l..iii .. -. . . .. - nasa · engineering experiment station ... laboratory set-upof lever arm...
Post on 08-Sep-2018
221 Views
Preview:
TRANSCRIPT
\.~~ i,i
., "- ..._..._--_....,-- ..--...._-~~~.'.' .~...,..-_.,~.-:"-~ ..... _.._-:.;......_ ..-...._..... _._ ........__..... - - ... - .-'------.. -_.- .- _._~.-:~-I -.. ~-
ENGINEERING EXPERIMENT STATION·COLLEGE OF ENGINEERING
THE UNIVERSITY OF ARIZONATUCSON, ARIZONA
i.
i'
'-i[
I- I
I
, I
I •
!i
;.. ~
,I
'-',-
j1j.
')~
i;,~-.,
N73-12497
I\ .I
III'l..__.... .._-. . .__ __ .._
./
CALIBRATION OF COMBINED BENDING-TORSION FATIGUE
RELIABILITY RESEARCH NACHINES AND
RELIA3ILITY DATA REDUCTION
by ,
Dr. Dimitri Kececioglu and Jeffrey B. McConnel:
._. - -- --~._- -- - .- -". _._-------- ----_.•. ~------ _."------------
(NASA-CR-72837) CALIBRATION OF COMBINEDBENDING-TORSION FATIGUE "RELIABILITY DATA
=REDUCTION D. ~ececioglu. et-al (ArizonaH~IUniv•• Tucson.) 31 Jul. ,1969 166 P CSCL
14B
https://ntrs.nasa.gov/search.jsp?R=19730003770 2018-09-08T03:15:39+00:00Z
NASA CONTRACTORREPORT,
/
'1;213JNASA CR-~
CALIBRATION OF COMBINED BENDING-TORSION FATIGUE RELIABILITYRESEARCH MACHINES AND RELIABILITY DATA REDUCTION
by Dr. Dimitri Kececioglu and Jeffrey B. McConnell. ;10.1--' .
Prepared under Grant No. 03-002-044 by The University ofArizona
Tucson, Arizona
for
NATIONAL AERONAUTICS AND SPACE ADMINISTRATION - WASHINGTON,D. C.
CALIBRATION OF COMBINED BENDING-TORSION
FATIGUE RELIABILITY RESEARCH MACHINES
AND
RELIABILITY DATA REDUCTION
by
Dr. Dimitri Kececiog1u and Jeffrey B. McConnell
Prepared For
'NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
, July 31, 1969
Contract NG~03-002-044
Project ManagementNASA-Lewis Research Center
Cleve[and, Ohio
Vincent R. Lalli
,
College of EngineeringTHE UNIVERSITY OF ARIZONA
Engineering Experiment Station, Tucson, Arizona
\
\
i
ii
ABSTRACT
The combined bending-torsion fatigue reliability research ma
chines, conceived, designed, and built at The University of Arizona~
are described. Three such machines are presently in operation at
The University of Arizona. The calibration of these machines is
presented in depth. Fatigue data generated with these machines for
SAE 4340 steel grooved specimens subjected to reversed bending and
~teady torque loading are given. The data reduction procedure is
presented. Finally, some comments are made about notch sensitivity
and stress concentration as applied to combined fatigue.
/
\
iii
'TABLE OF CONTENTS
:Page No.
Abstract . . . . . . . . . . '. . . . . . . ii
List of Illustrations
List of Tables . . . . . . .
. . . . .- . . . .3
1
xi
vi
ix
. . .. .
.... .
. . . . . ... . . .
. . . . . .I Summary
II Introduction
Symbols
III Description of Combined Bending-TorsionFatigue Reliability Research Machines 9
IV Calibration of Research Machines . . . . 13
A. Calibration Requirements • . ./
,13
B. Bending Calibration 15
.1. Visicorder output from bending straingages on toolholder versus true stressin toolholder ..' • '. • • •• • • • •• 15
2. Visicorder output from bending straingages in specimen groove versus truestress in specimen groove ••••• '.. 20
2.1 Experimental determination ofaverage stress concentration~actor in specimen groove . . . 23
3. Torque interaction into thebending bridge output ••••••••• 25
\4. Relationship between toolholderbending
stress and specimen groove bending stress 30
1. Effect of. axial load on bendingbridge output . . . . . ... . . . . 44
/
: 2. Measurement of the axial stress .. . . . . 45
Page No.
.' ,
Research Data Reduction
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
48
. iv
54
51
52
51
42
45
47
44
38
33
37
37
31
. . ..". .
2.1 Effect of bending on torquebridge output - additionaltest • • • •
4.2 Toolholder bending stress versus .specimen bending stress •
Torque Calibration • •.
1. Torque load versus visicorderoutput . . . . . . . . .. . • . . . .
2. Bending interaction into the torquebridge output • • • •
2.2. Measurement of the axial stressas a function of torque load • • • •
2~1 Measurement of the axial stressas a function of bending load . • • • • •
. 4.1 Toolholder strain gage output .versus specimen strain gageoutput • • • • • • • • •
E. Summary of Calibration Results •• ',' ••••
A. Conversion of Visicorder Outputs to StressLevels .- .. . . . . . . . . . .. .. '. . . . .
B. Reduction of Endurance Strength Data Forr = Q). and r =0.90 • • • • •
s s
C. Generation of S-N Diagrams Using Distributional Stress Levels and Mean Cycles-To-Failure . . · ... . . . . . . . . . .,. .'. . . .
D. Axial Effects
C.
v
v'Page No.
,D. Generation of Theoretical S-N Diagram · . • 55
E. Comparison of Theoretical and Experi-mental S-N Curves . . · · . . · · 58
· .VI Stress Concentration and Notch SensitivityEffects'. • . • • • • • • 59 '
A. Stress Concentration · . . 59
B. Notch Sensitivity Survey · . . 60
C. Proposed Notch Sensitivity Determination 62
VII Conclusions. •
VIII Recommendations.
Acknowledgements . • • • .'
References • • • • • • • • •
• • •
· . . . . . . . .· . . . . . ~
63
66
67
68
./
Appendix A. Data Reduction Results for r = cos,
Appendix B. Data Reduction Results for, r =0.7 and0.9 sr = . . · · . . . . · · · .' .
<('.
133
139
Appendix C. Operating Checklis~ for Combined BendingTorsion Fatigue Research Machines . . • •• 144
Appendix D. Procedure for Operating Combine4 BendingTorsion Fatigue Research Machines with .Combined Bending and Torque Loads . • • • • ' 147
Appendix E. Calibration of New Torque Gages forMachine #3, June 1, 1969 • . • • • • • • •• 149
Appendix F. Maintenance Checklist for CombinedBending-Torsion Fatigue Research Machines 151
LIST OF ILLUSTRATIONS
Figure No.
1 Mabie-Gjesdahl fatigue testing machine
2.- Complex-fatigue reliability research machine front view
3 - Complex-fatigue reliability research machine top view
4 - Loading frame analysis and test specimen
5 - Bending load lever arm
6 - Strain gage and slip ring arrangement
7 - Bending moment instrumentation
B - Torque instrumentation .
9 - Torsional stresses for one gage
10 - Test set-up for the bending cantilever calibration
11 - Test set-up for torque interaction into bendingbridge output
12 - Torque interaction into bending - Machine #1
13 - Torque interaction into bending - Machine #2
14 - Specimen groove output versus toolholderoutput - Machine # 1
15 - Specimen groove output versus toolholderoutput ~ Machine #2
16 - Specimen groove output versus toolholderoutput - Machine #3
Page No.
71
72
73
74 .
75
76
77
78
79
80
81
82
83
84
85
86
vi
Ij
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
..
Figure No. Page No.vii
17 Free body diagram of toolholder 87
18 - Laboratory set-up of lever arm ,pivot point,resisting moment measurement experiment
19 - Scale graduations of measurement experiment
20 - Torque output versus pan weight- Machine #1
21 - Torque output versus pan weight - Machine #2
22 - Torque output versus pan weight - Machine #3
88
89
90
91
92
23 Bending interaction into torque - Machine #1 93
2~ - Bending interaction into torque - Machine #2
25 - Bending interaction into torque - Machine #3
26 - Bending interaction into torque at high torquelevel - Machine #2
./
27 - Axial output versus .bending load - Machine #1
28 - Axial output versus torque load - Machine #1
29 - The calibration flow chart
30-- Data reduction program
31 - Pan weight versus true bending stress forendurance strength for r = 00
s
32 - Pan weight versus true bending stress forendurance strength for r = 0.90
s
33 - Cycles-to-failure distributions and endurance strength distribution for r =~. 5
34 - Cyc1es~to-failure distributions forr s + 0.70 "and endurance strength distri
.bution for r s =0.90
. J.
94
95
96
97
98
99
100
103
104
105
106
Figure No.
35 - Fatigue strength versus meancycles-tofailure and endurance strength for r ofro, 0.70~ and 0.90, and the theoretic~lS-N diagram for r =ro
S
36 ',- Three-dimensional distributional Goodmanfatigue strength surface based on researchresults, reported here for 10 7 cycles oflife
37 - Notch sensitivity of normalized steelspecimens
38 - Notch sensitivity of quenched andtempered alloy steel specimens
. 39 -Curves of notch sensitivity versusradius for steels, bending or axialloading
Page No.
107
108
109
109
110
viii
40 - The fatigue notch factor
41.- Notch sensitivity charts for steelsand 245-T wrought aluminum alloyssubjected to reversed bending orreversed axial loads
/' 110
III
42 - Notch sensitivity curves for materialsin reversed torsion
43 - Bending interaction into torque Machine # 3, June 1969
44 - Torque output versus pan weight Machine # 3, June 1969
\
III
112
113
/
LIST OF TABLES
Tabie No.
1'- Instrumentation
2 - Too1holder output versus pan weight datafor bending bridge calibration
3 - Toolholder output versus pan weight data:in incremental form for bending bridgecalibration in terms of change in toolholderoutput divisions for 10.27 lbs. of pan weight
4- Calibration specimen groove strain gage outputversus pan weight data
5 - Reduction of Machine #2 data given in Table 4
6 ~ Reduction of Machines #1 and #3 data given inTable 4
7 ~ Torque output versus bending output data from~orque into bending interaction calibration
8 - Too1holder versus groove bridge output datafor quasi-static calibration of groove stress
9 - Visicorder divisions converted to strain
10 - Torque output versus pan weight data for torquebridge calibration
11 - Torque reduction table
12 - Bending output versus torque output data forbending into torque interaction calibration
13 - Bending output versus torque output data at hightorque level for bending into torque interactioncalibration
14 - Axial output versus bending output for axial tobending interaction calibration
, 'Page No.
114
115
116
117
118
119
120
121
122
123
124
125
126
127
ix
List of Tables
15 - Axial output versus torque output for axial loadinto torque interaction calibration
16 - The calibration coefficients for each researchmachine
17 - Stress levels and visicorder divisions forr =3 and Machine #1s
18 - Torque output versus pan weight data for newtorque gages - Machine #3, .'. June 1969
19 - Bending output versus torque output data fornew torque gages - Machine #3, June 1969
/
/
--Page No.
128
129
130
131
132
x
Symbol
C
div
E
F
G
I
in
k_ a
Kave
-~
~GR
~/T
~TH
xi
SYMBOLS
-English
Radius of shaft, in
Divisions
Yo~ng's Modulus, psi
Force, lbs
Strain gage factor
Moment of inertia of cross-section,in1+
Inches/
Surface finish factor
Experimentally determined staticstress concentration factor
Size factor
GroQve bending calibration coefficient
Bending into torque interaction -coefficient
Toolholder-bending calibration coefficient
Temperature factor
Stress concentration design factor
Dynamic stress concentration factor
Miscellaneous effects factor
Symbol
KGR- TH
L
lbs
M
N. cal
N •v~s
q
r, s
Rcal
Rgage
Se
w
w .bear
xii
English
Groove~toolholder calibration coefficient
Toolholdertorque calibration coefficient
. Static stress concentration factor
Static stress concentration factor inshear
Torque interaction into bending calibration coefficient
Lever arm distance, in
Pounds
Moment, lb-in .
·Number of active arms in strain gagebridge
Visicorder calibration divisions, div/"
Visicorder output, div
Notch sensitivity factor
Stress ratio
Calibration resistance, ohms
Resistance of strain gage, ohms
Mean of endurance strength, psi
. Endurance strength, psi
Fatigue strength for 10 3 cycles, psi
Ultimate strength, psi
Pan weight, lbs
Loading, lbs/in
Weight of the bearing, lbs
.~
Symbols
wcol
wcoup
Wnut
W 'slip
v
Symbols
T
xiii
En~lish
We,ight, of the collet, lbs
,Weight of the', coupling, lbs
Weight' of the collet assembly nut, lhs
We,ight of the slipring assembly, lbs
Coefficient of Variation, %
, 'Greek
Normal strain, micro in/in
Mean
Normal stress, psi
, Standard deviation of normal distribution
Standard deviation of endurance data,psi
Shear stress, psi
Symbols Subscripts
B 'Bending
GR Specimen Groove
out Output
TH Toolholder
T Torque
vis Visicorder
- '--~.- ... -- -- ~----- .-.
1
I SUMMARY
The calibration effort and data reduction techniques for the
Combined Bending-Torsion Fatigue Reliability Research Machines is
presented here. Because of the complexity of the problem of deter
mining the true bending stress and the true shear stress present in
the groov~ of the test specimen, eight distinct calibrations had to
be performed on each machine. A description of each calibration
test, the test setup, the procedure, the data, a~d the data reduc
tion are given for each calibration test. Then the need for each
test is presented in Section IV. E where a calibration flow chart was
developed to aid in the data reduction procedure. Specific cali
bration parameters for'.each machine were determined and their needs
demonstrated. The calibration equations in bending were shown to\ '. . . .
" /be ',,
N N' G R,cal a cal= ---:::-:::------,-- <1 IE R
gageout
TH
and the calibration equations in torque were shown to be
<1trueGR=r 13s
.. l,
2
... -. -,.- .."~ -- -=- ....
';_':,' "t\':.. '": "';~-.-:-.:.~. ':.~."'::-~ _'0 ' ...
- - In -add~t~on, the S:tClt~c stress concentration factor for the notched
U;~sp'ec{~e'~" was' :det-ermined to be 1.28.
:C:"~<'~~-For data reduction, the calibration flow chart was generalizedc..~ _._~ ...-,- ,', .- " - - '. . .
·,··arid'computerized. The data reduction program as well as all the
"'='datagenerated to date are presented. S~mple calculations for the
"data reduction technique are given. Cycles-to-failure data reduc
~:tion is not included in this' report and is the subject of another
"report.
The results of the data reduction are briefly presented in the
form of S-N curves.
Lastly, the problem of notch sensitivity is discussed and a
proposal made.
_.-"..-.'-'~-"'" ~
3
II INTRODUCTION
Prior to this research effort, the basic methodology for
designing reliability into mechanical components by consideration
of the interf~rence of their stress-strength distributions was dis-.'. .
cussed by Kececioglu and Cormier (I)". Included in this paper was
a discussion of Monte Carlo techniques for determining stress and
strength distributions, given the distributions of the factors
affecting them.
Freudenthal (2) wrote a paper in which structural unreliabil
ity was considered to be the probability, or risk, of failure. The
safety factor was shown to be a distribution function which is the
quotient of the strength to the stress, where both strength and
stress are considered as statistical variables. Freudenthal,
Garrelts, and Shinozuka (3) prepared a comprehensive report, along
the same lines, which discussed in more detail the mathematical
techniques,required, the appropriate statistical distributions .~
involved, and problems which remained to be solved. Several exam~
pIe problems in structural reliability were worked out, an exten
sive bibliography was given. These efforts concentrated on simple
fatigue and structural reliability.
The Batteile Memorial Institute and its Mechanicai Reliability
Research Center presented studies (4, 5) which described some of
the fundamental problems in mechanical reliability and suggested
methods for their solution.
Mittenbergs (4) discussed the fundamental aspects of reliabil
ity engineering as they pertain to mechanical devices. He stated
that the failure modes of ~echanical elements were basically:
icNumbers in parentheses refer to those under References.
1. Deformation.
2. Fracture.iI
4
3. . Instabili ty .
He also asserted that many factors combi~e to determine the
reliability of a mechanical part under such failure modes. The
interaction of strength and load distributions· was discussed. The
Sixth Progress Report of the Mechanical Reliability Research Center
(5) summarized a two-year research effort. This extensive research
effort contained a thorough discussio~ of mechanical reliability,
and attempted to quantify the relationships of various factors on
.such phenomena as creep and fatigue. An extensive bibl~ography was
included.
The· lIT Research. Institute conducted·a program in "Methods for
Prediction of Electro:"Mechimical Systems Reliability" (6). The
p~ogram was concerned with three major areas:
l.The study of prime mechanisms of failure in mechani-....cal design. Specific items included fatigue, surface. . \
. fat.igu~, .wear, creep,. and corros ion. ~vd i(-... 2 i . The application of failure mechanism and design \lY)c.~d
information for the reliability evaluation of speci- . 0ficmechanical pa~ts. Parts included were gears,
bearings, springs, and shafts.
3. The determination of mechanical .system reliability in
terms of individual part reliability figures.
A paper by G.Reethof, M.J. Bratt, and G. W. Weber of the
Large Jet Engine Department, General Electric Company, entitled "A
Mod~l for Time Varying and Interfering Stress-Strength Probability
Degradation" (7), provided a computer approach towards.the solution
of the time variant strength distribution case.
An extension of this study was made by Lipson~ et al (28), who
conducted an extensive literature survey,. gathered available fatigue
5
data, and developed an analysis of the stress-strength interference
theory using the Weibull distribution extensively.. . i
The above works provided some interesting and valuable contri-
butions to the problem of designing specified reliabilities into
mechanical components. However; a number of important aspects of
.. this problem remained to be investigated. The problem of time
variant stress andstr~ngth distributions needed further treatment.
The effects of various factors, which are themselves distributions,
on the distributions of the failure-governing stress and strength
had yet to be fully explored; The development of a formal engineer-.
ing design met~odol.ogy for designing mechanicai components had yet
to be developed. Finally, much of the work .in mechanical reliabil
ity theory suffered from a lack of statistically adequate data, due
to a lack of test results on a large number of identical mechanical
components.
The purpose of tl:J.e current invest.igation is to fill in the
gaps in the above-mentioned areas, with th~ following specific
objectives:
. 1. Develop a formal engineering methodology for design
·.ing into ~echanical comporients, subjected to combined
·stress fatigue wh~ch involves time-dependent ~trength
distributions, specified reliabilities.
2. Explore the methods of functions of random variables
as applied to structural reliability.
3. Explore the methods available for determining failure
governing stress and strength distributions and
develop new ones.
4. Explore the methods available for calculating the
reliability once the failure-governing stress and
str~ngth distributions are known and develop new o~es.
5. Develop and fabricate fatigue testing machines for
reliability research, so that the explored and
ii·
6
developed methodologies described above can be demon
strated.
·6. Pursue a test program with a statistically signifi
cant number of test specimens to obtain data from
which these methodologie·s can be demonstrated .
. A literature survey was made in order to locate fatigue test
ing machines to generate the desired, combined bending-torsion
·fatigue data. References on Fatigue was surveyed from 1955 to
1963. The only paper of interest was. the "Symposium on Large
Fatigue Testing Machines and their Results ll (9).· No testing
machines capable of handling combined steady torque and reversed
bending moment were found in th·e paper. ' Other ref~rences (10, 11)
were reviewed; information concern~ng combined-stress fatigue
machines was not found.
The Proceedings of the Society for Experimental Stress Analy
sis (12) from 1945 to 1960 and Experimental ,Mechanics (13) from
1961 to December 1965 were reviewed in an attempt to locate a com
bined steady torque and reversed bending moment testing machine.
Several fatigue.test~ngmachines were found, but only one was of
direct interest to the NASA contract, a testing m~chine built by
Mabie and Gjesdahl (14). This machine used the four-square prin-
. ciple for applyi,ng a steady torque while the rotating beam principle
was used to produce the bending moment.
The four-square principle is not a new principle for develop
.ing steady torque. Industrial corporations, such as gear manufac-
turers, speed reducer manufacturers, and coupling manufacturers,
all use this principle to evaluate their products (15).
In the Mabie-Gjesdahl Machine this principle was used to
develop a maximum steady torque of 6,000 in-lb; however, the
machine was only operated at a maximum of about 2,000 in-lb of
torque (14, p. ,86). At this loading the machine produced a high
pitch whine (16), a result of the pitch line velocity of the spur
).
7
gears being 3,000 to 4,000 feet per minute (17).
The pre-set torque could not be maintained. The steady
torque, four-square principle was coupled with a reversed bending
moment, as shown in Figure 1. The desired bending moment was
applied to the test piece so as to simulate 'a simply-supported
beam. Thr~ugh the use of a hydraulic cylinder and associated
equipment the required bending moment load was developed (16). A
reduction in bending moment occurred during testing as a result of
hydraulic cylinder leakage. The bending moment was con-
stant along the length of the test piece for a specific value of
the bending load. ,The machine was designed for 5,000 in-lb and
operated at a maximum of about 3,200 in-lb of bending moment. The
reversed bending moment was gained through the rotation of the test
piece in the four-square mechanism.
The Mabie-Gjesdahl test machine operated at 1,200 rpm. The
machine was driven by a 3hp, 1,200 rpm induction motor (18).
Mabie (16) furnished two assembly drawings (19, 20) and additional
design information as to the problem areas in his test machine.
Mabie indicated that the disadvantages of this machine were that it
was difficult to hold the torque and bending moment, and noise and
vibration were present. However, the machine did not dissipate. .
energy to apply torque to the specimen and operated on a proven
principle.
The exact instrumentation on the Mabie-Gjesdahl test machine
. is not known. However, the torque values were measured and checked
only in a static situation. The ~ending moment values were checked
and related to the pressure gage on the hydraulic equipment. The
load was applied statically and the pressure noted. Strain gages
were used for the static torque measurements and also for the bend
ing load. The bending load strain gages were mounted on the load
i,ng bar.
'I
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
j
BThe test machine was calibrated dynamically with suitablY
mounted strain gages and slip-ring and brush assemblies. The exact
equipment" is not knoWn •.Correlation of'these dynamic tests were
made to the stresses obtained through calculations and an 8-12%
error was noted (14). Correspondence with Mabie (12) indicated
that the commercially purchaseable components exceeded $5,000.00.
With these thoughts in mind, test machines similar to the
Mabie-Gjesdahl ~rincipie were conceived, designed and built at The
University of Arizona, starting the fall of 1965.
Three combined bending-torsion fatigue reliability research
machines ar~ presently in 'operation at The University of Arizona.
A research program is being conducted for the National Aeronautics
and Space Administration under the direction of Dr. Dimitri
Kecec~oglu at The University of Arizona, Tucson, Arizona, and
Mr. Vincent R. Lalli at the NASA Lewis Research Center, Cleveland,
Ohio. The description of these test procedures, test data and
their reduction presented here are part of this research effort.
The objective of this repQrt is to obtain the calibration param~
eters' presented in Figure 29. The experimental tests which must be
run to determine these parameters are described. The data reduc
tion technique isals~ given. The problem of stress concentration
in the notch of the test specimens and the associated notch sensi
tivity are discussed, and recommendations for future work are made.. .
1'1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
I
':i."
9
III DESCRIPTION OF COMBINED BENDING-TORSION FATIGUE
RELIABILITY RESEARCH MACHINES
The combined bending~torsion fatigue reliability research
machines are designed to simulate a shaft in service. The objec-
tive of the immediate research program is to examine the fatigue
life of specimens made of SAE 4340 steel under combined loadings.
The specimens are subjected to reversed bending and steady torque
applied to a rotating specimen with a stress concentration~ which
produce combined bending-torsion stress~ or combined-stress~
fatigue.
General Description of Fatigue Machines
.Each fatigue machine consists of a two-section~ rotating shaft
with a test specimen locked in the center~ as shown in Figures 2
and 3. The horizontal shaft is coupled at each end to allow for
relatively free deflection when the specimen is loaded. A seven
and one-half horsepower~ 1~800 rpm motor powers the shaft. .The
bending load is applied to the specimen by means of two yokes~ one
each on two bearings located symmetrically about the specimen on
two commercial tOOl-holders. Below the shaft~ the yokes are con
nected by a horizontal link, which concentrates the load at a
single vertical link in the center. The vertical link is then con
nected to either a long or a short loading lever arm. These load
ing arms make possible the application of a great range of bending
stresses in the specimen groove, by means of pan weight applied at
the end of the loading arm. One pound of pan weight is approxi
mately equal to two thousand psi in the groove. The torque is
applied by means of a commercial Infinit ~Indexer which is located
on the back shaft of the machine.
11
Flexible Couplings
::'":': "Sier-Bath, all-steel,- flexible coupli.ngs are used because of
r,'tfi~ii- ~biiityt~C~r~~s~ltt~~qu~and allow relative movement of
S§h~fts-h6lding th~ie~t'~pe~imen for proper transmission of the
~beE~i~g_~~~~~t. In addition, they are small in size and relatively
low in cost. '.. -There are two Sier-Bath couplings located at either
ietld·c::of the : two halves of the front shaft and a larger one on the
::-backishaft.' There areshrurik-fit on the shafts.
""-GearBox
...'..
• r;: "-7""".'-':::"
A Falk C6rp6ratI~ngear reducer box is used. It has a mechan-
ical horsepower rating of2iOhorsepower and a thermal rating of
272 horsepower, with'cooling fans. Its speed is 1,800 rpm, and it
has a AGMA gear ratio of -'1:84 .
cMethod of Torquing , '
:-G~__;=~h:e_ to!,q~eis applied, to the specimen/by means of a HDUI-200
;~~r~nit-Indexer mad~ by the Harmonic Drive Division of United Shoe
Ma_chinery Corp. ,Beverly, Massachusetts. The Infinit-Indexer has a• :~.~ .. - - . - - • <-' -.'
,flexible circular gear rotating within a slightly elliptical flexi-
?~~:e, spline-like,' outer gear. At the major axis the gear teeth do
not mesh. ,When the shaft turns, the.inner gear advances very
?lightly ~ith respect to the outer gear inducing a steady torque.,". ,- >". • • • ~ - ••
r~e torque level is adjusted by turning a large hexagonal shell on:::r c. _ -_ ... • - ~ • -:,.••- '-
.tl1e outside housing of the Indexer with respect to the shaft.
Loading. Frame
: The loading frame is capable of producing a 3,540 in-Ib bend
ingmoment in the specimen groove. There are two bearings located
on each side of the front shaft. They are spherical, roller bear
ings with a tapered inside diameter capable of a maximum of 3°"mis
alignment~The bearings require adapter sleeves and are SKF from
service catalogue No. 450. The bearing housings. are DIN UANASA
6700-E-006 type and are press fitted on to the bearings. Below
12
the front shaft is located a T-shaped frame which joins the bearing
housings to the essentially horizontal, loading lever arm. The
specifications of the lever are given in Figure 5.
Instrumentation
Strain gages are used on the tool-holder to monitor the bend
ing and torque loads. They are not located in the specimen groove,
but rather on the tool-holder ~irectly behind the collets as shown
in ~igure 6. The reason for this is tw6~fold:
1. It is extremely difficult to mount strain gages in
the limited space of the specimen groove.
·2. Since the specimens are not r~usable, the gages are
not reusable also.
The positioning and electricaicircuitry for the four-gage bending
bridge is given in ~igure 7. The strain gage bridge arrangement
for torque is shown in Figure 8. Torque is measured by the method
shown in Figure 9. These are double-gages, 90 0 apart and all in
one piece. They are so mounted on the tool-holder surface that the
~w~ gages make 45 0 with the tool-holder axis of rotation. Table 1
contains the specifications of all the gages, as well as of the
other electrical components. The slip-ring assembly is located
adjacent to the strain gages, as shown in Figure 6. The slip-ri.ng
and brushes used are Breeze AJ-8005-A8 type. The slip-rings are
counterbalanced with an aluminum collar of equal weight and nearly
equal dimensions located on the other tool-holder, as shown in Fig
ure 6. The amplifiers, galvanometers and recorder are matched
units consisting of a Honeywell Model 119 carrier amplifier,
Ml650 galvanometers and a Model 906 C-l recorder (Visicorder). This
is the equipment used to amplify and record the output.from the
bendi.ng and torque. gages.
