PB85-161255
Dynamometer Simulation of Truck and Bus Road
Horsepower for Transient Emissions Evaluations
Southwest Research Inst., San Antonio, TX
Prepared for
Environmental Protection Agency
Research Triangle Park, NC
Jan 85
U.S. Department of Commerce
Nations! Technical Information Service
-------
PB85-161255
EPA/600/D-85/011
January 1985
DYNAMOMETER SIMULATION OF TRUCK AND BUS ROAD HORSEPOWER FOR
TRANSIENT EMISSIONS EVALUATIONS
by
Charles M. Urban
Southwest Research Institute
San Antonio, TX
EPA Contract
68-02-3722
EPA Project Officer
Frank Black
\
ATMOSPHERIC SCIENCES RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NC27711
-------
TECHNICAL REPORT DATA
fflreif trail Insiniciioni on Ihe merit bejort committing)
1. REPORT NO.
EPA/600/D-85/Q11
2.
3. RECIPIENT'S ACCt6S»OM *p»f
PB8 5 1 n< 1 2^ 5 /AS
4. TITLE AND SUBTITLE
DYNAMOMETER SIMULATION OF TRUCK AND BUS ROAD
HORSEPOWER FOR TRANSIENT EMISSIONS EVALUATIONS
5. REPORT DATE
January 1985
6. PERFORMING ORGANIZATION CODE
7. ALITHOR(S)
C.M. Urban
8. PERFORMING ORGANIZATION REPORT NO.
S. PERFORMING ORGANIZATION NAME AND ADDRESS
Southwest Research Institute
San Antonio, Texas 78284
10. PROGRAM ELEMENT NO.
C9YA1C/01-1275 (FY-8S)
11. CONTRACT/GRANT NO.
68-02-3722
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Sciences Research Laboratory - RTP, NC
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
13. TVPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCV CODE
EPA/600/09
15. SUPPLEMENTARY NOTES
The relationship between engine power and speed associated with vehicle operation
on a roadway (speed-power relationship) was developed for two truck tractqr-trailers
and one city bus. Results of these determinations, along with data reported in
the literature, were used to determine the power to be absorbed by a chassis
dynamometer to simulate on-road driving of trucks and buses. The chassis dynamometer
simulations are being used in tests to characterize emissions from heavy-duty
vehicles
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lOENTlFIERS/OPEN ENDED TERMS C. COSATI Field/Group
8. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (fins RepOrll
UNCLASSIFIED
21. NO. OF PAGES
19
20. SECURITY CLASS tTlnt pagei
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (R«». 4-77) PREVIOUS EDITION is OBSOLETE
-------
NOTICE
This document has been rex'iewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade nanes
or conaercial products does not constitute endorse-
ment or recommendation for use.
-------
Dynamometer Simulation of Truck
and Bus Road Horsepower for
Transient Emissions Evaluations
Charles M. Urban
Southwest Research Institute
San Antonio. TX
ABSTRACT
Appropriate ch2-_3is dynamometer simulation
of road jx>wer for truck tractor-trailers and
buses wore required for emissions evaluations.
To establish such simulations, the power re-
quired to operate vehicles over a roadway
(speed-power relationship) was determined for
two truck iractor-trailers and one city bus.
Results of these determinations, along with
data reported in the literature, were used to
determine the power to be absorbed by a chassis
dynamometer to simulate on-road driving of
trucks and buses. The chassis dynamometer is
being used in the subsequent phases of this
study involving emissions evaluations of heavy-
duty vehicles.
THE PURPOSE OF THIS PAPER is to describe the
findings associated with road power determina-
tion and simulation for heavy-duty trucks and
buses. Included is a general discussion of
road power, along with the results of evalua-
tions on the road and on the chassis dynamo-
meter. The results have been used to define
the appropriate amount of power to be absorbed
by a chassis dynamometer to simulate on-road
driving of trucks and buses. Appropriate road
power simulation is important for meaningful
emissions evaluations on a chapsis dynamoneter.
Subsequent studies, associated with this pro-
gram, involve emissions evaluations on a chassis
dynamometer.
PROGRAM DESIGI!
The objective of this study was to estab-
lish appropriate dynamometer simulation of road
power for use in emissions evaluations of heavy-
duty trucks and buses. The approach toward
meeting this objective involved review of the
relevant literature, evaluation of several
vehicles on the road, and tVic simulation of
the resultant speed-power relationships on
the chassis dynamometer.
The primary method used in this study for
determination of power requirements on the road
was the "coastdown" method. This involved
determination of the speed-time curve during
coastdown of the vehicle and then using this
relationship, along with the total inertia of
the vehicle, to calculate the speed-power
relationship. Special emphasis was applied
to appropriate interpretation of the resulting
values.
The vehicle:? tested on the road were then
installed onto the heavy-duty chassis dynamo-
meter to establish dynamometer simulation of
the speed-power relationship. In addition,
some effort was made to determine the relation-
ship between tire rolling resistance on the
road and on the dynamometer.
CONDUCT OF THE PROGRAM
This study involved a literature review
and evaluations of three vehicles. The vehicles
evaluated were a single-drive-axle tractor-
trailer, a tandem-drive-axle tractor-trailer,
and a city bus. Evaluations were conducted
with these vehicles operating on the road and
on a programmable heavy-duty chassis dynamo-
meter. These evaluations primarily involved
"coastdowns" of the vehicles.
Two of the vehicles were tested under as
near ideal conditions as practically attain-
able. Analysis of road power data obtained
under ideal conditions (i.e., zero road grade,
zero wind, standard temperature, standard road
surface, etc.) were found to be straightforward
and relatively simple, with one of these
vehicles, good data illustrating the effects
of sidewinds were also obtained.
The waterbrake power absorption units on
the tandem-axle Clayton heavy-duty chassis
-------
I
dynamometer were replaced with eddy current power
absorbers. Electronic programming of t"ie system
enables obtaining essentially any spced-j>ower
curve. By utilizing an electrical signal from
the vehicle braking system, electrical braking
of thj dynamometer rolls is also provided.
Each of tiie absorption units in tandera have
dual rolls that are eight and five-eights inches
in dianeter. Inertia simulation is provided by
an appropriate combination of direct connected
inertia wheels. The inertia wheels and eddy
current power absorbers are shown in Figure 1.
