170
I GN I TI ON
of ignition . Nevertheless the point in the cycle mustcorrespond to the speed at which the engine is operating
,
at any particular instant, in order to secure smooth running.
I n starting the engine the spark is set at a point slightlybeyond dead center and in the expa nsion stroke. As soon asthe explosions commence to occur regularly the spark isgradually advanced past dead center and in to the compressionstroke, thus giving the engine time to gather speed . I f thespark is rapidly thrown over dead center into the compressionstroke the engine will either stop or pound badly until i tatta ins suffi cient speed to carry itself from the point ofignition up to dead center before the burning gases reachtheir maximum pressure .The make-and-break system of electrical ignition consists
in causing,by mechanical means , two points or electrodes
located in the compression space to close and then open theelectrical circuit. This may be accomplished by causing thepoints to rub together and then separate, producing what isknown as the “ wipe spark, or by forcing the points togetherand causing them to separate by means of a spring
,some
times called the “ hammer break .
” When the points breakcontact an intensely hot spark or a re is produced
,due to the
inertia of the electri c circuit producing, momentarily, a veryhigh potential .The make-and-break Spark is much hotter than the j ump
spark and on reasonably slow Speed engines is the mostsatisfactory form of electrical ignition, it being almostcertain tha t, if the points are in good condition, a spark hotenough to ignite the charge wil l be produced at everycontact . On the other hand, the points are subj ect tomuch wear
,especially with the wipe spark, and couse
quently deteriorate qui te rapidl y. I f platinum alloy isused for the points they are usually quite expensive, and thenecessity of replacing them is troublesome as well . Withinthe past few years other alloys , which it is claimed by theirmanufacturers give better satisfaction , have been placedon the ma rket. “
Casa lloy” or “ meteor-wire ” is one of
these substitutes .
I N TERNAL COM BUSTION ENGI NES
The points of a make-and—break ignition may, with propercare and a current of proper strength
,be mad e to last a long
time . I f the electri c pressure is too low,unsatisfactory
ignition will result, while , on the other hand , should thepressure be too high the plugs wil l require adjustment orrenewal in a very Short time . W ith a battery Of low internalresistance
,as a storage cell
,the difference of potential at the
termin als should be much less than with cells of high internalresistance. I n order that a sufficiently large number Of prima ry cells may be carried for all emergenc ies and to allow for
Fig . 69 . The Non -I nduc t ive Resistan ce and t he Condenser .
depletion,the destructive action at the points may be reduced
by placing a conden ser in parallel with the poin ts or byintroduction in to the circu it of a non - inductive resistance .
Fig. 69 illustrates clea r lv these two methods . A noninductive resistance thrown in to the circuit in series causesa fall in potential w ithout producing a ny unbalan cedelectromagnetic action wh i ch would affect the sparkin g coil .Such a resistan ce coil is made by doubling a wire , placingthe closed end on a bobbin of wood or other non-magnet iz
able substan ce , a nd w indin g the wire about the bobbin a s
indicated at A,Fig. 69 . The electromagnetic action in one
wire is thus n eutralized by an exactly sim ilar and equalaction in the other . I t is a pparen t that i f this method isused a suitable resistan ce box with varying resistan ces toaccommodate itself to varying condition s must be provided .
I f the condenser method is used and a condenser large enoughto meet all requirements is prov ided , no further adjustmentor attention will be n ecessary .
1 74 I NTERNAL COM BUSTI ON ENGI NES
at p ( the condenser is not essential and many coils aremade and operate successfully without it); G the primarycoil of heavy wire carrying battery current ; H the secondarycoil of heavily insulated fine wire ; I the spark plug, to oneterminal of which the secondary coil is connected, the otherterminal being connected through the ground to the oppositeend of the secondary coil
,and J the iron core of very soft
annealed wire.Fig. 7 1 shows , diagrammatically , the wiring connections
for a four-cylinder engine with j ump-spark ignition. The
Fig. 71 . W iri ng Connec t ions for Four Cyl inders.
coil consists of four unit coil s identical with the single coilpreviously described a nd all using the same ground connection . By means of the commutator the primary currentis made to pass alternately through the units 3 , and 4,producing in each
,in turn
,a high—potential current which ‘
fires the charge in the cylinder to which it is connected .
The system as shown is wired for two sets of batteries,
B and B By mean s of the three—point switch A either setof coils may be thrown into the primary circuit
,so that if
one is found to produce an insuffi cient spark the other setmay be used .
The coil is a very important element in the successfulOperation of a j ump—spark ignition system . Numerousmakes of coils are marketed
,many of which are apparently
very reasonable in cost, but it is doubtful if it is advisable ,
in any case,to purchase a coil and consider its low cost as
I GN I TI ON 1
a recommendation . The coils which give best satisfactionare rather expensive
,but it will be found in the long run that
they are most economical . To give good results,especially
on high-speed en gines,a coil must be fast
,that is
,it must
charge and discharge very rapidly . I n order to accomplishthis result the core must be of best annealed iron wire andthe coils well insulated
,especially the high-tension windings .
With a sluggish hungry coil the best of engines will give poorresults .
Every coil is rated to give a certain length of spark,but
the points of the spark plug must be set very much closer thanthis maximum
,Since it requires much greater pressure to
spark across a gap under high compression than in a vacuumor at atmospheric pressure . The strength of current mustalso govern the spark gap . For average practice 1 11; in . isabout right for this gap .
The vibrator coil is not suitable for a make-and-breakignition system . The vibrating high-potential curren t willcause the ignition points to spark as they come together
,or a
Spark may be produced at some equally undesirable poin tin the cycle . The ordinary single induction coil , or booster ,
”
produces the best results, unless a sparking dynamo is used , inwhich case no coil is necessary
,the current from the dynamo
being suffi cient to produce a good spark. I n using the
make-and—break system of ignition no commutator is used,a s the current is in terrupted at the ign ition points and thetim ing of the ignition is secured by changing the point inthe cycle at which they separate .
I n the jump-spark ign ition system the spark is producedwhen the circuit is closed
,this result being accomplished by
the vibrator interrupting the current and producing thenecessary high potential many times a second , while in themake—and-break system
,this high potential being produced
only once,a s the points separate , it is at that point that the
spark occurs . Fig. 72 illustrates a make—and-break systemin which the points p and p, are forced apart suddenlywithout the wiping effect . I n the figure , A is a quickreturn cam which in the position shown is about to trip
I N TERNAL COM BUSTION ENGI NES
push rod B . On this push rod is located the stop andadjustment collar C which impinges against the fiberwasher D when the circuit is broken . The flat compen
sating spring E takes up the motion of the push rod B afterthe points p and p , are in contact . Coil spring acting
Fig. 72 . H ammer-Break I gn iter .
on lever G,as shown
,throws the sparking points sharply
apart as the point of the cam A leaves the push rod . The
plug H,which carries the battery term inal must be con
structed of porcelain,or other insulating material equal ly
a s good,and brass . I n the plug as shown the parts marked
b are brass and those marked i are insulation . The entiresparking device is set into the cylinder head on the plate Iand held in place with cap screws . The wiring connectionsare exceedingly simple
,the coil J and battery K being
connected up in series . The free terminal of the coil isgrounded on the engine frame and the free terminal of thebattery connected to the plug H as shown . I t is apparentthat the time of ignition may be varied by causing cam Ato trip early or late
,thus advancing or retarding the spark .