\
13
.r·,e' 'eIV.··· CALIBRATION OF RESEARCH MACHINES....,.,.---..- "'-~~ - _.~ _..- ~.~. . '-.-. -
.. .-V."2:~:·';:~-":::': ':.':.. :' __ '::.>:.. ~',:, ."-',.- :...- .... -~-.
A. Ca.lib~ation Requirements.: ..
I~ wa~cd~si~ed tha~~hebending stress and shear stress in the
specimen ~oove.be .accurately known for each specimen. These two
. stresses cannot be monitored directly because it is not possible to
locate strain gages. directly in the specimen groove since each fatigue
failure would destr.9y.the gages.< This would necessitate replacement
of the gages afte~.each. ,test .r.un. Therefore, the shear and bending
stresses must be determined indirectly through the use of strain
gages J9cate<;l_o!1 :the to.o.lholder shaft adjacent to the specimen.
The compl~te calibration procedure takes into account the fol
lowing: .c;:o,mplis:atJon§l..:.ee .....:::. c.e :::':...
7:':";> .1. Since the strain gages are located on the toolholder
rather than in the specimen groove, the groove stress must
be calibrated against the strain gage output.
2. The specimen contains a groove, which introduces an addi
tional unknown, the stress concentration factor. This
value can either be taken from published data or determined
through additional calibration ...
3. The torque and bending. gages may be damaged during instal
lation and require calibration against a standard.
4. The torque and bending gages may be slightly misaligned
during installation, and there may be interference or .
interaction between the torque and bending outputs.
5. There may be an axial force present in the specimen due
to the geometry of the couplings and the loading frame.
This axial force could be a function of the torque or the
bending load, or both.
The above suggests that the following calibrations be performed:
1. Calibration of the bending gages in terms of bending stress
versus visicorder output~
I
2. Calibration of the bending'stress in the specimen groove
versus visicorder output.
3. Torq~e interaction into the bending bridge.
4. Toolholder strain gage bending output'versus specimen
strain gage bending output.
5. Calibration of the torque gages in terms of shear stress
versus visicorder output.
6. Bending interaction into the torque bridge.
7. Axial interaction into bending of the specimen groove
gages.
8. Measurement of any axial force from torque and bending
moment.
These calibrations are explained in detail on the following~~
pages.
/
14
15
B. Bending Calibration
1. Visicorder output from bending strain gages on toolholder
versus true stress in toolholder
Description of Test
This involved the calibration of the visicorder output in bend
ing against the true bending stress. The true bending stress was
obtained by introducing a known bending moment at the toolholder
strain gages. 'In order that the bending moment at the gages be accu
rately known, it was desirable that the test set-up be as simple as
possible in order that the error introduced was small when calculating
the bending moment, and subsequently, the bending stress.
Test Set-up
The laboratory set-up is shown in Figure 10. The slipring side
of the toolholder was removed from one of the fatigue machines. The
toolholder was locked down, cantilever fashion, at the coupling end
of the shaft. This was accomplished by gripping the toolholder shaft
with a torque clamp-screw device between the loading bearing housing
and the coupling. The screw end of the torque device was then locked
in a vise on a laboratory table, thus securing the toolholder in a
horizontal position. Next a small notch was machined near the end
of a test specimen, and the specimen was installed in the collet of
the toolholder. The purpose of the notch was to hold the wire which
supports the loading pan in position. Therefore the specimen was
positioned so that the notch was on top.
Electrical connection between the toolholder bridge and the
amplifier was made with the us~ of thin uninsulated wire which was
wedged between the silver plates and the di-electric on the appro
priate arms of the slipring assembly•. The opposite end of the wires
were soldered to the correct leads of a bending bridge amplifier
cable, after having been removed from the brush terminals of one of
the fatigue machines.
16
The toolholder was removed from the clamp-screw device and placed
in a vertical position, resting on the coupling. This insured that
there was no bending on the strain gages while the electrical equip~
ment was zeroed and balanced. The visicorder was zeroed and the bend
ing bridge balanced according to the standard laboratory procedure.
A five-hundred-thousand ohm calibration resistance was used for
twenty-five visicorder calibration divisions. These calibration
divisions were established left and right of zero bending in accor
dance with standard procedures.
Next the toolholder shaft was again clamped in the cantilever
po~ition~ this time making sure that the bending gages were located
directly on the top and bottom of the toolholder. This was accom
~, plished in two ways:
1. A visual check to see that the gages were in the proper
~sll~ns. ,
2. With the toolholder gripped loosely and free to rotate,
the shaft was moved slightly to see where the visicorder
bending output peaks and then clamped in that position.
It was necessary to have the gages aligned in this manner because the
bending load was to be applied vertically, and this was the only
position in which the bending ga~es will record full output. Finally,
a stout wire was hung in the specimen notch and was attached to a
loading pan.
Test Procedure
Weight was added to the loading pan in ten-pound increments until
sixty pounds was reached. Then weight was removed in ten-pound incre
ments until zero pan weight was again reached. The static strain
gage output was monitored at each bending level. The visicorder out
put was carefully watched to see that the visicorder output returned
to the same ievel each time zero pan weight was reached. This insured
that there was no electrical drift occurring in the amplifier. Since
the lever arm distance for application of the bending moment to the
o
17
toolho:lde:r~ gage~ was. abo~t seven inches and the maximum pan v7eight
. wp§~~~~y pounds, the. bending moment only reached about 420 inch. -. _ ..
P9.un<is.!, This was .<1.. good deal below the operating bending moment
al1<i-1:herefore the amplifier strain gage system was extremely sensitive
in this" low range ,of ope:ration. Because of the great sensitivity it
wa_s_n~_cessary to alter the test plan for some runs because of bending
z~r~ drift and o~her small problems. _The sequence of the pan weights
W?_s:~I1ot Jmporta~t.a.s..l()ng ~s_ at least twelve data points were taken.
Afte:r:'.chapging t!l~ pan weight,. a wait of up to ten minutes was some
t.:im.~I?~n~c.essarY.. !:lE~;f9r_e.. _taki~g the visicorder run in order for the
electrical system to reach equilibrium.
This test procedure was ,applied to the toolholder arms of all
three fatigue machines •. - _
It was important that the lever arm distance, the distance between
the loading wire and the geometric center of the toolholder gages,
be recorded before the test system was torn down.
The D~ta Reduction
The standard procedure for reducing data of the kind presented<- - •• - ~
in Table 2 is~? pl~ttoolholder output versus pan weight and fit
a straight line to the poi~ts; then convert the panweight axis to
true bending str~~'; -~t the toolholder and the output axis to apparent, . - I _ - -
bending stress. However, this data was reduced using an analytical
incremental method. This method determines the average increase in
visicorder output per unit increase in pan weight. This is the
slope of the above-mentioned curve. Since strain gage outputs are
linear and are zero for zero_load, it was not necessary to calculate
the bending moment produced by the dead weight of the toolholder, thus
eliminating a major source of error.
Looking at the data, it can be seen .that the pan weights were
not taken in ten-pound increments as previously stated. When the
w~ights were calibrated against a standard, they were all found to
be 10.27 pounds. These same weights were used throughout the entire
calibration program.
18
First, the data were ~etabulated in incremental form as shown
in Table 3. Note that the size of the strain increments is roughly
constant, not a function of panweight; and therefore, the assumption
that the data is linear is a good one.
The arithmetic average of the data was determined for each
machine. These averages have the units visicorder divisions per
10.27 pounds: They had to be converted to true stress per visicorder
output stress, a useful calibration parameter, which was the slope
of the desired curve. The toolholder output can be converted to in
cremental output bending stress by
60 =o
E RgageN G R 1a ca
liN •Vl.S
Ncal(1)
output stresses for each machine were: .
the incremental visi~order output,
visicorder calibration divisions.
/
incremental output bending stress,6Young's Modulus for SAE 4940 steel =30 x 10 psi,
resistance of the bending gage = 190n,
number of active arms in the strain gage bridge = 4,
gage factor =3.23,
= the calibration resistance = 500 kn,Rcal
6N. =Vl.S
Nca1 =resulting
where 60 =0
E =
R =gageN =a·
G =
The
60 = 135.9 psi,0
1
60 =129.9 psi,02
60 = 135.6 psi.0
3
where the subscripts indicate the machine numbers. These values
were the apparent increase in bending strain when a 10.27 pound weight
was loaded on th~ pan. It remained to calculate the actual bending
19
stress increase for 10.27 pounds of pan weight. The equation was:
L~WC=I
where
60'T = incremental actual bending stress,
L = lever arm distance,
6W = incremental pan weight .- 10 •2 7 pounds ,
C = radius of the toolholder =1.0 inch,
I = moment of inertia of the toolholder cross-section =0.74 in4 •
·The lever arm distances for each of the three machines were:
/
Ll = 8.745 in,
L . = 8.199 in,2
L3 = 8.237 in.
Performing the calculations, the true stresses for each machine per
10.27 pounds of pan weight were:
60'T = 140.0 psi,1
60'T = 131.5 psi,. 2
6aT = 132.2 psi.. 3
Forming the ra:tio,· ~TH= ~OUTPUT (apparent)· STRESS
~TRUE STRESS
= 0.967,
.c::: 2. .~ "~-: ~'.' -::-' . ~ . ~~~:c.: ~ .', -:- :-:. ", ~ .~.~
y,ir'.:: ~£~:TH2 :=~:~;,:~?2_,
.~:S:':::,~;~3c.~: ;:::~;~9Z_~~:'" :'~-
20
D~i~gThese~pa~~~~t~~~~~iil~~o;beused directly in the calibration
~r9ge~~~ pr~~ented in Section IV-E but have a variety of uses in
the daily operation'of the fatigue machine. This parameter can be
~~ed.to determine: true stress at the toolholder gages and is extremely
\l§.~ftll to have on hand •.' ... ''--
p:<>j-''-2 .2·::·::Vi§idbrd~r' 'ou-tp{l.f; ':' fu.Om bending strain gages in specimen
.'-C,:>:iE~.,.:::-"~ -gro;~~ -&~f~u~ =.fr~e::~tress in specimen groove
l)'escr{pti6~'6f Test
,;:=-::c Ina 'previous section, it was pointed out that the fatigue test.__~-.--:-_._.- .._...
specimen does not 'ordinarily contain strain gages in the groove.
But fo~ calibration purposes, two test specimens with bending gages
in~he grooves 'were prepared. Before these specimens can be used
i~ the calibration of the fatigue machines, their bridge outputs must
i'irst J;>e compared with the analytically determined true bending
stress in the specimen groove.
"~' .:: Since this calibration is identical in nature to the toolholder
strain gage calibrationpresenteo in the previous section, the same
test set-up was used. Indeed, it was possible to run the two tests
simultimeously.
Test set-up
Instead of using an ordinary test specimen in the bending canti
lever test set-up, the specimen with the gages in the groove was
used. It was locked in the collet with the bending strain gages
exactly on the top and bottom of the specimen. Therefore, the gages
on-the toolholder and ~n the groove lined up. Once again, a notch
was machined at the extreme end of the test specimen in order to guide
the loading wire and maintain a constant lever arm.
21
A second amplifier cable was removed from the brush ,terminals
of a fatigue machine and positioned near the cantilever system. ThinI ,
wire leads were connected from the strain gage terminals to the cable
in the proper arrangement for a two-gage bridge. The amplifier is
designed to accommodate a two-gage bridge as well as a four~gage
bridge. All connections were soldered and insulated.
Test Procedure
The test procedure was identical to that in Section IV-B-l;
however, instead of monitoring just the toolholder output, the groove
output was also recorded. The calibration resistance used for the
groove amplifier channel was 30 kn with the visicorder set at 25.0
,divisions. ,'Upon completion of the test, both lever arm distances
were I'ecorded.
Data Reduction
A glance at the data in Table 4 shows that the groove Otitputs
for Machines #1 and #3 are not presented in form of visicorder divi-/
sions. The reason for this is that at the time of the tests, some
difficulty was being experienced in balancing_the specimen groove
'bending bridge with the Honeywell amplifier. Therefore, a static
strain indicator was briefly substituted for the amplifier and visi
corder. The strain indicator aliows for the setting of the gage factor
and bridge size and then gives strain directly in microinches per
inch.
The data from Machine #2 was reduced by the incremental method
which was presented in Section IV-B-l. The change in visicorder out
put for each 10.27 pound pan weight increment was determined. The
changes in output were then averaged. The results are given in
Table 5.
The incremental output stress in the groove was then calculated
as follows:
E RgageN G R 1a ca
6N •v~s
Ncal(1)
strain gage indicator outputs.
Then the incremental output
The results are given in Table
22where
E = 30 x 106 psi,
R l20n,gage =N = 2,a
G = 2.08 ,R = 30 kn,cal
N = 2.5.0 divisions,cal
AN. = 2.358 divisions,v~s
.1 /).0 = 2718 psi.O2
'-~
For Machines #1 and #3, the static
were averaged for each panweight level.
strains were determined and averaged.
6. The result is an overall average of the increase in strain in the
groove for a 10.27 pound increase in pan ~eight. This average is then
converted to incremental output stress.
2560 psi
2589 psi
The next step was to determine the true bending stress in the
groove analytically using
Me°GR= I (2)
But the use of this equation is deferred until the next section,
(IV-B-2.l),because of the complication which is introdu~ed by the
presence of stress concentration in the ·groove. This stress
23
the pverage stress congr'oove
'Kt~ the stress concentration factor
KBGR , the visicorder output stress ver
Since the procedure in Section IV-B-
Experimental determination ofcentration,factor in specimen
two unknowns:There are
concentration factor, not yet determined so far, must be included in
Equation 2.
,2.1
for the specimen groove, and
sus true stress curve slope.
2.2 was the only experimental test, it was necessary to somehow
extract both of these unknowns from this single test. A rigorous
determination of Kt and KBGR is not theoretically possible, but it
wa's~:fE!lt-: that 'even aies's pr~cise development is more desirable than
:resor'ting'to:': :'tabular valuesof theoretical stress concentration
factors .'~ ....' .. .;, 2::':: The: 'methocf us ed for det~rmining K
tand KBGR was the following:
Fo~' each machine~ the inc~em~ntal output bending stress in the groove
wasc.s~trequai to the incremental true stress·. i,
- .._'".'~'--...... "';",':" ....... -- '"... _ ....-"... ArT - ArT,-.~ ._ ,,'';... Llv. - LlV
T·0
or,
60 = K 6WLCo t I
Solving for Kt
"--.- ..... '-"'"_ ...'---
(3)
The solution of Equation 3 for each machine gives
24
Kt" = 1. 240l'
. K . = 1.160t
2,
iI
III
.: ....- ~..-- 1.445
'Since' the' same specimen was used for each of the three tests, the
~tress-concentration'factor must be the same. Its value was taken
;i:~: be'· :the-~rithmeti:cC. :~ve~age of the three values. Therefore, the
~ffe;t'{ve7~t~ti~"str~ss:'concentration factor for all specimens is
Y.2S" . This' is' astati:dstress concentration factor because the"'.'~.- . _. -
,c~olibra:t:ron 'procedure'used:w~s static in nature., -'
:~T":~: ~~~-xt ~ itw~~ n~c~s;ary to allow, for' variation from machine tot"~:; _~ _ .; . :_.-.:- - ..~_._. .~_ ~- _ .- - -.- _.. .
mach1ne. Th1s 1S reflected in the value of KBGR for each machine.
_F~rs'~~' 60i'wascalculafed using the newly determined value of Kt
, or:o..:C- ..:.. __'." - - ..... '.: p ~_ _ ~_ .... ~ __.: •• __ _.
.......;:-.:.:. -
60 = K 6WLC::0',0. "T' ave I._ .. - - ' .- :.---
~~~en'~GR.was calco.iated using the experimental results from Section'-iV::"B:";'2~"-"" -.'- ,,:
--'-"~ .~ -- - ~ ,;.~.:: .;.::.
. -"""'-- . ~ -". -,.-
.- -- '. - - .",~ .__ . -,,"'
=6butput Stress6True Stress
\.
-=.The ~esults for,e.ach machine were
.::.: - ._~. ~.::...::-
_. ~-.., . "- .. - :-,-' '-" '-" _. . .:: -
25
0.902
.... ,
= 0.972
~GR will be used in the data reduction technique of Section IV-E.
3. Torque Interaction into Bending Bridge Output
Description of Test
For the bending strain gages on the toolholder to perform satis
factorily, they must be in the proper position on the shaft and must
have the correct orientation. No matter how much care is taken, due
to human error during installation, the ~ages will always be slightly,,'
out of position. The effect of such misalignment is interaction of
th'e'torque load with the bending bridge, o~tput . The extent of this" -
int'eraction and its direction must be determined. .The procedure used
was to maintain a constant bending load and vary the applied torque.
A set-up was used, whereby the change in the bending bridge output
was only due to torque interaction.
Test Set-Up
-- ,Because of its simplicity ,'a cantilever-type set-up was again
used. A torque arm was needed to apply torque to·the toolholder.
This device was a standard test specimen with a steel bar welded
perpendicular to the specimen at the groove. The bar was.approxi
mately thirty inches long and has a small hole near the far end
through which a wire was strung to support a loading pan, as shown in
Figure 11. The toolholder was cantilevered in exactly the same way
as in previously described calibrations. Once again, the toolholder
ben~ing gages should be directly on the top and bottom of the shaft
for full bridge output •
. ·In this test, it was necessary to monitor bending and torque
toolliolder outputs; therefore, two amplifier cables were used and
26
iwere connected to the slip-ring 'by fine wires in the manner described
in the earlier calibrations.
Next the specimen with the torque arm was placed in the tool
holder collet. A key was placed in the keyway on the specimen so
that the collet could support the torque without the specimen slip
ping. The specimen was then tightened in the collet with the loading
pan, and the torque arm was placed in a position slightly above the
horizontal. The reason for this was that the torque arm deformed
elastically during loading .. Using a level, the torque arm was posi
tioned so that it was about one half a ~egree above the horizontal
in the no-load configuration and one half a degree below, the horizon
tal when the maximum load was applied. It should be noted that the
error introduced by the change in lever arm distance due to elastic
deformation of the torque armis a function of the cosine of the
change of the angle and is negligible. .--
Once the torque arm was ,positioned with respect to the tool~
holder, the entire assembly was removed from the screw device and
placed with the toolholder in the vertical position. Then the torque
, and bending bridges were balanced; the visicorder outputs zeroed, and
the calibra~ion r~sistances (five hundred thousand ohms for bending
and three hundred four thousand ohms for torque) and the calibration
divisions (twenty-five divisions for bending and forty-five divisions
for torque) were set on the visicorder. Then the assembly was
returned to the test configuration. Next a bending load pan was sus~
pended directly below the specimen groove. Also a torque load pan
was hung from the end,of the torque arm. A plane connecting the two
pans must be perpendicular to the l~ngitudinal axis of the toolholder
and specimen. This was checked through plum lines and squares. Four
ten-pound weights were added to the bending pan and the test set-up
was ready for torque interaction calibration •.
Test Procedure
Weight was removed in ten-pound increments from the bending pan
and added to the torque pan. Both the.bending and the torque channels
27
were monitored. When the weights were changed, there was a tendency
for the torque arm to vibrate. This was dampened with the use of
the hand. It was necessary at times to' wait as long as ten minutes
between data points, depending on how much drift there was in
the amplifier and how l?ng the amplifier took to reach equilibrium.
When all the weight was in"the torque pan, the weights were removed,
ten pounds at a time, and pla ed in the bending pan. When all theweights
were back in the bending pan, the two visicorder outputs for zero
torque pan weight were compared. If they were identical, then it was
conCluded that no amplifier drift occurred during the data taking
period and the data was~ good. If they did not compare favorably,
then the data was scrapped, and the test was re-run. At least two
good sets of up-and-down runs are needed for calibration purposes.
, Data ·Reduction
The calibration data are presented in Table 7. After the test
for ~achine #3 was ru?, the visicorder output was closely inspected.
It was found that there was no measurable change in the bending
bridge output with torque load. Therefore, it was not necessary to
_~arry theyeduction any further; it was immediately concluded that
there is no detectibletorque interaction into bending for Machine
#3.
\
(10.0, 11.912)
(40.0, 10.425)
The data from Machines #1 and #2 were reduced graphically.
First, the torque bridge output in divisions was plotted versus the
bending bridge output in divisions, as shown in Figures 12 and 13;
then a straight line was fitted to the points~ Next the slope of
the curve was determined by selecting two points for the curve, and
the rates of change were determined. For Machine #2, there are two
sets of data and two slopes. The average slope was· used. From
Figure 12:
Machine #1
Point A
Point B,
D
.. I
~N . = 30 div- ~·-v.J.s .. torque
28
~N .= -1.487 divvJ.sb d'en J..ng
From Figure i3:
Machine #2
Point A
Point B
...~ ...~: -
(36~25~ 18.70)
(5.00, 17.72)
~N . =31.25 divvJ.storque
~N . = 0.98 div",.J.Sbending
Next the visicorder outputs in divisions were converted to strain-
using Equation 1. The conversions to strain gave:
M:.· = 1. 75'1l in/in. B1
·M:. = -31. 9 II in/inT .. 1
~£ = 1..15 II in/in. B2
• .
,_ ....' ....~£T . - 33.5 II in/in
2
29
Conversion to stress instead of strain could have been used also
:~he~e_the subscript denotes torque or bending for Machines #1 or #2.
For torque, the parameter~ are:
....-" - . - ~
.:.;. _.:.. ~:.:._--
R = 120 11..." -. :. . ;: . __ . ~gage
:- c.:" ':":. :::.::-:::.:":::-::_;.:. --.::~ .::.:-.::.:=.-- ~N = 4a... ,.....,... ..:-; ._::,"~-.,; -'::.:' -=::.2. ~~_~._, ... __ ~.,...,
.- ~- _._- ~.~.'_0-'_' _,..:..:- ,. __
'~::.:-.G ::: 2.06
R =304 kr2cal
/.
The slopes in terms of strain would be:.- ...- . ..-~. ~ ......_.---_.
Sl= -;i~~ ::: -0.0548 =
S =1.15 - 0.0343 =2 33.5-
Since stress is equal to the strain multiplied by Young's ModulUS,
the slope of the bending stress versus shear stress. curve is identi
cal to_thestrain·curve slope. This slope is given the designation
Kr/B and will be used in the calibration procedure described later.
Since there must be no interaction into bending for zero torque
load, a plot of bending stress interaction versus shear stress could
be constructed by drawing a line through the origin with slope Kr/B
and label~ng the axes appropriately.
30.
. -_.~ -- ~ - ~ ." ~_ ...
N'ote' :that the interaction for Machine #1 is negative, the inter-
~~~'~~~[on for Machi~e #2'is positive, and the interaction for Machine #3
is zero.
4. Relationship between toolholder bending stress and specimengroove bending stress
All previous tests involved the calibration of only the bending
strain gage bridge outside of the fatigue machine. In this test, the
. bending bridge will be calibrated in the fatigue machine.
'~-:-~~,-,~:,:ror:anapproximateanalysis, it can be assumed that the bending
:': -moment along 'the toolhblder shaft between the two loading bearings is
cc__ , :constant. 'This 'would be true if the toolholder was weightless. Then·
~~he-relationship between toolholder bending moment and specimen groove
bending mom.ent·· i~~ ,-
- -: -'. • ....... '.:~ -- _.; •. '.- e-' M'-· .'.'; ~.. . = M •toolholder grooye/
. .... ..-_4 ~ -.:;.... __ ... "......
·-':"T:he--bending stresses are given by
,Combining these gives
. 'OTHITH_ °GRIGRCTH - KfCGR
31
Solving for the bending stress in the specimen groove, the groove
stress'isobtained as ..
_.... , . ''"' ~ - '- --.......... '~" _. ---_..- '::,~
. .
-_. :·.. :,.·.·[C., ~ [I jGR THC1 . ··:: .. K -- -- cr'GR" f CTH I GR TH
- -Howeye~, in a more sophisticated analysis, the dead weight of the
'.- --',-
toolholder must b~ considered. Two approaches are given. In Section'- -. ~
IV-B-4 .l~ an expe~ime~tal analysis is conducted. The results of". - -. _. ~. _.
these ,tests ~~e applied t~ t'h~ calibration procedure presented later.
Th~:~~o;d'-a;;r~a:~h i~'-anaiyt'ical and is presented in Section, IV-B-4. 2..
'~-.:' 4.i'T~-01ho'lder 'st~~in g:lge output versus specimen strain... c, . -= '. .c :',"gige output .,.
Description of Test
-. 'This test involves the caiibration of the bending bridge of the
toolholder against the bending bridge of the: specimen groove.
Test Set-Up
For this calibration to have any usefulness, the set-up must be
the normal running mode of the fatigue machines. Therefore, a speci
men with bending gages in the groove was installed in the toolholder
collets according to the actual test procedure which is used in run
ning specimens. For the full procedure, the laboratory checklist
given in Appendix C should be consulted. The leads from the specimen
were soldered directly to an amplifier; leaving some play in the
leads so that the shaft can be rotated by hand a few times without
causing the leads to wrap tightly around the specimen. The normal
brush and sl~p-ring arrangement was used for monitoring the toolholder
output.
The strain gage bri.dges were zeroed and balanced according' to the
checkl~st procedure. The toolholder shaft was rotated until the spe
. cimen groove gages were along the neutral axis, when balancing the
32
bridge. The toolholder 8,ageswill then be r.andomly oriented, Hhich
is perfectly all right since the l~ading frame: is blocked up (see
checklist procedure). The calibration resistances were 190 kQ for
the toolholder and 11 kQfor'the groove bridges, while theyisi-'
corder calibration divisions were both 25 divisions.
Test Procedure
A lIquasi-staticll test was run by rotating the fatigue machine
shaft by hand rather than being turned full speed or left at rest,
:while~readings were taken. ,
c.,,:~"·:n~leights were placed in the loading pan in two-and-one-half
cpound increments until fifteen pounds of total weight was reached.
Then the weights were removed from the loading pan at the same rate
until the pan weight was zero. At each level, the toolholder shaft
was rotated a few"times by hand while the visicorder recorded the- - - +.
-outputs"froin' b'~th b'~hding'bridges. It was made sure that both out-
'puts snowed an'upper and'lo~er peak so that the total bending width
;'>coui~ be' dete~i~ed. ' At least twelve data points were taken.
, ",', "The- test ~as 'rep~ated for the other two fatigue machines." . ..;.......:. ....
Data Reduction
", __,,:.:.!?e calibration data are given in Table 8. The graphical reduc
_:tion technique ,of the data was used . For each set of data, a plot of~ ".- , ~ - -- . . ,
, toolholder output in divisions versus groove output in divisions was
, made" as shown in Figures 14, 15, and 16. Straight lines were drawn
,to fit the data. Two points were selected on each line and the
6N . and 6N . were determined. Next the outpu,tsVJ.s ' , vJ.stoolholder groove
-.. - - _.
'were converted from divisions to strain by using Equation 1 with
"Young's Modulus removed. The results are given in Table 9. Lastly,
the slopes of the toolholder bending strain versus groove bending
" 'strain plots were determined and given the designation KGR- TH , with,
the following results:
·"
34
~~~~r~ng.~~u~ingscanbe expressed as a function of the pan Height if
.(..:~I1 ~nalys~s. 0(. the l?~dingframe is undertaken.
_:.. ;''':::. -.To c()J!lpl~t~ the mathematical model) the operation of the cou
~.:.p~i!1gs. a~.~~tl1~r end of the toolholder had to be fully understood.
-It was.believed .that ~ .s·ingle-~alued resisting moment) reaction force
'~~J?~ pi'{0"1:: . p,?int could be associated with the coupling mechanism. An
.:.~xperimen1:.wasproposed .,?-nd initiated to determine these three unknm-Tn
.. p~;,ame"1::~rs. .. f.: .. specil1len was placed in the machine and the machine was
:s~"1::_~p.as ~f_~t ':las about to be run •. However) the chuck on one side
.::w.~s: l].~l~.only at 011~.·e.nd., .' Then the opposite side of the machine was
~.s.':lpported in: ~_ ~evel position so that the linkage would exert an equal. .