Haxinuir. i..ertia simulations readily attainable
are 49,000 pounds for single-drive-axle vehicles
and 76,000 pounds for tanden-drive-axle vehicles.
DISCUSSION OF ROAD POWER
Proper understanding of the data generated
in this study, or in other related studies, re-
quires some general understanding of road power,
and of the factors associated with its deter-
mination. This section provides a brief over-
view of the subject, and is intended as a
ready source for a general understanding of
road power determination and simulation.
The forces or resistance involved in
operation of a vehicle on the road are as
follows:
Total Resistance* = T + R + A + I+G
Where: T - Transmission and Driveline
Losses
R - Rolling Resistance of the Tires
A - Air Resistance
I - Inertia (affects accels and
decels)
G - Road Grade
•Engine friction can have some
effect—see text.
ENGINE FRICTIO.'J - In coastdown testing, the
tr\rsmission is placed in neutral and engine
friction is not a factor. In a1.! other operation
of tie truck, the engine friction is a factor
if the engine speed on the dynanometer differs
fron that on road. Since s;ippasj and the
effective rolling radius ol the tires, on the
dynanoneter and on the road, do differ.(by an
apparent two to three percent), this is a factor.
In d/'n.inometer operation, relative to operation
on the road, one can run at the same engine
speed or at the same vehicle speed, but not both.
This is a relatively minor factor relative to
total road power requirements, but does become
significant relative to some component parts
of the total road power.
TRANSMISSION LOSSES - The engine speed to
vehicle speed relationships, as discussed in
the previous section, apply to the transmission,
but the affects are of much less significance.
No significant difficulties are foreseen rela-
tive to transmission and driveline losses. When
attempting to actually measure these losses,
however, there are difficulties in actually
simulating tae same conditions that exist in
actual operation on the road.
ROLLING RESISTANCE - The losses due to tire
rolling resistance are a major contribution to
total road power requirements. Difficulties
relative to tire rolling resistance are that
tire rolling resistance on the dynamometer
apparently differs from that on the road, and
that agreement has not been reach ?d on the
relationship between rolling resistance and
vehicle speed. Reported relationships between
total rolling resistance (i.e., tire plus drive
train losses) and vehicle speed are expressed 11.
the equations given in Table 1.
Figure 1 - Chassis dynamometer inertia wheels and eddy
current power absorption units
-------
Table 1 - Equations for Determination of
Rolling Resistance
Equation
Ko.
Equation in form of
F,(c, + r2y)M/c?
(1)
(2)
(3)
(4)
(5)
(6)
(7)
R =
R =
R =
R =
R =
R =
R =
F(10 H
0.70(10 <
F(10 i
F(10 H
0.66(10 ^
0.60(10 ^
0.76(10 ^
^ O.OOOV)W/1000
^ 0.047V)W/1000°
^ 0.050V)W/iOOOb'°
>• 0.100V)V1000C
^ 0.117V)W/100O
I- 0.117VJW/1000
^ 0.118V)W/1000
Refer-
ence
1
1
2
3
4
1
5
Where: R = Rolling drag in pounds
W = Vehicle veight
V = Vehicle velocity in cph
C = Some constant
F = Some factor
3
Based on data from several tire manufacturers
Derived from data given in a Figure
Linear approximation
The effects of vehicle speed on rolling
resistance, as given by these equations, is
illustrated in Figurt 2. As shown, the re-
ported effects of vehicle speed on rolling
resistance are not very consistent.
Rolling resistance is affected by tire
type, temperature and pressure, and by the
condition of the road. Specific, definitive
references have not yet been found for the
effects of these parameters on trucks. For
cars there are considerable data available con-
cerning the effects of type of tire. It is not
k.~.own, however, how these data relate to truck
tires.
It appears that the range of differences
between available tires for trucks is similar
to that for.cars. With cars, the rolling
resistance is reported to decrease by about
0.0 percent for each i"F increase ir. antoient
temperature. (6)* Experiments from tins study,
along with data given in the references,
ir.dicate that the preconditioning operation of
the tires may be an even more important factor
thin ambient temperature. With the equivalent
of about half vehicle payload on a truck tire.
stabilized tire temperature increased by about
5T /or each 10 mph increase in vehicle speed. (1)
Wet roads are known tu increase rolling
resistance, probably due to the cooling effect.
(3) A probable;, but currently undefined,
factor is the effect of sunshine on the tem-
perature of the road surface and the air
ineacdiately above the surface of the road.
AIR RESISTANCE - Air resistance and rolling
resistance are the two primary contributors to
total road power (when operating on level road).
Air resistance is significantly affected by a
r.unber of factors (i.e., vehicle shape, gap
between tractor and trailer, air deflectors,
wind speed, wind direction, air turbulence,
air density, and possibly others). Wind speed
and direction are especially troublesome, since
they have a large effect, they are not con-
trollable, and there are apparently no fully
satisfactory correction factors available.
Also, even if one finds sufficient "no-wind"
days in which to conduct the road testing, the
resulting values will not be representative of
noraal operation. The normal situation is for
there to be wind. (4,7)
2.Or
o
CC LU
<-> 1 5
i.i rr ' • J
I— '-O
e; a
i.oL
EQUATION 5.G.&7
EQUATION 4
EQUATION 2J3
EQUATION 1
20 40
VEHICLE SPEED, flPH
60
Figure 2 - Rolling resistance vs vehicle speed
*NuECers in parentheses designate references
at end of paper
-------
Vehicle Shape - One cab-over-engir.e (CCE)
configuration had 13 percent more air resistance
than a simplified conventional cab. Other COE
configurations had 2 to 14 percent more air
resistance than a simplified conventional cab.
(8) Definitive data, however, does not appear
to be available in the literature.
Gap Between Tractor and Trailer - The effect
of gap between the tractor and the trailer is
significant. This effect on air resistance
appears to be equal to or greater than a 0.4
percent increase in air resistance per 1 inch
increase in gap. (2,8)
Air Deflectors - At zero yaw angle (yaw
angle is a measure of sidewind) , a "standard"
type air deflector is reported to reduce the
air resistance by a little over fifteen percent.
As the yaw angle (sidewind) increased, the air
deflector was less effective. The overall
average for all tests conducted was a five
percent reduction in air resistance. (4) In
an evaluation with the nractor-trailer road
tested in this project, cursory analysis in-
dicateJ a reduction in air resistance of about
ten percent under conditions that approached
the national average wind speed and direction
relative to the road.