I NTERNAL COM BUSTI ON ENGI NES
engaged by A and only a slight movement of the piston willbe required to cause it to slip off of A and produce a spark .
Under these conditions the ignition will be advanced . I f,
however,plunger C is pushed farther in
,the contact break
can be made to occur in the explosion stroke, the Spring Bbeing too long to slip past the incl ined position of A which ismerely pulled away from B a s the piston recedes. W ith the
Fig . 74 . W ipe-Spark I gn iter.
ignition advanced it is apparen t that th is mechanism willproduce two sparks , one as the finger B Sl ips off A into thestirrup and another as it snaps out on the return stroke Ofthe piston .
The commutator or spark advancer,as used in connection
with the j ump-spark ignition system,is made in a variety
of di ff erent forms and a description of one typical formshould be sufficient for all purposes . For four-cycle enginesit is either mounted on the cam shaft or geared to it with1 1 gearing, it being apparent that, since but one explosionoccurs in each complete cycle , the spark as well as the valvemechanism must be operated by a half-time shaft . I n
I GNI TI ON 1
two-cycle engines the commutator is either mounted onthe crank Shaft or geared directly to it without any reduetion in time .
A commutator consists essentially of a piece of fiber,or
other tough insulating material,carrying one terminal , and
another terminal,connected to
the ground , which once in everycycle comes in contact with theinsulated terminal thus moment a rily closing the electrical circuit .
The Penn ington Fig. 76 . Commuta tor for TwoI gn iter. Cylinder Engine.
Fig. 76 shows a common form of commutator for a twocylinder engine . I n the drawing
,A is a fiber ring mounted
as shown on a flanged sleeve B ,which may be rotated about
shaft C by means of the lever D . Shaft C, which may be. thecam Shaft or an auxiliary timing shaft
,carries the ground con
tact mechanism E. As the Shaft C revolves , and the contact E engages the shoe F
,the circuit is closed through the
primary coil , and the current from the battery causes anelectromotive force to be generated in the seconda ry winding, producing a spark. I t is apparent that by shifting the
1 80 I N TERNAL COM BUSTI ON ENGI NES
sleeve B around the shaft C the ignition may be advancedor retarded by varying the point where the ground contactE meets the shoe F. A commutator similar to the onedescribed may be made for any number of cylinders byincreasing the number of contacts accordin gly .
For two- cycle engines a Simple and inexpensive commutator may be made of a flat fiber disk moun ted on the end ofthe bearing next the flywheel
,the ground contact point being
a spring pin in the wheel . 77 represents this form of
Fig. 77 . Crank Sha ft Commuta tor for Two-Cyc le Engine.
commutator . Mean s for advan cing or retardin g the sparkis provided by lever A
,by which the fiber m ay be moved
about the shaft C.
Fig. 78 represents a form of commutator,of cheap con
struction , which m ay be used to advantage in single- cyl inderfour-cycle con struction . The make and break is secured bymean s Of the cam C,
moun ted on the cam shaft,actin g on
the spring S which is carried on the fiber F as shown .
T iming is secured by moving the fiber around the shaft bymean s Of the rod R.
Fig. 79 illustrates a more complicated a nd expensivecommutator .
I NTERNAL COM BUSTI ON ENGI N ES
Fig. 79 . Commut a tor w ith H ammer Break Con ta cts.
FI G . 80 . The Pogn on Spa rk Fig . 81 . Spark p lug ofPlug . Ord ina ry Const ruct ion .
I GNI TI ON
Raj ah plugs,although the porcelains being light they a re
subj ect to breakage . Figs . 8 1,82
,83 , and 84 represent
several well-known makes of plugs .
Dynamo ignition ha s become quite popular in the past fewyears
,and were it not for its rather high first cost it would
undoubtedly be the universal electrical ignition . I t is
absolute in its operation , and used in conn ection with a
Fig . 82 . Plug w ith Fig . 83 . Sta -Ri te Fig. 84 . Sta-Ri teSpring Cl ip Con Plug . Plug .
nect ion .
storage cell , the operator need give no attention to the matterof batteries . W ithout the storage cell it becomes necessaryto use two or three cells for starting
,if the en gine is too large
to turn over quite rapidly by hand , after which , by means ofthe dyn amo
,i t generates its own Sparking current .
I N TERNAL COM BUSTI ON ENGI NES
Of the dynamo igniters probably the one most commonlyin use is the Apple made by the Dayton Electrical Manufa cturing Company , Dayton , Oh io . I t is a neat and compactdev ice
,the reasonable first cost of which , as well as its
effic ien cy , recommends it . The parts are entirely enclosed ina water a nd dust proof case . I t is provided with a centrifugalfriction clutch governor , as shown in Fig. 85, the Shoes ofwhich release as the engine Speed increases, thus causing it to
run steadily . The brushes a re so placed as to enable the
dyn amo to run in either direction , the field magnets beingperm anen t . These igniters a re m ade either for j ump-Sparkor make—a nd-break ignition , and generate a constant e .m . f . of
Fig. 85 . The Apple I gn iter .
from 4 to 5 volts a t from 1000 to 1200 rev . per min . Storageba tteries may be connected up to these clyn amos and thesurplus energy not used for sparking stored up for starting .
Fig. 86 shows the wiring diagram for a single cylinder withjump— spa rk ign ition . For a multiple cylinder the system isiden tical except that there are extra spark plugs and commutator connections to be made . See Fig. 87 . The storageba ttery
,Fig. 88 , always furn ishes the current to the primary ,
a nd by means of the four-point automatic cut-out switchS
,see Fig. 89 , it is possible to read on the volt-ammeter the
I N TERNAL COM BUSTI ON ENGI NES
other terminal of which and the moving arm of contact breakerF are grounded . Sleeve B is slotted , and when the slotscome opposite the poles of the field magnet G, the armature
Fig. 88 . Storage Ba t tery . Fi cr . 89 . Four ‘ Poin t Cut -Ou t .
D
windings momentarily cutting the magnetic lines,a powerful
e .m . f . is generated . The con tacts of the contact breaker Fare held together by the action of the disk D during thein terval between sparks ; the low-tension winding a thusbein g closed on itself , there is a high—tension current through
Fig . 90 . W iring Connect ions for Make-and-Break I gn it ion .
it a s the magnetism of its core is varied by the rota tingsleeve B . When one of the notches shown on B comesunder the movable arm of the contact breaker F the poin tssnap open
,owing to the action of the spring
,and break the
primary circuit a . This generates a high e .m . f . in thesecondary winding b, the cur ren t being further increasedby means of the condenser H . The seconda ry windingbeing connected to the prima ry a s described a nd connected
I GNI TI ON
through it and the ground to the several spark plugs as shown ,sparks are produced in the four or more cylinders
,the tension
of both the primary and secondary coils being utilized .
