550/9-74-005a
BACKGROUND DOCUMENT/
ENVIRONMENTAL EXPLANATION
FOR
PROPOSED INTERSTATE RAIL CARRIER
NOISE EMISSION REGULATIONS
June 1974
OFFICE OF NOISE ABATEMENT AND CONTROL
WASHINGTON, D.C. 20460
-------
FOREWORD
The study of railroad noise is relatively new. Most of the information and d:ita contained in
this report has been generated during the past year. It is important to note that this report and the
proposed regulations are an initial step in a continuing effort to understand and reduce railroad
noise.
The Agency wishes to acknowledge the cooperation of a multitude of parties and to extend
its appreciation for their efforts. Those parties include, but are by no means limited to, The
Department of Transportation, Association of American Railroads, the Department of Commerce,
and the National Bureau of Standards.
-------
TABLE OF CONTENTS
Section Page
1 PROLOGUE
Statutory Basis for Action 1-1
Internal EPA Procedure 1-2
Preemption 1-2
2 DATA BASE FOR THE REGULATION 2-1
3 THE RAILROAD INDUSTRY 3-1
Economic Status 3-1
Employment 3-3
Health of the Industry 3-4
Growth 3-7
4 RAILROAD NOISE SOURCES 4-1
General 4-1
Consideration of Railroad Noise Sources for Federal Regulation 4-2
Office Buildings 4-3
Repair and Maintenance Shops 4-3
Terminals, Marshaling Yards, and Humping Yards 4-4
Track and Right-of-way Design 4-5
Horns, Whistles, Bells, and Other Warning Devices 4-5
Special Purpose Equipment 4-6
Trains 4-7
Character of Railroad Noise Sources and Abatement Technology 4-7
Locomotives 4-7
Diesel-Electric 4-7
Locomotive at Rest 4-8
Locomotive in Motion 4-14
Locomotive Noise Abatement 4-1 6
Abatement by Equipment Modifications 4-16
Noise Abatement by Operational Procedures 4-25
Electric/Gas Turbine 4-26
Wheel/Rail Noise 4-26
Whed/Rail Noise Abatement 4-29
Retarder Noise 4-31
Retarder Noise Abatement 4-33
-------
TABLE OF CONTENTS (con't)
Section Page
Benefits 4-33
Costs 4-33
Car-Car Impact Noise 4-34
Warning Devices 4-37
Public Address Systems 4-37
Maintenance and Repair Shops 4-37
Refrigerator Cars 4-37
5 SUMMARY OF WHAT THE PROPOSED REGULATIONS WILL REQUIRE 5-1
"Application of Best Available Technology Taking Into Account
the Cost of Compliance" 5-1
Levels of Train Noise Control 5-2
Locomotive Noise: Vehicle at Rest 5-2
Locomotive Noise: Vehicle in Motion 5-3
Railcar Noise: Vehicles in Motion on Line 5-4
Railcar Noise: Vehicles in Motion in Yards 5-4
6 GENERAL PROCEDURE TO MEASURE RAILROAD NOISE 6-1
Introduction 6-1
Measurement Instrumentation 6-1
Test Site Physical, Acoustical, Weather and Background Noise Conditions 6-2
Procedures for the Measurement of Locomotive and Railcar Noise Emissions 6-5
Introduction 6-5
General Requirements 6-5
Locomotive Load Cell and Self-Load Noise Emission Measurement 6-6
Locomotive Pass-by Noise Emission Measurement 6-6
Rail Car Pass-by Test Noise Emission Measurement 6-6
7 ECONOMIC EFFECTS OF A RETROFIT PROGRAM 7-1
Introduction 7--
The Impact on the Railroad Industry 7-1
General Impact 7-1
The Impact on Marginal Railroads 7-15
The Impact on Bankrupt Railroads 7-18
The Impact on Users of Rail Transportation 7-18
The Effect on Railway Freight Rates 7-18
-------
TABLE OF CONTENTS (con't)
Section ' Page
The Effect on Quality of Service 7-24
Summary and Conclusions 7-25
Impact on the Railroad Industry 7-25
Impact on Users of Rail Services 7-26
8 ENVIRONMENTAL EFFECTS OF PROPOSED REGULATIONS 8-
Introduction 8-
Impact Related to Acoustical Environment 8-
Case Studies of Railroad Lines o
Analysis of Train Noise Impact 8-
Impact Related to Land • 8-10
Impact Related to Water 8-10
Impact Related to Air 8-10
Enclosure A: "Day Night Equivalent Noise Level" (LDN) 8- 2
Enclosure B: Excess Attenuation of Railroad Noise 8- 2
9 SELECTION OF THE PROPOSED REGULATIONS 9-
Problem Addressed and Approach 9-
Problem Addressed 9-
Approach 9-1
Regulatory Approaches Considered 9-1
"Status Quo" Regulations Alternative 9-1
Future Noise Standards Regulations Alternative 9-2
Noise Reduction in Combination with Status Quo Regulations Alternative 9-2
Regulatory Approach Selected by EPA 9-2
Discussion of Proposed Regulations 9-3
Locomotive at Rest , 9-3
Locomotive in Motion 9-3
Rail Car 9-3
REFERENCES R-l
APPENDIX A: MAJOR TYPES OF DIESEL-ELECTRIC LOCOMOTIVES
IN CURRENT U.S. SERVICE A-l
APPENDIX B: REVIEW OF THE USE OF AUDIBLE TRAIN MOUNTED
WARNING DEVICES AT PROTECTED RAILROAD-
HIGHWAY CROSSINGS B-l
111
-------
TABLE OF CONTENTS (con't)
Section Page
B. 1 Requirements of Use of Audible Devices B-1
B.2 Railroad-Highway Accidents B-8
B.3 Impact and Effectiveness of Train Horns B-l 2
B.4 Prohibition Against the Use of Audible Devices B-20
B.5 Judicial Background B-23
APPENDIX C: OPERATING RAILROAD RETARDER YARDS
IN THE UNITED STATES C-l
APPENDIX D: SUMMARY OF YARD NOISE IMPACT STUDY D-l
IV
-------
SECTION 1
PROLOGUE
STATUTORY BASIS FOR ACTION
Through the Noise Control Act of 1972 (86 Stat. 1234), Congress established a national
policy "to promote an environment for all Americans free from noise that jeopardizes their health
V
and welfare." In pursuit of that policy, Congress stated, in Section 2 of the Act, "that while pri-
mary responsibility for control of noise rests with State and local governments, Federal action is
essential to deal with major noise sources in commerce, control of which requires national uniformity
of treatment." As a part of this essential Federal action, Section 17 requires the Administrator to
publish proposed noise emission regulations that "shall include noise emission standards setting such
limits on noise emissions resulting from operation of the equipment and facilities of surface carriers
engaged in interstate commerce by railroad which reflect the degree of noise reduction achievable
through the application of the best available technology, taking into account the cost of compliance."
These two sections of the Act establish the criteria the Administrator has followed in the
development of these proposed regulations. Section 17 does not contemplate the promulgation of
regulations covering every aspect of the massive, complex interstate railroad industry, but only
those on noise emissions from particular equipment and facilities of that industry. The types of
equipment and facilities to be covered by Federal regulations are those that are "major noise
sources in commerce," which require "national uniformity of treatment." The need for national
uniformity of treatment depends largely upon interference with interstate commerce that would be
caused by the lack of national uniformity. Regardless of whether or not there are Federal regula-
tions on noise emissions from any type of interstate railroad equipment or facility under Section 17,
the states and localities are barred by the Commerce Clause of the Constitution from imposing any
regulations that would constitute an undue burden on interstate commerce.
Regulations under Section 17 are to be promulgated after consultation with the Secretary
of Transportation in order to ensure appropriate consideration for safety and technological avail-
ability. They are to take effect after such period as the Administrator finds necessary, after con-
sultation with the Secretary of Transportation, to permit the development and application of the
requisite technology, giving appropriate consideration to the cost of compliance within such period.
Final regulations are to be promulgated within 90 days after publication of the proposed regula-
tions and may be revised from time to time in accordance with Subsection 17(a)(2) of the Noise
1-1
-------
Control Act. These regulations under Section 17 of the Noise Control Act shall be in addition to
any regulations that may be proposed under Section 6 of the Act.
Section 17(b) of the Noise Control Act requires the Secretary of Transportation, after con-
sultation with the Administrator, to promulgate regulations to ensure compliance with all standards
promulgated by the Administrator under Section 17. The Secretary of Transportation shall carry
out such regulations through the use of his powers and duties of enforcement and inspection
authorized by the Safety Appliance Act, the Interstate Commerce Act, and the Department of
Transportation Act. Regulations promulgated under Section 17 shall be subject to the provisioi
of Sections 10, 11, 12, and 16' of the Noise Control Act.
INTERNAL EPA PROCEDURE
The rulemaking process of EPA started with the publication of an Advanced Notice of Proposed
Rulemaking in the Federal Register. At that time EPA informed the public of the requirement that
regulations be developed and requested that pertinent information be submitted to the Agency for
consideration. In the case of interstate rail carrier regulations, a task force was formed about the
same time and was composed of Federal, State, and local government officials and consultants.
The Office of Noise Abatement and Control considered recommendations of the Task Force with
the recommendations of the EPA Working Group, which is comprised of represent atives from
various parts of the Agency, in developing the proposed regulation. After the Deputy Assistant
Administrator for Noise Control Programs approved the proposed regulations, they were submitted
to the Assistant Administrator for Air and Waste Management Programs, who has responsiblity for the
Noise Control Program as well as several other programs. Following the Assistant Administrator's
approval, the proposed regulations were submitted to the EPA Steering Committee, which is com-
prised of "' .he Deputy Assistant Administrators of EPA. Upon the Steering Committee's approval,
the proposed regulations were forwarded to the Office of Management and Budget, and other
interested Federal agencies, for review. After these comments were analyzed and satisfactorily
addressed, the proposed regulations were submitted through the Assistant Administrator for Air
and Waste Management Programs to the Administrator for final approval and ultimate publication
in the Federal Register. The resulting public comments will be analyzed and a recommendation
for the final regulation will be prepared by the Deputy Assistant Administrator for Noise Control
Programs. The review process followed in the case of the proposed regulation will then be initiated
again, culminating in the promulgation of the regulation.
PREEMPTION
Under Subsection 17(c)( 1) of the Noise Control Act, after the effective date of these regula-
tions no State or political subdivision thereof may adopt or enforce any standard Applicable to
noise emissions resulting from the operation of locomotives or railroad cars of surface carriers
1-2
-------
engaged in interstate commerce by railroad unless such standard is identical to the standard pre-
scribed by these regulations. Subsection 17(c)(2), however, provides that this section does not
diminish or enhance the rights of any State or political subdivision thereof to establish and enforce
standards or controls on levels of environmental noise, or to control, license, regulate, or restrict
the use, operation, or movement of any train if the Administrator, after consultation with the
Secretary of Transportation, determines,that such standard, control, license, regulation, or restric-
tion is necessitated by special local conditions and is not in conflict with regulations promulgated
under Section 17.
Conversely, Subsection 17(c)(l) does not in any way preempt State or local standards appli-
cable to noise emissions resulting from the operation of any equipment or facility of interstate
railroads not covered by Federal regulations. Thus, under the proposed regulations, the States and
localities will remain free to enact and enforce noise standards on railroad equipment and facilities
other than trains without any special determination by the Administrator. Only after a Federal
regulation on noise emissions resulting from the operation of a particular type of railroad equip-
ment or facility has become effective must the States and localties obtain a determination by the
Administrator under Subsection 17(c)(2) where it is believed that special local conditions
necessitate particular consideration.
Some types of railroad equipment and facilities on which no Federal noise standards or regu-
lations have become effective, and which may, therefore, be subjected to State and local noise
standards without any special determination by the Administrator, may include other types of
equipment or facilities that are covered by preemptive Federal regulations. Railroad maintenance
shops, for example, may from time to time emit the noise of locomotives undergoing tests along
with noises common to many industrial operations such as forging and grinding. Also, railroad
marshaling yards include locomotives among their many types of noise sources.
In most instances, State or local standards on non-Federally regulated equipment or facili-
ties of railroads can be met without affecting the Federally regulated equipment within them.
Standards on noise emission from repair shops, for example, can be met by many measures includ-
ing improved sound insulation in the walls of the shop, buffer zones of land between the shop and
noise-impacted areas, and scheduling the ^operation of the shop to reduce noise at those times of
the day when its impact is most severe. Standards on railroad marshaling yards can be met by
a variety of steps including: reducing the, volume of loudspeaker systems by using a distributed
sound system or replacing speakers with two-way radios, reducing noise emissions from equip-
ment not covered by Federal regulations, installing noise barriers, acquiring additional land to
^
act as a noise buffer, and locating noisy equipment such as parked refrigerator cars or idling
locomotives as far as possible from adjacent noise-sensitive property. Since State or local regu-
lations on noise emissions from railroad facilities that the railroad can meet by initiating measures
such as these are not standards applicable to noise emission resulting from the operation of
locomotives or railroad cars, they would hot be preempted by the proposed regulations. Thus
1-3
-------
no special determination by the Administrator under Subsection 17(c)(2) would be necessary.
State or local noise standards on facilities involved in interstate commerce such as railroad
marshaling yards are, of course, subject to Constitutional prohibition if they are so stringent as
to place an undue burden on commerce.
In some cases, however, a State or local noise standard that is not stated as a standard appli-
cable to a Federally regulated type of equipment or facility may, in effect, be such a standard if
the only way the standard can be met is by modifying the equipment to meet the Federal standard
applicable to it. This would be the case, for example, if after the proposed regulations become
effective a State or locality attempted to adopt or enforce a limit on noise emissions from railroad
rights-of-way in urban areas that could not reasonably be met by measures such as noise barriers.
Such a standard, would, in effect, require modifications to trains even though they met the Federal
standards, and would be preempted under Subsection 17(c)(l). It could not stand if it differed
from the Federal standards, unless the Administrator made the determinations specified in Sub-
section 17(c)(2). The same would be true of any State or local standard on railroad yards that
could not reasonably be met except by modifying locomotives or railroad cars subject to the
Federal standards.
State or local use or operation regulations directly applicable to noise emissions resulting from
the operation of Federally regulated equipment and facilities can, of course, stand if the Adminis-
trator makes the determinations specified in Subsection 17(c)(2) regarding them.
State or local noise emission standards directly applicable to noise emissions resulting from the
operation of Federally regulated equipment and facilities may also stand without any special deter-
mination by the Administrator if those standards are identical to the Federal standards. By adopt-
ing such identical standards, States and their political subdivisions can add their enforcement
capability t^ aat of the Department of Transportation. The Environmental Protection Agency
recommends and encourages such adoption of standards identical to the Federal standards.
1-4
-------
SECTION 2
DATA BASE FOR THE REGULATION
The program for compiling data on train noise began with a search for already existing data.
By compiling the existing data, it was possible to avoid repeating the few measurements completed
by others, and the limitations of the existing data indicated what measurements needed to be made
to extend the data. Technical journals were searched for reports of pertinent measurements. Pub-
lished accounts of measurements in Europe and Asia were considered along with the accounts of
measurements in the United States and Canada. A bibliography of relevant articles appears after
Section 9.
Much of the needed data was obtained by the EPA Regional Offices and under contract by
acoustical consultants. Some data were obtained through informal communication with members
of the acoustics community to obtain unpublished accounts of measurements and proceedings of
appropriate seminars. Leaders in the engineering departments of the two locomotive manufacturers
that remain in business (Electro-Motive Division of General Motors—EMD, and General Electric—GE)
were also interviewed in order to ascertain the extent of their data files, as well as to determine what
problems may be created by attempts to control locomotive noise. At a meeting hosted by the
Association of American Railroads, EMD and GE engineers reported measurements of locomotive
noise and discussed some possible effects of locomotive noise controls. Three leading muffler manu-
facturers (Donaldson, Harco Engineering, and Universal Silencer) were contacted in order to evaluate
the feasibility and the impact of fitting locomotives with exhaust mufflers.
Railroad company personnel who worked in various capacities at various levels were contacted
in order to determine the mix of equipment used by railroads, the configurations of properties and
equipment, scheduling of operations, and modes of operation. In particular, yard masters, yard
superintendants, or engineering personnel were contacted to obtain information about yard configu-
ration, layout, and equipment. Railroad personnel were asked for information related to schedules
and speeds of trains. The railroad companies that participated are listed in the bibliography at the
end of this report.
2-1
-------
SECTION 3
THE RAILROAD INDUSTRY
ECONOMIC STATUS
There are currently 72 Class I railroads in the U.S.* These tend to break down into two
groups: large transportation companies such as the Union Pacific or the Penn Central and railroads
that are owned by large industrial firms such as U.S. Steel. The latter roads primarily provide trans-
portation services to the "parent company." Since railroads are regulated by the Interstate Com-
merce Commission (ICC), the degree of competition is also regulated. The size of the firms has in
many cases been determined by whether the ICC has allowed or disapproved mergers. Most large
roads have grown through mergers. In addition, the favorable financial position of some roads
results from their nontransportation activities.
The total tonnage of freight moved in the U.S. has been rising over time, but the transportation
sector of the economy has declined in relative importance. In 1950, 5.6% of national income
originated in the transportation sector; by 1968 this figure declined to 3.8% and has remained at
about that level. This trend reflects the higher relative growth rates in those industries that require
a smaller transportation input.
The rail industry has declined more rapidly than the transportation sector. In 1950
the rail sector constituted 53% of the national income originating in the transportation sector. By
1968 it had declined to 25.8% of the transportation sector and has remained relatively stable since
then. Table 3-1 summarizes these statistics.**
Accompanying the decline in the rail sector's share in national income originating in the trans-
portation sector, the proportion of total freight hauled by rail has declined. In 1940 the railroads
hauled 63.2% of all freight, dropping to 44.7% by 1960 and 39.9% by 1970. Motor carriers and oil
pipelines have rapidly increased their share during this period. Air freight has increased more rapidly
than either motor carriers or pipelines but it accounts for only . 18% of total freight. In spite of the
decreasing proportion of shipments by rail, the total volume of freight hauled by rail increased from
41 1.8 million ton miles in 1940 to 594.9 in 1960 and to 768.0 in 1970. Table 3-2 summarizes
these trends.
*Class I railroads are those having annual revenues of $5 million or more. They account for 99%
of the national freight traffic.
!*Unless otherwise stated, the data presented in Tables 3-1 through 3-6 were obtained from the
Statistical Abstract of the United States (1971 and 1972).
3-1
-------
TABLE 3-1
NATIONAL INCOME ORIGINATING IN THE TRANSPORTATION AND RAIL SECTORS
($ In Billions)
Year
1950
1960
1965
1968
1969
1970
National
Income
$241.1
414.5
564.3
712.7
769.5
795.9
Transportation
$13.4
18.2
23.2
27.1
29.2
29.5
Transportation
as % of
National Income
5.6%
4.5
4.1
3.8
3.8
3.7
Rail
$7.1
6.7
7.0
7.0
7.4
7.2
Rail as 7r of
Transportation
53.0%
36.8
30.2
25.8
25.3
24.4
TABLE 3-2
INTERCITY FREIGHT (In Millions of Ton Miles)
Year
1940
1956
1960
1965
1968
1969
1970
Total Freight
Volume in
106 Ton Miles
651.2
1376.3
1330.0
1651.0
1838.7
1898.0
1921.0
Rail Freight
in lO6^
Ton Miles
411.8
677.0
594.9
721.1
765.8
780.0
768
Rail
%
63.2
49.2
44.7
43.7
41.2
41.1
39.9
Motor
Vehicles
%
9.5
18.1
21.5
21.8
21.6
21.3
21.44
Oil
Pipelines
%
9.1
16.7
17.2
18.6
21.3
21.7
22.4
Air
%
.002
.04
.06
.12
.16
.17
.18
Inland
Water
%
18.1
16.0
16.6
15.9
15.9
15.8
15.98
Rail passenger service declined from 6.4% of intercity travel in 1950 to less than 1% in 1970.
The real impact of railroads on the national economy is in terms of freight rather than passengers.
The decline of the rail industry's share of the transportation sector is less dramatic when passenger
service (air, local, suburban, and highway) is eliminated from calculations. Table 3-3 gives the
transportation sectors' percentage contributions to national income, less the passenger sectors men-
tioned above, and the rail industry's percent of the transportation sector.
From comparison of Tables 3-1 and 3-3, it can be seen that the freight sector has declined more
rapidly than the total transportation sector. It can also be seen that the railroads' decline is some-
what less dramatic in terms of freight alone than in terms of both freight and passenger service.
3-2
-------
TABLE 3-3
PERCENT OF NATIONAL INCOME ORIGINATING IN THE
TRANSPORTATION SECTOR (LESS AIRLINE AND LOCAL
SUBURBAN AND HIGHWAY PASSENGERS) AND THE
RAIL SECTOR AS A PERCENT OF TRANSPORTATION
Year
1950
1960
1965
1968
1969
1970
Transportation* (Adjusted)
as%of
National Income
4.8%
3.7
3.3
3.0
3.0
2.9
Railroads
as % of
Transportation
(Adjusted)
61.7%
44.1
37.6
33.0
32.3
Not
Available
^Transportation minus air carriers and local suburban and highway passengers.
EMPLOYMENT
The railroads' importance as a source of employment within the economy has decreased along
with their share of the nation's transportation output. In 1950 the railroads accounted for 2.7% of
all employees in nonagricultural establishments. By 1970 this had fallen to less than 1%. Not only
has the relative importance of railroads declined but also the absolute level of employment from
1950 to 1970 decreased by over 50%, as shown in Table 3-4.
Wages in the rail sector have consistently been above the average of all manufacturing employees
and this differential has increased over the years. In 1950 the average hourly compensation in the
rail sector was $ 1.60, which was 110% of the average hourly compensation in manufacturing. In
1968 average compensation was $3.54, or 118% of that in manufacturing. By 1971 rail compensa-
tion had increased to 126% of the average compensation in the manufacturing sector.
Increases in wage rates in the rail sector have been greater than the increases in the wage rates
in the manufacturing sector. Using 1967 as the base (= 100), the index of wage rates in manufac-
turing in 1970 was 1 21.6 while the rail industry index was 125.6. Over the same period the increase
in productivity in the rail industry has been less than productivity increases in manufacturing. In
1970 the index of output for all railroad employees was 109.9* while in manufacturing it was 111.6
(using a 1967 base of 100). Table 3-5 summarizes the wage and productivity data.
''Computed on the basis of revenue per man hour.
3-3
-------
TABLE 3-4
EMPLOYMENT IN THE RAIL INDUSTRY
RELATIVE TO THE NATIONAL ECONOMY
Year
1950
1960
1965
1968
1969
1970
National Employees
in All
Nonagricultural
Establishments
(1000)
45,222
54,234
60,815
67,915
70,274
70,664
Railroad
Employment
(1000)
1220
780
640
591
578
566
Railroad
as%of
National
2.7%
1.4
1.1
.9
.8
.8
TABLE 3-5
INDEX OF OUTPUT PER MAN HOUR AND WAGES
(1967=100)
Year
1950
1960
1965
1968
1969
1970
Rail Wage
41.5
74.3
88.9
106.3
113.6
125.6
Manufacturing
Wage
44.7
76.6
91.2
107.1
113.9
121.6
Rail
Productivity
42.0
63.6
90.8
104.4
109.3
109.9
Manufacturing
Productivity
64.4
79.9
98.3
104.7
107.7
116.6
The fact that productivity increases have not kept pace with wage rate increases indicates
that unit labor cost is rising.
In the years since 1970, wages in the rail industry have, as in most industries, increased rapidly.
The index of wages in 1971 was 136.8;in 1972, 136.8;andin 1973, 165.4 (estimated).
HEALTH OF THE INDUSTRY
There are a number of measures one might use to judge the "health" or financial stability of
the rail industry. Two of these are the rate of return on stockholders' equity and the percent of
3-4
-------
revenue carried through to net operating revenue. Shareholders' equity is the excess of assets over
liabilities, which is equal to the book value of capital stock and surplus.
In 1971 the rate of return on stockholders' equity for all manufacturing firms was 10.8%. The
rates of returns in some selected industries are as follows:
instruments, photo goods, etc. 15.8%
glass products • 11.1%
distilling 9.9%
nonferrous metals 5.2%
The return for the total transportation sector was 3.1%. Railroads showed a 2.1% on stockholders'
equity, slightly above the airlines' 2.0%.
The rate of return on stockholders' equity increased from 1.3% in 1971 to 3.0% in 1972. The
use of industry data, however, tends to give a misleading impression of the industry.*
The Eastern District had a negative rate of return for the three years from 1970 to 1972 while
both the Southern and Western Districts had positive and increasing rates of returns. The Southern
District showed an increase from 5.2% to 6.1% and the West from 3.7 to 5.1%. The rates of returns
in these districts are well above the 3.1% for total transportation and are about equal to the textile
and paper industries.
These trends indicate that the problem in the rail industry is not with all districts but primarily
with roads in the Eastern District. Using operating ratios** as the measure of financial stability,
one draws the same conclusions.
The historical trends in the profitability of the industry can be measured by the percent of
gross revenue that is carried through to net operating income before Federal income taxes. This
measure is similar to the rate of return on sales before taxes. For the industry as a whole, the per-
cent of gross revenue carried through has been declining. This is also true of each district, with the
Eastern being the worst. Table 3-6 summarizes these trends.
The performance of the Southern and Western Districts is much better than the Eastern.
In fact, one would conclude that compared with nonregulated industries such as steel, the
Southern and Western roads are reasonably good performers. Compared with other regulated
industries, such as public utilities (10.5% return on stockholders' equity) and telephone and tele-
graph companies (9.5% return on stockholders' equity), the railroads' rate of return is low. One
point that should be made is that railroads follow a "betterment" accounting procedure, which
tends to overstate the value of their assets. We have not attempted to adjust rate of return in the
rail industry to reflect this.
* Because the railroads use a nonstandard accounting procedure (the so-called betterment tech-
nique), the rate of return is low relative to what it would be if they used a procedure comparable
to those used in the nonregulated sector.
s*Opcrating ratio equals operation expenses divided by operating revenues.
3-5
-------
TABLE 3-6
PERCENT OF GROSS REVENUE CARRIED THROUGH
TO NET OPERATING INCOME BEFORE FEDERAL INCOME TAXES
Year
1950
1960
1965
1968
1969
1970
1971
All Class I
RR's
17.3%
8.3
11.0
6.9
6.6
4.2
4.0
Southern
District
20.1%
10.7
12.1
11.0
12.1
11.8
10.3
Eastern
District
12.0%
2.1
10.0
3.7
2.7
Nil
0.5
Western
District
19.8%
10.0
11.6
8.4
8.0
7.7
7.2
The historical decline in the profitability of railroads came as a result of a decrease in the
relative importance of high-weight, low-value cargo, which has traditionally been handled by rail.
The increased competition from motor carriers and pipelines has further reduced the relative
importance of railroads. Federal and State funding of highways has improved the competitive
position of trucks and has led to the diversion of high-valued freight to motor carriers.
In 1935 when motor carriers came under Interstate Commerce Commission regulation, the
value-of-service rate structure applied to railroads was also applied to motor carriers. (The value-of-
service rate-making policy was originally applied to railroads in order to favor agricultural products.
Under value-of-service rates, low-valued products have a lower rate per ton mile than do high-value
products.*) his measure reduced intermodal price competition and in fact gave an advantage to
trucks in carrying high-valued freight when they could give faster service. Railroads were unable to
lower prices on this type of freight, which could have offset the faster service offered by trucks.
The decline of some manufacturing industries in the East has led to a more intense financial crisis
among eastern roads. Also the capital stock of these railroads tends to be older than that of the
other roads. They spend a larger portion of total cost on yard switching than do either southern or
western roads, due to shorter hauls and a larger number of interchanges among roads. Since shippers
pay for movement from one point to another (i.e., rate per mile), the competitive position of rail-
roads tends to be diminished if these nonline-haul expenses rise. The greater yard-switching results
These points are examined in an article by R.H. Harbeson in the 1969 Journal of Law and Eco-
nomics, pp. 321-338.
3-6
-------
in having rail cars sit in switching yards waiting for a train to be made up, thus resulting in longer
time in transit and higher comparative costs.
GROWTH
In projecting growth rates in any industry, it is assumed that historical trends and relationships
will continue to hold in the future to some extent. If these relationships do continue, then rail
freight can be projected based on projection of other figures. For example, rail freight service on
the basis of population or gross national product can be projected. If the population continues to
consume similar commodities, if these commodities move by the same modes of transportation, and if
increases in income are ignored, then projections based on accurate population projections will be
valid.
The ton miles of railroad freight per capita in the U.S. has remained quite stable over the past
five years. It was 3.73 in 1965, 3.77 in 1968, and 3.75 in 1970. Given this stability, short-run
projections b:>sed on population growth may be quite accurate. Based on the population projec-
tions for the U.S., about a 1% annual increase over the next 5 years is estimated. This would mean
an increase from 768 million ton miles in 1970 to about 822 million ton miles in 1975.
The rail industry's contribution to national income has remained relatively constant over the
period from 1968 to 1970 at about 1%. The long-run rate of growth in GNP has been about 3.5%.
Again, under the assumption that these historical relationships hold, the long-run growth should be
around 3.5%.
One factor which may reverse these trends is that rail movement uses less energy than other
forms of freight movement. A ton mile of freight moved by rail requires 750 British thermal units
(BTU), while pipelines require 1850, trucks 2400, and air freight 63,000. The only mode of freight
movement more efficient (in terms of energy) than rail is water, which requires 500 BTU.*
Energy may come to be an important factor, but it seems unlikely that rail freight will increase
more rapidly than the growth in national income. The factor militating against a more rapid
increase is that consumption patterns have continued to move toward more services and fewer
manufactured products. This means a smaller transportation input. In addition, rising interest
rates and greater product differentiation have caused shippers to be increasingly concerned with
time in transit. The railroads' real advantage is in rates, not speed. However, the advent of trans-
porting entire truck trailers by rail has aided in reducing delivery time substantially in areas where
this is practiced.
*Bi.1-* >;'>t-'k, Mrdraw-Hill I.ic., September 8, 1973, p. 63.
3-7
-------
SECTION 4
RAILROAD NOISE SOURCES
GENERAL
Noise is generated by railroad operations in two basic locations: in yards and on lines. In
railroad yards, trains are broken down and assembled and maintenance is performed. Line opera-
tions involve the sustained motion of locomotives pulling a string of cars over tracks.
The hump yard is an efficient system for disengaging cars from incoming trains and assembling
them into appropriate outgoing trains. A locomotive pushes a string of cars up a small hill, known
as a hump, allowing each car to roll individually down the other side through a series of switches
onto the appropriate track where a train is being assembled. As each car rolls down the hump, it
is first slowed by the "master" retarder. The slowing, or retarding, is accomplished by metal beams
that squeeze the wheel of the rail car. After the cars leave the master retarder, they coast into a
switching area that contains many tracks. As each car is switched onto a particular track, it is
slowed by a "group" retarder. After a car moves out of a group retarder, it is switched onto one
of many (approximately 50) tracks in the "classification" area where the car collides with another
car. The collision causes the cars to couple, forming a train. In some yards, the first car that moves
into the classification area along a particular track is stopped by an "inert" retarder, so-called
because the retaining beam is spring-loaded and requires no external operation. Inert retarders
differ from the master and group retarders, which are controlled continuously by an operator or
automatically by a computer.
All three of the retarding processes described above produce noise. When the beam of a
master or group retarder rubs against the wheels, a loud squeal often is generated. The most
significant noise generated by inert retarders occurs when a string of cars is pulled through the
retarders. If the inert retarders are short and exert small forces, they may generate noise that is
negligible compared with the noise generated by the group retarders. Some yards are equipped with
inert retarders that can be manually or automatically released when a string of cars is pulled through
them thereby preventing retarder squeal. There are no inert retarders in some yards, so an
individual brakeman must ride some cars and brake them manually.
Noise is also produced when cars couple in the classification area of the yard. The impact
points, and thus the origins of the noise, are scattered over the classification yard. The noise is
impulsive, and sometimes it is followed by a thunderlike rumble that is audible for several
seconds after the impact.
4-1
-------
Locomotive engines generate noise as the locomotives move around or pass through yards.
When the locomotives are not in use, their engines are often allowed to idle continuously (even
overnight), which also results in significant noise. When the locomotives are in motion, their horns,
whistles, and bells may produce noise for warning purposes.
Some noise originates in the yard shops where locomotives and cars are repaired and maintained.
Power tools and ventilation fans represent such sources. However, the most readily identifiable
sources of shop noise are the locomotives themselves when undergoing testing.
Most yards are equipped with a number of loudspeakers that are used for conveying verbal
instructions and warning sounds to workers in the yard. The speakers are scattered about the yard,
and a given speaker issues sound on an unpredictable schedule.