Wind Speed and Direction - A more correct
term would be wind speeds and directions. In
real life, wind speed and direction is seldom
constant; this is especially true at very low
wind speeds. Almost instantaneous variations
of plus or minus 100 percent in wind speed and
180 degrees in wind direction were found to be
fairly common.
Wind parallel to the direction of vehicle
travel affects the total air velocity relative
to the vehicle. ' It appears that this effect
can be represented as follows:
Air Res. = Air Res. at 0 Wind*
Where: V = Vehicle velocity
// W = Wind speed parallel to road
This effect does not completely cancel out by
operation in both directions over the road.
With a vehicle speed of fifty mph and a wind
of five mph parallel to the road, the resulting
increase in air resistance (relative to a test
in both directions with zero wind) is about one
percent. The effect at other wind speeds is a
function of the square of the wind speed (e.g.,
four percent increase with a parallel wind of
10 mph, nine percent with 15 mph, sixteen
percent with 20 mph wind) .
A major difficulty associated vith wind
however, is t-'-°. apparent large effect of side
winds. (4,8) Side winds are generally defined
in terms of yaw angle; yaw angle is the direc-
tion of the effective air speed relative to
direction of vehicle travel. Yaw angle can
be determined as follows:
Yaw Angle - Arctan Ijw/ty ± //W)]
Where: 1W = Wind component perpen-
dicular to road
//W •=• Hind component parallel
to road
Based on available data for tractor-trailers,
the air resistance increases about 3.5 percent
(an apparent low of around 1.5 percent and a
high or around 6 or 7 percent) per degree in-
crease in yaw angle above a couple of degrees
of yaw angle. (4,8) At a vehicle speed of 50
mph, the yaw angle increases by approximately
one degree per each mph increase in the wind
speed component perpendicular to the road. As
an example, a 5 mph side wind appears to increase
the air resistance by about eighteen percent at
a tractor-trailer speed of 50 mph.
Using the overall nationwide average wind
speed of 9.5 nph, and assuming equal directional
distribution relative to direction of vehicle
travel, the overall average yaw angle cones
out to be about seven degrees at a vehicle
speed of 50 mph. • Kith tractor-trailers, the
overall effect of the nationwide average wind
speed would be about a fifteen percent increase
in the total power required at a vehicle speed
of 50 mph.
Since accepted correction factors for yaw
angle are not available, mathematical analyses
of road data requires negligible yaw angles
during the collection of the data (negligible
might be defined as less than a half mph side
wind).
Air Turbulence - Air turbulence can also
increase the air resistance of a vehicle. De-
termination of the effect of the atmospheric
turbulence on the vehicle, however, is extremely
difficult. The only presently available method
to account for air turbulence is to conduct the
road evaluations when it is nonexistent or neg-
ligible. In general, turbulence tends to de-
crease with decrease in wind speed. Turbulence,
however, is a function of more than wind speed.
Air Density - The air density is a function
of the temperature, pressure, and humidity. Air
resistance of a vehicle increases as a direct
function of increases in air density. Air re-
sistance data can be corrected to standard
conditions (light-duty applications utilize 68°F
and 29.0 in. Hg) as follows: (6)
Correction to Std. Cond. = (460 + T)/
528*29.0/Baro.
Where: T = Air Temperature, °F
Baro = Barometirc Preosure,
inches Hg
If the air resistance component of the vehicle
can be determined from the -road-test data,
correction to standard conditions is straight-
forward. Fortunately, around San Antonio, con-
ditions of minimum wind are generally associated
with atmospheric parameters that result in a
correction of less than one percent.
INERTIA - Inertia is important in acceler-
ations and decelerations (it does not affect
steady-state operation). The total inertia
(expressed in units of weight) of a vehicle is
equal to the weight of the vehicle plus the
equivalent weight of the rotating components
(primarily the complete wheel assemblies). In
operation on the road, all of the wheels on the
vehicle are rotating. In operation on the dyna-
-------
mometer, not all wheels rotate. The inertia
of a wheel assembly can be determined by labo-
ratory evaluation(9) or can be determined from
the available literature. (1,10) Some values
given in the literature are as follows:
Inertia of Wheel Assy.,
Tire Size lb-in-sec2
11:00X24.5 177(1) (198)(10)
10:00X20 129(1)
The equivalent weight is equal to I/R2, and the
equivalent weights are reasonably similar for
these two sizes of wheel assemblies. The
values result in an equivalent weight of about
150 pounds per wheel assembly (value varies
depending on actual rolling radius of the
tire). Using this value, the total inertia
of a 18-wheel tractor-trailer, with a weight
of 54,000 pounds, would be as follows:
Total Equiv. Weight = 54,000 -t- 2700 +•
Other Inertia
Where: 18><150 = 2700
Other Inertia is considered
negligible.
Of the 18 total wheels, eight rotate during
operation on the dynamometer. The remaining
ten need to be accounted for in comparisons
between coastdowns on the road and on the
dynamometer.
ROAD GRADE - Road grade is a very signi-
ficant factor in road testing of vehicles.
Road testing of vehicles (especially trucks)
required long sections of highway having
uniform grade: a maximum grade of 0.5 percent
and constant within ± 0.1 percent. (1,6) The
effect of 0.1 percent grade (5 feet per mile),
however, is equivalent to a ten percent change
in rolling resistance. Even th:'.s effect can
be significant. Sections of such highway are
reported to be rare(l); this holds true for
the area around San Antonio. This can result
in having to conduct the coastdowns in two or
more parts, which can affect the analyses. If
the road grade is small and constant, its
effects are essentially neutralized by operation
in both directions over the same section of the
road.
METHODS FOR DETERMINING ROAD POWER -
Application of a suitable torquemeter for heavy
trucks, is reported to be very difficult, and
no results lave been published in which a
torquemeter was used to determine aerodynamic
drag. (10) Other ir.ethods, such as measurement
of fuel flow, or other parameters related to
engine power output', does not enable deter-
mination of the actual power required. When
carefully conducted, however, such methods
can be used to transfer operation on the road
to dynamometer operation. Vehicle coastdowns
appear to be a generally accepted method for
determining the road power requirements of
heavy-duty vehicles. Essentially all of the
factors previously discussed equally affect
coastdown and steady-state evaluations.
