mTn^TTXTTrt . -r I-
1 TR-EN-78-04
'
r^* v
I I
OF A PROCESS FOR CREEP
, APRH1978
I:
I
II W. BERNER
S. BHATTACHARYYA
i
' , ,
TECHNICAl REPORT
,
IIT RESEARCH INSTITUTE
~ DISTRIBUTION STATEMENT
Approved for public release; distribution unlimited.
ENGINEERING
El DIRECTORATE .
.
;
' ^
.
.
L '' - :
. - - ;- "- ' - x
"- \
A-
% 1 ^\, y
'--, i A-' * - -
r-V'-
,
/;-
r V/ :
v '
PISTRIBUTION INSTRUCTIONS:
k
Dest
Destroy this report when ft is no longer needed. Do not return it to the
or,9iator.
CSCLAiHER:
^ \. .,:. /-'..
. ;
I
'
(- 1 , ,'
.
- ^
'
,
. , , * . ^ - ... - - .
^
>
>? t- T 1 - .r
V 7 - 1 ; i ; ^- r .^ -
f > -
! 1-
^^\ -t :
v3
/. -/ -1' ^
. -
UNCLASSIFIED
SECURITY CLASSIFICATION OF THIS PAGE fWtWI Dote Entered)
READ INSTRUCTIONS
REPORT DOCUMENTATION PAGE BEFORE COMPLETING FORM
2. GOVT ACCESSION NO 3. RECIPIENT'S CATALOG NUMBER
1. REPORT NUMBER
TR-EN-78-04
5. TYPE OF REPORT & PERIOD COVERED
4. TITLE (end Subtitle)
N. C. Birla S. Bhattacharyya
DAAA09-75-C-2056
K. M. KulkarnI
W. Berner 10. PROGRAM ELEMENT, PROJECT, TASK
9. PERFORMING ORGANIZATION NAME AND ADDRESS AREA & WORK UNIT NUMBERS
I IT Research Institute
10 West 35th Street AMS Code 3297.06.7^09
Chicago, Illinois 60616
12. REPORT DATE
11. CONTROLLING OFFICE NAME AND ADDRESS
CDR, Rock Island Arsenal April 1978
13. NUMBER OF PAGES
ATTN: SARRI-ENM
Rock island, Illinois 61299 128
15. SECURITY CLASS, (of this report)
14. MONITORING AGENCY NAME & *ODRBSS(ll dillerent from Controlling Office)
UNCLASSIFIED
15a DECLASS1FICATION/DOWNGRADING
SCHEDULE
19. KEY WORDS (Continue on reverse side if necessary end identify by block number)
20. ABSTRACT (Continue on reverse side if necessary and identify by block number)
This program was aimed at a novel manufacturing method for the reduction of the
fabrication costs of aluminum weapon components. Its technical goal was to
combine the ability of powder metallurgy processing to produce a well-
proportioned preform and of creep forging to make a precision part. The work
was done in two phases. The first phase was devoted to characterizing ele-
mental 7075 blended aluminum powder, establishing requisite process criteria,
and formulating a preliminary process specification. Five variables were
tmaatiaated "nHpr a staLlatlcaJ tkalaaed pxnprimp.ntal nronrnm. Thfv aaca-^
RM
DD , ^N 73 1473 EDITION OF 1 NOV 65 IS OBSOLETE I UNCLASSIFIED
SECURITY CLASSIFICATION OF THIS PAGE (When Data Entered)
UNCLASSIFIEP
SECURITY CLASSIFICATION OF THIS PAGEQWian Data Entered)
(1) particle size, (2) preform density, (3) sintering temperature, (M forging
temperature, and (5) percent forging deformation. Using the requisite process
criteria set up from these variables, a preliminary process specification was
formulated and further refined by isostatic pressing of trapezoidal preforms
and their forging into "T" shapes. The resultant forgings had elements of
forward extrusion, backward extrusion, and lateral flow and helped define the
optimized parameters necessary for creep forging.
1 1
UNCLASSIFIED
SECURITY CLASSIFICATION OF THIS PAGEfWhen Data Entered)
FOREWORD
This report covers the work performed under Contract No. DAAA09-75~
C-2056 by Dr. K. M. Kulkarni, Dr. N, C. Birla, Mr. W. Berner, and
Dr. S. Bhattacharyya of I IT Research Institute. The work was conducted
under the direction of the Engineering Directorate, Rock Island Arsenal,
Rock Island, Illinois, The report is designated Internally as IITRI-
B6137-6.
iii
TABLE OF CONTENTS
Page
1. INTRODUCTION 1
2. TECHNICAL BACKGROUND 6
2.1 Conventional P/M Processing, 6
2.2 Forging of P/M Preforms. .,.....,..,,,.,.. 10
2.3 Creep Forging-'-Basic Principle and Advantages. 10
2.^ Temperature and Strain-Rate Effects in Aluminum 13
2.5 Component Investigated , , l^t
2.6 Target Mechanical Properties ... 17
1 v
TABLE OF CONTENTS (cont.)
Page
vi
LIST OF TABLES
Table Page
1 Effect of Sintering Atmosphere on Strength. Properties
of Aluminum P/M Parts ,,,.,.,,,> , , , , , , , , , , 8
2 Properties of Forged P/M Parts. ,,,.,,.,,,,.,,, 11
3 Properties of Aluminum P/M Forgings Compared with
Conventional P/M Parts and Forgings , , . , , 12
k Typical Tensile Properties at Various Temperatures 1A
5 Chemical Analysis of 7075 Aluminum Alloy Powder 20
6 Sieve Analysis of 7075 Aluminum Alloy Powder . . , 20
7 Material and Process Parameter Levels ...,., 22
8 Statistical Design for Phase 1A Study ,.,.,..,,,.. 23
9 isostatic Pressing Scheme ,,..,.,,,,.,,.,,,, 26
10 Green Density as a Function of Compaction Pressure and
Powder Particle Size, Aluminum 7075 Alloy Billet 30
11 0,2 Percent Yield Strength of Forgings from Parametric
Study of Process Variables, .,,,,,,.,,,..,.,. 36
12 Ultimate Tensile Strength of Forgings from Parametric
Study of Process Variables, ,..,,.,,, , . , 37
13 Tensile Elongation of Forgings from Parametric Study
of Process Variables, ,.,.,,.,,,,.,,.,.,,, 38
]k Processing Combinations for Al 7075 P/M Alloy Forgings
with Tensile Properties Above Target Values , . . 39
15 Tensile Properties of Al 7075 P/M Alloy Forgings. ...... 41
16 Forging Data for Making T-Shapes , 50
17 Tensile Properties of Specimens from T^Shaped Al 7075
P/M Alloy Forgings 52
18 Simulating Material Forging Tests with Preforms
of Various Geometries .,,,,,,,,,,,,,, 73
19 Tensile Properties of a Forging Made Directly
In Finishing Dies ,,,,,,,,,,,,,,,,,,,,,, 96
20 Cost Comparison for Cams Machined from P/M and
Conventional Forgings ,,,.,,.,,,,,,,,,,,,, 109
1-1
Data about Preform Weights. ,..,,,,,,,,,.,,., 118
2-1 Forging Data. ,.,,.,,,,.,,,.,,,,,..., 120
VI I
LIST OF FIGURES
i
Figure Page
1 Philosophy of the Proposed Effort 2
2 Finish-Machined Drawing for the Cam 3
3 Summary of Program Plan 4
4 Compacting Pressure vs. Green Density of Alcoa 201AB Blend. . 7
5 Effect of Green Density and Atmosphere on Sintered
Dimensions of Alcoa 201AB alloy 9
6 Strain Rate/Temperature Effect on Flow Stress for Al 7075 . . 15
7 Two Views of the Conventional Forging and the Cam Finish-
Machined from It.. 16
8 Phase I, Task A. Powder Characterization and Upsetting
Studies ig
9 SEM Photographs of Elemental Blended Atomized Aluminum
Powder Made to 7075 Alloy Composition 21
10 Target Dimensions of the Forged Pancake for Parametric Study. 25
11 Isostatic Pressing Tools and Pressed Billets 27
12 Photomicrographs of Sintered Al 7075 Alloy 31
13 As-Sintered Microstructure of P/M Al 7075 Alloy 32
14 Sintered and Forged Microstructures of P/M and Wrought
Al 7075 Alloy Specimens 34
15 Cracking in Forged Pancakes 35
16 Phase I, Task B - T-Shape Forging and Its Evaluation .... kk
17 Schematic of T-Shape Forging for Phase I'-B and the Preform. . 45
18 Die Assembly for Forging of T-Shape 46
19 Tooling for Isostatic Pressing of Trapezoidal Billets .... 47
20 Comparison of IsostaUcal ly Pressed Cylindrical and
Trapezoidal Billets 48
21 Examples of T-Shaped Forgings , , 51
22 Micrographs of T-Shaped Forgings in STA Condition 54
23 Photomicrographs of Two Transverse Samples from a T-Shaped
Forging 55
24 SEM and X-Ray Spectra of Samples from T-Shaped Forgings ... 56
25 Appearance of Cracks in Hard Chromium-Rich Particles, .... 57
26 Phase II; Process Optimization for Production Run of Cam . . 60
VIII
i LIST OF FIGURES (cont.)
'"'gure Page
27 Target Cam Geometry for P/M Forging 61
28 Schematic of Blocker Bottom Die . 63
29 Schematic of Blocker Top Die 64
30 Schematic of Finishing Bottom Die 65
31 Schematic of Finishing Top Die. 66
32 Blocker Dies. 67
33 Finishing Dies 68
34 Supporting Plates and Ejection System for the Finishing Dies. 69
35 Finishing Die Assembly in lOOO'-ton Press 70
36 Hydrocal Dies for Forging of Simulated Material ....... 72
37 Simulated Material Preform Shapes and Creep Forgings 7^
38 Initial Design of Aluminum Form for Rubber Bag 75
39 Tooling for Isostatic Pressing Preforms for Cam and Bags
after Evacuation, Ready for Isostatic Pressing 76
kO Modified Alumfnum Form Design for Rubber Bag 77
41 Rubber Inserts Developed to Control Geometry of Preform
Made by Isostatic Pressing 79
k2 Isostatical ly Pressed and Sintered Preform . 81
43 Four Blocker Die Forgings 82
kk Two Views of Forging Nos. 1 and 3 after Finish Die Forging. . 84
45 Finish Die Forging No. 3 after Dye Penetrant Test 85
46 Finish Die Forging No. 6 at 1/4 in. from Full Die Closure . . 86
47 One-Step Forging in the Finishing Die 87
48 Preform for Forging No. 14 89
49 Four Preforms Prepared to Study Effect of Preform Geometry. . 90
50 Forging No. 14(Sectioned), 16, and Two Views of Forging
No. 20 91
51 Sections of Forging No. 14 92
52 Micrographs of Forgings with Different Processing and
Heat Treatments ..... 9^
53 Examples of Forgings Made Directly In Finishing Dies
and Preforms for the Forgings 98
ix
LtST OF FIGURES (cont.)