Line, or wayside, noise-the noise in communities from passing trains—is comprised of many
high noise sources. The locomotive engine and its components, such as exhaust systems and cooling
fans, and the interaction of railroad car wheels with rails results in significant noise. Wheel/rail
noise is caused principally by impact at rail joints, giving rise to the familiar "clickety-clack," and
by small-scale wheel and rail roughness. A severe form of wheel roughness that generates high noise
levels is caused by flat spots developed during hard braking. Also, wheels squeal on very sharp
curves and generate noise by flange-rubbing on moderate curves. The operation of such auxiliaries
as refrigeration equipment also contributes to the overall noise level. Horns or whistles are sounded
at crossings and are significantly louder than the other wayside noises. In addition, some crossings
are equipped with stationary bells that sound before and during the passage of trains.
The remainder of Section 4 treats each of the noise sources mentioned above separately and
in as much detail as the state-of-the-art allows. Included in the discussion of each source is a
description of abatement techniques.
CONSIDERATION OF RAILROAD NOISE SOURCES FOR FEDERAL REGULATION
Many railroad noise problems can best be controlled, at this time, by measures that do not
require national uniformity of treatment to facilitate interstate commerce. The network of railroad
operations is embedded into every corner of the country, including rights-of-way, spurs, stations,
terminals, sidings, marshaling yards, maintenance shops, etc. Protection of the environment for
such a complex and widespread industry is not simply a problem of modifying noisy equipment;
it also gets into the minutiae of countless daily operations at thousands of locations across the
country. The environmental impact of a given operation will vary depending on where it takes
place, for example, whether it occurs in a desert or adjacent to a residential area. For this reason,
state and local authorities appear better suited than the Federal government to consider fine details
such as the addition of sound insulation or noise barriers to particular facilities, the location of
noisy equipment within those facilities as far as possible from noise-sensitive areas, etc. There is
no indication at present that differences in requirements for such measures from place to place
impose any burden on interstate commerce. At this time, therefore, it appears that national
4-2
-------
uniformity of treatment of such measures is not needed to facilitate interstate commerce, and
would not be in the best interest of environmental protection.
However, since the national effort to control noise has only just begun, it is inevitable that
some presently unknown problems will come to light as the effort progresses. Experience may
teach that there are better approaches to some aspects of the problem than those that now appear
most desirable. The situation may change so as to call for a different approach. Section 17 of the
Noise Control Act clearly gives the Administrator of the Environmental Protection Agency authority
to set noise emission standards on the operation of all types of equipment and facilities of inter-
state railroads. If in the future it appears that a different approach is called for, either in regulating
more equipment and facilities, or fewer, or regulating them in a different way or with different
standards consistent with the criteria set forth in Section 17, these regulations will be revised
accordingly.
The Administrator has considered the following broad categories of railroad noise sources in
order to identify those types of equipment and facilities which reguire national uniformity of
treatment through Federal noise regulations to facilitate interstate commerce.
Office Buildings
Many, if not all, surface carriers engaged in interstate commerce by railroad own and operate
office buildings. These buildings are technically "facilities" of the carriers. Like all office buildings
they may emit noise from their air conditioning and mechanical equipment. But since each building
is permanently located in only one jurisdiction and is potentially subject only to its regulations, it
is not affected in any significant way by the fact that different jurisdictions may impose different
standards on noise emissions from the air conditioning and mechanical equipment of other build-
ings. At this time, there appears to be no need for national uniformity of treatment of these facili-
ties, and they are therefore not covered by these proposed regulations.
Repair and Maintenance Shops
Railroad repair and maintenance shops are similar in many ways to many nonrailroad indus-
trial facilities, such as machine shops, foundries, and forges. All such facilities can reduce their
noise impact on the surrounding community by a variety of measures including reduction of noise
emissions at the source, providing better sound insulation for their buildings, erecting noise barriers,
buying more land to act as a noise buffer, scheduling noisy operations at times when their impact
will be least severe, or simply moving noisy equipment to locations more remote from adjoining
property. Such detailed and highly localized environmental considerations are best handled by
local authorities. Like office buildings, shops are permanently located in only one jurisdiction and
thus are not potentially subject to differing or conflicting noise regulations of other jurisdictions.
At this time, therefore, there appears to be no need for national uniformity of treatment of these
facilities, and they are not covered by these proposed regulations.
4-3
-------
At times, railroad maintenance shops may contain major noise sources that do require national
uniformity of treatment, such as locomotives. But the fact that some such individual noise sources
within a shop may be subject to Federal noise emission regulations is irrelevant to the validity of
State or local noise emission regulations applied to the shop as a whole, as long as the State or local
regulation on the shop can reasonably be complied with without physically affecting the Federally
regulated noise source within the shop (for example, by installing sound insulation in the shop build-
ing). This will be discussed further in the section on preemption below.
i erminals, Marshaling Yards, and Humping Yards
Like office buildings and shops, railroad terminals and yards are permanent installations nor-
mally subject to the environmental noise regulations of only one jurisdiction. Noise emissions from
terminals and yards can also be reduced by many measures that do not require national uniformity
of treatment and that can best be handled by local environmental authorities. These include
measures such as placing noise barriers around such noise sources, for example, as retarders, acquiring
land to act as a noise buffer, locating noisy equipment as far as possible from adjacent noise-
sensitive property, and reducing the volume of loudspeaker systems or replacing them with two-
way radios. At this time, there appears to be no need for national uniformity of treatment of these
facilities, and they are not covered by the proposed regulations.
Some noise sources in railroad yards may at some point require national uniformity of treat-
ment through Federal noise regulations, even though such sources may be permanently physically
located in a yard. Such a circumstance could be occasioned because of the noise sources' intimate
relationship to the movement of railroad trains. Rail car retarding operations in humping and
marshaling yards, for example, produce individual peak noise levels of up to 120 dB(A) at 100
feet. Such retarding operations are an integral part of the movement of railroad trains. A number
of measures are now being investigated which may make it technologically and economically feasible
to control this noise at its source, i.e., the retarder. Such measures include lubrication of retarder
beams, changes in the composition or design of the beams, and changes in the method of application
of retarding force. At this time, however, it is the Agency's position that retarder noise is an
element of fixed facility railroad yard noise which, as such, can best be controlled by measures
which do not in themselves affect the movement of trains and therefore do not require national
uniformity of treatment. Such noise control measures might include, for example, the erection of
noise barriers. The Agency's study of railroad yard noise indicates that concern for noise from
railroad yards is more local than national. This is due in large part to the location of a number of
yards in non-urban areas. Accordingly, the establishment of a uniform national standard could
potentially incur significant costs to the railroads with only limited environmental impact resulting
in terms of population relief from undesirable noise levels. This subject is discussed in more detail
in Appendices C and D of this document.
4-4
-------
Like railroad maintenance shops, marshaling and humping yards contain some noise sources
that are covered by the proposed regulations. As is discussed in greater detail in the preamble to the
proposed regulations, a State or local noise regulation on a railroad terminal or yard is in effect a
regulation on the Federally regulated noise sources within the terminal or yard when it can be met
only by physically altering the Federally regulated noise sources.
Track and Right-of-way Design
Due to the intimate relationship between the track and the rail car wheels in the generation of
rail car noise, the proposed regulations must preempt State and local regulations specific to track.
However, some steps can be taken to reduce noise emissions from railroad rights-of-way that
do not in any way affect the operation of trains on the rights-of-way, such as the erection of noise
barriers. State and local governments are much better situated than the Federal Government
to determine if some noise-sensitive areas need such protection; and the existence of differing
requirements for such measures in different areas does not at this time appear to impose any
significant burden on interstate commerce. There is, at present, no need for national uniformity
of treatment of such noise abatement techniques; and they are, therefore, not covered by the
proposed regulations.
Horns, Whistles, Bells, and Other Warning Devices
These noises are different in nature from most other types of railroad noise since they are
created intentionally to convey information to the hearer instead of as an unwanted by-product
of some other activity. Railroad horns, whistles, bells, etc., are regulated at the Federal and State
levels as safety devices rather than as noise sources. Federal safety regulations are confined to the
inspection of such devices on locomotives, so as to ensure that, if present, they are suitably located
and in good working order (Safety Appliance Act, 45 USCA; 49 Code of Federal Regulation, 121,
234, 236, 428, 429). State regulations are oriented toward specifying the conditions of use of
these devices and, for the most part, do not specify any maximum or minimum allowable noise
level for them. A recent survey of the 48 contiguous States (See Appendix B) has revealed the
following:
1. At least 43 States require that trains must sound warning signals when approaching public
crossings.
2. 35 of these States specify some minimum distance from a public crossing at which a train
approaching that crossing may sound a warning signal.
3. 3 States specify a maximum distance from a public crossing at which a train approaching
that crossing may sound a warning signal.
4. 35 States specify that these warning signals must be sounded until the train reaches the
crossing.
4-5
-------
5. 3 States specify that these warning signals must be sounded until the train completely
clears the crossing.
6. 16 States provide for exceptions to their regulations for trains operating in incorporated
areas.
7. At least two States provide for exceptions to their regulations for trains approaching
public crossings that are equipped wiih satisfactory warning devices.
Two frequently proposed solutions to eliminate the need for trains to sound warning devices
v.uen approaching public crossings are:
I. Eliminate all public grade level railroad crossings,
2. Install active protection systems (e.g., flasher-gate combinations) at all public grade level
railroad crossings.
This first solution would be the most effective since it would eliminate the source of the
proble.n, the public grade level lailroad crossing. However, it would be extremely costly because
it would involve the elevating or depressing of either the railroad line or the public thoroughfare at
each public crossing. This solution may be infeasible for solving existing conditions but it should
be seriously considered in all future public thoroughfare or railroad line construction projects.
The second solution, although it does not attack the source of the problem, does seem to be
;j, -, icctive protection rnensvre in fnrt it could eliminate {he need for the sounding of warning
signals by trains approaching public crossings. This solution has its drawbacks, however. Flasher-
gate-type devices cost $30,000-$40,OQO with some installations costing up to $60,000. In the
State of Illinois there are 16,250 grade level crossings of which 1,625 have flasher-gate protection
devices. To outfit the remaining 15,000 crossings with these devices in that state alone would cost
$450 million or more. The nationwide cost of this solution would be prohibitive.
Since train horns, whistles, bells, etc., are designed to emit a great deal of noise in the interests of
safety, and since any regulation restricting the noise output of these devices could be construed as
contrary to these interests, no regulatory action affecting these devices is being proposed at this time.
Sjjeciai Purpose Equipment
Interstate rai! carriers operate a number of types of special purpose rail cars, including snow
plows, track laying equipment, and cranes. It if- not clear to EPA at this time whether such equip-
ment is uf e-'i in such a manner as to require national uniformity of treatment; or, if such treatment
is requisite, what noise emission standards should be applied to its operation. In any event,
there does not appear to be any conflicting State or local regulations on such equipment at present.
Accordingly, such special purpose equipment which may be located on or operated from rail cars is
not covered by the proposed standards. However, the raii cars themselves on which such special pur-
pose equipment is located or operated from are invluued under the proposed standards for rail car
operations, if in tne fucure K app.-ais that national uniforn>i',y of treatment of such equipment is
necessary, appropriate noise emission standards for it pursuant to .lection 17 will be proposed.
4-6
-------
Trains
Unlike the categories of railroad equipment and facilities discussed above, train noise is poten-
tially subject to the noise regulations of more than one jurisdiction. Trains are constantly moving
from one jurisdiction to another, and it is not feasible to have them stopped at political boundaries
and adapted to meet a different noise standard. Moreover, they constitute a major source of noise
to people close to railroad rights-of-way. The various sources of tiain noise (other than warning
devices) are therefore covered by these proposed regulations in order to facilitate interstate com-
merce through national uniform treatment of their control
CHARACTER OF RAILROAD NOISE SOURCES AND ABATEMENT TECHNOLOGY
Locomotives
Railroad locomotives are generally categorized as (1) steam, (2) ti;o.sd-electric, (3) electric or
(4) gas turbine. The few remaining steam locomotives in the United States are preserved primarily
as historical curiosities and are, therefore, not covered by the proposed regulations. In this sub-
section, noise associated with diesel-electric and elecuie/feus turbine locomotives are presented.
All measurements discussed in this section are A-wei^hrtx; levels obtained by means of a
microphone placed alongside a locomotive, and refer to 100 ft., unless otherwise noted. Details of
the measurements are given in Section 6.
Diesel-Electric Locomotives
Three types of engines are currently in use: 2-stroke Rootes blown, 2-stroke turbocharged,
and 4-stroke turbocharged. A turbocharged engine produces about 50% more power than does a
Rootes blown engine. The number of cylinders on - J:csei e ;g:ne may be 8, 12, 16, or 20, with
each cylinder having a displacement of 650 cu in. Each cyiirrLi produces 125 hp when Rootes
blown and 187.5 to 225 hp when turbocharged. These engine- -.re employed on the two basic
types of locomotive: the switcher, which is used primarily to sr. ...i cars around the railroad yard
and is powered by engines of under 1500 hp, and the road locomotive, which is used primarily for
long hauls and is powered by engines of 1500 hp or more.
A diesel locomotive engine drives an electric alternator that produces electricity to run the
electric traction motors attached to each axle of the locomotive. The rated power of the engine
is the maximum electrical power delivered continuously by the alternator. The engine has eight
possible throttle settings. As can be seen in Table 4-1, engine ;-'-wer and nmse levels increase with
throttle position. The data in this table are taken from a presentation given at an Association of
American Railroads (AAR) meeting in August 1973, by the Electro- Motive Division (EMD) of
General Motors Corporation, and were developed from a study of load cell information for a num-
ber of U.S. railroads. Of the approximately 27,000 locomotv/es in service on major railroads (see
Appendix A), about 20,000 were built by EMD. The percent of horsepower and percent of time
4-7
-------
TABLE 4-1
EFFECT OF THROTTLE POSITION ON
ENGINE POWER AND NOISE LEVELS
Throttle
Position"
Idle
1
1
L.
3
4
5
6
7
8
% of Rated
hp for
Diesel Engines
0.75t
5
12
23
35
51
66
86
100
% of .Time at
Throttle Position
Road Loco
41
3
3
3
3
3
3
3
30
Switcher
77
7
8
4
2
1
—
—
1
dB(A) at
100 Ft for
2000 hp Engine
69.5
72.0
74.0
77.0
80.0
84.5
86.0
87.5
89.0*
*Three cooling fans were operating during measurement for throttle position 8, only one
fan for other measurements.
t Locomotive auxiliary hp only-no traction.
given for each throttle position are typical of all locomotives. The dB(A) levels vary, of course,
from engine to engine. The example here is for a 2000 hp EMD GP40-2 locomotive.
Locomotive at Rest
During the course of this study, sound level measurements were made on individual locomo-
tives at different power settings during load cell or self load testing. The results of these tests are
shown in Table 4-2.
For purposes of separating the contributions of various components to overall engine noise
levels, the prediction schemes employed in the Department of Transportation Report of 1970 were
used. The predictions involve (1) determining the mechanical power and type of engine required
to perform a given task, (2) determining the throttle setting required to perform a given task, and
(3) converting from engine type and throttle setting to sound level. The expression for unmuffled
diesel exhaust noise is
dB(A) at 100 ft = 92 + 10 log (hp/1500) - 3 (8-throttle settings) - T
where T is 6 for turbocharged engines and 0 otherwise. As can be seen in Figure 4-1, the predicted
exhaust noise level for an EMD F7A locomotive at each throttle setting is very close to the measured
total noise level. This result agrees well with the assumption that engine exhaust is the dominant
4-8
-------
TABLE 4-2
STATIONARY NOISE EMISSION DATA FOR
GENERAL MOTORS AND GENERAL ELECTRIC LOCOMOTIVES
Locomotive
Identification
EMD-SW1500
EMD-F7A
EMD-SWI500
EMD-SW1500
EMD SD 9
SD 4328
EMD 25014
SD9
EMD-GP/SD38
EMD 5077
GP 38-2
EMD
GP 38-2 535
EMD
GP 38-2 535
EMD 41 15
72635-1
GP 38-2
EMD 41 11
72735-12
GP 38-2
EMD 4053
5806-4
GP 38-2
EMD 4050
5806-1
GP 38-2
EMD 4508
SD24
SD35 1921
EMD 29355
SD35
EMD 1952
29340
SDP35
EMD FP/SD-40
Horsepower
1500
1500
1500
1500
1750
1750
2000
2000
2000
2000
2000
2000
2000
2000
2400
2500
2500
2500
3000
Loading
Conditions
T
T
T
T
T
—
T
S
S
T
S
S
S
S
T
T
T
S
T
Aspiration
—
—
—
RB
RB
—
RB
—
TC
RB
RB
RB
TC
—
TC
TC
Throttle
0
66*
69*
—
68
70
—
65
67
66.5
66*
63*
62*
61*
68
69
68
70
72
Setting
8
84.5**
86
92*
93
89
—
91.5
91
88.5
88.5
91
90
88
89
86.5
86
88
88
89,5
Re fere
3
1
]
3
11
10
3
7
7
7
8
8
8
8
9
7
8
8
3
4-9
-------
TABLE 4-2 (Cont'd)
STATIONARY NOISE EMISSION DATA FOR
GENERAL MOTORS AND GENERAL ELECTRIC LOCOMOTIVES
Throttle Setting
is Aspiration 0 8 Reference
64.5 88 7
69.5 88.5 7
67 85.5 7
68.5 88 7
67 88 7
TC 69 92 8
TC 68 87 8
70* 88* 7
TC 68 90 8
TC 70 90 8
TC 72 94 11
90.5 3
86* 5
TC 72 — 10
TC 66* 89 8
TC 65* 87 8
Locomotive
Identification
EMD
GP 40 3049
EMD
GP 40 301 8
EMD
GP 40 3182
EMD
GP 40 3 195
EMD
GP 40 31 56
EMD 1559
32623
GP40
EMD 1562
32960
GP40
EMD-GP40-2
EMD 31 15
SD45
EMD 3 124
SD45
EMD
SD 45-T2
SP9212
EMD
SD45
GEU25
GE 38573
4300
GE 1472
38417
U30C
GE 1581
37970
U30C
Horsepower <
3000
3000
3000
3000
3000
3000
3000
3000
3200
3200
3600
3600
2500
3000
3000
3000
Load
Condi
T
T
T
T
T
T
T
T
S
S
S
T
T
S
S
4-10
-------
TABLE 4-2 (Cont'd)
STATIONARY NOISE EMISSION DATA FOR
GENERAL MOTORS AND GENERAL ELECTRIC LOCOMOTIVES
Locomotive
Identification
GE 1473
38418
U30C
GEU30
GE3811
U33C
GE8717
U36C
38879
GE U36B
1759
GE U36B
1825
GE U36B
1780
GE U36B
1855
GE U36B
1832
GE U36B
1815
GE1767
37430
U36B
GE 1796
37792
U36B
GE 1766
37429
U36B
GE 1771
37434
U36B
GE 1764
37427
U36B
Horsepower
3000
3000
3300
3600
3600
3600
3600
3600
3600
3600
3600
3600
3600
3600
3600
Loading
Conditions
S
T
S
S
S
S
S
S
S
S
S
S
S
S
S
Aspiration
TC
—
TC
TC
—
—
—
—
—
—
TC
TC
TC
TC
TC
Throttle
0
67*
—
68
72
68
67
66
66
65
64.5
66
67
67
67
67
Setting
8 I
87
86*
90
91.5
91
93
90.5
85.5
89.5
90
87
91
93
91
94
leferer
8
4
8
9
7
7
7
7
7
3
8
8
8
8
4-11
-------
TABLE 4-2 (Cont'd)
STATIONARY NOISE EMISSION DATA FOR
GENERAL MOTORS AND GENERAL ELECTRIC LOCOMOTIVES
Locomotive
Identification
GE1526
38048
U36B
Loading
Horsepower Conditions Aspiration
3600 T TC
Throttle Setting
0 8
66
90
Reference
rr-96
J36B
GE U36B
3600
3600
Sample Size
S
S
TC
68
64.5
47
92
90
51
8
7
S - Self Load
L - Load Cell
TC - Turbo Charged
RB - Rootes Blown
* Data taken at 50 ft.;
6 dB(A)added
**Pre-1960 muffler
Idle
Range 61-72 dB(A)
Mean 67.3 dB(A)
Standard
Deviation 2.45 dB(A)
Throttle 8
84.5-94 dB(A)
89.3 dB(A)
3.36 dB(A)
4-.12
-------
source mechanism in locomotive noise. A similar expression is used in Ref. 4 to predict the contri-
bution of casing-radiated noise.
EMD F7A
0500 hp. not turbochorged)
3456
THROTTLE SETTING
Figure 4-1. Measured Total and Predicted Exhaust Noise Levels
Table 4-3 gives the exhaust and casing noise levels predicted by the techniques in Ref. 4 for a
number of locomotives as well as total noise measurements made by BBN, EMD, and GE. The
measured data were gathered while the locomotive was stationary and under full load (throts'e
position 8) on a test cell. The engine was loaded by feeding the electric current into a resistor bank.
As can be seen in this table, the contribution of casing noise to overall level appears to increase
with mechanical power. Thus, for small locomotives where the level of casing noise is considerably
lower than exhaust levels, an exhaust muffler could provide substantial reduction in total locomotive
noise. For larger locomotives, exhaust muffling alone cannot reduce overall levels as much as the
small rootes-blown locomotives.
The average overall noise level for the EMD locomotives at 100 ft is 90 dB(A) ±4 db(A), where
the variance includes allowances for all possible measurement and locomotive differences, for
example, different observers and different test sites. The GE measurement for its 3000 hp loco-
motive is 86 dB(A) ±3 dB(A), again allowing for all possible measurement variations, slightly lower
than those measured by EMD. The reason for this difference may be that on GE locomotives, the
exhaust stacks rise about 6 in. above the hood, while on EMD locomotives the stacks are flush with
the hood and radiate sound more efficiently.
In addition to exhaust and casing noise, the noise from cooling fans may be significant. Fig-
ure 4-2 shows that the noise from an EMD GP-40-2 3000 hp locomotive measured 9 dB(A) higher
with three cooling fans running than with no fans running. Since it was necessary to open the
4-1*3
-------
TABLE 4-3
COMPARISON OF PREDICTED AND MEASURED NOISE LEVELS AT 100 FT
FOR VARIOUS EMD AND GE LOCOMOTIVES IN THROTTLE POSITION 8
Mechanical Power
and Type
EMD lOOOhp
Switcher
EMD ISOOhp
Switcher
EMD 2000 hp
Road Locomotive
EMD 3000 hp
, Road Locomotive
GESOOhp
Road Locomotive
EMD 3600 hp
Road Locomotive
GE 3600 hp
Road Locomotive
Predicted
Exhaust
db(A)
90
92
93
89
89
90
90
Predicted
Casing
db(A)
78
80
81
83
85.5
84
86.5
Measured
db(A)
_
93
89
89.5
86
89
—
No. of
Samples
0
2
2
1
1
4
0
Spread
db(A)
_
±1
±2
_
±3
—
Source
BBN
BBN
EMD
GE
BBN
—
engine access doors during the measurements, the recorded levels are somewhat higher than would
be generated under normal operating conditions. However, there is little doubt that cooling-fan
operation can contribute significantly to overall levels. The fans on GE engines run continuously,
thus contributing to total noise level under all operating conditions. Fans on EMD locomotives are
thermostatically controlled.
In summary, the major components of locomotive noise are, in order of significance, engine
exhaust noise, casing-radiated noise, cooling fan noise, and wheel/rail noise. Table 4-4 shows
average levels in dB(A) at 100 ft for each of these sources. Other sources, such as engine air intake,
traction motor blowers, and the traction motors themselves, have noise levels too far below the
cthei sources to be identified. Also, Rootes blown engines have a very unpleasant "bark" which
does not show up in any generally used method of measurement.
Locomotive in Motion
Another method of characterizing locomotive noise is as a locomotive passes by a fixed point
during normal operation. Levels recorded in this manner contain all sources of locomotive noise
discussed previously. Measurements of this nature are very meaningful, since this is the noise that
is emitted into the community. Unfortunately, the specific parameters that affect the level of noise
produced are not easily controlled. These include horsepower, velocity, throttle setting and number
4-14
-------
—OVERALL
WITHOUT FANS
31.5 SO 80 125 ZOO 313 500 800 ?Z50 JOOO 3150 5000 800012.500 20,500
25 40 63 100 160 250 400 T30 1000160025004000 63001CO<50I6$00
ONE-THIRD OCTAVE BAND CENTER FREQUENCY (Hz)
Figure 4-2. Effect of Fan Noise on the A-Weighted Spectrum of EMD GP40-2 Locomotive Noise
at 55 ft (Engine Access Doors Open)
TABLE 4-4
SOURCE CONTRIBUTIONS TO LOCOMOTIVE NOISE LEVELS
(Based on Prediction Techniques of Ref. 4)
Source
Exhaust
Casing
Cooling Fans
Wheel/Rail \
at 40 mph /
•
Locomotive only
Total train
dB(A)at 100 Ft
(Throttle 8)
86-93
80-85.5
80-84
78
81
4-15
-------
of locomotive units coupled together. However, by recording the sound levels of a large number of
pass-by events, typical levels may be established.
Figure 4-3 and Table 4-4.1 display the results of approximately 105 pass-by events. As indi-
cated, locomotive pass-bys range from 74 dB(A) to 98 dB(A) when measured at 100 feet.
Figure 4-4 shows, for the same events, the maximum sound level as a function of the velocity.
T! ij:e does not appear to be a relationship between speed and maximum locomotive noise
Figure 4-5 relates, again for the same events, the maximum sound levels as a function of
velocity and number of locomotives. There does not appear to be a definitive relationship between
the number of coupled locomotives and the noise emitted.
The measurement of locomotive pass-by events is explained in Section 6.
Locomotive Noise Abatement
Locomotive noise abatement may be grouped into two broad categories (1) Abatement By
Equipment Modification and (2) Abatement by Operational Procedures.
1. Abatement By Equipment Modifications:
Mufflers
Since locomotives contribute most of the noise of railroad operations and since exhaust noise
dominates locomotive noise, the first step in reducing locomotive sound levels is to require that
locomotives be fitted with an effective muffler. This section contains muffler manufacturer's
estimates of various factors affecting the feasibility of supplying both new and in-service loco-
motives with mufflers.
One such factor is the amount of back pressure a muffler creates. Back pressures on the engine
may affect its performance and life to a small extent. The engine must pump against the back pres-
sure, thereby reducing the power that can be distributed to propel the train. Normally, this degrada-
tion in performance is about 1% when back pressures are held within manufacturers' limits. Back
pressure may shorten engine life because when gases with increased temperature and density exhaust
into a region of high pressure, they raise the temperature of exhaust valves and turbochargers. The
following information on back pressure and its effects was determined by muffler manufacturers.
Engine Type
Back Pressure
Effect
Rootes Blown
Turbocharged
47.5 in. H20 measured at engine
exhaust port
5 in. HbO measured at exhaust
stack
10°C rise in turbocharger
temperature
20-hp loss on 3000 hp engine
< 0.6% increase in fuel
consumption
4-16
-------
-DIESEL-ELECTRIC LOCOMOTIVES
100
90
80
70
UJ
VE PASS-BYS EXCEEDING LEV
Ul CPi
0 0
§
§ 40
u.
O
K
Ul
O
S 30
20
10
105MEASU
X
xx
CEMENTS
X
X
X
X
'•
xx
X
X
)
X
X
X
X
xx
X
70
75
80 85 90
.PEAKdB(A) AT 100 FT.
95
100
Figure 4-3. Diesel-Electric Locomotive Pass-Bys
4-17
-------
TABLE 4-4.1
LOCOMOTIVE PASS-BY NOISE EMISSION LEVELS MEASURED AT 100 FEET
(see Figure 4-3)
Road Noise Studies ^
III IV TOTAL
74
75
76
77
78
79
80
81
82
83
84
85
dB(A) 86
87
88
89
90
91
92
93
94
95
96
97
98
I
1
1
2
2
4
3
3
1
2
1
2
4
2
3
4
3
1
1
II
1
1
1
1
1
2
3
1
1
2
1
1
3
2
2
4
2
3
2
2
1
1
1
2
2
2
3
4
2
3
1
2
1
1
1
1
5
2
2
4
8
7
8
1
8
5
4
7
6
74
6
7
7
1
3
2
I. Department of Transportation - Office of Noise Abatement
11. Department of Commerce - National Bureau of Standards
III. Wyle Laboratories
IV. Environmental Protection Agency — Office of Noise Abatement and Control
4-18
-------
0
UJ
UJ
ft
en
_i
UJ
LU
UJ
z
5
z
UJ
LU
Q.
I
5 o
1
\
\>
^Si cc
VL LLJ
H \\ 0
< \\ m
0 \ < ~
OJ » 0.
CC \ Q
O \ UJ
K \ UJ
' \ I
-1 \ i ^
XN j £
\i §
/• oc
/i i
/ 1
[ i
i *
\ i
\ !
,'' 1
C i
"*"***v 1
\ 1
x» *
<'' •— -
"~\ s
""""> \ UJ
x • cc
X 1 u.
X \ Q
^--^^ ! <
^ *• i s
**•%. ' i
r> y / &
,' £' 2
/ 3 * ~
/ °
v ^
N. £
1 >
' z
1 0
L— i
_J 1
UJ x^
UJ X
°xx
1 " 1 1 1
—
8
8
S
S
I
§ 0
UJ
UJ
Si
s
0
°
s
n
ass? a1"
8,
O
2
w
-
O
S
O
O
O
•«1-
•4
§
'J.JOOI J.VH3A3T
4-19
-------
o
o>
o
co
x
CO
LU
I O 3 z
u. (- u. P O
II
I
I
i
8
g §
Id 001 JLV 13A3T>IV3d
4-20
-------
Mufflers have no appreciable effect on exhaust emissions; muffler-equipped locomotives give-
off insignificant incremental amounts of NOX, CO, and smoke [EMD (1973)]. One potential prob-
lem manufacturers want to investigate further is that condensed, unburncd hydrocarbons might
give rise to a stack fire. This has never occurred on locomotives having mufflers, although it has
happened on stationary installations.
Three manufacturers with experience in fabricating mufflers for locomotives have indicated
that their products will materially assist the railroads in complying with the proposed regulations:
Donaldson of Minneapolis, Minn.; Harco Engineering of Portland, Ore.; and Universal Silencer of
Libertyville, 111. The following are these manufacturer's estimates of the attenuation that could be
achieved with their mufflers, the approximate cost of the mufflers alone, without any allowance for
installation, and the amount of back pressure they create,
Donaldson has had experience with the Chicago and Northwestern Railroad in equipping a
locomotive with an off-highway truck type of muffler. The results were:
Muffler Cost - approximately $800 for two mufflers
Back Pressure - further testing necessary
Harco Engineering has achieved the following results for a switcher locomotive. The muffler is
fitted to a Harco spark arrester.*
Attenuation - approximately 5 dB(A)**
Muffler Cost - $75
The results for road locomotives are:
Rootes Blown:
Attenuation — approximately 10 dB(A)**
Muffler Cost - $750
Turbocharged:
Attenuation — approximately 10 dB(A)**
Muffler Cost - $1000
Back Pressure - 13-20 in. ^0 (EMD claims that the back pressure is too high)
Universal Silencer has built mufflers for EMD locomotives (3 DRG and 40 Amtrack). Accord-
ing to EMD (presentation at AAR meeting, 1973) these mufflers achieved:
Attenuation - 9-10 dB(A) at full power
Muffler Cost - approximately $ 1200
Back Pressure - 3 in. ^O
The estimated overall noise that would result from equipping various locomotives with muf-
flers that give 5 dB(A) and 10 dB(A) attenuation in throttle 8 is indicated in Table 4-5.
Muffler manufacturers have said that they could supply fully developed and tested muffler
systems for all locomotives by the following dates.
*From EPA Docket 7201001, No. R007.
**This measurement was performed by the manufacturer.
4-21
-------
1 January 1974
1 January 1976
1 January 1976
HARCO
Switchers
Road
DONALDSON
All types
UNIVERSAL SILENCER
Turbocharged Locos 1 January 1976
Rootes Blown 1 January 1977
Switchers 1 January 1978
EMD and GE have said that they could fit mufflers on new locomotives by the following dates.
EMD
1 January 1976
1 January 1977
1 January 1978*
Turbocharged
Rootes Blown
Switchers
Road
GE
Turbocharged
1 January 1976
TABLE 4-5
LOCOMOTIVE NOISE LEVELS EXPECTED FROM EXHAUST MUFFLING, THROTTLE 8
Locomotive Type
EMD 1000-hp Rootes Blown
Switcher
EMD 1500-hp Rootes Blown
Switcher
EMD 2000-hp Rootes Blown
Road Locomotive
EMD 3000-hp Turbocharged
Road Locomotive
GE for Alco) 3000-hp
Turbocharged Road
Locomotive
EMD 3600-hp Turbocharged
Road Locomotive
GE (or Alco) 3600-hp
Turbocharged Road
Locomotive
5 dB(A) Exhaust Muffling
Total Noise
Level
[dB(A)l
86.0
88.0
89.0
86.5
87.5
87.5
88.5
Total
Attenuation
IdB(A)]
4.0
4.0
4.0
3.5
3.0
3.5
3.0
10 dB(A) Exhaust Muffling
Total Noise
Level
[dB(A)l
82.0
84.0
85.0
84.5
86.5
85.5
87.5
Total
Attenuation
IdB(A)]
8.0
8.0
8.0
5.5
4.0
5.5
4.0
*Because of problems integrating with spark arrester.