Fig . 9 1 . Sect i on through Apple I gn iter.
Distribution is accomplished by means of the commutator E which opens and closes the secondary coil .The Motsinger Auto-sparker is shown in Fig. 93 . I t
Fig. 92 . Bosch Type of Dynamo.
runs at from 700 to 1200 rev . per min . and is automaticallygoverned to maintain a uniform speed .
A cut of the Remy Magneto , made by the Remy ElectricCompany at Anderson , I nd.
,is shown in Fig. 94 . This
I NTERNAL COM BUSTI ON ENGI NES
ign iter is so designed as to produce two electrical impulsesfor each revolution of its armature
,and must consequently
Fig. 93 . The M otsinger Sparker .
Fig. 94 . The Remy M agneto .
be in step the engine . I t will del iver a Spark atfrom 25 to 30 rev . per min . a nd it is easy to sta rt an enginewith this magneto by cranking it in the ordinary way .
I NTERNAL COM BUSTION ENGI NES
alternately open to ports a and d,and _g is a gas j et con
t a ined in the hollow pet cock c as shown .
I n starting,the engine is turned over until ports b and d
correspond and gas j et e is lighted . Jet g is then turned onand lights from c. Now i f the en gine is turned over untilports b and 0. correspond as shown in the figure it is apparentthat a bare flame will be exposed to the contents of the cylin
H ot Tube I g n iter .
and that the charge will ignite . The pet cock mustturn quite rapidly, otherwise the flame 9 will be extinguishedowing to lack of air when ports a and b are both closed to b.
When the explosion occurs the flame 9 will be extinguished,
but on turning the pet cock back until ports b and d correspond i t is again ignited by flame e. The time of ignitionmay be varied by suitable arrangement of the half-timedevice .
The hot-tube ignition is well illustrated by Fig. 96 and issimilar to those in use on the Robey engines .
I n the figure , a is the igniter tube , which should be of ni ckela lley or porcelain ; b, the chimney wit-h lining of asbestos ;
I GNI TI ON
c, the asbestos lining d, the bunsen burner which heats a , andd, a set screw by means of which tube b may be raised orlowered to regulate the degree of heat to which a shall beraised .
On the compression stroke of the engine the charge isforced into tube a
,through ignition port f , and a t
'
the properinstant , determined by the compression , is exploded . The
adjustment of flame and tube to secure the proper degreeof heat , that the charge may be ignited at the proper instant ,is a deli cate operation and one that is determined only byexperiment for the different makes of engines . I t is tobe remembered that the products of combustion tend toremain for some time in the ignition tube and that the nextcharge of gas must force these products ahead of it beforereaching the hot part of the tube .
The hot-tube igniter,in connection with a timing valve
by which the ignition may be more readily varied,is also
used to some extent. The tube, in this style of igniter, isShorter and of larger bore
,and is kept at a temperature
high enough to insure the charge igniting as soon a s itcomes in contact with the heated surface . The ignition tubeof a hot-tube igniter Should be of
,some substance not
easily affected by the severe heat to which it is continuallysubj ected . Ni ckel alloy has been found to give the bestsatisfaction
,in metal tubes, a tube of this material giving
service for a long time, unless subj ected to the action ofsulphurous gas or excessive heat
,while a tube of wrought
iron pipe must be replaced every few days. Porcelaintubes
,although more difficult to set, give excellent sa t isfa c
tion for long periods . The R. Thomas Sons Company,
East Liverpool , Ohio , furnish porcelain tubes ; and rodsfor the making Of nickel-alloy tubes may be secured fromH . Boker CO.
,10 1 Duane Street, New York .
Fig. 97 shows a good method for setting porcelain tubesin a socket that may be screwed into the cylinder . The
tube may be pa cked in with asbestos washers and moistfire clay or with asbestos cement .
Auto- ignition , or ignition secured from the heat of com
I N TERNAL COM BUSTION ENGI NES
pression,is found
,in more or less modified form ,
in severalmakes of engines . Chief among these is the Diesel motor .I n this motor the compression is carried to about 500 lb . persq . in . above atmosphere
,and the temperature in the
combustion chamber reaches nearly1000 deg . fahr .
,which is sufficient to
ignite the charge,as soon as it is forced
into it . The ign ition of this engine aswell a s its fuel valve are fully describedelsewhere in this work
,under their
respective heads .A modified form of this type of
ignition,and one in which the pressure
is not carried as high,is found in the
H ornsby-Akroid a nd the Miet z WeissOil engines
,the former being of the
four-cycle type and the latter a twocycle engine . Fig. 98
,A and B
,Shows
the principle Of these two igniters . I n
both,ignition is Obtained by forcing
the charge,during the compression
cycle , into a small chamber a,the
walls Of which are not water— j acketed
Fig . 97 . M ethod of I nbut are surrounded by an outer casing
sert ing Porcela in Tube. e. The heat of compressi on,together
with the residual heat from previousexplosions
,in this small insulated chamber is suffi cien t to
ignite the charge . I n A,the four- cycle type , the oil is sprayed
directly into the igniter,thus utilizing the heat in this cham
ber to vaporize the oil . The same mechanism,operating the
oil spray b, Opens the air port 0 on the suction stroke , at theend of which stroke the ign iter chamber contains vaporizedOil and products of combustion , while the cylinder containsalmost pure air . On the compression stroke
,the air
being forced into chamber a , the temperature , due to compression
,rises un til at the proper instant ignition occurs .
On the exhaust stroke , the exhaust valve d is opened and theproducts of combustion are forced out of the cylinder .
I N TERNAL COM BUSTI ON ENGI NES
compression and make the explosion occur late . The speedvariation is obtained by varying the fuel charge .
There is no fixed method of determining at what point inthe cycle ignition should occur ; with diff erent fuels and undervarying conditions this point will be diff erent with nearlyevery case . Gasoline vapor ignites and burns very rapidly ;gaseous fuels consume more time in the process . The theoretical condition is to ignite the charge such a point
Four-Throw Cranks for Differen t Fir ing Orders.
compression stroke that it will attain its maximum pressurej ust a s the engine passes dead cen ter into the expansions troke . Approximately it may be said that the lead shouldbe increased 5° on the crank circle with every 100 rev . permin . increase in speed ; thus at 200 rev . per min . the leadshould be about at 300 rev . per min . about and soforth . The point at which the engine runs best should bedetermined by open ing the throttle wide and then advancingthe spark un til with further advancement no greater speedis obtained or the engine commences to knock .
I GNI TI ON
The firing order of engines with more than two cylindersvaries to some extent with the manufacturers . A three
1 2 3
cylinder engine should fire in order 1 — 2 in fact it can befired no other way
,the cylinders following each other from
left to right or from right to left . A four-cylinder engine1 2 3 4 1 2 3 4 1 2 3 4
may be fired (A) 1 — 2— 3— 4, or (B) - 2—4,or (C) 1— 3—4- 2 ,
the cranks, of course , being made to correspond . The ordermay be taken from either the left or right . For firing orderA the crank throws must be as shown at A
,Fig. 99. For
firing order B the crank throws must be as shown at B,
Fig. 99,and for firing order C the crank throws must be as
shown at C,Fig. 99 . I t is apparent that the cranks for
firing orders A a nd C are identical , the wiring connectionsonly being changed . Order B is probably the best practiceand as widely in use as any .