Figure Page
This program was divided into two phases, I and II, and is summarized
in Fig. 3. The objectives of Phase I were to characterize the work mate-
rial establ i sh the requisite process criteria, and formulate a process
specification for prototype production. Phase I was divided into two tasks
Task A involved powder characterization, isostatic pressing, sintering,
forging, and evaluation of round pancakes so as to select preliminary pro-
cess specifications for Task B. Task B concerned forging of "T" shapes
from wedge-shaped preforms to determine forward and backward extrusion and
lateral flow properties of the preform. From Phase I, process specifica-
tions for Phase II were prepared.
Phase If started with the design and construction of the tooling for
both isostatic pressing of the preform and precision forging of the final
part. After the tooling was fabricated, the subsequent work concerned
optimization of processing parameters for the cam using the process spec-
ification developed in Phase I as a guideline. Several variations in pre-
form design, processing parameters, and die modification were required
prior to a prototype production run of the component. Phase 11 also
O 1 JZ
O
Q.
O
in
O
en
c
en
in c
in
c
in 8
*
0)
o * d) in
o JZ 4-"
!_
i
L. o o
in 10 0) <u
z:
1_ ->
u l_ a> o D-
<U 10 <u TO C I-
a 4-1 4J Q_ 4->
i c c
O E ro c E TO
TO D. o to > o D-
C -o O
o < 4->
o 0)
3 c 4-J
-o =3 4->
a) 0) <U
>
c
o
l_>
E
c
5
3 m
Cn
o
TO
PHASE I Powder Characterization
TASK A (Blended 7075 Al Alloy)
Prelimlnary Sintering
and Forging Tests
and Their Evaluation
RepIicate Tests
under Optimum Conditions
(Revised) Process
Specifications
Cam Forgings
and Thei r Evaluat ion
Prototype Production
Figure 3
The powder mass properties of major concern are flow rate, bulk and
tap densities, and compacting pressure-green density-strength relation-
ships. The temperatures and pressures required for compacting aluminum
alloys are fairly low. An example^.' for aluminum alloy 201AB which is
similar in composition to 2014 alloy is shown in Fig. 4. Notice that 90^
of the theoretical density is achieved at room temperature at a compacting
pressure of only about 25,000 psi.
(2) "Aluminum Powder Products for Powder Metallurgy Parts/1 Form F38~128l6
(Rev. 6-71), p. 8, Aluminum Company of America,
(3) "Improved Sintering Procedures for Aluminum P/M Parts," Form F38-12964,
p. 9, Aluminum Company of America.
5 10 15 20 25 30 35
Compacting Pressure. Tons Per Sq. In.
Figure ^t
Atmosphere
Strength Dissociated
Temper Vacuum
Properties Nitrogen Ammon i a
3
Alcoa 201AB, green density 90^ of theoretical (2.50 g/cm3)
0)
N +1
CO
o
E
O
s_
0)
c
TO
-C
TO
C
o
+
2.2 2.3 2.h 2.5 2.6 2.7
Figure 5
Work on alloys comparable to 6061 and 201^ shows ' that pressing
pressures of 20,000 to 50,000 psi were necessary in forging at 800oF.
Recent work(5) on alloys comparable to 7075 has shown that hot working
(extrusion and/or forging) of powder compacts can give properties gener-
ally better than the wrought product. These experimental 7075 type
forgings were up to 30 in. x 8 in. in an aircraft-related shape.
10
Table 2
11
Table 3
12
strength increases as it cools, realizing the benefit from the creep
forging process requires that the difference between the workpiece and
the die temperature be small, thus minimizing any temperature or heat loss
from the work material. When the die and the work material temperatures
are equal, the process is known as isothermal forging.
Aluminum and aluminum alloys are known to be strongly strain rate sen-
sitive at temperatures of forging interest.^-13) It is believed that the
13
strain rate sensitivity is due to the solutioning of second'-phase strength-
ening particles In the matrix. The effect of strain rate on flow stress*^'
of Al 7075 is shown in Fig, 6, and the effect of temperature on tensile
properties is shown in Table 4,^^'
Table k
Alloy Test
and Temp., 0,2% Y.S., UTS, Elong. ,
0
Temper F psi psi %
The preform for this forging will have a complex shape. Since it must
resemble, at least partly, the finished part In outline, Isostatic pressing
seems to be the only practical process for compaction of the preform.
(14) Aluminum Standards and Data, ^th Ed,, 197zt, The Aluminum Association,
tnc,, Table 2,2, pp, 31'-3/t,
]k
o
CM
in
o
04
I
O
o
(U
U
<u
in
i U>
l c
O (0 o
3
c o
TO
(U
I.
I.
<u
-a- Q.
i e
o
a:
+
LTV
h M i I iI- 4- I
o
in d LA
M
(9[eos 601) isd oi 'sisaj^s 01J
15
f
Figure 7
16
2.6 Target Mechanical Properties
17
3. PHASE I, TASK A; POWDER CHARACTERIZATION
AND UPSETTING STUDIES
The major variables and their levels selected for the evaluation are
shown in Table 7- The variables include particle size, preform density,
sintering temperature, forging temperature, and forging deformation.
Generally, the properties of a product are complex functions of the diffei
ent processing variables. Such interactions can best be identified by
multiple stepwise regression analysis.^'' For such analysis, a fractional
factorial plan is often adequate and minimizes the number of experiments to
be performed. In this program, a half factorial design was selected as
shown in Table 8, and this led to a total number of experiments of 36.
18
Elemental Blended - coarse (-40 mesh)
7075 Aluminum Alloy Powder - fine (-100 mesh)
- chemistry
Characterization - size analysis
- SEM
- bulk density
- compaction pressure
Parametric Study of Compaction, 20 and kO ksi
Sintering, and Upsetting
- sintering temp.
of Cylindrical Billets
900 to 1100oF
- forging temp.
650 to 850oF
- deformation 30, 50,
and 70%
- tensile and impact
Evaluat ion
of Forgings properties
- density
- micro- and macrostructures
Selection of Promising
Processing Parameters
Replicate Testing
Figure 8
19
Table 5
MIL-A-22771C
and Federal Powder Chemical
Specification Anal>'sis , %
Element QQ-A-367H, % Coarse Fine
na - Not analyzed.
Table 6
20
SEM No. 8226 200X
(a)
&~fc %& .
Figure 9
21
Table 7
No. of
Parameter Levels Code Remarks
22
Table 8
Particle
Size Sinter Forging Forging
Comb i- Distri- Preform Temper- Temper- Defor-
nation bution Density ature ature mation
1 la 1 1 1
2 1 1 2 3
3 1 1 3 2
4 2 2 1 1
5 2 2 2 3
6 2 2 3 2
7 2 1 2 1 1
8 2 1 2 2 3
9 2 1 2 3 2
10 1 2 2 1 1
11 1 2 2 2 3
12 1 2 2 3 2
13 2 2 2 2 1
]k 2 2 2 1 2
15 2 2 2 3 3
16 2 1 2 1
17 2 1 1 2
18 2 1 3 3
19 2 2 1
20 2 1 2
21 2 3 3
22 1 2 2 1
23 1 2 1 2
2h 1 2 3 3
25 2 2 2 3 1
26 2 2 2 1 3
27 2 2 2 2 2
28 2 1 3 1
29 2 1 1 3
30 2 1 2 2
31 2 3 1
32 2 1 3
33 2 2 2
3^ 1 2 3 1
35 1 2 1 3
36 1 2 2 2
23
3.3 Experimental Details
The size of the preforms for forging is dictated by the size of the
pancake desired after forging. In an open die forging, the outside layer
does not densify as well as the inner core. Therefore, to allow enough
sound material for test specimens, it was decided to forge the pancakes
to 3.8 in. diameter and 0.75 in. thickness, as shown in Fig. 10.
To obtain a 0.75 in. forged thickness under 30, 50, and 70^ deforma-
tions, the corresponding thicknesses prior to forging should be 1.07, 1.5,
and 2.5 in. For a 30% density sintered preform for 30^ deformation, its
diameter (d) will thus be given by equation 1:
(0 9) (d
- po.9) (1 07)
- = (3 5)
- pi.o (0 75)
- n)
or d = 3 09 in
p0.9 - -
The rubber bags with 1 1/6 in. wall thickness, rubber closures, per-
forated aluminum containers to support the bags, and aluminum sealing
rings were obtained from Trexler Rubber Company in Ravenna, Ohio. in all,
12 samples were prepared as shown in Table 9. The procedure involved
filling a bag to the required height, vibrating It, placing a filter
paper on top of the powder to prevent powder loss during evacuation, plac-
ing a sealing ring and clamp, evacuating the bag, and then tightening the
clamp. Then the bags were isostatically pressed in an oil system in a
6 1/2 in. diameter x 18 in. deep chamber. Pressurization to A0,000 psi
took about 6 to 10 min. The pressure was maintained for 2 min, the
chamber was depressurized, the bags were removed and clearted, and then
the billet was removed. The isostatic press tooling and typical com-
pacted preforms are shown in Fig. 11.
2A
0.5M
0.5"
Figure 10
25
Table 9
Billet
Powdt sr-FIlled Number of Powder and Pressure Serial
Bag D imensions Samples Combination No.
a
Compacted billet loose inside bag. Small amount of oil
leakage at top and bottom.
L.
Slight trace of oil on compact. Bags deteriorating after
second pressing.
C
Rubber was perforated by protruding metal tube while closing
cover; sample discarded.
26
Neg. No. 423^8 Neg. No. 42349
(a) (b)
27
For replicate tests, the isostatic pressing procedure was very simi-
lar to that described above except that in these tests plastic tape was
used to seal the rubber bags instead of aluminum sleeves used in the early
part of the tests. This change was made to minimize the possibility of
any tearing of the rubber bags during isostatic pressing.
Since Al 7075 alloy contains zinc which melts at 790oF and forms a
eutectic with aluminum at 720^, it was necessary to perform preliminary
trials to select sintering conditions and experimental combinations for
the Phase IA parametric study. The sintering for this investigation was
done in a small vacuum furnace at about 50 ym pressure. Small pieces
were cut from the top and bottom discards of the pressed billets. The
temperatures selected for this study were 900, 950, 1000, 1050, and
1100oF. The sintering parameters were evaluated on the basis of optical
metallography and scanning electron microscopy. The results of this
work are presented in Section 3.k.Z.
Sintering for all the forging work was conducted using a large Inconel
muffle placed inside a large electric furnace. The inside dimensions of
the muffle were 13 in. width, h in. height, and 58 in. length, and it was
designed for heating with any atmosphere. The samples to be sintered
were placed in a stainless steel tray,with a long handle and could be
inserted to any desired location inside the muffle. A chromel-alumel
thermocouple was inserted to the center of a 1 lA in. diameter x 1 lA in.
high piece of aluminum which was placed next to the samples being sintered.
This thermocouple was used for temperature control. All the samples (pre-
forms) for the forging trials were sintered in nitrogen.
The preform dimensions and forging experimental plan for the para-
metric study were discussed in Sections 3.3-1 and Table 8, respectively.