4-22
-------
EMD and GE agree that mufflers can be incorporated in new locomotives. The cost of instal-
ling mufflers on locomotives must be compared with a total cost of $300,000 to $400,000 per loco-
motive (GE and EMD presentations to AAR meeting, 1973). The following methods would be used
by each locomotive manufacturer in fitting mufflers on new engines. *
New GE Road Locomotives
Mufflers would be installed above the engine and the hood roof would be raised 8 in. A loco-
motive would still clear the required 15-ft, 7-in. gauge. Cost = $1500 per locomotive.
New EMD Road Locomotives
Turbocharged: The muffler would be installed over the turbocharger. Mountings would have
to be changed as would the roof structure, brake cabling, and extended range dynamic brakes.
Cost = $2500 per locomotive.
Rootes blown: The muffler would be integrated with the spark arrester. There would be
changes to the dynamic brake contactors, roof structure, and coolant piping. Cost = $3000 per
locomotive.
New EMD Switchers
The muffler would be integrated with the spark arrester, but EMD is not quite sure how.
Cost = $200-5500 (estimate based on Harco figures).
Retrofitting Older Locomotives
Retrofitting mufflers on locomotives involves finding out how many of each type of loco-
motive are still in service and adopting muffler installation procedure to the peculiarities of each
model.
Table 4-6 illustrates the distribution of switchers in service, categorized by manufacturer.
Very few new switchers are being built, only about 120 per year, since switchers appear to
run indefinitely. Furthermore, old road locomotives can be downgraded for switching use.
Most switching locomotives built before 1960 were equipped with mufflers, but after 1960
railroads generally fitted spark arresters instead.
In general, there does not seem to be any difficulty in fitting a muffler to the exhaust stack
above the hood of a switcher. This has already been done in many cases which spark arrester, result-
ing in some loss in visibility for the driver. Harco has designed and tested a muffler that integrates
with its spark arrester. The Harco muffler costs $75. However, this unit may have inadequate
muffling for the regulation or too high a back pressure. Keeping this in mind, EPA estimates the
cost for other spark arresters to be $200 to $500 plus 1 man-day labor for installation.
4-23
-------
TABLE 4-6
SWITCHER LOCOMOTIVES IN SERVICE
Manufacturer
EMD
ALCO
GE
Baldwin, Lima Hamilton
Fairbanks Morse
Year Built
1940-59
1 960-present
1940-61
1940-58
1946-56
1944-58
No. in Service
3200
1100
950
116
415
220
TOTAL 6000
The 8758 EMD Rootes blown road locomotives built before 1 January 1972 have less space for
mufflers than the new model GP/SD 38-2. Care must be given to the siting of mufflers, but installa-
tion is considered to be possible. The dynamic brake grids will have to be resited, and the roof
structure will have to be modified. Railroads might have changed exhaust systems on rebuilding.
Discussions with a representative from Penn Central have led to the following cost estimates for
fitting each of these older models with a muffler.
Muffler = $1500
Labor = 25 man-days ($/man-day = $46.40) (see Section 7)
Parts = $200-$500
Labor covers the resiting of dynamic-brake grids, plumbing and cabling, modifying the roof struc-
ture, and installing the muffler.
Thus, we see that mufflers can be fitted to new locomotives for less than a 1% increase in cost,
and a retrofit program for mufflers is practical inasmuch as no locomotive has been identified that
would be unduly difficult to retrofit.
Mufflers that product 5 to 10 dB(A) of exhaust muffling are currently feasible. It is important
that a muffler be designed to give as good muffling at idle as at full power, since locomotives idle
much of the time. Unless other noise sources on the locomotive are also treated, the net locomotive
quieting will be only about 6-dB(A) due to contributions from these sources (see Table 4-4).
Mufflers could be developed and ready for production by 1 January 1976. The manufacturers
have sufficient capacity to produce the mufflers required.
Cooling Fan Modification
The next contribution to locomotive noise that may be treated is the cooling fan. This compo^
nent noise is essentially aerodynamic noise resulting from the air movement created by the fan. «
4-24
-------
Methods of treatment include increasing the diameter of the fan, adjusting clearances between blade
and shroud, and varying the pitch of the blade. Although fan modifications are feasible, the appli-
cation of fan retrofitting has not been developed for locomotives. Further, the impact of such a
requirement could not be assessed with regard to cost and the effect of the total noise.
Engine Shielding
The vibrations of the engine casing is a significant component of the total locomotive noise.
On a limited basis, work has been done to reduce the noise from this source by adding acoustic
panels to the engine, stiffening the engine casing, and using sound-absorbing materials. This tech-
nique has not been developed to the extent that it could be applied to locomotives at this time.
2. Noise Abatement By Operational Procedures:
Parking Idling Locomotives Away from Residences
One of the most frequent complaints about railroad noise is that locomotives are left idling
overnight. Railroads are reluctant to shut down locomotives because (1) shutting down and start-
ing locomotives require a special crew, (2) engines do not contain any antifreeze in their cooling
systems and would have to be heated in cold weather, and (3) locomotive engines are likely to leak
cooling fluid into the cylinders, which could damage an engine on starting if precautions were not
taken to drain it. Therefore, locomotives are usually shut down only during their monthly inspection.
Railroads are sometimes rather careless about where idling locomotives are left; frequently
they are parked on the edge of a rail yard close to residences. With a little effort, locomotives could
be parked near the center of a rail yard where they would be less troublesome to neighboring homes.
Speed Reduction
The power needed to pull a train increases almost directly with speed, but the noise of a given
locomotive increases very rapidly with speed. Thus, one could achieve some noise reduction by
lowering the speed limit for trains passing through residential areas. For example, the throttle
settings of the locomotives of passing trains would generally be lower, and hence the locomotive
noise would be reduced. Further, other noise sources, such as wheel/rail noise, would also be
reduced.
This noise reduction method may not be practical generally, except perhaps in special urban
areas, since the net effect would be to slow the movement of train traffic. The cost to the railroads
of lower speeds has not been calculated.
A Ban on Night Operations
Many freight trains, particularly in the eastern United States, operate at night. Their noise is
most disturbing at this time, since the background noise is lowest and people can be awakened from
sleep. Thus, a significant impact on the annoyance resulting from train noise can be made by banning
4-25
-------
night-tirne operations. However, such a ban on night operations would frequently be impractical,
since trains are scheduled for markets that open in the morning and the trains are loaded during the
previous day. The resulting burden on the flow of interstate commerce could be extensive.
Use More or Larger Locomotives for a Given Train
One paradox emerged from the model of locomotive noise presented earlier. A large locomo-
tive in a low throttle position develops less noise than a small locomotive in a high throttle position,
even when the two develop the same horsepower. For example, a 3600-hp locomotive in throttle
4 generates 15 dB(A) less noise than a 2000-hp locomotive in throttle 8. Thus, a considerable noise
reduction is achieved by using a 3600-hp engine to haul a train requiring only 2000-hp. Similarly, a
9 dB(A) reduction could be obtained by using four 3600-hp locomotives with lower throttle settings
to pull a train that normally requires two 3600-hp locomotives, but which operate at high throttle
settings.
This noise reduction technique is considered to be impractical in general, since the extra haulage
power required is quite large. However, this method could be used in some situations such as switch-
ing operations. Locomotive engineers could use low throttle positions rather than "gunning" the
engine in throttle 8.
Electric/Gas Turbine Locomotives
There are other means of train propulsion, apart from diesel-electric, currently in use on American
railroads. All-electric and gas turbine locomotives are becoming more popular, particularly in the
Northwest corridor. Rickley, Quinn, and Sussan have measured the wayside noise levels of the Metro-
liner, Turbotrain, and electric passenger and freight trains. The levels at 100 feet are given in Table
4-7. In general, levels do not exceed 88 dB(A). For those trains, namely two Metroliner trains and
one standard passenger trains, exceeding 88 dB(A), it is felt that the exceedance was caused by
wheel/rail interaction phenomena as opposed to locomotive engine generated noise, per se, since these
vehicle travelled at rates of speed where rail noise is likely to predominate (see wheel/rail noise section)
Thus, in general, the non-diesel-electric locomotive noise is well below that of diesel-electric
locomotives and the former are likely to comply with any regulation written for the latter.
Wheel/Rail Noise
Rail car noise includes all sources of train noise other than that produced by the locomotive.
These sources are wheel/rail interaction, structural vibration and rattle, and refigerator car cooling
system noise.
Of these sources, the interaction of the wheel and rail is the major component. As discussed
in the Bolt, Beranek and Newman Report No. 2709, "Railroad Environmental Noise: A State of the
Art Assessment," this source is generated by four mechanisms. These are labeled "roar," "impact,"
"flange rubbing" and "squeal."
4-26
-------
TABLE 4-7
NOISE LEVELS FROM ELECTRIC AND GAS-TURBINE TRAINS
Train
Metroliner
Electric Pass
Electric Freight
(2 Locos)
Turbotrain
No. of
Cars
4
4
4
6
4
6
6
3
5
5
3
3
Direction
South
South
North
North
North
North
South
South
East
West
East
West
Speed
(mph)
106
110
106
110
80
84
84
49
97
91
89
104
SPL[dB(A) 100ft]
89
89
84
84
78
80
90 (wheel/rail)
88
85
85
84
88
"Roar" describes the noise that predominates on welded tangent track. It is believed that roar
is due to roughness on the wheels and rails.
"Impact" noise refers to the noise produced by wheel and rail discontinuities such as wheel
flats, rail joints, frogs and signal junctions. This noise is characterized by a "clickety clack" sound
and may cause significant increase in wayside noise.
"Flange rubbing" describes the sound made when the flange contacts the rail and squeal does
not occur. This noise is characterized by a low-frequency grinding sound. It could be caused by a
stick-slip phenonmenon or by roughness on the flange and rail head.
"Squeal" is a very high pitched noise produced when a train negotiates a tight curve. Three
possible ways in which squel can occur are: 1) differential slip between inner and outer wheels on a
solid axle, 2) rubbing of the wheel flanges against the rails, and 3) "crabbing" or lateral motion of
the wheel across the top of the rail.
Structural vibration and rattle emanate from from the car bodies and couplings. Noise from
these sources may be distinguishable in a slowly moving train. Normally, however, this noise combines
with the other sources of car noise and is not readily distinguishable.
Refrigerator cars are railroad cars used to transport perishable freight that requires refrigeration.
It is necessary for the cooling equipment to operate continuously when the car is loaded, and also
4-27
-------
when the car is empty but a lead is anticipated. This cooling equipment usually contains an unmuffle
diesel engine to drive a compressor. These engines are similar in size and performance to engines used
in other applications in a muffled configuration. It is believed that the muffler industry could supply
the additional muffler requirement for rail refrigerator cars. However, application consideration woul
also have to include space availability and installation and replacement costs. The maximum noise
level from this source is approximately 75 dB(A) at 50 ft. (Wyle Laboratories, 1973). When a train
is moving, the noise levels emitted from a refrigerator car cannot be distinguished from overall train
noise; however, if the train stops or if the cars are held over, the continuous operation of the compres
engine may be a source of undesirable noise.
Refrigerator cars parked with their cooling systems running, as they often are in marshaling and
humping yards, may cause noise problems but only in places where refrigerator cars are parked near
noise-sensitive areas. At this time, such localized problems can best be controlled as a part of railroad
yard noise control, through measures such as parking refrigerator cars away from noise-sensitive areas
or installing noise barriers, rather than by requiring modifications to the entire refrigerator car fleet.
Typical measured levels of rail car noise are illustrated in figures 4-6, 4-7 and 4-8. Figure 4-7
indicates that the A-weighted wheel/rail noise level varies as 30 log V where V is the train velocity.
This relationship primarily describes the "roar" component of the noise. The higher levels present
are most probably indicative of "impact," "flange rubbing" and "squeal" noise.
CO
o
10
UJ
a
z
i
I
o
95
90
I
85
80
75
70
65
60
I I I I
• ZERO GRADE
O ZERO GRADE, EMPTY CARS
A UP 1 % GRADE
A DOWN 1% GRADE
1 T
SPL [(JBiA)] AT 50 FT* 75+30 LOG [v(MPH)/20 ] ~
I
I
I i A
I
I
10 15 20 25 30 35 40
SPEED
50 60 70 80 90 100
Figure 4-6. Wheel/Rail Noise Measured on Level Ground and on a 1% Grade
4-28
-------
100
-------
o
UJ
cc
a
O
03
oc
o
ui
O
ii
i I
ui o
\\
§2
i§
\°
< U)
S <
^o
<
ai <
00
Ul
a.
x a z
<> «
X U
< >
ui <
a.
s
a
g
ui
u
•*-*
I
•o
c
•o
-s
I
o
U
"S
2
1
13
o
'S.
.2
•o
I
en
•o
c
I
S
'S
I
I
od
4
8
i33d 001IV (V)8P 13A3T QNOOS 031H9I3/W-V
4-30
-------
3. "SAB" resilient wheels, marketed in the U.S. by American SAB Company, Inc., Chicago,
Illinois
4. "P.C.C." wheels, made by Penn Machine Co., Johnstown, Pa.
The Penn Cushion and Acousta Flex wheels are similar in principle. Both utilize an elasto-
meric ring between the rim and the hub of the wheel. The SAB and PCC wheels also are similar to
each other in principle. In these wheels, the rim is part of a steel disc, and the hub assembly consists
of one or more parallel steel discs. The rim disc is connected to the hub assembly via rubber elements
which deform as the wheel is loaded radially. The experimentation and data for resilient wheels on
rapid transit cars indicate that such wheels would be of negligible benefit for reducing railroad
freight car noise (Bolt Beranek and Newman 1974). Freight cars operate principally on tangent
track where resilient wheels are least effective.
Another technique which has been explored is "wheel damping." B.F. Goodrich Company
constructed a wheel with a layer of viscoelastic damping material bonded to the inside of the wheel
rim and covered with a bonded steel "constraining layer." This treatment is said to have eliminated
screech, reduced farfield noise obtained on tangent track by up to 2 dB(A) at high speeds, and also
attenuated rail vibration. Some limited experiments by B.F. Goodrich showed that use of an
"unconstrained" viscoelastic layer resulted in no significant noise reduction. However, the Toronto
Transit Commission found a 12 to 15 dB(A) squeal noise reduction when applying unconstrained
damping layers. Use of a four-layer damping configuration on a BART prototype car had no signifi-
cant effect on interior and wayside noise on tangent track, but eliminated some screeching on
curved track. Reductions of 20 dB(A) in screeching noise and 4 dB(A) for nonscreeching noise were
realized for curved track.
Rail welding is a method that can be used to reduce the noise caused by the discontinuities at
rail joints. On the average, it can be expected to reduce wayside noise by as much as 3.5 dB(A).
However, maximum levels are as high on welded rail as on bolted rail (see figure 4-8). Other
advantages of welded rail are the potential for less maintenance and a decrease in average rolling
resistance. Both are due to the absence of rail joints.
Rail damping is a technique which has undergone very limited testing. A damping compound
is applied to the nonrunning surfaces of the rails which should shorten the length of rail that vibrates
when a wheel passes over it. At this time, experimentation is so limited that no conclusions can be
reached as to the effectiveness of this technique.
In summary, although there are some new techniques and systems which show a degree of
promise, the only available methods today for reducing moving rail car noise emissions is through
the maintenance practices of car wheel and rail grinding in addition to the use of welded rail.
Retarder Noise
Within railcar classification yards, several thousands of cars are moved in each 24-hr period,
as trains are assembled/disassembled. Two general methods are used for car movement, (1) small
switcher locomotives are used to maneuver (one or more cars) and to create railcar vehicle velocity
4-31
-------
prior to release for selfanovement to pre-selected tracks, or, (2) heavy duty pusher locomotives
push rail cars up an incline and over a "hump" where the cars are released to travel on their own
to pre-determined yard locations. As a result of the techniques used in hump yards a single railcar
vr several railcars coupled together may be traveling at 10 to 15 mph and accelerating while moving
down the h^mp.
To manage the rail car(s). retarders are used to reduce car(s) speed or to stop them. In the
process of slowing or stopping the car(s) intense noise, characterized as a squeal, is often generated.
Figure 4-9 shows the amplitude distribution of noise associated with railcar movement through
retarders. Noise levels as high as 120 dB(A) at 50 feet have been observed.
Although studies (Ungar, Strunk and Nayak, 1970; Kurze, Ungar and Strunk, 1971) have been
conducted to determine the mechanism of wheel/retarder noise generation, a thorough under-
standing of the phenomenon is not yet at hand. It is thought that the intense wheel squeal is the
result of excitation of the railcar wheel at its resonant frequencies. Apparently, the noise levels
emitted by the car wheels are influenced by car type, car weight and loading, type of wheels, the
structure and composition of the retarder and the decelerating force that the retarder applies to
moving cars.
According to the Federal Railroad Administration there are approximately 130 hump yards
in this country. A listing of the current in-use hump yards by location, railroad, and number of
classification tracks is shown in Appendix C.
*w
30
to
z
UJ
UJ
0 20
„
UJ
00
10
-
-
-
-
-
80
1 1 1 1 I
1
1 I
1 1 1 : I I
90 100 110
i 1
H
m m
—
r i
i
20
W/
Z
UJ
UJ
15 <
5
U.
0
10
f^
UJ
U
(T
UJ
_ Q.
5
«
120 130
SOUND LEVEL AT 50 FEET [dB(A)]
Figure 4-9. Retarder Squeal Amplitude Distribution
4-32
-------
Retarder Noise Abatement
Though the mechanisms of wheel/retarder noise are not fully understood, several methods to
control the noise are thought feasible. One method, namely the use of barriers would control the
noise once generated, i.e., minimize the noise propagation efficiency; while four methods, (1) retarder
lubrication, (2) use of ductile iron wheel shoes, (3) use of releasable inert retarders and (4) retarder
control by computers would control noise at the source; i.e., minimize noise generation efficiency.
While the five methods cited are thought to be possible alternatives for retarder noise control,
much further study is required to assess the benefits and costs associated with each method. To
date, known benefit and cost information associated with the aforementioned methods are sum-
marized below.
Benefits:
The only study that has been completed which models the impact on retarder noise reduction
on people was of the Cicero Yard outside of Chicago. (See Appendix D.) The results of that study
showed that the reduction of retarder noise levels by 20 dB(A) allowed about 200 more people
to be exposed to less than an Ldn of 65 dB(A). The maximum reduction that would be experienced
by any of the 200 people would be a 2 dB(A) change in Ldn. If retarders were completely silenced
the noise reduction would benefit only 200 more people (total of 400) as per the above criteria,
according to the study.
Although it is not altogether accurate to project a study of a single yard to a national impact,
if the assumption was made that Cicero Yard is typical of all rail yards, approximately 26,000 more
people would be exposed to less than an Ldn of 65 dB(A).
By reducing locomotive exhaust noise by 10 dB(A) in the Cicero Yard, approximately twice
the benefit was realized (400 people less than 65 Ldn) than with the 20 dB(A) reduction in retarder
noise, according to the study.
Costs*:
A. Barriers (material costs of initial installation only)
1. $50 to $70 per linear foot.
2. $50,000 - $ 100,000 per yard.
3. $6.5 — 13.0 million for railroad industry.
4. Maintenance/replacement costs unknown.
5. Space and safety hazards unknown.
6. Down time and track modification costs are unknown.
*The cost of shutting down a yard or part of a yard during installation or maintenance of these
systems could double or triple the estimated costs.
4-33
-------
B. Source Control
1, Lubrication Systems (excludes maintenance/operation costs).
a. Specific costs unknown, estimated by industry to be $250,000 to $500,000
per retarder system (master plus 4 to 8 group retarders) or 5 to 10 percent of
total capital investment.
b. Estimated initial cost of new equipment on basis — $ 100 million (assuming
200 retarder systems)
c. Maintenance and operational downtime, and modification costs to track
system, are unknown.
2. Ductile Iron Shoes
a. Initial Cost ($25 per foot) cost is twice that of regular retarder shoes.
b. Ductile shoes wear 10 times faster than regular retarder shoes.
c. Estimated additional cost for using ductile iron shoes to replace present shoes
is $ 100,000 per retarder system.
d. Estimate of national cost impact to industry is $ 100 million (assuming 200
retarder systems)
e. Yard down time is not included in this cost estimate.
3. Releasable Inert Retarders
a. Conversion of non-releasable inert retarders to releasables cost $5,000 per
retarder, not including labor, down time, or operation costs.
b. The number of non-releasable inert retarders in use is unknown. Gross esti-
mate is 20,000.
c. Estimate of national cost to convert $ 100 million.
4. Computer Control of Retarders
a. Computer control of retarders seems practicable only at the newer yards where
computer control systems were installed when the yard was initially built.
b. There are approximately 40 computer controlled yards.
c. The cost, during new construction of a yard, for computer control of a retarder
system is $ 1.5 million.
d. Cost of feasibility of retrofitting a yard with computer control is unknown.
e. If hardware installation costs were assumed to triple the new installation cost,
the national cost impact for retrofit of existing yards for computer control
would be 540 million dollars, assuming 120 retarder systems.
Car-Car Impact Noise
The time histories of car-car impact noise illustrated in Figure 4-10 show some features of the
physical phenomena that accompany car-car impact. The initial impact of the car couplers causes
a "crack," as illustrated by the sharp rise in sound level in both parts of the figure. The high-frequency
4-34
-------
- o
« -g
o
1— "o
.?£
Q. LJ
00
c
o
o
0)
m
Ul
O
IO
o
CM
o
o
o
o>
aj(v)8p] 13A31 QNOOS
0)
o> o
_—I *^_ ^»"
o
0^=
CD U>
o
CM
o
o
00 T3
C
o
o
0)
U> (A
LU
CM
O
O)
13A31 QNOOS
C/3
'5
Z
•4-4
O
w
a
E
6
o
4
4-35
-------
O)
0>
2 sii*
O LUOO 4S
O
O
CO
k.
O
o
•o
O)
CD
JE"°
QJ O O
o *~1
O
to
O
ex)
in
•O
c
Ul
2
w »-
O
o>
o
00
T3A3T QNOOS
00
u>
CSI
(A
•o
o o
00
<0
CVJ
O
O
[jDqT/2000'0 9MV)8PJ 13A31 QNOOS
Figure 4-10. (Cont.)
4-36
-------
portion of the mechanical energy ged into couplers often excites an entire ear bod /. The second
time trace in the figure shows how, as the resulting vibrutional energy decays exponentially, the
radiated noise falls off proportionally. The time trace for a tank car hitting two loaded flat bed
cars shows the noise sometimes generated by secondary impacts as cars pull away from each other
and coupler slack is subsequently taken up. The time trace for the noise measured eight cars away
from a point of impact shows how the energy from an impact can propagate along a chain of cars.
Warning Devices
This source of noise includes bells, horns, and whistles, which are sounded to warn pedestrians
and motorists that a train is approaching a grade crossing. The noise level at 50 ft due to either a
horn or a whistle is 105 dB(A) ± 10 dB(A). Of prime consideration in addressing hose sources of
noiso is tho measure of safety that Ihey provide.
Methods of noise abatement for warning devices have not been fully evaluated. Some localities
hjve required that the devices not be sounded, while others have required just the opposite. Various
alternatives for controlling their noise include requiring reduced levels, specifying directionality, or
limiting the times and areas in which the devices should be sounded.
PvhSic Address Systems
Although tin frequency of occurrence of noise from loudspeakers in railroad yards is sporadic
a:v! unpr?UirUibr;\ the level of the noise from speakers is comparable to the level < >f noise from
other sou "us ;n m1 y:,: 's, Wb.iv abatement is desired 01 necessary, more speakers could be stra-
tegically ideated so i'
-------
REFERENCES TO TABLE4-2
!. R. A. Bly, "Measurement and {/valuation oi the impact of Railroad Noise Upon Communities,"
BBN Repoit No. 26: i, August !<<73.
2. E, K. Bender and R. A. Ely, "Noise Measurements In and Around the Missoi ric Pacific
Centennial Yard, Fort Worth, Texas." BBN Report No. 2648, October 1973,
3, I'lectromotrve Division of General Motors, presentation to American Associa ion of Railroads,
Augusts, 197,1.
4. General Electric, prtseiuation to American Association of Railroads. August 3, i973.
5. J. W. Awing a.f.d D. B. Pies, "Assessment of Noise Lnvironmcnts Around Railroad Operations,"
WyJe laboratories Report \VCR-73-5, July 1973.
6. 0. J. Rickley, Department of T~ansp irt-'.fion, Transpoitatirn Sysior.ia Center unpublished data.
7. M, Alakc-J, C M;dme, M. Rudd, Bolt Ceranek anJ Nev/jnan inc., unpublished data.
8. EPA Region IV study of locomotive noise, unpublished data.
9 EPA Region VII study of locomotive noise, unpublished data,
10, FPA Region VIII study of locomotive noise, unpublished data.
1 i. EPA Region IX studj oflocor^'^ivc noise, unpiib-'isK-ii tfofa.
4-38
-------
SECTION 5
SUMMARY OF WHAT THE PROPOSED
REGULATIONS WILL REQUIRE
"APPLICATION OF BEST AVAILABLE TECHNOLOGY TAKING INTO ACCOUNT
THE COST OF COMPLIANCE"
Section 17 of the Noise Control Act requires that the proposed regulations . . . "reflect the
degree of noise reduction achievable through the application of the best available technology, taking
into account the cost of compliance." For this purpose, "best available technology" is defined as
that noise abatement technology available for application to railroads which prod ices meaningful
reduction in the noise produced by railroads. "Available" is further defined to in ;lude:
i. Technology which has been demonstrated and is currently known to be feasible.
2. Technology for which there will be a production capacity to produce tl e estimated num-
ber of parts required in reasonable time to allow for distribution and installation prior to
the effective date of the regulation.
3. Technology that is compatible with all safety regulations and takes into account opera-
tional considerations, including maintenance, and other pollution control equipment.
The "cost of compliance," as used in the proposed regulation, means the cost of identifying
what action must be taken to meet the specified noise emission levels, the cost of taking that action,
and any additional cost of operation and maintenance caused by that action. The cost for future
replacement parts was also considered.
As discussed in Section 4 of this report, the only source of railroad noise proposed to be
regulated by the Federal government at the present time is trains. Therefore, the ioUowing pages
will discuss the noise abatement technology for trains, in consonance with the statutory require-
ments and interpretation presented above.
Train noise is composed of locomotive noise and car noise. The latter is primarily the result
of wheel/rail interaction and wheel/retarder interaction. The locomotive noise is :omposed of
noise from the engine exhaust, casing, cooling fans, and wheel/rail interaction. Tl e technology for
treating casing, fans, and wheel/rail noise is in the early development and research stages and thus
not "available" for application at this time. However, at the present time, the technology for
exhaust silencing has been found to be "available." Further, the locomotive noise is dominated
5-1
-------
by ihe engirt,' ^-xhuusi' noise and, there '"ore. the application of exhaust muffler te :h lology is the
'•»•'«; i!ljvct'"'t ii'Hi..:! >>fep to require to locomotive noise abatement, The consequences of estab-
Uj.,i'.;g u iiauciurd fr.u wcuju require '.nodifkjjtion of engine casing, cooling fans, and wheel/rail
if'tor»..ron iuve net bet-p assessed m detail. J( is c'ear, however, that without first reducing exhaust
ST.--;'-,' 'sc-'.tnv-i-t o! la^se errnpoiMijb woulH result in little1 or no perceptible noise reduction.
\'!nV'f-.r tecl\r,o?ogy h; svoj' .:, have ne?n precii-rte^ and J!i the, judgment of fhe Agency constitutes the "appli-
Ci'tRT; of bt^t aiai!'uv,».- iccrsclogv inking inio ;ic^outU the cost of compliance."
J ,.vn,S OF TT;A?N
i;) tuns S"'<:tK>.', t'O'se wvels Uia' .• 1,1 M; rtx,: i nabiy all lined wi.h appropriate n aintenance of
, AtiS.ng e 4''!.j: ;7vr;t i"d b".' -li;? appljCjiion of the host available technology are discussed for loco-
:n;>nves both r,: rest and iu- motion and for railcars in motion.
Locomotive Nois?: Vehicle a,' ,Rt\;t
As discussed iri t>> o-ion 4, "ocoiu'oiw. noise is dominated by (he exhaust of di 'sei engines, which
opentc .1! eitfht possible speed and power output levels. One way to attain environmental noise
i^n'ivi wou'.j be to !»mit tl"- r^oisc .-,' 'il o* ihi sc throtM«' settings, hcive/er. this could lead to cum-
b-Tsot.ie iM.iv';cetn".jit practi/ta'- Por taj>c ofenfoax-iurnt. pcimis&ible noise could Jic specified at the
tr«a rtlj v.ning witii the most noise throttle 8. However, this approach may lead muffler manu-
facture's to u* ;.ign rriiif{\vrs that are tuned to the engine speed corresponding to that throttle setting.
Sui •!) iauJfteJi> cuuid he effective, at the design setting and ineffective at other settings. Obviously,
A io.irpr ><---ist. soiut.an is 10 convrcl lovoiaotive no\se ai two conditions: idle and full power.
i«iL anJ fa-' PDA-,:/ ^ppl- '••-* fr;.qj-;;. 'V UVM! au^aJc sct'.ing.^. Specif>ir.g two thro tie settings will
prcfhibiy rieci sd-j ri.a j«Iin of ;,po,:u»ily tuned iifuffleru Rather, it is anticipated that mufflers that
viii be UiiJi'o-rrtiv tiT,^;five at all r.irottl-j settings will rest..!,,
Although {•! is i'nreaiiitic ?o assumt- that niafflcrs can be designed, fabricated, and installed on
,»-.'..unuij-cs &s -toon .fS «. i-jpuittior* ^ n- M^ul^ateH, ,^ ii, noi unreasonable '.J hold noise at the level
n>f '-x,;,an;.». %v \?-n;;,'i»tained "q..ipi''':'""* Hat;-, -o,' .:•• ; - 'h-.^tiv^ at tlirottle setting 3, indicate that
.alri'u:-? -1*' Kvor.'.c^y.'S exceed ().j JSfA; at iOv) It. likewise, data iridicatd t-'iat locomotives at idle
can bv :-x;v\ ic.-d not u en. s ?iore /Ji 'n '/? c'B! ^,1 'if 100 ff. Accordingly, the following levels have
hc;n idrv'^ii'.'rJ .s -n-jic^tivc of prese'-t no.se emissions:
'/.»
Section 4 indicates v.ji Ui>i,",;w'r. •-.:», uolj «.-i ttfiluciu" •, .-..t^:,,;,i nois-c by 10 dB{A) are feasible.
. pen-ji."*,-: upo-i th-: rv^n-. cofan'r.ui. - •-; ;..v . ;;;;jus» '.oiss 10 ih-, ciormnent sources of locomotive
-------
noise, this reduction may produce a 4 to 8 dB(A) reduction in the total noise (see Table 4-5). It
is believed that the noisier locomotives have a higher exhaust noise component and, therefore, may
achieve greater overall reduction in total noise by reducing exhaust noise. When exhaust noise is less
dominant, smaller reductions in total noise will result. However, in this case, overall noise seems to
be initially lower. Based on the considerations of available empirical data, an overall noise reduction
of 6 dB(A) for the noisier locomotives seems reasonable. Accordingly, the application of an exhaust
muffler can be expected to permit all locomotives to achieve the following levels:
Idle 67 dB(A)
Overall Maximum 87 dB(A)
The exhaust noise is primarily a function of the diesel engine horsepower and the method of
engine aspiration. Rootes blown engines would have higher exhaust noise than an equal size turbo-
charged engine. Also, a larger engine has higher exhaust noise than a smaller engine if the aspiration
is the same.
However, the larger engines are generally turbocharged, while the small engines are rootes
blown. This leads to a partial cancellation of the effect of power and aspiration on the exhaust noise.
It may be feasible in the future to establish separate standards for different types of locomotives,
depending upon power or method of aspiration. This is not possible with the present data, however.
Section 4 also shows that muffler manufacturers could supply the needed hardware after
approximately 2 years for design, development, and testing. Allowing another 2 years for installa-
tion (see Section 8 of this document for a discussion of installation costs), a 4-year program for
completion of muffler retrofit appears reasonable.
Locomotive Noise: Vehicle in Motion
In addition to the stationary locomotive standard a pass-by standard which relates directly to
the manner in which locomotives operate in the environment is also desirable. Such a standard also
could be a useful tool for adoption and enforcement by local and State governments.
Based on available train pass-by data (see Figure 4-3) 96 dB(A) measured at 100 feet is achiev-
able and represents the status quo for current locomotive noise emissions. As discussed above, a
reduction in overall locomotive noise of 6 dB(A) for the noisier locomotive through proper muffler
application is considered reasonable. Therefore, using the same projected design, development,
testing, and installation times mentioned above a 90 dB(A) noise emission level measured at 100 feet
for all locomotives during a pass-by test would be required in four years.