I n a six-cylinder four- cycle engine the explosions overlap,
one impulse being given every third of a revolution . The1 2 3 4 5 6
order of firing should be 1 — 3— 5— 2— 4— 6,with cranks set 60°
apart . For6
an eight-cyl inder engine the order of firing is1 2 3 4 5
1 — 5— 3— 7— 2— 6— 4— 8,with the cranks set 45° apart , that is ,
crank number 5 i s located 45° behind number 1 , crank number 3 is located 45° behind number 5 , and so forth .
I n the above firing orders the sma ll figures represent the cy linders ,ca lling the one to the left 1 . The la rger figures below indicate the orderin which that particula r cy linder is fired .
CHAPTER XXI I .
ENGI NE TEST I NG.
TH ERE are but two general methods of testing engines todetermine their b . hp .
,although there are many diff erent
designs and makes of in struments used in the running a nd
recordin g of these tests. The two methods are ( 1) Byabsorption of the power ; (2) By transmission Of the power.The instruments used in the first case a re known a s absorp
tion dynamometers and those used in the second case astransmission dynamometers . The absorption dynamometerdiffers from the tran sm ission dyn amometer in that itsobj ect is to absorb the delivered power at its poin t of deliverya nd to record its force and velocity ; while the transmissiondynamometer measures the actual net differential tension ofthe belting or gearin g a nd
,with its velocity known
,deter
m in es the amount of power del ivered by it. Transmissiondynamometers are Of two kinds — the self—recording andthose in which the readin gs must be constantly taken bythe operator from a dial and the mean pressure obtainedfrom a number Of these readings .Fig. 100 represents the Simplest form of absorption
dynamometer or Prony brake . I n the figure,A is the radius
of the drum or pulley to which the friction is applied a s
Shown,the bolts C and C clamping the device as tightly
a s may be necessary . The lever arm B delivers the absorbedforce to the Sprin g balance E
,where from time to time
,at
regular in tervals,readin gs are taken by one Operator while
an other,by means of a tachometer, determines at wh a t
speed the pulley of radiusA is revolving . Now it is apparen tthat if the end of lever a rm B were un restri cted it wouldcon tinue to revolve a s a part of the pul ley A and that thepoint X
,at which it is shown attached to the balance E
,
196
I N TERNAL COM BUSTI ON ENGI NES
Trials , tells us that the number obtained by multiplyingthe width of brake W by the peripheral speed V and dividingby the horsepower H shall not exceed 500 to Call ingthis value K ,
we have the formula
WV
H
From which the unknown value K may be readily found .
I n running a brake test on a gas or gasoline engine, especia lly the latter , where the speed is high
,difficulty is en coun
tered in the heat produced by the fri ction of the brake on
Fig . 101 . Arrangement for Brake.
the wheel . For this reason a method should lie providedsupplying water to the rim . One method is to make
the flywheel rim in the shape of an inverted U ,pouring
in water a s the engine is runnin g . The centrifugal forcecarries the water out in the rim and keeps it there . Thismethod is satisfactory for comparatively short tests
,but
for longer running tests a more efficien t device should beused . A good way is to so design the wheel that running
“ Engine a nd Boiler T ria ls, R . H . Thurst on ,pages 272 to 279 .
ENGI NE TESTING
wa ter may be kept flowing into it . Fig. 101 i llustrates thedesign of such a wheel . The construction of the brake is asfollows : the small blocks a
,made of fiber or tough wood
,are
securely riveted to band b which passes around the wheeland through blocks 6 which are fastened to the lever arm B .
Pressure of the brake on the wheel is obtained by means ofthe clamping nut d. The wheel has a rim section
,as Shown ,
through which -in . holes are bored at random . Watermay now be kept running in a steady stream into the rim andout through these holes onto the brake blocks . AS the
Fig. 102 . Brake Shield.
rapidly revolving wheel throws the water and makes thetesting an exceedingly unpleasant operation
,a shield of
sheet metal Should be placed over the brake,as shown in
Fig. 102 . The addition of a little soap is a good precaution to prevent the brake sticking and causing j erky anduneven action of the scale beam .
Discussi on of the Formula .
— The formula for the Pronybrake
,as above derived , admits in all cases of a constant
being determined which Simplifies the solution . There a re
in the formul a four known quantities : a R the
radius of the lever arm as designed in any particular brake,
the foot-pound equivalent,and the figure 2 . Sup
pose now that the brake were designed with a lever arm
the formula gives us H WN . I t is apparent thatfor any given length of brake arm a similar constant may bederived a s shown in Table X I I I .
200 I N TERNAL COM BUST I ON ENGI NES
TABLE XI I I .
PRONY BRAKE FACTORS .
From the table it is readily seen that the most convenientlen gth for R is ft .
,for the constan t then is and the
solution of the formula on ly entails moving the decimalpoin t one place to the left in order to Obtain the horsepower .
The horsepower for any one revolution may be obtained bysimply multiplying the scale reading at any instant by thecon stant .The brakes most in practical use are modifications of the
ones described,but there are some specially design ed brakes
which Obtain the friction by hydraulic action,the water
producing the pressure acting as a coolin g agent a s well .The Alden brake is so con structed . For a more exhaustivediscussion of the Prony brake the author would refer thereader to Carpenter ’s “
Experimental Engineering, pages 207to 2 16 .
The Belt Dynamometer . Amodification of the Pronybrake whi ch m ay be used in the testing of small en gines isthe belt dynamometer , in which the brake a rm is the radiusof the pulley or flywheel . The belt to be used is providedat both ends with a ring
,to which weights may be fastened
,
a s shown in Fig. 103 . The acting force,a nalogous to W in
the formula,in this case is equal to the diff erence between
the weights A and B when the apparatus is in equil ibrium .
2 7zR ( . l — B)NThen
202 INTERNAL COMBUSTI ON ENGI NES
smaller pul ley is thencator drum
,a s Shown .
tus. A reducing device
,consisting oi pul
leys or a pantograph
(“ l azy tongs
”
) mustbe rigged up . The
device best adaptedto the testing of agas en gine is the re
ducing pul ley . Thisapparatus is shown inthe diagrammatic layout
,Fig. 105, a t b, and
consists of a pair ofpulleys a s shown in
detail at l . The pul leywheels should be a s
light a s possible , preferably Of aluminum orwood
,in order that
the inertia may be reduced a s much as possible . The device is
tapped in to the top ofthe water j acket
,a s
shown,and two or
three wraps of strongcord a re taken aroundboth pulleys , the endof the cord being madefast to the rim of thewheel . The cord fromthe larger pulley isthen run to the pistonend and fastened thererigidly ,
a s shown at 2 ;the cord from the
fastened to the cord on the indiFor engines operating at a high
ENGINE TESTING
rate of speed the indicator drum spring must be set up tohigh tension in order to absorb the inertia of the part s asrapidly as possible . The cord must a lso be free from meta lhooks in order to minimize the tendency to vibrate . For an
engine running at 200 revolutions per minute the cord tension should be about 2 pounds . For 300 revolutions per minute it should be pounds ; for 400 revolutions, pounds
,
and for 500 revolutions, 14 pounds . The Prony brake C isset up substantially as diagramed . A thermometer in abrass oil bath is inserted in the discharge
,as shown at e,
and another in a similar oil bath is placed in the in let,as
shown at d. The temperature and pressure of the gas entering the engine are taken at g and f respectively . The
pyrometer is located in the exhaust passage, a s at h, in order
to Obtain the temperature of the exhaust gases . Athree—wayvalve is placed at i so that the discharge water may beturned on or off the weighing tank j as required . The
barometer and room thermometer may be placed at anyconvenient point near the test . The air meter is notshown but should be placed at some point in the airsuction pipe .