.. force on each. section of the shaft when the machine was loaded. Next)
;~::1?ala!1c~.systemwas.rigged above the machine) as shown in Figure 18.
Care: had to be taken ..toinsure that the wire connected to the specimen
-was vertical at all times. .' Also) the balance bar had to be horizontal •
.~ bubble-type level was attached to the t.op of the chuck with a rubber
band.' This was u~ed t~ tell when the shaft was in a level position.
Then a pointer was attached to the chuck. A scale was then connected
to the. opposite safety bridge and the pointer adjusted so that it
would read out· increments on the gage. The graduations on the scale
stood for no physic~l quantitie~; they were only for reference.
Data Taking Procedure
First the machine was loaded in the lower weight pan. Five data
points were .taken.starting at zero loading and increasing in five
pound increments to twenty pounds. Each load was carefUlly centered in
the loading pan so .thatthe load was balanced at all times. Then the
balance pan was loaded until the force in the wire was great enough to
lift the shaft to approximately a level position as indicated by the
bubble level. Then the weight in the balance pan was carefullycen
tered and the positions of the balance pins adjusted so that all forces
acted at their measured distances from the fulcrum. At this time)
'us~ngthe level) a zero mark (level shaft position) was recorded on the
35
scale. Before and after each data point was taken, the zero mark
was-checked to see that {thad not moved. Now, with the observer in .
a seated position so'thathe could line up his eye with the top of
the backshaft and~the pointer'to maintain the same parallax throughout
the experiment,-the system'was' displaced from the equilibrium posi
tion .and allowed-to' return. ': Small weights were added or subtracted
on the balance pan to reach an equilibrium point when the shaft was
horizontal. Since the pivot point in the coupling was stiff (large
resistihg in6~ent),-there'was little sensitivity to weight added to
tHe' 'balance 'pan. :Thus it -was" found that sensitivity could be
iti~reased by~foliowing the procedure just mentioned, that is, displace
th~'~' shaft- up -~nd down'c from the' equilibrium position and allow is to
return freely. An example is now given to help clarify the procedure.
F,igure 19 shows the scale graduations. The dotted line, B, is
the position the pointer would indicate when the shaft is level. If
the shaft was displaced upward and allowed to return to the equili
brium position, it would stop, say, at line A. Now if the shaft was
pisplaced downward, it would come to an equilibrium position at point
C, or two full graduations below A. However, lines A and C are equi
distant above and below the level point, line B, and therefore the
observer considered this set of ,circumstances to be the sought after
equilibrium position and the weight in the balance pan at this time
was considered to be the weight necessary to bring the shaft to a
horizontal position for a given weight in the loading pan. The prob-'
lem was further complicated by the fact that the pointer would not
always return to line A when tha shaft was displaced upward. It
would be very close to line A, and the average of the displacements
upward would give line A. The same is true for line C. This would
indicate that a great deal of time and trials were necessary to pro
cure each data point. This was, in fact, the case.
When the data from the preceding experiment was reduced, it was
found that the lever arm distance calculated was 'physically impossible.
36
Since it was impossible to increase the accuracy or sensitivity of
the experiment;. it was concluded that:
.:1., The.ope:ration of the coupling was erratic and no valid
=---:..:. ._~_:._ ..__~I1!.athematical mode of its operation could be determined.
". 2 ... Since the coupling operation vias indeterminate, a full
... -- - . mathematical model could not be determined and some of
'.'_. ~"" '--'- ~..:._~~_the. _.object~ve~ of this study would have to be compro-
--:"-- .. ---,' '..-::._-::; .~ .' .. 'mised o. :
-~_::::.- The results of this .approach indicated that the exact resisting
=momentat _.the. coupli.ng, . the lever arm, and the reaction forces could
-:-i\ot--be determined with the d~sired accuracy. This conclusion lead to
_.~ ~he pursuit of the different calibration procedure presented in this
report ° _. _.. . _.- '. ~ - --=---.. ~:.:' - -"
- - - '-~ - P':". ._". '" "_ ~ _: •
_/
..~ ...:_.~ -:- "_.-
. \,
. C. .Torque Calibration
L Torque load versus visicorder output,
Description~of T~st-- -
As was the case in bending, it was necessary to calibrate the
torque bridge ~n the~o~l-~?lder against a known torque loading.
Wi th this c~·iibr~t·i~~;-~~y'visicorder output of the torque channel
can immediately be converted to shear stress in the tool-holder.
Since the applied .:.1;0.r'q~~ i,s..constant along the tool-holder shaft,
shear stress in the tool-holder can be converted to shear stress in
the specimen groove without further calibration.=- -~. -
37
Test Set-Up
The test set-up was identical to the system used in Section IV
B-3 to determin.e the. torque interaction into bending. In fact, if
th.e. torque pan weights for each visicorder run are recorded at the. /
time the torque interaction into bending calibration is run, then
the same data can be used for both tests. This procedure was fol
lowed for all three machines.
Data Reduction
The calibration data are given in Table 10. The pan weight
versus torque output plots are given in Figures 20,. 21, and 22. A
straight line was fitted to each set of data and two points are
selected off each curve. The delta values were determined; the pan
weight was converted to true .torqueusing:
liT = lIWL
where
liTtruelITC=
J
38'6W = increment of pan weight,
L = torque leverqrm distance,
6T = increment of torque,
.C =radius of toolholder at.the torque bridge.,
'.
J = polar moment of inertia at the torque bridge.
The torque output was converted to apparent shear stress using
reduction Equation 1. Lastly, the slopes were determined and given/
the designation Kr' The results appear in Table 11.
2•. Bending Interaction into the Torque Bridge Output
Description of Test
If the torque strain gages in the toolholder torque bridge were
slightly misaligned 'during installation, then there will be an
interaction between the torque bridge output and the bending load.
This interaction must be determined experimentally for each torque
bridge. The method used in Section IV-B-3 to determine the effect
of torque on the bending bridge output was used again. The torque
load was varied, and the change in the torque bridge output recorded.
Test Set-Up
If the torque load applied to the torque bridge is allowed to
be zero, then the set-up, Figure 10, used in Sections IV-B-l and
IV-B-2 can be used. Thus, three different calibration· tests can be
run with the same set-up, greatly reducing the time necessary to
complete the calibration of the fatigue machines.
39
In this -wst the torque bridge and bending bridge outputs
needed to be ~qnit'?.!'.~<:l; _Th~ cantilever system was connected elec
trically as de~cribed'in Section IV-A.
Test ProceduY'.£. . .
Wei~h.~ !t<;;~_.pJ,§_~~(t iDthe loading pan in ten-pound increments,
as in Figure 1.0.-· 'A visicorder run was taken after each weight
increase. One hundred ten pounds was the maximum pan weight used.
From twelve ~o thirty data points were run depending on the repro
ducibility of "the data. _ ta~ch time zero pan weight was reached, the--~~ .
visicorder pocition was noted and compared with the previous one.
If the differ'ence was greater than O.l.divisions, it was concluded
that ~~pl~f~~~.~i!t has occurred and the points between the zeroes
were emitted. .
Data Reduct iC..itl
The calib~ation data are given in T~ble 12. For Machines #1
and #2, the data of bending bridge output was plotted against the
torque bridge output', as shown in Figures 23 and 24. A straight
line was fitt~d to the data and the slope of the line determined by
taking incremental changes in torque and bending outputs and con-_
verting them ~ostrain. Then the values were ratioed.
For Machine #1, two points on·curve are:
point A
Point B
Using Equation·l,
(8, 0.15)
(0.35, -0.1)
6N . = 7.65v~sb d'en ~ng
.~ . =0.25v~storque
I:,,;:;..:c-. ::::"-"'-6£' ;c.' .... (7.65)(190) = 8.97 f.I in/in
c.~. -: :>;" --: :.' ~:-.-:: .:::( 2 5 •1 ) ( 4 ) ( 3. 23 ) ( 6 . 5) (106
)
," .. .:...~;.~ ~'.~':'~~-._-~.-~.
40
= 0.267 1-1' in/in
,,-.- ~ -~ - - ............- .......-...
.~ :::._-- - :- ...• '- "-." '.- .
~e:T 0.267Slope = ,~e:B = 8.97
,..' -.'-
= 0.0298 .
For Machine #2, two "p'ointson curve are:
·Point A
Point B
(40 :0,- 1. 8 )
(12.-5, 0.45)
. /
"'_ ... _. -_._---"_ ..--
. '::AN; : =1.35'Y.LS. torque
';' ~_e:B =(27.5)(190)
= 33.58 in/in. (25.0)(4)(3.23)(500)(10a)
f.I
- . -r ~ .... • ~
.... . ,::..._-' . . . .." "-"
~e:T( .135)(120)
= 1.255 in/in-(44.8)(4)(2.06)(350)(10 3 )
f.I
Slope = 1. 255533.58
= 0.0374
41
Since for zero bending', the' bending interaction into torque
_, .,__ must also be zero; the slopes are all that are needed to completely. .
define the interaction. The slope was given the designation ~/T.
:.;,'..:- ':'.:','-The Machine #3 data were reduced slightly differently. The
~~xabulation of the data iri Table 12 shows, that the datum of the da~a
:,':: :s.hifted with each subset. Therefore, each subset was given a dif-
.::~ ferent plotting symbol; and the data was plotted as shown in Figure
--:-:·25'.-·' A· -line was· fitted to each subset, apd the slopes were deter-
",,::mined. . ~-' ~ -- - ......' - - ... - . - .
, --
_..- -." ~ -~._.:..... _ ,1:: ._. ~ ,.
.":" .-. -~.~,:-.- ~
_', ~ _.-- .
,Symbol
...:- ~.- -- • q •
Slope
--0.02835
-0.0409~ .. -.
. .,.0.0591
-0.0543 /
The dot subset was thrown out because it is far out of line. This
----was--justified by the fact that this data was taken first, and equi-
,librium of the'electrical equipment may not have been established.
Th~ 'remainin~ f~~ slo~e~:were averaged. The average slope is
-0:0538 • Then the conversion to strain was made.
.:-.--- ,-.- --_:_[ J' J1 N ]R N N GR visSl = gage· Jcal a cal torque ==
ope N N GR R 5N .. cal a cal gage -2 v~sb d'en ~ng
-0.01698
The coefficient of the bending visicorder divisions is a correction
factor ~ade nece$sary bya change in the attenuation on the bending
channel of the amplifier.
42
.2.1 Effect of bending on torque bridge output - additionaltest . .
Description of Test --
From the definition of interaction, it is evident that the bend
ing interaction into torque should not be a function of the mean
torqlle ~o~.~~ ~hat is, _~~=-_ITl_~~~lignment of the torque gages does not
vary with torque load; therefore, the interaction would not vary. A
spot check was necessary to show that this reasoning is correct.\ .
The result can also be extended to the torque interaction into the
This test was only applied to Machine #2.: ..... ''"-r,. ,.
Test Set-Up' v __
For this test, a high mean torque was applied to the canti
levered toolholder.The torque arm was required to apply this
torque. The test set-up was identical to that used to determine the
effect of torque on the bending bridge output and is that of Figure
11. /
Test Procedure
Twenty pounds was placed in the torque loading pan. This
weight w~s left untouched for the remainder of the test. The
remainder of the test proceeded in the manner described for the pre
vious bending into torque interaction calibration.
Data Reduction
The data are given in Table 13 and plotted in Figure 26. The
reduction of the data was the same as that given for the previous
bending into torque interaction calibration:
From two points on fitted line:
Point A (25.D, 26.1) .
Point B (10.0, ~5;25)
6N . . = 0.85_v~storque
N . Rv~s gage
N N G Rcal a cal
fiN . = ·15.0v~s ..
bend~ng
(15.0)(190)= (25.0)(4)(3.23)(500)(103) - 17.65 pin/in
43
N • Rv~s gage
N N G Rcal a cal
= (0.85)(120)7(4:-:4-.-::8~) "7":(4~)'(-=-2-.0~6-o-)"7"(3-:"":5=-:0"'7)-;"'(1:-:0:-:3~) .. o. 788 pin / in
Slope = 0.78817.65 = 0.0447
/
. This compares favorably with the slope presented for the zero
mean torque interaction and therefore, for Machine #2, th~ average
of the two, 0.0410,is taken to be ~/T.·
44
D.:" 'Aii~l"'Effects" '''- .,
':=-'~':"~l:' :Eff~cit-~f ~iai load on bending bri.dge output
It was not known whether'the loading configuration of the
f.~:t;Jm-!~_..mSi,¢hJnes impariS an axial stress in the test specimen. If
there~is such,a:force,then it must either be eliminated or
accounted for in the determination of the stress ratio. The axial
stress could be zero, a constant, a function of bending load, a
function of' torque' load, or· a" function of both bending and torque
loads. This axial stress must be accounted for in order to com
plete~thecalibration of the .fatigue machines.
The tool-holder bending bridge was set up so that it measured
. the bending strahl."and canceled out any axial .force applied. If-" _.~ -- ."
two-of--the-arms"on" the bridge are reversed , the bridge will now
measure' the -axial ibaci","cancel out any bending load. A preliminary
step In the' measurement of-the axial stress was to determine if any
axial i~ad:inte~action into the bending bridge output was present.
If there -Is' no"'interaction, then any change in the axial stress
bridge" when a be~ding load-is applied, will be a true measure of
the axial" stress in the specimen groove." If there is interaction;
:~hen t"he -correcte<f output will 1:?e a measure of the axial force.
,:,,·;":-~':::-.The"test to be performed was not bending interaction into the
axial 'bridge, but the axial interaction into the bending bridge.
since' any interaction is due to misalignment of the bending gages,
if one form of interaction is present, then the other must also be
present.
Test Set-Up
The tool-holder was removed from the fatigue machine and., .
placed in a vertical position, resting the coupling end on a hard,. "
level surface. A specimen with gages in the groove was installed'
in the collet •. The bending bridge of the specimen was connected
to the amplifier. Then the amplifier channel was balanced .. A
"~~eight pan was next balanced on the end of the specimen and a visi-. _.._-----~-----_ .....-corder~run taken. Then weight was stacked in ten-pound increments
~on":the._p~n until sixi;ypounds was ,reached. The bending bridge out
;. P':1t,V!as.recorded after. each. addition of weight.
;'J:jeita'Reduction .~. :.:.'=.' ~-_.: .. , .. , .
>:",,:, The visicorder output showed that in all cases there was no
:measurable axial interaction into the bending bridge. From a pre-,--~-~-~-_._-_.
vious, di~cussion,it can also be concluded that there was no bend-
:.i.,ng .interactioninto the axia·l bridge' of the gaged specimen.
~2 ..o. .Measurement of the Axial Stress
.. -..:->:~~:~:::..• _,-:;::.,~2.~ 1: Heasu):'ement of the axial stress as a function of. . Pending load
<".L •• ~'-..: _0 .... ~ •• -. :---:.. .-.. ::" ~ -' ~ _." ~.: ~. •
Description of Test
Since it was established that there was no bending interaction
into the axial stress for the gaged specimen, the axial stresscouldbe--meastired"~·_."--- _._.._p~_._'---
In this first test, the axial stress was investigated as a
function of the bending load. The test was conducted only on
Mac;:hine.#l and the results were reduced before any decision was
made about extending the analys~s to the other two machines.
45
Test Set-Up
The specimen which was used in the axial interaction test was
installed in Machine #1 according to the installation checklist.
The brush assembly. which monitors the toot-holder torque output was
,lifted from the slip-rings so that they were inoperative. The long
leads from the specimen were connected to the torque terminals on
the brtlsh support so that the bridge formed would measure axial
.stress and cancel out bending. The tool-holder bending bridge and.
the s.pecimen axial bridge were zeroed and balanced ,according to the
checklist.
Test Procedure
With the long arm in use, weight was added to the loading pan
in five-pound increments. Between each increment a visicorder run
was taken while the tool-holder was turned quickly by hand. Weight
was added until twenty pounds was reached, and then the test was. .repeated.
Data Reduction
The data are given in Table 14. A plot of the data is given
in Figure 27. It shows a non-zero axial stress which is a linear
function of the bending load.
The magnitude of the axial stress was 4etermined next. For a
twenty-pound range in pan weight there was a 1.7 division range in
axial groove gage bridge output. Converting this to strain,
46
£o
N. R=~ gageN N G Rcal a cal
(1.7)(120)= =(50.0)(2)(2.09)(290,000) 3.38 pin/in
or in terms of stress
a . - E£.0 0
6 -6= 30 x io x 3.38 x 10 =101.4 psi
This is a negligible amount of axial stress. Since the pan weight
range was twenty pounds on the long arm, an axial stress of about
100 psi would be the maximum that would ever be encountered in any
test run. Therefore, the axial stress due to bend~ng load can be
omitted. The analysis was not extended to the other two machines
because the axial stress was found to be so small that a value ten
times as great would still be negligible.
· 47
,_ '2.2 Measurement of the axial stress as a function of":. ".:- -, -~ :torque load
Description of Test
'Next,' the axial force 'was measured as a function of the torque
16ad 'with the bending load held constant. This test was performed
o'rily :fo~ Machine #1. '
Test Set-Up
~~e torque brushes were placed in contact with the slip-rings
;~d_'th~,be~di'~g,_gages'-iif.fed. The axial bridge of the specimen was
~ir~dt~'th~'bending terminals on the brush mount. The specimen
~~s~~instaiied'~;cord{I1gto checklist procedur'es. The torque bridge
;~d-the ~p~cim~~--axia'l'br.i.dge~ere zeroed ,and balanced. Five
;~~~d~:'~er~'piac~d in the loading pan at the end of the lO,ng arm.:.- -'
,,,,,Test Procedure.-.~; ." ~--
~~ =~Torque load was var~e~ randomly through the Infinit-Indexer/
.. and, ,a yisicord,ep_ .run taken at each level. Fourteen levels were
~- .:...:: .--:. . .. ..
'Data Reduction'
The data is given iri'Table15. The specimen groove gage out
-put 'plotted against the torque output appears in Figure 28 and is a
~iinearly-vary{ng axial stress. The range of the stress is 3.7
divisions. The strain for this output is,
e;o
N.. ,R= 'V1S gageN N G Rcal a ' cal
(3.9 )(120)= =(49.8)(2)(2.08)(290,000) 7. 78 ~ in/in
J
J
" which is 233.4 psi in terms of stress.
This value, though larger than the one for bending load, is
:_:still, negligible. The small value of the axial stress over a large
range of torque indicated that the test did not have to be repeated
for the other two machines.
~I
48
E. Summary of Calibration Results
The calibration of the complex fatigue reliability research
machines was thus completed. Bending, torque and axial interaction
tests, calibration of' the bending, torque and specimen groove
bridges against a standard, and tool-holder bending output versus
specimen groove bending output tests were thus performed for each
machine. The relationship between the. variables in each case
proved to be linear and a slope could be associated with each cali
bration. Since for all cases, the functional relationship begins
at the origin, the slope of each curve completely defines the func
tion. These slopes have been given calibration coefficient desig
nations and appear in Table 16. The two footnotes of the table are
explained in Appendix E.
Now, it is necessary to combine the calibration results into
an orderly method of determining the true shear and bending stresses
in the specimen groove given a visicorder output, or vice versa.
Actually, the calibration procedure requires that the calibration
method be run in both directions. First, in any test run, the
desired stress ratio and bending stress level are selected. From
these, the shear stress level can be calculated and, through the
use of the calibration coeffici~nts, the number of divisions of
bending and torque on the visicorder can be determined given the
machine to be used, the calibration resistance, and the calibration
divisions. Next, the test is run and a visicorder record taken.
The record is then reduced. The four experimental parameters,
bending divisions, torque divisions, bending calibration divisions,
and torque calibration divisions, are measured. Using the calibra
tion procedure in reverse direction, the actual bending stress,
shear stress, and stress ratio are determined. When an entire
stress level is run,some statistical comments can be made concern
ing the stress levels and stress ratios achieved. This is discussed
in Section V-A.
49- ..,..
... J.• -•• ,_,
Jhecalibration procedure will now be described in detail with__ • ~. - .,:. -:...::,;. - , ..•,:. .~ _'.0 • . ~ ~_ •
the aid of the Calibration Flow Chart of Figure 29. Steps 1 and 2
on the flow chart require a selection or a stress ratio and a bend
ing stress level. The true shear stress in the specimen groove to
give this ratio is found in Step 3 from the relationship
( .
where C1 . is the true bending stress in the groove and r is. . ~~ s .
the desired stress rat;io~ Next, in Step 4~ the true bending stress
in the groove is converted to output stress in the groove using
~GR and then KGR- TH , the relationship between output bending stress
in the groove,is employed. On the torque side, Step 5 gives the
equation for converting shear stress in the groove to shear stress
in the tool-holder. This is a purely geometric relationship employ
ing no calibration parameters. However, on the bending side an
empirical relationship was nece~sary because the bending moment
along the tool-holder shaft is not constant. For torque, it is a
safe assumption ~hat the torque applied along the tool-holder shaft
is constant. The static stress concentration factor is shear,Kts '
is included in this term. A published value of Kts ' 1.22 (26),is
used here. Also, in Step 5, the true shear stress in the tool
holder is converted to outputstress. Steps 6 and 7 are the correc
tions for interaction. The shear stress value is used to correct
bending stress and vice versa. The calibration c~efficients are
KT/ B and KB/ T• Steps 8 and 9 convert the corrected output stresses
to visicorder divisions, completing the procedure.
;II
. I·I
"V RESEARCH DATA REDUCTION
51
A.. Conversion of Visicorder Outputs to Stress Levels
Once the visicorder divisions for bending and torque are deter
mined according to the methods of Section IV-E, the test runs are
mape. A visicorder record is made and kept for each specimen. When.
a stress level is complete, the shear and bending divisions, and the
shear and bending calibration divisions, are measured on each visi
corder record. These four values, along with the test number,
specimen number, the calibration resistances, the pan weight and the
. machine number,are punched on IBM cards.for computer analysis. A
computer program in Fortran IV has been prepared which gives the
actual shear stress, bending stress and stress ratio for each run,
as shown in Figure 30. It also determines. the mean and standard
deviation of the shear and bending stresses and the stress ratio for
each bending stress level assuming a normal distribution for this
data.
The program contains a number of "IF" statements which select
the proper machine number and mode for a data card·. The mode is
determined by the date on which 'the specimen was run.
MODE 1: For 26 July 1966 to 10 August 1967
MODE 2: For 11 August 1967 to 1 June 1969
MODE 3: For 1 June 1969 to next gage failure
A new mode is initiated whenever a strain gage on any machine fails,
necessitating recalibration. Thus, the computer program dif~erenti
ates between past and present machine configurations. The program
is set up So that a new configuration can be integrated into the
data reduction by adding three "IF" statements and listing the new
calibration parameters. In addition, the program can accomodate
the stress ratio infinity as well as all finite ratios.
/
Since the mean, S ,e
strength were deter-
52
B. Reduction of Endurance Strength Data for r = 00 and r = 0.90. s s
For stress ratios of infinity and 0.90,the increment used in
the staircase method (27, pp. 14-17) was one pound pan weight on
the short arm. Therefore, the results of the staircase method,the
mean and standard deviation of the bending stress for endurance, is
in terms of pan weight. It is now necessary to convert pan weight
to true stress.
Just like the finite times-to-failure runs, visicorder records
were made for the endurance limit runs. The visicorder records
were measured, the data placed on computer cards, and a computer
. run made, just as described in Section V-A'. Only' the points which
were used in the staircase method (successes or failures) were used.
The output contains the bending stress levels, torque stress levels
(r = 0.90 only) and the stress ratios (r = 0.90 only). Also thes s
/
mean and standard deviation of the data 'set as a whole is outputted;
however, these are meaningless for endurance runs, hence are dis
carded. However, ~he stress ratio is needed.
For a given pan weight, all the bending stresses for that pan
weight were averaged. This was done for each pan weight. The
result was the mean true bending stress for each pan weight.
Next, the true bending stress in the groove was plotted against
pan weight for both r = 00 in Figure· 31 and r = 0.90 in Figure 32.. ·s·s
Straight lines were then fitted.to the points.
and standard deviation, aS ' of the endurancee
mined in terms of pan weight by the staircase method (27, pp. 14-17),
Figures 31 and 32 made it possible to relate pan weight to t~ue
stress. The results are as follows:
. For r = 00s_.S =' 25.5 lbs= 61,500 psi.eaS = 1.34 . lbs = 3,500 psi
e
.~:c<::c- :For ..r:'· = 0; 90''''.'," .'., .~_:'. .~. 5.. _ ...
,_.. ,,~ ..... :S = .26.0 lbs.= 61,000 psi.. "~'" . -~·;"--e· ..- -
:::=~,~:~:,.~~ ..:;:-'_;':; :~:-...::.:,\~_:..::,_aS". ~ .2 •..~.~~_lbs ~ 3,000 psie
. !Jz;;T'G:'The:mean· and :standard deviation of the stress ratio for r s =
53
~'~'9~ 'fr~m the _dat.a reduction computer program,. given in Appendix B,
iho~'gh printed ou't, are -ri'ot intended for any use . The program is
.l~~::rs~d fo~- iinite-'life data; hence it automatically calculates
/'
C. Generation of S-N Diagrams Using Distributional Stress Levelsand Mean Cycles-to-Failure
The reduced data of Appendices A and B provide the mean and
standard deviation of each stress level. With these distributional
parameters and the corresponding mean cycles-to-failure data ( 27,
pp. 52-54 ) it was possible to construct S-N curves for r = ~ andsr s = 0.70, with the 3cr limits of the distribution of the actually
applied bending stresses also given, assuming the stresses applied
are normally distributed.
The S-N curve for r s =~ appears· in Figure 33 and for r s = 0.70
in Figure 34.
/
54 '
55
D. Generation of Theoretical S-N Diagram
It would be interesting to compare the·S-N diagram results gen
erated in this research with a theoretical S-N diagram constructed
~ccording .to the procedure recommended by Shigley (23, p. 162).
The fatigue strength of specimen,Se,at 103 cycles = 0.9SU'"
Su = 178 ksi , for unnotched· specimens
./
.'
The endurance strength is given by
For the specimens used in this research
(4)
ka = surface finish factor = 0.89 for ground finish (23, p. 167)
kb = size factor = 0.85 for D = 2.0 in. bending .
.. Kd = temperature factor· = 1, no temperature effect
(23, p. 168)
f.
.. '.~'
""':, -_c:k ~,' :=-:-'~tr-~ss concentration design factore
- . , - . - .. '
~ = 1 + q (Kt
- '1)
56
(23, p. 170)
(23, p. 170)
-q = 0.92 -.< '. '.. -':
, (23, p. 171)
,Kf
= 1 + 0.92 (1.45 - 1)
Kf = 1.41
/
k1
0.709=---, e' 1.41
kf = miscellaneous effects factor = 1
(26, p. 49 )
S t
e= theoretical endurance str~ngth = 0.5 S
,U(23, p. 162)
s =178 ksiU
Therefore
S t = 0.5 x 178 = 89 ksie
\
Using Equation 4
S- (0.89)(0.85)(1.0)(0.709)(1.0)( 89) = 47.7 ksie
,With Sand S thus determined, an S-N d~agram'was plotted ande - e
103
is given in Figure 35.
.-
57
59
VI. STRESS CONCENTRATION AND NOTCH SENSITIVITY EFFECTS
A. Stress 'Concentration
The best published value of the static stress concentration
factor for the notched specimens used in this research is 1.45 (26).
In Section IV-B-2.i, an attempt was made to isolate the effective
static stress concentration factor from ~~. The resulting value
was 1.28. Since the experimental value was arrived at by an averag
ing technique rather than by a rigid procedure of measurement, there
is room for error in this result. Likewise, the published values
leave much to be desired since they do not take into account the
test conditions. It was felt that the experimental value was a
definite improvement over the theoretical value and was used in
the calibration procedure to determine t~e bending stress in the
groove. The accuracy of the result could be improved upon by out
fitting more specimens with strain gages and repeating the cali
'bration. The same reduction technique would be used but more data
points would be had to average, thus increasing the accuracy of
determining the effective static stress concentration factor.
