0001 001

Download as pdf or txt
Download as pdf or txt
You are on page 1of 87

AN EMPIRICAL MODEL FOR THE PREDICTION

OF THE TORQUE OUTPUT OF COMMERCIAL


VEHICLE AIR BRAKES

MVMA P r o j e c t #1,36

L. K, Johnson
P. S. Fancher
T.D. G i l l e s p i e

F i n a l Technical Report

December 1978

Highway S a f e t y Research I n s t i t u t e
The U n i v e r s i t y o f Michigan
Ann Arbor, Michigan 48109
T u L i c a l Rapwt Docurmtoti'a Page
1. R . r o r t N a 2. G o v e r n m i Accession Ho. 1. Raclsarnt's Cotelop No.
Ell- tIS2I-75--53
I
4. Title rd Subtiile 5. R e r t Date
Deceeber 197;
7,i; EMPIRICAL MQDEL FOR THE PREDICTION OF THE
TORQUE OUTPUT OF COMtlERCIAL VEHICLE AIR BRAKES '. pwfo"inqO'""""C'*

L. Johnson, P . Fancher,, T . Gillespie UM-HSRI-78-53


9. Pwkrminq Olgairdiocr MH md Addmra 10. WorC Unit No.
Hi~hwaySafety Research I n s t i t u t e 361 506
The University of Michiaan 11. Cantrma w claw NO.
Huron Parkway & Baxter Road MV!?A Proj 1 .36 .
n 43109 17. 7vm of RW m d Pwiod G..I~
12. krnluiar A- N u a d Addrams
Final
Mctor Vehicle Manufacturers Association 711 177-10/15/78
3C0 Nev; Center Bui 1di ng
Cetroi t, Mi chi gan 48202 14. S C ~ S ~ ~Agency
I I ~ tode

16. Absirocr
An empirical model f o r representing the torque capabi 1i ty of commercial
vehicle brakes i s described including discussions of the fcrm of the model,
the required t e s t data, and the regression method used t o f i t the data.
Inertia1 dynamometer data f o r six pneunatical ly-actuated brakes are
employed in analyses i l l u s t r a t i n g the application of t h i s modeling technique
t o brakes of both wedge and S-can design. The results presented show ( 1 )
brake torque divided by actuation e f f o r t as a function of slidinq speed,
interface temperature, actuation force, and work history and ( 2 ) the good-
ness of f i t between the empirical model and the oricinal t e s t data. Aopli- /
cation of the brake model in a detailed vehicle simulation i s discussed.
The report concludes w i t h remarks summari zing the properties of the
empirical model and recommending research on the influence cf work history. I
I

17. K.r Wwda 18. Oistr~butimStot-mi


brakes, a i r brakes, interface
temperature, work history, brake
godel ing, nrocessi ng brake data Ui;L IF1ITED

19. 5 w r i t y Clmaaif. (of h i s mpu4


II
ID. k e n i i y Clmssif. (04 this pr).) 21. No. of Paps 2% Prtcr
iICi.IE :iOl!E
I , 83
TABLE OF CONTENTS

1. INTRODUCTION ....................... 1

2 . THE BRAKE MODEL . . . . . . . . . . . . . . . . . . . . . 5

2.1 The Concept o f Brake E f f e c t i v e n e s s . . . . . . . . . 5


2.2 Measuring E f f e c t i v e n e s s f r o m Dynamometer
Tests ........................ 7
2.3 Representing t h e E f f e c t i v e n e s s F u n c t i o n ....... 9

3 . TEST PROGRAM ....................... 13

3.1 I n e r t i a Dynamometer Tests . . . . . . . . . . . . . . 13


3.2 Processing o f t h e Raw Data . . . . . . . . . . . . . 17
3.3 Torques Achieved D u r i n g Each T e s t . . . . . . . . . . 17

4. DATA ANALYSIS . DERIVING THE EFFECTIVENESS


FUNCTION ......................... 30

C a l c u l a t i n g E f f e c t i v e n e s s f o r Wedge
and Cam Brakes ................... 30
Determining t h e Force Produced by t h e
A i r Chambers ,................... 30
Return S p r i n g E f f o r t s . . . . . . . . . . . . . . . .
C a l c u l a t i n g S l i d i n g Speed . . . . . . . . . . . . . . 34
C a l c u l a t i n g I n t e r f a c e Temperature . . . . . . . . . . 34
Computer Program DYNA-DRUM I 1 1 . . . . . . . . . . . 37
Comparison o f Measured and C a l c u l a t e d
Temperatures . . . . . . . . . . . . . ....... 37
........
C u r v e - F i t t i n g t h e E f f e c t i v e n e s s Data 41

5 . RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . 43

5.1 E f f e c t i v e n e s s F u n c t i o n . . . . . . . . . . . . . . . 43
5.2 S i m u l a t i o n o f Time H i s t o r i e s of Torque . . . . . . . 49

6 . EMPLOYING THE BRAKE MODEL I N VEHICLE SIMULATIONS . ..... 53

7. COKCLUDING REklARKS . PROPERTIES OF THE ERAKE MODEL


AND RECOMMENDATIONS FOR RESEARCH ............. 56

REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . 58

APPENDIX A . I n e r t i a l Dynamometer T e s t Procedure . . . . . . . 59

APPENDIX B . E f f e c t i v e n e s s Functions . . . . . . . . . . . . . 62
ACKNOWLEDGEMENTS

The help and support of the following organizations and


people are gratefully acknowledged:

The Motor Vehicle Manufacturers Association for


project support.
Greening Tes t i ng Laboratories and Mr. Robert Hannon
f o r dynamometer t e s t s .
Dave Velliky and Sandy Milazzo for digitizing
osci 11ograph records.
The mechanical technicians a t the Highway Safety
Research I n s t i t u t e f o r instal 1 ing thermocouples.
Jeannette Nafe for typing.
Mr. Percosh Jain of Ford Motor Company and
Mr. Thomas Mahaffey of Rockwell International
f o r aid in procurring brake linings.
1 .0 INTRODUCTION

Predicting the braking performance of a commercial vehicle


in an emergency stop i s a very difficult task, requiring detailed
information on the vehicle's brakes, antilock system ( i f i t has one),
load distribution, suspension characteristics, and t i r e properties.
To aid in predicting and understanding the factors contributing t o
stopping capabi 1 i ty, large-scale, computer-based mathematical models
have been developed for simulating emergency braking of commercial
vehicles [ I ] . These large-scale models contain representations of
vehicle components (such as the brakes) in separate sections of
computer code. The empirical model presented herein i s intended t o
be used as part of a large-scale simulation for studying the
infl uence of measured brake properties on vehicle response during
a single stop.
Nonetheless, another important use of an empirical model of
brake torque capabi 1 i t y i s simply t o provide a comprehensive summary
of the results of numerous dynamometer tests. The form of an
example torque versus time curve obtained from a single dynamometer
t e s t a t a fixed brake 1 ine pressure i s i 11 us trated in Figure 1 . 1 .
In the past, data from a series of tests of the type illustrated in
Figure 1 . 1 have been summarized in graphs of average torque versus
i n i t i a l speed and brake line pressure as shown in Figure 1 . 2 . How-
ever, much of the detail of the variation of torque during a stop
i s lost in working with average torque. I n particular, i f the
conditions for wheel lockup during a stop are t o be evaluated, the
type of data illustrated in Figure 1.2 i s inadequate. Accordingly,
values of "peak i n i t i a l torque," "minimum torque," and "final tor-
que" are sometimes read from curves of the type shown in Figure 1 . I .
In this regard, the empirical model described here i s intended t o
contain enough information about the brake t o essentially reproduce
the original torque versus time characteristics.
Figure 1.1. Example torque time history from a dynamometer t e s t .

.- . -.-
LINE ?RESSOR~=

Figure 1 . 2 . Typical summary of average torque versus initial speed


and 1 ine pressure.
This model was developed following work by H . E . Cook, e t a l .
[2] on passenger car brakes. The independent variables used t o
predict torque for a breke with a specified prior work history are
sliding speed, actuation force (or torque), and interface tempera-
ture. The relationship between torque and the independent variables
i s developed empirical ly from dynamomgter t e s t resul t s .
I n addition t o predicting stopping distance, other potential
applications of the model pertain t o predicting (1) brake torque
distribution among a vehicle's axles so as t o optimize brake system
design, ( 2 ) wear balance, and (3) mountain descent performance.
This report contains a technical description of the brake
model, including discussions of the form of the model, the required
t e s t data, and the calculation and regression techniques used t o
f i t the data. Inertial dynamometer data from six pneumatically-
actuated brakes are presented and then employed in analyses i 1lustrat-
ing the application of this modeling technique t o brakes of b o t h
wedge and S-cam design w i t h various types of linings. Example
results are presented indicating ( 1 ) the variation in the "effec-
tiveness function" ( i .e., torque o u t p u t divided by actuation force)
as a function of sl iding speed, interface temperature, actuation
force, and work history and ( 2 ) the goodness of f i t between the
empirical model and the original t e s t data. The body of the report
concludes with a brief discussion of the application of this brake
model within a large-scal e vehicle simulation. Concluding remarks
summarizing the properties of this brake model and recommending
specific research on the influence of work history appear in a final
section.
I t i s worth noting that the modeling techniques described
herein are quite general and they have been employed in another
study addressing the torque capabi 1 i ty of hydraul ical ly-actuated
brakes for heavy trucks [3].
2.0 THE BRAKE MODEL