There seems to be a general relationship between the load cell and pass-by levels prescribed.
The maximum levels observed differ by approximately 3 dB(A). This relationship cannot be
definitively stated since measurements comparing the two procedures have not been conducted
under controlled situations. However, by proposing both pass-by and load cell measurement tests
in the proposed standard the public is allowed the opportunity to comment on both.
5-3
-------
K sucar Notes: Vehicles in Motion on Line
Figure 4-8 shows that at a given speed, railcar noise ranges ±5 dB(A) above or below a mean
valc.e. At 45 mph the mean is approximately 83 dB(A). At 60 mph the mean is approximately
?* '15^ 4). As such, the following status quo standard measured at a 100 ft. distance for railcars
••.jp >:» considered appropriate:
Railcar Speed (v) Noise Level
mph dB(A)
V<45 88
V>45 93
Raiicar Noise: Vehicles in Motion in Yards
As discussed in Section 4, railcar passage through a retarder causes the emission of noise levels
as high as 120 dB(A). Further discussed, were five possible methods of retarder noise control that
might conceivably be employed individually or in concert. With such information it might be
argued that a status quo level of 120 dB(A) may be appropriate at this time and subsequently
reduced to approximately 80 dB(A) as the technology of retarder noise control advances over the
next few years. At this time, however, it is the Agency's position that retarder noise is an element
of fixed facility railroad yard noise which, as such, can best be controlled by measures which do
not in themselves affect the movement of trains and therefore dc not require national uniformity
of treatment. Such noise control measures might include, for example, the erection of noise
barriers. The Agency's study of railroad yard noise indicates that concern for noise from railroad
yards is more local than national. This is due in large part to the location of the number of yards
in non-urban areas and the relatively small number of hump yards (130). Accordingly, the estab-
lishment of a uniform national standard could potentially incur significant costs to the railroads
with only limited environmental impact resulting in terms of population relief from undesirable
noise levels.
In summary, the principal reasons for proposing to not regulate retarders at this time are:
1. The technology and cost information on retarder source control is not adequate at this
time to justify inclusion in proposed regulations.
2. Application of barriers (which is a general technology applicable to many noise sources)
to reduce retarder noise is more appropriately handled by local or State jurisdiction on
a case-by-case basis.
3. EPA studies of models of environmental benefit resulting from reducing retarder noise
imply only a small benefit on a national basis. This is due largely to the relatively small
number of hump yards. (See Appendices C and D).
5-4
-------
SECTION 6
GENERAL PROCEDURE TO MEASURE RAILROAD NOISE
.INTRODUCTION
In developing this Background Document/Environmental Explanation and proposed standard.
EPA has reviewed several methods which may be used to measure railroad noise emissions. The
procedures used by EPA to measure railroad noise conform in general with the measurement pro-
cedures described in this section. The Agency believes this procedure to be reasonable for the
purpose of measuring railroad noise, and suggests it for use by other parties in the measurement of
such noise emission.
If issue is taken with the data supporting the railroad standards proposed by EPA, such data
as may be submitted to the Agency in support of the respondent's position should be based on
similar measurement methods or procedures. The equivalency or correlation between different
measurement practices must be clearly explained in order to permit adequate comparisons with
the data and levels in the proposed regulation.
It is recommended that technically competent personnel select the equipment to be used for
the test measurements. Proper test instrumentation and experienced personnel is essential to obtain.
valid measurements. Operating manuals or other literati ,e furnished by the instrument manufac-
turer should be referred to, for both recommended operation of the instruments and precautions
to be observed.
MEASUREMENT INSTRUMENTATION
A sound level meter that meets all the requirements of American National Standard SI .4-1971
for a Type I instrument and all requirements of International Electrotechnical Commission (JEC)
Publication 179(1965) should be used with the meter set to "fast" response. Alternative/additional
measurement instrumentation, such as a magnetic tape recorder or a graphic level recorder, may be
used for conducting the measurements, provided that the overall performance of the measurement
system conforms to the requirement of this Measurement Instrumentation Section over the
frequency range from 25 Hz to 1 OK Hz. In conducting the measurements of sound level, the
general requirements and procedures of American National Standard SI .13-1971 should be followed.
These publications are available from the American National Standards Institute, Inc., 3420 Broad-
way, New York, New York 10018.
6-1
-------
A wind screen that does not introduce measurement uncertainties in excess of plus or minus
0.5 dB(A), should be used at all times. No sound level measurements should be taken when the
wind speed near the microphone exceeds 20 km/hr (12 mph).
The sound level meter, or other measurement instrumentation, should be calibrated (e.g., by
;neans of a pistonphone) at one or more frequencies, at the beginning and end of each series of
measurements. The calibrator should produce a sound pressure level, at the microphone diaphragm,
that is known within a precision of plus or minus 0.5 decibel. The calibrator should be checked
monthly to verify that its output has not changed.
A complete frequency response calibration of the instrumentation over the entire frequency
range of 25 Hz to 1 OK Hz should be performed at least monthly using methodology of sufficient
precision and accuracy to determine compliance with American National Standard SI-4-1971 and
!EC 179. This calibration shall consist, at a minimum, of an overall frequency response calibration
and an attenuator (gain control) calibration plus a measurement of dynamic range and instrument
noise floor.
TEST SITE PHYSICAL, ACOUSTICAL, WEATHER AND
BACKGROUND NOISE CONDITIONS
In general, the test site should be selected such that the locomotive or train radiates sound
over the ground plane of an open space free of large, sound reflecting objects, such as barriers, hills,
signboards, parked vehicles, bridges or buildings withiri the boundaries described by Figure 1 (rail
car or locomotive noise pass-by test) and Figure 2 (for stationary test). In addition, the following
specific conditions are also suggested:
1. The track bed within the test site described in Figures 6-1 or 6-2 should be visible by
direct line of sight from a position 4 feet above the ground at the microphone loca-
tion, which is also described in Figures 6-1 or 6-2.
2. The terrain between the vehicle under test and measuring microphone should be relatively
free of ground covering having excessive sound absorption characteristics.
3. The ground elevation at the microphone location should be within plus or minus 3 meters
(10 feet) of the elevation of the track bed at the location in-line with the microphone.
4. Within the test section, the track should exhibit less than a 2 degree curve [or a radius of
curvature greater than 873 meters (2,865 feet)]. This does not apply during a stationary
test. The track should have tie and ballast in good condition and preferably welded rails,
and be free of special track work such as turnouts or crossovers, and bridges or trestles.
5. Measurements should not be made during precipitation.
6. Maximum background noise at the microphone location of Figure 6-1 or 6-2, immedi-
ately before and after the test, should be at least 10 dB(A) below the level measured
during the test. Measurements should be made with the sound meter set to fast response.
6-2
-------
Microphone
Location
Figure 6-1. Test Site Clearance Requirements for Wayside
Rail Car Noise Pass-by Test
6-3
-------
Figure 6-2. Test Site Clearance Requirement for Locomotive Stationary Test
-------
7. Corrections for measurements at varying altitudes should be made in accordance with
recommendations of the instrumentations manufacturers for altitudes greater than
1,000 meters (3,000 feet) above sea level.
PROCEDURES FOR THE MEASUREMENT OF LOCOMOTIVE
AND RAIL CAR NOISE EMISSIONS
Introduction
One procedure for the measurement of locomotive noise is to connect the locomotive to a
load cell where it can be loaded by feeding its electrical power into resistor grids. Since a load cell
may not always be available or conform to test site requkements (see Figure 6-2), alternative ways
of measuring locomotive noise often are used. These include stationary self-load testing for loco-
motives which are so equipped, and pass-by measurements of locomotives. The procedures relating
to rail car noise emissions are for the pass-by condition.
General Requirements
The noise emitted by the locomotive should be measured from both sides when connected to
a load cell or under self-load test, if possible. The test site should be selected in accordance with
requirements of the previous section on physical, acoustical, weather and background noise condi-
tions. Measurement on both sides of the locomotive would not be done for uncontrolled pass-by
measurements.
For the stationary locomotive tests, the microphone should be positioned at a point on a line
normal to the track and 30 meters (100 feet) from the center of the locomotive.
For moving rail car and locomotive tests, the microphone should be 30 meters (100 feet) from
the track center line.
In all cases the microphone should be positioned 4 feet above the ground, with its diaphragm
oriented toward the source in accordance with the manufacturer's recommendations to provide the
most uniform frequency response.
The observer should be at least 3 meters (10 feet) from the microphone. Under no circumstances
should an observer stand between the microphone and the source whose sound level is being
measured.
To assure that adequate information is collected for each test it is recommended that the
following data be recorded:
1. Name and precise location of test site
2. Locomotive: manufacturer, type, model, serial number and horsepower rating
3. A-weighted sound pressure levels as determined in the test described below
4. Altitude, above sea level, of the test site
5. Prevailing wind speed and direction at the time of the test
6-5
-------
6. Date and time of day of the test
7. Name and identification of the person(s) making the test
8. Model and serial number of test instrumentation.
Two types of sound measurement tests seem particularly applicable for rail carrier noise
emissions. These are the load cell test for stationary locomotives, and the wayside test for moving
locomotives and rail cars. For load cell tests, measurements should be repeated at least three times
for each side of the locomotive which is measured. The highest of the two arithmetic means, of the
sound levels observed for each side, should be the sound level recorded. This is not possible for
uncontrolled pass-by measurements. Only one measurement need be made for the uncontrolled
wayside noise pass-by test for locomotives and rail cars.
Locomotive Load Cell and Self-Load Noise Emission Measurement
Measurement should be made at several throttle settings, with engine cooling fans operating;
however, as a minimum, settings corresponding to idle and maximum engine power should be
mandatory. The maximum engine power setting for most locomotives will correspond to setting
eight. The sound level meter should be observed for thirty seconds after the test throttle setting is
established. The maximum sound level observed during that time should be recorded,
Locomotive Pass-by Noise Emission Measurement
Locomotive noise measurements should begin when the locomotive, or combination of loco-
motives, is within 60 meters (200 feet) of the measuring position (as measured along the track) and
continue until the last locomotive has passed at least 150 meters (500 feet) or is 10 rail car lengths
away from the measuring point. The maximum sound level observed in this manner should be
recorded. Locomotive acoustical warning devices such as horns, whistles and bells should be
excluded in selecting the maximum sound level observed. *
Rail Car Pass-by Test Noise Emission Measurement
Rail car noise measurements should begin when the locomotive or combination of locomotives
has passed a distance of 300 meters (1,000 feet) or 20 rail cars beyond the measuring position.
There should be no other locomotives within 300 meters (1,000 feet) or 20 rail car lengths from
the measuring point. The maximum sound level observed in this manner should be recorded.
6-6
-------
SECTION 7
ECONOMIC EFFECTS OF A RETROFIT PROGRAM
INTRODUCTION
The imposition of a railroad muffler retrofit program will affect both the railroads and the
industries that purchase transportation services. Minimal changes in transportation patterns may
be expected as a result of a retrofit program since increases in cost per ton mile of freight moved
are estimated to be small. The purpose of this portion of the background document is to examine
the possible magnitude of such effects; their consequences in terms of railroad viability and the
transportation of commodities; and techniques by which adverse economic impacts might be
avoided.
The study presented here relies on a number of information sources and makes a number of
assumptions in the course of arriving at quantitative estimates of impact. Data on costs of materials
and labor for retrofit program were obtained chiefly from muffler manufacturers and railroad
personnel. Information on locomotive maintenance requirements was likewise obtained from the
railroads. Operating and financial statistics for individual roads and the industry as a whole came
from reports of the Interstate Commerce Commission. To project the ultimate economic effects
of incurred costs, assumptions were required concerning future trends in railroad activity. In some
cases for which a range of assumptions was possible, the alternative least favorable in terms of
impact was chosen; in this sense, the analysis represents somewhat of a "worst case" approach.
Wherever assumptions are made, however, they are substantiated to the extent allowed by existing
data.
THE IMPACT ON THE RAILROAD INDUSTRY
General Impact
The engineering data gathered from discussions with various manufacturers and railroad oper-
ating personnel were used to estimate the direct cost of muffler retrofit by locomotive type and
manufacturer. The differences in construction between switcher and road locomotives required that
these be treated separately. The three categories of direct cost are mufflers, additional hardware,
7-1
-------
and labor. Since each make of locomoti /e is somewhat unique, it was necessary to make separate
analyses of each type. The costs are shown in Table 7-1. The retrofit costs associated with the
various types of locomotives are based on the designs of several common types, which make up
about 90% of the population. For some locomotives, retrofit costs may be significantly higher
ch ..» the figures shown here. This may be the case, for example, for several hundred units which,
"Ul> oi.gh originally conforming to one of the common designs, have been heavily modified during
service so that their configurations now present difficult hardware problems to a muffler installer.
Also, there are some 1,000 older road locomotives manufactured by Alco and Fairbanks-Morse
and owned by a total of 22 railroads, the design of which may render muffler installation difficult.
The Agency has been advised that these units are, in fact, in the process of being replaced. Thus
this discussion assumes that such units will be retired from service during the compliance period.
The estimates of the direct cost of mufflers and additional materials were gathered from
locomotive and muffler manufacturers; the sources of the data on required labor input were loco-
motive manufacturers, muffler manufacturers, and management personnel of selected railroads.
An hourly wage rate of $5.80 per hour was arrived at by taking total compensation of main-
tenance personnel as reported in annual ICC summaries and dividing by total hours worked.*
Although this wage rate probably includes some overtime compensation, it may be an accurate
TABLE 7-1
MUFFLER COSTS* PER LOCOMOTIVE
(Source: Manufacturers' and Operators' Estimates)
Time of Installation
New Production
Muffler Only
Additional Hardware
Labor @ 5.80/hr
Total
Locomotive Manufacturer and Type
GM
Road
$3000 (RB)
2500 (TC)
1500
200- 500
464- 1163
$2164-3163
GM
Switcher
$200 - 500
200 - 500
46
$246 - 546
GE
Road
$1500
1500 .
1500-2500
187
$3187-4187
Other
Road
1500
1500-2500
187
$3187-4187
Other
Switcher
500 - 800
46
$546 - 846
(RB) = Rootes Blown
(TC) = Turbocharged
*A1J railroad data presented in this section come from Interstate Commerce Commission,
Transportation Statistics in the U.S. (1971) unless otherwise specified.
7-2
-------
reflection of the true labor cost, since some retrofitting may be done at the overtime rate. We
assume that the current mix of straight time and overtime will be used in the retrofit program.
No capital costs for maintenance facilities were assigned to the retrofit program. Annual
compensation statistics and discussions with the Association of American Railroads indicate that
the roads have been generally cutting back their maintenance staff over the last decade, while
not necessarily reducing the size of their plant.* Frequently, therefore, excess physical capacity
would be available for a retrofit program. In an economic, although not necessarily an account-
ing sense, such excess capacity can be utilized at zero cost.
The next step was to determine how many of each type of locomotive are in service. The
May 1973 issue of Railway Locomotives and Cars lists the make and horsepower of each loco-
motive in service by railroad. In most cases, the horsepower of the engine could be used to deter-
mine whether it is a switcher or road locomotive. General Motors (GM) produces both a 1500-hp
switcher and a 1500-hp road locomotive, but because road locomotives outnumber switchers by
about seven to one, we assumed all General Motors 1500-hp locomotives to be road locomotives.
This biased the cost estimates upward by a small amount. Table 7-2 shows the distribution of
locomotives by type and manufacturer both nationally and for each of the three ICC regions.
TABLE 7-2
DISTRIBUTION OF LOCOMOTIVES BY MANUFACTURER, TYPE, AND REGION
(Source: "Railway Motive Power, 1973," Railway Locomotives and Cars, May 1973)
Manufacturer
and
Type
GM Road
GM Switcher
GE Road
Other Road
Other Switcher
Region
Total
16,155
2,811
1,930
1,737
1,504
East
(29 Roads)*
7,006
1,462
878
1,052
734
South
(8 Roads)*
2,026
304
230
289
139
West
(22 Roads)*
7,123
1,045
822
396
631
*Number of roads in each district obtained from ICC, op. cit. Other listings of roads may not tally with
this one, due to varying methods of accounting for mergers, subsidiaries, etc.
*Sources in the AAR state that this may not be the case for roads which have recently modern-
ized their plants and which may have divested themselves of some unneeded facilities. In these
cases, according to the AAR, the cost of installing or renting the needed plant and equipment
may significantly increase retrofit costs. Unfortunately, precise estimates of capital stock in
maintenance facilities do not exist.
7-3
-------
Total direct cost of the retrofit program was obtained by multiplying the cost per locomotive
by the number of locomotives.* This is given in Table 7-3 in terms of minimum and maximum
costs for each region and for the entire nation.
TABLE 7-3
TOTAL DIRECT COST OF RETROFIT PROGRAM
(Millions of Dollars)
Region
East
max.
min.
West
max.
min.
South
max.
min.
National
max.
min.
Locomotive Manufacturer and Type
GM
Road
$22.160
15.161
22.530
15.414
6.411
4.386
GM
Switcher
$0.798
0.360
0.570
0.257
0.166
0.075
GE
Road
$3.676
2.798
3.442
2.620
0.963
0.733
Other
Road
$4.405
3.353
1.659
1.262
1.210
0.921
Other
Switcher
.$0.621
0.401
0.534
0.345
0.118
0.076
Total
$31.660
22.073
28.735
19.898
8.868
6.191
69.263
48.162
The annual direct costs in Table 7-4 were derived from Table 7-3 by dividing total cost by the
number of years allowed to complete the retrofit program. In addition, the annual cost for 2- and
5-year compliance periods is shown as a percentage of the 1971 net operating revenue. It should be
noted that we are assuming 2 and 5 years beginning at the time the muffler becomes available.
*Normally, some locomotives would be retired during the compliance period and, therefore, would
not incur retrofit costs. (Their replacements would presumably have been quieted at the factory.)
This consideration has not been included here, because it is difficult to forecast replacement rates
in the light of an endemic shortage of motive power such as presently exists. If we assume instead
that past retirement rates (about 2000 units per year from 1965 through 1969) are cut in half due
to the shortage of locomotives, this will result in 5000 fewer units needing muffler retrofit for a
5-year compliance period and 2000 fewer over a 2-year period. The total cost estimates projected
above would then be high by about 20% and 8% for the two compliance periods, respectively.
-------
oo
a.
LU
OS
w
S Q
U.
O
H
on
8
W
oi
2
<*-!
O
u
1? §
4-> 7
S S
g s
^^
a."
-*-> ^
s
u
M
g|
2-§
+-* (
o 'S
Si
*H c/)
s^
r-j nj
C<3 [«
•*-• 3
^1
•4— >
t^
5«
u
>r
in
!3
4)
>r
(N
t-i
cd
D
><
m
S3
o>
>-
C4
c
S
^
M
s
c
i
X
03
s
c
s
X
s
c
• »"l
s
X
cd
2
_O
^
oo t— rn O
ro in *• m r-
d — ^ d d
m Tf
-------
Generally, mufflers will not be available until 2 years after the regulation is promulgated, so that
the 2-year program will not be completed until 4 years after promulgation, and the 5-year program
until 7 years after promulgation.
It appears that the direct cost of a retrofit program will not constitute a significant burden on
i.e railroads. Total direct cost is invariant with respect to compliance period, although annual cost
is not. Annual cost is, therefore, probably a more relevant measure of the financial impact on the
railroads.
The direct cost of retrofitting mufflers is only part of the total cost, however. If retrofitting
requires that locomotives be taken out of service and if the railroads have no excess capacity with
respect to locomotives then there will be some loss of revenue. At present, most railroads are oper-
ating at full capacity. The number of locomotives has decreased slightly from 1965 to 1973 (from
27,988 to 27,041) although total horsepower did increase from 52 million in 1971 to 55 million in
1973. It appears, therefore, that capacity has remained about constant or decreased slightly while
demand has increased. It seems unlikely that the present high volume of grain shipments will con-
tinue beyond a year. Other factors, however, indicate that the current high levels of capacity utili-
zation will probably continue into the future.
One of the developments that will tend to keep rail transportation at a high level of capacity
utilization is the "energy crisis." A general fuel shortage favors the railroads over other modes of
transportation. An increase in coal output, which seems inevitable, would stimulate rail freight
volume. Coal, because of its low value per ton, is hauled almost exclusively by rail.
A further impact of a general fuel shortage would be to potentially degrade the quality and
cost of truck transport relative to rail service. Restricted speed limits could induce delays and
uncertainties in truck schedules. Fuel price increases would have a greater adverse impact on trucks
than on rail, since trucks use 3.2 times as much diesel oil per ton mile of freight. As a result, trans-
portation demand would tend to shift from trucks to rail. The net effect of these considerations is
to support the assumption that railroads will be operating at close to full capacity for the next 5
or so years. This means that locomotive downtime due to retrofit may likely result in lost revenues.*
The time lost may be significantly reduced by scheduling retrofits during regular locomotive
maintenance. Nationally, the average maintenance cycle is 4 years for an intermediate overhaul and
8 years for a heavy overhaul. The length of the cycle for an individual railroad is a function of
*One way in which operators may overcome this problem is to buy new locomotives to take the
place of those being retrofitted. Such a procedure would virtually eliminate the indirect cost
associated with the retrofit. This is an option, however, only if the locomotive manufacturers
can produce the extra units. At present, according to locomotive manufacturers, locomotive pro-
duction is below demand even though production facilities are operating at full capacity. It is
reasonable to assume that conditions of motor power shortage relative to demand for transporta-
tion will persist throughout the compliance period, resulting in lost revenue when units are
removed for retrofit.
7-6
-------
locomotive mileage. Table 7-5 shows the national average adjusted regionally to retleet different
average locomotive miles per year. The maintenance cycle is shortest in the West where locomotives
travel more miles per year and longest in the East where miles per year are lowest.
TABLE 7-5
AVERAGE MAINTENANCE INTERVAL BY DISTRICT (years)
(Source: 1971 ICC Statistics and Operators' Estimates)
Type of
Maintenance
Intermediate
Heavy
Regional Average Maintenance
Interval (Years)*
National
4.0
8.0
i
East
5.5
11.0
South
4.0
8.0
West
3.5
7.0
"These figures do not include the effects of deferred maintenance as practiced by some roads in financial distress.
An intermediate overhaul generally takes about 2 to 3 days, while a heavy overhaul takes about
14 days. The estimated time required to retrofit a muffler ranges from 3 days for a General Motors
road locomotive to 1 day for a switcher. Table 7-6 shows the number of lost locomotive days
"charged" to retrofit under different conditions. Line 1, for example, gives lost days by type of
locomotive if the locomotive is taken out of service specifically for retrofit. One can see that there
are no lost days for any type of locomotive if all retrofitting is done during heavy overhaul.
TABLE 7-6
DAYS LOST DUE TO RETROFIT
(Source: Manufacturers' and Operators' Estimates)
Basis of Retrofit*
If done by itself
If done during regular
intermediate overhauls
If done during regular
heavy overhaul
Locomotive Manufacturer and Type
GM
Road
3
1
0
GM
Switcher
1
0
0
GE
Road
2
0
0
Other
Road
2
0
0
Other
Switcher
1
0
0
'Assumes no lost time due to travel to and from shop and no muffler retrofitting done during emergency repairs.
7-7
-------
As is shown, the total lost locomotive time due to muffler retrofits depends on how many
locomotives can be treated during the normal maintenance cycle. Table 7-7 shows the expression
used to compute total lost days for each line or district. The first term represents the time lost by
GM road locomotives undergoing intermediate overhaul. The remaining three terms account for
time lost by those locomotives that will not be due for routine maintenence during the compliance
period and which, therefore, must be specially called in for muffler retrofit. (Recall from Table 7-6
that, except for GM road locomotives, units undergoing intermediate or heavy overhaul will exper-
ience no extra time lost due to retrofitting a muffler.)
TABLE 7-7
EQUATION FOR TOTAL LOST TIME PER DISTRICT
LT =
NGM x
2T
NSW X
m
X Y X 1 day
X 3 days
m
X 1 -
m
T
1
m
X 2 days
|X 1 day
for
> 0
Lm
where
2 NGM x
dav
for
N
SW
Lm
= number of years allowed for retrofit
= number of GM road locomotives
= number of GE and "other" road locomotives
= total number of switchers of all makes
= time interval for "Intermediate" maintenance
Lm
7-8
-------
The equation in Table 7-7 has been used to compute lost locomotive days for each region.
These have been summed to give a national total. The figures are shown in Table 7-8. Two
compliance periods are used to illustrate the decrease in lost time with a longer retrofit period.
We see from the table that increasing the period from 2 to 5 years results in a decrease of the lost
locomotive days per year by 70 percent.
A change in the compliance period affects only the number of lost locomotive days; the direct
cost of the retrofit program does not change. If we take the total number of lost locomotive days
resulting from a 2-year period and assign it the number 1, then the total number of lost days for a
3-year program is 0.76, the total of a 4-year program is 0.52, and the total of a 5-year program is
0.29. As the compliance period is lengthened, lost locomotive days decrease; thus, the indirect
cost of the program decreases.
The calculations of lost locomotive days must be translated into dollar costs. A number of
problems arise in calculating the value of a locomotive. First, should a distinction be made between
road locomotives and switchers? It seems desirable to treat the transportation revenue earned by
rail service as being earned by both road and switch engines, since the lack of either (if both are
used to full capacity) would cause a reduction in service. We have therefore assumed that each has
the same value per day.
Secondly, what value should be assigned to a locomotive day? If all roads are operating at
full capacity, then removing a locomotive causes a daily loss of revenue amounting to the value
of one locomotive day. A locomotive day is thus evaluated at the value of the average product.
This technique is further justified in capital theory, which states that the value of a piece of
capital is the present value of its discounted future stream of earnings, that is, the present value
of the marginal product.
TABLE 7-8
LOST LOCOMOTIVE DAYS BY REGION AND COMPLIANCE PERIOD
Compliance
Period
2-year
program
5-year
program
Lost
Locomotive
Days
Yearly
Total
Yearly
Total
Region
National*
17,048
34,096
2,044
10,220
East
(29 roads)
9,252
18,504
1,129
5,645
South
(8 roads)
2,143
4,286
203
1,013
West
(22 roads)
6,378
17,048
712
3,562
* Locomotive days lost nationally is not the sum of the three regions, since the national was calculated
using an average maintenance cycle and the regional was adjusted to reflect different utilization rates.
7-9
-------
Given the conditions stated above, the value of a locomotive day was calculated by taking
total transportation revenue and dividing by the total number of locomotive days available.
Table 7-9 shows these calculations nationally and regionally. Table 7-10 gives estimates of
the indirect costs of a 2- and 5-year retrofit program by incorporating the lost locomotive days
from Table 7-8 and the value of a locomotive day from Table 7-9. Note that the shorter the
compliance period the larger the total indirect costs. This is a function of the increase in the
number of lost locomotive days as the compliance period is shortened.
TABLE 7-9
REGIONAL ANNUAL REVENUE PER LOCOMOTIVE DAY
Total tranportation
revenue (millions of S)
Transportation revenue
per locomotive day ($)
Region
National
$12,417
1,251
East
$4,497
1,186
South
$2,121
1,256
West
$5,799
1,304
TABLE 7-10
ESTIMATED LOST REVENUE DUE TO RETROFIT
(Thousands of Dollars)
Region
National
East
South
West
2- Year Program
Per Year
21,982
10,973
2,692
8,317
Total
43,963
21,946
5,383
16,634
5-Year Program
Per Year
2,557
1,338
254
928
Total
12,785
6,690
1,270
4,640
7-10
-------
Table 7-11 arrives at the annual net retrofit cost by combining the direct and indirect costs
and subtracting the reduction in operating costs that would occur as a result of a reduction in traf-
fic. Cost reductions were determined from the ICC detailed accounts and include the following:
Account No.
Description
365
367
368
371
373
374
382
383
387
388
395
Dispatching Trains
Weighing, Inspection, & Demurrage Bureaus
Coal and Ore Wharves
Yard Conductors & Brakemen
Yard Enginemen
Yard Switching Fuel
Train Enginemen
Train Fuel
Trainmen
Train Supplies and Fuel
Employees' Health and Welfare Bureaus
The estimates of cost reductions used here are much lower than those used by the ICC.*
They have claimed that 80 percent of costs are out of pocket or variable costs. This might be
true if railroads were curtailing service in the face of falling demand. Variable cost may constitute
80 percent of total cost, but the situation dealt with here is an unplanned reduction in capacity
in the face of full utilization of equipment. Under these circumstances, it seems unlikely that the
railroads would curtail other operations but rather that they would attempt to offset locomotive
shortages by changes in labor and equipment usage patterns. In addition, if there are adjustment
costs and since the cutback in capacity is temporary, the railroads would be expected to respond
differently from a situation in which the reduction was anticipated to be of longer duration.
Table 7-12 gives the total net cost of the 2- and 5-year programs. Again, it points up the cost
differential associated with different compliance periods. Much of the computed retrofit cost is
the result of lost revenue to the railroads. Figure 7-1 shows the breakdown of annual cost into
direct and indirect components for compliance periods of 2 to 5 years.
*See U.S. Interstate Commerce Commission, Bureau of Accounts, Explanation of Rail Cost Finding Procedures
and Principles Relating to the Use of Costs. St. 7-63, Washington, D.C., 1 November 1963 and U.S. Interstate
Commission, "Rules to Govern the Assembling and Presenting of Cost Evidence." Docket No. 34013,321 I.C C
238 Order of April 16, 1962.
7-11
-------
TABLE 7-11
ANNUAL NET COST OF RETROFIT
(Thousands of Dollars)
Direct Cost
2-year program
max
min
5-year program
max
min
National
$34,632
24,082
13,853
9,633
East
$15,830
11,037
6,332
4,415
South
$4,434
3,096
1,774
1,238
West
$14,368
9,949
5,747
3,980
Indirect Cost
2-year program
5-year program
21,982
2,557
10,973
1,338
2,692
254
8,317
928
Reduction in
Operating Costs
2-year program
5-year program
4,964
597
2,748
335
555
53
1,856
207
Net Cost
2-year program
max
min
5-year program
max
min
51,650
41,100
15,813
11,593
24,055
19,262
7,335
5,418
6,571
5,233
1,975
1,439
20,829
16,410
6,468
4,701
7-12
-------
TABLE 7-12
TOTAL NET COST OF RETROFIT PROGRAM
(Thousands of Dollars)*
Compliance
Period
2 years
3 years*
4 years*
5 years
National
Max
103,300
95,221
87,143
79,065
Min
82,200
74,121
66,043
57,965
East
Max
48,110
36,675
Min
38,524
27,090
South
Max
13,142
8,875
Min
10,466
7,195
West
Max
41,658
32,340
Min
32,820
23,505
"These represent linear interpolations of the 2- and 5-year programs.
The annual costs shown in Table 7-11 are best understood in the context of total operating
revenue for each region. Table 7-13 shows that the eastern roads would pay a higher percentage
of total revenue toward a retrofit program than would the other regions.
Annual retrofit cost as a percentage of net operating revenue* gives the best indication of the
rail industry's ability to pay for a retrofit program (see Table 7-14). Retrofit constitutes a small
percentage of net operating revenue both nationally and regionally. As we have seen earlier, how-
ever, the eastern railroads will pay the highest percentage of net revenue for the retrofit program.
This partly reflects the fact that eastern roads as a group tend to earn less profit than roads in
other regions.
TABLE 7-13
ANNUAL RETROFIT COST AS A PERCENTAGE OF 1971 TOTAL
OPERATING REVENUE
Compliance
National
T'enod I Mu> ' Mir
2 years
5 years
0.42%
0.13%
0.33%
0.09%
East South
Mr* '
0.53%
0.16%
I mv ' M,*> ' Mu-
0.43%
0. 1 2%
0.31%
0.09%
0.25%
0.07%
West
' JVL; ^ .v:.
0.36%
0.11%
0.28%,
0.08%
'Net operating revenue is defined as transportation revenue minus variable transportation costs. Subtracting
rents, taxes, and interest payments from net operating revenue gives net operating income.or profit from
freight operations.
7-13
-------
60
50
40
z
o
30
t/i
a
o
I 20
10
Q TOTAL ANNUAL COST
DIRECT ANNUAfCOST
COST OF LOST LOCOMOTIVE DAYS
COMPLIANCE PERIOD (YEARS)
Figure 7-1. Cost of Retrofit Program as a Function of Compliance Period
7-14
-------
TABLE 7-14
ANNUAL RETROFIT COST AS A PERCENTAGE OF 1971 NET
OPERATING REVENUE
Compliance
Period
2 years
5 years
National
Max
1.96%
0.60%
Min
1.56%
0.44%
East
Max
2.48%
0.95%
Min
0.31%
0.70%
South
Max
1.22%
0.38%
Min
0.97%
0.27%
West
Max
1.58%
0.49%
Min
1.24%
0.36%
Bankrupt roads constitute a special subset for which financial and operating problems are
substantially different than for normal roads; these will be treated elsewhere.