Testing wi th the Prony Brake.
— I n running a test withthe preceding arrangement
,the first consideration is the
weight which the brake itself will impose on the scale,a
constant which must be subtracted from all scale readingsin order to find the net value of W or the actual force beingexerted on the scale by the brake arm . To determinethis constant loosen the brake on the wheel or pulley andby means of a Spring balance attached to the point at whichthe weight reading is taken
,raise the arm a foot or more
,
reading the scale while doing so . Then lower the arm to itsoriginal position , taking another reading during the operation .
As there is some friction on the wheel at any time, these tworeadings will be different, the first being the larger . The
avera ge of the two is the value of the constant C.
204 I NTERNAL COM BUST I ON ENGI NES
Constant C having been determined , prepare a logfor the test substant ially as follows .
NAME OF ENG I NE .
MANUFACTURER
TEST MADE BY
AT
DATED I AM . PI STON . STROKE CLEARANCE .
Before commencing the actual test two or three assistantsshould be secured in order that the different readings m ay
be taken a s nearly at the same time a s possible . One readin gof the barometer a nd the room thermometer is gen erallysufficien t for a n hour’s run . The en gineer in charge of thetest should handle the speed indicator while an assistan ttakes the brake scale readings
,keeping the scale beam
con stantly floating a s the power fluctuates . An otherassistan t will be able to handle the indicator and to takethe tempera ture of the in let and discharge water
,while a
fourth assistan t will be able to take care Of the weighin gtan k a nd to take whatever other readings are necessary .
Readings should be taken at five—minute in tervals whilethe test lasts
,and a n hour’s test is gen erally suffi cien t for
any one adjustment of the engine . One run should be madeat maximum power
,one at the rated horsepower
,a nd a third
at no load with the brake removed,the latter being merely
a n indicator test . Run s may also ‘ be made at quarter,
half,a nd three-quarter load if desired . I n order that the
readings may be taken a s n early a s possible at the sametime the engineer in charge should be provided with a
I N TERNAL COM BUSTI ON ENGI NES
or by weights,suitable recording arrangements , either auto
matic or dial,being provided .
The method of engine testing as described applies to themaking of a complete test of a gas engine
,in every detail .
Such a complete test is not usually run ex cept in caseswhere a record of the actual performance of an engine isrequired in every such detail as a basis for a guarantee orto detect faulty design . A description of a complete testnaturally takes into consideration all the smaller details ,but it is an easy matter to perform any part of the testrequ ired . The most common tests performed are the brakeand indicator tests to determine the horsepower and themechanical effi ciency of the engine .
Ga soline, Alcohol , a nd Oi l Engines — I n the testing ofengines operating on liquid fuel , while the indicator andbrake tests
,as well as a maj ority of the other tests
,will be run
in the same way,the tests for fuel con sumption wil l be differ
ent . I n place of the column cu . ft . of ga s,” a column read
ing gal . of fuel ” should be substituted . The column“ press . of gas in in . of water ” is omitted
,and the column
reading the “ temperature of the ga s”Should be replaced by
one reading the “ temperature of the fuel ” ; in all otherrespects the log sheet may be used as it stands . The tem
pera ture of the fuel , if a carburettor is used , should be takenat the carburettor ; i f a j et or m ixin g valve is used the t em
pera ture should be taken j ust before the fuel reaches thevalve .
I n a gasoline automobile motor,while many manufacturers
make talking points of their low fuel consumption,the
question of the amount consumed is secondary to the powerderived
,flexibility of control
,a nd weight
,and
,a s a matter
of fact,with the varying loads and speeds to which such an
engine is constantly subj ected it is impossible to Obtain anyvery great fuel economy , and as a consequence a brake andindicator test is all that is usually required for such an engine .
CHAPTER XXI I I .
REPORT or TESTS .
TH E d ata having been obtained by the method discussed inthe la st chapter , a report of the conditions found must bemade .
When a number of tests are being conducted a printedform of report blank Should be made as fol low s :
PLACE . .
ENG I NEER I N CH ARGE .
ASSI STANTS .
NAME OF ENG I NE . MANUFACTURER
RATED H P REV .
“
PER M I N .
DI MENSI ONS .
Diem . p istonArea pist onPiston displacement
Compression spaceTheoret ica l eompression .
DATA
Length of testRev . per min
Rev . per hr
Explosions per min . i
Explosions per hr . .
Air per hr
Rati o gas toWater consumptionTemperature inlet waterTemperature dis . waterRange water temperature
Temperature exhaustAv . lb.
208 I NTERNAL COMBUSTION ENGI NES
PRONY BRAKE .
Length lever a rm (R) .
Constant brake (Correction)Brake load avera ge (Gross)Brake load average (Net )Weight Of gas per cu . ft
Weight of ai r per cu . ft
Average m ixture Ra tio gas to airWeight Of m ixture per cu . ft
Sp . heat gas .
Sp . hea t a irSp . hea t m ixtureH ea t va lue cu . f t
RESULTS .
I ndica ted m .e .p
I ndicated hpGas per i .hp
Gas per d .hp
Mech . efficien cyFrict ion loss
H EAT PER H OUR .
Supplied by fuelAbsorbed by wa teI n exha ust gases B .t .u .
Absorbed in workRa diation a nd f rictionTherma l efficiency Per centB .t .u . per i .hp .
As there will probably be some items in the report not
readily understood by the reader, an explanation of thesewill be given .
The ratio Of gas to a ir is the quantity of gas c onsumed perhour divided by the air consumed . I n the en gines whichtake a charge of air into the cvlinder, whether gas is takenor not
,a s in the type governed by mean s of the “ hit or
miss ” cut-Off type of governor , the exact ratio cannot beObtained , except a s the engine takes an impulse at ev erycyc le . An a pproxima te result may be Obtained however
,
Av . ft .
- lb . per min .
Av . ft., -lb . per hr .
AverageAV . lb:
Averagecu . ft .
cu . ft .
Av . d .hp . Av . i .hp .
i .hp . d .hp .