The application of the coastdown method is
straightforward, provided you have a long section
of level highway (zero grade) and ideal atmos-
pheric conditions (primarily no wind). Obtaining
all of these conditions simultaneously is re-
ported to be extremely difficult. (1,4,7) This
was also found to be true for the evaluations
conducted in this study. Ironically," such ideal
conditions produce results that are somewr.at
incongruous to actual vehicle operation.
In forty cities at seven o'clock in the
morning, the average wind speed was about 7
mph(7). Daybreak is generally the calmest part
of the daylight hours. Assuming a normal wind
distrubtion, that seven minus zero mph is equal
to three standard deviations, and no bias in
wind direction relative to the road; the wind
perpendicular to the road would be less than
one-half mph about two percent of the time
(i.e., about one day cut of fifty would have
ideal wind conditions). Experience around
San Antonio in the spring make the two percent
appear optimistic; ideal conditions for a long
enough period to obtain data were essentially
nonexistent. In summer, however, such periods
of "no wind" occurred a little rrore often than
two percent of the time.
One appr .ach toward defining road power
is to measure all of the individual components,
except for air drag, and to conduct coastdown
tests to determine the total power required.
The air drag is then determined by subtracting
ths power required by the components from the
total. The expected results are illustrated by
data given in Reference 1. In tl'at reference,
data was obtained on a vehicle at several
different vehicle weights. One w-.uld expect
the air drag to remain nearly constant, or
at least consistent (i.e., configuration having
highest air drag at high speed to have highest
air drag at slower speeds). The actual results,
however, were a plus or minus seven percent
variation in air drag at 60 raph and a plus or
minus sixteen percent variation at 30 mph.
ROAD POWER EVALUATIONS
Road power evaluations were conducted with
two truck tractor-trailers and one city bus.
The more extensive and better data were generated
with the second truck tractor-trailer and the
bus.
TANDEM-AXLE TRUCK TRACTOR - Initial road-
load data were obtained using an IHC truck-
tractor with a 44 foot long, tandem-axle, en-
closed trailer. This unit is shown in Figure 3
and described in Appendix A. The trailer was
loaded, as required, with concrete blocks to
simulate an empty trailer, half payload (70
percent of the rated GCW), and rated GCW. These
initial road-load data were obtained under less
than ideal conditions due to the inclement
weather that occurred during the evaluations.
The evaluations on the road with this
vehicle consisted of steady-state operation and
coastdowns. On the-basis of dynamometer settings
-------
• _
required, the steady-state and the coastdown
data generally agreed within seven percent or
less. The results obtained, however, are
higher than the reported or calculated results
based on ideal wind and weather conditions.
These data uncorrected for wind, along with
data from two references, are given in Table 2.
'
• ! ]
1.
Figure 3 - Tandem-axle truck tractor with
tandem-axle trailer
Table 2 - Road Horsepower of Tandem-Axle Truck Tractor
Total Road Horsepower
Data From this Study3 Ref. 11 Ref. 2<3
34000b5400078000 5400QC 32000d (54000)e(54000)*
30
40
50
49
61
130
60
95
145
77
122
180
135
70
113
92
145
88
135
Cab-over-engine tractor with wind deflector and tandem
.axle trailer
Vehicle total weight in pounds
Based on equations in the Recommended Procedure
Thermostatically controlled radiator cooling shutters
covered due to considerable effect on drag
Data adjusted to 54,000 pound truck weight
Estimate for with wind deflector
The data generated on this vehicle were with a
wind speed of around ten miles per hour blowing
at an angle of about 45 degrees relative to the
road. Interpolation of the empty, half, and
full load data, results in a horsepower value
at half load between 145 and 150. This value
is about 10 percent higher than the 135 horse-
power values derived from References 2 and 11.
This difference is likely due to the wind that
occurred during these evaluations.
Based on the experience with this vehicle,
it was primarily concluded that the following
were either very desirable or necessary: a
test road less than one hour driving time from
,
-------
the laboratory; more stable weather conditions,
(i.e., summer); and provisions for conducting
the evaluations at a time of essentially zero
wind speed. All of these criteria, along with
a number of other equipment and operating im-
provements, were incorporated prior to con-
ducting the evaluations with the other two
vehicles.
TEST ROAD AND CRITERIA - The initially
selected road for vehicle testing was located
about ninety miles south of the chassis dyna-
mometer laboratory. This made it difficult
to assure the atmospheric conditions in that area
and resulted in a major effort even when testing
had to be aborted because of unfavorable con-
ditions. Therefore, a determined effort was
again made to locate a suitable road for the
vehicle testing. Airport runways and other
non-road surfaces had already been excluded
during the initial search as not being available
or suitable.
A section of the access road to an inter-
state highway (Interstate 10) was subsequently
selected. This section of road is located
about thirty-five miles east of the laboratory,
a distance just long enough to provide sufficient
warm-up of the vehicle. The roadway was accept-
ably flat, the surface was in good condition,
and there was essentially no traffic. The absence
of traffic was because of the low population
density in that area and the fact that this
access road dead-ends about a mile from the
section utilized. The average grade (or slope)
of the 0.9 mile section used was 0.11 percent,
and the maximum variation was plus or minus 0.11
percent (0.11 percent is equal to six feet per
mile).
Since the minimum wind speed generally
occurs around daybreak, provisions were made
to begin the actual testing at daybreak (safety,
annoyance, and operating considerations pre-
cluded testing while it was still dark).
fleasurement of fuel flow at constant vehicle
speeds, or of some parameter representative of
fuel flow, is time consuming and does not
enable calculation of the total road power
requirement. (It can, however, provide useful
data for setting a load into the chassis dyna-
mometer). Therefore, with the very minimal
amount of "no-wind" test time available, only
coastdowm evaluations were subsequently con-
ducted with the two remaining vehicles.
A "fifth-wheel" was used to measure the
speed of the vehicle. Three precautions were
found to be essential in using a fifth-wheel:
absolutely rigid mountinu to the vehicle fraae
(vibration shows up as speed variation), elim-
ination of electronic noiso (without excessive
damping of the response), and frequent cali-
bration checks. Calibration of the velocity
of travel was conducted using distance versus
time measurements. Distance measurement was
calibrated prior to this, with the "fifth-wheel"
mounted on the vehicle, using a precisely
measured distance on the road.