In order to give a more detailed picture of the industry's ability to pay for a retrofit program,
program cost as a percent of net operating revenue has been computed for each Class I railroad
(including bankrupt roads but excluding those with negative net revenues). Figure 7-2 shows how
the railroads are distributed with respect to cost-to-net revenue ratio. The figure shows that the
impact of a 2-year program is much greater than that of a 5-year program.
The Impact on Marginal Railroads
The adverse effects of extra operating costs is greater on firms in financial distress than those
that are healthy. This is of concern in the case of the railroads, because a number of them face dif-
ficulties in maintaining profitable operations. It is important to estimate the number of railroads
that may have trouble paying the cost of a retrofit program even though the magnitudes of the
expenses involved in such a program are small relative to other expenses faced by the railroads.
(For example, a 30 percent increase in the price of diesel fuel would increase operating costs by
roughly $125 million.* This would represent from 2.5 to 12 times the annual cost of a muffler
retrofit program, depending on the compliance period allowed.)
This section attempts to gauge the extent of the problem posed in paying for a retrofit pro-
gram by determining how many railroads are in financial distress. This ib done by computing
for each road, several financial ratios that are generally accepted as> iodx.ahng the tmanuitl wmh
lion of a business enterprise. A summary' of the number of i<,:,<\-. with <>:,\-i/'/raMe •' -Hit* M r-i- f.
ratio is then given. This technique does not give a quantitative definition of which railroads cannot
afford a retrofit program. At best, it gives a rank-ordering. The cutoff value that determines
"financial distress" is arbitrary.
*This figure is computed by using as a baseline the total cost of fuel for all Class I railroads in 1971, which was
$417 million (ICC, op. cit.)
7-15
-------
ta
a
o
cc
cc
Ul
CO
S
25 -
20 -
15 -
10 -
5 -
5-YEAR PROGRAM
2-YEAR PROGRAM
1.0 1.6 2.6 2.5 3.0 3.5 4.0+
PERCENT OF NET OPERATING REVENUE
Figure 7-2. Distribution of Railroads by Retrofit Cost as a Percent of No Operating Krv-ma-
for 2- and 5-year Compliance Periods. (Maximum Total Cost Assumed. Bankrupt
Roads Included; Made with Negative Net Operating Revenue Excluded.)
7-16
-------
The following financial ratios were computed:
a. Current assets/total assets
b. Operating ratio (operating expenses/operating revenues)
c. Total liabilities less stockholders' equity/total assets
d. Income after fixed charges/total assets
e. Retained earnings/total assets
f. Net income/total assets
g. Net income/operating revenue
All bankrupt roads are excluded from this discussion, which is concerned only with roads
that have not been declared bankrupt but which may be in financial distress.
In most cases these ratios parallel those used by Edward Altman (1971). Ratios a and b are
measures of the liquidity* of a railroad, while b, d, f, and g are measures of profitability and effi-
ciency. Ratio c measures solvency.
With respect to ratio a, the analysis seems inconclusive. A large number of roads had ratios
of current to total assets in excess of three standard deviations from the mean. This indicates that
the distribution of values of this ratio did not approximate a normal distribution. This being the
case, ratio a does not constitute a valid indicator of which roads may be in distress.
The analysis of ratio e (retained earnings/total assets) indicated that 14 railroads have negative
retained earnings, while two have zero, showing that these roads lack liquidity. While internal
financing may not be important in the rail industry, the negative retained earnings indicates that
these roads are drawing down cash reserves.**
The most commonly used measure of profitability is operating ratio b, the ratio of operating
revenue to operating expenses. Three roads have operating ratios greater than 1, indicating that
expenses exceed revenue. An additional seven roads have operating ratios more than three standard
deviations higher than the mean. Certainly the three roads and possibly some of the seven must be
considered to be in an adverse position. Ratios/ and g are similar measures, in that a road with a
negative net income will have a negative ratio for both / and g. Six roads have negative net incomes.
In addition, two other roads must be considered to be poor performers as measured by the
ratio of net income to total assets (f).
Ratio d indicates that nine roads have negative income and two have zero income after fixed
charges. These roads are unprofitable by definition. The ratio of total liabilities (less stockholders'
equity) to total assets c appears to have also yielded inconclusive results. One road stands out as
being extremely poor by this measure, and there are four other roads for which this ratio is greater
than 1.
* Liquidity is the ability of a firm to convert assets into cash.
**This may also represent an insufficient amount of funds allocated to depreciation.
7-17
-------
A word of caution should be issued in the interpretation of any ratio that uses total assets.
Under the "betterment" accounting procedure, total assets tend to be inflated. However, to the
extent that this bias is uniform throughout the industry, it is possible to compare different roads.
It is not possible to compare these ratios with other firms outside the rail industry.
Table 7-15 summarizes the above findings with respect to the named ratios. As was mentioned
before, the table lists "worst performers" as indicated by each ratio, the cutoff point being rather
arbitrary. More significant is Table 7-16, which shows how many of the railroads contained in the
previous table appear under more than one ratio. Table 7-16 shows that 12 roads are in distress
with respect to three or more indicators; it can reasonably be presumed that these 12, at least,
could have difficulty in financing a retrofit program.
The Impact on Bankrupt Railroads
Of the 71 Class I line-haul railroads in the United States, seven are bankrupt: Boston and
Maine, Central Railroad of New Jersey, Erie Lackawanna, Lehigh Valley, Penn Central Transporta-
tion Co., the Reading Co., and Ann Arbor. These seven railroads operate about 20% of the loco-
motives owned by Class I railroads in the U.S. Not surprisingly, the total cost of retrofit for these
roads (see Table 7-17) is about 20% of the total cost for the entire muffler retrofit program.
These railroads will have difficulty financing the cost of a muffler retrofit program. There is
no question that the financial positions of these roads are bad. All six have negative net income,
and are currently meeting their deficits in part by drawing down cash reserves. Many of these
roads are currently receiving some form of subsidy, and all are in default on interest payments,
bonds, and/or taxes.
THE IMPACT ON USERS OF RAIL TRANSPORTATION
The effect of a muffler retrofit program may be felt by the railroads' users in either or both
of two ways. First, the possibility exists that the railroads may try to recover their retrofit
expenses through a rate increase. Second, the withdrawal of locomotives from service could
result in reduced hauling capacity and a consequent decline in the quality of service. Either
of these developments would tend to encourage some shippers to seek elsewhere for trans-
portation services. This section examines the possible magnitude of these effects.
The Effect On Railway Freight Rates
The ability of the rail industry to recapture the cost of a muffler retrofit program depends
on the characteristics of the market it faces. The establishment of Amtrak and the low volume
(and high price elasticity) of passenger service probably precludes the railroads from recovering
any of the retrofit costs through increases in passenger fares; rather, increased revenues would
be more likely to come from increasing freight rates.
7-18
-------
TABLE 7-15
NUMBER OF RAILROADS IN UNFAVORABLE FINANCIAL
POSITION RELATIVE TO EIGHT INDICATORS
(For Each Indicator, Railroads Listed in Order of
Increasingly Favorable Position)
Indicator
A. Current assets/total assets
B. Operating ratio
C. Total liabilities (less stockholders'
equity )/total assets
D. Income after fixed charges/
total assets
E. Retained earnings/total assets
F. Net income/total assets
G. Net income/operating revenue
Number of Roads in Unfavorable Position
Inconclusive
4 roads' greater than 1 (expenses > revenues)
4 roads' between 1 and .85
3 roads' greater than 1
2 roads' equal 1
2 roads' between .99 and .71
8 roads' negative
1 road's zero
13 roads' negative
1 road's zero
4 roads' negative
4 roads' ^ero
2 roads' positive but less than .011
4 roads' negative
2 roads' zero
2 roads' positive but less than .031
7-19
-------
TABLE 7-16
NUMBER OF RAILROADS DESIGNATED AS BEING IN FINANCIAL
DIFFICULTY BY ONE OR MORE FINANCIAL INDICATORS
Number of Financial Indicators,
N, in Table 7-IS
Number of Railroads Appearing
under N Indicators in Table 7-15
2
3
4
5
7
2
6
3
TABLE 7-17
NET COST OF MUFFLER RETROFIT PROGRAM FOR THE
SEVEN BANKRUPT CALSS I RAILROADS
Length of
Program
Annual Cost
Max
Min
Total Cost
Max
Min
2 Years
5 years
$10,569,000
3,197,000
$8,393,000
2,326,000
$21,139,000
15,984,000
$16,786,000
11,631,000
7-20
-------
Freight rate increases must be approved by the Interstate Commerce Commission. Inquiries
to the JCC indicate that the Commission places no a priori limits on the magnitude of rate increases
that may be requested. It is entirely the railroad industry's prerogative to decide if requests for rate
increases are to be submitted to cover the costs shown in Table 7-12. Any cost factor could form a
legitimate basis for increasing rates to recover costs. Furthermore, the Commission is considering
environmental aspects in its rate determination. As a result of litigation involving the environmental
effects of various rate structures, the ICC has prepared several Environmental Impact Statements
showing their concern.*
In summary, there are strong indications that the rate increases that could be requested by
railroad companies to defray the costs of noise reduction would fall within the practice of the ICC.
No a priori bias would be applied by ICC agents, and they could be expected to act with a positive
attitude toward the objective of improving the quality of the environment.
To place the level of expenditure and possible freight rate increase in perspective, previous
cost increases and subsequent rate increases may be used for reference. In the ICC report served
4 October 1972, in Ex Parte 281, a rate increase for railroad freight was authorized. The railroads
claimed in their rate request that expenses had increased $1.312 billion from January 1971 to April
1972. The authorized rate increases were
National Average 3.44%**
East 3.60%
South 3.10%
West 3.44%
These increases, if fully applied, would have increased revenue by $426 million; however, the most
usual case is that they are not fully applied. The industry estimates that only 85% or $349 million
will actually be realized.***
Since the rate increase of September 10, 1972, costs have risen by $930 million. About 80%
of this rise has stemmed from wage increases and increased payroll taxes. In light of these higher
costs, in April of 1973 the railroads applied for a 5% rate increase. The maximum cost of the 2-
year muffler retrofit program is about $51 million, which is only 5.5% of the $930 million cost
increase that led to the request for a 5% rate increase. The rail industry claims that if the entire
$930 million cost increase is to be recovered, it will require a 7.5% increase in rates.****
*See ICC Docket, Ex Parte 281 and Ex Parte 344F, Supplement 927.
**The national average was calculated by using regional data.
***These figures come from estimates made by the rail industry. They assume that the elasticity
of demand is zero—an unlikely situation. The question of elasticity is considered later in this
section.
****Again, this estimate assumes that the elasticity of demand for rail service is zero.
7-21
-------
The amount of the recoverable costs and the attendant freight rate increase necessary will
depend on the elasticity of demand for rail freight service.* The annual (maximum) retrofit costs
for the 2-year program represent about 0.4% of 1971 freight revenue, while the 5-year (minimum)
program represents only about 0.1% of freight revenue (see Table 7-13).
Data from Friedlander (1969, p.73) for 1961 have been used to calculate an overall rail freight
demand elasticity of-0.7. Using this elasticity, we can estimate the increase in freight rates neces-
sary to offset the increased costs. The freight increases are shown in Table 7-18. Also shown in the
percent these increases would represent of the 1971 average rate per ton mile, which was $.01 594.
TABLE 7-18
RATE INCREASE THAT WOULD ENABLE RAILROADS
TO RECOVER RETROFIT EXPENSES
2-year
max
min
5-year
max
min
Rate Increase
(Cents per Ton Mile)
.0232
.0184
.0076
.0057
Percent of 1971
Average Freight Rate
1.46%
1.15
0.48
0.36
These rate increases must be interpreted carefully. They were calculated by using demand
elasticities derived from 1961 data; since then a number of changes have taken place that would
probably increase the elasticity of demand for rail service. First, the near-completion of the inter-
state highway system has improved the service rendered by trucks and has reduced operating costs.
Second, the rise in interest rates has made the cost of holding inventories higher and might have
made shippers more sensitive to other service characteristics, causing a downward shift in the de-
mand curve and potentially increasing its elasticity. Third, shifts among the various commodity
classes of freight might have resulted in an increase in the elasticity. For example, if the price elas-
ticity of demand for rail service is higher for mineral ores than for manufactured products and if
the share of mineral ores has increased relative to manufactured products, then the overall elasticity
would have increased.
* Elasticity of demand is the ratio of the percent rise in quantity demanded to the percent rise in
price. An elasticity coefficient of-.l, therefore, indicates that a 10% price increase would result
in a \% decrease in demand.
7-22
-------
We have attempted to make some estimates of the new elasticity, taking into account the shift
in the distribution of commodities. The results should be interpreted only as tentative. We have
used the 1961 elasticities for each commodity group but have weighted them by the 1971 commod-
ity distribution.
Data from Friedlander (op. cit., p. 73) have been used to obtain the following elasticities for
the five major commodity groups:
Commodity
Agriculture
Animal products
Products of forests
Products of mines
Manufacturing and other
Elasticity
0.5
0.6
0.9
1.2
0.7
These figures represent the pre-1964 commodity classifications used by the ICC. In order to deter-
mine the current elasticity of demand, we used these commodity group elasticities and weighted
them by the current distribution of freight within these groups. These weighting factors are as
follows:
Commodity
Weight
Agriculture
Animal products
Products of forests
Products of mines
Manufacturing and other
.097
.0002
.144
.420
.387
To determine the distribution, it was necessary to take the current freight classifications and assign
them to one of these categories.
The overall elasticity was calculated to be -0.953, significantly more than the esti-
mate of -0.7 obtained from Friedlander's data. Even more interesting is the distribution of elastic-
ities by district. To arrive at these estimates, it was necessary to assume that the rate per ton mile
for each of the 1971 commodity classifications was equal for each of the three districts. Although
this is not the case, we believe the errors to be quite small. The estimated elasticities are:
East -0.99
South -0.95
West -0.83
These figures indicate that the eastern roads, which are in financial difficulty, would have the most
trouble recovering the cost of a retrofit program. The western roads, which as a group are the most
profitable, would recover the cost of a retrofit program most easily.
7-23
-------
Given the energy crisis, however, even this tentative analysis may not be valid. As discussed
earlier railroads use less energy per ton mile of freight moved than trucks, pipelines or airlines.
As a result, railroads would be impacted less than these other competitive modes by increases in
fuel costs.
It is not possible to predict accurately at this point, the effect of any rate increases the ICC
might grant to the railroads to recover the costs of a retrofit program. The possible effects of
increased rates on demands for rail service are directly related to the energy situation. If compe-
titive modes of transportation (i.e., trucks, pipelines, and airlines) are more severely impacted by
increased fuel rates, the fact that railroads increased their rates to cover the costs of a retrofit
program might well be insignificant.
The Effect on Quality of Service
It has been shown above (see Introduction) that, in order to accomplish a retrofit program
within a compliance period of 5 years or less, some locomotives would likely have to be withdrawn
from service in addition to those undergoing maintenance by the usual schedules. The number of
locomotive days taken up in this manner is given in Table 7-19, in absolute numbers and as a per-
centage of locomotive days available. If, under normal conditions, the railroads are operating at or
near full capacity, then the figures shown in the table represent the upper bound of lost freight-
hauling capability.
TABLE 7-19
ANNUAL LOCOMOTIVE DAYS TAKEN UP BY RETROFIT PROGRAM
Compliance
Period
2-year-
\
5-year
Locomotive
Days
Absolute
% of Total
Available
Absolute
% of Total
Available
Region
National
17,048
.194%
2,044
.023%
East
9,252
.225%
1,129
.027%
South
2,143
.197%
203
.0187%
West
6,378
.174%
712
.0195%
The impact of decreased hauling capability on the various commodities shipped by rail depends
on how the railroads react to the capacity decrease. There are two ways in which demand for rail
service can be made to equal the available supply: non-price rationing or price rationing.
7-24
-------
In the case of non-price rationing, the railroads could simply allow service to decline in quality
while maintaining the same rates. The resulting delays and uncertainties in the transportation net-
work would have differential impacts on the various commodities being shipped; those items highly
sensitive to the quality of service will tend to be diverted to other modes of transportation. Com-
modities in this category are high-valued products, for which transportation charges are a small
fraction of total value, and perishables.
Price rationing involves raising the price of service (with the approval of the ICC) in order to
decrease demand to the level of the new, reduced capacity. Such a policy would affect commodities
sensitive to freight rates; examples of these would be mineral ores and semifinished products. Such
goods would tend to be shipped by other modes, or the quantity shipped would be reduced.
The probable magnitude of the effect of price rationing can be estimated. Table 7-19 shows
that, in the worst case, capacity would decline by about .2% nationally. Assuming (from p. 7-22)
that the elasticity of demand for rail transportation is about -.7 gives a price rise of .28% necessary
to effect the required reduction in demand. This amounts to an average increase of .004 cents per
ton mile relative to the 1971 average freight rate. This increase is fairly small, so minimal changes in
transportation patterns may be expected as a result of the retrofit program.
SUMMARY AND CONCLUSIONS
Impact on the Railroad Industry
Cost. The cost of a muffler retrofit program is highly sensitive to the compliance period
allowed. Maximum total cost for a 2-year program is estimated to be $103 million. Allowing 5
years for compliance would reduce the total cost to approximately $79 million.
Change in net revenues. The impact of a 2-year program would be to reduce overall Class I
railroad annual net operating revenues by about 2%.
Effect on prices. For the railroads to recover the expense of a retrofit program would require
an average freight rate increase of approximately .023 cents per ton mile in the 2-year case and
.008 cents per ton mile in the 5-year case. These figures represent, respectively, 1.46% and
.48% of the 1971 average freight rate.
Effect on capacity. A 2-year retrofit program would result in an annual loss of as many as
17,000 locomotive days, or about .2% of the total available, for the duration of the program. This
would drop to about .02% for a 5-year program.
Impact on marginal railroads. Approximately a dozen railroads are in financial difficulties, as
indicated by the computed values of a number of standard financial ratios. These roads may have
difficulty in raising the funds necessary to pay for a retrofit program.
Impact on bankrupt railroads. Six roads are presently bankrupt, and may not be able to
finance a retrofit program without an external source of funds. The total program cost for these
roads would be $21 million for a 2-year program and $16 million for a 5-year program.
7-25
-------
Impact on Users of Rail Services
Prices. Increases in freight rates would tend to encourage some shippers to seek alternate
modes of transportation. This would occur primarily among shippers of commodities whose price
is sensitive to transportation cost, such as semifinished products. It is not likely, however, that the
small rate increases foreseen by this study would cause any major hardships or dislocations.
The energy crisis may make any railroad rate increases insignificant compared with compe-
titive modes of transportation, which would be more severely impacted by rising fuel costs.
Quality of service. A decrease in the haulage capacity of the railroads may result in the diver-
sion of some freight to other modes of transport. Which commodities would be affected depends
on how the railroad decided to reduce demand to the level of supply. If rates were raised, the
effect would be the same as discussed in the previous paragraph. If rates remained constant but
shipping delays were allowed to develop, commodities sensitive to transit time (such as perishables)
would be most affected. Such diversions, however, will tend to be localized and on a small scale
in view of the small reductions in capacity anticipated.
7-26
-------
SECTION 8
ENVIRONMENTAL EFFECTS OF PROPOSED REGULATIONS
INTRODUCTION
The proposed regulations will immediately stop the noise emitted by railroad trains from in-
creasing and over a 4-year period will progressively reduce the noise presently emitted by railroad
locomotives. As a result, the number of people currently subjected to annoying levels of railroad
noise will be reduced.
IMPACT RELATED TO ACOUSTICAL ENVIRONMENT
Several studies have been conducted to estimate the reduction in noise levels, and the number
of people who will potentially benefit as a result of the noise control standards proposed.
Case Studies of Railroad Lines
Ten cities with widely varying populations were selected to make detailed comparisons of
train traffic with population densities near railroad tracks and with the type of land use adjacent to
tracks (see Table 8-1). Such comparisons provide a basis for determining how many people are
exposed to railroad noise, how often they are exposed, and what activities they are engaged in at
the time.
The schedules of trains moving over the railroad lines were determined from The Official Guide
of the Railways, July 1973, or from employee timetables. Estimates of speed maxima and minima
were taken from employee timetables or obtained from railroad employees. Speeds for AMTRACK
trains were not obtained. The period between 10:00 p.m. and 7:00 a.m. was designated as "night,"
and the rest of each 24-hour period was designated as "day," Table 8-2 summarizes the results of
the ten case studies.
Analysis of Train Noise Impact
There are three major noise sources that contribute to LQJ^J (see Enclosure A for definition of
Lrjj^j) at points along and away from railroad tracks: locomotives, wheel/rail interaction, and horns
or whistles.
8-1
-------
TABLE 8-1
LAND US1: NEAR RAILROAD LINtS
City and State
Newton, Mass.
Boston, Mass.
Valparaiso, Ind.
St. Joseph, Mo.
Akron, Ohio
Somerville, Mass.
Michigan City, Ind.
Kalamazoo, Mich.
Altoona, Pa.
Ft. Lauderdale, Fla.
Lewiston, Maine
Denver, Colo.
Cheyenne, Wyo.
Cambridge, Mass.
Macon, Ga.
Average
i
Land Use Within 500 Ft of Track
(Percent)
Residential
75
59
43
42
40
30
29
22
16
1 >
12
12
9
8
6
28
Business
21
9
8
13
23
18
15
; 5
18
22
19
3
11
24
4
14
Industrial &
Other
4
32
49
45
37
51
56
73
65
66
68
85
79
68
90
Mileage
Studied
6
7
9
26
25
7
17
20
6
21
11
51
15
9
25_
58 Total 255
8-2
-------
00
J
Q
•J
<:
u
£
H
Ci
<
tti
Z
00
CM U
oo p
w £2
U
a
z
<
o a
5 (/)
3 *
^5
S 3
u. O
O fs
2 £
uj
S ._,
^
s"
1 1
3 8
s £
-I
oe g.
NUMI
PASSENG
Ii
il
si
at £
u j_
gs o
ui
ol
u.
§
P
1
1
Q
2
2
H
S
X
S
a:
w
b
I
ffl
_J
p
z
u
O
£
J3
0,
Q
LU
(A
|
>
£
I
Q
U
S
VI
1
*
>•
o
z
2
UJ
•«•
CITY & ST
0 __(/__
'oav^Dr~r^rtjoOrJrN
r-u"iri(^--vi O'tfON
r^i ^c i~o r- oo m i^-
0;OON^^ 2^ ^JC^
O ( r- ^ O ,
O^-lOOOOoOOO
O *"-< O rj o -t O "" I O O
1
^0?, 00^000
^Ci^ioor — o — r I ~ O
r| l- O r — rf — V, l~~ O",
M'I 'A, — *t — — tt ON \O O
•o
S. ' 1 » •§
o SJ^—CUC1^ —
^ °- 3 „• ^ ,° Z C S d
° «' - ^ S • c c c I
c'o§SgS™Soj;
l-lll=flif
8-3
-------
Figure 8-1 shows sonic LQ^ profiles that were calculated by applying the prediction techniques
to actual operations on a specific railroad line. The profiles shown in Figure 8-1 were calculated
from the following data supplied by I'enn Central:
10:00 p.m :md 7:00 a.m.
6 freight trains
each 14 loaded cars and 10 empty cars
40 mph
and
7:00 a.m. and 10:00 p.m.
36 passenger trains, each
40 mph
^assenger trains with eight cars correspond to the national average passenger loading of cars (Moody,
)971). The curve for two i -s is displayed in order to demonstrate the influence of the number of
cars on the results.
Since there are no crossings along the branch picked for this study, no whistle noise was con-
sidered. In addition to the usual geometric attenuation, atmospheric absorption and ground surface
attenuation (Beranek, 1971) were included in the calculation for Figure 8-1 (See enclosure B to
this Section.).
Figure 8-2 shows Lrjjsj profiles that were calculated for the average of all the train movements
in the U.S. The profiles were calculated from the following data (Moody, 1971);
Urban Areas
4 freight trains by day, 2 by night, each 33 mph, 40 cars 3800 tons
2 passenger trains by day, 0 by night, each 36 mph, 6 cars
Nonurban Areas
3 freights by day, 2 by night, each 33 mph. 40 cars, 3800 tons
0 passenger trains
Figures 8-3 through 8-6 provided examples of the impact on the community of a program to
equip locomotive exhausts with mufflers. Figure 8-3 shows that a muffler that provides 10 dB(A)
of quieting will nearly halve the distance to which people are exposed to LQN of 55 or more by
train traffic on the Dorchester Branch of Penn Central (assuming that no other sources of locomo-
tive noise produce levels comparable to exhaust noise levels). Figure 8-4 shows that there is a reduc-
tion of 24,000 people exposed to LTJN of 55 or more by train traffic on the 7.2-mile-long Dorches-
ter Branch. Figure 8-5 is based on national average train traffic and also shows that a muffler that
quiets locomotive exhaust noise by 10 dB(A) will more than halve the distance to which people are
8-4
-------
UJ
ui
U.
o
cr
L.
m
o
O
oo
o
to
o
in
O
10
CL>
U
c
— 0-
V" <*-
" X
CC c
2
CQ
C/5
-------
UJ
u
u.
o
Q:
o
tr
u.
UJ
o
2
H
g
'«
H
0>
00
KI
u
(U
Cfl
C
oo
-------
CO
_l
UJ
UJ
Q
IU
s
o
o
z
o
UJ
CO
20
19
18
17
16
15
14
. 13
j
' 12
11
10
_
Q
x z 9
O
<
cr
O
QC
LJ
O
I
CO
o
10 dB
LOCOMOTIVE
QUIETING
5 dB LOCOMOTIVE
QUIETING
PRESENT
EQUIPMENT
40 50 60 70 80
LON (EXCLUSIVE OF WHISTLES AND HORNS)
Figure 8-3. Distance From Track at Which Various
the Penn Central
Occur Around the Dorchester Branch of
8-7
-------
80
2
O
O
UJ
<
O
O
2
O
2
O
UJ
X
UJ
to
_J
UJ
O
UJ
UJ
UJ
_>
a.
s
to
o
2
O
I
70
60
„
50
40
30
20
10
10 dB LOCOMOTIVE
QUIETING
5 dB LOCOMOTIVE
QUIETING
PRESENT EQUIPMENT
40 50 60 70 80
LDN (EXCLUSIVE OF WHISTLES AND HORNS)
Figure 8-4. Thousands of People Exposed to Various LJ^XJ by 7.2 Miles of Track on the Dorchester
Branch of the Penn Central
8-8
-------
W)
u
>
UJ
_l
o
UJ
20
19
18
17
16
15
14
O
O
S~ 13
guj
B£ 12
UJ u-
UJ
11
to
QUJ 10
-J CC
o
5i 9
5
5
o
o:
o
cr
UJ
o
I
10 dB LOCOMOTIVE
QUIETING
5 dB LOCOMOTIVE"
QUIETING
PRESENT
EQUIPMENT
40 50 60 70 80
LDN (EXCLUSIVE OF WHISTLES AND HORNS)
Figure 8-5. Distance From Track at Which Various LDN Occur Due to National Average Train
Traffic i
8-9
-------
exposed to Lj^ of 55 or more (assuming that no other sources of locomoti/e noise produce levels
comparable to exhaust noise levels;. Figure 8-6 shows that there is a corresponding 5.1 million re-
duction in the number of people exposed to LJ^M of 55 or more based on national average train
traffic.
Population densities used to construct Figures 8-3 and 8-6 were obtained from the U.S.
. partr'ent c f , i. "KTCC, Bureau ui ;lic ( , ; is. 1 he census results show 28,098 people living
within 1000 leei of ihe 7.2 miles of track comprising the Dorchester Branch of Penn Central. The
population density in the first 500 feet next to the 1m - \vas taken to be one-half of the density for
the entire region, in keeping with national trends.
Die figures for the number of people exposed to noise from national average train traffic were
based on estimates of 30,000 miles of railroad rights-of-way in urban areas in the U.S. Urban areas
are defined as the 40 Standard Metropolitan Statistical Areas (SMSAs) having average population
densitites in excess of 500 people per square mile and a total population greater than 250,000. The
40 SMSAs defined above have a total land area of 58,200 square miles and a total population of
71,082,000, for an average population density of 1220 people per square mile. This figure must be
modified, however, as there tends to be a concentration of industrial, commercial, and other non-
residential activities in the vicinity of rail lines. Land use and zoning maps indicate that the residen-
tial population density in the vicinity of a railroad line tends to be about 50% of the average density
for the entire region.
IMPACT RELATED TO LAND
These regulations will have no adverse effects relative to land.
IMPACT RELATED TO WATER
These regulations will have no effect on water quality or supply.
IMPACT RELATED TO AIR
The use of more efficient exhaust muffling systems can cause a change in the back pressure to
the engine and may result in a change in the exhaust emissions level. The data, at present, are insuf-
ficient to make other than a general statement concerning the directions the various emission levels
take when a different back pressure is applied, since the behavior of the various engines and exhaust
emission control systems vary widely. However, internal combustion engine exhaust emissions are
affected by changes in exhaust system back pressure, as evidenced by the tests of gasoline engines
at the University of Michigan (Bolt, Bergin, Verper, 1973), and they must be considered. It is
important to note, however, that motor carrier exhaust emissions are approximately 3.7 times
higher than rail carrier exhaust emissions per ton mile of goods transported (Battelle Laboratories,
1971), indicating that in the overall balance rail carriers are already more efficient than motor
carriers, from an exhaust emission standpoint.
8-10
-------
11
co
_j
LU
>
8
UJ
X O
UJZ
uio
—J 5"
O. ±
Q.UJ
ULX
ou
CO
z
O
r
T
TO dB LOCOMOTIVE
QUIETING
5 dB LOCOMOTIVE
QUIETING
PRESENT EQUIPMENT
40 50 60 70 80
LDN(EXCLUSIVE OF WHISTLES AND HORNS)
Figure 8-6. Millions of People Exposed to Various Lp^ by National Average Train Traffic
8-11
[....
-------
It must also be noted that promulgating stricter rail carrier noise regulations at this time may
inadvertently divert cargo traffic from the rails toward motor carriers due to difficulties in com-
pliance with regulations, thereby causing an increase in total exhaust emissions to the atmosphere,
as well as increasing noise emissions. Based on the analysis presented, problems such as this are not
expected to arise as a result of the proposed regulations.
ENCLOSURE A: "DAY NIGHT EQUIVALENT NOISE LEVEL" (LDN)
Lrjjsj is a modified energy-equivalent sound level. The energy-equivalent sound level LgQ is
the level of the continuous sound associated with an amount of energy equal to the sum of the
energies of a collection of discontinuous sounds. Lpn is defined by
LEQ= 10* -i- T'2 ,0NL/.Odt
'2 ,,ti
where NL is the instantaneous overall noise level in dB(A) at time t, and the time period of interest
is from time t j to time t->. LQN is determined precisely like Lpn, except that all noise levels NL
measured at night (between 10:00 p.m. and 7:00 a.m.) are increased by 10 dB(A) before being
entered into the above equation.
ENCLOSURE B: EXCESS ATTENUATION OF RAILROAD NOISE
Many mechanisms cause attentuation of sound beyond that caused by geometric spreading,
including molecular absorption in me air, precipitation, barriers, ground cover, wind, and temper-
ature and humidity gradients. The attenuation varies with location, time of day, and season of the
year. To account for the attenuation produced by these highly variable sources, it is necessary to
compile d .ailed records of wind, temperature, humidity, precipitation, and even cloud cover on a
statistical or probabilistic basis. The following discussion is directed at a base case that includes
two sources of excess attenuation that can be relied upon: atmospheric molecular absorption and
attenuation associated with variations in the physical characteristics of the atmosphere near the
ground. Both attenuations vary with frequency. The attenuation factors were evaluated for
reference conditions of 50°F and 50% relative humidity.
Figure 8-7 shows how atmospheric molecular absorption and variations of atmospheric char-
acteristics near the ground change the shape of the locomotive noise spectrum. The high frequen-
cies become less important as the sound travels outward from the source. The attenuation of the
overall sound level (logarithmically summed octave-band sound levels) was found to be about 2dB
per thousand ft out to 4000 ft. That value was used to calculate the propagation oflocomotive
noise described in this report. The value for the effective overall attenuation coefficient for loco-
motive noise is about the same for throttle position 8 and throttle position 1.
Figure 8-8 shows how the frequency-dependent attenuations change the shape of the spectrum
of wheel/ rail noise. Notice that here, too, the high frequencies become less important as the sound
8-12
-------
90
77 80
o
JD
5L
(VJ
o
O
9 70
O
o>
t_
CD
60
X
S2
UJ
£
I
50
UJ
>
UJ
ID
O
CO
CD
i
UJ
t- 40
O
O
O
UJ
30
20
Mill!