21 0 I NTERNAL COM BUSTI ON ENGI NES
Of the mixture is found by taking the weight of air to be1b . at 32 deg. fahr. and atmosphere The
weight of the gas being known,or having been determined ,
the weight Of mixture per cubic foot is found by the following formula :
Wm 3: a y
I n which a: is the percentage of air, a the weight Of the ga sper cubic foot
,and y the percentage Of gas .
The heating value of the gas Should always be determinedat a laboratory by an expert chemist . While Table I ,Chapter I X
,gives the heating values of various gases, their
composition in diff erent localities or under difieren t conditionsis subject to variation and for accurate resul ts should not bedepen ded on
, but on ly used a s a basis for computing probableresults . Samples of the gas Should be obtained at differenttimes and then mixed ; a sample of the mixture Should thenbe sent to the laboratory .
The volumes of ga s and air a s Obtained in the test andrecorded in the log sheet must be reduced to standardtemperature and atmospheric pressure in order to form abasis of comparison . The stan dards in use are the temperature of water at the freezin g poin t
,32 deg . fahr .
,and the
atmospheric pressure at sea level , which is equ ivalen t to30 in . Of mercury
“
. The following formula may be used in thereduction :
vp X
X (t1
I n which
t,
temperature at time of test .
p atmospheri c pressure at time Of test .I ) volume at this pressure and temperature .
V corrected volume at 32 deg . fahr . and pressure Oflb .
This formula , while derived on the basis Of the air thermometer , will give an approximately correct value when usedfor the gas . The gas pressure , as mea sured in the test , inin ches of wa ter may be reduced to inches of mercury by
REPORT OF TESTS
dividing by or if the pressure of the gas has been takenin pounds per sq. in . it may be reduced to inches of mercuryby multiplying by These ratios are for temperaturesof 32 deg. fahr . but will be found to give sufficiently accurate results ii used at any average temperature .The indicated work is to be computed from the indicator
cards taken during the test and is the product of the meaneffective pressure , the area of the piston in inches, the strokein feet
,and the number of explosions per minute . The
mean ordinate of the cards is obtained best by means of aplanimeter .* I f a planimeter is not available it may beObtained by the ordinate method
,see Fig. 106 .
From the atmospheri c line AB erect ordinates equaldistances apart, the first and last ordinates being ha l f a space
Fig. 106 . Engine Card .
from the ends of the diagram . Add the lengths of thelines contained in the diagram
,in this case ten , from AB to
the expansion curve ex, and add the lengths of the samelines from AB to the compression curve ay, subtract thesecond sum from the first and divide by the number Of
l ines. The result wil l be the approximate value of the
A plan imeter, as the n ame implies, is an instrument for computinga reas . I t may be so ad justed as to record, wi thout further ca lculat ion ,the mean ordina te of any irr egular figure of given length. See Car
penter’s“ Experimental Engineering,” page 81 .
21 2 I NTERNAL COM BUSTI ON ENGI NES
mean ordinate in inches . I f the expansion curve shou ld beirregular
,due to the vibration Of the indicator spring , a
mean curve may be draw n as shown to right Of Fig. 106 .
Whi le the ordinate method will give fairly close results ,it s use in computations requiring extreme accuracy is notadvisable .
The mean eff ective pressure is the result Obtained bymultiplying together the mean ordinate in inches and thescale of the Spring in lb . per sq . in .
,which is known . The
mea n effective pressure being known , the indicated horsepower is calculated by means of the following formula :
I n which P Mean effective pressure .
I Length Of stroke in ft .
0. Area Of piston in sq . in .
n Numbe r of explosion s per min .
NOTE .— The d ifference between n in the above formula and n in the
same formula as appl ied to the steam engine should be noted . I n steam
engine work n number of rev. per min . but in gas -engine work n the
number of impulses given per min . I f the engine “ h i ts ” regu la rlyevery cyc le there wou ld on ly be ha lf a s many impulses a s revolu t ion sin a four-cyc le engine . For a two -cyc le engine, however, the va lue of
n wou ld be the same as for a steam engine .
The quantity of heat supplied by the ga s per hour is theproduct Of the heating value per cu . ft . times the cu . ft . consumed . The heat absorbed in the water is the produ ct ofthe water consumed , as determined in the test , and the rangeof temperature :
H i (ta il )w
I n which i , The temperature of the discharge .
t2= T emperature of the inlet .
W Weight of the cooling water .
21 4 I NTERNAL COMBUSTI ON ENGI NES
I n which the numerator is r ecognized as the numerator offormula
Cu. ft . of gas per hr .
H eat value per cu . ft .
of engin e is found by means formula2 RTWn
01 ed i ther iv,or e
I n which the constants for different lengthsin the table , pa ge 200.
M I SCELLANEOUS .
The M afi a — The volume Of muffler necessary,for any
given size Of engine , can be only approximately determinedby formulas . AS a general proposition
,however
,the volume
of the muffler is from four to six times the total cylindervolume . While it is true that some engines
,especially
those in use on motorcycles,employ a muffler with a much
less comparative volume,their use does not produce as
good results from the standpoint Of quietness of operationalthough the back pressure is less .A muffler
,in its simplest form
,con sists Of an iron box or
drum into which the exhaust gases are discharged beforepassing to the atmosphere . While such a muffler deadens
,
to a certain extent,the noise of the explosion , the develop
ment of the automobile industry ha s demanded a moreefficient sound deadener . To produce this resu lt the gasesare made to pass through orifices or past baffle plates inorder to give them more time to approach the pressure ofthe atmosphere before discharging into it . One simplemethod is to enclose an iron pipe
,bored full Of small holes ,
in an iron shell and discharge the exhaust gases into the pipefrom which they pass through the holes to the Shell of themuffler and finally to the Open air .I t should be remembered that the more effic ient a muffl er
is the greater is the back pressure and loss of power ; hencethe use Of the simple drum type on industrial engines andother engines where quietness of Operation is a secondaryconsideration .
DEFI NI TI ONS OF UNI TS.
Work. The sustained exertion of pressure through space .
Uni t of Work. One foot pound, i .e. , a pressure of onepound exerted through a space of one foot .
216
21 6 I NTERNAL COM BUSTI ON ENGI NES
H orse-power .
— The rate of work . Unit of horse-powerft . lb . per minute, or 550 ft . lb . per secondft . lb . per hour .
H ea t Un i t — H eat required to raise 1 lb . of water 1 deg .
Fah r. (from 39 deg . to 40 deg ).
H orse—power expressed in heat units
heat units per minute heat units per second 2545
heat units per hour .ft . lb . per lb . of fuel .
1 lb . of fuel per hp —hr2545 heat un its .
1 000 ft . lb . per lb . of fuel = lb . of fuel per hp .-hr .
5280Veloci ty. Feet per second
3600 1
—
5X miles per hour .
TABLE XV.
W I RE AND SH EET — METAL GAUGES COMPARED .
inch inch inch nun .
. 500
. 464
. 438 . 432
. 406 . 4
. 14428
. 12849
. 1 1443 . 148
. 10189 . 1 35 . 19 1
.220 INTERNAL COM BUSTION ENG"NES
TABLE XVI I I .
