PAMPHLET 2

December 1970
TM70-975

Guy Bellows, Senior Manufacturing Engineer -

Advanced Manufacturing Process Engineering
Dean N. Tishler, Engineer -
Material & Process Technology Laboratories

A I R C R A F T ENGINE GROUP
C I N C I N N A T I , O H I O 45215
M A T E R I A L TRENDS
T R E N D S I N M A T E R I A L USAGE
Material trends for aerospace applications provide for greater strength
( W E I G H T P E R C E N T U S E D I N BASIC GAS T U R B I N E E N G I N E S )
and temperature capabilities. The improved material properties are
obtained along with increased difficulty in component manufacture,
GE DECISION AIIMg TI NdCo
including forming and machining. Machinability of the newer super alloys ENGINE YEAR COMPOSITES A L L O Y S A L L O Y S IRON SUPERALLOYS
can be as low as five percent of more conventional alloys.
J4 7 1945 0 22 0 70 8
J79 1955 0 3 2 80 15
T R E N D S I N GAS T U R B I N E R O T O R A L L O Y S TF39 1965 2 3 29 19 47
CF6 1968 4 2 27 21 46
FlOl 1970 3 4 20 15 58
XXX 1975 10 3 17 15 55
XXX 1980 15 0 15 15 55

Tensile
Strength, Ksi
DESIGN TRENDS
300

Trends i n engine component design are toward lowest welght and

maximum longevity and reliability. Alloys are required which are able
t o withstand high stress and/or temperature. These requirements create
a need for the stronger high temperature alloys. Mechanical properties
of these newer alloys vary greatly with different manufacturing processes.

MANUFACTURING TRENDS
I 1 I 1 1
1940 1950 1960 1970 1980 1990 The increased difficulty of machining the new alloys has created a need
Year Available
for better metal removal techniques. New cutting tool materials, more
machine power, and development of nontraditional processes (electrical,
thermal, and chemical) have been necessary t o machine the newer,

I MACHINABILITY INDEX
I stronger alloys. The effects of the conventional (grinding, milling) and
nontraditional processes on mechanical property are not fully known.

100
61112 (800 S F M I
+ -
Turning Tests The need is emphasized for determining the effects of material removal

80 - 30 rnin. Tool Life - processes on mechanical properties because of the sensitivity t o the
.015 in. Wear Land various removal processes. I n addition, the more prevalent use of thin
Machinability Carbide Tool sections, of which the affected depth layer of the removal process
Rating
Percent
-
6
60 - - (Altered Material Zone ')" is a greater percentage of the total thickness,
requires investigating the variability i n mechanical properties. One study2

40 '
4340
- examined the effects of metal removal methods on the fatigue life of
lnconel 718, Titanium 6A1-4V and 4340 Steel Rockwell hardness Rc=50.
The study described the effects of both gentlelfinish and abusive/roughing

20 -
Ti-6AI-4V

lnconel 7 1 8
- machining conditions on mechanical properties. The study further showed
that improper use of most processes or post-machining processes could

0 ,
I Rene 95
1 1,
cause degradation of cyclic life.
" References - on back of pamphlet.
2
SURFACE INTEGRITY FAILURE EXAMPLES TEST SPECIMENS AND EVALUATION TECHNIQUES

The following t w o figures show a failed M50 bearing from a J79 engine The various material testing techniques were reviewed in detail during
test. The bearing failed in slightly over four hours from improper or the W~ight-PattersonAir Force Base (WPAFB) study on "Surface Integrity
abusive grinding of the race. The surface had a .004 inches thick white of Machined Structural components".? Suggestions for a systematic test
or untempered Martensite layer and was harder by t w o Rockwell points. program were made as follows:
Cracking was found to a .016 inch depth. Proper grinding techniques Minimum Data Set (MDS)
would have prevented this failure. Standard Data Set (SDS)
Extended Data Set (EDS)

MINIMUM D A T A SET

The MDS includes the basic surface topogr-aphy that i s normal in

.I25 in. assessing a machined surface plus some of the simple metallurgical
measurements. These are:
Surface Finish-The profile trace shown can be measured by an instrument
such as a Profilometer which typically utilizes a .0005 inch radius probe.
The units of measure for A A (arithmetic average) or pzak-to-valley are in
MACROPHOTOGRAPH O F M50 BEARING RACE SURFACE SHOWING microinches (pin). A A readings are Center Line Average (CLA) readings.
CRACKS I N SURFACE CAUSED BY ABUSIVE GRINDING.
The A A i s an average of the heights of the peaks and depths of the valleys
as shown in the sketch. The equation mathematically describes the basic
method of computing a surface finish value.