50'
1000'
4000'
16 63 250 1000 4000
32 125 500 2000 8000
OCTAVE BAND CENTER FREQUENCY (Hz)
Figure 8-7. Influence of Frequency-Dependent Attenuations on Locomotive Noise Spectrum
(See Figure A.I.7h for Comparison)
8-13
-------
80
cc
<
CD
:t
cvj
O
o
O
c>
<
CD
_J
LJ
UJ
_J
O
2
O
V)
o
z
CD
UJ
>
I-
o
o
70
60
40
30
20
10
1 I I i i i I I
25'
1000'
3000'
16 63 250 1000 4000
32 125 500 200 8000
OCTAVE BAND CENTER FREQUENCY (Hz
Figure 8-8. Influence of Frequency-Dei -mien! Attenuations on Wheel/Rail Noise, Train No. 6,
Region 2 (See Figure B. 1.13 lor Comparison)
8-14
-------
travels outward from the source. The attenuation of the overall sound level (logarithmically
summed octave-band sound levels) was about 3 dB per thousand ft out to 3000 ft. That value
was used to calculate the propagation of locomotive noise described in this Background Document.
8-15
-------
SECTION 9
SELECTION OF THE PROPOSED REGULATIONS
PROBLEM ADDRESSED AND APPROACH
Problem Addressed
The problem addressed in the proposed noise emission regulations is the development of noise
emission regulations that will control railroad noise and Federally preempt conflicting State and
local noise emission regulations, taking into consideration that (1) State and local governments have
the primary responsibility to protect the environment from noise and (2) Federal special local
conditions authorizations may be authorized in the case of use or operational regulations if the
State and local regulation in question is not in conflict with the noise emission regulations estab-
lished under Section 17.
Approach
In order to develop these noise emission regulations, the following approach, based on the
statutory requirements of the Noise Control Act of 1972, was utilized:
1. Determination of the sources of railroad noise to be Federally regulated
2. Determination of the best available technology to achieve nohe reduction
3. Determination of the cost of compliance to the railroad industry with possible noise
emission regulations
4. Determination of the environmental and economic impact of possible noise emission
regulations
5. Selection of the appropriate noise emission standards.
REGULATORY APPROACHES CONSIDERED
"Status Quo" Regulations Alternative
Status quo regulations for both locomotives and railroad car noise could be proposed that
would preempt State and local regulations. These status quo regulations would not reduce noise
but rather limit it to present levels and would have no financial impact on the railroads beyond
standard maintenance already required. The function of status quo regulations is, therefore, one
in which the intent of the Federal government to revise the status quo regulations is an implicit
9-1
-------
statement that such future revision will result in reduction in noise levels with probable concurrent
financial impact on the railroad industry. Thus, a status quo regulation placed on certain equip-
ment and facilities would establish the direction and intent of Federal regulation on those sources
in the future. The rationale for the issuance of status quo regulations would be that the financial
mpact of more stringent regulations at tlii^ ' - were not available, status quo regulations
could be estrbiishei'' to , ' JCB a ceiling on noise emissions and allow tune for further technology
development.
Future Noise Standards Regulations Alternative
The data gathered by EPA indicate that it is feasible to reduce railroad noise with presently
available technology at a reasonable cost. However, the shortest feasible time to apply this tech-
nology on a retrofit basis at a reasonable cost is 4 years. Thus, a regulation requiring the applica-
tion of this technology could be promulgated with an effective date 4 years in the future.
Section 17 provides for Federal preemption of State and local regulations upon the effective
date of the Federal standards. Therefore, during the 4-year period required for the application of
technology, State and local regulations could be established and enforced.
Noise Reduction in Combination with Status Quo Regulations Alternative
As pointed out in the previous alternative, if a regulation were promulgated with an effective
date some time in the future, State and local regulations would not be preempted until this date.
However, it is not feasible for a noise reduction regulation on trains to be effective in less than 4
years when based on available technology and cost. It, therefore, would appear unreasonable to
expect quieting of trains during this period. However, it is not unreasonable to expect that equip-
ment be maintained properly to eliminate unnecessary noise. To accomplish this, a status quo
regulation based on proper maintenance practice could be made effective earlier. This would not
have substantial economic impact, nor would it produce significant noise reduction. It would, how-
ever, ensure that noise will not increase during the period prior to the installation of noise abate-
ment equipment. Further, it would preclude the State and local governments from establishing
what might be unreasonable equipment standards during this interim period.
REGULATORY APPROACH SELECTED BY EPA
The Environmental Protection Agency has chosen to adopt the last alternative discussed. It
is believed that this approach is the most environmentally sound alternative and one that fulfills
the requirements of Section 17.
9-2
-------
DISCUSSION OF PROPOSED REGULATIONS
The proposed noise emission regulations will establish standards for noise emissions from
locomotives and railroad cars engaged in interstate commerce by railroad. The proposed standards
specify sound levels measured at a distance of 30 meters (100 feet) from the centerline of the rail-
road track Measurements will be made in decibels on the A-weighted scale, using the fast meter
response. The general measurement procedure used to obtain the data upon which the standards
are based is presented in more detail in Section 6.
All locomotives to which the proposed regulation is applicable are to meet the following noise
emission standards for the locomotive at rest and in motion:
v
Locomotive at Rest
Effective 270 days after promulgation of the regulations, under stationary test, 93 dB(A) at
any throttle setting and 73 dB(A) at idle, when measured over any surface.
Effective 4 years after promulgation of the regulations 87 dB(A) at any throttle setting and
67 dB(A) at idle, when measured over any surface.
Locomotive in Motion
Effective 270 days after the promulgation of the regulations, 96 dB(A) at any operating condi-
tion, when measured over any surface.
Effective 4 years after the promulgation of these regulations, 90 dB(A), at any operating condi-
tion, when measured over any surface.
Rail Car
Effective 270 days after promulgation of these regulations, all railroad cars or combination of
railroad cars operated by surface carriers engaged in interstate commerce by railroad are to meet a
noise emission standard of 88 dB(A) at speeds up to and including 72 km/hr (45 mph) and 93 dB(A)
at speeds greater than 72 km/hr (45 mph) when measured over any surface.
Based upon the strict language of the Noise Control Act of 1972, its legislative history, and
other relevant data, "best available technology" and "cost of compliance" have been defined as
follows:
"Best available technology" is that noise abatement technology available for application to
equipment and facilities of surface carriers engaged in interstate commerce by railroad which pro-
duces meaningful reduction in the noise produced by such equipment and facilities. "Available" is
further defined to include:
1. Technology which has been demonstrated and is currently known to be feasible
2. Technology for which there will be a production capacity to produce the estimated number
of parts required in reasonable time to allow for distribution and installation prior to the effective
date of the regulation.
3. Technology that is compatible with all safety regulations and takes into account operational
considerations, including maintenance, and other pollution control equipment.
9-3
-------
"Cost of compliance" is the cost of identifying what action must be taken to meet the specified
noise emission level, the cost of taking that action, and any additional cost of operation and maint-
enance caused by that action.
Currently existing technology known to reduce locomotive noise consists of (a) fan modifica-
(b) engine casing modification, and (c) muffler retrofit. Applications of fan modification and
casing modifvation were not included in establishing the noise emission levels in the pro-
posed regulations bccau, , of lack of equipment availability, prohibitive and limited cost data, and
low relative effectiveness in noise reduction. Muffler retrofit to the locomotive engine exhaust
system was determined to be the only method that meets ihe criteria established above for "best"
Currently existing technology known to reduce railroad car noise consists of (a) replacement
o, the bolted rail with the welded rail, (b) structural maintenance to railroad car bodies, and
(c) elimination of flat spots on wheels. The proposed noise emission regulation did not include
replacement of the bolted rail with [he welded rail and structural maintenance to railroad car
bodies because of prohibitive cost and lack of data. Elimination of flat spots on wheels and irregu-
larities on rails can be achieved through effective normal maintenance, without added cost for
compliance.
Conclusion. The only standards that can be adequately based on "best available technology"
and "cost of compliance" at this time are (1) the muffler retrofit to control locomotive exhaust
and (2) effective railroad car maintenance. The proposed regulations, therefore, require locomotives
to eventually meet a noise emission standard that results in significant reduction in noise which can
be achieved through the installation of exhaust mufflers. The proposed railroad car noise emission
standard is designed to ensure that railroad cars will be properly maintained so that train noise
levels will be as low as the available technology permits.
9-4
-------
REFERENCES
Anon., "Retarders Are Key to Yards," Railway System Controls, June 1973.
Altman, Edward I. (1971), "Railroad Bankruptcy Propensity," Journal of Finance, Vol. XXVI,
pp. 333-346.
American National Standard Specification for Sound Level Meters,SI .4-1971.
Bietry, M.(1973), "Annoyance Caused by Railroad Traffic Noise," Proceedings of a Congress on
Traffic Noise, Grenoble, France, Jan. 9, 1973.
DOT (1970), "A Study of the Magnitude of Transportation Noise Generation and Potential Abate-
ment, Vol. V, Train System Noise," U.S. Department of Transportation Report No.
OST-ONA-71-1.
DOT (1971), "Noise and Vibration Characteristics of High-Speed Transit Vehicles," U.S. Depart-
ment of Transportation Report No. OST-ONA-71-7.
Embleton, T. F. W. and G. J. Thiessen (1962),"Train Noises and Use of Adjacent Land," Sound,
I: l,pp. 10-16.
EPA Docket 7201001.
Friedlaender, Anne (1969), The Dilemma of Freight Transportation Regulations, Brookings Insti-
tution, Washington, D.C.
Kendall, Hugh C.(1971), "Noise Studied in Retarder Yards," Railway Systems Controls, July 1971,
pp. 9-13.
Kurze, U. and L. L. Beranek, "Sound Propagation Outdoors" Noise a~id Vibration Control, edited
by L. L. Beranek, McGraw-Hill, 1971.
Kurze, U. J.,E. E. Ungar, and R. D. Strunk (1971), "An Investigation of Potential Measures for the
Control of Car Retarder Screech Noise," BBN Report No. 2143.
Moody's Transportation Manual (1971).
National Railway Publication Company (July 1973), The Official Guide to the Railways.
Rand McNally & Co. (1971), Commercial Atlas and Marketing Guide.
Railway System Controls (1972), "BN Studies Retarder Noise Abatement," Railway System
Controls, November 1972, pp. 14-20.
Rickley, E. J., R. W. Quinn, and N. R. Sussan (1973), "Wayside Noise and Vibration Signatures of
High Speed Trains in the Northeast Corridor," Department of Transportation Report No.
DOT-TSC-OST-73-18.
Rapin, J. M. (1972), "Noise in the Vicinity of Railroad Lines. How to Characterize and Predict It,"
Centre Scientifique et Technique du Batiment, Cahiers, Building Research Establishment,
Garston, Watford, WD2 75R.
R-l
-------
Ratering, Edwin G.. "The Application of Vehicle Noise Test Results in the Regulatory Process,"
Conference on Motor Vehicle Noise, General Motors, April 3-4, 1973.
Rathe, E. J. (1968), "Effect of Barriers on the Noise of Railroad Trains," Eidgenossiche Material
PruTu«gs--und Versuchsanslatt fur Industrie, EMPA No. 38 155/2, Biihendorf (in German).
• igham, K F. and R. L. St;r-n to Environmental
or i'< y OHice of No>' '.: • nieiii ar. ' Control, San Francisco, Calif., September
Sch'.iltz, T. S. .1 r>72), "Some Sources of Error in C< ' '"'inity Noise Measurement," Sow a
Vibration, 6:, 2, p, . 18-27.
Schultz, T. S. (1971), "Technical Background for Noise Abatement in HUD's Operating Programs,"
Bolt Beranek and Newman Inc., Report No. 2005R.
I'ngar, E. F,., R. D. Strunk. and P. R. Nayak (1970), "An Investigation ' the Generation of Screech
by Railway Car Retard r,,' BBN Report No. 2067.
U. S. Bureau 01 ( ensus, Cc. .. . f Housing ( 1 970), Block Statistics, I 'inal Report HC (3).
U. S. Bureau of Census, U. S. Census of Population (1970), Number of Inhabitants, Final Report
PC(1) ~A1, United States Summary.
Wilson, G. F. (1971), "Community Noise from Rapid Transit Syste-.i^." in Noise and Vibration
Control Engineering, Proceedings of the Purdue Noise Control Conference, July 14-16, 1971 ,
p. 46, at Purdue University, Lafayette, Ind., edited by Malcolm J. Crocker.
Wyle Laboratories (1973), Preliminary Data from Wyle Laboratories Research Project No. 59141,
"Communities Noise Profiles for Typical Railroad Operations."
R-2
-------
Railroad Contacts
Personnel in the operations departments of the following railroads were contacted in the
comse of this study.
AMTRAK
Atchison, Topeka, and Santa Fe
Baltimore and Ohio
Boston and Maine
Burlington Northern
Chesapeake and Ohio
Chicago, Milwaukee, St. Paul, and Pacific
Chicago and North Western
Chicago, Rock Island, and Pacific
Denver and Rio Grande Western
Durham and Southern
Gulf, Mobile, and Ohio
Illinoise Central Gulf
Louisville & Nashville
Norfolk Southern
Norfolk and Western
Penn Central
Union Pacific
Yard superintendents, yard masters, or engineering department personnel with the following
railroad companies were contacted in the course of this study.
Chicago, Milwaukee, St. Paul, and Pacific Railroad \ards,
Bensenville, Illinoise
Chesapeake & Ohio/Baltimore & Ohio Railroad Yard,
Walbridge, Ohio
Illinois, Central and Gulf Railroad Yard
Markham, Illinois and Centreville, Illinois
Norfolk & Western Railroad Yard,
Bluefield, West Virginia
Penn Central Railroad Yard,
Elkhart, Indiana
Boston and Maine Railroad Yard,
Mechanicville, New York
R-3
-------
Southern Pacific Railroad Yard,
Roseville, California
Union Pacific Railroad Yard,
Cheyenne, Wyoming
Burlington N'ortl ^rn Railroad
Chicago, Illinois and M. Paul, Minnesota
Miscellaneous contact, in the railroad, or related, industry
Association of American Railroads, Research and Test Department
Washington, D.C.
General Electric Compan''
Erie, Pennsylvania
General Electric Company Sales
Chicago, Illinois
General Motors/EMD
Lagrange, Illinois
-------
APPENDIX A
MAJOR TYPES OF DIESEL-ELECTRIC LOCOMOTIVES
IN CURRENT U.S. SERVICE (1 JANUARY 1973)
-------
CL> »-•«
O >— <
> 10
S- ir>
2:
S-
1 0
^- >>
13
1 T?
O Q'
J3 n
S_ S-
t3 ''O
1 J— J^
0
a.
.
:r
,
OJ
•a
o
s:
O)
CL
>^
H-
S~
(U
s~
3
4_>
O
TO
l+_
3
C
m
7E;
f
, i
rH
CM
t—
JL
CT>
--r
1
O-,
on
O'i
rH
rH
•H
<
o
£•'
o
o
o
rH
OJ
*~^
f- +
r~-
-j
1
m
o
c\J
<
O
^
o
o
o
rH
in
n
ro
^
'.?,
A^
'-O
I '"\
1
o^
on
0
vo
1O
<
o
'/""
0
0
VO
r-\
CO
-T
If \
]
O
ir>
vo
O
on
<
o
^
o
o
CO
no
^S
CO
t --
, i
'O
CM
"•O
A
CM
VO
1
1
LP\
ir\
rH
<
0
"^2,
0
0
VO
o
o
vo
*~,~
CO
A
LPl
VO
1
--4-
ir\
o
VO
CM
<
0
"i^
O
0
cr\
O
0
OA
^r;
(O
r-l
in
1
--j-
on
rr\
_TT
<
O
'^
Q
O
CM
rH
t —
IS
CO
1! i
( )
( ' 1
1 1 )
r \
M l
r 1
JL
on
in
1
i -i
in
vo
a)
f~-
'-C
O
p^-*
o
0
CM
rH
OA
*V;
CO
~*
vo
VO
1
-r
in
t—
on
t —
<
O
^
O
o
CM
rH
O
0
CM
rH
"^
CO
1
1
It)
vp
rH
i
vo
vo
4-
co
VD
H
•^
O
2;
o
C3
0
rH
' O
O
0
H
CO
1
1
-r
\o
I
11 N
1
VO
VO
+
vo
-J"
in
<
o
'^
o
o
Ln
rH
o
o
in
rH
t^-
CO
£H
0)
.C
O
4-J
•r I
CO
C 1
0 )
, 1
( 1
u ^
m
C \!
-T
lf\
1
0^
•^
0^1
o
CO
CM
CQ
O
?-'
O
0
in
rH
rQ
t—
~\
f~-
Q
C/)
\
(^
O
, i
i\j
i>">
i '
o
V ^ ^
o\
Lf \
1
-3
If \
CM
r~
o
-r
ffi
O
r."
o
in
t-—
r-J
( 1
o~\
^
CT.
(' 1N
I 1
e~>
~r
!f\
lvl~)
1
en
ir\
vo
CM
~r
«
0
f^:
O
O
CO
rH
CO
CM
\
CO
rH
Q
CO
x^
O-i
C3
Q
CO
"\^
P-1
0
I
1""'
o
01
CM
VO
1
CT\
LTv
LP.
Oi'i
on
o
c^
CO
>^
0
O
'o
CM
O
CM
P<
co
1 1 )
o
i. ~>
vM
fO
VO
1
CO
ir\
-~j~
C\J
CM
o
CO
CD
>H
O
O
^r
CM
CO
-^T
CM
\
•^3"
CM
n
CO
i
i
V >
1
^ ^ ^
Ol
v^^
I
iH
vo
vo
-T
CT^
O
W
O
^^
O
in
CM
CM
ffl
CD
on
xx
0
CO
^
O
t * i
t\.
j
VO
VO
1
on
•-O
i r\
j-
vo
rH
CJ
to
0
>-<
c
c>
m
ex;
UA
on
x\
Q
10
\v
n,
o
(""O
^
, n
' i
t
i
vo
vo
H-
on
o
rH
rH
.Tl
O
'^,
C)
o
o
CM
CO
on
n
CO
i
"
CD
to
O £n
3 3 "o
flH Q -P
•rH
rH rH ^
a1 al c/i
Ji -H
(!) O Tl
f" a> co
Q) a, o
o co IT;
to >
J< •' (
O -1-''
4- 3 o
O ',:'. i'.
' •', 1 (")
0 -i (
•1 fl W
('( ^ •, 1
C i O 1 *
(!) (!) •, 1
i; i -i i t
?-\ i'- A-l
-------
O »-i
> in
i- VI
D
»»- r-
1 "O
o ,
1—
a>
j^
3
•+J
O
'•O
3
C
IT)
», -
I * >
V ' 1
, -)
t —
J
C-—
ro
O
to
0
>H
O
O
ro
CM
ON
00
CU
CD
r^
'o
P
•rH
'.>
6';
-c
ci
q
rt:
j}
o
p
q
^ -1
i i
IV<
j(
«i)
r'-'i
j
r' i
i- 1
t\i
1
vo
+.
r —
r-{
(M
CM
o
to
X
O
O
O
ro
o
Q
x^
CU
O
CJ>
•H
P
q
' r
o
^
p
C) •
' 1
1
1
-
-r-
rv
VI)
r;
i
in
+
CM
VO
f>-
r-t
O
to
CD
X
O
o
VO
ro
in
rj-
Q
00
r.
o
i
tO
;••
, i
1
I
I
in
r
1 ^
i
,' ,
in
-^
,0
CM
to
0)
><
O
0
o
in
03
in
ro
\
H
O
0
VO
vo
X
o
^f
c^
n
Q
i
i
a i
, I
*
/
'n
i
1
cr\
ro
vc>
o
rH
0
O
g;
O
in
ro
H
CQ
EH
f"T,
\
c3^
flH
P"H
TJ
OJ
n
•H
rH
[^
CtJ
0
^
P
C/J
"•»
VO
VO
i
CD
O
2
O
in
on
rH
CQ
CM
fo
~\
dj
CM
U,
SL,
0)
p
to
o
O
PQ
\s
O
cd
o
i
i
C )
-i
rT~>
^
1
in
— f~
— J
rH
0
CO
rH
CQ
O
^
o
o
in
H
CQ
CO
pr.
\
0 to
•H (t)
0) A<
J.<
to
I
I
f~_
o
CM
CO
in
1
""I"
J
C\J
co
CTv
ro
CQ
O
;?;
O
O
in
•H
CQ
t—
(x,
\
<3^
[ —
Pr,
/
A-2
i
i
i
in
o
CM
r~-
m
1
,:l"
in
CO
rvi
CQ
o
^
O
in
t —
rH
CQ
cr\
fj_.
\
W
f-™~"^
"1
VO
Lf ^
1
0
•3"
r>")
ro
1
0
^
O
o
^r
C
O
p
^Cf-
^
f_i
0)
r;
o
p
•H
>
(/)
rH
ft!
^
fl)
J|^
CI
O
' 1 >
(O
rH
TO
'n
1
v-°
ro
en
rH
1
to
0)
>-<
1 0
O VO
o vo
in
C
o
p
o
t-
o
•r!
£H
P
O
0
.H
I'-l
^
VO
in
1
^r
vo
=T
1
to
CD
X
1 0
O vo
O VO
-0
c
o
p
in
cr\
i
I
T
CM
in
vo
v O
I
c%
rH
n\
in
Q
CO
0
X
O
0
in
(M
O
\
CQ
in
eg
^
0
J3
0
p
•H
|g
CO
T3
a!
O
(r:
i
i
C'^
rH
(.M
VO
o\
rH
CM
Q
to
0
X
O
o
CO
CM
O
^-x
CQ
oo
CM
^
1
1
+
CM
rH
CM
1
no
M i
H-
CM
i— 1
CXI
Q
to
0
X
o
in
CM
CM
o
\
CQ
CO
CM
in
^
^
-------
CU ' — '
o >— «
> t/)
1- 1/1
O> 0!
I/O i —
t— i
S-
OJ c/1
ft I/I
E 13
3 r—
S- *-
^3 ro
i f—
1 — •*-
O
D-
.
31
d)
-o
O
yr
CU
o.
t—
O)
t-
4J
O
<0
*5
c
rO
•v
1
1
(~)
1
J
vo
vo
o
^j-
Q
n
0)
>H
o
o
o
on
O
CO
0
on
&
CU
,C
O
4-*
•rH
*<
CO
•o
cO
q
'co1
f,
a>
£*
ai
i
i
[-_
f A^
-•r
1
t--
£>
t~-
n\
^f
Q
w
cu
>H
o
o
on
on
o
\
m
on
on
^
C)
•rH
-P
C)
CD
i-i
ii
i
i
H-
IfA
r 1
|
vo
[-
in
t-i
Q
:/)
QJ
K- 1
O
CD
vo
on
0
\
m
on
^
1
1
VL ;
'.O
O
T
VO
VO
Q
nj
cu
><
o
0
0
in
0
\
CQ
o
LO
CD
• o
CM
C\J
t y(
„ +
t
r-n
LCA
i
o
-T
on
u~\
VT)
1
O
•s:
o
VO
VO
oo
\
rH
CO
£<
CU
,cl
C)
-p
•H
c>i
o
o
rH
*~.»-
1
O
1
in
in
o
0
H
w
CO
0
>H
O
o
o*\
VO
CO
(^
i
UO
in
in
!IA
W
to
CD
>H
0
o
o
1-1
VO
CH
on
o
OJ
r_t
CJ5
VO
r-l
1
CD
--J-
CM
H
O
OJ
W
to
CD
>H
O
O
o
rH
^f
\
CM
CO
">
O
1
rH
-3"
t^_
CTv
-=r
W
cu
>H
0
o
o
rH
1
1 1
O
CO
crt
\
rH
CO
cn
su
cu
,d
0
4-5
•H
|S
CO
rrt
nJ
q
A-3
in
V( )
t-
o
Y
~j
o
o
-^f
W
to
CD
>H
O
O
li ~\
rH
O4
co
«
r" )
ro
-t
vo
tn
J
r
vo
i_n
I
o
LTA
r\j
rH
on
rH
W
CO
0
>H
0
0
VO
rH
on
\
CM
CO
K
L
~*
vo
T
rH
ir\
ro
o
rvj
W
CD
>H
o
o
vo
rH
un
\
^3-
Q
co
K
r I
,H
no
-T
rv-)
VO
I
vo
LT\
VO
on
-3~
Q
CO
0
>H
O
0
oo
rH
VO
OO
OJ
rH
\
r-t
r-t
CO
K
I
I
VCJ
( J
'.•O
v O
I
VO
vo
vo
OJ
Q
to
CD
>H
O
o
tn
rH
l-Pl
rH
J^J-
O
, t
rH
OJ
r I
a ;
VC )
I
r t
VO
J3-
VO
rH
Q
CO
0
>H
0
o
0
OJ
o
OJ OJ
po^r
co i
K o
I
I
CT\
rH
rH
O
VO
I
^r
in
OJ
0
rH
Q
CO
0
^
0
o
^T
OJ
in
rH
\
t-—
Q
CO
K
*
I
I
^
vo
o\
in
o
CO
Q
CO
0
>H
O
0
-^r
OJ
f— -^3"
C\l OJ
Q -^
CO I
CC O
I
I
Ox
OO
vo
vo
-~3~
vo
rH
OA
Q
CO
0
>H
0
o
m
OJ
in
r\j
-^-
I
o
i
i
rH
F''\
CO
VO
I
on
vo
in
ro
H
75
CO
0
>H
O
in
t>-
OJ
-o
r\i
vo
i
o
^
-------
(11 «— 1
{ ) fr— H
> Wl
'.-- in
J) '°
LJ
*"•' «~
1
1
""
i
JL' i/l
f »0
—
O
co
!
in ! CO
i.
.13
c<
^>-
(U "O
.Q . —
F. 0
^
CJ OJ
r— Ci.
U" ^
s:
1 "CJ
c>
OL
^
r-
$_
$_
rs
•M
O
10
^
c
JO
;._;
VO
1
•o
vo
ro
cr\
Q
w
OJ
£H
O
O
O
ro
o
CO
vo
X
O
ro
.^j-
i
o
I
1
1
,-H
no
c- i
VD
1 1 f 1
1
1
I
_
\:~\
i. —
VO
^T
ro
O
:
to
(U
K*H
O
o
vo
oo
vo
CO
vo
1
o
^
CO
c;
p
•H
'-*
00
•a
a)
o
rr;
Vvl")
-"-J
rH
CO
in
I
w
o
J>-l
o
o
LfA
H
rH
CQ
Cr.,
x^
<£|
PH
• ! '
1
1
v.^
1
1
- 1
I ! I
f " ' f J
>n
rH
O^
--r
1
to
d>
>^
0
0
vo
rH
OJ
\
CQ
ji(
\^
«Tl
FT<
y
o
rH
OJ
1
1
1
1
1
in
• i
'U
M
V
I
-.
It \
1
r >
ir>
-T
"r^
i
I
CO
(W
f*^
o
o
o
OJ
rH
CQ
AH
^^
1 O
1
* ( J
:• )
r -1
., ! ,-
l; A
O
in
f- I
*• L >
O
^T'
O
O
CO
oo
1
CO
I f N
1
•-O
_-r
fO
en
"
w
0)
M
o
o
o
H
o
iH
1
_=r
1
^3-
1
CO
O
' i >
(n
r i
O'\
. t
1
d
( \ *
V'l
(' )
'0
in
t 4
in
Ot
'T
'''
CO
0)
^
o
o
OJ
'H
OJ
rH
1
CO
JLj
a)
.c
o
i-i
•H
'. ;
f'/J
•n
I
- 1
':'
'O
T
W
(D
>^
O
O
OJ
rH
OJ
rH
1
CO
re
in
i
L--
jy
r—
.a
"^
w
0)
!>i
O
0
vo
rH
-O
rH
! VD
^ rH
1 t-H
CO 1
iX ro
Q re;
^
0)
0
-P
•rH
s
w
•a
cd
o
it;
v. „
o
in
1
(. ~)
L'-\
O
vo
rH
H3
(L)
rH
o
o
VO
rH
CQ
vo
rH
X
VO
rH
|~r,
re"
aj
C
•H
rH
E
03
a;
f,
40
CO
'-O
C ")
J
-1
1
~r
^*
[--
o,
rH
0
id,
O
o
o
rH
_^J-
^3-
1
O
rH
X
< V,
j
VO
r 1
CO
LT\
1
O
_
vcj
o
rn
O
S
0
o
OJ
jg-
^3-
1
^-_
rH
X
£_(
(U
.c
'o
4-5
•H
',^
to
1
1
m
,-!
ro
VO
1
I f~*\
-T
CO
m
0
£S
o
0
vo
rH
VO
VO
\
-^r
•=T
I
vo
rH
K
1
1
rH
rn
vo
in
1
rn
in
in
o
rH
O
s
o
o
-3"
OJ
vo
vo
1
-^r
C\J
tt
P
J O
ni i i HI : •; x; r-i o
• 0 I') f^^\. TN !•: Pi C_>
-------
> l/l
S- in
O) rt3
LO , —
t_>
!T
>—(
»-H
!_
OJ l/l
n to
E fO
n c —
2: o
(/>
s_
ro
dl
>-
S-
CD "O
-O i —
E o
=3 <~r>
z:
i_
QJ OJ
r— CL
«*- >^
•+- h-
rs
s:
1 "O
0 <"
_- I—»S
^j Wl
s- s-
13 «
t /—
1 — -i-
o
D-
rn
QJ
•o
O
IE
0)
CL
>>
1—
S-
OJ
L.
3
4->
O
(O
If-
Zl
c
fO
1
1
(\1
o
t 1
^
o
r-—
— (-
— J
f ,
M
0)
r-*
c]
15
•f- '
•H
to
(O
s.;
•i t
!!
• •$
/ 1
1 1
o
0
1 f~\
Li i
r-
M
0)
r-*
O
•J- ^
•H
CO
J,
fl)
j )
r
> ^
o
1 ' 1
:>
A-5
-------
APPENDIX B
REVIEW OF THE USE OF AUDIBLE
TRAIN MDUNTED WARNING DEVICES
AT PROTECTED RAILROAD HIGHWAY
CROSSINGS
-------
REVIEW OF THE USE OF AUDIBLE TRAIN MOUNTED
WARNING DEVICES AT PROTECTED RAILROAD -
HIGHWAY CROSSINGS
B.I Requirements For the Use of Audible Warning Devices
The stopping distance of trains is much longer than
that of motor vehicles, they are much more difficult to
reaccelerate, and due to their length they often overlap
more than one road intersection at a time. Therefore,
trains have traditionally had the right-of-way at level
crossings, while motorists are expected to look out for
trains and give way. The burden is then placed upon
the railroad to assist the motorist in determining when
a train passage is imminent. The traditional method of
doing this is to sound a whistle and/or bell and keep a
headlight burning on the head ends of all trains, and to
mark the crossing in some manner so as to attract the
attention of approaching travelers.
Public Railroad-Highway grade crossings may be equipped
with one of the following, which are classified herein
into the three major headings shown:
(a) Unprotected
(1) Unilluminated stop-look-listen sign or
"cross buck" at the crossing generally accompanied by
striping and words painted on the road surface and passive
prewarning signs in advance of the crossing.
B-l
-------
(2) As above, plus continuous (night time)
illumination of the crossing and/or the signs.
(3) As above plus flashing amber caution lights.
(4) Any of the above, plus "rumble strips" on
the road surface.
(b) Protected (no gates)
"ihic group of systems may employ combinations of the
n i.gns, lights, mirkingn, etc. from (a) above, but is distin-
i.lished Hy the addition of:
(1) Flashing lights generally plus bells, which
are actuated upon the approach of the trains (s) by virtue
of automatic electrical signals attached to c.he tracks.
These systems are arranged to be fail-safe, in that most
internal failures cause the signal to indicate the approach
of a train.
(2) Traffic lights may be used in some places,
in lieu of the characteristic flashing crossing lights,
but also conveying the intelligence that a train(s) is in
fact in the vicinity.
(3) Watchmen, stationed at the crossing, or
trainmen walking with their train, will "flag" motorists
or may activate lights or other devices.
(c) Protected With Gates
In addition to active signals and advance warnings
dS ih (b) physical barriers are automatically dropped in
the motorists' path upon the approach of the train(s),
often with lights attached thereto.
B-2
-------
These gates may interrupt only the approaching highway
lanes (half gates) or both lanes on each side (to discourage
driving around) and may be supplemented by small
pedestrian gates at walkways. However, these gates are
not constructed so as to physically restrain vehicles, but
are really a type of "sign", intended to assure driver
attention and realization that a train is to be expected.
Gates are commonly used at busy crossings where there are
two or more tracks, to add a degree of protection against
motorists proceeding as soon as one train has passed, when
there may be one approaching on another track.
The cost of installation of crossing signals varies
widely and depends greatly upon particular local circum-
stances. Modest installations with gates average about
$30,000, and may be as high as $60,000. The annual cost
of inspecting, maintaining, and repairing protected
crossings is about $1,000 each, not including the cost
of roadway and track work.
Complete grade separations may cost hundreds of
thousands of dollars, or even millions, and while many
are being constructed, the number is not statistically
significant within the context of the overall problem.
(When separations are installed, it is usually possible
to arrange for the outright closing of a few nearby
crossings, thus expanding the safety benefit of this
large investment.)