2.1 The Concept of Brake Effectiveness


I t i s well known that the frictional characteristics of f r i c -
tion materials are influenced by the temperature, 8 , existing a t
the sliding surface, the sliding speed, V , and the pressure, P ,
acting between the two elements of the friction pair; i . e . ,
p = p(e, V, P ) . Thus, i t i s evident that the torque o u t p u t of a
brake will be dependent on these three factors. However, in the case
of a drum brake the dependence will n o t be 1 inear because of the
rather complex geometry relating the coefficient of friction of the
lining t o the frictional force produced. An example of the sensi-
t i v i t y , s , of brake torque, T, t o the coefficient of friction, p , of
the linings for various kinds of drum brakes i s shown in Figure 2 . 1 .
.. --.I

Figure 2.1. Variation of s w i t h P for different brakes. A:


duo-servo; B: two leading shoe (pivoted) ; C:
leading-trail ing (pivoted) from Reference [4].
One can create a relationship between the coefficients of
friction of the 1 ining and the torque o u t p u t of the brake per unit
force input which i s a function of the coefficient of friction and
brake geometry. Because p varies w i t h a , V, and P, the above rela-
tionship will also vary with e, V , and P . Thus, the torque o u t p u t
o f a drum brake can be expressed as follows:
where T denotes torque o u t p u t , F denotes the force actuating the
shoes, and e denotes torque o u t p u t per unit force input, which will
be called effectiveness for the remainder of this report. Effec-
tiveness, i s the parameter which links lining friction t o brake
torque, and i t i s a function of a, V , and P ; i . e . ,

However, the pressure acting between the 1ining and drum i s a


function of the actuation force, F, so that

There i s one other important factor, ignored u p t o this


point, which affects lining friction, and consequently, effective-
ness. That factor i s work history, H . Work history i s very
d i f f i c u l t t o quantify, b u t very easy t o observe. An example of
the influence of work history i s the change in the torque o u t p u t
of a brake a f t e r a fade and recovery t e s t compared t o the torque
o u t p u t prior t o the fade and recovery. Work history, however, can
be specified by referring t o some portion of a t e s t sequence, e.g.,
post-burnish effectivenss, f i r s t fade, f i r s t recovery, final
effectiveness, and so on. Therefore, in i t s final form, effective-
ness i s defined as

This concept of brake effectiveness, then, treats a brake


assembly as a black box with an actuating force being transformed
into a torque o u t p u t . The transfer function, e , i s complicated,
being a function of interface temperature, sl idi ng speed, actuation
force (i.e., pressure between drum and lining), and work history.
As a "black box" the brake may be diagrammed as shown in Figure 2 . 2 .
Figure 2 . 2 . A "black box" representation of the mechanical
friction brake.
I t should be noted that while effectiveness i s related t o
lining friction through the geometry of the brake, there are other
factors which play a lesser role. One of these factors i s the
mechanical efficiency of the actuation mechanism of the brake.
Another i s compliance of the brake drum and shoes which compliance
causes changes in the pressure distribution (across the width of
the lining) thus affecting the torque o u t p u t of the brake. These
factors are assumed t o be of second-order importance. They are
also d i f f i c u l t t o quantify, and, therefore, are not explicitly in-
cluded in the model as variables i nfl uenci ng effectiveness.

Measuring Effectiveness from Dynamometer Tests


Although the effectiveness function for a particular brake
assembly can, in theory, be calculated from known frictional pro-
perties of the lining and the geometry of the brake, i t i s straight-
forward to measure values of effectiveness from full-scale dyna-
mometer t e s t s . Measuring the performance of a brake as an
assembly has the advantage of including effects on the effective-
ness function which are inherent in the assembly. Such effects
could be systematic or random. For example, the mechanical efficiency
of the actuating mechanism would have a systematic effect on the
effectiveness function. On the other hand, changes in the pressure
distribution across the width of the lining caused by wear, mechanical
compliance of the shoe and drum, or thermal distortions of the lining
and drum could have b o t h systematic and random effects on the
effectiveness function. Thus, i t should be expected that measuring
the effectiveness in full-scale t e s t s will yield a more accurate
representation of brake performance than can be achieved by cal-
culating effectiveness from frictional properties of linings as
measured in t e s t s of lining samples,
Developing the effectiveness function from dynamometer t e s t s
requires detailed time histories of the variables involved, because
interface temperature and s l idi ng speed are continuously changing
throughout a brake application. Additionally, actuation force varies
during a constant actuation stop or snub, and brake torque varies
during a constant actuation force stop or snub. During a single
brake application, effectiveness can be calculated as a function of
time with each point in time corresponding t o a different tempera-
ture, sliding speed, and [possibly) actuation force. Thus, by per-
forming many brake applications over a range of temperatures,
sliding speeds, and actuation forces, i t becomes possible t o map
the effectiveness function.
Not only must effectiveness be evaluated a t many points during
a brake application, b u t the corresponding values of sliding speed,
actuation force, and interface temperature must be known. Whereas
sliding speed can easily be computed from instantaneous values of
the rotational speed of the brake, the remaining two variables re-
quire special attention, For example, in the case of the wedge
brake, the force applying the shoes can be computed from known out-
p u t force characteristics of the a i r chambers and the wedge angle.
On the other hand, the force applying the shoes in a cam brake i s
not easily computed. However, i t i s not necessary to know that force.
Because the effectiveness model t r e a t s the brake as a black box,
the actuating force for a cam brake can be considered to be the tor-
que actuating the cam. Cam torque can be computed from known a i r
chamber output force characteristics and the length of the slack
adjuster. (Note, then, t h a t effectiveness f o r a cam brake i s
computed as e = TIT where T denotes cam torque,)
The temperature that exists a t the interface between the
lining and drum i s extremely d i f f i c u l t t o measure, HSRIts experience
with measuring temperatures in brake drums a1 so indicates that the
temperature can vary greatly from stop t o stop even when stops
are made from seemingly identical i n i t i a l conditions. Additionally,
temperatures can vary markedly across the width of the lining.
For these reasons, a calculated interface temperature i s used in
the determination of the effectiveness function. Nevertheless,
measured temperatures of the drum a t the beginning of a brake appli-
cation are used as a starting point for calculating the interface
temperature. Once a brake application has begun, the interface
temperature i s calculated using measured values of the instan-
taneous rate of heat generated a t the drumllining interface. The
rate a t which heat i s generated i s merely equal t o the product of
brake toroue and rotational speed. Also, as a simplification, the
calculated temperature i s the average across the entire width of the
lining.
A flow diagram of the procedure used to calculate instantaneous
values of effectiveness during a brake application i s shown in
Figure 2.3.

2.3 Representing the Effectiveness Function


Two methods are avai lable for representing the effectiveness
function once a set of data has been generated from dynamometer
tests. The f i r s t i s tabular representation. This method has the
advantage of being easily interpreted i f a suitable table can be
developed. The bounds on the domain of the function are also easily
determined by the endpoints for each line of the table. However, a
table takes much room t o store and, more importantly in this case,
can be d i f f i c u l t to construct. Past experience in trying t o construct
s i ngl e-val ued tables of effectiveness from results of dynamometer
tests has proven t o be frustrating. Variability in brake t e s t data,
inadequate temperature information, and lack of data repeatabi 1 i ty
due t o work-history effects have confounded the situation t o the
extent that the construction of meaningful tables i s extremely
difficult.
F(t e(t) = .m
T(t) e(t)
T(t)
. ,-
r

e(t>
~ ( t )\ . 1:
- F(t) e ( e , V , FL
/

LC,
.
- i
V(t)
8

t.
Heat Transfer
Oo r
Calculation

Figure 2.3. Flow diagram for calculating values of


effectiveness from dynamometer data
(8, denotes i n i t i a l brake temperature).
The second method, and the one chosen for this study, i s
t o employ a curve-fitting process t o derive a mathematical equation
describing the data.
Since the end use of a brake model i s t o simulate vehicle
braking behavior on a computer, a mathematical description of the
effectiveness function i s very attractive, because i t i s better
suited for implementation on a computer than a table. Whereas a
table requires considerable storage space, an equation only requires
storage of some coefficients. The computer can also perform a cal-
culation more quickly than i t can a table lookup. Curve f i t t i n g
the data also smooths the scatter in the data, although sufficient
care must be exercised t o ensure that the data i s n o t smoothed
t o the point of eliminating genuine variations.
Curve f i t t i n g can also provide two valuable numerics. The
f i r s t of these i s the coefficient of determination, r 2 , which
represents the fraction of the behavior of the data which the curve
f i t has described. For example, i f a curve-fitting operation pro-
duces an r 2 value of 0,9, then the resulting f i t t e d equation explains
90% of the variation of the data. The remaining 10%of unexplained
behavior i s composed of random errors and errors resulting from
systematic trends in the data which the curve f i t has ignored.
The second numeric i s the standard error. This quantity i s a
measure of the scatter of the differences between predicted and
observed data points. Ideally, i t i s a measure of experimental
error. Thus, the standard error can be used to estimate the range
of torques a brake might produce under identical conditions of inter-
face temperature, sl idi ng speed, and actuation force.
A disadvantage of the curve-fi ttinq method i s the difficulty
of determining the bounds on the domain of the derived effectiveness
function. However, this determination can be done, and i t i s very
important t o do so as extrapolating a curve-fitted equation can be
dangerous indeed.
No attempt i s made t o represent work history in numerical
terms. Rather, for each work history ( i . e . , post-burnish effective-
ness, f i r s t fade, e t c . ) a separate curve-fitting operation i s per-
formed. For example, a dynamometer t e s t might consist of several
different phases which might be grouped into, say, five work histories.
Then., five sets of effectiveness data would be produced and a separate
curve-fitting operation performed for each data set. The result
would be five curve-fi tted equations , each representing the perfor-
mance of the brake during each of five work histories.
3.0 TEST PROGRAM

