“'0'

in
O
(I!
M
m
9—3
«’3
31
TC
,2?)
H
Ti

WI

‘

\
\

1W1

SIHIIWHNHIMHIHHH

 

THS_

A STUDY OF THE CAUSES OF
VARIATWNS IN THE
PHYSICAL PROPERTIES OF
CARBON STEELS FOR CRANKSHAFTS
THESIS FUR THE DEG-BEE OF M. 3.
Robert Guthrie
1932

   

veal“; Vb-

.j 'I‘|U£-\‘

 

‘ . , ‘.
l.- {1.41 7'_” '

 

   

.l’o?1.oea....f)..i6.l i: .v . . v.1..l1s q ... . .. .. .A . . , . . ...
v. . , V . .

 

A STUDY OF THF CAUSFS OF VARIATION IN THE

PHYSICAL PROPFRTIES OF CARBON SIRELS

FOR CRANKSHAFTS.

A Thesis Reapectfully Submitted to the Faculty of
Kichigan State College for Partial Fulfillrent
of the Requirements for the Degree of

Master of Science.

Robert Guthrie.
w

1932.

 

b"

u.

TABLE OF CONTENTS.

Introduction

Purpose

Method of Study
Discussion of Results
Conclusions

Bibliography.

.
V'
“*9"
‘5 6
up»
D
fl
J1

I E‘ITRODUC‘TION .

With increasing power and speed in automobile engines there
has developed greater and greater demands on the steel of the
crankshafts and connecting rods. These parts have had to reet
higher physical phyeieal property requirements, making it essential
that the manufacturer study all the possible factors which affect
the strength and toughness.

Strength, or hardness, which is the ability of the steel to
resist deformation from external forces, may be measured by the
normality of the steel and indicated by the expansion. The
normality is a measure of the ability of the steel to harden. A
normal steel being defined as one nich will harden all the way
through from the surface to the center.1 The expansion of the
steel during hardening and tempering is proportional to its
hardenability; a steel with the maximum hardenability will show
the maximum eXpansion.

Toughness, which may be defined as resistance to fracture
after deforration has begun, way be determined in two ways: by the
grain size‘by the NcQuaid-Ehn test and by the value obtained on
the impact tester. The McQuaid-th test consists of classifying
specimens nhich have been carburized at 1700 deg. Fahr. for eight

3 . . .
hours and allowed to cool in the furnace. The cla551fication is
carried out at a magnification of 100 diameters. The chart consists
of eight grain size numbers running from number one with one and

one-half grains per square inch up to nurber eight with ninety~six.

grains per square inch. It has been found by Grossman, VcQuaid,
and others that a coarse grained structure based on this classification
is usually an indication of brittle steel while a fine grained
. 4
structure su*gests toughness- The impact value is a measure
of the toughness in that it gives the amount of energy required

to fracture a standard test bar. The tougher the steel the higher

is this value.5
OBJECT.

The purpose of this paper is to study the relationship
existing between normality, dinension changes, grain size under the
YcQuaid-Thn classification, and Izod impact values of pla'n

carbon crankshaft steels.
APPARATUS.

The instrument used to measure the dimensional changes of the
bars was a Gaertner Comparator (Fig.L) which had a range of 200
millimeters.

The micrometer head is about eighteen centimeters in diameter
and is divided into one thousand parts, exery ten of which is

1:} l

,_,,
marked sowreadings of one micron (.0001 or.) can be made directly

, .

and fractions can easily be estimated.

The instrument used for breaking the bars was a standard Izod

Impact Tester with a range of 120 foot- pounds. The scale of this

instrument could be read accurately to one-half of a foot pound.

Comparator.

Gaertner

_

 

 

M
m
.
.
m
.
~
“
~

,‘

 

LIIZOSa

A standard metallurgical microscope was used to detemrine
the grain size and the normality, using a magnification of 100

and 500 diameters respectively.

PRO u KKK-1’"? .

tandari S.A.E. 1045 steels used in crankshafts were obtained

free various manufacturers in order to introduce variables of

’3)

(f)

neciwens cut from the center of

manufacture as much as p: sible. ,

V

each bar were then carburized in the furnace at 1700 xcfl. Fahr.

(927 den. Cent.) for eiyht hours, nolishei, etched, and rated

-0 g e s 1 u f‘ ‘u .
accoriiinr to the TcQuaid-Enn let. Che opec1r~ezs were. also

1‘
A

I. In hath o~

Ll.

examined for abnervalit ttese classi icctions the
listinrs accentei were t‘e chew? results of two independent
observers.

The rerainder of the bars were then cut into specimens six
inches long and turned to a diameter of .500 in. with an allowance
of plus or minus .002 in. These limits were chosen inorder to
have the minimum variation in the rate of heat absorption due to
variations in the dimensions of the piece. Grooves were out about
four inches apart around each bar while the work was revolving in
the lathe.

The distance between the inner edges of the grooves was
measured accurately with the Gaertner Comparator.

The bars were then heated to the prOper hardeninr temperature
in an electric muffle furnace, held at heat for thirty minutes,

and quenched in water. After leing allowed to stand overnight

they were terpered to a Rockwell Hardness of 20-25 on the C-scale

Safiple a. T. "arderirg Temnering
NJ. Yo. ”reatcert Treatnert
“eh. fshr. Deg. Farr.

1 1C45 1500 water 1000 air

2 lrn; 1800 water 1090 31?

3 174‘ 1’0“ watrr lolO air

4 l‘fib 1500 water 1000 air

5 1’45 150? vzter IFFC air

6 1045 1500 wzter 1000 air

7 1045 lSPO water 1000 air
8 lCfib lECO water 1030 air
9 1045 1500 water lCOG air

10 10:0 1525 water 1000 air

15 1C45 1500 water 1030 air
,3 1040 15?5 water 1000 air
24 1050 1500 water 1000 air
25 1040 1535 water 1000 air
26 1040 1525 water 1000 vir.