.004 in.
f-. 0 0 0 0 2 tn,

-a

MICROPHOTOGRAPH O F M 5 0 S T E E L BEARING CROSS-SECTION SHOW-

I N G ABUSIVELY GROUND, "BURNED", SURFACE O F UNTEMPERED
MARTENSITE. Typical Surface F~ntshProfde

A premature failure of a Titanium 6AI-4V blade in a lab fatigue test is an

example of abusive post-machining processing. The microphotograph
shows a contaminated layer produced during a brazing operation.
where A A i s the surface finish value, X i s the distance between two
readings, n is the number of readings and a is the magnitude of the curve
from some reference"point. I n the sketch, the peak-to-valley reading would
be the max. peak (height) A t o the deepest valley B.

Metallographic techniques are useful for determining visual changes in

the surface layer at high magnifications (1000X preferred). Good edge
retention i s mandatory and requires careful mounting and polishing
T I T A N I U M 6A-4V BLADE SURFACE SHOWED F A I L U R E I N T H E CON-
T A M I N A T E D L A Y E R WHICH WAS PRODUCED D U R I N G BRAZING
OPERATION. 1OOOX
techniques. Contamination of the surface during mounting by any
process such as plating must be avoided. Changes in microstructure,
I
plastic deformation, heat affected layers, microcracks, processing inclu- cantilever bending (completely reversed bending, A = a ) . A i s the
sions, surface roughness, pitting, tears, laps, and many other effects can symbol for stress ratio of alternating stress to mean stress. Car-e must be
be readily identified. taken w i t h specimen edge preparation to prevent degr-adation of cyclic
life by edge failure. A n acceptable method of edge preparation is to
polish the edge followed by shot-peening of the edge radii. The HCF
test i s generally considered to be over a cyclic life range of 105 to 107
cycles. The fatigue strength or fatigue limit at 107 cycles i s usually
considered the endu~ance limit. A simple form of a fatigue curve or
S-N cul-ve where S i s the symbol for stress and N i s the symbol for number
of cycles i s shown below.

1
. . ,' ..
(a) (b)
S U R F A C E C H A R A C T E R I S T I C S O F RENE 4 1 ( A G E D , 4 5 Rc) P R O D U C E D Stress S
B Y ECM Or O
(a) Gentle Condition - No significant effect. (b) Off-Standard Condition - Same
general appearance. Surface Finish: (a) 1 0 A A . (b) 5 0 A A .

Mlcrohardness tests at various depths beneath the surface are conducted ,lo5 1 lo6 ,lo7
t o determine hardness profiles. The profiles help to identify changes in Cycles to Failure - Nf
the wrocessed surface.
'YPICAL F A T I G U E OR S-N C U R V E
STANDARD D A T A SET
Residual Srress tests are conducted using a specimen length - 4.0 inches
The SDS provides more in-depth data. In addition t o all investigations width - .75 inch, and thickness = .060 inch and a deflection measurement
listed under the MDS, High Cycles Fatigue (HCF) tests and residual fixture shown below.
stress tests are included in the SDS.

High Cycles Fatigue tests can be accomplished using a specimen with

stress concentration or K t = l .O shown below. The triangular specimen
Locating
shape provides a uniform stress field and i s tested at room temperature in Pins Specimen
I 1

Measurement

-
2
Radius 3.5 in. Gage

Section A A
D E F L E C T I O N M E A S U R E M E N T F I X T U R E A N D SPECIMEN

C A N T I L E V E R B E A M H I G H C Y C L E F A T I G U E SPECIMEN, Kt=l.O
5 6
 L o w Cycle Fatigue tests can be axial or bending; however, the bending
The technique involves successively etching away thin layel-s of the
test i s more sensitive to surface changes. One test utilizes a rectangular
stressed specimen surface (only one side has specific process surface
cross-section specimen loaded i n pure bending and cycled at stresses
preparation.) Deflectioris are measured between etching operations and
sufficient to cause failul-e in less than 105 cycles at room and elevated
the results are utilized in an equation developed by F. steblein3.
temperatur-es.
Because of the complexity of the equation and involved mathematics,
the process has been cornputel-ized for ease i n obtaining residual
Specially selected tests depend upon design needs. Creep tests in which
stress measurements.
a specimen i s subjected to a constant load at elevated temperature are of
interest to designers when considering stressed parts subjected t o elevated
THE EXTENDED D A T A SET
temperatures. A typical example of a creep curve is shown below. The
The EDS includes both the MDS and SDS. The information generated relaxation test i s also of interest for elevated temperature service for
i s generally design data, and the category would only be engaged if such stressed parts such as a bolt in tension. A typical stress relaxation curve
data were necessary. In addition to the previously listed tests, stress i s shown below.
corrosion, low-cycle fatigue ( L C F ) , and other specially selected tests
such as creep, relaxation, stress rupture, tensile, oxidation, etc., as
required for a particular design problem are conducted. Tests at both
room and elevated temperature are included.