3.1 Inertia Dynamometer Tests


Six t e s t s of ai r-actuated friction brakes were conducted on
an inerti a1 dynamometer a t Greening Testing Laboratories, Detroit,
Michigan. Three t e s t s were conducted on a Rockwell 15 x 6 RDA
wedge brake, each t e s t with a different type of lining. The remain-
ing three t e s t s were performed using a Rockwell 16 112 x 7 S-cam P
brake, also w i t h three different types of lining materials. All of
the t e s t s were conducted a t a nominal wheel loading of 9000 1 bs and
a t i r e radius of 20.2 inches. The inertial load used in these tests
was 793 slug-ft2.
The original-equipment linings used in b o t h the wedge and cam
brakes were AEB 693-551D and MM8C5. These lining materials were
selected because of their widespread use in FMVSS 121 a i r brake
systems. Other 1 inings tested were the Euclid E84 (in the wedge
brake) and Euclid E80 (in the cam brake). These l a t t e r linings are
aftermarket linings, and although they are slightly different formu-
lations, they are recommended for equivalent appl ications.
The wedge brake was actuated by two Type 16 a i r chambers and
used a 10" wedge angle. A Type 30 a i r chamber and a 6-inch slack
adjuster length was used on the cam brake.
The brakes and linings tested are summarized in Table
3.1.
All six tests followed the same procedure, which was similar
t o an FMVSS 121 dynamometer t e s t . The t e s t consisted of three parts:
burnish, post-burnish effectiveness, and fade and recovery. Effec-
tiveness data i s thus acquired for three work histories: ( 1 ) post-
burnish effectiveness, ( 2 ) fade, and ( 3 ) recovery.
The effectiveness t e s t differs somewhat from the 121 procedure
in that i t i s designed to cover a broad range of temperatures and
sliding speeds so that a sufficient amount of data could be generated t o
Table 3.1. Brakes and Linings Tested.

Brake Assembly Li ni ngs

Rockwell 15 x 6 "RDA"
Wedge ABB 693-551D
Guni t e 2046A Drum MM8C5
Type 16 Air Chambers
10" Wedge Angle

Rockwell 16 1/2 x 7 "P"


S- Cam ABB 693-551D
Guni t e 31 66 Drum MM8C5
Type 30 Air Chamber
6" Slack Adjuster

Wheel Load - 9000 1 bs


Tire Radius - 20.2 in
Inertia - 793 slug-ft2

determine an effectiveness function. For t h i s reason, the effective-


ness t e s t consisted of five stops made from 30 mph a t i n i t i a l brake
temperatures (IBT's) of 150, 200, 250, 300, and 350°F, followed by
f i v e stops from 50 rnph a t the same i n i t i a l brake temperatures.
These stops were run i n a constant pressure mode a t a pressure which
was determined to yield an average deceleration of 12 fpsps f o r
stops from 30 and 50 rnph a t an i n i t i a l brake temperature of 150°F.
Therefore, data was only generated for two levels of actuation force.
A1 though i t would have been desirable to t e s t over several levels
of actuation force, insufficient dynamometer time was available f o r
t h i s purpose.
The fade and recovery t e s t , also, differed somewhat from a
121 procedure. Instead of constant deceleration appl ications as
specified by 121, constant pressure appl ications were used. Other-
wise, the procedure was the same: ten fade snubs starting a t 175"F,
followed by a h o t stop, followed by 20 recovery stops. The details
of the t e s t procedure are contained in Appendix A.
The variables recorded during each t e s t were: ( 1 ) brake tor-
qure, ( 2 ) wheel speed, ( 3 ) line pressure, ( 4 ) a i r chamber stroke,
and ( 5 ) drum temperature a t four locations. These variables were
recorded on an osci 1lograph for each brake appl ication.
Three thermocouples were instal 1 ed in each drum with their
measuring junctions positioned approximately 0.040" be1 ow the
braking surface. One thermocouple was placed in the center of the
rubbing path while the other two were placed one inch in from either
edge of the lining. These thermocouples were used in an attempt to
measure the temperature a t the drum/l ining interface. I t i s necessary
t o place a thermocouple in the drum rather than the 1 ining to meet
this objective because approximately 95% of the heat generated dur-
ing braking flows into the drum with typical organic linings and
cast iron drums [4]. The standard SAE 1 ining thermocouple i s a poor
indicator of interface temperature. Placing three thermocouples
across the w i d t h of the lining also allowed observations of tempera-
ture variations across the 1 ining. The thermocouples worked very
well throughout a l l of the t e s t s .
The construction of these drum thermocouples i s shown in
Figure 3.1. Originally designed a t the University of Illinois for
passenger car brake drums [3], they were adapted a t HSRI for use in
commercial vehicle brake drums. They are of a 1/4" diameter plug-
type construction which i s pressed into a through-hole drilled in
the drum, and afterwards ground flush with the braking surface.
The fourth drum temperature was measured by a thermocouple
welded to the periphery of the drum opposite the center of the
rubbing surface.
Figure 3.1 . Schematic of p l u g - type thermocouple
construction.
3.2 P r o c e s s i n g o f t h e Raw Data

To d e r i v e t h e e f f e c t i v e n e s s f u n c t i o n f r o m t h e t i m e h i s t o r i e s
r e c o r d e d d u r i n g t e s t , t h e o s c i 1l o g r a p h r e c o r d i n g s o b t a i n e d d u r i n g
each brake a p p l i c a t i o n were semi-automatical l y d i g i t i z e d on an
X-Y t a b l e . A c u r s o r on t h e t a b l e i s moved by hand o v e r each d a t a
t r a c e w h i l e p o i n t s a l o n g t h e t r a c e a r e sampled and d i g i t i z e d i n t o
X-Y coordinates. The d i g i t i z e d p o i n t s a r e a u t o m a t i c a l l y w r i t t e n
i n t o a computer f i l e . For each brake a p p l i c a t i o n , t h e f i l e c o n t a i n s
a d i g i t i z e d r e c o r d o f each d a t a t r a c e , a l o n g w i t h t i m e s c a l e i n f o r -
mation, b e g i n n i n g and e n d i n g p o i n t s f o r t h e a p p l i c a t i o n , and zeros
f o r each t r a c e .

The i n f o r m a t i o n i n each f i l e i s subsequently processed b y


computer program TRANSLATE. TRANSLATE reads t h e i n f o r m a t i o n s t o r e d
i n a f i l e and t r a n s l a t e s i t i n t o p h y s i c a l u n i t s . The program p r o -
duces two o u t p u t s . The f i r s t i s a p r i n t e d r e c o r d f o r each brake
a p p l i c a t i o n c o n t a i n i n g a r e c o r d o f t h e a p p l i c a t i o n a t 20 e q u a l l y
spaced p o i n t s . Also, some summary i n f o r m a t i o n i s p r i n t e d such as
average t o r q u e and pressure, i n i t i a l and f i n a l speeds, e t c . A
sample page o f o u t p u t i s shown i n F i g u r e 3.2. The second f o r m o f
o u t p u t i s a v e r y d e t a i l e d r e c o r d o f each a p p l i c a t i o n which i s
w r i t t e n on magnetic tape. T h i s t a p e i s l a t e r analyzed by program
DYNA-DRUM I11 which c a l c u l a t e s t h e values o f e f f e c t i v e n e s s a t a s e r i e s
o f p o i n t s d u r i n g each a p p l i c a t i o n .

3.3 Torques Achieved D u r i n g Each T e s t

The minimum and maximum values o f t o r q u e achieved d u r i n g t h e


e f f e c t i v e n e s s p o r t i o n s f o r each t e s t a r e shown i n F i g u r e s 3.3a-c
and 3.4a-c f o r each brake and l i n i n g c o m b i n a t i o n t e s t e d . The range
o f t o r q u e s experienced d u r i n g t h e f a d e and r e c o v e r y c y c l e i s shown
i n F i g u r e s 3.5a-c and 3.6a-c, a g a i n f o r each brake and l i n i n g
combination.

The f i g u r e s show t h a t t h e v a r i a t i o n o f t o r q u e w i t h i n a s t o p i s
much g r e a t e r f o r t h e wedge brake t h a n i t i s f o r t h e cam brake. This
ROCKWELL 1 6 - 1 / 2 X 7 S-CAM B R A K E AHd 6 9 3 - 5 5 1 D L I N I N G
MVYA PROJECT b1.36 GQEENIIdG TESTING L A 8 5 TEST M2-13-15 1 3 J.JLYv1378
POST-BURNISH EFFECTIVENESS

STOP 8 5 0 YPH 2 5 0 DEG F


I T E H P E R A T U R E S
TIHE PRESSURE STROKE SPEED TORO(IE (JEG F 1
(SEC] (PSI) (IN) (CPH) (K I N - L B S ) SURFACE 1 SURFACE 2 SURFACE 3 PERIPHESY

STOP T I M E : 5.84 SEC AVG PRESSURE: 33.5 PSI

SPEED AVG STROKE: 1.32 IN


INITIAL: 51.2 MPH
FINAL: 0.6 HPH TEMPERATURES ( D E G F )
INITIAL MAX RISE
TORQlJE SURFACE 1 224 336 A2
MAX: 78.3 K IN-LBS SURFACE2 250 402 152
HIN: 68.1 K IWL8S SURFACE 3 224 358 134
AVG: 69.9 K IN-LBS PERIPHERY 242 336 54

F i g u r e 3.2. Sample o u t p u t from program TRANSLATE.