Of the 36 tests planned (Table 8), eight were aborted because of sinter-
ing furnace malfunction. The results of the parametric study are discussed
28
in Section S.1*,2*. The procedure for replicate forging tests was identic
cal to that fop parametric study, and the results of the former are pre^
sented in Section 3,^,5,
The sintering study was conducted with both fine and coarse powders
and, as expected, ease of diffusion for the fine particle size was quite
evident. A typical example of comparison between coarse and fine powder
sintered compacts is shown in Fig. 13. Both the compacts were sintered
at 950 F (2 hr) in argon atmosphere.
On the basis of this study, 950oF and 1025oF were selected as the
two sintering temperatures for the more detailed forging parametric study
planned according to Tables 7 and 8. These temperatures are high enough
to provide extensive diffusion but not so high that much zinc would be
lost. For the forging trials, nitrogen was selected as the sintering
atmosphere and, as discussed in Section 3.4.4, the preforms sintered in
nitrogen and then forged led to good properties.
29
Table 10
30
mamm
31
'.. 4:
\*. <
Figure 13
32
forged P/M parts. The ends of cylindrtcal btllets tsostatleally pressed
at 40,000 psi were used as preforms. In all, 13 pieces were forged rep-
resenting various combinations of powder size, sintering temperature
(950 to 1050oF) and forging temperature (650 to 85Q0F). The deformation
was 50^ In height on all preforms. A few wrought preforms were also simi-
larly forged for comparison.
The density of the P/M forgings was 100 percent of the theoretical.
Microstructures were also studied for various combinations of processing
conditions. The samples forged 50% at 750oF (Figs. 14a and b) did not
reveal any remnant effect of hot forging In the form of elongated grains.
However, with 650oF forging temperature (Figs. 14c and d), the typical
elongated appearance of grains can be clearly seen. The Irregular-shaped
areas are particles of much higher hardness, e.g., chromium, which had
remained unreacted and undeformed at these temperatures. The mlcrostruc-
ture of the forged wrought products shown In Figs. l4e and f is similar
to that in Figs. 14c and d.
Visual observation of the forged pancakes shows (Fig. 15) that the
high deformation of 70^ invariably led to severe cracking, whereas it was
much less at lower reductions. The extent of deformation was the dominant
factor and other process variables showed little systematic effect on the
surface cracking in the forgings.
Tables 11, 12, and 13 present data on 0.2^ offset yield strength,
ultimate tensile strength, and tensile elongation, respectively. By
simple comparison of these data with the minimum values required by the
specifications. It is observed that there are several processing combina-
tions which meet the minimum requirements, and they are shown In Table 14.
Tables 11, 12, and 13 clearly Indicate that none of the preforms obtained
by Isostatically pressing at 20 ksl met the minimum requirements, suggest-
ing that a minimum Initial consolidation Is required to obtain the desired
properties In the final product. This investigation showed that the
33
; -
.
!
f
1
^'^ cy /'.
I M.'
t 0m
Neg. No. 42356 100X Neg. No. 42354 100X
(a) (b)
fcnlh'S
.<*
4
- *
v *"
.. . ^
34
Neg. No. ^25^4
(a)
35
in
LU
I Csi CM
CO CO LTV TO 1_
< r^
vO
r^
vO
cn
vO
O
oo cn *. ">v ^ -a-
< LTV
CO
O
r^-
CM
r^ t^. en
IS) vD sO vD (0
oo csl CO
LU o CM en CO O
o LA
O 01 o CM
CC OJ "i a)
CL. o 4)
JC CD
(0
3: oo -r en E
> 3
Q \) r~- CO r- CT\ C o
-a- CO vX) -a- vO vD vO sD c
P vO '.D > s X. s. V.
s. V. o -a- -a- ^ CM r- o
-3- LA vD cr
u r - r^ "^ r^ OO CO CO
^ ca r<. vO
^>. ^-^
vO vD MD
^^
vD
i vO MD s*. *m. y--^ c
vO CO t^ o oo en
CM CO
3- CM -D TO
3
< O
< ^II
a. (U
jr
13 L.
0) Si
I-'
Ol CM
DC 0) (U
u M
-3" CO U-
a) t/> | i MD o 01o
\ .
"%
C3
r "s. c O
CM vD CP\ XI 0) O
jQ
(D
Z
I 1 -T c #E CTl
|_ C3 \ LO <N CO J- "
I vO MD \D o II
o ^^ ^s ^-^, ^^s ,N 01
o co LO -a- r^- LA 0) D-Xl
Csl CM O) 01 <U
^3- 3 3 0) cn
> O TO
l/> j-J 2 c
in 1/1 CL J-J -o
Q. Q- a) c
O i_ i- TO UA
o
CM -C o (D O CM
Ul o CL >+- TJ
0) 0 vD 0O 0) <U O
4 C (U 4->
t- LO o \0 1- TO U
TO 0) <U
o vD
v \0 CM LA
^ ^ V0 vD II 01
u 01
U
+J
4-1
(U
vD CM
* E <U 0) E
(- r^ ro I 3 -Q 3 C TO
VD CO LA E fr O
^O
,^ ^^ I vD 13 "TO -o
c
>
t CSI C 4-1
CO OJ co -3- vO LA 3 <U
aE OJ
x:oo
ls 0) Ol
-C O TO
c h- 1- CO C3
o V
*
!-> . J3 >
fD . TO CM co-a-
o O
.. h-
^~^
en
Oi
c < M-
LU
0-
>_ 01 ^
014- <^e o O O O O o H) 0)
csl l_ 01 ro LO LA 0) u "O
o o Q. o o
O LL. OO z: E
36
vO CM ro
CM cn a-
^ Q < LA
1-
'OJ
CQ
< ^D 00 LA
-C > r- -a- rA
< I- S <
-a-
in CM
<s> CM v^)
LU tNI u
O CO oo OA
O -a- >D r. vO -a- oo 4-1 o
a: s. ^^ ^^ r ro r^ r--
v O
cn \ -, cn
* oom
LA LA ^v. LA
> o oo -a- > CM
CTl O rr\ < <
-a- r^ r~~ vO -a- r-- 0)
,^ ,^ r- cn
o h.
,r-~^
^s CM
>- CM CA -a- ^~^ ^^ - 0) TO
Q CM CM CM -a-' S. <r> oo vD
vO r-. CD E
> > -C
< in < c o
Q. D-l c
\D rA -a- vo CM CM c 0)
cr\ o~ -3- o 3
m cn O rA rA r-~ r^ 0A cr
LPv -J- vD LA 0) P^ LA r~. r*. r^. r- f-s
CM VD r^ 2: ^^ ^^ >s "v 03 !_
37 **, ^~ ^^ cn -a- rA oo -a-
O CO
CD C (1)
< CO r*. LA > > 4-1
&. < cn LA <n-a- 1--00 < TO
< CM vXJ LA M3 vO so r. r- r^. 3
a. CM vD r^ 0A ^^^v t*^.**** ^^ ^^ LA
^^ ^^ -. s * vO 0O r^ o oo cn O
CNJ z i CA CM LA CM rA 1 ^ CM i
37
O o 1
O O O ia o
CM LA ' * u o
DO ^^ ^^ VO LA MD
"^^^
in LA
< LA o O
O
-.
\ "N, CM
(N x. OP O O -
O o CM -T m m O (U
< CM
^
C-l
s ^s
<r
vO
,^
LA
^v,
r**
^S, i_
CD
(TJ
CM CM CM r^ cr* OO a)
CO 0 "-' ^w^
co a. 0 E
UJ to in 0 3 O
o CL c c
o "^ 0
Qi jr -3- c
Q. en 0 cr
a; m <U 0 0 0 O O 0 .
z. O m 2: 0 - V 4J
c VM
O CNJ CM SO LA (0
<N J- T ^ ^ N. v^ "S. C
>- 00 v. ^s O p O
-
O
O
- O
' (13
Q O o O x J3 2
s. ^ O . CN LPi \D
o * MD ^* ,% ,^-s. ^. s-. ^^ O
co ^-^ ^D OO r^ O OO en u
^ ^
^ro- CM CM 1 (A CM
r ?
^ v ^ ^-> V 1 n ^ ' s ^ >*--' Q a) x:
^:
4-1 U> CM
0
4-1
< ro (/> o
0 u
~>
c o
< (U o
a. 0 O ts -0 E CT.
x: o 0 t
0
^ ^^
0 .c O II
o *K ^^ "S. ( N. (U
-O a: o 0 O 0 O CA Q--O
TO ^^ rf^, ^-v >^-N ^^ ^-^ 1/1 in a)
o CA UA -T r^ LA 0) en
co 0 r CM r CM i .n Q "j
^' >
s in parent
CD 0
'St 0 4-> -o
CL a c
C3 N. 0 l- TO LA
-C CM O CNI
O in M- TJ
0) O 0 U O
3: O o 2: O 0) 4->
O O rA
8). S_ TO 1_
CM tv-i 1 (0 <U 0)
II
o o o o "^ \ u in 4-1 <u
(3-9 \ >s. O O O B (U 0) <U E
o o CM rA >. 3 J3 3 c ' (o
^~~* ^ ~^ ,> ^-^, O P ^ S E -Q > O TO
en -- CM ^- ^ D (0 TO (U T3
o rri CM r^v -a- s> LA c C H-^ > +J
3 O (U
2: (U O en
< J: x; o o TO
c (- (U h- try o
z 0
o |
4-> E TO
m PBB CM CA J"
0 ^
en >
c
Mr
M- 0
n
1 . in
OH4- oN? O o O 0 O 0 0 0) 4-1
u t) ro ca r-~ OA LA r^ a) 4-1 (n -3"
UJ O Q ex O CM
U- in Z +J
38
C71
c o tn o LTV LA tn
*e o
in LTN vo \o LA tn -a- LTV
0)
to
a) <r\ <r\ urv >-
QjCO'
o LA vD r-. -3- r^ CA oo
h rx* i^. a)
to
(N CM CO OO f-V CM
to >- in
.2 r^. r*- cn p-- r>. LT* <T>
a* VO vO vO vO VO vO vO
CM
CC (/> O
O UJ
li- 3 ro
_J
>- < c
o > IA O LTV o o o o
CM LA CM O O O LA
u p vo r. oo oo oo vo c
< UJ HO o
C3 ID
u
v. <
a. I- M
TO
LA LU u E
r^ > (U O LA LA LA O LA O 3
o o QJ LA CM CM CM LA CM LA
r--. co E CA O O O <r> O CT^ E
~
< <u
c
< to E
LU to
QC
J3 O (-
u. ac
UJ c w
to a. o *? rc
z o U
o en M * .