For the torque strain gage 'bridge a published value of the
static stress concentration was used. There is an error involved
in this also, as the exact stress concentration factor in shear is
not known at this time. It is believed that any corrections would
not be in excess of 5 percent. Nevertheless, further investigations
should be initiated to obtain a better value for the static stress
concentration factor in torsion.
60
f: ¥:.e, :~o!ch Sen13 givity Survey._
>·,,,",.In.fatigue. tests tl1e reduction in fatigue strength is found to<-... ~ ~, ••-' '--. • - _ •• _. -
-, be_.l~ss. than that predicted. by the static stress concentration factor,~.- .:. ~ ...... -- . ,- . - .. -- - .- . -
rJ\ f~4-, p. 249t· :;-,The ratio of the nominal fatigue strength of an
:-~.m!1otched specimen to that ~ of a notched specimen is defined as the........- --.- - -. -..;. - - _.
T:f~t~g;ue stress-cc>D<;:eI!tration factor, Kf . Kf has also been called the
strength reduction factor, the fatigue notch factor, and the effec
tive stress concentration factor. Therefore, Kf can be experimentally
determined by forming the ratio (23, p. 170)
= endurance limit of notch-free specimensendurance limit of notched specimens (5)
where S' is then
(22, p. 128).
The reduction in fatigue strength is explained
.the theoretical peak stress K S is lowered to KfStn, nnominal stress in the,corresponding
by the fact that
by plastic flow,
unnotched specimen
It is also useful to define the notch sensitivity factor as
(23, p. 170)
Kf
- 1
q =15t
l'
If q = 0, then there is no sensitivity'for the notch; while if
q is unity, Kf = Kt , and there would be no observed redyction in the
fatigue strength of notched specimens. This would be the case for
perfectly elastic materials (25, p. 244).
The notch sensitivity is a very useful but also elusive para
meter. As the notch radius increases, q tends toward unity as shown
in Figures 37, 38, 39,41, and 42. q also tends to unity for fine
grained, relatively homogeneous materials (23, p. 172) as shown in
Figure 40. Notch sensitivity is also a function of the size of the
part. Kf may be higher for large parts than for small parts (24,
p. 251). Steel generally has a higher q than lower grades of iron
i 1
(23~ p. 172). This can be explained by the relative grain sizes.
At low cycle life~ fatigue strength in notched speci~ens can be a
little higher than notch-free specimens (25~p. 246). Finally~q
may also be a function of the stress amplitude (25~ p. 246).
Peterson~ (26) a widely used reference on notch sensitivity~ does
not take this into account.
;/
61
~.
62c.' Proposed Notch Sensit i vity Determinat ion '
Unlike the value of the static stress concentration factor, the
notch sensitivity of the test specimen is critical. The calibration
method which was used to determine the stress concentration does not
incorporate notch sensitivity because it was a static test •
. In the calibration procedure, the notch sensitivity effect was
not included. The reasons for this are:'
1. 'An experimental notch sensitivity determination has not
as yet been carried out, .and,
2. The published data of Section VI-Bindicate that the, notch
sensitivity effect will be very small, perhaps 2 or 3 per
cent. Such minor adjustments to the stress levels can be
made at a later date if deemed appropriate.
Due to the irregularities and apparent contradictions in the
no~ch sensitivity, the most accurate method of determining the appro
priate value of q for the SAE 4340 notched steel specimens of this
research is to do it experimentally. ' Then shape, size, and material
errors would be eliminated. Kf
, 'and consequently, q, can be determined
experimentally by Equation 5.
It is proposed that unnotched specimens similar to those used
in the r = 0 studies for this research be prepared. The numbers
needed would be approximately 38. This is the amount needed with
the staircase method of endurance strength distribution determination.
Only the unnotched specimen need be run since notch endurance strength
for r =~ has already been generated. The proposed test would gives
an accurate value for the notch ~ensitivity of these specimens.
\.
63
VII CONCLUSIONS
1. The fatigue machines are extremely difficult to calibrate
dynamically. Attempts were made over a period of one year to obtain
valid dynamic calibrations but were abandoned because of large incon
sistencies in the results. Finally, the static calibration procedure
reported here was embarked upon and was successfully completed.
Since all calibrations except one are'gage calibrations, there is no
error introduced by not calibrating dynamically. The determination
of exact specimen groove stress from the toolholder gage bridge out
put does, however, require dynamic calibration 'so that the true'
fatigue stress concentration factor in bending can be determined.'I '
The determination of this factor requires extensive research and the
minor improvements achieved mayor may not be of the desired quality"
in view of the difficulties involved.
2. The calibration procedure is set up so that if a ~age fail
ure occurs, the fatigue machine may be recalibrated and put back into
operation in three days.
3. Machine #1 has the followipg calibration equations:
N NGR [. ' , '.= cal a cal (0.967)(0.0203)0 " +
E R trueGR
'. ,gage . , '
(O'.0548h t ]. ·ou TH
= NcalNaGRcall<o.0157) . '. E R (0.882) 'true
GR' ,gage' \+ (0.0298)0 t "]
ou TH
4. Machine #2 has the following calibration equations:
N N GR [ '. .= cal a cal (0.902)(0.0198)0E,Rgage .. trueGR
+ (O.0343h t ], ou TH
64
N N GR [ . ,= cal a cal (O.0157)T +E R (0.840) true
GR,
.' ,g,age(0.0410)0 't ]
ou TH
5. Machine #3 has the following,calibration equations:
N ' N GR [ " 'J'-cal a cal= E R (0.972)(0.0215)0,~age . trueGR ,
(-0.0238)0 t ]. ou TH
Note that there is no torque interaction into the bending channel for
Machine #3 and that the bending interaction into the torque channel
is n,egative.
6. The axial stress in the specimen was found to be negligible.
7. The effective static stress concentration factor of the spe
cimen groove in bending was determined to be 1.28.
8. The data reduction shows that the machines were very success
ful in maintaining and reproducing the torque and bending loads. With
other past machines difficulty has been experienced in maintaining
constant bending and torque loads.
9. From ~igures 33 and 34 and Appendices A and B, it may be seen
that the fatigue reliability research machines maintained the desired
bending stress levels with a coefficient of variation of less than
2.7% for r =w.and less than 5.6% for r = 0.70.,s s
65
'10. From ~igure 34 and Appendix B, it may be seen that the coef-
ficient of variation did not exceed 8% and for the most part was less
than 6% for the stress ratio.
L...:.\.
....-~.: ... .:~ ..... '..
-=.::~ .. -
- ~ .--:::~,:",:~.: .:-:. ~:: ..;:..;=:: 3: l(ji'.':.:=-'::_ ::..:.:. .~ .':;";".•:" '_.' .!.~",.
- -- - ~._':..! ':.".. -_:-:::-: .~' ..::---_::'...
-. _.."':",- _ •. ~_-__ .~._._•. _,•. ,. _. J
.Y',
....... ,',. -.~ ••- _.- , .••• ...to
66
VIII RECOMMENDATIONS
1. Calibrations, test runs, and data· reduction would be
greatly expedited if the amplification system was improved upon.
Presently, amplifier drift must be carefully taken into account
on each run, consequently, a relatively drift-free amplifying and
recording system, with a very short warm tip time, should be ob-
. tained.
2. A pan weight versus true stress in the groove calibration
should be conducted. Although, like the determination of KBTH '
this relationship would not be an integral part of the calibration
procedure, it has a wide variety of uses during daily operation
of the fatigue machines; e.g., in determining the pan weight
directly from the required bending and shear stresses.
3. The notch sensitivity determination proposal for bending
should be undertaken, thus insuring that this dynamic phenomenon
has been accounted for in the calibration procedure.
4. Now that experience has been gained in running the ma
chines, the staircase method for determining the endurance strengths
can be improved. Increments of bending stress rather than incre
ments of pan weight should be used because they are more reproducible.
5. Research with these machines shouid be continued and ex
panded to include other stress ratios, notch geometries, and mater
ials.
".
68.... < •• " .
REFERENCES
L "Designing a Specified Reliability Directly Into a Component,"Dimitri 'Kececioglu and David Cormier, Proceedings of the ThirdAnnual SAE-ASME-AlAA Aerospace Reliability and MaintainabilityConference, Washington, D. C., June 29-July 1,1964, pp. 546-565. '
2. "Safety ,Reliability and Structural Des ign, '.' A. M. Freudenthal,"American Society of Civil Engineers Transactions, Vol. 127, 1962,Part II, pp. 304-323. ' ,
3. '!The Analysis of Structural Safety," A. M. Freudenthal, J. M.Garrelts, and M. Shinozuka, Institute for the Study of Fatigueand Reliability, Columbia University, New York, N. Y., June1965, 129 pp.
4.
. .
5.
6.
7.
8.
,9.
tiFundamental Aspects of Mechanical Reliability from MechanicalReliability Concepts," A. A. Mittenbergs, ASME Design Engineering Conference, New York, N. Y., May 17-20,1965, pp. 17-33 •
"Sixth Progress Report to MRRC Members," A. H. Clover, A. R.Rosenfield, A. A. Putnam, R. E. Mesloh, L. M. Cassidy, R. Simon,and A. A. Mittenbergs, Battelle Memorial Ins~itute, ColumbusOhio, March 31, 1965.
"Methods for Prediction of Electro-Mechanical System Reliability,"T. L. Bush,A. P. Meyers, and D. F. Semonaitis, lIT ResearchInstitute Technology Center~ Chicago, Ill., May 1965, 173 pp.
"A Model for Time Varying and Interfering Stress-Strength Probability Density Distributions with Considerations for FailureIncidence and Property Degradation," M. J. Bratt, G. Reethof,and G. W. Weber, Proceedings of the Third Annual SAE~ASME-AIAA
Conference on Aerospace Reliability and Maintainability, Washington, D. C., June 29-July 1, 1964, pp. 566-575.
"References on Fatigue," ASTM STP 9G-9N, American Society forTesting Materials, Philadelphia, Pa., 1955-1963.
"Proceedings of the Symposium on Large Fatigue Testing Machinesand Their Results," ASTM STP No. 216, American Society for Testing Materials, Philadelphia, Pa., 1957, p. 151.
69
::;].p.. "!S,eventy~~even Year ,Index," American Society of Mechanical'.J.:pgipeers " :N " Y., 19.57.
=-;J;~l,. '''Applied Science and: Technology Index," The H. W. Wilson Co.,;,N.-. Y. ,,1958 ,to .Dec. ,1965 ••• -_ "••• '. __ __." 0. _.,. _ ,
~12.. -"Proceedings of. the Society for Experimental Stress Analysis,"is,9.ciety for Experimental Stress Analysis, Cambridge, Mass.,1943-1960.
13.
":".- .14.-
-L ; ..
15.
16.
~ .... :::.- ...-.~",--- -.~-'-
:,iExp'erimental Mechanics ," Journal of the Society for Experi'nientai 'Stress 'Analysis', taston, Pa., 1961 - Dec., 1965.
, -- - - . - -- _. - . --
:"A'~Fatig'ue 'Testing Machine for Reversed -Bending and Steadyiforque," H. H. Mabie and M. S. Gj esdahl, Proceedings of the-S9ciety for Experimental Stress Analysis, Vol. 14, No.1,',Ca~bridge,.Mass.,,1956,pp. 83-88.
:Ph9tographs 'and correspondence submitted to Dr. Dimitri B.Kececipglu, NASA NGR-03-002-044, University of Arizona, Tucson,!~rizona,Oct., 1965,by,the Falk Corporation, Milwaukee, Wis-:.qo~sin. ,
C~~~e~~o~dencewithH~ H. Mabie, Professor of Mechanical Engi;neering, Virginia Polytechnic Institute, Blacksburg', Virginia,(~ept~, ~965. . - ,; -, -' ---_.. ~. - -"-.
17.-.Correspondence'with Arthur H. Burr, Professor in charge, Depart'ment 'of Machine Design,Sibley School of Mechanical Engineering,Cornell University, Ithaca, N. Y., Oct. 1,1965.
18.
. 19.
20.
21.
Correspondence with Arthur H. Burr, Professor in charge, Department of Machine Design, Sibley School of Mechanical Engineering,Cornell University, Ithaca, N. Y., Sept., 1965 •.
"Four-Square Test Machine," DIN 100, 8-10-51, by H. H. Mabie,Department of Mechanical Engineering, Pennsylvania State College.
"Four-Square Test Machine Modified," DIN 100, 9-13-57, SibleySchool of Mechanical Engineering, Cornell University, Ithaca,New York.
Correspondence with H. H. Mabie, Professor of Mechanical Engineering, Virginia Polytechnic Institute, Blacksburg, Virginia,Jan., 1966.
~-.;--.•..f·'.~.- ...~.: 70
22.' "Metal Fati~'~;~"""Theory and Design," A.. F. Madayag, John Wileyr-,..and-.Sons, Inc'.·', New York ,"1969, 425 pp.IL23,._..._.~~Mechanical Engineering Design," J. E. Shigley, McGraw-Hill
Book Company, Inc., New York~' 1963, 631 pp.
24. "Stress, Strain, and Strength," R. C. Juvinall, McGraw-HillBook Company, Inc., New York, 1967,580 pp.
25. "Strength, Fracture, and Fatigue of Materials," T. Yokobori,P. Noordhoff, Ltd., Groninge'n, The Netherlands, 1964, 372 pp.
26. "Stress Concentration Design Factors," R. E. Peterson, JohnWiley and Sons, Inc., New York, 1953, 155 pp.
27. "Design and Development of and· Results from Combined BendingTorsion Fatigue Reliability Research Machines," D. B. Kececioglu,M. J. Saroni, H. W. Broome, and J. McConnell, NASA, Washington,D. C. ,J'uly '15, 1969, ~7.
~ •..-
28. "Reliability Prediction - Mechanical' Stress/Strength Interference,"C•. Lipson , N. J. Sheth, -and R. L. Disney,·Rome Air Development:Center-, Griffiss Air Force 'Base, New York, 1967, 450 pp.-----_.""._..,-_...
29.
....-
~UInteraction'AmongThe'Va;ious Phenomena Involved in the Designof Dynamic and Rotary Machinery and Their Effects on Reliability,"D. B. Kececioglu andE. B. Haugen, Office of Naval Research,Washington, D. C. 1968, 379 pp •
·~·--""--'-+.T .
.~~"".",."",=~.-=,,_ ...._.~.---~-_ .... "''--~-~'''''''''_'''''=-''"~~'''''''~~~-'.''..
~'_I
\._
/,'
..,_A
~.
•I
"_
,::;'
l-cill
GE
AR
S
I·"~t
II
!I,
•II
'""! ::
i; I I; !'
,,:.
','
i
;1;:
,I
,i
FL
EX
IBL
EC
OU
PL
ING
\ \ ~{\"
t;',:
,:Ji',
;,!i
,I"
':I"
I',:1
,1
,!;":
;:1i.i
I,"f
II
'!i
flr
.'.
CH
UC
K:':
\\
/./
,/('
~\'::.
.,,
',
toG.T
OR
QU
EC
OU
PL
ING
),I
![.l
f',
i•
'..,
'
i
..l'.
";;.,
ii'::
iTESTPIECE~
,.
,
ST
RA
ING
AG
ES
ST
RA
IN'
GA
GE
S
:.
.\
,. , I:
RO
LL
ER
BE
AR
ING
S
ji
GE
AR
Sb
l
,;
t4H
YD
RA
ULI
CC
YLI
ND
ER
FIG
UR
EI.
MA
BIE
-G
JES
DA
HL
FA
TIG
UE
TE
ST
ING
MA
CH
INE
..,J ....
\".
'-.-
/'-
oJ'
3uo
(2)
[[g
jJ
116
OJ
.~
1.
Mot
or2
.G
earb
oxes
1.L
oadi
ngFr
ame
Bea
ring
9.T
est
Spe
cim
en1
0.
Sli
pR
ing
san
dB
rush
es1
1.
Co
un
ter
12
.C
oupl
ing
13.
Saf
ety
Bri
dge
1u.
Bru
shH
olde
r15
.M
Oto
rB
rack
et1
6.
Mou
ntB
olt
s1
7.
Loa
dine
Fra
me·
18.
Wei
ghts
19
.~1achine
Nou
nt2
0.
Flo
or
Mou
nts
-..3
!'.)
'.;'
\,
FID
UR
E.
2.
CO
MB
INED
BE
ND
ING
-TO
RSI
ON
FAT
IGU
ER
EL
IAB
ILIT
YR
ESEA
RC
H.M
AC
HIN
E-
FRO
NT
VIE
W I
~}..
i.f
.r:,
.J'
,1.
Mo
tor
2.
Gea
rbo
xes
3.
Cou
plin
g4.
Infi
nit
eIn
dex
er5.
Bac
ksh
aft
6.C
oupl
ing
7.L
oadi
ngFr
ame
Den
ring
8.st
raig
ht
Sha
nkT
oo
lho
lder
9.T
est
Spe
cim
en10
.S
lip
Rin
gsan
dB
ru5
hes
....,~
...~. ..
I·.
IC
)IO
I:..
.:.'I .
I
'1)
,j j
.27
-..J
W
FIGU
RE'3
.CO
MBI
NED
BENDING-TORSIO~
FATI
GU
ER
ELIA
BIL
ITY
RESE
ARCH
MAC
HINE
-TO
PVI
EW
16,. - ~- ... '---';_...- .... ";" .
-:.' "~." . -. - , ..: . _. - . -. -. - _.. .
a). UANASA TES T SPECIMEN.. -'.~.;~:--->~ ..:: -- 6700 - B-002
. . ·C··'-<>
-. ~~",,.......... ~ .._: ~ ,.... -"""". "-. ~ ...', .
. '-.i.
. _:' -. .-~-'" . . . :. ....i...
1-.•
: I. '...~
I
-f'------j-.-----~ BENDING MOMENT;',
'b) . LOA.DING FRAME ANALYSI S
. FIGURE 4 •. ;:. LOADING FRAME ANALYSIS
AND TEST SPECIM EN
'-'.'.'.;
75
/
:~ "
-'I,,;.,..-'
:'.. -'
·t-'
C"t~ -,- .'~~-'
'1 c·.' c·~-~.,.
'.~'. -1;' " .-,~t_~'r-
<,-.C:'
':'"..-~;,..:.....
~~" r-.i"...- ;(/i
!~
"-.lr.~
..:'-- j:..~'
.C) " i
~
:-tJ~: ....,.-:(;£1
.~_. ~
'."..
36.0"
tw- Bending
Load"
FIGURE 5. BENDING ' LOAD LEVER ARM.\,
\,
-"
,•••.
>
....j
., I,
":..,;
....,,
'1""
""'"
BE
ND
ING
MO
ME
NT
-H
OL
DE
R'STR~IN
GA
GE
Si
(\
,,
"....
,~.
,.,__
,',.,
;..~
~,.:
.."....
..l~._"..
.....,~,
t'\
.i"\r
TO
RS
ION
::1
~.G~,GES
" ,(.}
,
'J;-
..'
(:1,.'-
...';
",(
';
:':''ii;
.!'
~..,~:
,,!
",~
,,~
SL
IPR
ING
AS
SE
MB
LY
.~1 -I I'
-..<
~'I,
..'~
BE
ND
ING
MO
ME
NT
-G
RO
OV
ES
TR
AIN
GA
GE
S
'.
I.
-:C
OUN
TERB
ALAN
CE
,-,;:~,:·.-;v,··1
I I"'--J
,.'.
'.•
':..:
:,~J
.I--
----
---
I
0,'
,
FIG
UR
E6
.'S
TR
AIN
GA
GE
AN
DS
LIP
RIN
GARRANG~MENT
-..J en
:;~,.'
77
- .~ -'R", R//;/. It 3
;
"_._-_.. "
-~----+---I---
... :tc.: -:... .. " (0) BENDING MqMENT STRAIN GAGES
"IC
' . . . ~;.'."
>'.', '.:; .
STATIONARY
: SLIP' RINGS... co " -,;, ' •
. ;~._" RO'TATINGj
,/..-,,-,f
/ ....,.
}HONEYWELL__ 119
-L'o.~--- AMPLIFIERL-....--;-------~--__t a RECORDER
". '-,: ...
. !. ",
'. -.~.' . :.
. (b) STffAIN GAGE BRIDGE ARRANGEMENT
FIGURE 7 ... ,· BENDING MOMENT INSTRUMENTATION.\ , . • J
T
,78
.
, (a) TORSION ST~AIN' GAGE
," • r
.", ,
: ....... ~ ..
STATIONARY
I ,
}
HONEYWELL119
~---- AMPLIFIER, a RECORDER
/
SLIP RINGS.' "1' -f'" l! ROTATING
" \ .I, :.'
'7 ~.~,::,::,-~.;;.-~:~-----==-=--. -,
'1 ",
'j ..
..
. '.~
i
i,
:1':.,:.
. ~ ..
(b) STRAIN GAGE BRIDGE ARRANGEMENT
FIGURE 8. " TORQUE INSTRUMENTATION. ,," ~ .'
, .
T
. \
~~ -
Gage Measures Strain Ina Directions
;::---.•. ~_,.:.., , __".w__ •
• ""/::;""<' ..~ -,
{_.----_.- ..
. I-'loo-:-_'"
.'....,-"~.;-
~ ..
-,--
..
= -T
,
.' .
. \
T
79
. ~.' !i ~.\
' ...., ..i
FIGURE 9. TORSIONAL STRE$SES FOR ONE GAGE.
Pan
wei
gh
t
Loa
ding
pan
...,,,'..
,'"
\11,
Jl.t
J\
....
'i
'lIi.
..:~,
gage
s.
\~;
,..
.\
/",tl
..#'
'~
II
[
Gro
ove.._~;"~i~~;g{1
.~.
.'specime~:j
~.."""
""'"
.'L
;',
.")
\.
--..
...;
......-
-!,~'.~?)
,'f
...
\
.-'i
i-J''~
1..';:
';~
"Too
1hol
der
stra
in
Co
llet
s
\.
'1.<
'
Toam
pli
fier
I~~
",,"
...._
oi'·
.!,
'"1;~
c·.,
"1
-'.\'~
~1*
:';:-"
"'l'''
~'i;1
',·:;'
,~.'rr
-',;~
Sli
pri
ng
Loa
d,in
gbe~ing
/(
I f f I ,
clam
p-sc
rew
dev
ice
Toam
pli
fier
.,
(":"
\1!r\
";.
FIG
UR
E1
0.
TE
STSE
T-U
PFO
RTl
m-:E
'EN
DIN
GC
AN
TILE
VER
CA
LIB
RA
TIO
N.
(X)
o
..:"-"
.~~.
~T
orqu
e
Vis
e
clam
psc
rew
dev
ice
.,
Bea
ring
hous
ing
\
Co
llet
To
amp
lifi
er.,
(
,C."
,'
"pa
n.,
',
pan ....
.,
.'
FIG
URE
11
.TE
STSE
T-U
PFO
RTO
RQUE
INTE
RACT
ION
INTO
BEND
ING
BRID
GEO
UTP
UT.
i,
CD ....
"
12
.0
. > ori~
.. gli
.5o
ri +oJ
C) "'H ell
+oJ~ H tlO
11
.0~ ·ri~ ~ Q
l~
10
.5
o o "
.,
.'\
"
..... .'
".,'
.'
o
~
,.
'.,-
."'>.
..."
o
.:
-:. o
II
I0
II
I1
0.0
II
'I
I
o5
1015
202~
30
35
40
45
Tor
que
Lev
el,
dive
FIG
URE
12
.TO
RQUE
INTE
RACT
ION
INTO
BEN
DIN
GM
ACHI
NE#1
.
CD~
I::-
§e
M "Cl
+J
I::,I
/)0
':St'
Mt)
o.-f
(1)
4-l
I/)
.~.
.1
4.0
+1
9.
> .... "Cl .. I:: 0 o.
-f +J t) ro
13
.5M (1
)+
J I:: H tlO ~ o.-f
"Cl~
13
.0T
18
.(1
)p:
:j
12
.5T
17
.'>
---
:
. "
'"
,.
..,
•....
".~
"..
"',
'
:.,' "
,'
.1
:'.,
,"'"
:~)
,
o o o
o
',:
12
.01
"1
7.0
;I
II
II
II
I
o5
,10
15
20
Tor
que
Lev
el,
div
.
2530
35
40
n '\'JFI
GU
RE
13
.TO
RQU
EIN
TE
RA
CT
ION
INT
OB
END
ING
l::::.:'
.M
AC
HIN
E#
2.
co (.l)
.~.;
f
-",-
;v""
·.
f);;t-
.i'·~l
,i·
",.,1
.";'
."v.
;I•..
,r,.
.".(
';'
.".
,."
,;
~
·'·I.
:
",
..;
.,.,. "
\.
.','
,,-'
<1
" .'
"
~. JI ~ "~ ::
1530
20
~ ~ .. +-'~
,0
.25
·'~ o ~ o o 19
2025
3035
,40
45
~oo1ho1der
du
tpu
t,d
iv.
FIG
URE
14
.SP
ECIM
ENGR
OOVE
OUTP
UTVE
RSUS
TOOL
HOLD
EROU
TPUT
-MA
CHIN
E#1
.
0) -'=
.;.
'-,
'.~-'
...-'":
'...
,
-..'.
~
,,
"
"
-.:
'.
\.
.'-.' .\
.....,
....
.\'"
.
..,.
.':.
.,",
.;"
,.~'.
".
',1
,
~":...
.'
'./ . .',-,
"--,
.
,
,-..
o1060 ~50
.'40
=> ..-(
''0 • +'
::J Pc
+J ::J 0
30
Q) > 0 0 ~ C) .
20
o5
10
1520
2530
354
0
To
olh
old
erO
utp
ut,
div
.
FIG
URE
15.
SPEC
IMEN
GROO
VEOU
TPUT
VERS
USTO
OLHO
LDER
OUTP
UT-
MAC
HINE
#2.
CIO
<.n
60t
'f'
.. "
f',~',.,
.'.'
.:
.:-'
:\':
~'" ..
,..
:
.,'
1'i
U ,t.! "", ,.. '.::
.10
'(I
'"~..
:
-..,'
.,.\~;
"
..'~.
.r...t
.L-...
.:...,.
.:..,:
~'..:-
1"1.1'
.i
I'"
"..
.--,'
,...
..".,
"\'
'..
I"f"
0"
.-_
'.'
',"'~
':7
"'''';'';.-
...!
'"
I:I·
.··
L!'
)·::.
1-.:-
-1~
i
t.i
,/1 I l
..
.I."
·_·:·
·..:···
.:I··.J·
~'"',,
:,:-
.'/'.'
'I.
';'
..
-"~"1'
.,
..
-..,
I:'
..:....
.1
.:'
..'
.:.:
",,
1';.
"1
.'.-
'A••
'
."".
"'"'
f~:
<"':
:;'
''1,
"
e, .\':.
-,.,
~.
.
, ..."
'...
.....'
.'
..'~.
,.
.....
.~r
iIi
~..,
./,,~,
+L,"
~,,1':';1
'..'I'
"I'
,.\.,
.",.
,I,"j'
".r'.
t·,
h':-
f'!..
'1."
".
"I
I•
Ii',
"I·
,'"
ro·.'..
..!
,j.
fI
,~.
/i'
;""'
f1"
iI
!'.
'!
I!.
;,.
.'1
'1'.
,.I
'j".
1I
I\
I~
..'
f;,'::'
,I·:;..
.'..
I!
II·
·,I
"'.,.
.'.':.a·.
.'
.,
f.'..
iI
..:
II
..<
:::
..".
!;"
..
:'
i!
'jII
It:
..,:'
~'.."
'1"
"",:--
1-1
,.1
1'\I'J
."....,',...
.).
'j,'
.l'.;
:~.I!
'~..;
'!:'"
.,'.
...."i
\1
.I
t'1
".'
!1
'1
:
I.r;
.,,',:
1r.'
1.
1 :'
..J,~
:.la
I.
'i.
:'
I,.\
.\..1
.·..
't.
·.,
7"•
...
.!.'
~'.',.
'"
.".II
..iI
.,
1!
.i
·1
.I"
.i'
"~
\y
.,.;.!
f"··t{
,.
45 35
505S . > or{ 't:' ..