Wind speed and atmospheric temperature a'ld
huaidity were measured at the test site. Baro-
metric pressure was measured at the laboratox/,
after 't was determined that this method pro-
vided t.ie required accuracy. The rotating vane
wind speed instrument used had a readability of
one mile per hour and an accuracy of better
than one mile per hour at low speeds. Tempera-
ture and humidity were determined using wet
and dry bulb thermometers.
SINGLE-AXLE TRUCK TRACTOR - The single-axle
truck tractor and single-axle trailer evaluated
are shown in Figure 4 and described in Appendix
B. The tractor was of conventional design and
the tractor-trailer had a GCW of 42,000 pounds.
The evaluations were conducted at tractor-trailer
combined weights of approximately 25,COO and
41,200 pounds. These weights were selected
on the basis of the maximum variation that could
be obtained without repeated rearrangement of
the inertia wheels on the chassis dynamometer.
Based on an equivalent inertia of 165 pounds in
each wheel assembly on the truck, the total
equivalent inertias were 27.3QO and 42,900
pounds.
At the lower vehicle loading, coastdowns
were conducted from an initial speed above 60
mph to a final speed below 20 mph. At full-
load, two coastdown segments were required,
and speeds above 60 mph were not practical to
obtain. Therefore, the combined coastdowns at
full-load were from about 55 mph to below 15
raph. An attempt was made to obtain five coast-
do'~-ns in each direction with each vehicle
loading. The mimimum number was three coast-
downs producing repeatable results.
It was determined that the available methods
for processing the data, such as that given in
Reference 9, did not account for side-winds or
for variation in road grade (slope), and there-
fore, were not directlv applicable. Therefore,
a method was developed that involved calculation
of the road power at each five mile per hour
increment in vehicle speed; this enabled incor-
porating side-winds and road grade into the
calculations. The method is summarized in
Appendix D. Results of the data obtained, along
with calculated horsepower values, are summarized
in Tables 3 and 4.
To put these data into perspective, the
following discussion is presented. About five
percent difference in horsepower at 50 mph (e.g.
105 vs 100 hp) results from any of the following:
• 0.1 j% Road Grade
• 2 -h Side Wind
• f. n Head Wind
• ..iph Vehicle Speed
Therefore, a five percent difference in road
horsepower at a vehicle speed of 50 mph is
actually not very significant. Kith light-duty
vehicles, the EPA generally accepts data which
differs by less than seven percent (A/C No. 5^j)
16) in the direction opposite to that produced
by these previously listed factors.
-------
FJ.gu«.o 4 - Single-drive-axle truck tractor with
single-axle trailer
Table 3 - Road-Power Evaluatior.s-TT Road 2
Vehicle
Speed
mph
50
45
40
35
30
25
20
Calculated Hp Using
Road Hp
41200
127
103
83
65
4G
34
25
from Coastdowns
25600
105
80
60
47
35
26
18
A
22
23
23
18
14
8
7
Recommended Procedure
41200
120a
96
76
60
46
34
25
25600
iooa
79
61
46
34
25
17
A
20
17
15
14
12
10
8
1
Values calculated at 50 mph using EPA Recommended Procedure
and assuming constant rolling resistance:
Total Hp = 0.67*(Height - 0.75)xwidth*(V/503 + 0.00125*weightx(V/50)
Table 4 - Comparison of Road and
Calculated Horsepower
Vehicle
Speed
mph
50
45
40
35
30
25
20
Avg.
Road
41200
1.06
1.07
1.09
1.08
1.06
1.00
1.01
1.05
Hp/Calculated
25600 A
1.05
1.01
0.98
1.02
1.02
1.07
1.07
1.03
HD
vg.a
1.06
1.04
1.04
1.05
1.04
1.04
1.04
1.04
Provided as a matter of potential
interest
The data in Table 3 indicate that for the
tractor-trailer tested, the EPA Recommended
Practice(11) provides a good estimation of the
road horsepower required at a vehicle speed
of 50 mph. Also, these d
-------
ta.icv (•;.«•!» all of t.'^r effect of •:;«s fci>.-;r-«•
i.jic. 1J:1» va!u«r atti-*.-* • it:.
i-J t;.r«c A:..! >.*.»!-.• il s «.rr«rf.' f
titijutcd oo tf>e «.i»i* of t:»-
i--. t.'.v Ktrf
TV! jc j - f f feel vt Sliie ».r.}» or K>*4 lv*xr r
At A '.,.'_::« '•• i ::.t •, ? ."« ?
-.Xr.ii-le
h4 «•.
M
4.
!
-
e^
:.',
t •
JS
.6
Avq.
; ••»
l •*
: . -»
:: »
i -•»
1..%
or, th<
iw t r*r*-tr »vtor *••• ev«:^it»j c^« t.
./ *nwt foj vO^lir.'j of tn« tiiv* oe t.Nt drlw
it i-. ?»r. in f*i<;ar« *. A!*o ».VJ«TI IB
of in* lir«*
Figur* S - Si&9l»-r>»l*t«r.t
rty mr.-jt«« at 4v »th, or tii*
**ch **r.c* of co«»t icwr.i,
TTw r««vilt»
cr> tft
d 7.
indicate Ihat «ouiw, it.-jt n>!ativ«ly r^r.or. dif-
f«r*rvcr»«f»J t
of
in
if t^
oAJy-current
On «.*•* c'-.aitu* £-fr.*im?'.fT, tiie "rolling
cv* d>»s ro* a;f«ar to i< Iin*ar> it
to lr.cr*a»« »* t-'-c velocity lr.cr*<>«*.