D. . C oi" oo -
E u.c o o o o o o o
< UJ r> r<^ r-. IA r^ r*. r~. (U u
Z -I O
o? J3
(U
Q.
oa to <u o <TJ CO
3: z Q Ll. h-
o UJ
o H-
c w
O X o M c
O) c <u
4-1 c 0) B
o 3 nj #M o o o o o o o >
to 4-1 tn in ^r -a- -a- -a- -a- -a- -a- u
UJ in W J^ CD 0)
o O 0) Q.
o ifl u 0) l/l
0C
a.
a. l_ W
(TJ o
5
^
(/> +J o
*
x: 0) o o o o o U
w o o o o o o o 0
0) N
_ -a- '- . -a- ^3
>4-
O r-.
x: oo i i i i i i i E
3 U) ~
C i) (O
3 t-
ro c Z3
en
c o
c
0 (D
> "
M
to M <0 0) ui
.c cn-^:
l/l (13
<u C CM r^- c en oo -a- <n ro
u CM CM -Q v. o
O -n 0)
g o
>
<CKO
o n -Q
39
preform density should exceed 90^ of theoretical density to meet and
exceed the wrought product tensile properties, Also, the coarse powder
required 70^ deformation during forging irrespective of sintering or
forging temperatures, while the fine powder sintered at 1025oF met the
specifications at all the deformations and forging temperatures under
consideration. Lower sintering temperature of 950oF for fine powder
required 70% deformation regardless of forging temperature.
From the above discussion and the fact that the amount of deforma"
tlon in the actual part varies from one section to another, ft is apparent
that the processing combinations 7, 8, and 9 are the best suited for fur-
ther evaluation. Thus, the optimum processing conditions are those In-
volving usage of -100 mesh powder, 40,000 psi pressure for cold isostatic
pressing, and 1025oF sintering temperature.
40
Table 15
0 ,2% Offset
Sample Y.S., UTS, Elong., R.A,,
Identification ksl ksi % %
41
the sample was cut and two letters, The first letter Js T, M, or B, which
stands for top, middle, or bottom, respectively, and shows the location of
the sample within the isostatically pressed billet since three samples
were cut from each billet. The second letter^B, C, or B^roerely distin-
guishes the three different tensile test specimens cut from each forged
pancake. Thus, at each forging temperature there were essentially six
identical tensile test specimens, and the uniformity of the properties
achieved within these six samples at each of the three forging tempera-
tures is quite remarkable. These tests conclusively show the reproduci-
bility of the results obtained by using optimum processing conditions and
further confirm the selection of the conditions as optimum.
The entire work of Phase I, Task A, thus leads to the following pre-
liminary process specification for creep forging of P/M preforms for 7075
aluminum alloy;
A2
k% PHASE t. TASK B r T-SHAPE FORGING
The work done in Phase [t Task A, was conducted prlmartly with cyltn-
drfcal preforms, [n actual practice of making a complex forging, the metal
flow is considerably more nonuniform than in the case of upsetting of
simple cylindrical pieces where there is perfect circular symmetry. There-
fore, the purpose pf Phase \, Task B, was to test and evaluate the forgings
made with a special die incorporating elements of forward and backward
extrusion along with longitudinal flow and, if necessary, to modify prelim-
inary process specifications. The detailed scheme is shown in Fig. 16.
The sample for evaluation and the corresponding preform shape selected
is shown in Fig. 17. The preform selected had approximately the same width
as the forging so that there would be little transverse deformation during
forging. The deformation was primarily of extrusion type in the longitud-
inal direction as well as in the vertical direction to form the short legs
of the "T" by backward and forward extrusion. Keeping the above-mentioned
factors in mind, the dies for the T-shape forgings were designed,and the
die assembly is shown in Fig. 18. Since only a few pieces were to be
forged from this die, it was made from a common die stee1--A. Finkl and
Sons Grade FX heat treated to RQ 38, These dies were mounted in a standard
die set and assembled in the 1000-ton press,
43
PrelIminary Process Specifications
from Pfxase I, Task A
Sintering of Preforms
., - longitudinal
Evaluat ion - T
Tensile
- transverse
of "T" Shape Forgings
" Microstructure
" Product Integrity
Pre!iminary
Specificat ions
for Phase 11
Figure 16
kk
1.25'
Forging Direction
Width - V
0.75
Flash
1.0
0.20
Figure 17
45
Neg. No. 43351 Approx, 1/6 scale
Figure 18
A6
Neg. No. ^3353
(a) Tooling for trapezoidal billets
Figure 19
Tooling for Isostatic Pressing of Trapezoidal
Billets
^7
Neg. No. ^2760
Figure 20
48
Note thgt during forging the preform was on Its stde with the long
arm of the ^T1 formed (n the horizontal plane and the short arms formed
by backward and forward extrusion In the vertical direction towards the
back of the die system. The press was moved In a slow manual mode with
the approximate speed estimated to be of the order of 3 fpm or less.
In the first forging test, the preform had excessive material and
the stop blocks were not set precisely. This caused excessive thinning
of the web and extrusion of thin flash between dte clearances which made
removal of the forging difficult, in all the subsequent trials, the pre-
forms were machined to a width of 3 3//t In, and a weight of ] ,k to 1.5 lb.
With these optimized dimensions, no difficulty was encountered In the
forging tests, Two preforms were overheated because of a furnace mal-
function but did not appear to be damaged and were used In the tests.
These preforms were Nos, 5T and 7T. Apart from the initial test on pre-
form No. 6B, all the remaining forgings were made with a work material
and die temperature of 750oF (Table 16). Figure 21 shows some of the
forgings made,
As shown in Fig. 17, samples 1 and 2 were cut for the transverse
properties, and samples 3 and k for longitudinal properties. There was
a significant amount of extrusion flow in the direction of the sample
length for the longitudinal samples, whereas the metal flow was trans-
verse to the length of samples 1 and 2. Because the width of the trap-
ezoidal preforms was very similar to the width of the T-shaped forging,
there was practically no metal flow along the length of the samples 1
and 2. Creep forging was conducted at 750oF die and workpiece tempera-
tures, and tensile specimens were taken from six T-shaped forgings. The
average values are shown in Table 17 along with the ranges of the indi-
vidual values. Averages are given for six tests except for location 1
where three of the specimens had forging defects and had consequently
poorer properties. The defective specimens were not included in the
averages given for location 1,
The results of the tests were truly remarkable in that the tensile
properties in the longitudinal direction are, In many cases, in excess of
the minimum specified properties for the wrought materials In the direc-
tion of the grain. Indeed, In view of the high yield stress of the
samples. It is conceivable that with somewhat different heat treatment,
all the specimens would have exceeded the minimum specified elongation of
7% set by the Federal Specification Q,QrAT-367H (see Section 2,6), The
transverse specimens 1 and 2 exceed the specified properties for the
transverse direction In most cases, and the maximum elongation Is as high
as S%, Considering that the Federal property specification was for
wrought material, this combination of high strength and satisfactory
elongation formed In the P/M processed samples Is quite remarkable-^-
this, in spite of the fact that in the lateral direction, the trapezoidal
h3
Table 16
50
::
Figure 21
51
Table 17
0.2% Offset
Specimen Y.S., UTS, Elong,, R.A.,
No. ksi kit ; %
7T-lb
*0.4 + 1,6
Avg 67.2
0,2 70.5 1,6 1.7 +0.3 2.6 +1.0
1.3
>0.7
4B-2d 67.6 73.8 3.0 4.0
^T-2 68.5 75.4 5.0 6.8
5T-2 68.4 73.8 2.0 6.8
5B-2 68.2 71.6 2.0 3.1
7B-2 68.3 73,0 2.0 4.7
7T-2 67.1 70.2 2.0 2.4
0+0.5
Avg 68
-0.9
72 q+2'Z| 2 7+2'3
"V.l
4B-3a 69.7 77.8 7.0 10.1
i.T-3 71.5 78.8 6.0 10.8
5T-3 69.7 79.0 8.0 11.6
5B-3 72.A 79.0 6.0 10.1
7B-3 72.5 78.4 7.0 9.4
7T-3 71.A 81.2 10.0 15.2
,+1.3 79.0:j 7 3+2'7
+k 0
11 z
7 '
Avg 71
' -1.5
n - -1.8
/tB^a 70.6 78.0 7.0 10.1
kl-k 70.8 78.4 6.0 10.1
SJ-k 69.7 78.2 7.0 11.6
5B-I> 72.4 78.8 5.0 9.4
7B*-k 71,3 77.8 6.0 10,1
7T-A 71.6 80.4 8.0 13.0
,+1,3 78 6 8 6 5+1'5 10,7+2'3
Avg 71 ' tJ: 8 "1,3
a
'-l and '2 are transverse specimens:
r ^3 and T<4 are longitudinal
specimens,
Failure before yielding,
Notes; 1, Powder mesh size, '100 mesh; isostatic compaction,
40 ksJ; sintering temperature, 1025oF, forging deformation, 50'-
70%; Isothermal forging temperature, 750oF, 2. The tensile
test specimens were In STA condition and were tested at room
temperature.
52
preforms were made 50 wide as to have practTcally no flow In that direc-
tion, Even in conventional forging, the lateral direction is subjected
to some metal flow.
The data obtained thus far clearly indicated that, if the cam were
made with the optimum processing conditions selected in this work, it
would have excellent properties, The properties should exceed the spec-
ification used for this program based on the specifications for the
wrought material,
Note the rounded structure in Fig, 22c, specimen 5B-A, and similar
structures in Fig, 23t Observation of similar structures through SEM is
shown in Fig, 2k, along with an X-ray spectra of the whole surface as
well as of the rounded phases. The X^ray spectra suggest that these
spherical particles are chromium-rich and, as discussed earlier, remain
in the matrix because of the poor diffusibi1ity of the chromium into the
aluminum matrix at the low processing temperatures. The fact that the
particle does not show elongated shape In Fig, 22 (specimen 5B-^) also
indicates that, because of the hardness of the particle, it does not
deform with the matrix.
In many specimens, such as those shown In Fig. 25, some cracking was
noticed In the hard chromlumTlch particles, However, since such cracking
appeared In the particles In both the longitudinal and transverse sections,
it is probably not related to the relatively low ductility in the trans-
verse specimens.
53
r
.4' V ^ ^i..
I
V
%* t f
i<~
-ij v * */
200X
^
-:.J
T^
I*
r'im
3fb
#|W
5P
- -^k
LV^J, . - 1 1
s
v
tC
:K
.'r^i'-'i
Jfefw**-.
Figure 23
55
c
E
O TO
<u -
Q.
10 c
0)
w c
- S
u o
> JI
if) I/)
c
(D 0)
s_
4-1 O
0
m
(U Q-
O
ro a)
S_ 4-J
u
TO
L,
U
I
M- X
o
JD |
o
. (U
(^ Q-
U en <"
c
3 .- <u
Dl E
10
U. O
X
o
a o
0)
Q- X
m
-C 1-
CO -M
I TO
I- E
E 0)
O
i- O
5
0) 0)
-C
a. +-1
E
co o
M- C
O CD
o
CO l_
4-) >-
CNJ O fD
en (U
Q.
I_
I
00 X
>-.