B-3
-------
The level of crossing protection installed at a
particular location is determined by the hazard involved
which is effected by the amount of road traffic, the
number r<\ > --eed oi trains oassing and topography. This
may be determined by the judge::unL of local officials,
the vailroad i,.inagencats, or both and is often establishe<
simply by a past record of accidents at a crossing in
question. The investment in crossing equipment may be
the responsibii of the railroad, the State or local
government, the Federal government or any combination
thereof. This question has been the subject of much
controversy in the past, and is in a state of flux
at present, with the trend being toward greater govern-
ment responsibility although some railroads continue to
spend large sums of their own money on new systems every
year. Automatic signal system maintenance has always
been the responsibility of the railroad.
Train-born signals to warn motorists and pedestrians
of the approach of trains are required by most States.
Federal safety regulations are confined to the inspection
of such devices on locomotives, to the end that - if
present - they shall be suitably located and in good
working order (Safety Appliance Act, 45 USCA; 49 Code of
Fed. Regulation 121, 234, 236, 428, 429). The Federal
government has shunned greater regulatory responsibility
in this field in the past. There is a very significant
B-4
-------
Federal research and promotional effort underway to
improve grade corssing safety, however.
The State laws requiring train-born signals do
not quanlify their loudness. It is common for the State
laws to quatify the requirement to apply all public
crossings except in municipalities, leaving the use of
horms or bells in towns and cities to local discretion.
A survey of the 48 contiguous States yields the
following summary of information regarding their
regulations:
.. Requirements for sound signals at public crossings
imposed by:
Statute 38
Public Utility Commission 1 (Calif.)
Common Law 3
Penal Code 1 (N. Y.)
None or no information 5
48
Requirement at private crossing: - if view is
^obstructed .... 1
Signals to consist of:
Whistle or bell 24
Whistle and bell 7
Whistle 6
Bell only 2 (Fla. & R.I.)
-------
Distance at which signal is to be sounded:
Beginning at a minimum of distance (35 States
varying from 660 feet in Michigan to 1500
feet in South Carolina, with an average of
1,265, the most common being 1,320 feet
(80 rods).
Beginning at a maximum distance (3 States) :
Montana 1,320, Ohio 1,650, and Virginia
1,800 it.
To continue until train:
Reaches crossing 35
Is entirely over crossing 3
Exception of some form provided for incorporated
areas in at least 15 States:
California, Lowa, Indiana, Kentucky, Michigan,
Minnesota, Missours, New Jersey, New York,
Nevada, Utah, Virginia, Washington, Wisconsin,
and Florida.
.. Exception provided at crossing with:
Gates and/or watchmen - Delaware
Flashing lights and bells - Illinois
(More is said about exceptions in a later section of
this report.)
B-6
-------
Railroad operating rules reflect the ordinances in
effect in the areas through which they pass, generally
encouraging the use of warning signals at the discretion
of the operator to avoid accidents, but admonishing
against unnecessary soundings. Specific supplementary
advice is contained in Standard Rule 14, which is adopted
by many carriers, requiring the sounding of signals in all
situations where two or more trains are at or approaching
a crossing simultaneously, due to the extra hazard con-
sequent to the limited view and preoccupation of approach-
ing motorists and pedestrians when they see or hear just
one of the trains.
Two good examples of State requirements for the
sounding of warning signals at crossings are those of
California and West Virginia, attached hereto as Appendix
Al, A2, and B, respectively.
Over and above statutory and regulatory requirements
for the use of warning signals on trains, the judiciary
and juries have tended to assume that there is a burden
upon the operators of railroads to employ such devices.
Numerous judgments have been made against railroads in
court cases wherein the sufficiency of warnings were
questioned, particularly by juries and seemingly to a
relatively greater degree in California. As a result,
railroads are reluctant to dispense with any ordinary
action which might be construed to be a contributing factor
in crossing accidents. More will be said on this topic
B-7
-------
in a later section.
In addition to requirements for warning travellers
at level crossings, the State of New Jersey Public Utilities
Commission has ordered that passenger carrying railroads
operating in that State sound a horn or whistle prior to
stopping at or passing through a passenger station on
a track adjacent to a platform. (January 20, 1972,
Docket 7010-525) Subsequent modifications limited this
requirement to one long blast, during daylight hours, and
then only when the engineer has reason to believe persons
may be in the vicinity of such platforms.
I
B.2 Railroad - Highway Accidents
There are over 220,000 public rail highway crossings
at grade in the United States, of which 22% are actively
protected (Categories 2_ and _3) . (There are also about
150,000 private crossings.)
In 1972 there were almost 12,000 public crossing
accidents, resulting in 1,260 deaths. These totals have
been decreasing slowly since 1966. In 67% of these accidents
the train struck a motor vehicle, in 28% a motor vehicle
struck trains and in 5% trains struck pedestrians or there
NOTE: Figures in this section are taken from references
(4) and (5). Accident figures sometimes differ
between references due to the $750 cost baseline
for reporting accidents to the Federal Railroad
Administration. Crossing figures may differ due
to the inclusion or exclusion of private crossings.
B-8
-------
were no trains involved. 39% of the collisions occurred
at crossings provided with gates, watchman, audible and/or
visible signals, while 61% were at crossings having signs
which did not indicate the approach of trains (Category !_) .
63% of the collisions occurred during daylight, and
37% at night. It is believed that about 67% of motor
vehicle traffic flows in the daytime, 33% at night, suggest-
ing a slightly higher crossing hazard at night (37 £ of
the collisions with 33% of the traffic).
Automobiles constituted 73% of the motor vehicles
involved, trucks 25%, motorcycles 1.3% and buses 0.3%.
When motor vehicles struck sides of trains, they
usually contacted the front portion thereof, particularly
during daylight; the propensity to strike elsewhere in-
creases at night. The side of train category appear to
be twice as hazardous at night, in that 53% of them occur
then, with 33% of the traffic, with the peak occurring
between midnight and 2 a.m. In fact, when these are de-
ducted from the total, the train-strikes-vehicle collisions
are in about equal proportion to the traffic distribution,
day and night.
The propensity for accodents at actively protected
crossings is also greater at night than in daylight, per
unit of traffic, perhaps indicating that driver alterness
is a more significant factor in these cases.
B-9
-------
TABLE 1. SUMMARY OF PUBLIC CROSSING TYPES,
LOCATIONS AND ACCIDENTS (1970)
URBAN
RURAL
TOTAL
5970
18050
4240
28260
(3624)
50860
(3827)
79120
(7451)
2970
14620
2680
20270
(1533)
12385
(3428)
144120
(4961)
8940
32670
6920
48530
(5157)
17471
(7255)
223240
(12412)
GATES (category 3)
SIGNALS (category 2)
OTHER OR MANNED
TOTAL ACTIVE
(ACCIDENTS)
PASSIVE (category 1)
(ACCIDENTS)
GRAND TOTAL
(ACCIDENTS)
There were 70 fatalities in 1972 at gates, and
440 total at all active crossings, somewhat less than one
per 100 crossings.
Accident rates and severity are significantly higher
at actively protected crossings, indicating that the
greater hazards where they are installed are not fully
compensated for by the increased protection. The rates
are also higher in urban areas than rural, for both
active and passive crossings, so that in the very areas
where noise exposure is greatest, the safety situation
is worst.
B-10
-------
It could also be argued that the accidents which
occurred in spite of the active protection demonstrate
the ineffectiveness or waste of warnings such as train
horns in such areas.
While vehicle traffic, train traffic and speed
continue to increase, protection installations are also
increasing, and the total number of crossings is de-
creasing. The 1973 Highway Act provides a total of
$175 million over a three year period for crossing safety,
on a 90/10 Federal share basis, or a potential total of
$193 million, of which at least half is to be spent on
active protection systems. Gate installations constitute
about 30% of all new protection, and since such systems
cost about $30,000 on the average, approximately 1,000
more gate installations should occur during this three
year period, in addition to those installed at railroad
initiative. The Northeast Corridor is already on its
way to being totally without level crossings of any kind.
NOTE: Reports of crossing statistics vary from year to
year, are often based on different reporting
criteria and may be for either public and private
crossings.
B-ll
-------
B.3 The Impact and Effectiveness of Locomotive Horns
Acoustical Chaiacteristics and No so Impact
The audibility of air horns, the predominant warning
devices which are the subject of attention herein, has
been investigated (1) as part of a DOT program to make
crossing warning systems more effective. It was found
that the horns which are presently employed are not very
effective, and to be so it would be necessary to increase
their loudness, "warbling" and/or the use of as many as
5 chimes (pitches) have been recommended. Obviously,
since the whole purpose is to gain attention and instill
a sense of imminent danger and alertness in persons
located at 1/4 mile distance, such signals are bound to
be disturbing - by definition.
Figure 1 shows the approximate noise pattern of an
average locomotive horn. In order to increase motorist
impact to a degree sufficient to be of real value, the
loudness would need to be increased as much as 23 dB,
resulting in a loudness of 128 dB at 100 feet. (The
A and C weighted loudness of the common air horns are
almost identical; no distinction is made in the literature)
Loudness at 90° from the direction of movement is
5 to 10 dB less than straight ahead and it is possible
B-12
-------
that this pattern could be improved somewhat, but the loud-
/•\
ness should be substantially maintained to at least 30
each side of center due to the variation in angle of approach
of railroads and highways.
This problem of audible warning is shared with emer-
gency vehicle sirens. Fire, police and rescue units have
a parallel problem. With motor vehicle windows closed,
in modern, acoustically well constructed vehicles, and
with road noises and/or air conditioning, radios, etc.
competing with the warning devices, at least 105 dB is
needed outside a vehicle in order to gain the attention
of most drivers. Research is underway to determine the
feasibility of installing warning devices inside motor
vehicles, which would be actuated by the approach of a
train or an emergency vehicle.
In Figure 1 are shown the acoustical characteristics
of the common railroad air horns, the orientation of
train and vehicles in a set of relatively high speed en-
counters, such that the motor vehicles shown would have
a reasonable stopping distance to the point and instant
of train passage at a crossing. Table 2 lists the required
noise levels at vehicles travelling at various speeds
(exterior background noise assumed dominated by running
noise of vehicle) to gain the attention of the drivers;
the 50% attention column nearly corresponds to the average
B-13
-------
110
100
dB
300
600
900
1200
1500
70 mph
ILLUSTRAT uiST OF HORN
LOUDNESS VS DISTANCE &
EXAMPLE OF DISTANCES TO
APPROACHING VEHICLES
] RECOMMENDED
1800
nrah
TABLE 2
VEHICLE SPEED
>_ 35 mph
36 - 50 mph
51-65 mph
dB OUTSIDE VEHICLE'FOR
50%
83
87
91
FOR DRIVERS TO NOTICE
98%
101
105
109
(SOURCE: REF 1) STANDARD DEVIATION - 6dB
-------
situation today. To alert 98% of the drivers at (B)
it would be necessary to increase the sound levels by
about 30 dB, resulting in a level at 100 feet abreast of
the locomotive of about 130 dB.
Figure 2(a) illustrates the noise pattern which
characterizes most horns in use today, and Figure 2(b)
depicts the areas lying within an envelope in which the
noise from a horn being blown for a crossing will equal
or exceed 77 dB for some period with each train passage.
The 77 dB figure is chosen rather arbitrarily, largely
because it corresponds to a 1,000 foot boundary adjacent
to the track, which is compatible with the modest data
available on residential population alongside railroads.
It is also a reasonable number as regards nuisance levels
of intermittent noise intrusion, being used herein
merely for the purpose of approximating the scope of the
impact of warning device noise.
Some 202 miles of railroad route in 12 areas of 10
cities of varying overall size, selected randomly, have
been reviewed. The population within 1,000 feet of the
railroads in this examination average 2,410. Therefore,
in urban areas, about 600 persons are usually exposed to
77 dB from an instant up to 10 or 15 seconds each time a
train passes a level crossing.
B-15
-------
LOCOMOTIVE HORNS - .AVERAGE NOISE PROPAGATION UNDER
IDEAL CONDITIONS
800
1000'
a) 77 dB Profile
o
o
o
fM
4000'
b) Area subjected to 77dB level or more
Based upon extension of profile along route
FIGURE 2
-------
Table 3
% of Population
1. Unprotected 33.0 million 16
2. Signalled 13.7 6
3. Gated (3.7) (2)
Total 46.7 million 22
(Signalled includes gated)
This vieHlld indicate that one-fifth of the total
!|»S;
population ;Ip "within hearing" of a grade crossing. In
fact, the n^ise patterns are probably much less severe
than shown here, due to topographical features, and many
of the projt^Sted as well as some of the unprotected
&
crossings &ita covered by restrictive ordinances, so that
probably incite like 10-15% of the people are exposed to
-*!?/
the 77 dB or greater level used here for illustration
(exterior to dwellings, etc.).
If the use of horns was prohibited at all actively
protected crossings, 30% of these exposures would be
avoided. If such a restriction was confined to crossing
with gates, 8% of the exposures would be avoided. These
abatement measures would be noticeable to about 3% or 1%
of the population, respectively, allowing for attenuation
B-17
-------
.locally and background noise and the fact that many
crossings are already c^~ ^reH by such rules.
AfaS'"ning that the UFJ o f. signals and gates corresponds
to fir i-.gh- ' hazard levels or volume classes as depicted
by the Depar'-in^nt of Transportation, the number of daily
traxri and vehicle passages at the crossings in question
has been estimated as shown in Table 4.
Table 4
Daily Trains Daily Vehicles
Total over signalled
crossings 950,000 160,000,000
Average per signalled
crossing 20 3,300
Total over gated crossings 200,000 70,000,000
Average per grated crossing 22 7,800
If the average train sounds its horn over a period of
12 seconds, the average citizen within 1,000 feet will experi-
ence the noise at 77 dB or more for an average of 8 seconds.
At gated crossings where horn blowing occurs 22 times per day,
the equivalent energy produced (L ) is 50.1 dB, whereas at
signalled crossings where it occurs only 20 times per day, the
equivalent energy would be 49.7 dB.
People residing within hearing of grade crossings
are generally conditioned to the sound, which tonewise
B-18
-------
is not particularly disturbing. The most common casual
notice of the use of horns at crossings is expressed by
persons staying at motels, which are not infrequently
located on highways which parallel railroads and are near
road crossings. Being otherwise unaccustomed to the sound,
it is quite noticeable, particularly at night.
Warning^ Effectiveness of Horns
As noted above, at present only about half of all
motorists can notice the sound of a train horn when they
are driving and their windows are closed, even under ideal
conditions. And the alerting capability - even if the
horn is noticeable - is still less. It is impossible to
determine how many accidents have been prevented by the
routine sounding of horns, although it is apparent from
the experience of train drivers that many accidents have
been averted by the ad hoc sounding of horns, while an
even greater number have occurred in spite of it. However,
these comments are directed to all crossings, passive
(unprotected) as well as active (protected). It is unlikely
that either routine or ad hoc use of horns.at crossings
where lights are flashing and bells are ringing at the
crossing significantly improves ordinary driver attention,
particularly where gates are lowered as well. On the other
hand, some drivers and most pedestrians can hear the horn
when it is sounded. Also, in those occasional incidents
where a vehicle is stalled on a crossing the horn may serve
B-19
-------
to divert people from continued efforts to move their
vehicle and to depart forthwith on foot. But in the latter
case, sounding on a routine basis is probably not necessary.
Attached hereto as Enclosures C, D, and E are (abridged)
reports on three rather typical grade crossing accidents
wherein the accidents occurred in spite of crossing signals
and the sounding of warnings by the train. These are
selected somewhat randomly, to illustrate by example a
kind of crossing accident which is all too common.
B.4 Prohibition against the use of audible devices
It is already quite common for the routine sounding
of horns or whistles to be prohibited, except in emergencies.
It is also common for these prohibitions not to be enforced.
A careful search for cases where such prohibitions appeared
to, or were claimed to contribute to an accident has not
yielded evidence of a single such situation.
Among the localities which restrict the use of horns
are those listed in Table 5.
B-20
-------
Table 5. Some Localities with Restrictions
Notes
The State of Florida (2)
The State of Illinois (1)
The State of Massachusetts
Chicago, Illinois (1) (2) (3)
Houston, Texas (1) (2)
Minneapolis, Minnesota
Buffalo, New York (1) (2)
Philadelphia, Pennsylvania
Knoxville, Tennessee (1) (2)
Durham, North Carolina (2)
Mason City, Iowa (3)
Warren Pennsylvania
Elkhart, Indiana
Toledo, Ohio
Columbus, Ohio
Akron, Ohio
Lynchburg, Virginia (1) (2)
San Bernadino, California (1)
South Holland, Illinois
Elmhurst, Illinois
Lockport, N.Y.
Rochester, N.Y.
(1) Contacted local authorities in course of this study.
(2) Specific Information contained in Enclosure F.
(3) Not enforced. B-21
-------
The 15 states where requirements to use horns are
excepted, but not necessarily prohibited, in incorporated
.-sreas are:
Table 6.
California''1 New Jersey
^Torida New York*
Iowa* Nevada*
Kansas Utah
Kentucky* Virginia*
Michigan* Washington
Minnesota Wisconsin
(*also have local-option provision)
In 4 additional states there is a local option provision,
allowing cities and towns to relieve requirements:
Table 7.
Illinois North Carolina
Indiana West Virginia
Two states permit silent running at crossings with
certain protection systems:
.. Delaware: warning requirements do not apply when
crossing is protected by watchman or gates.
.. Illinois: requirements do not apply when crossing
is protected by automatic signals (with or without
gates).
B-22
-------
One of the most comprehensive Noise Control Regulations
thus far drafted in the United States is that of the State of
Illinois. As it stands, its property line limitations would
affect the use of audible crossing warning devices except that
its Rule 208, Exceptions, states: "Rules 202 through 207
inclusive shall not apply to sound emitted from emergency
warning devices and unregulated safety relief valves."
Thus, it can be seen that there is considerable
precedent for placing constraints upon the use of audible
warnings, with no apparent adverse effects. However, they
are not uniformly enforced, and where enforced, the carrier
generally receives written instructions from the constraining
authority, and is nevertheless impowered to sound warnings
"in emergencies"..."in the event of impending accident"...
etc.
B.5 Judicial Background
Tort litigation constitutes the bulk of the legal or
judicial history of grade crossing safety responsibility.
Abstracts of 2500 cases throughout the United States during
the period 1946 to 1966 have been surveyed (3), checking
into 300 possibly related to the question at hand.
In addition, 5 cases were cited by a cooperating
railroad as illustrative of the railroad liability question.
One of these was found to be inapplicable to the question
at hand, three were decided in favor of the railroad. In
the other, a jury found for the plaintiff, although a
B-23
-------
whistle had in fact been sounded. Of these, 21 appeared to
be somewhat related and the case records were reviewed.
Nothing was unearthed which would appear to deter Federal
or local constraints on audible traincarried devices at
protected crossings.
Several themes are woven through the opinions rendered
in the many cases on record. These are certainly not
uniformly respected, but they are sufficiently common as
to be noticeable:
.. Safety provisions, including warnings, should be
compensurate with the specifics of local conditions.
.. The railroad is expected to give "adequate and
timely" warning of the approach of a train. The railroad's
case is often intended to show that their warning could
have been heard by an attentive motorist.
.. To be cause for placing liability, an omission on
the part of the carrier generally must be shown to have
contributed to the event in question.
.. Motorists are generally expected to be cautious
at crossings, to the extent even of stopping or look
"and listen".
.. Contributory negligence on the part of a motorist
is generally taken into account.
The fact remains, however, that courts, especially
juries, have extracted severe payments from railroads,
B-24
-------
seeming usually to give plaintiffs the benefit of all doubt.
For this reason, railroad companies are understandably at
pains to make any changes which could conceivably be con-
strued as a reduction in safety precaution (or increase in
hazard). Also, the employees charged with operating trains
are usually subject to prosecution under criminal law if
negligence and/or violation of a statute might be involved,
and are thus inclined to err in the direction of sounding
their warning devices, not to mention their sincare personal
desire to avoid injury to even the negligent public, as
well as themselves. (Collision between trains and large
trucks, especially those carrying hazardous materials, are
very dangerous to the occupants of the train.) A possible
fine for violation of a noise ordinance is not nearly as
imposing a threat as the liablility, criminal action and con-
science which accompany the threat of collision.
B.6 Summary
One of the railroad noise sources which has been
commented upon in the course of interstate rail carrier
regulatory development by this Agency's Office of Noise
Abatement and Control, is that of railroad train horns
which are sounded routinely at grade crossings. It has
B-25
-------
been suggested that such sounding be prohibited in cases
where automatic, active protection is in operation at
the crossing itself, particularly where this protection
includes gates,
However, it remains that the routine sounding of horns
mignt be contributing to the prevention of some accidents.
Certainly, a small segment of the population is exposed to
serious noise intrusion thereby and a reduction in their
welfare, particularly at night. But it is the Agency's
position at this time, that it would be imprudent to single
out and restrict night time use of horns, since the crossing
hazard with regard to driver behavior is, if anything, worse
at night.
In view of the questionable value of train horns for
warning highway drivers, particularly at locations having
active crossing signals, it may be appropriate to encourage
the abolition of routine use of horns at crossings so
equipped, particularly but not necessarily only those
with gates. The circumstances which determine hazard
levels as well as noise intrusion vary widely and are
peculiar to local circumstances. It is therefore concluded
that regulation of railroad warning be best left to the
option of local authorities at this time, recommending
thereto that consideration be given to restrictions upon
the routine sounding of train horns at protected crossings.
B-26
-------
REFERENCES
1. The Visibility and Audibility of Trains Approaching
Rail-Highway Grade Crossings; J. P. Amelius, N. Korobow;
NTIS-PB-202668.
2. Driver Information Systems for Highway-Railway Grade
Crossings; K. W. Heathington, T. Urbanik.
3. American Digest System, 6th and 7th Dicennial Digests.
4. Rail Highway Grade Crossing Accidents from the Year
1972, Department of Transportation, Federal Railroad
Administration.
5. Report to Congress on Railroad-Highway Safety, No. II,
Department of Transportation, FRA/FHWA.
B-27
-------
ENCLOSURE A
Public Utilities Cco'e Annotated of the
State of California
Adopted May 31, 19bl . •• ,
f . *•* /~\ *
Page 734
• .—', • *'.• - , ARTICLE 8 ..'.; v . : . .. -
; .•';'"• '..'.' :.-•' -; . CHIMES . ; '^ .• ';.••.;•-. .'
Collateral Heferences , ... < - .
z> 7678. Omission to sound boll or whistle.' hvery person in chs;:r;3 c
a loccmotivo-enginc who, before crossing any traveled public v;ay, emit;
to cause a bell to ring or steam whistle, air siren, or air whistle- to
sound at the distance of at least 80 rods from the crossing, and up to
it, is guilty of a misdemeanor. . • ; ..•-''•• ,-_ -
Legislative History
••'n?ot
-------
PUBLIC UTILITIES CODE, STATE OF CALIFORNIA
(Abridged)
7604. A bell, of at least 20 pounds weight, shall be placed on
each locomotive engine, and shall be rung at a distance of at
least 80 rods from the place where the railroad crosses any
street, road or highway, and be kept ringing until it has
crossed the street, road, or highway; or a steam whistle, air
siren, or an air whistle shall be attached, and be sounded
except in cities, at the like distance; etc.
B-29
-------
ENCLOSURE B
THE WEST VIRGINIA CODE
(At> ridged)
1-?
V/srnmg of approach
jv.0.u'-acl tracks.
of train at crossings; ~crossing
Aj?eH_or steam whistle shall be placed on each locomotive engine, whicS |
n|i."r*be"rung or whistled by the engineer or fireman, at a J^tancej)f_ at^'
least sixty 'rocls_f rotr the place where the railroad crosses any publicjtreet
or highway, and be jcept ringing or whistling for a time sufficient to fe'ive
due notice of the approach of such train before such street or highway is
is'ajcnisdemear.or punishable by a fine of
anyJfellHI£_so *°
not exceeding one huniur5cl_for all Damages
^-^'
I. Scop* of Statute as to Warnings.
A. General Consideration.
B. Does Not Apply to Trespassers.
C. Does Not Apply to Employees.
II. Failure to Give Warnings as Negligence:
111. Evidence.
I. SCOPE OF STATUTE AS TO
WARNINGS.
A. General Consideration.
Michit*s Jurisprudence.—For full treat-
ment of accidents at crossings, see 15
M.J., Railroads, §§ &-101. As to duty to
Kive signal by bell or whistle, see 15 M.J.,
Railroads, §§ 81-83.
ALR references. — Railroad company's
'"Cgligence in respect to maintaining flag1-
mur: at crossing, 16 ALR 1273; 71 ALR
1160.
Duty cf railroad company to maintain
Bagman at crossing, 24 ALR2d 116L
Acrr.issibility of evidence of train speed
••r;.,r to irnuJe-crnssing accident, and coni-
' i'< >. nrnn-l.nv r.-i'iti!ri.'>nPit as to
'''-'':--''-• 1* fully a.-, uvatti:i;< as the statu-
'.',.-}• fliitr. What the notice and warning to
• •" P'."'» «: shall be depends, under the
','''".m-'D ;.iWi upon the circumstances of
- vh c.-'ic; but some adequate methods of
Contributory Negligence.
apprising travelers of the crossipff muat
be practic.qd.j K'iland v! Honongahela &
West Penn Pub. Serv. Co., 106 W. Va. 528,
147 S.E. 478 (1928).
Both bell and whistle are not required
without statute. — There is no absolute
requirement upon a railroad company to
blow a " whistle and ring a bell at a
crossing unless made so by statute.
Xilana v. Monongahela & West Penn Pub.
Sew. Co., 106 V/. Va. o23, 147 S.E. 478
(1928).
The methods of apprising travelers of
a crossing almost universally adopted
art by the ringing of a bell or the
sounding of a whistle, but in order to
make both obligatory, tho ««e of both
r. .?r !v (••!:;•! .'".:• by a ?-_-:.' T.e. Ni'v '. v.
y..,-~. -.'..'<<. e' . '; V. -.-•:. ;'... -. ?ub. o."/.
Co., 10'' r<". V.i. -jJ:', : \~. i ',. -'7S (IDJS).
, .^re. minimum re-
__,, .
J=5-T*he provisions of this sec-~~
tion as to warning signals are of broa~d
application and are minimum ' require-
ments, j;nd in every case the compliance
B-30
-------
with this statute, plus the presence of ai?
efficiently operating crossing-bell •.:'! not
(apart from the ou^siioi: of contril.u:ory
negligence of the plaintiff) constitute an
ironclad defense to the railroad, ir.djr all
cnc'jnistances. Baltimore & O.P..R. v.
Dcneen, 161 F.2J C7J (4th Cir. li>-:7).
^ravelors haye_the FJillil.J0 R'sume
_jjjajT trains will gi*e thA-iiailArsTgr.ais at
^rosshnjs.^Morris v. Baltimore & O.R.R.,
107 \V. Va. 97, 147 S.E. 547 (1920).
But railroad only CM es duty to signal
as required by statute-—The driver of an
automobile on a public crossing is an
invitee, and the railway company is bound
only to use reasonable care not" to collide
with the automobile, and owes only the
duty to give the signals provided by stat-
ute. Chesapeake & 0. Ry. v. Hamveli, 142
W. Va. 318, 95 S.E.2d 462 (1056).
\Yhere the only ewvJence was that the
warning signals rpqi.irej by this section
were not given, and that the failure to
do so constituted negligence on the part
of defendant, it was held that notwith-
standing defendant's negligence, if d»-
cea.-ecl's contnbatory negligence is es-
tablished as a neater of la^.. p!;i as the Ir.v,- iv
ihi-i section is intended to protect^
'/r.< '••T-£u;hway .— I'he duty imposed
'• -.'. .:•-• loTound a~bell or whistle when
- '.•.', P. j.-'jblL crossing does not.
- !~'''r':rl conoany to q;ive such
•(-• 'J-e railroad tracks f.s parts
\.- .'ii^'r,-.'. r.y Jones v. Virginian
11. FAILURE TO GIVE WARNINGS
-\S NEGLIGENCE; CONTRIBU-
TORY NEGLIGENCE.
Violation of section is negligence.—The
failure to give proper signals of the
approach of a train to a -ailroad crossing
as required by this section would consti-
tute negligence on the part of a defendant
railroad. Cavendish v. Chesapeake &
0. Ry., 05 \V. Va. 400, 121 S.E. 498
(1924").
But does not impose liability unless it
proximately causes injury.—Liability for
injury to baby of 13 months could not be
based" on failure to give signals since the
failure was not the proximate cause of
the injury. Virginian Ry. v. Armentrout,
158 F.2d 358 (4th Cir. 1946).
Failure to ring the bell or blow the
whistle at crossings, though required by
law, will not render the company liable,
unless that be the provi: :ate cause of the
injurv. Bevel v. Nev.port News & Miss.
Valley R.R., 34 W. Va. 53S, 12 S.E. 532
(ISDO").
'ThuSj^rirHrocidJs^otJJnjjle if coutiibu-
torr negligence is proximate cau^e. —
'"Where one is injured by clffelessiy driv-
ing on a railroad crossing in front of a
moving engine or train, the proximate
cause of his injury rr.ust be regarded as
his contributory negligence, and not the
negligence of the v«il:oaJ company in
failing to ring the bell or blow the
whistle. Cline v. McAdoo, 85 \V. Va. 524,
102 S.E. 21S (1920).
quires, to ascertain whetl.er a train is
approaching the crossing. Beyei v. New-
port News & Miss. Valley R.R., 34 \V. Va.
53tJ, 12 S.E. 532 (1890); Bassford v. Pitts-
burg1, Cincinnati, Chicago & St. Louis Ry.,
70 W. Va. £80, 73 S.E. 926 (1912); Cline
v. McAdoo, Sf> W. Va. 524, 102 S.E. 218
(1920); Robinson v. Chesapeake & 0. Ry.,
90 W. Va. 411, 110 S.E. "870, 22 A.L.R.
892 (1922); Cavendish v. Chesapeake &
O. Ry., S5 \V. Va. 490, 121 S.E. -193 (102-1);
Gray v. Norfolk & W. Ry., 99 W. Va. 575,
130 S.E. 130 (1925); Berkeley v. Chesa-
peake & O. Ry., 43 W. Va. 11, 26 S.E. 349
(1836>.
Though a traveler has the right to as-
sume that warning signals required by
this section will be given, failure to give
them will not excuse him from exercising
ordinary care, and taking the necessary
precautions for his safety. Arrowood v.
Norfolk & W. Ry., 127 \V. Va. 310, 32
S.E.2d 634 (1944).
III. EVIDENCE.
1?HIJ£
^Proving that signals^
— — -— ^
were not given rests upon the plaintiff.
Parsons v. Ne-.v York Cent. 11. R., 127 \V.
Va. 619, 34 S.E.2d 334 (1945).
No conflict 5n evidence where some
witnesses heard signals and some did not.
— The fact that witnesses have heard sig-
nals given by a locomotive approaching
a crossing warning travelers of danger,
is not necessarily in conflict with the e\i-
dence of other witnesses who did not hear
them; for the observation of the fact by
those who heard is consistent with the
failure of the others to hear them. Caven-
dish v. Chesapeake & 0. Ry., 95 W. Va.
490, 121 I.E. 498 (1924).
Unless witnesses not hearing had equal
opportunity to do so. — Testimony with
reference to the statutory warning signals
which only goes so far as to establish that
the witnesses did not hear the bell rung
and the whistle sounded is not in conflict
with the testimony of other witnesses who
tpifified that in fact the whistle v.as blown
B-31
-------
and the bell rung. An exception to tlie
foregoing- ru'o arises where there was
to.un!^pportunitv of a witness to h.vr th->
•t?g'."i:-< a-x! sp-idal circus •*.>:>(•:« ;:
•-•••• r- •.;: •'•<• •,': til'-' ;:"N;n:ioj! .' t' • • r.
~-'.s '. -,u,d i.'.iinre to ;riv- t!"'.". T: - i • •,:,
v. tJ.ih:-Ko-.-e •'- O.K.R., 1^7 \V. Va. <,'•;
74 £'.K.2d 767 (1053).
Witnesses in r"_Mtion to ohscrvc but not
hearing .-finals are entitled to peculiar
wtight.—Where th» witnesses were in a
f .,-,va to observe with unusual cnr*
the circumstances surrounding the acvi-
dent, their testimony as to the negloct to
sound the customary warnings by K-:I ,,r
whistle, or both, within a rcasor.af...
distance from the crossing, a duty d:
the sounding »' the Whistle. The dc:;:cl
by the one amd the affirmance by the other
produces a conflict ol evidence, which s:
is the province oi the Jury to detern-.ir.'!.
Tav.ney T.'Kirkhart, 130 W. Va. 550, •>!
S.E.Sd 634 (1947).
Whether a conflict arises between jn=:-
tfve and negative evidence of thii
character depends upon the facts ar.:
circumstances of each case, from which :'.
may be deterniir.ed whether such r.e.r.v.: ••
ev-dsnce has any probative value. Caver.-
d;5h v. Chesapeake & 0. Ry., 95 \V. \'.>
4r-0, 121 S.E. 498 (1924>; Tawney v. K::-.-
hart, 130 W. Va. 550, 44 S.E. 634 -
v. .Yorf jik iM \\. Jly., luO U". X':.. •- '•'-'• •
S.E. 563 (1026).