(It «7i«*r«. t-jt t>*»e data not [.rove, that
rolling ro»lKt «r.c* is «fMmtialiy conitant on
the ro*d and it a function of velocity ot. the
crvnamnrter. If t?.;» !• true, it would explain
th« faJKXj f c r lh* r«icirn of rolling rv«i»tanc« t.k.at «««r«
found :n thv lll«i »tirr ar.lpr»iri*trai*t that, if tt
-------
10
Table 6 - Dynanonuter Coast down Evaluations
26200 Inertia
^-•t Up Ca!c. froa the Coastdowns with 26200 Inertia
50 U>jd(TI'» 0 LoadtT.i-1 50-0* Dynanogetor (D)"
60
50
40
JO
20
10
".,6.5
42.6
30.9
21.0
12.7
5.7
53.1
39.6
28.7
19.8
12.2
5.7
3.4
3.0
2.2
1.2
0.5
0.0
11.5
8.0
5.6
3.4
1.8
0.7
attained with the truck on the dynanometer
Values outjined witliout a vehicle on the dynamometer;
diflurtncca at O and SO load settings were essentially neglicible
T*blo 7 - Power Absorbed by Truck Tractor Tires and Drive-Train
Xclativc Hp with 50 r.ph ° 1
Dyna.
Sj.ccd,
gj n
6O
50
43
Jj
20
10
Ket Calc.
(T) - (D
50 Load O
45.0
34. £
25.3
17.6
10. ->
5.0
Hp
)a
Load
41.6
31.6
21.1
16.4
10.4
5.0
(T) -
Calculated
50 Load 0 Load
1.30
1.00
0.73
0.51
0.32
0.14
1.32
1.00
0.73
0.52
0.33
0.16
(6.4 * 0.074 V) Linear
1.29 1.20
1.00 1.00
0.74 0.30
0.51 0.60
0.31 0.40
0.14 0.20
(T) is with truck on the dynamometer
(D) is withouv •* vehicle on the dynaooaeter
Tablu 8 - Comparison of Road and Dynamometer Coastdown Tines
Dyna.
Speed,
c^h
50
40
30
20
Road
Hp
37.2
53.1
35.3
13.8
Dynamometer Hp
S.R. *• Trie.-1
44.5
31.0
20.7
12.5
Absorbed Up
53.2
26.5
14.8
6.2
Tc^al
97.7
57.5
35.5
18.7
Coastdown Tine,
Sec.
Road
0
19.4
43.7
73.9
0
19.5
43.8
73.5
'calculated using extrapolation of the data obtained at 0 and 50
dynanoaeter load settings.
trailer on this ciynaraocieter. The dynanoacter
horsepower settings «;tSO cph for half-load are
57 based on coastdown evaluations and 71 based
c« the value calculated using the EPA Recora-
w;n<5ed Procedure.
Therefore, it was decided that data from
coastdowrs on the dynaaoseter be determined
and used with each vehicle evaluated. However,
data required to use the method given in the
EPA Reco^nended Procedure will be obtained
and recorded to enable subsequent; comparisons
of the nethods in a planned future paper.
-------
CITY BUS - Road horsepower evaluations
were conducted on a bus obtained from VIA
Metropolitan Transit in San Antonio. This bus
is shown in Figure 7 and is de-scribed in
Appendix C. In the road horsepower evaluations
with this bus, good data were obtained at two
load conditions under essentially no-wind
conditions. The results of the road-power
evaluations with the bus arc summarized in
Table 9. These data indicate that the EPA
Recommended Procedure for trucks somewhat over-
states the horsepower required with a bus. It
appears reasonable to assurx? that these dif-
ferences are due to the difference in air
resistance between a truck and a bus and that
the rolling resistance per unit of weight is
about the same for a truck and a bus. Using
such assumptions producas results given in
Table 10.
These data indicate that the force re-
quired to overcome rolling resistance can be
assumed to be essentially constant and that
U
the air resistance force with a bus is lower
per unit of frontal area than with a truck. The
air resistance force per unit of frontal area
with a bus appears to be about 0.85 as great
as that for a truck. This seens reasonable
since the bus is significantly nore streamlined
than a truck. Reference 3 reported an air resis-
tance coefficient of 0.6 to 0.7 for a bus, and
0.8 to 1.0 for a truck.
DYNAMOMETER SIMULATION
Using a prograsnable dynamometer, the pro-
cedure developed for road load simulation of
a vehicle on the dynamometer involves estab-
lishing the speed-power curve, dt-'tt'rnination of
inertia bimulation. and determination of system
friction.
SPEED-POWER CURVE - The equation selected
for calculation of the sj^eed-power curve to be
used for evaluations on the chassis dynanoneter
is as follows:
••'
Figure 7 - city Bus
Table 9 - Road-Power Evaluations - Bus Road 3
Vehicle
Speed,
vfh
50
45
40
35
30
25
20
Road Hp from
Coastdovnsa
31700
64.0
67.1
53.2
41.8
32.6
25.1
19.1
25700 '
73.9
59.O
46.6
36.3
27.8
20.9
15.3
A
10.1
7.9
6.6
5.5
4.8
4.2
3.8
Rec. Proc. for Trucks
31700 25700 A
88.5
81.0 7.5
Based on best curve fit throuoh the individual data at the
various speeds.
-------
12
Table 10 - Air and Rolling Resistance of a Bus
Rolling Resistance Hp
at 31700 Ibs
50
45
40
35
30
25
20
41.4
30.2
21.2
14.2
8. 9
5.2
2.6
iKR by
Difference^
42.6
36.9
32. C
27.6
23.7
19.9
16.5
Calc. by
EPA Proc.c
40.8
36.7
32.6
28.5
24.4
20.4
16.3
(iRR - EPA) *
Total H;.-*l','Oj
2.1
0.3
-1.1
-2.1
-2.1
-2.0
1.0
<£RR - EPA)
84.0-100e
2.1
0.2
-0.7
-1.1
-0.8
-0.6
•••0.2
tolling Resistance Up
Vehicle
Sp<-ed,
cph
50
45
40
35
30
25
20
a.
Average
for tho
Air
Resis.
Ho*
41.4
30.2
21.2
14.2
8.9
5.2
2.6
of the total
at
RH b
31700 Ibs-
y Calc. by
Difference"1 EPA Proc.c
32.5
23.8
25.4
22.1
18.9
15.7
12.7
rainus
two- loads evaluated
33.3
29.9
26.6
23.3
20.0
16.6
13.3
(iHR - EPA) i
Totil Kp»100»
-1.0
-1.4
-2.6
-3.3
-4.0
-4.3
-3.9
the calculated rolling resistance
and value adjusted
(ARK - EPA) f
73.9-!00\
-1.1
-1.1
-1.6
-l.o
-1.5
-1.2
-o.a
ho r ac (xwe r va 1 u« a
for best fit to a cubic c-quation
Total horsepower oinus air resictan-e horsepower.