(D -D
Qi "'
I
X -
-o
C I
TO 1-
56
200X
(a)
57
both specimens '"M and kl*2 have stmHar gratn size, but specimen kT*2
has a reasonable ductility of 5% whereas specimen hlv] failed immediately
on loading because of a forging defect, Generally, the finer grain size
is associated with better elongation but, ln;,tfie samples tested in this
program, the highest tens Ile elongation of lOfc was observed In specimen
7T-3 which, as can be seen In Fig, 22, shows fairly large grains. It
appears that the predominant factor affecting the elongation was the
orientation of the tensile specimen with respect to the direction of metal
movement. Understandably, If and when any forging defects appear, they
adversely affect the ductility,
58
5. PHASE M " PROCESS OPTtHMATION
FOR PRQDUCHQN RUN OF THE CAM
As a target, It was planned tc> make the Isothermal creep forging with
most of the surfaces net, Only a few portions of the forging will require
machining. The high precision cam slot for the roller will have to be'
machined out. The step <>abn (Fig, 27) will be forged net, but the "cd"
portion will be made with about a 1/8 tn, thick web which can be punched
out later to form the window. The surfaces (e,ffc,g,h,i,J and p,q ) and
59
Selection of Target ForgTng Geometry
I i ^"
Production Run
Figure 26
60
See Fig. 2 for detailed
d imens ions.
Frgufe 27
61
ribs will also be made net. However, the holes and the material (around
the areas SS In vtew BB) will require some machining,
The detailed design of the bottom and top dies for the first stage
are shown in Figs. 28 and 29, respectively, in the blocker die, dimen-
sional tolerances were not critical; therefore, low alloy steel SAE 'tl'tO
heat treated to RQ 35 was selected as the die material. The finishing
bottom and top die designs are shown in Figs. 30 and 31, respectively.
For the finishing dies the dimensions were critical. Therefore, H13 tool
steel heat treated to RQ 38 was selected. Because of the ejection system
necessary in finishing dies, the positions of the bottom and top dies
were reversed as compared to the blocker dies.
It was felt that, in view of the nonsymmetric shape of the cam, some
simulated material forging tests would be valuable for designing and pre-
form shape for the blocker die. Plaster (Hydrocal) molds were produced
using the conventionally forged cam as the pattern, The parting line was
the same as for the conventional forging. The pattern was placed in a
wooden core box and attached through the base by drilled and tapped holes
in the pattern. To form the parting line, the area around the pattern
was built up by wax. The plaster (100 parts plaster + 35 parts water by
62
-4- -)--
X
4--
*
5'
i r
7 |
fH f^ r
Figure 28
63
-t- "f- -f-
-4-
-f
-4- +- +-
\ 1
' !
\
n ii
r II
it
it
i
Figure 2Si
6A
1
SECTION B-B
Figure 30
5'
Figure 31
66
Neg. NO. kk752
(a) Top die
Figure 32
Blocker Dies
67
Neg. No. 44757
Bottom Top
Figure 33
Finishing Dies
68
Neg. No. '14750
(b) Bottom plate
Figure 3^
Supporting Plates and Ejection System
for the Finishing Dies
69
Neg. No. 45398
Figure 35
70
weight) was thoroughly mixed and then poured into the box and allowed to
set overnight. For reinforcing the plaster piold, steel wife was used tn
the cope, Then the cope was removed from the core box, whMe the pattern
remained in the plaster mold. The drag Glower dte) was produced in a
similar manner with nail and wire reinforcements, To prevent sticking of
the cope and drag, lubricating grease was used on the cope surface. Both
cope and drag were dried in an oven at 150oF for a couple of days, The
Hydrocal dies are shown in Fig, 36,
Play dough was used as the simulated material, After some initial
tests, graphite spray on wax worked quite satisfactorMiy as a lubricant,
and facilitated easy removal of the play dough forging. The results of
different shapes of preforms tried are summarized in Table 18, and some
of the preforms and final forgings of play dough are shown in Fig, 37.
Cylinder shape (No, 1) with 2,5 in, diameter showed complete filling of
the cavity but could not be considered for preform shape because of heavy
deformation (approximately 80% in height) which may crack the component.
Therefore, a semicircular disc with a varying height (Nos, 7 and 8) was
considered for preform shape. As can be seen in Fig. 37, they have good
reproducibility also, A modification was then made (Table 18, No. 9) to
accommodate the filled bag in liTRI's isostatic pressing facility. The
volume of the pressed preform (in this preliminary design) was estimated
to be about 43 in.3 whereas that of a trimmed conventional forging is
29 in.3 The excess is provided to allow some minor modifications of the
preform dimensions, to form a flash, and to allow removing cracks, if any,
after the first forging operation,
From the first-phase work, it was known that the density of sintered
product will be nearly 93^ and the -lOO mesh blended elemental Al 7075
alloy powder has a tapped (loose filled) density of about 55^ of theoret-
ical. Hence, for the preform shape No. 9 in Table 18, the design of the
bag should be as shown in Fig, 38, The volume of such a bag is about
80 in.^ |t has an extended portion and open end for filling and evacua-
tion. The cap for the bag has a ]/h in, diameter rubber evacuation tube.
The wall thickness of the tube and the cap were 1/16 in. As can be seen
in Fig. 38, the bag design does not have the semicircular detail (as in
shape No. 9 of Table 18) which is left out on purpose so that it can be
modified as necessary, Also, a bag with such detail would have made it
difficult to remove the compact from the bag, The rubber bags were fab-
ricated by Trexler Rubber Company, and the tooling for isostatic compaction
is shown in Fig, 39.
71
Neg. No. 43793
(a) Cope (top die)
mm lflPWWIPSJlf5pP iii|.iiiijmi4i., in
^ h rftmE
[
I 2 i
i n
4 5
T 6 9 10 II 12
IIT RESEARCH INSTITUTF
Figure 36
72
Table 18
Trapezoidal 8
o
(
2.5
ko
Ing of the cavity.
Incomplete fil1ing of
plate the cavity.
1
1,5
Semici rcular 32 Incomplete filling of
disc the cavity.
.5
h Cylinder 31 Incomplete fi11ing of
O the cavity and broke
during ejection from
the die.
5 Cylinder Same as H 31 Position of cylinder
was different than
position k. Broke
during ejection.
Trapezoida 1 39 Incomplete fi 1 1 ing.
semici rcular
d i sc
Note: Volume of the conventionally forged and trimmed cam is ^29 in.^
Play dough is the simulating^material.
73
i 1 : i 1
8 . . !
s
I [ PTI
H 12
fNSTlTuTI INCHES
Neg.,No. 43796
Neg. No. 43798
(a) No. 7
(b) No. 8
Figure 37
74
Neg. No. kkjM (b)
Figure 39
Tooling for Isostatic Pressing Preforms for Cam (a)
and Bags after Evacuation, Ready for Isostatic Pressing (b)
76
1
OJ
^
1
_
oo
^ -
en
CD
i_
0)
-O
o
c
o
m
0)
a
3 E
C7l i_
O
c
'i3
X)
0)
XI
O
Um p., iJ
77
5,3,k Development of Reusable Inserts
The Phase I work was conducted with powder supplied by Alcoa. Another
major supplier of the powder is Alcan, In Phase II, the two types of pow-
ders were compared to a limited extent. Many of the Phase II tests were
conducted on powder supplied by Alcan. The Alcan powder (-100 mesh) had
a bulk density of 1.18 (Alcoa 1.12), and the size distribution given below
with Alcoa analyses given in parentheses;
Size, Size,
U.S. Mesh % U.S, Mesh %
The Alcan powder had lesser (81%) fines ('"325 mesh) as compared to
Alcoa powder which contained 32% of -325 mesh, The chemical composition
{%) was 1,6 Cu, 2,5 Mg, 0,2 Cr, 5,6 Zn, and the rest aluminum, which is
within the specifications of Al 7075 alloy, The powder was checked after
sintering and had mlcrostructure similar to that of Alcoa powder. Thus,
from a technical viewpoint Alcan powder was deemed as good as Alcoa powder.
78
r
Figure 41
79
5.3.6 Isostgtic Pressing and
Sintering of Preforms
The processing conditions used for pressing and sinterfng were the
same as established during Phase t.
Except for forging No. 3, all the rest had some cracks at the places
shown by the arrows in Fig-. 43a, The data tn Appendix 2 show the trend
in load. With the press speed of 0.1 ipm the load was 350 tons, and with
the speed at 0.2 and 0,3 ipm the load increased to 400 ton. Except for the
80
Neg. No. kk753
Figure k2
81
Neg. No. 44893 (a)
Figure 43
Blocker die forgings became the preforms for the finishing dies.
The forging data for finishing dies are shown In Appendix 2, Forging
No. 1 had the same speed and temperature as In the blocker die forging.
It showed extensive cracking, Fig, 44a, With the same speed but with an
increase in temperature from 750 to 830oP, less cracking was observed
(Fig. 44c) on Forging No. 3. To check for smaller cracks, both forgings
were examined with dye penetrant tests and the results of a typical forg-
ing (No. 3) are shown In Fig, 45, Besides the cracks which could be
visually seen, two other kinds of defects were also observed. Defects
marked by the arrows In the left of Fig. 45a do not seem to be deep
enough and are most probably due to the lubricant entrapment, whereas
the defect to the right of Fig, 45a (shown by an arrow) seems to be a
"fold over" type of defect. Therefore, it was decided to forge preforms
Nos. 6 and 7 In steps (incrementally) to find the stages at which forma-
tion of defects starts.
The forging data for Nos. 6 and 7 are shown In Appendix 2 and one
of the isothermal forgings is shown in Fig. 46. The type of metal flow
that causes the formation of fold-over is quite evident and Is Indicated
by the arrow. The unsatisfactory metal flow suggests the need for mod-
ification in the preform design for more favorable material distribution.
From the studies in Sections 5,4,1 and 5.4.2, the following conclu-
sions were reached;
83
Neg. No. kk33h Neg. No. ^935
(a) (b)
Figure kk
8k
Neg. No. kkS83
(a)
Figure 45
85
Neg, No. 45388
Figure 46
86
Neg. No. hhStf
(a) As-forged
87
2. The original preform (bag) design No, 1 needs
(nodificati.on to impoye die filling and reduce
cracking,
Since powders from two manufacturers and one- and tworstage operations
were used for establishing initial forging parameters, a thorough Investi-
gation of mlcrostructures was done for the as^-forged (AF), solution treated
(ST), and solution treated and aged (STA, T6) conditions, Three forgings
were selected.
88
^^JfWi! m^^m*'
' ^
^
W "*,* "
^HK'
k*
k ^. >i t ^ m
m
JL
^^ ^
ji '-%
/
89
II oo
(U
Q.
un ro
oo 1 i i D-
1 1 i (
0
u >
c L.
ID 4-1
t)
r*. 1 N. i i Q,
_i 1. . i I
o 0)
4-1 C3
TO cE
^O o
i vD 3 d-
-d- -\ Jt U (i)
"N^ un V, V. ._ i_
^ i PO r^i T3 Q-
l/l C
CM CM (0 4-
CL O
u L
c M
VD Q. o
vJD 1 (U
^v^ ^r OO 0) in <+-
">, LH x. *"s u c M-
X r ro r-^ (0 O cn UJ
JT
<N in >
c C "O
o OJ 3
E 4J
\) 1/1 cn (JO
OJ oo oo c Q
1 "V \ "X v. 0) Lu O
X Lr\ LH m E 1 4-J
' 1 1 IN Q X "O
0)
L.
n
E r CM CL
L_ 0)
O . L.