I )ri ll .Actua l Ex Actua l I n
4 -500
222 I NTERNAL COM BUSTION ENGI NES
CI RCUMFERENCES AND AREAS OF CI RCLES — Con ti nued .
Diam . Circ . Area . Diam . Area . Diam . Circ . Area .
5
fi5131
4234}if9
l 10 . 75
224
CI RCUMFERENCES
Diam .
I N TERNAL COM BUST I ON ENGI NES
Area . Diam .
AREAS OF CI RCLES.
Circ . Area . Diam .
l 56 .z94
Area .
l876 . l
I 9S3 . 7
I NDEX
Abbé de H autefeu ille . 1
Acetylene, combustion , rateof . 82
genera tion of 81
heating va lue of . 82
pressure. of liquefica t ion . 82
y ield per pound of ca lc iumca rbide .
Air supply for horizont a l euTh e
Alcohol, combustion , a ir re
qu ired for . 76, 80
heat of 75
rate of 7 1
composition 76
hea t va lue of , computed 82
temperature form ixture, 56 , 7 1 ,9 , 80
therma l effic iency 83
vaporiza t ion ,hea t re
qu ired . 78, 80
Aspira ting va lve 49
Automa tic engines . 25
Avoga dro’s law 80
Ba ck firing . 1 6 , 23 , 34 , 35 , 38 , 1 27
Ba ffie pla te . 1 6 , 22 , 1 34
Ba la nce weights 154
Ba rber , John 1
Ba rnett engin 3
Ba rsa nt i a nd Ma tteucc i . 3
Ba ttery cel ls . 32,33 , 1 73
storage “ 1 87
Bea rings, adjustment of , inc losed ca se 145
connectin rod 1 57
cra nk sha t 1 56
length of , for cra nk sha ft . 152
l iners 144
lubrica tion 29
pound ing 35
setti ng, in horizonta l emgi nes .
setting , in vertica l engines, 144
studs 144
Beau de Rochas princ iple, 4, 6 , 1 7
PAGECa lorie 75Ca lorimeter, the 75Cams, classifica tion 105
double . 1 07
operation 1 09
layout 1 1 0
effect of , on card 89
exhaust 1 05
in let 1 07
ma teria l 1 1 1
off setting . 103
single, layout . .1 05
sparking . 1 77
sta rt ing 24, 39 , 40, 103
Cam mechan ism doub le camsha ft . 1 24
gearing 1 02 , 1 1 1
lost motion in . 107
tim ing of sha ft 1 02 , 1 1 1
transm ission of motion tovalves 1 02 , 103
Ca rburret tor , adjustment . 25 , 32
air supply 42
a lcohol 56,83
aux i liary a ir supply 53 , 54
design 57
effect on ca rd . 7
flex ib i l ity 49
floa t feed 43 , 48, 53
H olle 53 , 54
mechan ica l ebu ll ition type 46
primer . 53
Schebler . 52
spray type 48
surfa ce type 47
two-cycle engineCarburett ing, a lcohol .
petroleums ”
tempera ture of fuel a s a f
fected by continued
2 1
43,56
43
vaporization 46
Ca re of engine 27
Circu lation 27 , 30, 33 , 1 29
Clea rance of eng ine, determ ina tion Of .
Clerk, Dugald251
I NDEX
PAGECoal ga s, see Ga s. Crank pinCoil . Crank Sha ft
,a rms .
requ isites of ba la nce weightsCombustion bush ingair requ ired . capac ityheat . fin ish .
Commuta tor .
Single cyl inderthnnn;two cyl inderstwo-cyc le .
Compression , cha rt
Clerk enginecurveDegra nd engineD iesel engineedfect of va lveefficiency,
rela t ion to .
l i I I I it S
mean effective pressure,relation to
premature explosion produced by .
pumprela tiontofuelrela tion to speedrel ief cockrel ieved for sta rting .
Robson engine . 1 2
spa ce, ra tio t o st roke , 9 1 , 94 . 99
t able 85
tempera ture 84
two-cycle engine (cra nkcase)
two-cyc le engine (cylinder)
Condenser, electrica l .
Condensa tion , laten t hea tCon n ecting rod .
a djustment of bea rings .
formulaCooling , H ugon spra ytovver
water, effect of , in hot cylinder 30
hea t lost in 2 13
in let a nd outlet, loca t ion 132
in let a nd out let , S ize . 1 33
pressure . 27 , 96
regu la tion .
tempera ture .
Cost of fuel, a cetylenea lcohol .
oxygen andhydrogenproducer gas
Crank ca se explosions
PAGE153
1 53
1 54
1 56
1 51
1 56
1 53
1 51
Crude oi l, composit ion . 72
Cyl inder, a ir cooled 1 29
automobi le 1 3 1,1 34
bolts . 1 35
bore 98
bore , ra t io to stroke . 100
boring 1 33 , 1 34
care . 3 1
casting , cost of 136
effect of overhea ting 34
equation for . 99, 1 01
flooded . . 32
ga sket for 35
hot-spot in 33ma teria l . 1 36
23 , 33
proportiona te equa tion for 100
thickness Of wa lls . 1 30
water cooled 1 29
water in 35
wa ter j a cket . 1 3 1 , 1 32
Day engineDegra nd engineDiesel engine .
cycleeconomy of .
effic iency .
Draw ings for founda t ion sDynamometer, absorption .
belttra nsm ission .
Economy of Opera t ion . 66 , 67
a utomob i le engines 206
Diesel engine . 18, 22
four a nd two cyc le engines 18H ugon engine . 4
Lenoir engine . 3
Otto a nd La ngen engine 4
Effic iency of engine , 86 , 88, 96 , 97 ,10 1 ; 2 13
Electric ign ition,2 , 3 , 28, 1 70, 17 1 ,
Engines, Abbé de H a ute
Ba rber pa tentBarnettBarsa n t i a nd Ma tteucc i .
I NDEX
PAGEGasol ine (see Fuels, Fra c
t iona l disti lla tion , a ndCombustion ).
a ir required for combustion 77, 81
combustion , rate Of . 67 ,fra ctiona l d isti lla tion . 74
l ighting ga s from 74
low heat value of , com
puted .
mean effect ive pressureproduced . 7 1 , 98
qua l ity . 25, 33
stra in ing . 25
vaporiza tion hea t re
qu ired 78 79 , 8 1
weight 74
Gasometer . 6 7
Gea ring, pitch 1 3
reduction for cam-sha ft,1 02
,1 1 1
skew ratio . 1 1 3
Govern ing , c loseness of regula t io
methods ofGovern ors, ca re of .
centrifuga lelectrica l govern ingexhaust .
hit or m issinertia .
l ift of ballsloca tionmagneto forthrottle va lveun iform ity of speed re
quiredWright
Gunpowder a s mot ive power
H eat ba la nceH ea t losses 2 1 2
H ea ting of cy l inder wa ll . 2 1
H orsepower, computation of , 1 97 .
defin ition of 2 16
friction . 97
ra tio of i .hp . to b .hp . . 96 , 97
theoretical . 97
H ugon engine . 4
economy 4
H uygens . 1
I gn iter, care of 23 , 25 , 28
hammer break .