Stress Corrosion tests can be of the alternate immersion type ( i n salt water
for 10 minutes, out for 50 minutes and repeat) at a given stress level,
Strain, E
generally 75 percent of the yield strength, or of the hot salt corrosion
type. The type of test is dependent upon the base alloy, such as iron,
titanium, etc. The hot stress corrosion test i s conducted on a specimen
exposed to an elevated temperature while covered with a salt coating,
which i s subsequently tested for loss i n ductility. A fixture arrangement
for the alternate immersion test is shown below. In general, steels are
tested utilizing the alternate immersion test, and titanium alloys are
tested using the hot salt stress corrosion test technique. Stress corrosion Time, t
tests are not generally conducted for nickel base alloys. Primarily, nickel Typical Creep Curve
alloys are subject to stress oxidation which would be included in the
specially selected tests.

Stress. 0

I Fixture 1
Time, t
Typical Stress Relaxation Curve

STRESS C O R R O S I O N F I X T U R E A N D SPECIMEN
EFFECTS O F PROCESSES ON ALLOYS
I
A

Freqi!ently parts machined when utilizing the conventional metal removal

Recast Laver 1
processes show greatel fatigue life than do those processed by the non-
tr-aditional methods (Electrochemical Machining-ECM, Electrical Dis-
charge Machining-EDM). Post-machining PI-ocessing such as shot peening
can improve the cyclic life of machined parts. The fatigue life can be
eithel- equal to or greater- than the life of parts pr-ocessed without
shot peening.

Surface finish measurements do not necessarily correlate with fatigue

life. Data supporting the observation i s presented for lnconel 718 and
Titanium 6AI-4V in following sections. Material hardness appeal-s t o be
insensitive t o many processes. Ground specimens of Inconel 718 have
the same hardness for gentle and conventional conditions while the
fatigue stl-ength vat-ied from 60 ksi for the gentle condition to 39 ksi
I Depth Beneath Surface, Inches

for the conventional grinding condition. Residual stresses i n specimens

caused by the removal or post-machining processes have better correlation
Key: A - Roughing E D M and Shot Peen (Level 11
w i t h fatigue life. Tensile residual stresses are associated w i t h shorter
fatigue life while specimens w i t h compressive residual stresses show €3 - Gentle E C M and Glass Bead Peen

longer fatigue life. Applied tensile stresses on a specimen are reduced C - Gentle Grinding
where the residual stresses are compressive. On the contrary, where D - ECM
the residual stresses are tensile, the applied tensile stresses are increased, E - EDM
which accounts for the shorter fatigue lives.

The following sections l i s t some data for three alloys: lnconel 718,
Titanium 6AI-4V and 4340 Steel. Additional information can be found
by referring to reference 2.
SHOT PEEN CONDITIONS

IMPACT O F PROCESSES O N NICKEL BASE A L L O Y S

LEVEL 1 LEVEL 2

Manufacturing processes when applied t o components or specimens have Shot Size S110 S110
a great effect o n material properties. In one program2, various machining
Intensity .006 - .008A .006 - .008A
methods and post-machining processes were investigated. Some of the
effects on lnconel 718 in the solution treated and aged (STA) condition Coverage 300% 125%
are described below. Shot Hardness 50 - 55 Rc 50 - 55 R c

A summary of hardness results are shown in the next figure for

lnconel 71 8 STA plate.

A summary of HCF results i s shown in the next figure. I t can be seen

readily that shot peening provides a pronounced improvement in fatigue
life at room temperature. I t is interesting to note that Gentle and
Abusive ECM specimens both have the same fatigue strength (39 ksi)
and Finishing and Roughing EDM specimens have similar fatigue strength
(22 ksi) while the surface finish varied considerably.
10
 Curves of residual stress as a function of depth beneath the surface for
Inconel 718 STA for some processing methods are shown below. Note the
wide range of tensile to compressive rresidual stress between gentle and
8 0 k y G w conventional grinding achieved by varying operating parameters

Maximum
Stress. Ksi 60
A Conventional Grinding
B Gentle ECM
C Gentle Grinding
D Gentle ECM and Shot
Peening (Level 1 )

Tension

1 o6
Cycles t o Failure, N f

F A T I G U E C H A R A C T E R I S T I C S O F I N C O N E L 718 STA, C A N T I L E V E R
Residual Stress. Ksi
BENDING, A=@? Kt=l.O, R O O M T E M P E R A T U R E .