>^ o ^r SO r^ O " Q-
0) 2 i CM QJ
u l-J w
Cl_ o E
i-
O
4-
0)
L.
0.
90
ft
' 'Ax ' "" ""
V:
-aP i 14
91
Neg. No. ^5099 Neg. No. 45091 V5X
Figure 51
Sections of Forging No. 14. The black dots are most probably the unreacted
Cr-Rich Particles. (See Fig. 50 for key to part locations.)
92
Forging No, 1 - Made from Alcoa powder with the
two'-stage forgtng operatTon,
The hardness values of both Alcan and Alcoa powder forgings were
very close, e.g., forging No. 8 had an average hardness of RB 89, whereas
forging No. 17 had an average of Rg 86 in the STA condition. Room temper-
ature tensile tests for forging No. 17 in the STA condition show (Table 19)
properties closely approaching longitudinal wrought specifications and
exceeding those for the transverse direction.
Chemical Analysis, %
Forging Forging
No. 1 No. 16
Element (Alcoa) (Alcan)
Cu 1.53 1.59
Mg 2.31 2.14
Zn 5.85 5.52
0 0.48 0.37
93
Neg, No.
K
17 ^5156
- - -^
i. - -*' -*'
- '<'^
. .^.
-..?;,
: %- "*
^#-4
CD 45161
cn
L.
O
f":
45159
(
Figure 52
94
Neg. No.
\X& ^ \ . . ', ~ ) \-^;;--' V ^
45155
' I|
45160
.* -I
^>
45158
-:fc# -'P'T' ^rt
/' VA
c
u
o
17 WBSm V,4 45261
HHHB
*
^ ^* -
45157
I
200X
(c) Solution Treated 900oF, 2 hr, Water Quenched, and Aged 250oF, 24 hr,
Figure 52 (cont.)
95
T^ble 19
Forging No, 17
96
3, Modifications on preform bag shape are necessary
for complete die filling.
The modified bag design of Fig, ^0 in Section 5,3.3 was used for the
rest of the program. Several preforms starting with No, 22 in Appendix 1
were made with minor variations In the shapes by controlling the position
of the inserts during pressing, The pressing and sintering parameters
were kept the same as in the early part of Phase It, Preform Nos, 22, 23,
36, and 27 were forged directly in the finishing dies. The forging param-
eters and results are given In Appendix 2. The temperature and ram speed
were maintained the same. Figure 53 shows two of the forgings and the
preforms they were made from.
Forging No. 26 (Fig. 53) showed excellent die filling except for a
slight underfill at the tallest rib. Minor cracking also occurred over
a small length near the periphery. The latter did not occur on forging
No. 22 (Fig. 53), showing that it was Just a matter of providing enough
material in the preform. It was more difficult to improve the filling of
the rib detail. To achieve this, the flash geometry was modified by weld
depositing some material on the top die (Fig. 30), This restricted metal
flow into the flash near the rib and forced it into the rib detail of the
cavity in the lower die. As discussed later, this die modification im-
proved the die filling in the rib area.
In all, 21 more identical preforms were made using the modified design.
They were sintered and hand ground slightly to prepare them for production
run and for the series to study the effect of forging speed onload. The
latter forging series was conducted with a constant preform temperature of
830oF and with three press speeds of 0,1, 0.3, and 1.5 ipm. The effect of
forging speed on the load is shown in Fig. 5^*. As expected, the loads are
high at higher speeds, The maximum load applied was 700 tons and the
average projected area of the forging, including the flash, was about
70 sq in.; therefore, the maximum pressure applied was about 20 ksl. Near
the die closure the press speed showed less influence on load, Also,
usage of a slower speed makes the cycle time considerably longer. There-
fore, a press speed of 1.5 ipm was selected as the practical forging speed,
This speed Is significantly lower than 100 to 600 ipm speed commonly used
with hydraulic forging presses, That speed range can increase the forging
load substantially, Thus, the 1,5 Ipm speed gives an optimum combination
of low forging load and reasonable production rate.
97
Neg. No. ^5207 Neg. No. 45392
(a) (b)
S8
O- 830oF and 0.1 in/n^in
700
600
500
U1 400 -
c
o
"O
O
300
200
100
0 L
0.2 0.4 0.6 0.8 1.0
Distance to Die Closure, in.
Figure 54
99
5.5 Production Run Forgings
and Their fyjcil (ty
Some of the preforms used for the production run and for load vs
deformation study are shown in Fig. 56a in the as-sfntered condition. The
production run forgings are shown in Fig. 56b,
100
u
+ .08
A = 3.5^ F = O.Skt .09
.10
B = 1.08.08 G = 8.03*
.15
C = 0.66 06
-.10 H = 2.32^
02
E = 1.39.09
Figure 55
101
Neg. No. 453^5
(a) Preforms in as-sintered condition
Figure 56
102
Neg. No. 453^7 (a)
Figure 57
103
The weight of the flgsh was thus negrly 25 percent of the trtinmed weight,
This allowance was adequate to limit any cracking CFig, 56b) of the P/M
forging to the flash and give a sound trimrned forging.
The superior finish of the creep^forged P/M cam (fig, 58a) is obvious
when compared with the conventionally forged cam (Fig, 58b), The finish-
machined cam (Fig, 58c) is also shown to highlight the near^net shape con-
figuration of the P/M forging. Close-up views of the ribrweb areas from
both ends of the P/M and conventional forgi'ngs are shown in Fig, 59.
These figures clearly indicate the capability of isothermal creep forging
of a P/M preform to obtain a near^net shape component in contrast to the
oversize conventional forging.
104
TO
Q.
-a
<u
c
J=
U 1
TO *
SI
1 0)
-C 1
U1 ro
0
C U1
U. 14-
OJ n
JZ -C
4-1
T3 X
c 0
u
CD
Q-
^_^ a
J3 <
''
CD
C *~>
cn E
1- TO
O U
Li_
21
, ^
TO Q_
C
O T3
0)
CO J CD
LA C L.
a) O
OJ > 4-
1- c 1
3 O Q-
O) (_) 0)
0)
Ll. (D L.
O
JZ
4-1 0)
, JZ
M
S
^ 1*.
ro O
OJ
E a.
ro ro
(_> JZ
IS)
2:
"s. M
D. cu
C
XI 1
a) L.
CD It)
L. OJ
O C
Ll_
<u
CL JZ
<u 4J
u (U
1-
(_) +-1
0
OJ z
a.
E
O
(_)
105
Neg. No. 45^08
(a)
106
Neg. No. ^02
(c)
Figure 59 (cont.)
107
6. COST ANALYSIS
It is clear that the Isothermal creep forging process can lead to sub-
stantial reduction In production cost of complex aluminum alloy components.
The analysis is based on the work for the cam with linear dimensloas of up
to 10 in. and finished component weight of about 1,8 lb. The cost"
effectiveness of the process is likely to be valid for a wide range of
component size, but must be analyzed individually for each application
under consideration,
108
- LTV O LA O LA LA
+-) (0 CJ LA CM I O r^ r^.
ai u> U V *
c O ^ -3" O cn LA CM ^
0 -co- -T co
en
i_
O
ir> Ll_ 1-
o O x
3E -0 . O "S.
WM 0) m O O -Q LA U
o c c O - * CM a
OC O O LA c -^ W <u
o x: -"CO- O O X *
^> 1/1 0 * - * * * 3
Cu 0 ^ LA rrs ^r rr\ LA PA er
0 -co- CM LA 0)
o s: u
fsl o ^_
as av
0) u. c <D
o> in i_ c 4-)
-Q Q -a D> O (D D) JD -C 0 O
ro UJ (D 1- O J3 c r ~^ 4-1
\- z C O VD LA "V. "S* LA *J 0 O
. U. r^. CA en CM CM (D (X Q- in
X -C LA CM i_ (T^ CO 3 -^ *-
o 0 X w CO O W-CO O < O O LA u
< m "S. HT O 4- O 4-1 m vo **, TO
X X a_ (/> ~rA ~ -t-l co m > t. 0 0)
(0 a) ~ TO 0 a) PA x: CL >
to a CD ' r J_ r 4-1 u in CO- >*
10 -O <o- 0 tfl (0 CL TS en r c 0)
O M - en > C flj - c
o O W en > ID i_ 0 0 OJ-Q en fc E 0
U O c a) in -Q 0) a 0 c cn
a: u- O Q. in > V, O^LA e ^ M O L.
o x: to (U L- OJ -Q *v C N^. t- CO CL O a>
u. in -M i_ <u . 1- .- !_ 0) P V. CL >
D.J3 1_ x: x: JJ + LA v. 0
z c 0 0 -Q 0 LA c MW r>. *
o ^ in 3 4- vO ^ 4- S- \o * CM T)
/> U- z oc OA Ll. V> h- *' OA CO- 0)
N
DC
< 4J
a. cn cn
c
1_
2: C c -_J O
0 O C TJ - < E
TO
0
1
'W
Q. ro * O
O
U
4J
c
ID
cn
TO
a x
c
5h- cn
CO en ID M- (0 en c 1. 0 c
0 1_ c 0) V c h- TO
<_) O * u a. I- (D p z:
a) en E 4-1
U1
<U
p
O *J
TO
3
4-"
Q.
O 0)
in U
Q.
O -
i_ TO
u
0
"s. 1-
8
Q 0 0)
H X CO Q. U- t- X z ro
109
7. Fjm PROCESS SpECrFtCATtON
7.2 Material
Any conventional hydraulic forging press can be employed for the new
.,
process. The press should have adequate tonnage of about 2Q to 25 tons/
110
sq in, for the plan area of the forging includtng the flash, The press
should be capable of being operated at slow speeds of 5 ipm or lower, and
of maintaining the load for the desired duratton of 1 min on the forging.
The preform can be heated in air tn any fupnaee wtth suitable temperature
range. The dies can be heated hy any technique capable of heating the
dies to and maintaining them In the desired temperature range of about
750 to 830oF,
The properties of the P/M forging should meet those specified for
wrought 7075 aluminum alloy forgings by MIL-Spec. Q0-A-367H. The heat
treatment recommended on the basis of the wrought product specification
is solution treating at 900oF for 2 hr, water quenching, and aging at
250oF for 2k hr. For P/M forgings some modification of this practice
may lead to better properties. The processing leads to extremely fine
grain size and no specification for grain size is warranted.
7.7 Inspection
At this stage, the opening between the dies should be adequate to Insert
the preform. Too large an opening should be avoided to keep temperature
loss to a minimum.
Ill
8, PRtNCI PAL RESULTS AND CONCLUDING REMARKS
3, Tooling systems were designed and fabricated from H13 die steel
and showed to perform satisfactorily,
112
8, A process specification was prepared for tsothermal creeps-forging
of P/M preforms, The specification fs applicable to elemental powders
blended to 7075 composition,
113
REFERENCES
r,
2. Alumrnum Powder Products for Powder Metallurgy Parts,M Form FSS-
12816 (Rev. 6T71), p. 8, Aluminum Company of America,
114
REFERENCES (cont,l
1A. A1umtnum Stgndqrds and Data, ^th Ed,, 197^, The Aluminum Association,
FncTJ Table 2,2, ppVBl^V
115
APPENDIX 1
116
Table rl
Note 1. The preforms used for the production run were 31-35, 39, 41,
46, 49, and 51.