Figure 3.3. Minimum and maximum torques achieved during the post-
burnish effectiveness - 15 x 6 wedge.
Figure 3.3 (Cont.)
Figure 3.4. Minimum and maximum torques achieved during the post-
burnish effectiveness - 16 1/2 x 7 S-cam.
#

F i g u r e 3.4 (Cont.)
Q)
L
=I
C',
.r
LL.
behavior indicates t h a t the wedge brake i s more sensitive t o changes
in the coefficient of f r i c t i o n than i s the cam. This behavior i s
expected because the wedge brake has two leading shoes, which i s a
configuration more sensitive t o changes in p than the leading-
t r a i 1i ng shoe construction of the cam brake.
The effects of temperature on brake torque output are evident,
in particular, in the data obtained in the fade and recovery t e s t s
(Figures 3.5 and 3 . 6 ) . As temperature r i s e s , with each successive
fade snub, torque f a l l s . The E80 and E84 linings are especially
graphic in t h i s regard. And as the brakes cool during the recovery,
torque begins t o r i s e again. In f a c t , the brake may actually be-
come more effective toward the end of the recovery than i t was prior
t o the s t a r t of the fade. This behavior i s demonstrated by the
MMSC5 lining in b o t h brakes. The e f f e c t of work history i s i l l u s -
trated by t h i s p h e n o m e n o ~ a m e l y, the brake has changed between
the s t a r t of the fade cycle and the end of the recovery cycle.
4.0 DATA ANALYSIS - DERIVING THE EFFECTIVENESS FUNCTION

4.1 Calculating Effectiveness for Wedge and Cam Brakes


A t each instant during a brake application a value of effec-
tiveness may be calculated corresponding t o the instantaneous values
of interface temperature, sl idi ng speed, and actuation force
occurring a t that time. Recalling that effectiveness i s defined as
torque o u t p u t divided by actuation force, effectiveness for a wedge
brake can be calculated from Equation ( 4 . 1 ) :

where FAC denotes the force produced by the two a i r chambers,


132
Sw denotes the force required t o overcome the return springs, and
a denotes t h o included angle of the wedge.
For a cam brake, the force applying the shoes i s replaced by
the torque actuating the cam. Therefore, effectiveness i s calcu-
lated as

where FAC denotes the force produced by the a i r chamber, a denotes


the length of the slack adjuster, and Sc denotes the torque required
to overcome the return springs.

4.2 Determining the Force Produced by the Air Chambers


Although the force produced by an a i r chamber i s highly linear
with pressure, i t i s decidedly nonlinear with stroke. Graphs of
the behavior of the o u t p u t force for a typical a i r chamber are shown
in Figures 4.la and b. Because i t i s necessary t o know the force
Figure 4.1
produced by the a i r chambers for calculating effectiveness, the
o u t p u t force characteristics of the a i r chambers used in the
dynamometer t e s t s were evaluated with the aid of a t e s t fixture
previously constructed a t HSRI. This fixture uses a load cell t o
measure the o u t p u t force of the a i r chamber while the 1 ine pressure
and stroke are varied over the operating range of the a i r chamber.
A1 t h o u g h the ~erformanceof the a i r chambers tested was
virtually independent of stroke in the vicinity of the mid-range of
stroke, one cannot guarantee that an a i r chamber will always oper-
ate in this range. I n f a c t , the a i r chambers were not always
operating in this range during the dynamometer t e s t s . Therefore,
for purposes of calculating the o u t p u t force for the effective-
ness calculations, the a i r chambers were characterized over their
entire operating range. I t was found that the o u t p u t force could
be represented as varying in a linear manner with pressure and in
a cubic manner with stroke. Thus, equations of the form

(where p denotes pressure and x denotes stroke) were determined for


each a i r chamber by a least-squares regression. For the Type 16
chamber, a least-squares regression yielded the fol lowing equation:

and for the Type 30 a i r chamber:


These equations describe the measured behavior of each a i r chamber
extremely we1 1 .

4.3 Return Spring Efforts


The force or torque which must be exerted to overcome the
return springs i s a factor which i s d i f f i c u l t to estimate accurately
because of the difficulty of determining just when the linings are
brought into contact with the drum. Furthermore, i t i s a factor
which can vary somewhat because of ( 1 ) changes in thickness of linings
and ( 2 ) brake adjustment. However, i t i s n o t necessary t o know this
quantity with great precision because i t i s n o t a large factor in
the net actuation force or torque.
Return spring efforts were estimated from the oscillograph
charts by observing the line pressure and stroke a t the instant the
brake torque began t o rise. Then the return spring force or torque
could be calculated with the aid of Equations (4.1) or ( 4 . 2 ) . A
slight step in the pressure trace occurs a t the time the brake tor-
que begins to rise, and, typically, about 5 psi were required to
overcome the return springs. Table 4.1 l i s t s the return spring
efforts which were estimated for each of the t e s t s .

Tab1 e 4.1. Values of Return Spring Forces and Torques.


Return Spring
Brake Lining Force or Torque
15 x 6 wedge ABB 693-551D 90 l b s ( ' ) 80 l b ~ ( ~ )
MM8C5 80 lbs
E84 60 Ibs

16 1 / 2 x 7 ABB 693-551D 900 in-1 bs


S-cam MM8C5 750 in-lbs
E80 500 in-lbs

( l )post-burnish effectiveness
(elfade and recovery
4.4 Ca7culating Sliding Speed
Sliding speed can easily be calculated a t any instant during
a brake application from the wheel rotational speed and the radius
of the drum, i . e . , v = ~ r .In these t e s t s , the rotational speed
of the drum was n o t recorded, per se, b u t the equivalent road speed
of the wheel was. Therefore, sliding speed, v, can be calculated
as

where V denotes road speed in mph, r denotes drum radius, and R


denotes t i re radius.

4.5 Calculating InterfaceTemperature


A finite-element heat-transfer model of the brake drum i s
used to calculate the temperature a t the interface between the drum
and lining. The model assumes heat flow in radial and axial direc-
tions, b u t none in a circumferential direction. Also, a constant
heat flux i n p u t across the width of the braking surface i s assumed.
The heat transfer model adopted to represent the 16 112 x 7 drum
i s illustrated in Figure 4 . 2 . (The model adopted to represent the
15 x 6 drum i s essentially identical, except for different dimen-
sions. ) The mounting flange of the drums, which would extend down-
ward to the l e f t in the figure, has been ignored. Calculations
showed that very l i t t l e heat actually flows into the mounting flange
during a single snub or stop. Ignoring the mounting flange has a
miniscule effect on the temperature calculated a t the braking
surface.
The temperature calculation i s exercised by supplying a time
history of the rate of heat being input to the drum. Temperature
calculations are made a t time increments of 0.05 second for the
two drum models. The rate of heat input t o the drum i s defined as:
where G denotes the fraction of the heat generated which enters
the drum, and A denotes swept area. A value of 0.95 was used for
a . This value i s quite representative for brakes having organic
1 i nings and cast iron drums.
In addition t o the rate of heat input, the temperature calcu-
lation a1 so requires a starting poi nt-the temperature of the drum
a t the s t a r t of the brake application. The average of the three
temperatures indicated by the sub-surface drum thermocouples a t the
initiation of the brake a ~ p l i c a t i o ni s used for this value. By
the time the brake has cooled t o the point where another applica-
tion can be made, the temperatures measured by the three thermo-
coup1es are approximately the same. For the temperature cal cul a-
tion, a l l the nodes are initialized to this average temperature.
Values of the thermal conductivity and thermal diffusivi t y
of the drum material are also required by the calculation. The
value of thermal conductivity used was 30 BTU/hr-ftI0F, while the
value of thermal diffusivi ty used was 0.43 f t 2 / h r . These values
are similar to values used by other investigators [4,5,6], and they
do provide a reasonable match between calculated and measured
temperatures. Very precise knowledge of these parameters i s not
necessary because the temperature calculation i s not highly
sensitive t o changes in these factors.
The thermal model also allows convective heat transfer t o
take place a t the periphery of the drum. Heat transfer coeffi-
cients were therefore estimated from cool in! curves from the tem-
perature measured a t the drum periphery. For the wedge brake t e s t s ,
the heat transfer coefficient was estimated t o be 17 ETU/hr/ft2/"F,
and for the cam brake t e s t s , 19 BTU/hr/ft2/"F. The cooling curves
were obtained with the brake rotating a t 30 mvh. Calculations show
that very 1 i t t l e cooling takes place during a typical brake
a p p l i c a t i o n , and so a c c u r a t e knowledge of t h e h e a t t r a n s f e r c o e f -
f i c i e n t i s unnecessary. I n f a c t , w h i l e t h e r e may be some c o o l i n g
e f f e c t a t t h e p e r i p h e r y o f t h e brake d u r i n g an a p p l i c a t i o n , t h e
temperature a t t h e b r a k i n g s u r f a c e i s n o t a f f e c t e d .

4.6 Compu t e r Program DYNA-DRUM II 1

A1 1 o f t h e c o m p u t a t i onal elements d i s c u s s e d i n S e c t i o n s
4.1 t h r o u g h 4.5 a r e assembled i n a computer program c a l l e d DYNA-
DRUM 111. The program reads a l l o f t h e parameters d e s c r i b i n g t h e
brake, and t h e n p e r f o r m s c a l c u l a t i o n s o f drum temperature, s l i d i n g
speed, a c t u a t i o n f o r c e / t o r q u e , and e f f e c t i v e n e s s f o r a s e t o f
brake applications. The d i g i t i z e d t i m e h i s t o r i e s produced b y
program TRANSLATE a r e used as i n p u t .