-+J [ -+J
'.:::J o lU ~ o ~
40
......
30
'e
II
II
II
II
II
II
II
II
Ife--+
--o
f---
20'.
22
24
2628
303
234
363B
I40
To
olh
old
erO
utp
ut,
dive
FIG
URE
16
.SP
ECIM
ENGR
OOVE
OUTP
UTVE
RSUS
TOOL
HOLD
EROU
TPUT
MAC
HINE
#3.
{i }~.
(X)
O'l
.J
::
';.~
!.<
.
..
Coupling
earing Housing
Slliprings
Bending- Giges
T6rque~~ages
87
, ': ..... ' ...: ~
." .'
...... :
.. : ~;
·._.·.• 1. /.,
Fx
Wnut
ws
IC--- L3 ----J~
L4-----~
LS------oDof
LA--------JiiJilf
.....----- ~ ----------.1L7
L8....--------Lg--------------.t....---------L10 -------------..tJoCI---------Lll-----------------.-.t
. . ~.
FIGURE 17. trREE BODY DIAGRAM OF TOOLHOLDER.
II
iI
88
, .. '.
.. ,
Balance pan
. .
Coupling Housing
.'. : .. , ~
o
Balance bar
Soale-·,p'!l
'1-,
......._.; I., .... :,~..... ,' \ I .•_..l~ . \- I
WoodBlock
\ .
Loading pan
FIGURE 18. LABORATORY SET-UP OF LEVER ARM, PIVOT POINT, RESISTING. MOMENT MEASUREMENT EXPERIMENT •
..
89:\'. . .,
'., -\. . ~
.\•. -'1."("'••, "
\. \
\\\"\~.\
\
\\
'\\\
\..C,
\"
---.,.--------- SCALE GRADUATIONS\
A ----------------------------------------
B.---------------------------------------- LEVEL SHAFT POSITION
c ----------------------------------------
. FIGURE 19. SCALE GRADUATIONS OF MEASUREMENT. EXPERIMENT/
.,'
. i
.,.i... .;, .
"
"
.r
.~..
"'.'
,-
.-:..-...
~,
..~'
..;-
......
....
,-':~.
-.;'. -.(
l..'"
-":-.
',.,.:
::'.'.
...~
..•
f
.,.....
FIG
UR
E2
0.
TOR
QU
EOUTP~
VE
RS
US
PAN
WE
IGH
T
51
01
52
02
530
35
40
45
Pan
Wei
ght,
1b
s.
MAC
HINE
#-1
.ID 0
:~
.~'-
~"';cr
:
./
.'.
,-'"
.....,
..'.~'
....,~:.:,
",""
........
..
':" .>.:
\..
"..'-
...
.' ,;-'''
'.~.
-":-
.
.,.,:
..:..
.:
;.:,
i,.~ "
.. ".
".;
..
..J
'.
'.',
.':'.'
":~
....:;:'
.....
.~
.
.:.'
'::.
- .;'.
-.'.' ,'-
\":
..,
.., .•.. -..
.f
51
01
520
Pan
Wei
gh
t,'l
bs.
25
30
35
40
45
..~.
FIG
URE
21.
TORQ
UEOU
TPUT
VER
SUS
PAN
WEI
GHT
I
MAC
HINE
#2(0 ....
, ,
".
.,
-d
.:'.
';"".
'..
..
:.
"..'.
"
......
.
(:'.,'
....
.:;
"
.'
",,/
.iI
.r~
•
///'
"''''
>.,
,/"'
'''.
.:,..
;,>
/.t·
.1",
r>
~""j""
~"'"
),
_;/~
~A~
:,>",;
//'
,/.,
,,
/:/~
>'.
..../>;
?."
>...
:'
j'.
'L
,.';
':_.
,'.:
'
..':
'f'1
...",
.::
~..
.,::
"
'-.
1
/
30
.I 35
'.~"> .....
.~
...
-i,'
0
'1''~40
.:.15 10 5
,,>.
''''l'
~.l
~"\,
.
",' n I j' r
."(\,
i"\ "
.,..
.-.+-
'.
::S'0
.+-
'.::
:6 .¢
)25
,::s cr
.•H
'0
):'~
~\20
';"
o-I
.,,
I.
I,
II
II
454
03
53
020
15
10
25
Pan
J'le
ight
,ib
s.
TORQ
UEOU
TPUT
VERS
USPA
NW
EIGH
T~.
MAC
HINE
#3.
5
FIG
URE
22.
o·
.U)
I\)
i
"..
3?
+0
.5
1.0
to>
18
o
o0
0
16
14
,"
..~'.'
.".
,,:,
'
,,'
.'>
..!
12
"
""
",'
10..
.~;
"':
'".-
.
;::;v ,,
;.,
8
,..
~o0
00
Ben
ding
Ou
tpu
t,d
ive
6
"'.
','
4
.:.'
BEND
ING
INTE
RA
CTI
ON
INTO
TORQ
UE-
MAC
HINE
#1.
2
"
FIG
UR
E23
.
o
+0
.4
~ ~~~:~-
:~---;
---~=~
=:~=--
:---~~
--~~~
I~:I
II
II
II
~I
~O.~
+0
.3
. > or!~
...fJ g,
+0.
2'.fJ ::1 0 ~ ,g. ~.
.~0
.+
•.>
','
,
......
'"0
.0
t t",~
"
?
;.'
.;;'.
"
:',
.'.~
'"
"40~
....,:,;'.
.!
.;.01-
;',
".."
.~
".~
..".t
""
,',,;-
....':
.:.-
.' > "'"''t:J
,3
~;..
'.-
.~'.
,'.:
.
~;.-
'..\
-~~_.-
-,-,
..':"
~~"".
''..
'1
'"
.... ,-.':
.:'
,.,'
.~.,
.',,:
",...
.
..
,•.
..
,',<
",
/.
1
• & +>"-
6 III
':'g.
2'M
"'.0
•,r:
-",
o5
10
15
202
53
03
54
04
5
Ben
ding
Ou
tpu
t,d
ive
FIG
URE
24.
BEND
ING
INTE
RACT
ION
INTO
TORQ
UE-
MAC
HINE
#2.
\.0 .s::
......
..':~
"..,'
.:/,
..:.-<
~~:
",,'
,g (.
0.2
<....~\
-.'.;
....•,
..~
,4'°
0,
':'.
'
.'-
....;
'.
,'..,
".-..'
'...
.•,
.~.
:-~'-~~'-"
.~-,
...
":"'
,
..,
'.
...•.
.~.
:."...
...;.:.
":..
...
',:::.:.
"'j•
......
.•f'"
:".
.,-
..,..
..
--
....•."
.
..... ..,
':."
.'.:
,',
'.;.:
..
x
"
,_..
-_
.f',
:,,
,'.
","
.'
~
/.'::
::.:,
:,
-0.3
.;'0
.2
-0.1
-0.'
8
.,"..
-0.7
. '.
0.1
':"0
.9
0.0
!':":'.~'.
.... 5.,
-;;
-0
.4o (I
) g. M.
0"
E:-o•
.'
.
•. .:; ~
•:-ll
..0
-1I
II
II
II
II
II
II
II
II
I"'
I~
0I
I
o1
23
45
67
89
10
11
12
13
14
..1
51
61
71
81
920
2122
Ben
ding
Output~
div
.lO (J
1
FIG
URE
25.
BEND
ING
INTE
RACT
ION
INTO
TORQ
UEM
ACHI
NE#3
.i . t.__
CD
,<".....
....
.'
.......
.
.'
"
26
"
"
..)
"
.'
-';
25I
I.c;
II
II
II
I
o5
101
520
253
035
40
Ben
ding
Ou
tpu
t,di
ve
FIG
URE
26.
BEND
ING
INTE
RACT
ION
INTO
TORQ
UEAT
HIGH
TORQ
UELE
VEL
-M
ACHI
NE#2
.1.
0O
l
\N...
...; t::.,
.~.;
.,'-
+1
.0
··-'
.f:;-
"
+0
.5
'.",
> -..i "d
0.0
s:: ttl 4> ::c r-I ttl or4 X < lI)-0
.5g ~ (!)
..1.0
i:~
J'"
I
".~> ~'.~
.
......
.,
,.-'
,.
o o
", ,""
...
o
.:.
,"
"....; \
.!
....
":'(
" ......
.",
o o o-1
.5I
II
II
II
II
I
o5
10
1520
2530
354
04
5
Ben
din
gL
evel
,d
ive
FIG
UR
E2
7.
AX
IAL'
OU
TPU
TV
ERSU
SBE
NDIN
GLO
ADM
ACHI
NE#
1.
U)
~
..,.
,'.
~
....
"
o
,.... o
'
15
"'.:
"':.:'"
\
.... 20
...<
'.;
':':
'.
".:
o
....:
.'.1
••
'
',-;
..........
.i'
..~
.'
"
25
,
,'....
~
'/
',.I
':' 30
o
", f'·
.:,
35
o 40
o
:;:.
'
45
Tor
que
Lev
el,
div
.
FIG
URE
28.
AXIA
LOU
TPUT
VERS
USTO
RQUE
LOAD
MAC
HINE
#l~
•'>~
.!"\
.:,.';
l.O ex>
~~nding stress level in groovE
,,~
Stress ratio
99
2- C1trueGR
1--
Shear stress
TtrueGR =
r s
level inC1trueGRr s fi
groove
4--
Output stress in grooveC1 =K a 'outGR BGR true
GR
Output stress in toolholder
a = K C1outTH GR-TH outGR
Shear stress level intoolholder
1 CTH J GRT =--'---- T5-- trueTH Kts J TH C
GRtrue
GR
ToutTH =TtrueTH/Kr
Output stress in toolholdercorrected for torque
6-- interaction "a' =a + K_ T. outTH outTH -lIB out
TH
Shear stress level intoolholder corrected for
7-- bending interactionT' = T + K aoutTH outTH BIT out
TH
iI
8--
Visicorder bendingoutput
N' N G R_cal , a calN. - E Rv1sB gage
level
'\
9--
Visicorder. torque leveloutput
N N G R= cal a cal~,N • E •V1ST ' Rgage out
TH
\.
FIGURE 29. THE CALIBRATION FLOW CHART.
·100
\
.. --_.-. ~ ---- -_.. -- .... -- -_ .. _.,. -_._---- ----- ..... __ .. _---_ ...-----_ ... -----..._--- - .....__ .._--_.. __ .._---...... -----_ ..._----------_.
.... ' .. - ... __....._'-_.__._._--_._---
-----~···---1)~~frUOft·_STRORt!O.t· TAUlJR"1501.-1ttSOt-·-···-·--··-----.----·---··--------:NCOUhT • 0
--···C------··RfAO i" THE NUMBER OF STRESS LEVELS' IN RUtor-'R~AO 161, NOSTLV
---=-16"'"'1'--:'''':;:-ORM AlCTs-.----·-___1t.9.__..~JHNr '0 _.,
90 FORMAT Ill'll/It___. .~CQU~t. NC~~N.T~.• '
PRINT 60.C !'eRMAT FglL2.~l~U.1.. ~e:AO']N~S_ .. __. ~. --=-o-=-,,------:~____=:_:___,__~__=:_:_----
. 60 IfORNAT (JX,4MT£5T,3X.6HSPECIMEN,3X,7HMACHINE.5X,3HPAN,"X,4HRCAL,7X '.'.
·----------~·~~~~-~~~:1 ~ :·~~~T,· ~.~~-~::~~~~~: ::~~~~~~~.:.;~; ;~~~~-~~.:~ ;;~=~~~af~-~~-:~-~ ---. -..-- .. -------. '... .--------lIa~~~g~~~~::~~~~~~~~~:~~~~~~~~~~ ~~~·:;~~:~Ig~·~Y~ t~~·.~~T9A~U~.t4.XL -...-... _. -._-~~-
:c R!.AD~~~...NU~J!~.!L()£._~~JlDSJN TH.e:_~lR~~!.l.!YiI.~O THE M~O~OEIL- _~ Olf OPE~AIION OF THE MA~HINE .
READ lOa, NCARDS,MOOE------rOCf·'ORMjYTzIsy··.. -. -... ,
Do lZ0 I ~ 1, NCARDS---:e---'---'RE'AD -Yh TM~' T£51 NO~, SPECIMEN' NO, •. MACHINE NO,. ·PANWEIGHT,ANO'·tHt--· .. ---------.
:C CALIDRATION RESISTANCE, VISICOROER CALIBRATION DISTANCE AND VISICOAOEA·c OuTPUT DlvnloN--SFOR'lKNon~-a-ANO-,.o"RQUE-'---- . . .'.. . 80 READ 10,NOTEST,NOSPEC,~ACHNO,NPANWT,ACALB,ENCALB,ENVISB,RCALT, .'.-----------n:f-iCAL"T,ENVrsr----- -...... . .' .. ' - --.-~----------:-
10 'OR~AT (4I9,6r10.2t .. - ·.,.----------'"tA·AOl"4:-.-·tiiCAlllrS··· -- ---. . - -. -"'.- ..--- ..-- -- ..~..-- --- - .--..- ---•.-----------------.... '-'. .....~e DEF~~! TME ELASTIC MODULUS, NO, OF ACTIVE ARMS OF THE BRIDGES, THE
.t Rln-STiNCrs-or-rRr'-g£RDlffG-lNO'-TORQij[-Gll1tlEstANO THE B[ND1N1fAmNO...=·:...-------c' . TORQUE GAUGE 'ACTORS . . '.-----------rii"j-O·o--O 000'0 ~ .----..•..--..---- --.: .. -.-..------..~- - ----..•...•..--.--..-.--------.. --.---.--
. ENA-4o' .'---------lfGAGl!B.19~'~"....' .. - ...--.----..•.....- .-..•.. ,.. -.... - --.-- --_ ... --. _..•- .•...-,--....•-- ...-.-....-------------------. RGAGETo120,
._---------_ ..- .__ _---_ - .. _ .. ..,___~_- _;__--.-._ .•..a--- _
• ,J.'
J
101
I •
; :
., ;.._ -- .._-~ ------_ ---- ~-.. ------ -_ .._---_.
. . '. -' .....·4_.·. __· ..·
: ~ ..
..............., ..".', .
. -----_.---_..._----.-------------
.. _.__ .._....__._----_. -----------
GO 'flO 50·-------2.0. CB-aif••U't-
·CQRTr.. ".02OJ..----. -'" -'e Tii~88~
CTO-.0948.,..,---celi;'Oz9'S-' . _.. . ----- - __ u _. :',' .:••-- -.------
. GO 'flO50' . ,.---.. "30" eeGA- .902 .',.. ' . - -- ...•.,- -
CGRTotolo.0198 .._-_ --~eT ••a40 -CTBo.0143 .
----.~C~8Tl)o04TO--·-···-..-- ..-..-·-···---·..,-..·-:.... "':~'-"-'-'----:-'-'-------------GO 1050 - '
L-----'--·40--CBGAIi ~912 ... .. .... . '," '.CORl~•• 02i~ . .'----·--------C'-a.-94'·-·---·· -- .._..- _.. ~ _ :... -,.,'CTBa,OCBTa-,0169's
.' .--..---.-~~-~AbJi~~i~.! ~~~_!,9.,.O) 90 TO l~O .;C'CAI.OULATtQN Of' B!NOING STRESS LEVEL, ,OA IH,INlTV RATIO
·------160-·S0lJll~ii'{EN9! SBc.[*RGAG£B II tENCAle*£NA'ttG8*ACAL.9)· . - .... .".----.... - .._-- --.---•. -. - ' .•STRe~(I)_ SOUTH I CCGRTH .·CB~RJ .,
. TAUoN Cit" 6";-0-·---·---'-----------·------ ._- :,' .:'j
'PRINT 71'NOTE$T,NOSPEC,MACHNO,NPANWT,RCAL8,ENCALB~ENVISB,RCALT, .!
~-----~----I"E'~C'A~Ti"EN\lrsT'STRGRor.TAUGIHn-·- '-" .' ",," -...---- . "...... -----.------------ ,j
.. "h ., ;ORMAT (U, IJ '6)(, 13,0X, 11 ,81(,12,'14, ifF!O,Z"11.2,'12,ltF9,2,'iot2,·· .,.,r;·----------1F11.TiTn;T~'2x.6HTNFIN./)· -'., -.. ,.- -' "., - ,.---------------------
. . GO '0 120 .·t cALCULAtIoN OF aENDINGSTR[SS, .. SH!AR~STR[SS·ANO STRESS RATIO FOR:C' ALL Ht\UE RATIOS· . ". ....,,~
---·--':':-sT-"S001'~lJ'rENvlsa-filroROAuE'81-,-Cl!NCAl;B'*EliAlfGtJ·ReAtB1--·~-·--··.'''--.. -.---------,-,---------------- .. . .-. TAUll"Pa (ENVIST*ElIIFlOAGEl) I CENCALT*ENAeOT*ACALn . . - .--..:------..'!'OlJ1lffIiSOUTHP-CnrO,.-Aunw·-' -.-- .. - -- - -.. --.-- .- ----.--':-----------.~.
~TAU'~8TAUTHPDcaTeSOUTHSTRGIH h 0 SOUnf"7~llG'R}---~-·
TAUGPClluCTo1YAUTH/.0157) .'--~--------ATrr-a--STR'GRT I f7-nA'UGR rn .. '.~ -1·~-732r'-'·· .--- '---".- ----- ...---- --------------------------~
'PRINT 70;NOTEST.N05PEC,MACHNO,NPANWT,ACAL8,ENCALB~ENVISB,RCALT.---------ll:UCAl;ToE'NVlST,STFfGlnrr;l'AUGRCl) ,RH)'--" --'- . ... --,... --,,-.. ---------..---~,,------
. '70 FORMAT C4X, 13, 6)(,13,8)(, I1,8X, I2,n.,l,F10,~,'!h?FU.l'F9,2,,.i0.2, ';' .
.. ,1~11.1,Fl1.1"8,3/)120 ~ONTlNU! . .....
--.,c------~-eALOUL"A·nOtrOF-'ME'AN'·ANO- ..STANOARD ' DEYIAllOH '0,. 'BEND IHO' -STRE"SS';··'SHEA'R----- ....----~-·,~ - ,S'R~5S,ANO STRESS RATIO
-'--------tifCl-'-H£A-N--(STROA,....· CAROS,NCAADS', XMEAN, . O!V)·. ", PRINT 3 .
;) FORNH «lHO) -"-----PRINT 130~ XMEAN, DEY
·-1'3'lf----.,O'RRilr-U~)f,31HMEAN BENDING STRESS IN GROOVE-,F10.1,SHPS1.II11X-'·:- .. ---------. i39H5TO. DEV. OF BENDING .STRESS IN GROOV~ -,F10.2,5M PSI./J------TFT2)~vrsT;EQ.o-.-O)60 "TO 200
CALL MEANcTAuaR, CARDS, NCAAOS,XMEAN, DEY)PRINT :I .__."....._-_... -..-.- ....... -.- ....-,..... .. ...- ..... - .... -...- .. ------
PRINt 140. XMEAN, DEY·--T41"-"--.,.ORRJl·r--CZOlf.30HMEAN 'TORQUE STRESS IN GROOVE • ,n 0.1 ,SH PSt .1112K, .--- .....
'. i38MSTO, OEV, Of' TORQUE STRESS IN GROOVE :o,F1002,5K PSI,') ..,·------tAL~--JilE"K.. r-:--·--'A. ·-"_tARDS,NCARDS. Xfl4EAN, DEV)' ... -.....-.;. ---,--....... -- ._.. ,--.--~
. ;
. ,
...' ,,.'
, .
. .
'...." '1;:
... ;..... ,
,--_._-_ ...__._. __.._---.... _..-.- ..-... - ...._._.-_..,.._----._- -~----.. ._ -0. _.•_ •.••.. :" _"__.••_.•' ._. • • . FIGURE 30. DATA REDUCTION PROGRAM (continued) •.
----_..._--------_._.~._.. _....... ... .. ......... -.' , .
.'.'le.
,.._-------.
!
102
PRtNT :I------.·- .. ·...,.RfNT.T50~ XMUN, orv
150 'ORMAT CllX,19HMEAN STRESS RATIO ~'F10.!lIlIX.32HS~D .. DEY. 0' M~~~._._.__ ._. .-"'---'-,' 'sTRESS ~UTIO -,"10.15)_--:2~O:..=.O_~.A~~~~J.NOSTLV)GO TQ J_1o ..
END______l .•. ...
FIQURE30. .DATA R&DUCTION PROGRAM .. (OOlIlt1nued).......•- -.-- --------.••.
------~-----_. __.._...--...--. _.•.
------_......__._-~-- -.., .....-
." . .----""*-.."---'''''__....-
___ -0' __ .. _ ......... _ ... .:.. ...... _0 _ •
. . .-----'._--------_.._-_.- -- .•. -. , ....-...-..-......;-- _..--.
. . ...• •• •••• _ • 4 _ h __ •
. . . ~. .___ u ~----------:.. - •••••••_ __ . __,_ ~_ .
.-_._-_. --_.__._----_._---------
._.._- .._----_._----'-----
----_._----~----
/
-----_._...~..;..;...;;.;.:.;;..:._.. :._-.;......:_:;..._::.. __. _.. _._:.....:.~;;:- ...•. __ .. __....__ ....
--....;...-------_..~ --_.._-_ ...
103
75
or-!1IlPo~ 70 ."~
>00M
(!)
s::. or-!
1Il 651Il~
M ,. ~ ~;.+'(/)
bO '.fs::or-!"CIs:: 60.~
~
~
~
55
50
22 23 24 25
Pan Weight, lbs.
• 26 27
-,.. ..-.
FIGURE 31. PAN WEIGHT VERSUS TRUE BENDING STRESS FOR ENDURANCESTRENGTH FOR r =~.
s
"" .
104
66
65
64
63
• .-jfI)
0-~ 62,.fI)fI)(])~ 61+J
(/)
bOl=l
'.-j
60."0l=l(])
.I:Q
(]) 59~
t:.58
57
56
55
24 25 26 27
Pan Weight,. lbs •
28 29
FIGURE 32. PAN WEIGHT VERSUS TRUE BENDING STRESS FOR ENDURANCESTRENGTH FOR r = 0.90.s
r I (~"'; -....'.~ I~: ../
154,000·psi~
as=
1,5
00
psi
,V
=1.
0%
..
.,
;~.->
,;..
.......
____
_~
=87
,000
psi
,as~=
1,0
00
psi
,V
=1.
1%.r
s
Vs
=10
4,50
0p
si,
as=
2,80
0p
si,
V=
2.7%
Mea
n,r
=·co
s
J~s=
12
1,5
00
psi
,as
=1
,00
0p
si,
V=
0.8%
*
lJ=
78,0
00~~
~
as=
1,9
00
psi
V::
2.4
%
/"
...../~'"'
Ap
pli
edb
end
ing
stre
ssd
istr
ibu
tio
ns
180
170
160
150
140
130
.... l1)
~12
0..
110
~ S r-l
100
'.-i III ~ 0
90+" l1
)l1
)
~80
-b en bO70
~ .....~ ~ ~ ~
60+" r-
l <
50 40
103
104
10
7
Cy
c1
es-
to-f
ail
ure
,\
FIG
URE
33.
APP
LIED
BEND
ING
STRE
SSD
ISTR
IBU
TIO
NS
FOR
STRE
SSRA
TIO
OF..
co.
~ o (Jl
76,5
00p
si,
oSB
=2
,80
0p
si,
I~B
=3
.7%
0.6
82
,o
Sr
=0
.04
2,
VI'
=6.
2%
,.:,
.~.~
~~.
0'
R·,q
~.:
;.,.
;;
yr.=-0
.90
..
.Su
bsc
rip
tB
:A
ppli
edb
end
ing
stre
ss
Su
bsc
rip
tr:
Str
ess
rati
o
....
~10~5qR
ps~,
~$
;~,~qR
p,s
i,,V
B~
t·~~
_.
00
0'.
~;.
o'l
~..'
~0B
._:
.,..
,_..
0.1
/.._
•_
(0
=9~
7@~,
O§r
::P;
·Q2
fJ·V
I.'~
~~:~~
;~
,~BS~':97,509
,p~f'
a.s'B~5,~pp
P?~,rYB
~ii
=0:
7()3
,O'
S--,
=c:r
O~OS
6°,
V=
's-:
O%
::
).lr
=:,~i.~1)3,
(J~r
=(:
..:':
;;6.
"r
=8.
0"5
r.....
.,-
rM
ean,
I'=
'b.7
0o'
s
"
App
lied
ben
din
gst
ress
dis
trib
uti
on
s
~BS=70,000
psi
,oS
B=
1,80
0ps
i,.V
B=
2.5%
~r
=0
.70
9,
oS
r=
0.0
31
,V r
=4.
3%50
.:.."
"180
170
160
150
"'"'14
0rI
)
.~
130
• Cl
120
J:.. ~ ..-I
110
"'"'rtl ~ 010
0+J en
.90=
en Cl b C
I)80
bO l: "'"'+J 1070
l: J:.. Cl
+J ..-I
.<
.·•.
60
......
40
.'10
310
410
510
610
7
Cy
cle
s-to
-fail
ure
,N
o..
FIG
URE
34.
APP
LIED
BEN
DIN
GST
RES
SD
ISTR
IBU
TIO
NS
FOR
AST
RES
SRA
TIO
OF0
.70
AND
ENDU
RANC
EST
REN
GTH
FOR
AST
RES
SRA
TIO
OF
0.9
0.
.... o 0'1
l.'
f.;.·~
":,,,
~-.
.~
~
t.:'"...<
l('
It~
"
'," ""...
"
':....
..~.
106
..~.
.........
'~..
t;,..
~<
,.......
r'
=eo
~r
s,r
r=
0.9
0..~~_
:LS
,",
I"
,I
'I
I"1
...
,I
"L
Th
eo
ret
ica1
,r
=""
,I,~_,
S,
I"T
--=
---
--.
;..
.I
I'
.----
I,
I'
II
I I
./
,,.,
1,
1;"
,.f,
)!':
'H
i''~
.:.·
t"r'
i~(I
(':l
[tI
::
~",~,,,,,.,
.~,...
"..
./
I,,/
,..'-..
105
r=
""S.,
."
~.."~.
~.(~~
'hl)
/....
=0.70~
,1
04
rs
Th
eo
reti
cal,
rs="
"
103
506080 70 4090
liO
,'
100
120
180
170
160
150
140
,130
;'o
ri,(
I)
;'f1
", I<...~
"i~
;.Q
l
'r'~
':'=' ~ ori III ~ o +.J
(I)
(I)
Ql~ +.J
(/) ~ •.-1 ti
.'s:: ~ Q
l ;t; <
Cy
cle
s-to
-fail
ure
,N
FIG
UR
E3
5.
FATI
GU
EST
REN
GTH
VER
SUS
MEA
NC
YC
LE
S-T
O-F
AIL
UR
E,A
ND
ENDU
RANC
EST
REN
GTH
FOR
r sO
F""
,0
.70
,~ND
0.9
0,
AND
'TH
ETH
EORE
TICA
LS-
NDI
AGRA
MFO
Rrs
="".
.... o ....:I
..
.. •,
•Q
o
~
..•.
..'8
•.....
'.
r8
0S
"
.
r•
0.9
0s
~'7
00
.E
Fo
r10
cy
cle
so
fli
fe
__
,L
"~
----
'.
..
r.
--....
.' -- ...... .....
. .... '".... " "
" "
---~---~-~-~---------
-......... '"
...." ", \ \ \ .
80
)s
o6Ls
80 40
GIll
Sot
4.1 as .d 4.1 ~ GJ Sot~ (I),
ClO~
'.. ~ e
,G)
4.1~ <
,40
'.
8012
016
020
024
0....
'Str
eng
tha
tr
•0 ,
kp
sis
FIG
URE
36
.TH
REE-
DIM
ENSI
ONA
LD
ISTR
IBU
TIO
NA
LGO
ODM
ANFA
TIG
UEST
RENG
THSU
RFAC
EBA
SED
ONRE
SEAR
CHRE
SULT
SRE
PORT
EDHE
REFO
R10
7CY
CLES
OFLI
FE•
".
".
.~I:' ".t !" r" !