2. Preforms used to study the effect of speed on forging load
were 28, 30, 36-38, 42, 43, 47, 48, 50, and 52.
3. Preforms through No. 25 were made while the proper rubber
insert geometry was being developed. Preforms from No. 26
onwards were made with the rubber inserts shown In Ffg. 41.
4. Good preforms that were not forged were 15, 18, 19, and 21.
5. Preforms rejected because of oil leakage or cracking during
isostatlc pressing were 4, 10-13, 25, 29, 40, 44, and 45.
"
117
APPENDIX 2
FORGING DATA
118
APPENDIX 2
FORGING DATA
This Appendix provides raw forging data about the forging trials
conducted to optimize the processing conditions and the preform geometry.
The details of preform geometry were covered in the text, and the data
about the preforms were summarized in Appendix 1. The preforms used for
the production run were forged under conditions described in the text in
Section 5.5, and the data about the forging series conducted to study
the effect of speed on forging load were summarized in Fig. 5^- The data
for these two series are not repeated in this Appendix.
The preforms for forging were heated for 1 hr after they reached the
desired temperature. The die temperature was, in all cases, in the range
of 700 to 800oF. The forging lubricant was Fiske's 604 for both the dies
and the preforms, and for forgings made from preform No. 22 onwards, the
lubricant was thinned with 50^ trichloroethylene.
Table 2-1
FORGING DATA
119
Table 2-1 (cont.)
120
Table 2-1 (cont.)
121
DISTRIBUTION
Copies
Department of Defense
Commander
US Army Materiel Development and Readiness Command
ATTN: DRCMT 1
5001 Eisenhower Avenue
Alexandria, VA 22333
Commander
US Army Materiel Development and Readiness Command
Scientific and Technical Information Team - Europe
ATTN: DRXST-STL, Dr. Richard B. Griffin I
APO New York 09710
Commander
US Army Armament Materiel Readiness Command
ATTN: DRSAR-RDP 1
DRSAR-SC 1
DRSAR-QAE 1
DRSAR-IRW 1
Rock Island, IL 61299
Commander
US Army Armament Research S Development Command
ATTN: DRDAR-PMP, Mr. Donald J. Fischer I
Dover, NJ 07801
Di rector
US Army Materials and Mechanics Research Center
ATTN: DRXMR-M 2
Water town, MA 02172
Commander
US Army Maintenance Management Center
ATTN: DRXMD-A 1
Lexington, KY A0507
Commander
US Army Electronics Research and Development Command
ATTN: DRSEL-PA-E 1
Fort Monmouth, NJ 07703
122
DISTRIBUTION
Copies
Commander
US Army Missile Research and Development Command
ATTN: DRDMI
Redstone Arsenal, AL 35809
Commander
US Army Tank-Automotive Materiel Readiness Command
ATTN: ORSTA-Q
Warren, Ml ^8090
Commander
US Army Tank-Automotive Research and Development Command
ATTN: DRDAR-UL
DRDTA-RKA
Warren, Ml 48090
Commander
US Army Aviation Research and Development Command
ATTN: DRDAV-EXT
P.O. Box 209
St. Louis, M0 63166
Commander
US Army Troop Support and Aviation Materiel Readiness Command
ATTN: DRSTS-PLE
4300 Goodfellow Blvd.
St. Louis, M0 63120
Commander
Army Ballistic Research Labs
ATTN: DRXBR-X
Huntsville, AL 35804
Commander
Harry Diamond Laboratories
ATTN: DRXD0-RCD
2800 Powder Mill Road
Adelphi, MD 20783
Commander
New Cumberland Army Depot
ATTN: SDSNC-QA
New Cumberland, PA 17070
Commander
Pueblo Army Depot Activity
ATTN: DRXPU
Pueblo, CO 81001 123
DISTRIBUTION
Copies
Commander
Red River Army Depot
ATTN: SDSRR-QA
Texarkana, TX 75501
Commander
Sacramento Army Depot
ATTN: SDSSA-QA
Sacramento, CA 95813
Commander
Seneca Army Depot
ATTN: SDSSE-R
Romulus, NY ]kSk]
Commander
Sharpe Army Depot
ATTN: SDSSH-QE
Lathrop, CA 95330
Commander
Sierra Army Depot
ATTN: SDSSI-DQA
Herlong, CA 96113
Commander
Tobyhanna Army Depot
ATTN: SDSTO-Q
Tobyhanna, PA 18A66
Di rector
US Army Industrial Base Engineering Activity
ATTN: DRXIB-MT
Rock Island Arsenal
Rock Island, IL 61299
D i rector
USDARCOM Intern Training Center
ATTN: SDSRR-QA
Red River Army Depot
Texarkana, TX 75501
Commander
US Army Tropic Test Center
ATTN: STETC-TD
Drawer 9^2
Fort Clayton, Canal Zone
DISTRIBUTION
Copies
Commander
Anniston Army Depot
ATTN: SDSAN-QA
Anniston, AL 36202
Commander
Corpus Christ! Army Depot
ATTN: SDSCC-MEE
Mai 1 Stop 55
Corpus Christi, TX 78^19
Commander
Fort Wingate Army Depot Activity
ATTN: DRXFW
Gallup, NM 87301
Commander
Letterkenny Army Depot
ATTN: SDSLE
Chambersburg, PA 17201
Commander
Lexington-Blue Grass Army Depot Activity
ATTN: SDSLX
Lexington, KY 40507
Commander
Tooele Army Depot
ATTN: SDSTE-QA
Tooele, UT QhOJk
Commander
Holston Army Ammunition Plant
ATTN: SARH0
Kingsport, TN 37660
Commander
Indiana Army Ammunition Plant
ATTN: SARIN
Charleston, IN 47111
Commander
Iowa Army Ammunition Plant
ATTN: SARI0
Middletown, IA 52601
125
DISTRIBUTION
Copies
Commander
Joliet Army Ammunition Plant
ATTN: SARJO
Joliet, IL GO^k
Commander
Lone Star Army Ammunition Plant
ATTN: SARLS
Texarkana, TX 75501
Commander
Louisiana Army Ammunition Plant
ATTN: SARLA
P.O. Box 30058
Shreveport, LA 71161
Commander
Milan Army Ammunition Plant
ATTN: SARMI
Milan, TN 38358
Commander
Radford Army Ammunition Plant
ATTN: SARRA
Radford, VI 24141
Commander
Riverbank Army Ammunition Plant
ATTN: SARRB
Riverbank, CA 95367
Commander
Scranton Army Ammunition Plant
ATTN: SARSC
Scranton, PA 18501
Commander
Twin Cities Army Ammunition Plant
ATTN: SARTC
New Brighton, MN 55112
Commander
Volunteer Army Ammunition Plant
ATTN: SARVO-T
P.O. Box 6008
Chattanooga, TN 37^01
126
DISTRIBUTION
Copies
Commander
US Navy Materiel Industrial Resources Office
ATTN: Code Okk, CPT L. C. Dictmar '
Code 227 '
Philadelphia, PA 19112
Commander
Air Force Materials Laboratory
]
ATTN: LTM
Wright-Patterson AFB, OH kSMl
127
DISTRIBUTION LIST UPDATE
COMMENTS
Date: Signature:
Technical Report #
li -a
O. a) o
o 4J *-
o c ^ +J t- -o e
CTi o 1- a) m
c 4) w c_^ O c
-. i, c n a. O
o O -O o -^ > i_ 4J o) o m
. 41 4 V ^ *0 U a> 0) re
4) E a) a
i O E
i-
1 3 M s
0) E h 3 -C
c
n O h
3
O) Q. l-
-Q - lt-
ra o a
O
E u- < s c
C *J
4) Q CL
O
) 3
~- Q. I_ 3 C J
i- m i > C TD L X Q-
:3 ro .a
w
o j c ra QJ
e a) <u . : E Q >
O 3
i ^ ^ nj JD
0) E
r
'- t CJ jr
1. D "O (S
fU O O D Q. - o o ra O o D Q.
' ta w > c J-> re re > c
, ~ en - t ~ cn
. -a L- u. a - -a i- u.
OH) m
OJ
iu ra ~ u re ore c H- - -o m o re o to
-
E 0) (0 i- c
-o -u
a: CL J i c
s (J
e: co
V-
co
w
o c
>o o re en O l- -Q
C u
u O " o
WOO
<3
c
re reo en u u r-
o *J O LA
U- in cn r-^
u
(/) o (0>.>* CO
i_
m
reo o en
C VI O
o
L. U Q. '
tu Z u O U ID ^ CL o re re .o o o
c S cr. o
re Q.CT%
j>
^ m
Dl "O LU o.
E zr
'^1 o M a *-
Q
u i-
c -o
ra i-
i-
re re *J
x f5 u !_ C T3
CT>
3 re Q
co >
re E
o re 4-1
O C tJ .M \~ -a I 4J H -O
0 re o. o 04)-Q 0'J>>-*JOQa
O 41 J3 O *-< >
4J 4) re 5 ^ "o u 4) a re 3
4) E 1 a. re E
s:
_< J\-
C E 3 JZ E C0
(- a E w
Q. L
E
_ 1- =3
4-'
cn
C c
4) O 3
Q. 1-
0 -o
'-
<^ C E 4J 5
JC)
IB
L.
O O
c e 4)
c O 3
s-
gtt- 0 e
a. 1-
u 3 n. 3
(J 3 e -o
q c
X n a 3
^: L.
3 re JD u
(0
c re ro C -CT 1- X
0
CL
0 1 re > 4) E 4) 9 E re >
O O v- a. 3
re a CM
j: ' C1 0
O I- O O 1
(D O O 3 Q.
re re > c
c:.
O
0 U
^5
3 Q.
re re. u> c
0
u-
^ U
c
3
c
reE
C
L.
10
u T) V- U. 0
a.
re 0
(3
a 4)
0 re
"a
c:
z O 4>
*-
0
a-: Q. 3 w E 4) CJ u c
L>
CC
L
OO >4)O c
0
Dl O H XI
C 4-1
0 0
O
O
O
cC as 0
CM
c *J 0
>.re>
O J-" 4J LJ- at r-. 0 0 iJ 4-
O O m c w in u
0 0
W 4) ._ O h* oi-
< 4) .- a r-. a
>. 14- Q CO
<c O
"O
<u
4-< 4)
C 4-1
O.
W
^
t>
D. -O E a o >- a o a O- J3 E O 0 >- a. O O "
<L 0
< : -c o 01 a 5 E < O
4-. V. < 4-' u
s: 0 cn 4)
U 4)
E in tn
L
4)
1-
4-. VO
to re o^ O u.
a 0 0 O
a.