hot tube . 1 90
materia l . 29,19 1
James-Lunkenheimer m ixing valve
Johnston engine
Keroseneign ition tempera ture
Keys, table of
I gn iter, Penn ingtonporcela in tube for .
setting .
w ipe sparkI gn ition
,eff ect on ca rd .
auto-ign ition,H ornsby
Akroid 19 1
M ietz a nd We iss 1 9 1 , 1 93
Diesel ign ition 1 7, 1 9 1
electric ign ition,jump
spark 170, 1 75
Lebon patent 2
Lenoir ign ition . 3
make-and-break ign ition . 28, 17 1 , 1 75
spark ing points ma teria l for .
prevention of destruc
tive a ctionflame ign ition ,
Ba rnett ign it ion cock 2 , 189
hot-tube . 24,29, 34 , 189
lead of 1 94
magneto ign it ion . 183
apple 183— 1 85
Bosch . 185
Mots inge 187
Remy . 187
Stepha rd 3
t ime Of, 7, 23 , 24, 25 , 28, 34 ,
I ndica tor, computa tion of
ca rdDiesel ca rd . 1 9
for ga s engine . 201
formu la for ca rd 86 , 90
four-cyc le ca rd . 6
purpose of card 89
Spring effects on ca rd, 9 , 1 0 , 2 1 2
two-cyc le cra nk ca se ca rd, 1 3 , 1 5two—cyc le cyl inder ca rd ,
1 3 , 1 5
I nd icated work 2 1 1
I n jectors, Diesel method . 17, 44H ornsby—Akroid . 44M ietz a nd We iss .
I nsula tion of engine foundation .
I NDEX
Lebon-Phil ipLenoir engineeconomy .
L iqu id a nd a ir explosivem ixture . 2
Lubrication . 23 , 24 , 26 , 29 , 1 56
Marine engine 25
Mea n effective pressure . 94, 2 1 2
Mechan ica l ebullition . 45
effic iency . 96, 97
mult iple cyl inder eh
g I neS .
M ix ing va lve , d iffi culty fOrtwo-cyc le engi nes
fuel most desirable .
James—Lunkenheimer design 49
throttle connection . 51
M ixture, effect on explosion . 97
explosive 23 , 24, 25, 26, 3 1 , 34
lack of . . 36
lim its of explosive 1 62
weight . 2 1 0
Muffler . 2 1 5
Natura l ga s, see Ga s.
Non- I nductive resI sta nce .
Oil gas, see Ga s.
Oil ringsOtto cyc leOtto and Langen engine
economy
PantographPa pinPermanent gasPetroleum disti lla tes (Table)
combustion hea trate .
Pi l ing for foundationP iston , design “
double actingexpansion .
pin .
rings .
speed .
taper .
two-cyclePlan imeterPorts, eff ective open ing .
lead of exhaust, for twocyc le 1 27
location of , in two—cyc le . . 1 1
Opera tion of , in two 1
2
5
1,
o o o o o o o o o o o o o o o o
0 0 0 0 0 0 0 0 0 0 0
o o o o o
6 6 6 6 6 6 6 6 6 6 6 6 6
o o o o o o o
t t t t t t
o o o o o o o o o o o
Ports, period of open ing intwo-cycle engine
proportion .
th ird , in Da y engine .
Power, comparative production in two a nd fourcycle engines 2 1
Premature explosion , 18, 34, 170Pressure requ ired for air
start ingPrim ing of ca rburettor .
cyl inder .
Producer, a ir pressure re
qu ired .
charging hopper forcost Of Operat ion .
d isti lling,temperature
effic iency Of plant as compared w ith steam
fuel avai lable .
suction , Operation ofProducer ga s, see Gas .
Producers, d isti ll ing (Riche) 60Dowson . 58inverted 62
pressure 58suction . 62
Prony brake, constant of 202
cooling . 198design . 197 , 1 99
formu la, derived 1 96, 1 97
d iscussed . 199hydraul ic brake 200
length of bra ke a rm 1 99
(Table) 200
wheel, design . 197, 1 98
k, formula 90, 92
Radiation , loss byReducer for producer ga s .
Reduc ing motions,pan to
gra hreduc I ng pulley .
Report Of test, form OfRing cuts .
Robson engine .
Ruhmkorff coi l .
Schebler carburettorScrubberSed iment in j a cket .
Smoke from cyl inderexhaust .
Spark , ca use of fai lure .
length of gap .
weak, cause of .
256 I NDEX
PAGESpark plug . 181 183
t ap for 13
Spray ing 52 , 53
Sta rt ing devices, a ir starters, 23 ,39
aux i l iary explosion chamber .
aux i l iary storage chamber .
ca rtridge starter ”
compression of first cha rgeby means of ha nd
pump 38
ext ernal ly applied energy . 37
hand starting ” 37
match ign iter 38
ret a rding spark ” 37
Starting of autom obi le orma rine engine .
stat ion ary engineStarting troublesSteam , condensation of , to
produce va cuum .
Stockport engineStopp I ng of automobi le or
ma rine engine .
station ary engine .
Street , RobertStroke Of engine .
rat io of , to bore
Tables, c ircumferen ces a nd
areasof c ircles . 22 1
compressi on temperatures 85
cyl inder d imensions . 1 01
flywheel coeffic ients 139
keys 14 1
logarithms 233
mach ine screw 2 19
Prony brake factors . 200
tap drill 2 18
trigonometric function 225
valve 1 19, 1 20
weight a nd spec ific heatof gases
wire a nd sheet meta lgauges
wrought I ron pipeTempera ture, cyl inderDiesel combustioncompression
Tests, appa ratusarrangement .assistants .
methodsl iqu id fuel enginesloads for
o o o o o o o o o o
o o o o o o o o o o
6 6 6 6 6
o o o o o o o o
PAGE
Tests,l og of 204
read ings, interval of 204report . 207
Thermal effic iency 213Throttle . 26Two-cycle engine, 5, 1 1 , 1 2 , 20, 67per cent increase of powerover four -cycle
ports for .
Valves, a ir , for vaporizer . 43angle seated ,
advantage . 1 1 5arrangement 1 2 1— 1 25aspirating , for carburettor 49care of . 29cool ing 28, 1 22 , 1 30design and proportion . 7, 1 16diameter 120Diesel . 44
dimensions (Table) 1 19effective open ing 1 17, 1 19
(Table) . 1 20fia t seated , advantage Of 1 15
24
material . 1 1 6mechan ica lly controlled , 20, 2 1m ix ing 43needle, for va porizers 43Robson engine 1 2size . 7 , 9stems . 1 21
tim ing 105timing , for hot tube ig
n itertwo-cycle, see Ports.
Va por, pressure of gassaturat ion pressure .
(Table)Vaporization .
Venturi tube in carburettordesign
Volati l ity a s aff ected byvaporization
Water gas, see Ga s .
Wa ter1jlacket , copper .
deptdrain inglength
WattW irin connectionsfor our—cyl inder
Wright engineWrist pin .