~ o (0
•~.-
•••_
_._
."
••.•
'_
_.,
-•••••
~_
_•
~----:
.•-,_
_.._
_•~_._
_...'~T-
..:;
...~
__
"I j" I"
,.,'
.
109
"
_••.•" •••,. ; • "~' ....;- • ~ ., ........ ~l.
, ,. I
x••
,Holes rId = 0.045-Holes rId = 0.125'Fillets - rId = 0.15
(DId = 1.5)(t/r=1.66 )
r
SAE 1020 Steel(Normalized)
Ni'- Mo Steel(Normalized)
•,0 0.04 0.08 0.12 0.16(1)~.24 0.28 0.32 0.36' 0.40
Fillet or Hole Radius, r, in./
o...' ,
1-, '
o., ," O.
at..~0f.I0IlJ
'J&.
~....> 0....f.I 1.....0s::G>
(I)
'5f.I
, , '~
':'
FIGURE 37. NOTCH SENSITIVITY Or. NORMALIZED STEEL SPECIMENS. .';" ".
'. ~ . '". ;" .
....
1.0
0.01 0.02
;.".,... -
"Notch" Radius, r, in.
FIGURE ~e.NOTCH SENSITIVITY OF QUENCHED AND TEMPERED ALLOYSTEEL SPECIMENS.
0.14
.Sir- ~t~Based on data
tlr Jt4
Quenched & Tempered
0.06 0.10Notch radius, r, in'
~ Annealed or Normalized
0.02 .
oP· 8
l.0
...Hot o.6
~
~.... 9.4>....
.j.I .....~ 0.2~
:'5 0.j.I
~ 0
• FIGURE 39. CURVES OF NOTCH SENSITIVITY VERSUS RADIUS FOR_STEELS, BENDING OR AXIAL LOADING.
j -....)
'. .. · •.._..t.·f ";:".1.•
..... /.
- : '~ ',. ", "-' .
" ,- ,.
'. t:J4;..:.
" .'.. '
'~. -._ ...: .
. .- " '.. ." . ~0.8 .. l.00.4 0.6
did'
0.2 --
r--..,.--.,.-r----r--r--r----,--~-'"T'-..,.._......, Tensile Strength,psi '.": - ..;:;;.-.......---lo<...........J.oIo".~----r-13B,000 - -
..........._=--=::0...;=-----+- 149,000 ._~~_180,OOO
'-...:;..;;~ +_-75,000
"",--:~~~':";"'---=t__52,000
...2,2H
0'. ' .j.I
2.00Ill-~
t' l.8....l.6>....
.j.I....1.4l/)
r::Q)
1.2C1J
.c:o ·l.0.j.I0 0Z
\ '
.-
.. .' .. -
FIGURE l!M). THE FATI(,;UE NOTCH FACTOR. . ".~: - ..'.
", .-:
'. ,
0.16
-------
0.ili2
_Steels---Alum. allo
0.080.04
Notch Radius, r, in
----...---...'""'....,,;1'
/'~
/
at 1.0...~0 0.8• .f-I0
~.>, 0.6.f-I.,..>.,.. O.G. .f-I
0'"(f)
s::OJ 0.2(f)
.c0.f-I 0.00:z;
0
-:
'.
FIGURE ~L NOTCH SENSITIVITY CHARTS 'FOR STEELS AND 245-T WROUGHT. ALUMINUM ALLOYS SUBJECTED TO REVERSED BENDING OR RE
VERSED AXIAL LOADS../
'Quenched and drawnsteels (Bhn>200)
. . 1\,
Annealed steels (Bhn>200)
A;hlminum alloys
0.04 0.08 0.12 0.16
. . Notch Radius, r,. in .
FIGURE 4a, NOTCH SENSITIVITY CURVES FOR MATERIALS IN REVERSEDTORSION.
•
o,. ,
....
"....
.:
:,-
~,.'
....
";
..,
.~.:
~
, ..."'
:
,.
.....
-:
o
".0 .0
.\
':",
-'.
.-
.'.'
~:
•>
"....
Ben
din
gO
utp
ut,
div
e\
51
015
2025
3035
40~:
t..:P
FIG
URE
43•.
BEND
ING
INTE
RACT
ION
INTO
TORQ
UE-
MAC
HINE
#3
,JU
NE19
69.
'.:-
"'"
.
"
.~
r"••'"
•
45
403
530
.~
..
".":-~
.
.'.....
:",
25
~....
."
....
..
....
,..'
,
':.
.•..
.
20.~,..
'~'"
..'
15
,I"'
i
.1
05
SO·
45
.40
> "".
"d ~·3
5·
..~
'..:
:s Po.
..
~ ::s 030
'Q) ::s 0'.
M 025
E-t
20 15
10 5 0,.. 0
Pan
Wei
ght,
1b
s.
FIG
UR
E:4
4.TO
RQ
UE
OU
TPU
TV
ER
SU
SP
AN
WE
IGH
T,..
MA
CH
INE
#3
,JU
NE
19
69
'•t:: (j
,)
:..' >
;';"
'J_
....
.~
...'-
'
!
.! 114,
I,
,INSTRUMENTATION..-------'----_._..--- -
,~--------'------'-Part-·---·--------_.~--- - Catalog,:';;:=.-Name"::'-" ~:Number: . 'Manufacturer' 'Numbe'r ":~ Remarks.,:::-- ~< ---
A;,;LMetalfilm 321+B,...190, The Budd Co.~-strain-- -.-----,-,----Instruments
~gages with Division,leads _ _, _", _:,:..
BG2400 4 Bending momentshank of toolholder
, '
..~
'.".rr ,' ...":i,"!-
Bending momentgroove of specimen
'Torque-shank of'tool holder
2
2
".. :
" II
. ,':: . 'i
, " II,.
. ".
-~ : ...,"
. ~ .
.~,,--, -..-
II
Metalfilin '~traingages with.leads
Metalfilin- ~ ':3 X 4 '~.-
strain ,- M15E-240, ~,_ '. 'gages with 'r,.. ; "
leads, ' ;0:',' • ,'-:
:::!.i. :.: ':C6-121~1 "R2VC
~.,.Slip ,AJ-8005--,Breeze Corp- 66SR\,'rings 'andi, A8 ,.oration~ Inc.~hrushes ' '-.. , .
1 Transfer of, 'data
'Visicorder 906 C-l ,~,Honeywell , D-:2009 1, With grid linesystem, 14 magnetic assemblychannels
'Amplifier 119
Galvano':", meter
M1650 ," II D-2007
D':'2005
6,
6
0-5000 cps
Carrier andlinear/integrating system withcarrier channels0-5000 cps
,'.
----_ ...- - -__~ ..-. -----.~--"f"..--~ ..--". : - ~. -... - -- - ~
" ' "
TABLE 2
_m,,,,,,,.~';:..,.n,,J.'O.oLHOLDEROUTPUT VERSUS -PAN WEIGHT DATA,;'J.):~:"";':::: .. " .. "FORBENDING BRIDGE CALIBRATION
.:.r-·t}~:~ :.:._~' ..- _ -'-'
115
MACHINE '#1 MACHINE #2 ... MACHINE #3
Data Too1ho1der Pan Too1ho1derPoiti1:---. ":Output· ""'--'Weigh1:"'-""O,itput
Number~'" 'div. . '·llis. . .div•.
PanWeight'.1bs •.
Too1ho1derOutput
dive
PanWeightlbs.
1"2--:2-------0-.--·-·---·3.5 0 2.4
.- .. c: "f ",: •
2 2_~3. 9 30.81 6.8 10.27 14.4... ""'.'
3 .:~5. 6 61.62 10.3 20.54 23.6' . .:::- ..
4 17.8 41.08 13.5' 30.81 '18.6'L.- :: '- .. ,
5 . 9.9 20.54 17.2 41.08 22.4......-.-
6 21.3 51.35 21.0 51.35 7.0 . ,
7 ,. 5.8 10.27 24.9 61.62 10.8
8 2~1 0 20.8 . ~1.35 3.0-... - -
9 9.7 20.54 17.1 41.08 10.7
·10 17.3 41.08 13.4 30.81 22.1, .
11 25.1 61.62 ·10.0 . 20.54 18.5-..'
12 21.2 51.35 .6.0 .10.27 26.1
13 13.3 30.81 2.1 0 14.4
14 5.8 10.27 6.7.,.._.
15 2.0 0 2.• 7..16 5.6 10.27. 18.1
17 ... 17.1 41.08 25.8
18 ."' . 24.8 61.62 6.5
19 13.2 30.81 14.3..
20 21.0 51.35 . 10.4
21 9.5 20.54 21.8
22--· ...--.--. 1. 7 .. ·..·.. ··--·0 "2.5
Visicorder Calibration Resistances:Machine #1 - 500 kfl at 24.9 divisionsMachine #2 - 500 kfl at 25.2 divisionsMachine #3 - 500 kO at 25.0 divisions
o30.81
61.62
41.08
51.35
10.27
20.54
o20.54
51.35
41.08
61.62
30.81..
10.27
o41.08
61.62
10.27
3.0.81
20.54
51.35
o
r .
',.7'i-'.
J .~ :,
!: :.~ ....,..
; .
,
I
!'.I,I .i.iI·Ii
.' .1: '
.,, .
,:{
!
; ...•
TABLE 3
TOOLHOLDER OUTPUT VERSUS PAN WEIGHT DATA IN INCREMENTAL FORMFOR BENDING BRIDGE CALIBRATION·IN TERMS~£.CHANGE
IN TOOLHOLDER OUTPUT DIVISIONSFOR 10.27 LBS. OF PAN
WEIGHT
Data Point:·· .Machine #1 Machine #2. ' Machine #3, .Increment ," . . . . . . . . . . . . . . ......
, 1-2 ' 3.90 3.3 4.00..'·2-3 3.90 3.5 3.96
3-4 3.90 3.2 3.85
lJ-5 ... 3.95 3.7 3.80
5-6 3.80 3.8 3.85
6"7 3.87 3.9 3.80
7-8 3.70 4.1 /, 3.90
8-9 3.60 3.7 3.85
9-10 3.80 3.7 3.60
10-11 3.90 3.4 3.60'.
11-12 3.90 4.0 3.80
12-13 3.95 ' . " 3.9 3.'90
13-14 3.75 3.85
14-15 3.80 4.00
15-16 3.60 3~85
16-17 3.83· O' 3.85I'
17-18 3.85 3.86
18-19 3.86 ·3.90
19-20 3.90 3.90
20-21 '3.83 3.80
21-22 3.90 3.86,
Average ' 3.838 3:683, ' ,3.850,..
116
. . ~
, .~ ' .... -,
117I,
I) . ,"
TABLE 4
CALIBRATION SPECIMEN GROOVE STRAIN GAGE. OUTPUT VERSUS PAN WEIGHT DATA
. . . . . . . . . . . . .... . . . . . .
.. MACHINE, #1 ' " MACHINE· #2 MACHINE #3
Data Groove Pan Groove Pan Groove PanPoint Output Weight Output Weight .Output Weight
.__Number lJin/in Ibs. ,dive , . ' .. Ibs~, , .. lJin/in, ' .llis.
1 39670 0 0.1 0 39655 0
2 40190 30.81 " '2.3 10.27 ' 40174 30.81
3 '., 40703 61.62 4.6 20.54 40695 61.62 ,.1.' "'" .. ,.~.
! 1 4 40357 ' 41.08 6.9 30181 40346 41.08
5 40015 20.54 9.4 , 41.08 40518 51.35:1'::,
6· 40539 51.35 11.6 51.35 39825 10.27 . ..... . ~.' -: !: ,... .''; ",
.. : ... #'~,,!-
7 ~9838 10.27 '14.2 61.62 39998 20.54
8 39668 0 11.4 51.35 39655 0
9 40013 20.54 9.1 41.08 39998 20.54
, 10 ,40357 '41.08 7.0 30.81 40516 51.35
11 40697 61.62 4.8 20.54' ' 40345 41.08
12 40527 51.35 2'.5 10.27 40785 61.62
13 40182 30.81 0 O. 40172 . 30.81
14 39834 10.27 39825 10.27
15 39661 0 39655 0
16 39834 , 10.27 40345 41.08
17 40351 41.08 40687 61.62
'18 40691 61.62 39824, 10.27
19 40181 30.81 40171 30.81
20 40522 51.35 39995 20.54
21 39005 20.54 40514 51. 35 .
22 39661 o· .. 39654 0
Visicorder Calibration Resistance ,:
Machine #2 - 30 kn at 25.0 divisions
TABLE 5
REDUCTION OF MACHINE #2 DATA GIVEN IN TABLE 4
118
L. .,.Change in Too1ho1derOutput in Divisions--- ------_.._-- -- ..
" . per 10.27 lbs. of Pan... '.'~ :~.': .•.: .. ~ ~' We.ight· .
; .....
~ :..
.:.:
2.2
2.3
2.3
2.5
2.2
2.6
2.8
2.3
2.1
2.2
2.3
2.5
:/
, .
--------..---_., ..
::~:::~.: .. :.~ ..;~:~: ~~
.... 4 • .~ ~ .~ "'._
' ...3-:-4"-i '.~. '.~ .. ~". <' :
.. 4.,.5
.5-6.~:: :_: ': - .
6-7
.. .7-:-8.~......~ --....-
·---,~-_:_·.:.,-S-9.-~~---·.--.~_ .... ·.
9-10' ':':'~ ,
. -10-11---,-~-----_._·--, .....
11-12
12-13
. ---------,~-
-.
Aver,age 2.358 divisions
' ..'-
TABLE 6
REDUCTION OF MACHINES #1 AND #3 ..DATA GIVEN IN TABLE'4
....... ','
MACHINE #1 ' , 'MACHINE #3
Pan Average - ,bStrain Average bStrainWeight Strain Strain
--, _.1bs. llin/in -}Jin/in, ' , ,pin/in, , pin/in
0 39666.3 '169.0 39654.7 170.0
10.27 39835.3 175.7 39824.7 - 172.0..- '.
20.54 40011.0 173.3 ' 39996.7 , 175.6
30.81 40184.3 170.7. 40172.3 173.0
41.08 40355.0 174.3 ' 40345.3 171.7
51.35 40529.3 170.7 40516.0 173.Q
61.62 40697.0 40689.0
-Average 170.7 172.7
I.
. "
119
i. ,1·, . ;
, ;
. "' ..... ..J..,
TABLE 7
TORQUE.OUTPUT VERSUS BENDING OUTPUT DATAFROM TORQUE INTO BENDING INTERACTION CALIBRATION
> .
; .
....." . . ............. ',' ....
. .MACHINE. #1 MACHINE. #2 .... .MACHINE #3
Data Torque Bending Torque Bending Torque BendingPoint . Output Output Output .Output Output Output
Number· dive dive dive .. .. div. ....... .div. dive
1 B.B 11.6 5.6 17.62 41.9 10.0 14.9 IB.33 8.8 11.8 23.8 18.54 '41. 7 10.2 32.4 18.95 33.9 10.6 40.4 19.06 25.7 11.1 . 32.5 18.67 17.4 11.6 23.9 IB.3 .8 B.8 12.0 14.9 18.09 . 17.3 11.5 5.7 17.6
10 25.3 11.3 14.9 17.9 . SEE DATA11 34.0 10.8 23.7 18.1 REDUCTION12 41.9 10.4 . 32.3 18.313 8.8 10.9 40.3 18.8 IN SECTION14 32.4 1B.5
. 15 23.8 18.2 IV.-B~3
16 14:9 17.917 5.6 17.618 5.3 12.919 14.5 13.120 23.4 13.321 32.2 13.6
. 22 40.2 13.823 32.2 13.624 23.5 13.525 '14.6 13.226 5.3 12.8
Visicorder Calibration Resistances:Machine #1 - Bending 500 kn at 25.0 divisions
.Torque304 kn at 45.0 divisionsMachine #2 - Bending 500 kn at 25.0 divisions
Torque 30 kn at 44.7 divisionsMacaine #3 - Bending 500 kn at 25.0 divisions
To~que 304 kn at 45.0 divisions
, ,".,,"
" i. J
..., I
"'.;. '.
121
TABLE ' 8
rOOLHOLDER VERSUS GROOVE BRIDGE OUTPUT DATA. FOR"QUASI:-STATIC CALIBRATION OF GROOVE STRESS
MA~.l!~~~ __.#:!:., ._- __. MACHINE #2---_._--_.
,43.9 31~7 34.7
39.9 29.3 30.0
35.9 26.4. 27.7·
, 32.7 23.9 23.7
, 29.2 21.5 20.1
25.7 18.9 '23.1
22.0 16.4 27.0
25.8 19.2 30.0
28.8 21.3, 33.9
33.3 24~3\ ... 36.2,
36.1 26.7
40.2 29.3
44.0 31. 7
i:~', \
j
I. !
['j'{ 1
i'
,.
.. :.;:• lil';' .;.; ~
. :".
," :,.~ ~::'::':'..:'-~~_"-,_• .1 ...
MACHINE #3
Toolholder GrooveOutput Output
. .div .. .div•
25.0 26'.8
28.6 ,41.8
32.0 47.3
39.1 57.0
21.6 31.9
25.1 36.3
'28.2 40.9
32.4 46.2
40.0 56.0
21.7 31.3-
25.2 ' 36.3
28.8 41.3
. 32.0 46.6
',35.3 50.3
3Ll
'36.2
41.3
45.3
56.0
52.8
5p.7
44.6
. 40.6
34.3
30.5
34.8
39.9
43.6
49.3
52.2
20.0
23.9
27.2·
" 30.3
37.6 '
16.2
18.6
21. 7...........
24.2
26.1 -
Groove Toolholder GrooveOutput Output Output.. div·.·, .. , .' .. diVe '~" , , ,div..
2 25.1
3 .' ..29.1-l~2,:::: :'-;,'.~ . :"
4 33.1
, 5 36.1
~ ~_.__ 40. _~_'_..:.__~_~_.:3 :_._--- 36.6
7
8
9
10
11
12
13
14
15
16
i718
19
Data. ToolholderPoint Output
Number ... ' . ,div •
Visicorder Calibration Resistances:Machine #1 Toolholder 200 kn at 24.9 divisions
Groove 22.22 kn at 24.9 divisionsMachine #2 Toolholder 190 kn at 24.8 divisions
Groove 11 kn at 24.6 divisionsMachine #3 - Toolholder 190 kn at 25.0 divisions
Groove 11 kn at 25.0 divisions'
. "
, /
I
TABLE' 9
TABLE 8 VISICORDER DIVISIONS CONVERTED TO.STRAININ TOOLHOLDER AND SPECIMEN GROOVE
122
, , .• ,1.•
6Nvis 6EToo1ho1der 6NvisToo1ho1der groovediv. , ,pin/in div.
Machine #1
Machine #2
Machine #3
1 :,'
22.2
40.0
10.7
.,
,I I
.,
32.85
62.40
16.55
15.75
59.0
15.0
:-~ .
M.groove
pin/in
1620
3145
786
1:;-;,:
. j:'., t'
I'i'
.1•!
t .'.~
i. i.
'.
,i
TABLE 10
TORQUE OUTPUT VERSUS PAN'WEIGHT DATAFOR TORQUE BRIDGE CALIBRATION
," 123
MACHINE #1 MACHINE #2. .MACHINE #3 . .""'.
Data Torque Pan . Torque Pan Torque PanPoint Output Weight Output Weight Output Weight
Numbeza div. lbs. . .div•. .... l.bs •. . div•. lbs.
". ",
1 8.8 0 5.6 0 5.4 - 0·2 41.9 41.08 14.9 10.27 13 ..2 10.27
3 8.8 0 23.8 20.54 21.1 20.544 41.7 41.08 32.4 .30.81 28.6 30.815 33.9
_.30.8. 40.4 41.08 ' 35.9 41.08
6 25.7 20.54 32.5 30.81 28.5 30.81 .... ____ ":1., ..,~7 17.4 10.27 23.9 20.54 20.8 20.548 8.8 0 14.9 10.27 .. 12.9 10.279 17.3 10.27 5.7 0 5.1 0
10 25.8 20.54 14.9 10.27 5.2 011 34.0 30.81 23.7 . 20.54 13.1 10.27
· 12 41.9 41.08 32.3 30.81 20.9 20.5413 8.8 0 40.3 41.'08 28.6 30.8114 32.4 30.81 35.6 41.08
· 15 23.8 20.54 28.3 30.81.16 14.9. 10.27. 20.8 20.54 .17 5.6 0 . 12.9 10.2718 , "
5.3 0 5.4 019 14.5 10.2720 23.4 . 20.5421 32.2 30.8122 40.2 41.08
· 23 32.2 30.8124 23.5 20.5425 14.6 10.2726· 5.3 0
Visicorder Calibration Resistances:Machine #1 - 304 kn at .45.0 divisionsMachine #2 - 304 kn at 44.7 divisionsMachine #3- 304 kn at 44.5 divisions
MachineNo.
2"
'3
TABLE 11
, TORQUE REDUCTION TABLEFROM THE RESULTS OF TABLE 10
Point Point 6W 6Nvis
6'[ , 6'[ .A B lbs. out True
.... ",' ... . , , ,', , , , , ,dive ' ,psi, , , ,psi,
(30.(),'33.0) (10.0,17.0) ,20.0 . 16.0 513 " 452
"
(40.0,40.0) (5.0, 10.5) , 35.0 29.5 943 ' 792
(35.0,31.5) (15.0,16.5) 20.0 15.0 480 '452
124
Slope
I<.r
0.882
0.840
0.944
-,:.
. ,~
"
_. ... :.•. ~ •..,..._ ..~ ...... _.a:.
J
TABLE 12
BENDING OUTPUT VERSUS· TORQUE OUTPUT DATA FOR BENDINGINTO TORQUE INTERACTION CALIBRATION
125
. . . . . . . . . . . . . . . . . . . . .
MACHINE· #1 MACHINE #2......
MACHINE #3
Data Torque Bending. Torque Bending Torque Bending *Point Output Output Output Output Output Output
Number div .. div•. div. . . . . :div•. .... - .. . .div • div •
1 ~0.1 0.2 0 2.7 -0.3 2.42 +0.5 17.3 0.8 21.0 -0;5 9.0 .,
3 -0.1 0.8 0 3.2 -0.8 20.14 +0.4 17.1 0.1 6.9 . -0.3 9.15 +0.2 8.2 0.3 10.5 0 1.1 ,,": ;
6 -0.1 0.5 0.6 14.1 -0.1 8.97 . +0.1 8.0 0.7 17.1 -0.7 20.2 . ,
I ......._ ....
8 +0.4 16.9 0.9 22.6 -0.3 9.09 +0.2 .7.8 1.0 25.2 +0.1 1.1·
10 -0.1 0.3 0.8 21..6 -0.5 8.911 +0.1 7.7 0.6 18.0 -1.0 20.212 +0.4 16.5 0.5 14.4 ~0.4 8.813 -0.1 0.1 0.3 10.7 +0.2 0.914 0.1 7.1 -0.4 8.815 0 3.3 -0.9 20.016 .)..0 21. 5 -0.3 8.917 ·1:2 25.2 +0.1 1.118 1.3 29.0 -0.4 8.819 1.5 32.8 -0.9 20.120 1.7 36.0 -0·.3 9.021 1.8 40.022 2.0 44.0 ..
23 2.0 . 44.224 1.8 40.525 1.6 36.026 1.4 33.227 1.2 29.528 1.0 25.829 0.8 22.030
*A11 bending data appears at 0.2 amplifier attenuation, therefore a 5/2. .correction factor is required.
Visicorder Calibration Resistances:Machine #1 - Bending 500 knat 25.1 div.; Torque 304 kn at 44.9 div.Machine #2 - Bending 500 kn at 24.0 div.; Torque 350'kn at 44.8 div.Machine #3 -Bending 500 kn at 25.0 div.; Torque 350 kn· at 44.8 div.
" .
. TABLE 13 .
. BENDING OUTPUT VERSUS TORQUE OUTPUT DATA AT HIGH TORQUE LEVELFOR BENDING INTO TORQUE INTERACTION CALIBRATION
126
Data PointNumber
. ,. MACHINE #2
Torque Output Bending Output. .",div ~ . .. " , , , , " , , , , , ; 'div • .: . '.
123'456789
10111213 '141516171819
·20,21222324252627282930
25.2 9.425.3 11.825.5 14.125.6 16.325.7 18.725.9 21.126.0 23.425.8 21.125.9 18.8 , ... _u
25.6 16.625.5 14.225.4 . .J 11.825.3 9.525.5 13.525.8 21.026.0 23.526.1 25.826.3 28.026:4 30.226.5 32.826.7 34.826.5 32.826.4 30.5,26.3 28.026.2 25.826.3 . '28.0
,,26.2 25.826.1 23.525.9 21.025.4 13.7
Visicorder Calibration Resistances:Machine #2 -, Bending 500 kn at 25.0 divisions
Torque 350 kO_at 44.8 divisions
TABLE 14
AXIAL OUTPUT VERSUS BENDING OUTPUT FOR AXIAL TOBENDING INTERACTION CALIBRATION
MACHINE #1,
127
Data Point,Number
Bending Level " Groove Bending, ,'div., , .. , .. , .Mean ' ' div. '
1
2, 3 '
4
5
6
7
8
9
10
11
12
13
20.8
'24.2
27'.9
31.5
35.0
'38.7
41.9
19.6
42.2
19.7
,41.9
19.4
41.8
:\
-0.2
-0.5
-0.5
-0.6
-0.9
-1.1
-0.8
o-1.4
+0.2,
-1.0
+0.3
-1.0
Visicorder Calibration Resistances:
Machine # 1 - Bending (toolholder) 190 kn at 50.0 divisionsBending .(groove) 290 kn at 50.0 divisions
. - '.
TABLE 15
AXIAL OUTPUT VERSUS TORQUE OUTPUT FOR AXIAL LOADINTO TORQUE INTERACTION CALIBRATION
I' MACHINE #1
Data Point Torque Level 'Groove Axial
,Number, ' div. Mean dl.v.... -.- . - ....
1 0 -0.6
2 11.8 +0.33 ' 36.7 +1.8
4 27.1 +0.8
5 1.5 .-1.2
6 40.0 +1.5
7' 21.5 +0.5
8 7.8 -1.5
9 1.1 -1.9
10 31.3 +2.0,11 41.5 +0.8
12 27.6 +1.0
13 15.6 0
14 0.5 -1.8
Visicorder Calibration Resistances:
Bending (groove) 290 kn at 49.8 divisions
Torque 304 kn at 45.0 divisions
,'--
128
,-i
!.I
:. -
(,
J,iifiIi,ii-I,i ,
I'
129
TABLE 16
.THE CALIBRATION COEFFICIENTS FOR EACHRESEARCH MACHINE
Machine .. l<aGR :KGR- TH .' ..~ K.r/B ~/T
1
2
3
.. 0.967
0.902
0.972
0.0203 .
0.0198
0.0215
0'.882 0.0548
0.840 0.0343
0.944ft 0.0000
0.0298
0.0410
-0.0170**
,i
i.• 1. i. !
....,.,' II, .
. ' ~. I,.· ...1~ \. __ .
*After June 1, 1969
**After June 1, 1969
= 0.842 because of gage replacement
'~/T =-0.0238 because o~ ~age replacement
.III!I,.II.
.. ,psi, , , , , , , ' ,div.
'i :,TABLE 17
TCF:.QLC /::~:~STRESS LEVELS AND VISICORDER DIVISIONSFOR"r = 3 AND MACHINE #1s
.. '
Alternating Bending Calibration Mean ShearBending Divisions Div/Resistance . Shear Stress DivisionsStress .-'.--.. ,- --.-.- - ---- -.".-..,..- -.-.-- .
psi div. . div/kQ
130
CalibrationDiv/Resistance
div/kn
154,800
121,800
104,350
86,800
77,700
43.5
34.2
44.5
37.0
33.2
50/125
50/125
50/190
50/190
50/190
29.800
23.400
20,080
16,700
14,950
19.4
15.3
13.1
10.9
9.8
45/304
45/304
45/304
45/304
45/304
iTABLElb
TORQUE OUTPUT VERSUS PAN WEIGHT DATA FOR NEW TORQUE GAGES,MACHINE #3, JUNE 1969
131
. Data Point TOl....que Output Pan WeightNumber .div. dive
1 3.3 0
2 13.1 10.27
3 22.6 20.54
4 31.3 _30.81
5 40.6 41.08
6 31.9 .'30.81
7 22.6 20.54
8 13.2 10.27
9 3.4 /
·0
10 13.2 / ':'0.27
11 JJ. '7 .',54
12 31.9 30.81
13 .40. '! :';·:"09
14 31.9 20.54
15 22.7 '" ~ -4~t: .. :>
16 13.3 1.0.27
17 3.4 ,.,v
......- ...._-,.._.._...