3 L
c
^3 4) CJ 0
-a 0 u. 1 0 a. o c cr o xi 3 13
c
4)
1 1 0
l-
00
O
O
re *fr-
4J cr 0 0 3
4) O c
c O
fc;:
CO
r-.r- L.
a H
r*. Fv
<n
i_ 3
c w 0 4> C
CTI U L. 3 >-
CL 4)
0 1 0^ Z 6 ta z c CTI
O LO Z fM "O a. LU
CM 0
0 reE 0 4}
E "a O N C >- O H
-0 JC
^
0-0
01 T) (X LU (^ 0) u +J
c c s: C O 4->
c cre
::
i 4) - JK; to as 4) u- cn u t- fD L- x: ro a
^f
u t.
X 0 t-
t "O o 3 c o o u
0
0 K a O 3 c O
*J
O in c
re CT>
<-c4)
0
a u m O t- 4) -a E E
^: c E re*-*-
^ O in 0 -O
or
CJ in
(1) -1 u 4-t 0 in
I-TS 4)TJ
a: a CJ d. a> C 3 v- 3 0 ^- "i +J ro
c ro xi ~ in L.
c
J* <Hcn < 0
D_
4)
CO
v.
0 vi
re x.
s: cr, U
<\- < 0
D-
L.
O
Q u O O O
Q.14- L. ro re cn
-
or o> 0
0
. 0
Q.3:
a <
0
c <-)
Q C O IS,
s3
Dh
4J <
S O
c
4J
JZ X O
1-
E V L ro 4-> re > 0
JC w 1- C L
O a. 0 re o. Q.
H *- o O a. O U 1 UJ ct: OC => O
~ ^ _^
r* 01
a
3n
O
Q_
TD
A
M -J O
-n-o o
o
n
*
-i
-a TU
T
o n
T
H
re
3-
3 a
X
rr
TD V>
re-n ore
-1 X3
c ^
to W
rt 0) X)
3- 3 "I
re a. o
n reo r TT D.
c
-h
O
A
C o
3
(D
W> W 3"
Q.
c O
3
0 0 3
~
(D
^
rt ^ 0 3
n
-i -n A rt 3
rt
re~ rtre
Q r 3 CJ 3 rt 3 -.1J3 3 O ; rt 3
n o
T)
T O
3
O vi
3 - O
3
A
ZJ
=! <
o
CL
-i O
3 b
-t
O to
O (U 3 -
3
O
3 3
3
a.
i O 3" rt
< **-<
n
| (O
o
o
-t 3 H A
o
-h r+
Q.
"O
ZT
? cr
a
o. o
-i
3 E
JO
M
M
3 -1 A
O
CL
o
3"
Q. o =r zr a O c -C zr re O A
a a zr g
3
Q. o
Q) CB n
O
3
o
A O
A
o
re ^ U) A
B
CO O
3
A
0
o
A
o1 o"
o o
C--3 CN
A 3
3 (A
< =i
A
a a.
-ft 0
re re
^ XI
(J, -I A 3
3
<o
3
3 t
a
o
VA> (0 3
C <
a.
o
o
a.
re
O o -
X!
-i g
Qi
^J rt tO
C < 3
o D.
w < <
o 3 o X -1 r+ TT Q- 3 Q A
s o kQ O 3 (Q C 0 B to O 3 t to c o c:
o
^s
a
l
3 T B
IT CU
n
o <a
to T
C 3 A
O
re re T -1 "D
O to
A
O.
i o -i
: 3 CJ re re
TJ
n
EP
1
o
3 A A
3 A
A
3 3
Q. TT A
3
-h
C
'^
3
O
IQ n-
Q
Q.
01 o 3 3
. O Q-
o o Q- 13 3 3* 3 3
o a CL fU CO
( 3 i Q. o a ia 3 3 CL x re
3 a a
3
(O a 3 T3
fl) o
n o
3
T
QJ s a. x n
3 N.
-o *
n
A
3
O. -1 -ft
rt o.
re o
-h
3 3 re m n
rt < fl> "D -h
3 ^ (B z
I -h t -i
re o
Q.-D O. kQ 3 o :c Z C IQ
a rt S CO a m -.
o
a O
A A
CL *< QJ in
-i
rf 3
n o Q.
o
O
o
B
I-J re
=r ^ D 1 re ii re re
o
3"
n
Q.
3* 3
o
o a,
A
CJ O
-j re
M H
N
O
-i
CJ
T3
A
tn
1
-i B 3
re i re rr 3- 3 -l X) - rt i
CJ n A O 3 A -o -o i a. o 3 - rt
3 A "a O -n re x) o
n <u 3 co n
D.
*< O
O
A re o"
O
::
0
-h t
-) 3 -1 A
0-1 Ul -t! -ti O
O
"D
0
a
"O
^ re
PJ
CL
r+ -1
n- A
r^-
o
A
D- re
re _
O rr
3 C
: A O at o.
_
3" 3 -t er
rr re
ix 3 3 0 A "O Q. A *< 3-
a
.^ Q 3- i
A
o
oo
j
A
re
3" C
A
tf> A
3
U3
ft 3 -^
n 0 n> o
3 o
\J-I re T3 r+ O
-o A (A
re 3
< a.
o
3 < W
(O A O A 3 rt 0) --K
ri
"O
T
O
o
c
o
3-
3 re
n Q. rr CJ
rt
TJ
3
a. A
-hlfi rr
O
QJ
(A re
CL
o n
D)
crx)
IQ 3 C T l. -h CJ
> o o cr-o
nn
a.
,0
(O
3
O 3
_ J
a. =) T
(O ^
A
re -i
h T
- to X3
A
A -\
o
0) a. T
> A
tfl A rt
3" "I
A
N 1
A U3
O
3
A
T
3 re - re 3 ti ro re o <
3 01 o> A T 3 Q> rt ^ 3 A
o o
o 0 Q.
ON
-J
3
Q. ae 1
3
rt 1 J
A J 3 O -ft
3
<
3 n z
3
*e
-J
A a.
O A m tu
3" 3"
T A
0
o
CJ 03
CL
-i
A
o
O
1
O to
3 a> o A o re n -H - Q -ti
o
<tu *< n
O
3 3* re a.
n ^ 3^
3 3 - 3 3
fD rt IQ
3 3
to
o
3 O
A
! A
A
A -<
3 - 3
A A u o. 3-13
O CJ 0) CL
D)
0
0)
3
Q.
c T
to cu
A
A XI 3
3
rt A c
A
CL
3 XI
rt o -< -ft
T A
O
cjioatDOrtretn
3
O O.
C T -'
tO 0) {U X3 3
3 XJ 1 A
- rt O < -h
W N O
3
to
=r
-\
A 3
rj A
-i
A
-i
A
A
^ o
1
U3 ^-<
o
-1
o A
T
w O
* -n
3"
QJ re re
-i
A
rt o
re 3
3 tO
rt Qj TJ 3 A X) W C *-^
3-3-1
A a. o o
A 1
re A
3
i
-I
O
1
A
ST
Q Are -
to 3" CJ c O -COW
I A rt 3 -i -h
I W ID A 3 i o -i re
rt E O
n
l
I O
O
CJ
O in
3 - *<
T O
3
A
to re re to
3 o
'i
A
3
VI O A
n O
3 0
3
A 3
3 Q.
o
3
rti D n 3 K> Q.
O tr O A
o. o -n
T
n
Q.
3 -Q
C o
<
VI
-1
3- -< A-
M
A
A
O
rt
O A
X>
3"
re 3 o j- 3 A
a. o 1 Oi tn
("l! A Q. n
A
t^ O 0
O
A
A S 3 A A A 3 O
to O t A T T "j to 0 A 3
o a.
j
)
os -. -i D ;
c *< (
o
P 0 n o
A
\
A
o o
-J 3
Q."0
O c
1 A
3
<
3 to
O a. 0
<
3
> sr a. 3 c A o 7r a.
) w to n ;
13 -J 1
O X
T
to c- o
o
C A
3 1
3
o
o
3
-o
A
< O X
-1 rt Id C
M 3 I
O C
rt
rt a
C A A
t - CJ
_. CJ n ;
3- A C
3
3
to
A VI 3- A 0 to o.
> A o. TT ( to
o O A 3 r> 3
-h re re 1 3 A O 7?a.
o
-i
o
o
O
3 3
O.
A -1
o A A
O
O 3 A
3 A
Q. 3 3
Q. A to
3
3
rt c
-1
to 3 ii
I O
O
A
' rt < 01 A
3
O 3
Q. rtre o-h
3
3
3
3
A A
a. >t
O
A 3
"<
A TJ
3 *s.
-h A
A O -l A
i xi re 3 A 3 CL : a. -i A
ux) o. to X) 3 A 3 OL O
3 re
i
- < rt rt
C
A
to
3
< Q.XI
A t
O. o a.
lO 3 A
rt <
CJ I/I _
<
o
XI
* w 3 3
--j re
3* 3"
A A A
T D
re O O O. O rt (o cu re re
A A
-\ c 3 ffi o tfl re 3- rr
S rt -ft n o
-n o. o 3 > A X) "O
A X) -t -I D rt
co -i oi ~l
re~ a.
O H 3 A
Ui O CJ i-t -i
C rt ifl X) O
o. re <
n cr re rr
T
_A
.3OO30J QX)
O
CL 3 n O 3 A A "a rf o. a> < zr"
-i zr
c
re n o re o
3"
A
rt r e0a.
A
3 9 t
A O
3
O
O
o
o
VI A X)
tO f) QJ 3 -i o 3 o
o ~. -. 5- ~ A 3 o
XI
o "
A
3 O
IQ 3
rt
C
-O
M
3
-hlO
"O
o
1
A O
3
to
0 XJ
3-
-h
3
3
^ re
-h' X) CL Qj - O n crx) *o
3 C
Q.
3"
o A O O crX>
-I A -i rt O ~t n 3 A
(=,0 3 rt a x) A -1 A -l (O 3
O O
3 "
A
D
-^1
0
CL i
>
CJ
O
m re
3" -1
O. 3
rt Q tO tn
O
a
rt d c lO
o.
>
3 o.
A
D O.
o n
I
3
3 1
ON 3
A (D
3
3
to
A
(/.
QJ
3
rt-
a. -i
3
3
A L^
tU
3-
A A
3
0
3
<
-4 a.-< 3 A o 3
o. re re a.
3- 3 XI -n o Q- ^J CL-< 3:
--J rt o re
-i O
rt A Q.
3
-i tN? 3" 3"
a.
A
X
tr
A
< *<rt
o o A A
o a.
1 DJ n A -, A
O
3
1re
O -i 3 0)
O
A
3
CL 3 0
-i
-I
3 111 -< *<
o- cr 1, 3
A CJ
s oi "<
3 O rt re in 3 XI ~l
C -i
tO 0> 0) "D 3
3" 3
3 "I IQ n x) A
O O
3
CL
3 o
c n
A
A
A o 3
rt
3 3 A
A <
A
O
J> rt tn Q. O QJ -i
-fto ex oo
"O
-1
I
3
S > < 3r 3 3
A.
(O
o A
"O
-i 0
O X)
re re
< Xi A 3 O O
-*i 3 re A
fto reQJ r+ o 3
*p
3 A
a A 3 o o
O
rt -ft re
a. 0
n
w
oi re
-i
re re -i u
O
3
-1
3
A
1
A
o
-< o 0 o
Ato
re 3
o o
"O
3
re A
t/i
3