Calculated at 50 cph using EPA Rccoraended Proceduio and assucing co.ista.it
.rolling resistance forc«.
Difference in (road) and calculated rolling resistance horsepower divided by tho
total horsepower at the respective vehicle speed.
Difference in the rolling resistance values divided by the total horsepower at
50 Ef>h.
RU> m pxo.67(H-0.75)Kx(v/50)3 +
0.00125XLVW/50
Who re«
RLP • Road Load Power in horsepower
F • 1.00 for trartor-trailer and
0.85 for city bus
H » Average oaxinun height in feet
K » Average raaxissun width in feet
LVW * Loaded vehicle weight in pounds
On ti.e clayton dynar»aeter with eight and
five^eights inch dianetor rolls, the equation
for determination of dynaaoncter torque «nu
load are as follows:
Dynanonoter Torque » Kpx134.8/Eph, foot-rounds
Dynanoneter Load • Torque'<12/ (Load Ana in
inches), pounds
1IER7IA SIMULATION - In keeping with the
general provision in the EPA Recoiaaended Pro-
cedure, (11) the equivalent inertia to set in
the dynanosieter system for evaluation of a
tractor-trailer is to be equal to 70 percent
of the gross coiabined weight. For buses the
equivalent inertia is to be equal to the sua
of the enf-ty weight, half passenger load plus
the driver (at 150 pounds per person), and the
equivalent inertia weight of t:vc- nonrotating
vehicle wheel jSicrJDl les. A deviation equal
to one percent of the total iiivrtia, rather
than the 250 pounds specified in the EPA Fecos-
rended Procedure, will 1» allcr.-t;d.
For actual inertia simulation on the
chassis dynarost;ter, the inertia of the •aeel
assemblies on the vehicle have to bo accounted
for. The resultant dynanor»?ter inertia is as
follows:
Total Inertia - F.ID + EIK
Where i
EID * Equivalent inertia of dynar»»-ter
system, pounds
EIW e Equivalent inertia of rotating
wheels
This total inertia is to be used in the deter-
aination of system friction.
SYSTE.X FRICTION - With the vehicle installed
onto the dynanoneter and with the appropriate
inertia wheels connected, the total system
absorbed horsepower will be determined using
coastdovns. This can be accomplished by
-------
13
obtaining repeatable 60 tc 5 nph coastdown speed
vs time data and then solving for the instan-
taneous decelerations. From the instantaneous'
decelerations, the power absorption of the
vehicle-dynamometer system can be determined
as a functior. of vehicle speed. The speed-
power curve for programming into the dynamometer
controller can then be dete.Tained by difference
between the total power required on the road
and the power absorbed by the vohicle-dynamo-
raeter systen.
The method is briefly described as followsj
(1) Obtain 60 to 5 nph coastdown data
(2) Obt< in acceleration using the
following equation:
Where:
ao and
resistance
represent rolling
a2V2)
a,V represents air resistance
V = velocity of vehicle
Note: An acceptable alternate method is to
graphically determine and calculate the accel-
eration at each five nph increment in vehicle
speed.
(3) Calculate the power absoroed using
the acceleration values, K = ea, and Hp =
F*oph/375.
(<) Develop the specd-pover curve for
programming the dynamometer by subtracting
the power absorbed by the vehicle-dynamometer
systea from the total power n.-quired on the
road.
(5) Calculate the speed-load curve to
progran. into the dynamometer.
SUMMARY AND CONCUJSIOHS
An improved road-load simulation method
has been developed for use in operating large
trucks on a chassis dynamometer. The purpose
of these improvements is to permit more real-
istic laboratory simulation of the way Class
VII and VIII diesel tractors and city buses
perform on the road. These improved pro-
cedures will be used in a laboratory investi-
gation of regulated and unregulated emissions
from large diesel trucks and buses operated
over a transient driving cycle.
Analytical and experimental studies were
performed to determine mathematically, under
essentially ideal environmental conditions,
the truck or bus power-speed characteristics.
A city bus, a single-drive-axle tractor-trailer
(Class VII truck), and a tandcm-drive-axle
tractor-trailer (Class VIII truck) were oper-
cted on-the— road under as near— ideal environ-
mental conditions as possible. The 'coastdown"
method (time to decelerate froa one speed to
a lower speed) was used to compute road horse-
power. With the. tractor-trailer trucks evalu-
ated, the power required on-the-road at fifty
miles per hour generally agreed with values
obtained using the appropriate portion of the
equation given in the EPA Recommended Procedure
for heavy-duty vehicle testing on a chassis
dynamometer. Tne equation given in the EPA
Reconoended Procedure does.not define the power
requirements at other vehicle speeds.
From the road evaluations, a generalized
expression for determining road horsepower at
various vehicle speeds was developed. By use
of tho proper vehicle weight, frontal area and
coefficients, a road-load can -be computed.
Next, the test vehicle is operated on the
chassis dynamometer to determine the power
absorbed by the drive train, the tires, the
dynarorwter bearings and by tire and inertia
system windage. This is accomplished by running
repetitive coastdowns. This absorbed ,.over is
then subtracted from the total power required
on-the-road, to determine the power values for
programming into the controllable power absorp-
tion unit on the chassis dynamometer.
A major finding of the study was the signi-
ficant effect that r.on-ideal environmental
conditions have on road-power. Side winds are
especially significant, and morely operating
both ways over a level course does not cancel
out the effect. Since ideal conditions are
not the norm, it is concluded that road-power
relationships currently used arc conservative.
From the limited Uata obtained with side winds
present, it appears that the use of ideal
conditions (i.e., no wind, etc.) results in
horsepower values that are ten to fifteen
percent lower for tractor-trailer truck* at
half payload.
Developing a generally accepted solution to
the question of the effects of non-ideal con-
ditions associated with operation on-the-road
was beyond the scope of this study. Therefore,
the primary evaluations involved data obtained
ori-the-road under essentially ideal operating
conditions. The data obtained on-the-road and
the method developed for programming the speed-
power relationship into a controllable chassis
dynamometer will be used in subsequent emissions
studies. Results of these emissions measure-
ments will be the subject of a future technical
paper.