The o u t p u t e x i s t s i n two forms. One i s a p r i n t e d r e c o r d o f


each b r a k e a p p l i c a t i o n f o r 22 p o i n t s i n t i m e d u r i n g each a p p l i c a -
tion. A sample o f t h i s o u t p u t f o r one a p p l i c a t i o n i s shown i n
F i g u r e 4.3. The second o u t p u t saves, on magnetic tape, v a l u e s o f
e f f e c t i v e n e s s , temperature, s l i d i n g speed, and a c t u a t i o n f o r c e a t
21 p o i n t s f o r each a p p l i c a t i o n ( a v a l u e o f e f f e c t i v e n e s s a t
time=O i s n o t d e f i n e d ) . The e f f e c t i v e n e s s d a t a can t h e n l a t e r be
analyzed.

Two v e r s i o n o f DYNA-DRUM I 1 1 e x i s t . One i s a p p l i c a b l e t o


wedge brakes and t h e o t h e r t o cam. Two v e r s i o n s a r e r e q u i r e d
because each t y p e o f b r a k e uses a d i f f e r e n t a c t u a t i n g mechanism.

4.7 Comparison o f Measured and C a l c u l a t e d Temperatures

I t i s , o f course, i m p o r t a n t t h a t c a l c u l a t e d temperatures
c o r r e l a t e w e l l w i t h t h o s e t h a t a r e measured. Comparisons o f c a l c u -
l a t e d and measured temperatures f o r two o f t h e t e s t sequences a r e
shown i n F i g u r e s 4.4a and b. Because t h e thermocouple j u n c t i o n s
were l o c a t e d about 0.040" below t h e b r a k i n g s u r f a c e , t h e c a l c u l a t e d
temperatures i n t h e f i g u r e s a r e t h o s e f o r t h e second node. The
second node c o n t a i n s t h e thermocouple j u n c t i o n s .
* C
aG
x m w
Figure 4.4a. Comparison of calculated and measured temperatures -
cam brake with MM8C5 linings - post-burnish
effectiveness.
Figure 4.4b. Comparison of calculated and measured temperatures -
wedge brake with ABB 693-5510 linings - fade.
I t i s dramatically evident from the figures why measured
temperature cannot be used in deriving the effectiveness function.
The temperature r i s e s vary considerably across the width of the
lining during a brake application. Thus, attempting t o measure
an average temperature across the w i d t h of the lining would be
qui t e d i f f i c u l t , requiring many, many thermocouples or some other
very sophisticated measuring scheme. A1 so, the thermal model shows
t h a t the temperature measured by the sub-surface thermocouples lags
the temperature a t the braking surface. So, even i f there was no
variation in temperature across the w i d t h of the lining, the tem-
perature measured by the sub-surface thermocouples would s t i 11 have
the shortcoming of differing from the temperature a t the braking
surface. The difference i s substantial during the early portion of
a brake application.
With the measured temperatures being a t such variance with one
another, i t i s d i f f i c u l t to assess how we1 1 the calculated tempera-
ture matches the average measured temperature over the width of the
lining. One can only say t h a t the calculated temperatures appear
quite reasonable. Therefore, the va1 idi t y of the thermal model
has been accepted.

4.8 Curve-Fi t t i n g the Effectiveness Data


The final step in deriving the effectiveness function i s curve-
f i t t i n g the effectiveness data produced by DYT!A-DRUM 111. In t h i s
study, the data were f i t t e d by a l e a s t squares prccedure t o a
multinomial of the following form, viz.:

Note t h a t i f two of the independent variables are held constant,


the mu1 tinomial reduces t o a polynomial i n the third variable.
Thus, the values of n i , n and n k can be guessed by plotting the
j'
effectiveness data against one of the independent variables whi l e
holding the other two fixed.
The curve f i t i s performed using a s t a t i s t i c a l analysis
package called MIDAS (Michigan Interactive Data Analysis System)
which resides on the Michigan Terminal System, the computer system
a t The University of Michigan. I'lIDAS possesses the capability t o
perform a mu1 ti-variable 1inear least-squares curve f i t . Because
Equation (4.6) i s nonlinear, i t must be transformed i n t o a multi-
variable 1 inear equation. This i s accomplished by expanding the
mu1 tinomial and making each term of the summation a new independent
variable. For example, i f the values of n i , n and n k are 3, 3,
j'
and 3, respectively, the multinomial will consist of 27 terms.
Thus, the equation can be expressed as a linear equation w i t h 26
independent variables and one constant term. After the data has
undergone t h i s linear transformation, MIDAS can readily calculate
the coefficients which provide the least-squares curve f i t t o
Equation ( 4 . 6 ) . In addition, MIDAS computes the coefficient o f
determination, r 2 , and the standard e r r o r , se. These two numerics
are also very helpful i n determining the values of n i , n and n k .
j,
The values f o r these numbers are picked so that the data i s ade-
quately described with the simp1 e s t equation.
5.0 RESULTS

5.1 Effectiveness Function


Examples of the effectiveness functions which have been cal-
culated from the dynamometer t e s t data are shown graphically in
Figures 5.1 and 5 . 2 . (Detailed presentations of a l l of the
effectiveness functi ons which have been generated are contained in
Appendix B. )
Some observations which can be made for the cam brake w i t h
ABB 693-551D and MM8C5 linings in the post-burnish condition are:

-effectiveness i s most dependent on actuation torque,


followed by sliding speed, and then temperature,

*temperature has very l i t t l e effect, and

*the brake i s more effective a t high sliding speeds


than low sliding speeds.

For the wedge brake with ABB 693-551D and MM8C5 1 inings
during the fade:
*sliding speed has a great influence on effectiveness,
as does temperature,

-again, the brake i s more effective a t high sliding


speeds than a t low speeds, and

the MM8C5 1 ining fades with i ncreasi n~ temperature u p


to about 500°F, and then begins t o get more effective.

Ccefficients of determination and standard errors for a l l of


the regressions performed are shown in Table 5.1. The curve f i t s
describe the data very well as i s demonstrated by the high values
of r 2 and low standard errors. The standard errors are about 3-5%
of the average value of effectiveness for each of the regressions,
ROCKWELL 16-1/2 X 7 S-CRM - RM 693-5510
,POST-9URN
.w -I
ISH EFFECT I VENESS

d
204 300. 400. 500. 600. 700.
CRLCULATED TERPERRTURE (DEG Fl
. .

, R O C K W N 16-l/2 X 7 S-CRM
POST-BURNISH EFFECT IVOESS
- MMBCS

,
(Y*

, u l Run u4.5 P311


sm
8.- +XA 25FP3
IS FP3 9 8 U IN43 M18 PSI1

d 4 1
204 300. 400. 500. 600. 700.
CRLCULRTED TUlPERFlTURE (DEG Fl

Figure 5.1. Effectiveness functions for the can brake with


P.EB 693-551D and MYEC5 linincs post-burnish -
effectiveness.
LD
I atom I

F i g u r e 5.2, E f f e c t i v e n e s s f u n c t i o n s f o r t h e wedae brake with


ABB 693-551D and rlP1EC5 l i n i n r ~ s- fade.
Table 5.1. Values of the C o e f f i c i e n t of Determination and
Standard E r r o r s f o r the Regression Analyses.

C o e f f i c i e n t of Standard
Brake Li ni ng Determi n a t i o n , r2 E r r o r , S,

ABB 693- Pos t-Burni sh 0.980 1.00


551 D Eff.
Fade 0.951 1.20

15x6 MM8C5 Pos t-Burnish 0.984


Wedge Eff.
Fade 0.751 3.02
Post-Burni sh 0.997
Eff.
Fade 0.937

ABB 693- Post-Burni sh 0.951


551 D Eff.
Fade 0.949
MM8C5 Post-Burnish 0,912
16 1/2x7 Eff.
S-cam Fade 0.915
Post-Burni sh 0.980
Eff.
Fade 0.927
which i s extremely good. (The regression for the wedge brake
f i t t e d with MM8C5 linings for the fade data resulted i n a poorer
f i t than the other regressions. I t i s suspected that one of the
thermocouples was not working properly ~ u r i n gt h i s t e s t , so that
the temperatures calculated were in error. )
Attempts to regress the recovery data resulted in mixed results.
I n some cases, the regression was acceptable, in others not. The
reason for the lack of success can be seen in Figure 5.3. Here,
effectiveness i s plotted against stop number a t fixed conditions of
sliding speed and temperature (actuation force i s constant for a l l
of these stops). As can be seen, effectiveness increa;es as the
recovery progresses for about the f i r s t seven recovery stops. Thus,
there are some rapid changes occurring within the brake during this
time which cannot be explained by temperature and sliding speed
alone. This behavior i s an outstanding example of the effect of
work history.
I t i s very d i f f i c u l t t o try t o explain with any confidence
the mechanism which causes the above phenomenon. However, some
reasons can be conjectured. I t i s t h o u g h t that during the burnish,
two processes occur. The f i r s t wears the 1 ining so that i t con-
forms t o the curvature of the drum. The sec~ndestablishes a stable
chemical composition of the lining surface. iiow, i f during the
fade, the temperatures occurri n a a t the braking surface ~ e very t
much higher than they d i d during the burnish, the lining will be-
come chemically active again and i t s surface chemistry will change.
The amount of this change will depend on reaction rates and phase
transformation rates which are i n turn dependent on a time a t
temperature relation. Any changes i n the chemical composition of
the linings will affect the frictional behavior of the linings.
Thus, a fade t e s t can produce a change in the frictional charac-
t e r i s t i c s of the brake. Although this effect i s n o t obvious during
the fade, as evidenced by the very successful regression of the
effectiveness data, i t becomes obvious during the f i r s t part of the
recovery.
Figure 5.3. Effectiveness vs. stop number during recovery t e s t
constant temperature and rubbing speed - 15 x 6
-
wedge ABB 693-551 D.
A second mechanism i s mechanical and thermal distortion of
the brake drum. Distortion wi 11 occur because of temperature
gradients existing in the drum. Changes in the pressure distribu-
tion across the width of the lining will therefore occur which can
change the torque output of the brake. I t i s conceivable, then,
that as the drum i s cooling a t a f a i r l y rapid rate during the early
part of the recovery cycle, the pressure distribution across the
1i ni ng becomes more favorable and effectiveness i s increased,
These two mechanisms point out how sensitive the effective-
ness of a brake can be t o the r a t e a t which temperatures build
d u r i n g a fade cycle and cool during a recovery cycle. If vehicle
t e s t results from a fade and recovery cycle are t o be compared w i t h
dynamometer r e s u l t s , then the work done by the brake and the time
between stops must be quite closely matched between vehicle and
dynamometer t e s t s .