Key t o Figure, lnconel 718 S-N Curves

Endurance, Surface
Curve Condition L i m i t - Ksi Finish-AA Compression

A Gentle E C M and Glass Bead Peen

B Gentle E C M and S h o t Peen (Level 1) *
C Abusive ECM and S h o t Peen (Level 2)
D Roughing E D M and S h o t Peen (Level 1)
E Abusive ECM and S h o t Peen (Level 1)
F Finishing E D M and S h o t Peen (Level 1)
G Gentle Surface G r i n d

i
i Gentle E C M
Abusive E C M
Conventional Surface G r i n d

I Finishing E D M
Roughing E D M 0 .002 .004 .OM .008 .010 .012

Depth Beneath Surface-Inches

* See Page 10 f o r shot peen conditions.

R E S I D U A L STRESS O F I N C O N E L 718
IMPACT OF PROCESSES ON STEELS IMPACT OF PROCESSES ON TITANIUM

The manufacturing processes effects upon fatigue strength of AlSl 4340 The WPAFB Surface lntegrity2 study also investigated the processing
Steel, which are similar t o the results shown for lnconel 718, are effects on Titanium 6AI-4V Beta rolled plate. Again, the processes
shown below applied to specimens exhibited varied effects as shown below. Trends in
materia! property response seen fol both lnconel 718 and 4340 Steel
specimens were also found for Titanium 6AI-4V specimens. Generally,
mechanical removal processes done in a gentle manner provided longer
specimen fatigue life when compared to the nontraditional processes.
Unlike the nickel alloys, titanium alloys frequently show a greater effect
from surface roughness.
Maximum
Stress, Ksi

Maximum
Stress, Ksi

I T O~ I07
Cycles t o Failure, N f

F A T I G U E CHARACTERISTICS O F A l S l 4340 (QUENCHED A N D TEM-

PERED, Rc=50), C A N T I L E V E R B E N D I N G , A+ Kt=l.O, R O O M
TEMPERATURE.
I05 IO~ IO'
Cycles t o Failure. N
f
K e y t o Figure, 4 3 4 0 S-N Curves
F A T I G U E C H A R A C T E R I S T I C S O F B E T A R O L L E D T I T A N I U M 6AI-4V.
Endurance, Surface
C A N T I L E V E R BENDING, A=? Kt=l.O, R O O M TEMPERATURE.
Curve Condition L i m i t - Ksi Finish-AA

A Gentle Surface G r i n d and S.P. (Level 1) * 112 25

B Gentle Surface G r i n d 102 45
C Abusive Surface G r i n d a n d S.P. (Level 1) 92 40 K e y t o Figure, Ti-6AI-4V. S-N Curves
D Conventional Surface G r i n d 70 40 Endurance, Surface
E Abusive Surface G r i n d 62 50 Curve Condition L i m i t - Ksi Finish-AA

E n d M i l l i n p D u l l Cutter 77
Peripheral Milling-Condition Sharp Cutter 70
* S h o t Peen (S.P.) conditions. See lnconel 718 data. same as Level 1. E n d Milling-Sharp Cutter 64
Gentle Surface G r i n d 62
Gentle Hand G r i n d 57
Gentle Chemical Machining ( C H M ) 51
Abusive C H M 45
Peripheral Milling-Dull Cutter 32
Abusive H a n d G r i n d 30
Abusive Surface G r i n d 13
The milling conditions, end cutting-milling on center and peripheral
cutting-climb milling, show an interesting trend concerning the "gentle"
condition. Generally, a sharp tool i s considered for use in a "gentle"
condition and a dull tool for an "abusive" condition. The peripheral
cutting operation showed that the sharp cutter, "gentle" condition
produced much greater fatigue strength than did the "abusive" condition
using the dull cutter. The end cutting conditions showed for a sharp
cutter "gentle" condition that the fatigue strength was lower than
the strength for a dull cutter "abusive" condit~on; however, the 33/1 in. AA I
difference was very much smaller than that shown for the peripheral
technique. The data re-emphasized the necessity to investigate each
surface integrity situation separately for the effects of a specific level of 29C( in. AA 11
processing on a specific state of the material.