ACKNOWJJiDGEKEHT
This paper is based on work performed under
Task 1 of EPA Contract 68-02-3722. The assis-
tance provided by Sherrill Martin and Robert
Howard in developing the dynamometer control
system, and by Jitnmic Chessher and Ernie Krueger
in conducting the vehicle coastdown evaluations,
is gratefully acknowledged.
REFERENCES
1. J.W. Anderson, et al, "Truck Drag
Components by Road Test Measurement," Paper 881A
presented June 1964-at SAE Summer Meeting.
2. L.C. Montoya and L.L. Steers, "Aero-
dynamic Drag Reduction Tests on a Full-Scale
Tractor-Trailer Combination with Several Add-On
Devices," NASA Publication TM X-56028,
December 1974.
-------
3. J.J. Taborek, "Mechanics of Vehicles-
5," Townotor Corp., Cleveland.
4. F.T. Buckley, Jr., et al, "Analysis of
Coast-down Data to Assess Aerodynanic Drag
Reduction of Full-scale Tractor-Trailer Tvucka
in Windy Environments,"Preprint of SAE Paper No.
760107, 1976 Automotive Engineering Congress and
Exposition, Detroit, February 1576.
5. SAE Reconnended Practice J6B8, "Truck
Ability, Prediction Procedure," 1982 SAE Hand-
book Part 2.
6. OMSAPC Advisory Circular Mo. S5B,
"Determination and Use of Alternative Dyna-
moneter Power Absorption Values," December 6,
1978.
7. M.J1. Ingalls, "Estimating Mobile Source
Pollutants in Microscale Exposure Situations,"
EPA 460/3-81-021, July 1981.
8. H. Flynn and P. Lyropoulos, "Truck
Aerodynamics," General Motors Corp.
9. G. Thompson, "Prediction of Dynaao-
meter Power Absorption to Siculate Light-Duty
Truck Road Load," SAE Paper 770844.
10. W.H. Halston, Jr., et al, "Test Pro-
cedures for the Evaluation of Aerodynaaic Drag
on Full-Scale Vehicles in Windy Environratnts,"
SAE Paper 76C106, presented at Autoootive
Engineering Congress and Exposition, Detroit,
February 1J76.
11. C.J. France, et ,\1, "Recoesiended
Practice for Determining Exhaust Ecusnio.-.s
fron Heavy-Duty Vehicles Under Transient
Conditions," Technical Report SD5B 79-80,
Environmental Protection Agency, Ann Arbor, MI.
-------
15
APPENDIX A
DESCRIPTION OF TANDEM-AXLE TRACTOR-TKAII.OR
Description: 1981 IHC Transtar II Truck Tractor with
Cab Over Sleeper
Transmission: Eight Forward Gears
Chassis: Tandem Drive Axles
CCW: 78,000 pounds
Tires: 11R24.5X (measured 20.5 inch Rolling Radius
at Half Load)
Trailer: Utility Trailer VIN 7L8 1719 006 Semi
13.3 feet high by 8.1 feet wide by 44 feet long
Horizontal Ribs on the Sides
12 inch Radius on Front Vertical l.dges
60 inches Cap between Tractor and Trailer
APPENDIX B
DESCRIPTION OF SINGLE-AXLF. TRACTCT-TR/.l '.ER
Description:
Transmission:
Chassis:
CCW:
Tires:
Trailer:
1981 IHC S2500 Truck Tractor with
Conventional Cab
Fuller Road Ranger RT9509 - Mire Forward Gears
Single Drive Axle - 158 Wheel Base
42,000 pounds (with single Axle Trailer)
11R22.5 (measured 19.5 inch Roiling Radius at
Half Load)
13.3 feet high by 8.0 feet wide by 26 feet long
Flat Top-Smooth Sides ( no ribs)
12 inch Radius on Front Vertical Edges
38 inches Cap between True cor and Trailer
APPENDIX C
DESCRIPTION OF CITY BUS
Description:
Transmission:
Chassis:
GVW:
Tires:
1981 CMC RTS II Bus, Modc-l T70204
Three Speed Automatic
Single Drive Axle - Dual Wheels on Rear
36,000 pounds
11.00-22 (measured 21.5 Rolling Radius at 28,000 VW)
Goodyear City Cruiser on Fro.it, Good-fear Super
Hi-Mi ler on Rear
-------
16
APPENDIX D
METHOD USED FOR EVALUATION OF ROAD COASTDOWN RESULTS
1. Starting at the highest vehicle speed to the nearest 5 raph, determine
the time required, to decelerate in 5 raph increments. This was done
for each of the coastdown charts produced in both directions of
travel.
2. Taking into account the wind speed and direction during each
coastdown, determine and eliminate any outliers in the data.
3. Determine a mean value most representative of a no-wind condition
and of some specific ambient temperature.3 Then average the mean
values for the opposing directions of travel.
4. Determine total horsepower at each 5 mph increment of vehicle
speed using: Hp = (FxV)/375 = 1.215xlO~4xInertiaxAxV, "here
A * ((V + 5) - (V - 5))/At V + 5
5. Calculate approximate air resistance horsepower at 50 mph using the
formula given in the EPA Recommended Procedure and adjust to ambient
conditions experienced during the coastdowns. Calculate approximate
rolling resistance by difference. Assuming air resistance horsepower
is a function of V-* and rolling resistance as a function of V,
calcul-te air and rolling resistance horswpower at the other vehicle
speeds.
6. Determine horsepower corrections for air resistance corrected to 68°F
and 29.0 inches Hg, rolling resistance corrected to 68°F, and road
grade to zero net percent using:
Air Hp T & B Corr. = (((460 + T)/528x29.0/Baro.) - l)*Air Hp
R.R. Hp Temp. Corr. = 0.005XR.R.X(T - 68)
Grade Hp = 26.7xlO~6xWxVx% Grade
Where: T = Air Temperature in °F
Baro. = Barometric Pressure in "Hg
R.R. = Rolling Resistance
W = Weight of the vehicle
V = Vehicle velocity in mph
Z Grade = Net effective road grade
7. Add the horsepower corrections to the total horsepower values determined
in Step 4. These values represent the driving horsepower required to
operate the truck at the respective speeds under "standard" conditions
of 68"F, 29.0 inches of mercury, "no-wind", and zero road grade.
a
For data with a side wind, a mean value most reprcsentativp
some specific side wind was used.
------- |