5.2 Simulation of Time Histories of Torque


In order t o evaluate how we1 1 the derived equations can
rep1 i cate the original time histories generated by the dynamometer
t e s t s (from which the equations were derived), a computer program
named DRUMSIN was developed. This program uses a time history of
actuation force taken from a dynamometer t e s t t o calculate a time
history of torque. By means of an i t e r a t i v e process, values of
interface temperature, rubbing speed, and effectiveness are calcu-
lated. O u t p u t torque i s then simply the product of the actuation
force and the computed effectiveness.
Examples of two simulated torque traces are shown in Figures
5.4a and b. The agreement between simulated and measured torque i s
excellent in both cases. The agreement i s especially notable in
Figure 5.4a where the torque i s f a r from constant during the stop.
ROCKWELL 15 X 6 WEDGE - MM8CS
POST-BURN1 SH EFFECTIVENESS
STOP 10 50 MPH 350 OEG F

!i\
a nERs
st ..
C
a sm
L

d .,
gF I

J
.,
w
8 .I

%Is
3

a
*

8
?b

d+ t t t I , 9 t t

a.0 0.2 a4 0.8 0.8 to


NOF1MflLIZEO T IME
STOP TIME - 5.71 SEC
(a)
Figure 5.4. Comparison o f simulated and measured t i m e hi s t o r i e s
o f a brake a?plication on .the d y n a ~ o ~ i e t e r .
d-
4 ) o m
$4.
mm
II

5;
SS

m m
V u

I I
, . *
a2 Q4 Qb 08

m rn -
m 1 m T IK
s.55 SEC
(b)
Figure 5.4. (Cant. )
Torque traces were simulated for a number of other brake a p p l ica-
t i ons and showed excel 1 ent agreement w i t h measured values.
6.0 EMPLOYING THE BRAKE MODEL IN VEHICLE SIYULATIONS

This section i s intended t o provide an overview of how the


brake model described in Sections 2 through 5 would be incorporated
into a large, detailed vehicle simulation such as the one described
in Reference [ I ] .
Obviously, a description of the brakes on each axle would be
required. This description would consist of: (1) the effective-
ness function for the brakellining combination a t each axle for the
appropriate work history being simulated; ( 2 ) actuation parameters
for each brake, that i s , a i r chamber force characteristics and
wedge angle or the length of the slack adjuster arm; and ( 3 )
parameters describing the f i n i t e element model of the brake drum
used for computing the interface temperature of each brake.
During each time step of the simulation, a value of torque a t
each axle would be calculated from the effectiveness function, viz.:

where
T i s torque
F i s the actuation force (or torque)
and e i s the effectiveness function ( e = e(e, v , F)
(Note that v i s equal t o wheel speed, o , multiplied by drum radius,
- r . ) Evaluation of e a t each axle would have to be made a t each
time step. This evaluation would involve a reiterative process in
which values of e , U, and T a t the s t a r t of the time step ( i . e n ,
from the previous calculation) are used in c a l c u l a t i n ~new values of
e, e, and T a t the end of the time step, which i s , of course, the
s t a r t of the next time step.
The value of actuation force, or torque, i s available a t the
beginning of any time step because treadle pressure i s an input
to the program or, in case an antilock system i s used, brake chamber
pressure i s determined in another section of the computer program
dealing with the antilock system. However, assumptions or calcu-
lations concerning stroke in the a i r chambers a t each axle are
required t o determine actuation force.
Typically, the wheel rotational equations of motion are solved
in a separate section of the computer program, thereby providing
values of wheel speed, U, for use in ( 1 ) evaluating the effective-
ness function, e, and ( 2 ) calculating the interface temperature, e .
A major portion of the brake calculation would be concerned
with determining interface temperature. Since the brake model used
a calculated interface temperature in i t s development, i t seems wise
t o use the same temperature calculation method in the overall vehicle
simulation. Nevertheless, i t i s possible that simp1 ified or more
computational ly efficient means of calculating temperature would
be satisfactory as long as they produced results comparable to the
f i n i t e element model. Computer algorithms with simple, efficient
numerical integration methods for solving heat flow problems are
available and were used in DRUMSIN,
The following diagram i 1 lustrates the information flow
proposed for the computation of brake torque, T, a t the time t+a
based on values of T y U, and e evaluated a t time t plus F evaluated
a t t t a . (The quantity A represents the length of the time step in
the digital simulation.) I t i s of interest to observe that the
brake torque calculation uses wheel speed and actuation force as
the only input variables.
1

0.95T(t)w(t)
" Integrate" the heat
# flow equations for
the brake drum from
time t t o t + ~ . *
1

0.95* 8(t+d) e(t)


r t 9 I I
Mu1 t i p l i e r Store e for use
a t time t+a e(t)
h I
. -
* $ 7
Eva1 uate the
effectiveness F ( ~- + A ) from the
function** -
a
w
brake pressure
% subroutine
tion calcu- e(t+A)
1a t i on

t
Mu1 t i pl i er

T(~+A)

I
Store T T ( ~ + A ) t o the wheel
T(t) for use rotation cal-
a t time t + ~ culation
J

*0.95 i s the fraction of total heat flow that


goes into the drum.
**Note that this formulation uses the "old" value of wheel speed,
d t ) , t o evaluate the "new" value of the effectiveness function,
e(t+~r.
7.0 CONCLUDING REMARKS - PROPERTIES OF THE BRAKE MODEL
AND RECOMMENDATIONS FOR RESEARCH

In summarizing the findings of t h i s study, the following


points can be made:
*An empirical model has been developed which treats the
mechanical friction brake in a generalized manner. A so-called
"black box approach" has been used to identify an effectiveness
function which relates brake torque t o three independent variables,
namely, instantaneous values of actuation force [or torque for cam
brakes), sl idi ng velocity , and interface temperature,
*The empirical model provides a valid method for providing a
comprehensive representation of torque time histories obtained
from dynamometer t e s t s a t various i n i t i a l velocities, i n i t i a l brake
temperatures, and actuation pressure levels (see Section 5 , 2 ) ,
* I t should be emphasized that past attempts to organize
dynamometer data into tables suitable for predicting braking per-
formance in emergency stops of commercial vehicles has been con-
founded by (1) the variability of the b r a k ~t e s t data available, (2)
difficulties in assessing interface temperature, and ( 3 ) problems re-
lated t o work history. The regression analysis method used in this
study provides a method for f i t t i n g the t e s t data in a manner which
furnishes measures of the unaccounted for variability. Sets of
data which contain unexpectedly large amounts of unaccounted for
variability can be identified.
*The important influence of work history (that i s , the pre-
vious experience of the brake) i s a major pr~blemwhich affects the
interpretation and representation of data from brake t e s t s . I n
particular, analysis of data obtained during recovery tests indicates
unexplained changes in brake torque which are not readily understood
on the basis of instantaneous values of sliding speed, interface
temperature, and actuation force.
With regard t o future research, further work needs t o be done
on the effects of work history. For example, a post-burnish
effectiveness t e s t t o characterize a brake for stopping distance
performance could be influenced by the order i n which various level
stops were performed. Special t e s t s which randomize the order i n
which stops are made are called for t o evaluate any work history
effect here.
The effects of work history during a fade and recovery sequence
could be examined by varying the time between brake anplications
and/or inertia loading-that i s , anything which would influence
"time-at-temperature" conditions,
And finally, brake torque variability for one brake and 1 ining
needs examining by performing several repeats of a post-burnish
effectiveness t e s t , This could also point t o certain work history
effects which might occur as a brake i s repeatedly tested over the
same t e s t conditions .
I n conclusion, i t appears reasonable t o attempt to understand
and evaluate experimentally the apparent idiosyncracies of
pneumatical ly-actuated brakes using the ern?i,rical model developed
in this study.
REFERENCES

1. W i n k l e r , C. B., e t a1 . , " P r e d i c t i n g t h e B r a k i n g Performance o f


Trucks and T r a c t o r - T r a i 1e r s , " Highway S a f e t y Research
I n s t i t u t e , The U n i v e r s i t y o f Michigan, Report No. UM-HSRI-
76-26-2, a v a i l a b l e from NTIS, PB-263216, June 1976.

2. Cook, H.E., e t a l . , "The Development o f a C l a s s i f i c a t i o n


Scheme f o r Brake L i n i n g s . " F i n a l Report, C o n t r a c t No.
DOT-HS-5-01131, Report No. DOT/HS 8n2 244/MVSS 105-75,
N a t i o n a l Highway T r a f f i c S a f e t y A d m i n i s t r a t i o n ,
February 1977.