Visicorder CalibraLion Res~sLance:
Machine #3 -304 kfl at 44.8 divisions
,'c.:.-·- _
~~ ..:.. - .
," TABLE 19:: .:
BENDING OUTPUT VERSUS TORQUE OUTPUT DATA FOR NEW TORQUE GAGES,MACHINE #3, JUNE 1969
,.".: l-: -'-..
Data Point ~~BenC!ing Output Torque OutputNumber .div. div.
c' . --c· c·
1 :- .. 4.4 0;.. c
2 - 7.9 : -0.103 .;;: 11. 5 -0.204
c15.1 -0.40-~_:. :::
.<, -::~ ~
5:..., . ,
18.6 -0.45.... ~-.
6 22.2 -0.·707
.<~25. 8 -0.75
<
8'.- , 30.0 -0.90<" ':
9 .. - 33.5 -1.00-10 30.0 -0.9011 "'- .26.0 -0.7512
,.22.5 -0.75":. -- c
13 :- - 19.0 -0.60,-
14 15.4 -0.4015 11.9 -0.25
.C ;.
16 ., 8.4 -0.20..17 ,. i..- ~ ,
4.8 0,-. '.
18 4.9 019
~ ,33.5 -1.0020 5.5 021 ,. -- ,.. c :33.5 -1.0022 - 5.3:'· 023 33.5 -1.0524 5.5 0
Visicorder Calibration Resistances:Machine #3 - Bending 490 kQ at 25.0 divisions
Torque 304 kn at 45.3 divisions
.,'"
·132
I·
APP
END
IXA
.DA
TARE
DUCT
ION
RESU
LTS
FOR
r=_
..
...,
.s·
TfS
T"
!;PEC
IMEN
"AC~Jr~E
'PA
NR
CA
lN
eAL
N'I1
5H
eAl.
NCA
\.N
YU
tlfNDI~G
S..
EAR
STR
ESS
Nn.
/Ii.O
.N
O.
WEI
GH
TtlE
Nnl
NG
!!EN
DIN
G.
AEN
l1IN
GTO~QUE
TORQ
UE
TOkQ
UE
5T~ESS
STR
ESS
kA
TIo
1?5
166
~2'
Jl~ool}n.o
'~0.211
~9.30
-n.o
-0.0
0-0
.00
IS3~8,.2
o.t
I1N
F'II
II.
12
"20
4?
299
00
80
.0~O.IO
2H
.20
-o.n
-0.0
0-0
.00
1~44Q~.2
lI.n
INF
IN.
13~
no;
?29
"'00
110.
04
9.7
027.~0
_O
.tl
-0.0
0-0
.00
15
13
1R
.6(l
.1l
IIII
FIII
I.
13
'13~
?2'
1'1
00
00
.05
0.0
02
7.9
0-o~n
-0.0
0-0
.00
1~31SO;.4
o.n
INF
lN.
13
'1
9"·
22~
90
no
n.0
49
.4n
11l.
00-o
.n-0
.00
-0.0
01
55
57
1.2
n.1l
INFI
III.
\1
3)
220
229
·i,l
oooo
.o4
9.9
n21
1.20
-0.0
-0.0
0-0
.00
15
51
1,.
50
.0IN
nN
..-,
I~4
)93
?29
"'0
00
0.0
49
.60
18
.00
-0.0
-0.0
0-D
.OO
15
49
43
.9o
.nIN
FIN
.
13~
191
2'
29'1
01)0
0.0
50
'.40
28
.40
-0.0
-0.0
0.-
0.0
0)5
46
6".
(1n
.tI
INF
IN.
13
"1~3
22'
1'IO
I)O
tl.O
50
.20
28
.40
-0.0
-0.0
0-0
.00
15
52
79
.00.
1}IN
FlN
.
42'"
34
21
3015
01)0
0.0
~0.20
51
."0
-0.0
-0.0
0-0
.00
IS3
2H
·.2
0.0
INF
lN.
.4""
341
1.3
0IS
OO
OO
.o5
0.1
05
0.7
0.-
0.0
-0.0
0-0
.00
15
08
44
.3o
.nIN
FIN
•
4""
36
51
301
50
00
0.0
50
.30
51
.55
..15
27,,
).!'
>IN
F'IN
.-0
.0-0
.00
'-0
.00
o.n
.'.~EAN
RF.N
DTN
GST
RE
SSIN
bRO
OV
E•
15
31
45
.1P
ST
.
's-j
[.:-
nEv~
OF'
REN
DIN
GST~E'SS
'iN·~l'ioovi'~·'-i54".-7."PSI~
..--
-.-...
..
..... w w
t\j~
AI.
.JlEN
U!N
li
APP
END
IXA
.M
TAR
£DU
cTtD
MR
£SU
LTIl
roR
I'll
TEST
5P~rIMfN
M.eHI~f
~AN
HeA
L_
_N!
l.~_
._..
_N
il!
tlO
.."'E.IuH.rl:lt"nl~'G
e_
,
~I/l~
HFN
IJIN
G..e
~..
TO
HIJ
W.
Ne~L
TOH
UU
t.N
I/H
\To
"UU
EJi~~~I"If!
SfH
F.S
S§tj~~~
SH
IE5$
§ff'l~§~
IIA
TIQ
-0.0
0.1
02
93
4.0
·-o.
uo·to,~.,.0
1"
.
11 1"
.1~
I" r...·
.
2"
6..
~"l ." ..,
11 99
? , " :>
IS It.
15 It> 15
16
20,1
0001
1.0
2<
,10
00
0.0
29
00
00
.0
2~oiloll.u
C!'I
OO
OO
.o
29
00
00
.0
39
."0
"1
.10
38
.20
"0
.00
JtI.
20
'0.0
0
51
.00
..9.e
o
51
.00
51
.00
151.
00
'S1
l.S
O
-n.u
-o.n
-0.0
-0.0
.n.n
-I].
n'
-0.0
0
-0.0
0
-0.0
0
-0.0
0
-0.0
0
-.,.
00
-0.0
0
-0.0
0
.-0
.00
-Cl.O
O
99
79
9.0
10
32
12
.0
10
29
34
.0
10
86
"...
.1
o.n
INF
IN.
0.0
INF
IN.
0.11
INF
IN.
0.0
INF
IN.
0.0
INF
IN.
0.0
1N'1
N.'
-0
.0'
-0.1
10
'
....
.;;1
f:{1
....
.;Ir:
f(~
.
.'·0
.00
''1
05
14
1.9
n.n
1N
F1N
.
«!.II
11-4'
I~,
INF
aN.
IN"N
.
IN'I
N,
n.o
INI"
1N.
0.0
.IN
FIN
.
0.0
INF
IN.
0;0o.n
(10n
0.0
"IN
nN
.
-(f!
~"-
tt-l
r-I~
'f""
:.-
-..'0
.0IN
'IN
•
0.0
IN;'
IN.
l\ll'll~,,~
t01
39
4.0
106.7~."
10
13
94
.0
1110
",,,,
,.9
10
U)0
.t\
Ia.,I
1'\
.t,,
.",
101Z
7<;1
.9
iO~r.?4:~
10"t
li/"
'1I
10
70
11
..1
'
-\l,I
l11
-0.0
0
-0.0
0
-0.0
0
-0.0
0
-0.0
0
-(1
.00
-0.0
0
-11
.00
'-0
.00
..·...
;c·;&
H·
-0.0
0
-0.0
0
-0.0
0
-'h
l)1
I
-1I.
0U
-0,0
0
-0.0
0
-n.o
o
"·0
.00
,-O
,ClO
-II,
ll
-0.(
1
-0.0
-0.0
-n.n
-0.0
-n.a
·0."
•"0
.0
.-n.
0
50
.00
....,
I\'l
~l.oo
49
.50
'.
IU"'
(I
'So.~o
.11
"'0
!l0
.l0
'0.4
0
·!II
I.'3
0
51
.10
'siu
M
J8~.0
JI!.
"O
n.'
0
a~''
'11
JIi.
70
"0
.00
39
.40
:'Ul.
90
jif:I
:tH
"o.n
oJl
looi
l
'0."
0
i'll1
onn.
O
2'1
00
00
.0
2<
,10
00
0.0
P,1
;\lf\
htlo
II
j~onul1.o
2"'
00
00
.0
2'1
00
00
.0
t·/C
r(r(."
(f:l
i
~"'O(",n
..0
""'0
(10
0.0
'VO
Oll
O.O
11'#
111\
(10.
0
1~ 16
l~ Ie III
tll
III 1& I~ 16 t, M
~? , , It t ,"
M ,. U 11 I'fLl'
:n AS
II,)
~':l _II
no
10
4
H 17 II.,,,,n.,P \it 'f " U
....
p,"
--.."
_..,
...1(
.1'0
"'.
..nl"
'l)"Ht.~&
Iii"~oovl·.'
In"3
"0.:
Ip
SI
•.
'!I'
O,
."ot,
tl~
""'''
OIN
l\"I
,.~t
lllo
iN(.
III'
)OV
to..
'111
11,3
0"!
il..
•."._
,,
t'"._
_.;
_,.""
1..u.
~.,."','\,
~,
,",
.'1"'
1'.
."
~,....
•..
<!'.
~;
'.~
....
....
...,
.._-
...... Ul
VI
__
_~,~,_,."._
_,,,,,~,_,_,,~,"_'..
.--._
_~
.~_
....,J"'-
'-'
"'"
Ir---..,--.,;,;~~;..-._.,...
.I
Rep
rodu
ced
from
I,bes
tav
aila
ble
CO
pY.
yom
.:~.
-_..--
..,...:
-.~.~,'."
....._;
.-.....
..~'-~
.....---
....~....-
.....
TEST
NO
.
APPE
NDIX
A.
DATA
REDU
CTIO
NRE
SULT
SFO
Rrs
=-.
~PECIMEN
M~CHt~E
P~N
MeA
LNC~L
~O.
NO
.-f
lbM
1~ENuING
~ENUING
Nvi
sR~
lIIl
1IlI
IG
Mr,A
LT
Olll
.llJE
.PIC
AL
Tl,H\~IJE
NV
1ST
OH
QU
EM
END
ING
5TH
,;:SS
S...e
.A~
ST~ESS
51R
[SS
RA
llO
---'-'~Ii····_-·····9~
~o.oo
8b
8Z
1.9
'-0 5r
'1\\
'.
5'
_c_·
S'\ ...~ ..... 57
.
~s:. lor
.",
6'
'6"
"6
C
67
1Il
L'
.11
A.
'fI) )4 45
3'5
MJ
'70
'
<'5 31 11
4
:?9
"Ill 48 Rl
.t .1 1
III
.
.lU 10
10
10
lU lu lu lu 10 10 10
10
10
·10 10
2-"
00
11
0.0
l."O
OO
o.O
Z~Oallfl.O
2"'
00
0".
0
C!'i
OO
OO
.o
~'tOl)on.u
l~Utlotl.o
4!'tO
OO
U.o
~O;Oooo.O
7.'
10
00
0.0
iNO
aon.
O
l'lI)
OO
O.O
29
00
00
.0
"''''
00
00
.0
29
00
00
.'0
'NO"OO~O
45
.30
4!l
.f,0
46
.20
4S
.1n
'"!>
.20
"c.)
O
"~.!>o
"".7
0
"1).
(10
1i5.
!'>0
"c.l
0
45
.90
45
.20
46
•.7
0
4'.
90
....
.4
1)
51
.30
5).
30
52
.no
C:;0
.:?0
...1
.70
51
.70
!l11
.1I0
4'1
.00
Si.
l0
5.,
.40
51
.60
'52
.80
\5
1.3
0
5<
'.5
0
50
.60
'
50
.00
-11.
0
-n.o
-n.o
-1).
0
-n.\J
-0.0
-I)
.ll
.n.t
)
_".0
-1).
0
-n.o
"-0
.0
-0.0
-o~o
-0.0
-0.0
-0.0
0
-0.0
0
-0.1
)0
-0.0
0
-0
.00
-0.0
0
-0.0
0.
-1).
00
-0.0
0
-0.0
0
-0.0
0
-0.0
0
-0.0
0
-0.0
0
'-0
.00
-0.0
0
-0.0
0
-0.0
0
-0.0
0
-0.0
0
-0.0
0
-0.0
0
-0.0
0
-0.0
0
-0
.00
-0.0
0
-0.0
0
-0.0
0
-0.0
0
"0
.00
'-0
.00
87
31
1.5
tl67
31.1
'1
15
61
71
1.8
8581
F1.
2
tltl
le"'
.9
86~t-~.3
il60~n.,
84
51
"'.
3'
tl7
3?,.
1
15
54
0".
6
1162
QR
.O·
88
6l!
Q.7
"
87
5/1
4.6
"B6
b7
'i.l
.861
11l7
.3·
/)."
INF
IN.
n.o
INF
lN.
o.n
INF
IN.
0.1'
1IN
FIN
.
0.0
11II
FIN
.
'0.n
.1N
F11
01.
o.n
INF
IN.
n.o
INFl
N.
n.n
INF
IN.
0.0
INrI
N.
'0.0
IIIIY
IN.
.'0
.0IN
FIN
.
0.0
I"'F
IN·.
o.n
INP
'l.N
.
.:o
.n·
INf'l
N•
0.0
INF
I,...
.~. "
-0..
00
·---
1H
l7?.1
l·--
----
'--:
11.0
--iN
'I"'
....
~R
---
6"t1
'
19
30
"1
10 10
4!"'O
OO
O.O
Z"'O
OO
O.O
"'.
40
"3.<
'0
50
.00
41
1.9
0
-0.0
-0.•
0
-0.0
0
-0.0
0
"0
.00
"'8
68
21
.9"
O.l)
IN'I
...'
M~AN
HFNOt~r..~TRE5S
INbH
OO
vE·.
.------~rr".·-"E".
·vF
t'~"'OT"O
Sf~E5S
I'"
bMO
ovE
•
1161
">5.
3P
SI
•
'ib
t.7
3P
51
.·
----
----
-..---
-_...
.-.-'
.._-.
;..-
..•.._
_._...
......
.--'''':'
••
_•••
_••
_••._~
__
"--_
._
-_
••
-"0
,..
.•~
......
.__
._
1-'
w 0)
T£5T
111(
\.
APP
END
IXA
.DA
TARE
DUCT
ION
RESU
LTS
FOR
r s=
-.5s
:-e,
r.IM
fNM
"Ci-
iHE
"'AN
HC
AL
NC
AL
M'.
Nil.
"'F
:lIs
HT
'IS
E.N
oI"'
''t-
EN
OI"
GN
I/15
RF
NuI
NG
Rt:A
LIO
.lUU
E'lC
AL
·To
HU
UE
NV
ISTo
wUUE
FOfN
Ol"1
GST
llfSS
S"'EA~
STIlE
SS'S
TR
ES
SkA
no
1)Q
'
141'
]35 ''''
~ ;II
tl tl
I~OnOu.O
l'iO
OO
(l.O
49
.60
~O.20
3u.O
O
)'1
.40
-1).
0
-0.1
l
-0.0
0
-o.o
u
-0.0
0
-0.0
0
78
63
-..9
1813
"4.0
o.n
INF
IN.
lI.n
INF
}N
.
.'1
4'
'191
17.
tl1
"'0
0(1
0.0
~0.20
;0"'
.20
-0.0
-0.•
on-0
.00
7562
0:;.1
n.o
INF
IN.
14:a
14
C
"'14
11
5
;> ;II
6 b
-l'~00"0·0
IVO
noo.
\)
"'i.
tlll
""'.!)
o2
"'.4
0
01"'
.20
-n.o
-0.0
-0.0
0
-0.0
0
-0.0
0
·'0
.00
1f>
''54
.1
1f>
t>Q
6."
o.n
INF
}N.
n.o
INF
IN.
14
..."4
?6
1'1
00
0".
04
9.4
"3
0.1
0-n
.o-0
.00
-0.0
01'
i21
A.5
n.o
INF
lN.
14
7l;o
C;
7"o
l'to
oo
o.O
49
.('0
""'.6"
-rJ
.O-O
.UO
-0.0
07
11
21
9.2
0.0
INF1
...
"14
11.
';!\
"
"15
1
]"1;
>
'''~
1!'1
9.
141'1
I?!l
? :> .. ~
o o o tl
1"'
00
00
.'0
'
·1'1
00
00
.0
1"'
00
00
.0
1"'0
01)(
1.1l
~0.10
""'.!l
0""'.
90
~<?20
.'-
29
.80
.3".
4n
l"'.
60
.31.2"
-0.0
-0.1'
-0.0
-0
.0
-0.0
0
-O.O
U
-0.0
0
-o..O
U
-n.o
o-0
.00
-0.0
0
-0.0
0
71
3H
.I
19
11
4".
4
17
12
2.0
11
7"9
.0
o.n
INF
}N.
o.n
}I'<
FI"
I.
n.o
l"'F
lN.·
0.0
INF
ltoe
.
-~"---?19'"
.".•..
-...
O'-
"'-'l
"oon
n"0
'·-
.'t.
90
'0.6
0':
...
..--
1).
0··
··"--
-0.0
0'"
--
-0..a
O·--:-1~721.5--:------1l.(l·
__·I
NF
1N
.-,-
..-
.--'S~-·-
'-,4
5':
-"":
---1
.
'-"T~C;-"-""'T,n
..lY
Ooo
o..o
....
~O.O(J
..·.(
l.0
0--
"-1
'11
.1'.
8'
.:..
..-
.-:0
.0·-
·IN
f'IN
."
0.0
INF
IN.
80
96
....2
'18
47
".7
"-.
,o.
n'-
1N
"'IN
.
·0.0
0
'-0
.00
-0.0
0'
'-0
.00
--'
0"0
0
-0.0
-0;1
)
'-0
.0
2'1
.90
31
.20
'30
."0
"4
'1.'
70
.
!In.
to
1"'
00
00
.0'
1'1
00
00
.0tl & e-
i!
";
1..
3..
..1
54
...
·0.0
0.
·1Z
49
?'.
t·..
··"
li.o
}N
nN
.
'---T~7'•..
--..
"7
1
,-_..H
,"""
"-""
l!\&
'
'2
i1..'
.,-0
..
.tl
t."oo
oo.O
1'1
00
00
.0
49
.10
49
.·!l
0
.30
.40
?7.f
>O
..-0.fl
-0.0
-0.0
0.
·0.0
0'
'''0
.00
--1
95
25
·.·1
..•.
...·
.0
;0"·
INF
IN.
-l1o~-"tk~'--'_._~
.._-'
-_...
..-.:
-0--
...
'..
'-""'O
OO
thU
...9
.20
····
i?"'
.ln
.-.-"'.1
1·.;
.o..o
o·..·
···.
.~.oo
·:-··16&~A.-IO-·_-_·_·o.~·_·INF·I"'"'--
'M~AN
Rf~Dl~~
SlH
EsS
l~
bHO
OV
l•
7·(
tl7
3.e
PS
I.
---'---5'fh·-t~\I.·*
t1!:
;"O
I",G
sh~E~S
h.·
Ii>
Ho
ov
E·.
--~,-
_-__._~...
•..•
..
19
05
.1l1
PS
I;.
..... (.l) -.J
_._-
----
--..
.'
t'.'
,\'-
f,r
1',
<
","
,.(.
';;'
,
,11
"!
•.
\..
\,t;
.,Ill
"!~If;
,,;.
1J;'f
'ij
t~'l.
0.~hii
~~
~~iii.
•I~'
f.","
'""t~
i,.'l(
j,:'.
~'~f
,\I.
1~.,t~
."
.•~~l
·11J
,r.H
j~'
~'C
tq,
I,V
i,JI"
f
1.1'l
t,~
,'
1"J',
.l~.
;'l'
·Ii.:
!it
,it.
lf;;,
'i~,
r,~')
1;,
1;;.'
t'j
~•
~;.fl
·'t1
~',;
'.,\1
1'\
_••
0_
"_
'.'._
_•••••
_•••
•_
..,._
_••
_•••
_
APP
END
IXA
.DA
TA,R
EDUC
TION
RESU
LTS
FOR
r s=
-.
.T
(ST
-"S
VE
CIM
""'''
"I·II
,C..<t~·E
PA
r.'
'kC
AL
'"N
CA
L',
NV
lS';
__I
'l.ll.
~~~
....,.
..r:
:O~.
,._
"E.~\i.HT
.~E
."'I
J.IN
G"'
E~.u
l_~,
~__
.,:,~~~I~G
:'RC
At.·
I•
i ,
TO~"'UE
.·.
..··
...,·.·rl(
'.r,
rJC
'AL'
".
To
Ri,l
UE
..~
".
\"l'
",V
IS'\
TO
RQ
UE
.Fo
END
t",G
I,S
TR
ES
S.,..
"'--~"';---
SktARt.~
SlRESS~
ST
RE
SS
RA
TIO
...
:'f'
"-:
-,,-~
.",.
•..
-,Ii
.,--
·""'
14
''"
'l"··
-··
·..·
2!l
2~
0)
------".
,--....
'----
1I!
!..
11r.
-"U
l
'87 17
'20
2
-'1
12
"';> 1 1
Z~
2~
2:::.
~90000.0
,,<iO
Ooo
.'o
"90
00
0.0
3<io
ooo.
0
19
00
00
.0
'49
.4'0
!:I3
.40
,4
9.'
lYO
49.~0
49
.80
'37
.41
)
""'.8
0"4
8.5
0
41
.20
25
.90
;~,\
.0.0
~..'\\
'-n
.o
.~.'tl~n.o
·0.0
·0.1
)
"0
.00
·0.0
'0
-0.0
0
-o.n
o
-n.o
o
-0.0
0
-0.0
0
-0.0
0
·0.0
0
-0.0
0
:64
48
Q.2
;39
99
n.7
,49
tl9
8.e
...
,54
!lS
r..4
61
20
?;Z
-
.o.n
INF
IN.
·'o.n
INFIN~
j"•.
-o.
n.I
N"
1"1_
0.0
IN,.I
N'
o.n
INF
IN.
-12
?"--
"-
-,.1'1,"
"'"-.~-T-'-"---I:!~-
..1
90
00
0.0
'''-
"5
0.2
0'
27
.50
-1')
.0·0
.00
'·0
.00
.'6
44
6..
,.2
-'--
-.
'-'1r~-o
--tN
P'f
NO
--..
.
~-rs"---'-199
_...
"'~'-l'"
~O.O(r
'-"'
-'-2
11
'.'5
0-0
.0..
'-0
.00
'"-0
.00
,·..-·~f66?·~4·---:,-
..'
51
tl9
11
.t·
..'
..0
.0IN
"11<
1•.
.1
4]
.-'
14
9
If-I
''..~
.16
9
21
6
0'0
0;
20
9
11
7'.
1
26
24
2~'
..2~
24
·l<iO
OO
O.O
1"'
00
00
.0
.-1
'10
00
'0.0
19
00
00
.0
19
00
00
.0
"9.8
0'
"9
.60
49
.10
"9.1
'>0
2b
.20
'.2
2.3
0
'24
.00
24
.40
-n.o
-0.1
1
-0.0
-0.0
.
-0.0
0
-0.0
0
-0.0
0
-0.0
0
·0.0
0
"0
.00
·-0
.00
-0.0
0
61
91
1.1
52
90
7.8
·Sl!
,.Z
l.O
1).0
INF
IN.
o.n
INF
lN.
0..0
'IN
n"l.
0.0
INF
lN.
~'7I;-""'--1"8-----J'--"-'-cS
'-"
"-'1
90
1)1
10
'.0
..4
9-'
80
'-'·
'..
..1:!5·~80·
...
··_·_
..·.;O·;ll·~
'''-'
--''0
;00
''''''
'"·0
.00
"'-
-60
96
<;.
=-'
9--
--'-
--O
Ol'
--IN
rlw
•._
-
"'II
l'S
..1
90
'·2
26
l'iO
OO
O.O
50
.00
..--
..7
S·.
·40
-..-0
.0•.-0
.00
-0.0
0.··
·e6
0."
\;6
-··-
----
·-..
O'O
··lN
FlN
•."'-
'l!l~'
.,."'
21
17
"_.
.iJ
-:·
..•·
...7
7..
..··
19
00
0·(
\.0
·'~
o;I
n'.
,-'
-'2
'5.4
0...
..-,
.'-
0'.
0.'
-·..·0
.00
......
....0
·.0
0.
."6
59
14
;'8
-...----
-"0
;1I
.-1
Nf'
lI'J
'".-
,_.
-'"
9il--'-
.,''
Z..._
._
.....'
'2.-
..-.-
.c~'
·--·
----
no00
0·.
·0·-
--4
9;9
0'''--'-C~
.'8
o..--
·-·-
-..o
-;o
-·--
-..t
l·.O
O-.
.:.·
--...
.O-'1
70"--
·en
z,;-
.-.-
---
1T
'l,,
-'lN
rtn
;---
"'1
95
'..-.
...
'lo
R'.
•---
.:...
?2
3'-
-19
00
00
.0_
.~o~tO'
"'2
".4
0'
-0.0
"'·
'--"
0.0
0'
'·0
.00
""
6:"
lc r;7
"-'"
"0
.0-I
NI"
!N,,
"'"
....
-.'..
__
.
ENDU
RANC
E
f\\
....
•.._
._.•_
...;.~
_._r
._..
.__.•_
".•
,...__
•__•
.__.
....•--
----
---_
..._--
----_.._
--~ w ex
>
TEST
hO
.
APPE
NDIX
B.
DATA
REDU
CTIO
NRE
SULT
SrO
Rr s
~0.
7an
dr s
a~.9.
S~ECI~EN
~ACrit~E
PAN
HC~L
NeA
LN
YIS
IIll
l.N
O.
.."lb
Hl'
...l:
."'i
lIM
."F
.:'~
llI/
l.G
Rf,
NuI
NG
,"C
IIL
to""~'JE
~ICAL
TII
H\J
llt:
.N
YIS
T()
H\,l
IJE
..
RE
ND
ING
STIlI
':SS
SHEA
tfSf
Io1t
:.Sc;
stR
ESS
kA
TIa
loll
·1?
354
~~~
t,
·1
l'i
11
1"0
00
/1.0
1"0
00
0.0
:>0.
10
!)0
.3o
;
.7.7
5
''''.6
5
"0,,
00
1).
0
2ll1
l000
.0
.';.
20
.5.1
0
32
.50
3••
30
10
83
"'1
.6
11
20
\4.6
84
67
1.Q
'!l9••••".
.73
9
.12
3
-4
\3 .,<;
.,,,
"'-
'I?
,.,0
,"1
0
:w;
3:'1
5
~C;IJ
.nl!>
~"7
~56
Zl
U 21 1u il 21
1"0
00
0.0
1'1
00
00
.fl
l"'o
oo
n.O
1"'''''0
0.0
1"0
00
0.i
I.
1'",
,,nl
l.o
':10••
5
:In.
IO
..'i.'
Il5
!)".
",n
:OO
.lll
:>U
.4:>
....
90
"".3
0
."'.
45
....
60
4"'
.25
411.
7<;
20
no
oo
.o
?onO
OO
.o
'OO
UO
O.O
?onO
OI)
.1l
10
00
00
.0
?un
OIl
O.1
1
."'.JQ
4<
;.1
'"
.5.1
0
45
.10
.!).
Io
•.,
.30
33
.55
11
24
bn
.S
33
.30
11
Ift o
n.3
,
J2.9
0,
110Z
,H...
O
3".
30
10
9IQ
",.S
33
.60
I1P
3".7
35
.S0
l09
S?;.
9
117o
;,'1!
't.7
~bfl33.';
"s~ya.s
,Wh
3.,
1!\77~5.'"
ft2o
;,71.