POST PROCESSING TREATMENTS

Fatigue tests of specimens w i t h an unblemished stress-free surface will

generally exhibit lower fatigue life than do specimens with a mechanically
produced surface containing compressive residual stress. The stress-free
surface could be produced by the ECM process. Specimens w i t h mechanic- 7/l in. A A 1
ally produced (i.e., milling) surfaces, generally have a compressive
residual stress and a specimen w i t h compressive residual stresses i n the
surface will exhibit longer fatigue life than either specimens w i t h stress- 12C( in. A A I t
free or tensile residual stressed surfaces. Several post-machine processing
treatments are used t o develop compressive residual stresses or reduce
ECM-INCONEL 718 - COMPARISON O F SPECIFICATIONS A T
tensile residual stresses i n the surface. The most commonly used methods
1800°F S O L U T I O N H E A T T R E A T . 1OOOX
are shot and glass-bead peening. Roller finishing, tumbling, burnishing,
and post-heat treatments are also employed. Some fatigue resulrs of
shot and glass-bead peened specimens were given i n previous sections.

KNOWLEDGE OF MATERIAL STATE A N D

HISTORY NECESSARY

It i s necessary t o know the condition (prior history) of the material

upon which the processing operation will be performed. A n example of
ECM processing of lnconel 718 material supplied t o different specifica-
tions i s shown below (solution-heat treatment temperature was changed).
Using identical machining conditions, surface finishes varied appreciably.
Knowledge of the previous history was necessary to properly process
the material.

P O R T I O N O F ECM'D TEST POCKET WITH BOTH M A T E R l A L

PROCESSES (SEE PRECEEDING MICROGRAPHS). 3X
S U R F A C E C H A R A C T E R I S T I C S O F I N C O N E L 718 (STA. 44 Rc) PRO
D U C E D B Y E C M P L U S PEENING.
T h e effect o f shot peening is seen as strain hardening and twinning lines i n the
structure beneath the surface. lOOOX (a) Gentle ECM Plus Shot Peening
(Level 1). Finish: 7 4 A A ( b ) Gentle ECM Plus Glass Bead Peening. Surface
Finish: 6 9 A A .

S U R F A C E C H A R A C T E R I S T I C S O F A l S l 4 3 4 0 (QUENCHED A N D T E M -
PERED, 5 0 Rc) PRODUCED BY A B U S I V E G R I N D I N G PLUS SHOT
PEENING.
Abusive Grinding Plus Shot Peening (Level 1) - Untempered martensite f r o m
abusive grinding is evident, b u t w i t h o u t visible indication o f shot peening
effect. Surface Finish: 40AA.

(a) (b)
S U R F A C E MICROSTRUCTURES O F T I T A N I U M 6AI-4V ( B E T A ROLLED.
32 Rc) PRODUCED B Y "ABUSIVE" E N D M I L L I N G CONDITIONS. lOOOX
(a) Peripheral Cutting Surface Finish: 5 9 A A (b) End Cutting Surface
Finish: 67AA
18
REFERENCES AND GENERAL INFORMATION

1. "Introduction to Surface Integl-ity" - Bellows, G., and Tishler, D.N. -

Pamphlet 1 in Series of 4, GE Co., AEG (Cincinnati) TM70-974
October 1970.

2. "Surface Integrity of Machined Structural Components"-Koster. W.P.;

Field, M.; Kahles. J.F. - Metcut Research Associates for USAF;
MMP 721-8 Final Report AFML-TR-70-11.March 1970.

3. "Spannungsmessungen an einsertig abgeloschten Knuppeln"-Stable~n,

F., - Kruppsche Monatschefte, Vol. 12 (1931), pp 93 98.

4. "Surface Integrity of Electrochemical Machining"-Bello\\'s, G.--GE Co-.

Aircraft Engine Group, Evendale, Ohio, RGOAEG172, July 17, 1969.

5. "Impact of Nonconvent~onal Material Removal Processes on the

Surface Integrity of Materials" - Bellows, G. - ASTME Technical
Paper M R68-518, Januray 1968.

SOURCES FOR FURTHER INFORMATION

Air Force Machinability Data Center (AFMDC), 3980 Rosslyn Dr.,

Cincinnati, Ohio 45209. Phone (513) 271 -9510, Supervisor technical
inquiries.

Guy Bellows, Sr. Mfg. Engr., Advanced Mfg. Proc. Engineering,

Evendale Plant-Mail drop E-70, GE Co., Cincinnati, Ohio 45215.
Phone (513) 243-4976 or dial cornrn. 8-332-4976.

Dean N. Tishler, Engr. Material and Process Tech. Labs, Evendale

Plant-Mail drop M-88, GE Co., Cincinnati, Ohio 45215. Phone
(513) 243-6771 or dial comrn. 8-332-6771.