3. G i l l e s p i e , T.D., e t a l . , "Modeling t h e In-Stop Torque Per-


.
formance o f Hydraul ic Truck Brakes " Highway S a f e t y
Research I n s t i t u t e , The U n i v e r s i t y of M i c h i gan, Report
No. UM-HSRI-78-58, December 1978.

4. Newcomb and Spurr, B r a k i n g o f Road V e h i c l e s . Chapman and


H a l l , L t d . , London, 1967.

5. Rusnak, R.M., e t a l . , "A Comparison by Thermal A n a l y s i s o f


Rotor A l l o y s f o r Automobile D i s c Brakes." SAE Paper
No. 700137, 1970.

6. Rhee, S.K., e t a l . , "An I n e r t i a l Dynamometer E v a l u a t i o n o f


Three A1 l o y s f o r Automotive Brake Drums. " SAE Paper
No. 700138, 1970.
APPEPiDIX A
INERTIAL DYNAMOMETER TEST PROCEDURE

1. Test Set-Up
1.1 Ambient temperature
The ambient temperature shall be between 7 5 O F and 100°F
1.2 Cooling a i r
Air a t ambient temperature shall be directed continuously
and uniformly over the brake a t a sneed of 2200 fpm.
1.3 Instrumentation
Instrumentation shall be provided t o autographical ly
record the fol 1 owing data:
a ) 1 ine pressure
b) brake torque
c) dynamometer shaft speed
d) 1 ini ng temperature
e) temperature a t braking surface of drum a t three points
f) temperature of periphery of drum in line with the
center of the braking surface
1.4 Wheel 1oad
The wheel load i s 9,000 lb.
1.5 Ti re radius
The t i r e radius i s 20.2 in.
1 .6 Dynamometer speed
The shaft speed of the dynamometer i s given by the
relation rpm = 8.32 V
where V denotes vehicle speed in mph.
1 . 7 Dynamometer inertia
The dynamometer inertia i s 793 t 5% slug-ft2
2. Test Notes
2.1 Initial brake temperature (IBT) i s defined as the
lining temperature a t the time of brake application.
2.2 Specified decelerations are the average value computed
from the time of onset of deceleration t o the time of
completion of the stop or snub.
2.3 The specified line pressures shall be attained in n o t
more than 0.30 sec.
2.4 The brake temperature may be raised to a specified level
by making one or more stops from 40 mph a t a decelera-
tion of 10 fpsps. The brake temperature may be lowered
to a specified level by rotating the drum a t 30 mph.
2.5 Tolerances on the t e s t variables are as follows:
a ) dynamometer speed - +2 mph
b ) line pressure - t 2 psi
c ) deceleration - +I fpsps
d ) i n i t i a l brake temperature - +lO°F

3. Test Procedure
3.1 Instrumentation check stops
Adjust brake. Make 10 stops from 30 mph a t a line
pressure of 30 psi. Correct any malfunctions which occur.
3.2 Burnish
Make 200 stops from 40 mph a t a deceleration of 10 fpsps.
Initial brake temperature shall n o t be less than 315OF
nor greater than 385OF. Make 200 additional stops w i t h
an IBT of not less than 450°F nor greater than 550°F.
Adjust brake. Record f i r s t and l a s t stops in each tempera-
ture series as well as each twentieth stop in each series.
3.3 Post-burnish effectiveness
Establish the line pressures required t o generate a
deceleration of 1 2 fpsps from both 30 mph and 50 mph
w i t h an IBT of 150°F, b u t do n o t exceed 80 psi. Make
f i v e stops from 30 mph a t t h e p r e v i o u s l y - d e t e r m i n e d
l i n e pressure a t I B T ' s o f 150, 200, 250, 300, and
350°F. Then, make f i v e stops from 50 mph u s i n g t h e
l i n e pressure determined above a t I B T ' s o f 150, 200,
250, 300, and 350°F. Record a1 1 stops.

3.4 Fade and r e c o v e r y

E s t a b l i s h t h e 1 i n e pressures r e q u i r e d t o y i e l d decelera-
t i o n s o f 10 fpsps d u r i n g a snub from 50 mph t o 15 mph,
12 fpsps d u r i n g a s t o p from 30 mph, and 14 fpsps d u r i n g
a s t o p from 20 mph w i t h an IBT o f 175"F, b u t do n o t
exceed 80 p s i .

Make 10 snubs from 50 mph t o 15 mph a t t h e pressure


determined above. The i n i t i a l brake temperature f o r
t h e f i r s t s t o p i s 175°F. Make stops a t i n t e r v a l s o f
72 seconds measured between t h e s t a r t s o f successive
stops.

One minute a f t e r completion o f t h e t e n t h stop, make one


s t o p from 20 mph a t t h e pressure determined p r e v i o u s l y .

Two minutes a f t e r t h e previous stop, begin a s e r i e s of


20 s t o p s from 30 mph a t t h e p r e v i o u s l y - d e t e r m i n e d
pressure. The i n t e r v a l between s t a r t s o f successive
stops i s one minute.

Record a11 stops.


APPENDIX B

EFFECTIVENESS FUNCTIONS

T h i s appendix contains d e t a i l e d p r e s e n t a t i o n s o f t h e e f f e c t i v e -
ness f u n c t i o n s obtained f o r each brake. The e f f e c t i v e n e s s f u n c t i o n s
f o r t h e p o s t - b u r n i s h e f f e c t i v e n e s s a r e o f t h e f o l 1owing form:

where
e = e f f e c t i v e n e s s ( i n - 1 b / l b f o r wedge brake;
i n - l b / i n - l b f o r cam brake)

e = i n t e r f a c e temperature (OF)
v = s l i d i n g speed ( f p s )
F = a c t u a t i o n f o r c e / t o r q u e (1 bs f o r wedge brake;
i n - l b s f o r cam brake)

For t h e fade, they a r e o f t h e form:

The format o f each p r e s e n t a t i o n f o r each b r a k e / l i n i n g combina-


t i o n i s as f o l l o w s :

1) designation o f brakeand l i n i n g

2) post-burnish effectiveness f u n c t i o n

i) values o f ni, n and nk


j'
ii) c o e f f i c i e n t s ai jk

iii) d e s c r i p t i o n o f t h e bounds on t h e e f f e c t i v e n e s s
function
3) fade e f f e c t i v e n e s s f u n c t i o n

i) values o f ni and n
j
ii) c o e f f i c i e n t s ai

iii) d e s c r i p t i o n o f bounds on e f f e c t i v e n e s s f u n c t i o n

4) graphical presentation o f effectiveness functions


BRAKE: 15 x 6 Wedge

LINING: ABB 693-551 D

POST-BURNISH EFFECTIVENESS

A c t u a t i o n Force: 1900 l b s -
< F -
< 2900 I b s

S l i d i n g Speed: Lower v = 0 f p s
Upper m i s s i n g

I n t e r f a c e Temperature:
Lower e = 21.841 + 3 8 . 4 5 3 ~ - 4. 3639v2 + 0.076874~3
+ 0.11292F - 0.013242vF + 1.7944 x 10'3v2F
- 0.039306 x 1 0 - 3 ~ 3 F

Upper e = 177.06 + 5 0 . 6 1 3 ~ - 6. 9024v2 + 0.18223~3


+ 0.12349F - 0.016661vF + 2.6490 x ~ o - ~ v ~ F
- 0. O75605v3F
-
FADE

Bounds

A c t u a t i o n Force: F = 2040 I bs

S l i d i n g Speed: 8.4 f p s -
< v -
< 27.2 f p s

I n t e r f a c e Temperature:

Lower e = 507.73 - 2 9 . 0 6 9 ~ + 2.4788v2 - 0.063904v3

Upper e=816.20-31.078v+2.3994v2-0.057517v3
BRAKE: 15 x 5 Wedge

LINING: MM8C5

POST-BURNISH EFFECTIVENESS

= 3 n = 4 nk=2
i j

Bounds

A c t u a t i o n Force: 2200 l b s -
< F -
< 3100 I b s

S l i d i ng Speed : Lower v = 0 fps


Upper v = -9.1292 + 0.011563F

I n t e r f a c e Temperature:
Lower e = -77.632 + 5 2 . 4 1 0 ~ - 6.3180v2 + 0 . 1 4 0 2 6 ~ 3
t0.13541F - 0.017451vF + 2.3302 x 10-3v2F
- 0.056445 x 1 0 - 3 ~ 3 F

Upper e = 86.796 + 4 8 . 9 1 2 ~ - 5.2599v2 + 0.064536v3


+ 0.13481F - 0.015991vF + 2.0068 x 10m3v2F
- 0.033511 x 1 0 - 3 ~ 3 F
-
FADE

ni=3 3 = 3
j

Bounds

A c t u a t i o n Force: F = 1940 Ibs

S1 i d i n g Speed: 8.2 f p s -
< v -
< 27.1 f p s

I n t e r f a c e Temperature:

Lower a = 554.27 - 4 2 . 6 7 2 ~ + 3. 2781v2 - 0.079383v3

Upper e = 721.32 - 3 8 . 1 5 1 ~+ 3.0121v2 - 0.072265~3


BRAKE: 15 x G Wedge

LINING: E84

POST-BURNISH EFFECTIVENESS

Bounds

A c t u a t i o n Force: 2500 Ibs -


< F -< 4400 1bs
S l i d i n g Speed: Lower v = 0 fps
Upper v = 2.7941 + 5.4741 x W 3 F
I n t e r f a c e Temperature:
Lower e=100.88+23.554v-1.7407v2-0.017312v3
-
+ 0.054980F 5.7896 x 10-3vF + 0.58408 x 1 0 ' 3 ~ 2 F
- 3.3088 x 1 0 - 6 ~ 3 F
Upper e = 296.94 + 1 6 . 7 4 5 ~- 1.11 00v2 - 0.027729~3
-
+ 0.045276F 3.8345 x 10-3vF + 0.40654 x 10-3v2F
- 0.15175 x 1 0 - G V ~ F
FADE
-