A
.7.1
.7.4
.7"1
.10
3
.13
'"
.b9
.1
.-"
7'.
".-"
"o
T
••
•"
-••
•••.
l'lr
ll,-
'•.
..•,
~~
,or
~t
_..
-',•
.fl
'-"·
'1"0
11
/)1
1.0
.'''.'."."O---"-·".T~
.68
0
.73
5
.10
9
.72
.'
~t1149.6
1129
9"'.'
>
'1'0
267.
2II
Ob
ll1
'.t
10
96
0""·
3••
55
J5.5
0
33
.90
'.
11
2t'
tT.t
,·,
U.1
'5'·
'-tt
tt'.
i-3
···
.tl
d0
7tt
.'
4~.;jO
4!».
IfI
4o;.
1!)
?un
OO
O.l
l
ZO
:lO
OO
.O
-..G
IIO
on.o
•~O"lllln.o'
-'4
!).1
!»'
....
6")
.".flO
"".1
5
.-:;.
~.
'0.3
0
':10.
3")
:>0.
7'>
1""0
"".0
1"0
00
0.0
1"'0
0011
.0
fl II
.
u..1
··..
'..
"'n
""l
'
..
A',
.('4
1'''''
\lO
lt"
t..
"""1
0'1
(..
Jt0
n..
.6P
SI.
,','1
".
"t••~~
..~t""
,t"l'~
lei
OJf
tNJv
l:.
1')1
11.9
'1P
SI
•
...,,...
,...~"
"·t..
~I.
.tr
oft)
O"l
:•
_.,.,t
.-."y
.tA
··',c
a;.p
-r,lo
ot",.
I..
""00
""•
IH'~6'.'
It)J.
"'tt,.
••,.~t.
~a
..~l~~S
AaT
IO•
'-.-
-'11
'.rt.
,.n
rIf
,.a"
),..t
~')
Mall
O.
07
tlll
7
•OZ~'!l
'
--_.
'_..~--_._~-_
~..-
--.._
.,
".-.
--,-
---...
_.._-
--_.
_._-
----
--_.
~ (.I)
..-----
__
.'t
o
.J
140
I!!
oXl....,.,
co.0'....
e "II. C0{) ... .'"oC)-'\I'".r
o
...o
«a:......".
000N ~.(7\
• 0 0~ ...........~ "" r")
o
1/·...Xl
'"."..o
~ 0£. 0' •~ 1/"1 0 c.£I ..... ,.. .... . ." .
e.....o..
'",;
N·...orN...'"o
'"o.....
....41·
...o..
:r..,.,'"oXl.:Xl...
'"oC
'"',/1~..
.co...
01'0.....'"oo
..o......
'"~
It·"N•'"......·..~..No
o Cl,/I
o •
'" .c. ..
'"·..41-....lr'
o.0·..
.·~t.-lr'
..a:41·
o..·M•o,.,..'"..
tI"e>...·
0-o
a:
'"'"'"<>
..,•
..·'"..c'".c.
....o.........o
'"•
..
... .c ,.,· . .... " C.. ,., c.. J\ ~IV IV ....0' 0' 0'
o~
oN..
.... ..
... ..~ ~· .;... ". C.
• 0 0tI" tI· M.. '" .... ~ '"... IV ..~. 'l" C
....on:>-:> .~"X
ZO...
"'"Z .,.... ;.,::>zz ...w'".,
'"%'"C,J... .>:xIll'"
<><>·or..
'"o·'"•'"'".1/'1..
o...o
r• ·'"..e.-·'".. ·'"..
Xl :>r e· ... .f'.. . 4
J:.4..
'"e ::>X·It.. 4..
0<><>,. .c ::>.0... .. or.. 4,' ~
....J:l..C:;,vr:lrO..
<> c·O' C>o 0<> <:>.. ..<:> <:>... ...
<:>o
<>oo...o...
o·ooo..<>,.,
e·eoo..o
'"
<>·eee..C>..-
eo
eoo..o...
·C>o::>..o...
C>o:>..<>..,
co
ee::>..e...
e 0o 0
e 0e e::> '0.. ..::> 0... ,..
oo::>..o
'"
'" C-o 0
o. <=e <=o 0c' r:'
J\ '"I\. '"
co
ee::>
'"'"
oo
eo::>c
'"'"
co
ec<:c,/IN
•
....r
o...0..2:,.n.J:l
"Z.....~
...c
...o
'"~o.•
0{)
'"N'"....
.COVI
.~....o
•;,J~
ec')'...:>
. .......'" onQ. ..
0''''o '0
... 0no ...;II Xl::> ::>it: ..
.c...In
?
.......Z
,.'",.oJX
'"
'"....'\I
•
•c..
oil....::>/l
c'o
CooC,.-c.-
... c.l; ...,.... fP,
o 0• 0
Co C= 0o e'" :>,. ,.... -D 41......
... ..< c.. ..
o.C
ex·c..
oc
oe
'"
oe~...
o"..N..
<>•C'"c
c.
oo11
D...
,....
,.-
..,I\...
1/'1...·,.,..
,..
"..'"
'"o.:c'e:>:>
....
....I\.
n-
e..,o..
e
'"o,...
c0:<>..
eo..
eeo,....
..
eit:
o~It
....
o
'"o
oo
COocc,....
.... ,.,
o......'" '"
11...
c.c
o~....
oo
CC>Co,....
o..o
'"It
•
1\
.......I\.
..,.,'"
II...
eoeII
e....·
C>o
C<>eo,....
......-.'"
/l
..
'"-·....
oo
Ccec,....
,..
oe
o<>11
11
\1"....N
<>·Cceo,....
o..·M..
,.
I......'"
o·oeeo,....
/l-
C>1/'1·'"..C>~
~..
N...,.
e
'"·
<>·e-Ooe~...
C\1/'1
o0'..
oo
Cc·
,..
/l...
<>C>·£..
...'"I\.
o~....
'0o
ceeo0'...
...
Ib1\1
I
c....o.....
II...J\-
<>•~
o
'"o,.-
..£·
ooo
o."
,..
It\..'"
oo
. 0
o11
'"-
c.c
o-..
o·CC'"<>
'"-
.. 0
"'C""'Z...
Zl&IZ •"'0t.;Z'..C>.II>
•o~ z x<> ~.:>
;/3.:.•
·,.··
".'
APP
ElID
YX
B.
DAT~
Il!D
l.'C
7IO
NP.
ESU
LTS
l"O
ll"s
=O
.7,\
."ID
t"s
=0
.9.
-TES
T:S
PEeI
MtN
MAC
HINE
-PA~
RCAL
HCAL
NY
!5~
~,
___
NO
.vE
t'3>
IT_
B~NOING
BEN
DIN
GII~NOING
flCA
LTO~O~~_
HeA
l.TO
RQUE
:H
'f1S
TORQ
UEII[NOI~
ST
R!S
SSM
EAR
ST
RE
SS
STR
ESS
RA
TtO
'
--lt
lt--
--"
3ze
.--
---r
tr--
--t8
5-
-tn
i16
-:-t
tti4
i
---%
"13
---_
15
'--
215
Z18
I
--"2
07
----
109
21T
22
t
384
--3
ee
--3
87
390
--»1
----3
92
-393
ot33
in -zoo 217
28
5
363
U7
404
-
441
'16
405
38Z
-n
j-
1 1
----
-1 1 1
----
-I-- I -3 -3 3 1 1 -1 1 1 1 1 1
10 10 10 10 10 10 10 10 10 10
_10 10 1
0 10 10
10 10 -10
19
00
00
.0
1900
00••
1900
01'.0
19
00
00
.0
19
00
00
.0
19
00
00
.0
19
00
00
.0
1000
00-.
0
19
00
00
.0
19
00
00
.0
1'0
00
0.0
1"0
00
.0
19
00
00
.0
19
00
00
.0
1900
00.0
19
00
00
.0
1900
00.&
<
19
90
00
.0
49
.80
50
.10
50
.20
49
.80
49
.eo
50
.00
49
.90
49
.90
50
.00
49
.90
50
.30
50
.10
50
.20
50
.10
49
.10
50
.15
50
.15
.
.9".
81
f-
3Z
.U
34
.U
33
.90
33
000
29
.70
34
.00
33
.70
,4.7
0
35
.30
36
.50
30
.30
-33.
19
34
.90
34
.10
33
.45
34
.20
33.6
0-
33
."
30
40
00
.0
30
40
00
.0
30
40
00
.0
30
40
00
.0
30
40
00
.0
30
40
00
.0
30
40
00
.0
30
40
00
.0
3040
01>
.0
30
40
00
.0
30
40
00
.0
30
40
00
.0
30
40
00
.0
-30
40
00
.0
--3
04
00
0.0
30400~.0
30
40
00
.0
-30
40
0r'
0
•
44
.30
45
.10
44
.80
45
.00
44
.60
44.8~
44
.90
44
.00
44
.50
45
.00
45
.10
45
..20
45
.25
44
.95
45
.00
44
.90
4!?
00
45
.00
32
.59
36
.90
35
.10
36
.00
37
.00
35
.30
36
.00
36
.60
37
.50
35
.60
37
.50
-37
.65
37
.80
37
.eo
-37
.80
37
.40
31
.60
38
.50
-
13
5,.
.,
78
40
8.5
76
73
5.6
15
11
3.2
67
Z1
6.2
77
21
3.8
'
76
66
8.1
TT
Z65
.0
78
44
3.8
81
27
3.0
77
32
4.9
7634
Z.4
78
88
3.8
17
13
6.7
1624
1-05
11
32
4.3
1589
8..1
17
20
6.8
56
81
4.4
63
48
0.5
60
13
3.9
6211
8.9
64
76
9.9
610T
6.1
62
21
3.6
13
52
7.6
1.4
92
.8
T00
95.5
64
59
0.2
64
74
1.9
6485
10'J
65
35
9.3
65
31
3.Z
64
70
9.4
6496
5.-1
66
53
1.4
.14
1
.71
3
.12
9
.69
9
.59
9
.13
0
.7iz
.60
7
.60
8
.66
9
.69
1
.68
1
.70
Z
.611
1
.67
4
.69
0
.615
.67
0-
M!A
_B
END
ING
SlA
lSS
I~
GaO
CY
E.
16
57
5.3
PS
I.
---
-S
TD
.O
EV
.01
8[N
01N
aST
RE
SSI~
G'lo
ovt
..-
Z9
33
..46
PS
t.
;~fA~
TORQ
UEST
RES
SIN
GR
OO
VE
.6
S0
21
.SP
Sf.
--"
Sro
..-O
e'il'
r-0
'-T
OR
out
STR
US
ING~VI!.
42
75
.40
PS
t.-
MlA
NSTRE5~
RA
llO
..6
8Z
1i
------
-ST
t.O
r:v.O
P'-~EIIN
STI
'!ES
S-
RA
TIO
..0
42
00
..... ..;= .....
. I-
"'
-..
-.,>"---'--",~~
•.,~
'~----~-'
.,,,.,"
~.._
_,..,
-_
-:--_
-.-.
__,
"..-~.
..
-T[S
T'
"0.
APP
rND
IXB
.DA
1ARE
DUCT
ION
RESU
L1S
fOR
r•
0.7
AND
rs=
0.9
.,
sSPtCI~EN
'"Ae"I~[
PAN
RCAL
NCAL
NV
IS.~o.
,!O~
WEI~HT..,!S[NOI!l.6
BfNOI~G
B[f
d)I
NG
RC
AI.
TO
ilQU
,tH
CA
\.TO
RQU
E"'
VIS
TORQ
UE
B!N
OIN
8ST
R!S
SSH
tAR
STRES~
'TR
US
RA
TIO
-..
02
....
_·--
-"36
0
'"24
!l..
281
.-'
"252
22
8
25
3--
e3
l)'-
.
25
63
27
"'2
56
'3
06
..:.
...~
·5T-
'-..-
-·."
3n
--25
82
32
25
92
55
·--t
60
...-V
i
26
1Z
67
262
231
,·_
·t6
3.
"i2
9
--'-
16
4'
'ZeG
,':3
94'3
85
--3
96
'"3
31
30
.80
30
40
00
.0
30
.90
30
40
00
.0
30
.20
30
40
00
.0
30
.30
30
40
00
.0
31
.60
,304000.~
.10
8
.68
3
.70
2
.1i~
.73&
·
.10
3
.69
4
.70
6
.68
1
.10
1
.1iz
.12
4
.80
9
.121
.68
9
.66"
.114
-
.66
2'
513S~.5
51
19
3.3
56
88
9.0
56
47
1'..
1
51
90
5.3
57
27
6.0
56
64
6.9
52
18
8.3
.
55
08
9.1
51
05
7.8
55
45
3.&
5B
18
3.5
55
86
6.5
·
59
51
3.2
56
85
4.6
57
96
1.5
56
58
0.B
51
60
6.1
68
04
9.4
11
44
4.1
11
82
8.8
69
48
9.7
61
86
9'.5
,
70
24
9.8
'
10
53
0.1
1Z
14
4.6
.
13
91
2.6
68
17
1.9
-
6778
&.5
69
44
3.9
69
05
0.3
60
20
t.&
61
60
3.2
10
93
1.0
68
05
1.0
69
21
9.9
28
.60
29.2
0-
32
.30
.27.
&.
28
.10
,2&
.00
29
.20
28
.20
31)O1~
28
.70
29
.10
29
.10
28".
90
32
.10
33
.10
33
.35
33
.00
.-3
2'.&
Q'
44
.60
45
.00
44.&
0
44
.&5
'
44.&
0
4~~O
44•.7
0'
45
'.0
0
44
.60
44
.90
44
.80
4••
80
44
.50
44
.80
4".8
0
44.6~
44
.80
44
.60
30
40
00
.0
304000~0
30
40
00
.0
30
40
00
.0
30
40
00
.0
30
40
00
.0
30
40
00
.0
30
40
00
.0
-30
40
00
.0·
29
.10
'
30
.10
30
.50
30
.20
il.0
0
3O-e
1U
31
.05
31
.15
31•.1
0
31
.10
30
40
00
.0
32
.00
30
40
00
.0
30
.55
304q
OO
.O
30
.00
30
40
00
.0
49
.20
49
.&0
49
.50
49
.30
49
.50
49
.90
49
.60
49
.70
49
.60
49
.40
50
.15
49
.95
50
.10
50
.15
49.&
0
49
.60
49
.BO
.49
.50
19
00
00
.0
i90
00
0.0
19
00
00
.0
19
00
00
.0
19
00
00
.0
19
00
00
.0
19
00
0f)
.0
19
00
00
.0
19
00
00
.0
i90~00.0
19
00
00
.0
19
00
00
.0
19
00
00
.0
19
00
00
.0
19
00
00
.0
--1
90
00
0.0
19
00
00
.0
i90
00
0.0
7 T 7 T ., ., T T ., ., .,.1
., 1 1 7 1
;-1
'
1 t;, 3 '3 3 1 12 :3 '3 3 3 '3 3 3
.;,-
442
41
3
39
8
40
0
"
.MtA~
BEN
DIN
GST
RE
SSIR
GR
OO
VE
.69~7.~
PS
I.
----
...,-
"-sT
er.·
OrY
.--0
'-t
l[1'
iDtN
GS
TR
ES
Sti
tG
JlO
O"n
:•
tT6
9'e
29
'p
st.-
.-.
>'E
IlNTO
RO
U[
STR
ESS
I»G
RO
OY
!.5
6$
49
.9P
!-f.
,-,.
,TO
..D
!V.C
'T
OR
OI!
[S
T$l
[SS
ING
RO
OY
E•
14
53
.19
Pst•
.
---'-
--_
..--
--$
Tt.
-OfV
.O
f'"M
EliN
SlR
ItS
SR~TIo-."
--..
'..0
30
52
-
M[A
NS
TR
ES
SR
AT
IO.'
.101
195'
.._._
....-
.....•
....__
..-._
--_
.__.__
._-..
.__.__
.-..-
----
-_._
-_.
__._-
_.•.._
--_._
.-
~ +='
l\:)
I
APP
EN
!)IX
B.
DATA
RED
UCT
ION
RES
ULT
SR
ESU
LTS
FOR
rll
=0
jPN
Or
"0
.9II
TEST
SPEC
IMEN
roIA
C!1I
NEPA
NRC
ALN
CAl
NV
URC
ALNC
AI.
NV
ISBE
ND
ING
SHEA
RST
RES
S__
~p_!.-.
____
_f:l O
.•.·_
fiO
.V
EIG
HT
BEN
DIN
GBE
ND
ING
BEN
qIN
GTO
RQUE
TORQ
UE·
TORQ
UEST
RES
SST
RES
SR
ATI
O
280
248
321
39
00
00
.05
0.1
55
0.1
53
04
00
0.0
45.1~
24
.30
54
13
0.6
.47
66
2.4
.65
6
311
288
227
19
00
00
.05
0.1
02
5.9
53
04
00
0.0
44
.55
'2
5.1
56
57
83
.94
08
30
.9.9
30
-31
8.·
"2
67
'2
261
90
00
0.0
49
.70
24
.35
30
40
00
.04
4.4
5.
24
.30
62
18
9.7
39
59
9.0
.90
7
.'3
2130
02
271
90
00
0.0
49
.60
23
.15
30
40
00
.04
4.6
52
4.1
05
91
91
.73
91
83
.8.8
72
,.'3
2327
52
2619
00·0
0.0
49
.60
24
.85
30
40
00
.04
4.1
02
2.2
56
37
45
.03
62
97
.41
.01
4
-·~30
--40
92
251
90
00
0.0
49
.90
23
.90
-3
04
00
0.0
44
.95
23
.60
60
82
1.7
37
98
7.6
.92
4
339
364
224
19
00
00
.05
3.0
52
3.5
53
04
00
0.0
45
.00
22
.40
56
34
1.6
36
06
7.8
.90
2
348
372
226
19
00
00
.04
9.6
52
4.5
53
04
00
0.0
45
.20
25
.00
62
75
9.8
40
07
0.0
.90
4
.'37
7-.
-36
22
281
90
00
0.0
49
.80
25
.75
30
40
00
.04
5.0
02
6.3
06
56
12
.54
23
68
.6.8
94
378
42
92
271
90
00
0.0
50
.00
24
.75
30
40
00
.04
5.0
02
5.4
56
27
95
.54
10
26
.6.8
84
38
i35
92
271
90
00
0.0
50
.00
24
.65
30
40
00
.04
5.1
0'
24
.60
62
59
0.9
39
48
9.0
.9i5
-'"3
83'.
---4
11"--
--c'"
26-1
90
00
0.0
49
.55
23
.15
30
40
00
.04
4.6
02
4.3
05
92
38
.9-3
95
73
.3.8
64
388
"J36
.2
261
90
00
0.0
50
.30
22
.95
30
40
00
.04
5.1
5.-
26
.15
57
72
1.5
42
27
3.0
.78
8
395
435
.Z
26·i
90
00
0.0
50
.10
23
.65
30
40
00
.04
7.1
02
3.4
06
00
02
.23
58
50
.4.9
66
-;'9
7-.
-."-.
-"4
01
'---.
····z
-.
-25
.-."
,-"
1900
00.0
-..
49
.95
---·
·'-
-'24
.00
·-·304000.0···-··-'~'01l--c3.40---
..6
10
33
.5"
'--3
75
92
.7-
--.9
3i-
430
400
229
i90
00
0.0
50
.10
25
.30
30
40
00
.04
4.6
5'
"24.
756
41
25
.44
01
09
.9.9
23
-
431
38
6·
'Z'
2e
19
00
00
.0·5
0.1
0-
'1'.
65
30
40
00
.047
'015
---
-'-"
2!h5
0''&
24
75
.83
91
37
.7.9
22-
--432"---~02----
..-..
"2
-."2
7'-.
'-.-
-.,9
0000
.11'
·---
-50
.00-----~••
45-
-304000.~
--,:"
""47
.10
--2
4-.
-ll0
····
-6H
Z3
.r--
")8
05
2.9
-'-'~9U---'-
Mf~N
STR
ESS
RA
TIO
.'.8
91
01
ENDU
RANC
E
.---
---.
--..
,---
----
·ST
C.-
Dfv
.OF
14EA
NS
TM
S$
RA
TIO
•..0
7'8
2"
-....
....
...-
.._-
--...
.._--
..._..
_.,-,
_._.
._--._
.,._.•.
.._.
..,.__
.._._~-_
.._-_
._--
.-_...
_--
---_
.-
~ .r= w
APPENDIX C
OPERATING CHECKLIST FOR NASACOMPLEX FATIGUE RE'SEL\RCH MACHINES
TO INSTALL SPECIMEN
1. Turn power off at the wall - block up loading arm.
2.' Choose correct specimen for test using data book.
3. Prepare specimen for test.a) Inspect specimen for defect or damageb) Clean any oil or foreign matter off specimen
4. Clean inside of toolholders.
5. Clean collets.
6. Use WD-40 on collets - insure there is no binding.
7 •.· Block up one' s ide of loading arm with support: bolt.
8. Insert specimen and install collets.
9. .Block .uP other side of loading arm with support bolt.
lb. Level.the toolholder arms using the support bolts.
11. Place coupling gauges over th~ flex couplings. Be surethe coupling gauges fit snug.
12 •. Check the toolholder arms to see if they are still level.If not, adjust the support bolts to make the arms level.
13. Install keys in specimen.
14. Center specimen in toolholder.
15. Tighten outboard side (with strain gages) of specimen byhand •
. 16. Tighten outboard side of specimen using the wrench. Makesure it is very tight as it cannot be tightened again.after this step.
144
17. With the inboard side (without strain gages) still loos~~place brushes on the slip rings. They should be firmagainst the rings, but not bent tight.
18. Zero the instrumentation.a) . Turn milliammeter to "in"b)c)d)e) .f)g)
:·h)··i)
j)k)1)
m)
Check B+ (5 volts) - Bridge Balance Switch to "B+"Check GV (5 volts) - Bridge Balance Switch to "GV"Turn Bridge Balance Switch to "BB"Turn Channel Selector to desired channelSet Atten.Switch to desired levelUse C. Bal. to "dip" voltmeterUse R. Bal. to."zero" milliammeterRepeat (g) and (h) until minimum voltmeter readingand zero milliammeter readings are obtained simultaneouslyLock C. Bal. and R. Bal. controlsTake a short visicorder. runRepeat (e) throught (k) for each channel to beused in the testTurn milliammeter switch to "out"
19. Calibrate the instrumentation.a) Connect to Ext. Cal. the appropriate value of
calibration resistance·b) Using the gain control, obtain the appropriate
deflection on the visicorder (some adjustment ofthe R. Bal. may be necessary) - do. not changeC. Bal.
c) Take a short visicorder run for each deflection -be sure instrumentation is calibrated for both bend·ing and torque.
2~. Tighten the inboard side of the test specimen.
21. Remove flex coupling gauges.
22. Clean the specimen and specimen groove.
23. Lower support bolts. Make sure horizontal link is notresting on a support bolt (check both sides).
24. Lower the loading arm - no pan load.
·25. Check to see if pins are loose and all nuts are tight.
26. Check to see if bearings are vertical.
. I
27. Check to see if vertical link is vertical.
28. Check to see if brushes are still correctly on thesliP. ri.ngs.
29. Rotate machine .byhand.
'30'. Turn the power on at the wall.'
31. Clean the' slip ri.ng~. (with machine. runni.ng) •.
32. Turn the machine off.
33. Apply the appropriate load tbthe pan.
34. Set microswitch.
35. Record static bending.
36. If torque is to be applied, see Procedurer
for OperatingNASA Complex Fatigue Research Machines with Bending and
·Torque Loads.
37. Set the clock to zero.
38. Put on the bridges.
39. Record dynamic conditions.
40. Check lube and oil level.·
TO REMOVE SPECIMEN
1•. Turn power off at the wall.
2. Block up loading arm.
3.' Remove the brushes'from the slip rings - this step isvery important.
4~ Loosen both collets S9 the. specimen is free to slide.
5. Remove the bridges.
6. Remove the keys.
7. Raise one side of the toolholder using a support bolt.
8. Remove collets .and specimen.
146
147
APPENDIX D
PROCEDURE FOR OPERATING NASA COMPLEXFATIGUE RESEARCH MACHINES WITH
COMBINED BENDING AND TORQUE LOADS
1.' Install specimen and balance visicorder as per "Checklist.",
2. Apply the torque to the required number of divisions, includingthe bending interaction' divisions, using the Infinit-Indexer.
3. Stabilize the torque divisions as follows:
'3.1 Turn the machine by hand'for at least five cycles. Youwill probably observe a decrease in the torque divisions.
3.2 Re-torque machine to desired. divisions and repeat theprocess until the mean torque level remains constant.
4. Start the run and observe the mean torque divisions insuringthat it doesn't decrease. If no more than 2 divisions ofdownward shift in the mean torqu~ divisions is observed, continue the run.
'5. If more than 2 divisions of shift in the mean torque level inthe downward direction is observed or if excessive upwardsdrift in the mean torque divisions is observed, particularly10 or more divisions, stop machine, remoVe pan weight, blockup bending load train, block up loading arm until level, andremove torque by loosening in-board collet only (the sidewithout the strain gages). Check to see where the new torquezero is. If the shiftin the zero is within one division ofthe drift in the mean of the torque divisions, re-zero thetorque channel, re-tighten the specimen' and repeat Steps 3, 4,5, and 6 until torque is stabilized.
6•. If'the zero shift is in excess of the drift in the mean torquedivisions initiate an investigation as to its cause.
Probable,1.2.
3 •.
causes may be the following:.True change in torque beyond the intended level.Change in strain gage characteristics, perhapsindicating a strain gage deterioration.Deterioration of amplifier components.
7. If the drift in the mean torque divisions is not in excess of.10 divisions, and a few specimens at.the same stress levelcheck out favorably in Steps 1 thr~ugh 4, do not re-check thezero; finish run.
\
148
·.,
iAPPENDIX E
CALIBRATION OF NEW TORQUE GAGES FOR MACHINE #3, JUNE 1969
In May 1969, one of the strain ~ages in the torque bridge of
the toolholder of Machine #3 failed. The ~age was replaced. and
it was necessary to recalibrate. Since the bending bridge was
left untouched, the only tests which.had to be conducted were
the torque load versus visicorder output calibration and the
bending interaction into torque calibration.
The results of these tests are presented below•
. Torque Load Versus Visicorder Output
The procedure used is identical to the one presented in
Section IV-C-l. The data appears in Table 18.
Data Reduction
The same data reduction technique that was us.ed in Section
.IV-C-l is used. Figure 43 gives the plot of the data of Table
18.
149
Machine Point Point W N .VJ.s
Ibs div
Toutpsi
Ttruepsi
slope
K.r3 (40;0,40.0) (2.5,6.0) 37.534.0 1008 848 0.842
Therefore
= 0.842
150
Bending Interaction Into Torque
For the bend~ng interaction into torque procedure, consult
Section IV-C-2. The method given there was followed,exactly
for this test. The 'data appears in Table 19.
, 'Data Reduction
Us~ng the graphical reduction technique from Iigure 44,
,Point A = (25.0, 0.73)
Point B = (35.0, l.00)
N • = -10.0v~sB
N = + 0.27visT
N ~ RE:B = v~sB gage =
N N G RcalB a cal
(-10.0)(190) = 12.00 vin/in(25.0)(4)(2.23)(490)(10 3)
Therefore
RgageN G R 1a ca
=(0.27)(120)
(45.3)(4)(2.06)(304)(10 3) = -.2855 Vin/ip
~/T =-0.2855
12.00 =-0.0238 . ,
APPENDIX F
MAINTENANCE CHECKLIST
·EACH RUN~<, /
The level of the greas~' in·rthe front shaft couplings and 'thelevel of oil in the gear box shall be checked and lubricationadded if ne~ded•. The oil to be used,in, the gear box is Mobil D.T. E. Oil - BB.
.~
Log the hours each machine has been run.
EVERY MONTH
.Check front Shaft Couplings for proper amount of Marfak #1' ,. grease.
EVERY FOUR MONTHS
Front and back shaft couplings shall be cleaned, inspected.and regreased with Texaco Marfak #1.
\
\
151
top related