These specincn" were held at heat for thirty minutes in all

CQSPF.

which require} a reheutihg tevpcrature of 1030 deg. Fahr. (537 deg.6)
Care was taken at all tires that the bars were plrccd on an even
surface of the furnace floor in order to haihirin a rinimur
distortion.

After the ternering treatrent the length of the bars w;:

k .
A

‘

reasurei at four different positions around the brr, each position
being approxiretely ninety degrees Iron the precscdirc readin .
The per cent change in length as corpared to the original
censure; length was determined. Fach length change value, given
in Table II wrs the average of twenty four readings.

The bars were then nachined to the standard Izod Impact
round test bars and broken on the Izod Tester. The values
obtained are the average of twelve results for each type of bar

indie ted in Table 11.

3'3

From colurns two and four 0 Table II it will be seen that

x
the grain size as classified by the NcQuaid-th test decreases in(/ 2'

I
r I

alnost the same proportions as the Izod impact value increases. 3

-

Columns six and eight of Table II show the relation between
the normality and the dirensional changes. Those steels thich
showed a decrease in length after heat treatnent are marked
negative while those which increased in length are marked positive.
The degree of abnormality of each steel was determined by its
tendency to divorce cementite in the case of the VcQuaid-Ehn

test pieces. Those steels showing no divorced cerentite being

normal. In the case of sarples 2 and 26 there was a question as

.doapewnpdoo encoded“ uwufiowou o>apowon uncauaunxo epooacnfi awnacaon abapauom *

 

 

 

 

 

 

 

 

 

 

 

 

 

opoquOh
999.9- 9 49990994 hpo> s 9 a 99.49 om
994.4- 94 49399994 94 9 9 9.59 9
494.- 04 44990994 04 9 99 9s.s9 9
994.- 99 .94 s499w9am 9 4 9 99.59 99
944.- em .94 hap9w999 4 4 H 99.04 4
mmo. a .94 hap9wflam 9 4 4 9.04 4
«so. 4 .94 h999wflam 9 4 m 9.04 9
«mo. 9 .94 h499w9am ham» 9 4 9 99.94 9
999. 99 .94raaanoz 99 4 9H 9.44 s
049. m .94 99w499 m 4 04 9.94 a
o» Haanoz .
94m. 9 498902 4 4 99 9.09 04
459. 9 Haanoz 9 9 cm 0.99 a
904. 4 495992 99 9 9 9.99 49
494. 49 495902 99 919 s o.99 .94
999. s 9 aaanoz haauooH 4m 9 49 99.99 on
9999 99m .92 .02 .02 .oz .994399 .02
nodnaonxm oaqaum hpaauanoz oagsum ouflm manna oagfiom vomgaH couH oagsam
9 s. 9 9 4 9 m 4
.oz.qasaoo

« .oznmnmda

 

to just how they should be classified. These two specirens
seemed to be normal but at the same time, showed traces of
abnormal cementite. If the value of the length changes is taken
into consideration, however, specimen two seems to be listed

correctly while specinen twenty—six should be regarded as abnormal.
CONCLUSIONS.

The following conclusions may be drawn from the work of this
investigation upon the causes of variations in the physical
properties of S.A.T. 1045 steels:

l. The greater the expansion during heat treatnent, the
greater is the hardening ability. The normality, or
hardenability, is definitely related to the length
changes due to hardening and tenpering.

‘9. The grain size does not seem to bear any definite
relation to the nornality, or hardenability, of the
S.A.E. 1045 steels discussed in this paper. A
comparison of the results will show that the most
normal specimen had a fine grain size while the nest
abnormal specimen had a coarse grain size.

3. Variations in the toughness of S.A.E.1045 steels are

caused by variations in the grain size.
ACKNOWLEDGMENT .

The writer wishes to take this oppofiunity to OXPTCBS his
gratitude to Prof. F.G. Sefing of the Departrent of fechanical

Pngineering for his valuable surfiestions in regard to this work.

BIBLIOGRAPHY.

The Study of Nonnal and Abnormal Steel. Epstein and Rawdon.
Trans. A.S.S.T. Vol. XII September, 1927
Normality of Steel. John D. Gat
Trans. A.S.S.T. Vol. XII September, 1927
The Hardenability of Steel John D. Get
Forging-Stamping-Hest Treating - May, 1927

Austenite Decomposition and Length Changes in Steel.

E.C. Rain and W. Waring

Trans. A.S.S.T. Vol. XV January, 1929
Dilation of Steel on Quenching. G.h. Eaton
Trans. A.S.S.T. Vol. XVI December, 1929

Effect of Normal and Abnormal Steels on Case Carburizing.
McQuaid and th.
Trans. A.I.M.E. _ Vol. 67

Alloying Elements and Grain Size in Ketals. M.A. Grossmann.

A.S.S.T. Detroit Chapter April, 1932.

New Methods of Interpreting Notched Bar Impact Tests.
Trans. A.S.S.T. Vol. VII
Resume' of Impact Testing.
Trans. A.S.T.M. Vol. 22 Part 2
Endurance and Impact Tests of Petals.

Trans. A.S.T.Y. Vol. 16 Part 2

 

 

 

 

a o . ' . O I . v
1-51.5}!017. l2» 7...}. 3L. 1-.....

e

llfi‘l a

I‘ .0 .‘uflu.9|l ..

\ III .11:

 

.. like...
I}

.4-’ -~

QL... .

t A
.n

NW.

‘

,1??? 3.?

 

TATE UNIVERSITY LIBRARIES

MICHIGAN s
3 1

293 030l2 0201