Bounds

A c t u a t i o n Force: F = 3820 1bs

S l i d i n g Speed: 8.3 f p s -
< v -
< 27.4 f p s

I n t e r f a c e Temperature:

Lower e = 608.43 - 51 .824v + 3. 8026v2 - 0.086479~3

Upper e = 700.88 - 21.521 v + 1. 9087v2 - 0.049422~3


BRAKE: 16 1/2 x 7 S-cam

LINING: ABB 693-551D

POST-BURN1 SH EFFECTIVENESS

ni=3 n = 4 nk=2
j

Bounds

A c t u a t i o n Torque: < F -
4600 i n - 1 bs - < 5300 i n - 1 bs

S l i d i n g Speed: Lower v = 0 fps


Upper v = -18.580 t 8.9819 x I o - ~ F

I n t e r f a c e Temperature:

Lower e = -528.88 t 9 4 . 8 0 8 ~ - 1 3 . 4 7 4 ~ 2 t 0 . 4 3 8 8 2 ~ 3
+ 0.15770F -
0.017088vF t 2.5543 x 10-3v2F
- 0.084440 x 10-3v3F

Upper 6 = -
-222.81 + 7 8 . 9 6 4 ~ 72. 136v2 t 0.39732v3
+ 0.13193F -
0.014139vF t 2.3166 x 1 0 - 3 ~ 2 F
- 0.077037 x 1 0 - 3 ~ 3 F
FADE
= 4 n - 3
i j

Bounds
Actuation Torque: F = 4360 in-lbs
Sliding Speed: 9.3 fps -
< v -
< 30.0 fps

Interface Temperature:
Lower e = 516.09 - 3 4 . 2 8 4 ~+ 2. 3237v2 - 0.050372v3

Upper e = 707.20 - 2 5 . 9 9 2 ~ + 1 .8986v2 - 0.042721 v 3


BRAKE: 161/2x7S-cam

LINING: MM8C5

POST-BURN1 SH EFFECTIVENESS

ni=3 n = 4 nk=2
j

Bounds

A c t u a t i o n Torque: < F-
5500 i n - 1 bs - < 5800 i n - 1 bs

S l i d i n g Speed: Lower v = 0 fps


Upper v = -15.188 + 8.0468 x 10'3F

I n t e r f a c e Temperature:

Lower e = -1089.7 - 5 3 , 0 8 5 ~ + 8.9186v2 - 0.22660v3


+ 0.24346F + 9.1749 x 10m3vF - 1.5275 x 1 0 - 3 ~ 2 F
+ 0.038158 x 10-3v3F
Upper e=-206.76-176.17vt16.020v2-0.34450v3
+ 0.11751F + 0.030882vF - 2.7590 x 1 0 - 3 ~ 2 F
+ 0.058348 x 10'3v3F
FADE
-

Bounds
Actuation Torque: F = 4800 in-lbs

Sliding Speed: 9.3 fps -


< v -
< 30.3 fos

Interface Temperature:
Lower e = 451 - 0 7 - 2 2 . 4 9 7 ~ t 1 .6355v2 - 0. 037834v3

Upper e = 739.17 - 2 8 . 0 2 1 ~+ 1 . 9633v2 - 0. 042949v3


BRAKE: 16 1/2 x 7 S-cam

LINING: E80

POST-BURNISH EFFECTIVENESS

Bounds

A c t u a t i o n Torque: 7100 i n - 1 bs -
< F -
< 9400 i n - 1 bs

S l i d i ng Speed: Lower v=Ofps


Upper v=-3.1815+3.4741 x10'3F

I n t e r f a c e Temperature:

Lower e = -28.063 + 1 7 . 7 1 9 ~ - 2.3759v2 + 0.041770~3


+ 0.034759F - 1.6520 x 1 0 ' ~ v F + 0.28882 x 10-3v2F
- 6.2330 x 1 0 - 6 ~ 3 F

Upper e = 188.68 + 2 3 . 2 8 7 ~ - 2. 9367v2 + 0.063696~3


+ 0.028895F - 2.2759 x 10-3vF + 00.4011 x 1 0 - 3 ~ 2 F
- 8.2651 x 1 0 - 6 ~ 3 F
FADE
-
i
= 4 n = 4
j

Bounds
Actuation Torque: 7610 i n - l b s

Sl id i ng Speed: 9.2 f p s -
< v -
< 30.2 fps

Interface Temperature:
Lower e = 456.99 - 2 5 . 2 0 8 ~ + 1 .8790v2 - 0. 043370v3

Upper e = 635.01 - 1 4 . 0 0 4 ~+ 0.99238v2 - 0.023707v3


ROCKWELL 16-1/2 X 7 S-CAM - RBB 693-5510
POST-BURN ISH EFFECT IVENESS
$1 1

I + 155FPS
A
FPS
X 25 FPS
5.3 K IN-LBS (34.3 PSI1

-200, 300. 400, 500, 600. 700.


CRLCULRTED TEMPERATURE IDEG F1
ROCKWELL 16-1/2 X 7 S-CAM - MM8C5
POST-BURN ISH EFFECT l VENESS
4,
N

t +X 25155FPS
A
FPS 5.8 K IN-LBS (38.8 PSI1
FPS

k
W
0s"

4-

I I 1
I
I 1 I , I
d'r I I I I 8 I I

200. 300. 400, 500. 600, 700.


CRLCULRTED TEflPERFiTURE (DEG F)
ROWELL 16-1/2 X 7 S-CRM - RBE 693-5510
4 FADE (29.5 PSI)
N
T o w n

I
0 l5 FPS
A 20 FP3
* +wm
X30FPS

kld"
.I

4"

..
I I e I I I I e
d+ I I I I i I I I 1 1
200. 300, 400, 500, 800, 7W.
CRLCULATED TEMPERATURE (DEG F)
ROWELL 16-1/2 X 7 %CAM - M8CS
4 FAOE (31,O PSI)
rd
fllom
.. @Em
A 20FPs
+ 25 FPS
x aoRS
.b

4''

4.

d+
I
I
I I I
I
I 1 I
I
. *I
200. 3NL 400, 500, 600. 700,
CALCULATED TEMPERATURE (DEG F)
77
ROCKHELL 16-1/2 X 7 S-CAM - E8O
4 POST-BURNISH
., EFFECT IVENESS
N
RS 7.1 U 1N-m (420 PSI1
. 0 15 FPS
A 5 FPS
d.. +X 25FPS
15 FPS 3.4 K IWBS (55.0 PSI1

EP"
6
\
.-

4"

-
200. 300. 400, 500, 600. 700.
CALCULRTED TEMPERRTURE (BEG F)
d
I
I
.
*
I
I .
I I
I
I
I
,
I
I
I
I

200. 300, 400, 500, 600,


CALCULATED TEMPERATURE (DEG F)
ROCKWELL 15 X 6 WEDGE
POST-BURNISH EFFECTIVENESS
-
ABB 693-551D
d-
1900 LBS (28.3 PSI)
0 15 FPS

$I A 5 FPS
4- 15 FPS 2900 LBS t37.5 PSI)
X 25 FPS

fl
$00. 300, 400. 500. 600, 700.
CALCULATED TEUPERRTURE (DEG F1
ROCKWELL 15 X 6 WEDGE -
MMBC5
POST-BURNISH EFFECT IVENESS
FPs 2200 L83 129.5 PSI)
0 15 FPS
A 5 FPS
+X 25FPS
15 FPS 3100 LBS B7.8 PSI)

I I I I I 1 I I I
I I I

300. 400, 500. 600. 700.


CRLCULRTED TENPERFlTURE (DEG F)
ROCKWELL 15 X 6 WEDGE
FADE (26.8 PSI1
- RBB 693-5510
d~ [21 tO FPS I
t 0 WFPS

+A20FP3
25 FPS

300, 400, 500, 700,


CALCULRTED TEMPERATURE (DEG F)
ROCKHELL 15 X 6 WEDGE
FAE (26.1 PSI)
- MM8CS
&P
tI]!oFPS
Q Ism
" A 20FPS
i= 25 FPS
$*.

9 ..
zi*
I ,,
B
Fa*
.
dm-
..
I
1
I
I
I
,
I I
I
I
I
I
I
I
1
I
I I
%a. 300, 400. 500, 600, 700.
CALCULATED TEMPERRTURE (DEG F)
ROCKWELL 15 X 6 WEDGE - E8U
POST-BURNISH EFFECT IVENESS
dl

s
$
I + 155FF'S
A
FPS
X 25 FPS
4400 LBS (5l.3 PSI1

imH m p
-J
I "
Z
t!
kd-.
W
\
d .. d
I 9 I I I 1 I
I I 1 1 r I I I I

$00. 300, 400, 500, 600, 700,


CRLCULFITED TEUPERRTURE (DEG F1
ROCKWELL 15 X 6 WEDGE
FADE NU.3 PSI1
- E8U
d- a10m
"
0 WFPS
A 20FPS
+ 25 FPS
$-m
D

9 ..
a*B
I ,,

kid*'.
gqrn

# I * * 1 I I I
I I s I I I 1

%o. 300. 400, m. 600, 704


CFlLCULATED TEMPERATURE (DEG F)

You might also like