Question of traveler's contriSi^
'Tor' jury.—Sec Ar:3T
v. Norfolk & AV. Ry., 127 W. Va. 310, 2
S.E.2d 634 (1944).
K\idcnce held insufficient to submit
railroad's neRliKunce to jury. — In action
for injuries sustained in crossing1 collision
evidence was insufficient to justify sub-
mission to jury of question of railroad's
r.epligence in failure to comply with this
ccctton. Baltimore & O.R.R. v. Deneen,
ICl F.2d 674 (4th Cir. 1947).
Evidence held sufficient to sustain ver-
dict for either -party. — -Conflicting evi-
dence on question of whether railroad
statutory warning signals required
by this section was sufficient on both
sides to have sustained a verdict in favor
of either party. Tawney v. Kirkhart, 130
W. Va. 550, 44 S.E.2d 634 (1947).
Evidence held to favor railroad's com-
pliance with section.—In Krodel v. Balti-
more & O.R.R., 99 V,*. Va. 374, 128 S.E.
S24 (1925), there was some conflict of
testimony as to sounding the whistle and
ringing the bell at a railroad crossing1, but
it was held that the weight was in favor
that the defendant complied with the
statute.
B-32
-------
ENCLOSURE C
MULTIDISCIPLINARY ACCIDENT INVESTIGATION
Case No. UC852D
(Abridged)
Prepared by
University of California
Los Angeles, California
The contents of this report reflect the views of
the performing organization which is responsible
for the facts and the accuracy of the data pre-
sented herein. The contents do not necessarily
reflect the official views or policy of the
Departrr^nt of Transportation. This report does
not coastitute a standard, specification or
regulation.
B-33
-------
UCLA COLLISION INVESTIGATION PROGRAM
VEHICLE COLLISION REPORT
Prepared for the U.S. Department of Transportation
National Highway Safety Bureau/ .
Under Contract FH-H-6690
Certain information contained in this report is obtained from indirect sources,
The opinions, findings, and conclusions expressed in this publication are the.
of the authors and not necessarily of the National Highway Safety Bureau.
B-34
-------
U. C. 852D
!• STANDARD CASE SUMMARY
1.1 SUMMARY TEXT
IDENTIFICATION: This train versus automobile collision occurred on a Thurs-
day at 10:51 a.m. at a combination intersection/railroad
crossing in California. Maximum occupant injury severity: critical (06) Collision
causation: driver inattention.
AMBIENCE: Day; weather clear and dry; roadway dry.
ROADWAY: A straight, asphalt, undivided roadway, 75ft. wide with
curbs, in a suburban area with speed limit of 35 mph. The
collision site is at a railroad crossing, 25 feet before a T-intersection. The road has a
negligible crown, and is upgrade at the site. The roadway has three intersections within
one-quarter mile of this intersection.
TRAFFIC CONTROLS: The lanes are separated by broken white lines with opposing
lanes divided by double-double yellow lines. There is a
railroad automatic signal and a traffic signal at the railroad crossing. There were no
crossing gates at the time of the collision. Four auto/train collisions at this sire in past 3 yrs,
VEHICLES: Vehicle ^1: Freight train weighing approximately 400 tons.
Vehicle ^2: 1967 Cadillac Coupe de Ville two-door hardtop
with pov/er windows and seat. No apparent defects. Collision damage to right door
causing intrusion of 12". Occupant contact with intruding door and train. Deformation
Index: 03RPMW2.
OCCUPANTS: Vehicle #2: Driver; 59-year-old female, height, 64",
weight, 160 ibs. Lap belt in use. No HBD or drugs. In-
juries: fractured rib, lumbar back strain, abrasions, and contusion.
Right Front: 63-year-old female. No restraint
in use. No HBD or drugs. Injuries: compound, depressed skull fracture with cerebral
cci.njsicn, abrasions and contusions over body.
DESCRIPTION:
Fr-" - j'Misicn: Vehicle *2, the Cadillac, approaching the T-intersection,
failed to stop at tho railroad crossing in spite of the warning
I' •'•!=. CTJ L_l!. Slowing for the red light at the intersection, the Cadillac entered the
• ic.'-.'j, in:o the path of the train. The train was eaitbound at approximately 15 rnph,
•M- ;., caching the crcosirig. The train engineer was sounding the whistle and applied his
bra''"c> '/•!••:;, h..> jj,v Pie Cadillac in crc-sing .
B-35
-------
',', C. 852 D
Collision: The train struck the Cadillac in the right side, pushing it 150
ft. along the railroad tracks. The Cadillac remained in a
position at a right angle to the railroad tracks. Occupants of the Cadillac moved to the
right, and the right front occupant was struck by the intruding train.
Post-collision: Occupants were hospitalized. Railroad crossing gates were
later installed at the crossing.
1 7 CAUSAL FACTORS, CONCLUSIONS, RECOMMENDATIONS:
Matrix cell Explanation
("indicates positive factor)
1 Driver inattention and/or distraction appear to be
the chief cause of this collision.
4 Air conditioning on, with windows rolled up, makes
it difficult to hear train or warning bells.
5 Right door penetration of 12lvdue to side impact.
Door metal torn in area of hinges.
5 It is recommended that integrated side structures
be employed, combining strength of frame, door
sill, body pillars and roof.
5* • Right /door latch and hinges did not fail.
7 Driver's view of oncoming train partially blocked
by shrubbery along tracks.
7 Vehicles were allowed to stop on railroad tracks
while waiting to turn at intersection.'
7 It 5s recommended that visibility of oncoming trains
be maximized by removing obstructions. Vehicles
should not be allowed to wait on railroad tracks.
8* Railroad crossing gate was installed and light
locations were altered after the collision.
B-36
-------
U.C.#852D
1.3
N
CROSS-
WALK -
OLEANDER
BUSHES
BUILDING
OLIVE TREES
WARNINGJXK
LIGHTS S ¥3
^JL
TRAFFIC LIGHT _ .
FEET
20 40 60 80 100
' V—==Sl
SCALE 1" =40'
VI-FREIGHT TRAIN
V2- 1967 CADILLAC COUPE DE VILLE
B-37
-------
IDENTIFICA j 1.
ENCLOSURE D
SOUTHWEST RESEARCH INSTITUTE
CASE SUMMARY
(MV-TRAIN-INTfcRSECTION COLLISION)
^ CasoNo. 7173
(Abridged)
This accident occurred at the MKT railroad grade crossing on Eisenhauer Rd. at IH35 in San Antonio,
Bexar County, Texas, on Thursday, September 30, 1971 at 1335 hours, involving the collision of a diesel
f"ipht ..tine and a 1970 four-door station wagon with a lone driver. The westbound automobile was
struck on its left side by the northbound locomotive. The area is residential. The accident was injury-
producing; AIS Seventy Code No. 3.
AMBIENCE
It was daytime with partly cloudy skies, 85°F dry bulb, 57 percent relative humidity, 10-mph breeze
blowing from the southeast, the road surfaces were dry and clear of debris and loose gravel.
HIGHWAY
Eisenhjuer Rd. is a major access artery between the interstate loop expressway system and the
residential areas of northeast San Antonio. It is a 41-ft-wide, four-lane, two-way roadway with an asphalt
surface of the intermediate type in good condition. The road is divided at this immediate area of the IH35
access road-Eisenhauer Rd. intersection by 6-m.-high concrete channelizing islands. The traffic lanes are
10 ft wide. Eisenhauer Rd. runs east-west and is bounded on both sides by a 6-in. curb. The road is straight
and level. It is not crowned. The coefficient of friction on the dry surface was 0.61. A southbound, one-way,
two-lane 24-ft-wide frontage road runs 60 ft east and parallel to a mainline, single track railroad right-of-way',
both intersecting Eisenhauer Rd. at this point. An exit ramp from IH3S is immediately north of this inter-
section and an entrance ramp is immediately south. These ramps connect 1H35 to the frontage road.
TRAFFIC CONTROLS
The posted' speed limit on Eisenhauer Rd. is 30 mph. The speed limit is 40 mph on the frontage
road. A railroad company-imposed speed limit of 25 mph is assigned for 0.5 mile each side of the crossing.
Traffic control devices consist of pavement markings, b-in.-high channelizing islands, regulatory, warning,
and guide signs. There arc two flashing amber lights, 3<>-in.-diameter yellow railroad advance warning signs,
and black-on-white railroad crossbucks. There are neither traffic control signal(s) in the area nor a flashing
red light or bell warning signals, gates, or guards to provide immediate warning of an approaching train.
VEHICLES
No. 1. 1968 GP40 Electromotive diesel freight engine. The 3-yr-old engine is considered to be in good
operating condition with no indicated defects. Minor secondary damage includes bent brakeman's steps,
bent coupling actuator lever, and airhose torn loose, secondary vehicle deformation index 12FDLW1. The
retail repair cost was nil.
No. 2. I ^70 Oldsmobile Vista Cruiser, four-door, three-seat, yellow station wagon; odometer reading
22,224 miles; valid Texas Motor Vehicle Inspection sticker with a damaged illegible date; equipped with a
standard .i50-cu in. t.".:iit-cylinder gasoline engine, automatic transmission, power steering, and power front
disc-type brakes, radio, healer, air conditioner, and tape deck, padded armrests, sunvisor, seat back tops,
interior rearview mirror, windshield, interbeam. and instrument panel. Three seatbelts and two shoulder
straps for front bench-type seat and three seatbelts for the second bench-type seat. The shoulder straps
B-38
-------
were in the stored position. No defects were apparent or indicated. The last vehicle maintenance w
performed, at 13,663 miles on Jjnoiry 21, 197) and included lubrication and oil and filter change. Priinji
contact damage was 16-in. sheet metal and frame deformation to the left side, primary vehicle defonrut-,
index Os)LPAW5. Secondary damage was to the tires, rear bumper, and root'. The retail replacement vak
was $3075 (total less S200 salvage value). ,
OCCUPANTS
Vehicle No. 1. Engineer: 46-yr-old white male, 71 in., 155 Ib (estimated). An interview was no,
obtained. He was familiar with the vehicle and the route traveled. !
Injury. None.
Vehicle No. 2. Occupant No. 02. Driver: 42-yr-olJ white female of Latin-American extraction, 62 in.,
132 Ib. She has been driving 20 yr and currently drives approximately 9000 miles/yr. She was en routj
from her husband's office to home, a distance of 10 miles. The accident occurred 1 mile from her destma
tion. She had no definite F.TA. She was familiar with the vehicle and with the rouie traveled. She has had
no formal driver's education. Her physical condition was excellent. Her precrash state was rested with no
stress, she was inattentive to her driving task. Lap and shoulder restraints were available, but not in use.
Injury Severe (not life-threatening). AJS Severity Code No. 3.
STANDARDS
The following Highway Safety Program Standards (HSPS) and/or Motor Vehicle Program Standards
(MVPS) were relevant to this case
HSPS No. 4-Dnver Education Use o] Occupant Restraints, Radio, and Failure to Look for T ri
HSPS No. ^-Identijican.tn and Surveillance of Accident Locations
HSPS No I 3 - Trj/jtc (. ntrol Devues
MVPS No. 201 -Occ pant Protection in Interior Impact
MVPS No. 214-Side Door Strength.
DESCRIPTION
Precrash: The driver of vehicle No. 2 (passenger car) was traveling to her home trom her husband's office.
She had left northbound IH35 and turned west onto 1-isenhauer Rd.. passing under the IH35 overpass. She
crossed the southbound frontage road at a relatively low speed (estimated not more than 25 rnr-h) anu
drove in front of vehicle No. I (diesel treight engine), which was moving north at about 25 mph with its
horn blowing for the crossing There were no skidmarks from vehicle No. 2 prior to impact. The car radio
was in operation.
Crash Impact occurred on the left side of vehicle No. 2, centered approximately at the "A" pillar line, as it
crowed I'll; railroad track in tront of vehicle No. 1. The coupler of the freight engine forced in the forward
portion of the door btructure. firewall, cowl, and instrument panel structure Other portions of the front
structure of tin- engine brakeman's steps and brackets- forced in the doors, floor, and frame left sidn J'l t
a depth of 1 (•> nuhe*. The passenger vehicle was pushed northward on the railroad right of way. It ther
yawed left and came to rest 88 ft from the impact point, parallel to and 1 ft west of the tracks fa ::•:.; th •
crossing. The unrestrained driver was first thrown left against the incjvmg 5nlc structure of the car. 1;,.« sii.1
was thrown to the rij.'.''U \Vhicle No. 1 stopped 314 ft from the point of impact.
Postcrjsh Hi.- driver of vehicle No. 2 was not ejected from the vehKlt She was removed from vehicle
No 2 through the right front door without complications. She was taken to the hospital by ambulance
B-39
-------
•• -oprrxinately 20 min afler the crash, fi cau\e the automobile came to rest a considerable disttnvr fiom
tfc • roadway, there was no appreciable latcilVrciue with (rafiic. A wrecker had no complications in picking
•i the vehicle and lowing it away. Since the locomotive wa:> not significantly damaged, it was able to
proceed. Traffic on FiNenhauer Rd. was estimated at 15 vehicles/min; on the frontage road, traffic was
estimated at r v,-i,•.;!:• 'mm.
CAUSAL FACTORS, CONCLUSIONS, AND RECOMMENDATIONS
M-jtsix ^el!
(* Indicates
Positive Factors) Explanation
1 Driver No. 02 was inattentive and did not observe normal precautions when approach-
ing the railroad track.
1 Driver No. 02 had her radio on and windows up, which may have prevented or
seriously interfered with her ability to hear the train's signal horn.
1 The engineer may have been speeding, with respect to the company-imposed limit of
25 rnph, 40 to 50 rnph. This is the situation if the train brakes were adequate and if
the engineer maintained a locked brake mode throughout the stopping sequence.
2 Driver No. 02 was not wearing the available seatbelt or shoulder strap.
3 Driving in a veil of interior noise (radio, air conditioner, etc.) with the windows closed
should be discouraged in driver education programs.
4 The train should have been capable of stopping within 104 ft from 25 mph. The 314-ft
stopping distance, from the point of impact, suggests that either the driver did not
fully apply the brakes at some point during the collision sequence or that the brakes
were not performing adequately.
•5 Occupant injuries from impact against interior surfaces and protuberances were miti-
gated as a result of adequate padding and interior design.
7 This site has an extremely high accident rate; however, more adequate traffic control
by a train-approach signal system has not yet been authorized.
B-40
-------
n
2
U
o
a)
i
t-
o
o
2
JE
O
f-
v UJ
5 *
2 i-
V)
z
I
e
a:
Z
O
o
-------
ENCLOSURE E
Maryland Medical-Legal' Foundation
Office of the Chief Medical Examiner
State of Maryland
Truck/Train Impact
Caae # MMF 72-24
(Abridged)
MJLTIDISCIPL1NARY ACCIDENT INVESTIGATION SUMMARY
IDENTIFICATION OF 'COLLISION
The highway is a state road traversing, north and south in the south-
east portion of an industrial section of Baltimore" County. The accident
occurred in September of 1972 'at 0400 hours on a Friday involving a trac-
tor trailer and a freight train at a front to side impact. The accident
caused fatal injuries to the driver of the tractor trailer.
INJURY SEVERITY SCALE: Driver of Vehicle #1 FATAL-AIS-8
AMBIENCE
Night; no illumination; misty; 58 degrees F.; 607. relative humidity;
wind 10 m.p.h. from the northwest; visibility of 500 feet; road surface
was wet; coefficient of friction .55 dry (measured) and .45 wet (estimated).
HIGHWAY
The highway on which the accident occurred is a. major arterial state
road with a total width of 106 feet consisting of two 12 foot lanes going
north and two 12 foot lanes going south divided by a 48 foot grass median.
The roadway is of black top macadam with an 8 foot shoulder on the east
side and a 2 foot shoulder on the west side. The roadway is straight and
level. There is no artificial lighting and within % mile there are. two in-
tersections; one being 800 feet south of the railroad crossing and the other
being 600 feet north. There are 9 telephone and transit poles within %
mile. The accident: history at this point within a year previous is 6 pro-
perty damage and 3 personal injury accidents with an average daily traffic
of 22,500 vehicles.
B-42
-------
TRAFFIC .CONTROLS
The speed limit is posted at 55 m.p.h. and there are intermittent lane
lines with solid edge lines painted in the roadway." There are standard
railroad crossing signs and lights at the right side of the road with over-
head signals actuated by the train. \^/
VEHICLES INVOLVED
Vehicle #1 was a 1969 G.M.C. Tractor, two-door, red in color with an
odometer reading of 49,760 miles. There is no inspection data but the
vehicle was well maintained by the company garage. The vehicle was equipped
with manual steering, manual transmission, air brakes (drum type), seat
belts (being used by the driver when the accident occurred). There was no
previous damage noted'. Damage to Vehicle //I on impacting the train at an
eleven o'clock principal impact force was to the left front causing a sheet
metal crush of 38 inches. The bumper, grille, fender and hood deformed
rearward into the engine compartment whereby the engine separated from mounts.
The left front wheel and assembly moved rearward. The seats moved forward
and the driver impacted the steering wheel and column with his chest and
his head impacted the left A-Pillar as it was deformed inward and rearward.
After the initial impact a second impact of 06 hours principal force occurred
as the trailer sheared from the fifth wheel and impacted the rear of the cab
with a sheet metal crush of 18 inches compressing the cab interior by 50%
pinning the operator in.
VEHICLE DEFORMATION INDEX: Principal Impact - 11 FLAW-4
Secondary Impact - 06 BDHW-4 .
Vehicle #2 was a General Motors E.M.D. type locomotive pulling 47 box
cars and it sustained minor damage to the right front side.
VEHICLE DEFORMATION INDEX: 02 RFMW-1
OCCUPANT,. DATA ' • . •
The driver of Vehicle #1 was a 46 year old white male, 68 inches tall,
weighing 115 pounds having 30 years driving experience at approximately
15,000 niles per year. At the time of accident he was enroute froa his place
of employment with a delivery for a distant city expected to arrive 5 hours
after the accident occurred. The accident occurred within 5 miles from the
origin. He was familiar with the vehicle and the area having used both daily
for the past several years. His physical condition was normal as was ". '.s men-
tal condi. ;ion. There was no alcohol or drug involvement and scat belts were
avaiicDl-j and in use by the operator. During the accident the driver sus-
tained the following injuries: fractures of skull^ ribs, pelvis and extremi-
ties, contusions of lungs with hemothorax, laceration of heart, laceration
of liver and spleen with hemoperitoneum, rupture of bladder; and contusions
of hippocampi and temporal lobe of brain. '(AIS-8)
B-43
-------
The dtl.'r of VeMcl" #2 (train) was a 57 year old white male, weight
-'nd eight unknown having <•'0 years driving experience with 15 years as a
riit-ioad engineer. His driving record is good with 10,000 miles per year
plus rail i:,«age undetermined. He is familiar with the engine using same
three to four times weekly. At the tiae he was shifting cars along the
riilroad from yard to yard. His engineering ability was taught to him by -
"• r-a-nroad c-irrpany. Tfv-re were no drugs or alcohol involved. There were
no restraint-i available and no -injuries. There were three passengers on
Hie trair -T\, they -'ere not injured or restrained. Passenger #1 was a
Viixte male, 56 years of age and he was seated in the front center. Passen~
ger #2 was a white male, 36 years of age and he was seated in the front right.
Passency- #3 was a white male, 54 years of age and he was seated in the rear
left.
STANDARDS .
1. FHSPS #9 - Identification and Surveilance of Accident Locations.
The railroad crossing is well protected with traffic signals ac-
tuated by the train, but it is so Little used that drivers attempt
to beat the train. It is recommended that gates be installed at '
the railroad crossing.,
COLL IS ION .DESCRIPT ION
Pre-Crash
The driver of Vehicle #1 reported to work at the usual time, 0130 hours,
and had proceeded froa the terminal to deliver a load of hardware to a dis-
tant city. He was operating the vehicle northbound on a state road at an
estimated speed of 45 to 50 m.p.h. and when he approached the east/west rail-
road crossing he failed to stop for the signals and collided with the right
front side of a slow rr.oving freight train. The freight train was proceeding
eastbound at an approximated speed of 8 to 10 m.p.h. There is no evidence
to show that the driver of Vehicle #1 tried to take any evasive action, how-
ever, the operator of the train did apply his air brakes for an emergency
stop.
Crash
Vehicle /.-I inipacted the right front side of the train with its left front
at an eleven o'clock principal force impact with a secondary impact force of
06 o'clock when the trailer sheared off the fifth wheel and impacted the
rear of the truck cab. As the vehicle rotated 25 clockwise, and coming
to rest 42 feet east of the impact, the driver, who was restrained, moved
forward and to the left impacting the steering wheel and the left A-Pil-
lar and was impacted from the rear by the cab body and seat.
Vehicle #2 was impacted at the right side at front initial Impact
force at 02 o'clock deforming the entrance steps and the hand rail. The
unrestrained occupants were well to the rear of the impact point and suf-
fered no effects of the accident. The driver of the train applied his air '
brakes for an emergency stop and the train remained on the rails coming to
a stop 168 feet east of the impact.
B-44
-------
Post- Crash
1 * <
i
Vehicle #1 came to rest 42 feet east of the impact facing east off the
roadway and Vehicle #2 cane to rest" IJS'feet east of the ircpact, on rails.
The operator and passengers of Vehicl; $2 were unhurt. The operator of
Vehicle #1,- due to the compression of the truck cab from the front and rear
impacts, was pinned in the cab. Emergency rescue equipment of the Police
and Fire Depart- ints were called, responding within 10 minutes and pro-
ceeded to cut tM netal attesting to free- the driver. Due to severe de-
formation, extrication was difficult and took two hours to free the driver.
He was pronounced dead at the scene and was taken to the Office of the Chief
Medical Examiner. During the rescue operation, traffic was tied up in both
directions and r-iitable detours were maintained by the police. A two com-
pany was contacted to clear the scene of the truck and debris. The truck
was towed to th -• terminal and the train was moved under its own power. The
scene was cleared and open for traffic within four hours.
CAUSAL FACTORS ^ CONCLUSIONS AND RECOMMENDATIONS
ACCIDENT C- IS AT ION
Matrix Ccl
Explanation
Primary Cause
Driver of Vehicle #1 failed to perceive
the approaching train and danger of going
through signals. (Definite)
Severity Increasing
Driver of Vehicle #1 made no attempt it
evasive action. (Definite)
Relevant Conditions
Driver of Vehicle #1 was apparently pre-
occupied with thoughts of his trip. (Pro-
bable)
The crossing was well protected with ac-
tuate'd signals (at side and overhead) but
it allows room for passage. (Probable)
INJURY CAUSATION
Matrix Cell
Explanation
/
Driver of Vehicle #1 was wearing available
restraints but they were of no use in this
case, (Probable)
The collapse of Vehicle #1 from front and
rear impacts added to severe injury. (De-
finite)
B-45
-------
POST-CRASH TACTORS
Matrix Cell Explanation
Ambulance and rescue arrival within 10 tail.
utes, but extrication was difficult taking
two hours with netal saws. (Definite)
The load of Vehicle #1 shifted after the
initial impact. (Definite)
There were no fires or erolosions, detours
were set and r.aintair.ed adequately, and the
clean-up operation took four hours. (Defi-
nite)
B-46
-------
fflFDOL- LFG/U
I
attmc
B-47
-------
MFD/OH.- LFGRL
I
JHRTRIC PdLES
-€r
fiCTURTED
VFSH£flJ
SIGNALS
72-2
HCCIDFITT
B-48
-------
ENCLOSURE F
Durham City Code
Durham, N.C.
Ch. 18 § 9 Locomotive Whistle.
It shall be unlawful for any person to blow or allow to
be blown any locomotive whistle under his control within the city
limits. (Code 19^0, c. 28, § 8.)
Knoxville City Code
Knoxville,_Tenn.
Ch. 33 "§ 8 Blowing Whistles.
It shall be unlawful for any person operating or in charge
of a locomotive engine within the corporate limits of the city
to blow the whistle on the same except as may be absolutely
necessary in the use of the signals as laid down by the rules
and regulations of railway companies, or as required by the
laws of the state. (10-21-04.)
Houston City Code
Houston, Texas
Sec. 18^3 Blowing Whistles; Blowing out Boiler
All persons are prohibited from blowing any whistles on
any locomotive, or single blasts therefrom, within the limits
of the city, for a longer period of time than five seconds,
except when there is imminent danger of an accident. All
persons are prohibited from blowing off or blowing out a
B-49
-------
bonier when crossing any public street or other thoroughfare
within the limits of the city. Each and every person violat-
i ,? any provision of this section shall be fined in any sum,
upon conviction, not less than five dollars and not exceeding
fifty dollars.
Mas on_City, Iowa
26-29 Sounding of Locomotive Whistles
It shall be unlawful for any person to cause or permit
any locomotive whistle to be sounded within the limits of the
City except for the purpose of making necessary signals
required by law or required for the safe operation of the
railway, and where requisite signals cannot be made by other
means. (R fl6, Sec. 5^5.)
Chicago, Illinois
188-44. No person owning or operating a railroad shall cause
or allow the whistle of any locomotive engine to be sounded
within the city, except necessary brake signals and such as may
be absolutely necessary to prevent injury to life and property.
Each locomotive engine shall be equipped with a bell-
ringing device which shall at all times be maintained in
repair and which shall cause the bell of the engine to be rung
automatically. The bell of each locomotive engine shall be
rung continuously while such locomotive is running within the
city, excepting bells on locomotives running upon those
railroad tracks enclosed by walls or fences, or enclosed by a
B-50
-------
wall on one side and public waters on the other side, and
excepting bells on locomotives running upon those portions of
the railroad track which have been elevated. In the case of
these exceptions, no bell shall be rung or whistle blown except
as signals of danger.
Buffalo, New York
Chapter V. RAILROADS
#4. It shall not be lawful for any person in the employ of
any railroad company operating within the limits of the city
to permit the whistle of the locomotive under his control to
be blown, except for necessary signal purposes. Any person
violating the provisions of this section shall pay a penalty
of $25.00 for such offense.
NOTE: This restriction is generally associated with a train
speed restriction of 6 MPH and the use of flagmen.
Lynchburg, Virginia
CITY CODE SUPPLEMENT (Railroad)
Sec. 3809. Sounding whistles or horns.
The sounding or blowing of locomotive whistles or horns
within the corporate limits of the city of Lynchburg is hereby
prohibited, except as may be necessary for the transmission
of signals or in emergency to prevent accidents.
The provisions of this section shall not apply to the
two crossings of the tracks of the Chesapeake and Ohio Railway
B-51
-------
Company at Reusens, in the vicinity of the E. J. Lavino
Company, because of the lack of sight distance and warning
-cea at these crossings.
An" 1. lit1 . ii of thi; ordinance shall be punished by a
fine of not less ohan five dollars nor more than ten dollars
for each offense. (1931, §704; 6-8-42; 8=28-56; 10-9-56)
State of Illinois
Under authority delegated to it by the State Legislature
(11^-59), the Illinois Commerce Commission adopted General
Order #176 on August 15, 1957, excusing the sounding of horns
and whistles at crossings protected by flashing lights. This has
now been incorporated in General Order No. 138, Revised, August
22, 1973, Rule 501.
State of Florida
§351.03 limits signals to bells only in incorporated areas, with
an accompanying speed limit of 12 mph.
B-52
-------
I ^^ABOBCS'.
\J^r^ TO TH
V CALIFORN
ALL COMMU.«IL ATIOI.!.
COI-'M. ,IONLRS
• NONL nroRoeoN. r,..,»««Y /Ffc^r'."' . -• \ //N/~ TO THE COMMISSION
LtlAM SYMON», JK. Vl\~'vL^••'.'.'•/.- \V CALIFORNIA STATE i
P. VUKASIN, JH. ' XNfTs/"* SJ \ BAN FRANCIOCO, CALII 0.0.IA ..4101
DMAS MOHAN >i-V, '?:^/ A T*nr«0«n (*IO) 057-
W, HOLMU
STATE OF CALIFORNIA
November 10, 1972 F.LCNO. JQ 79403
Honorable Arlen Gregorio
The State Senate
12th District, San Mateo County ^,n
State Capitol ,- '^
Sacramento, CA 95814 (A\ \ D
Dear Senator Gregorio:
Subsequent to receipt of your letter of October 4, 1972, our representative
has discussed the use of train whistles approaching railroad grade crossings
with Kr. John Gilroy and Ms. Charlotte Schultz of your staff.
As discussed with them, it may be necessary to sound the train whistle
even at crossings equipped with automatic gates for the following
reasons:
1. Possibility of a malfunction of the automatic grade crossing protection
due to being struck by vehicles, vandalism or failure of track circuitry
or signal apparatus.
2. Rail highway crossings are frequently traversed by bicyclists and
pedestrians after the protective devices have been actuated by an
approaching train.
3. Impatient motorists sometimes ignore crossing signals and have been
known to drive around protective gate arms in an attempt to avoid
being delayed by a train.
4. Liability on the part of the railroads for failure to use every means
available to avoid an accident.
In view of the above, the staff feels that in the interest of safety, the
railroads should not be prohibited from using the train whistles to warn
persons that a train is approaching.
Yours very truly,
PUBLIC UTILITIES COMMISSION
WILLIAiM R. JOHNSON, Secretary
(
, Secret
B-53
-------
APPENDIX C
OPERATING RAILROAD RETARDER
YARDS IN THE
UNITED STATES
-------
OPERATING RAILROAD RET ARDER YARDS IN THE UNITED STATES
(CLASS I Railroads)
State
Alabama
Arkansas
California
Colorado
Connecticut
Florida
Georgia
Idaho
Yard
Birmingham
Birmingham
Sheffield
N. Little Rock
Pine Bluff
City of Industry
East Los Angeles
Los Angeles
Richmond
Roseville
West Colton
Grand Jet.
Pueblo
Cedar Hill (East)
Cedar Hill (West)
Tampa
Atlanta
Atlanta
Atlanta
Ma con
Pocatello
Railroad
L&N
Sou
Sou
M. P.
St. L. S. W.
S. P.
U. P.
S. P.
S. P.
S. P.
S. P.
D&RGW
AT&SF
P. C.
P. C.
S. C. L.
Sou
Sou
L&N
Sou
U. P.
Number of
Tracks
40
56
32
64
30
12
16
40
8
49
56
31
16
45
38
8
12
65
24
50
40
C-l
-------
State
Illinois
Indiana
Kansas
Kentucky
Louisiana
Maryland
Massachusetts
Yard
Bensenville
Blue Island
Chicago, Clearing
(East)
Chicago, Clearing
(West)
Chicago, Cicero
Chicago, Corwith
Chicago, 59th St.
E. St. Louis
E. St. Louis
Galesburg (East)
Galesburg (West)
Madison
Markam
Markam
Proviso
Silvio
Elkhart
Gary
Gibson (South)
Gibson (North)
Indianapolis
Argentine (East)
Argentine (West)
Armourdale
DeCoursey (North)
DeCoursey (South)
Russell
Stevens
Geismer
Cumberland (West)
Cumberland (East)
Boston
Railroad
C.M.S.P.&P.
I. H. B.
B. R. Chgo
B. R. Chgo
B. N.
AT&SF
P. C.
A. &S.
I. C. G.
B. N.
B. N.
T. R. R. A.
I. C. G.
I. C. G.
C. N. W.
C. R. I. P.
P. C.
E. J. &E.
I. H. B.
I. H. B.
P. C.
AT&SF
AT&SF
C. R. 1. P.
L&N
L&N
C&O/B&O
C&O/B&O
I.C. G.
C&O/B&O
C&O/B&O
B&M
Number of
Tracks
70
42
44
36
43
32
42
42
26
49
35
34
64
45
59
50
72
58
30
30
64
48
56
40
20
24
32
15
6
32
16
22
C-2
-------
State
Michigan
Minnesota
Missouri
Montana
Nebraska
New Jersey
New York
North Carolina
North Dakota
Ohio
Oklahoma
Yard
Detroit
West Detroit
Minneapolis
St. Paul
Kansas City (hast)
Kansas City (West)
N. Kansas City
Missoula
Lincoln
N. Platte
N. Platte (West)
Morrisville
Pavonia
Buffalo
Buffalo
DeWitt
Mechanicville
Hamlet
Minot
Bellevue
Columbus
Grandview
Marion
Portsmouth
Railroad
DT&l
P. C.
B. N.
C.M.S.P.&P.
M. P.
M. P.
B. N.
B. N.
B. N.
U. P.
U. P.
P. C.
P. C.
E.L.
P. C.
P. C.
B&M
S. C. L.
B. N.
N&W
P. C.
P. C.
E. L.
N&W
Portsmouth (West) j N&W
Sharonville P. C.
Stanley P. C.
Walkridge C&O/B&O
Willard j C&O/B&O
Tulsa
S. L. S. F.
Number of
Tracks
36
31
63
40
42
32
42
9
36
62
42
38
32
56
63
27
36
58
40
42
40
9
24
18
35
35
42
68
52
40
C-3
-------
State
Oregon
Pennsylvania
Tennessee
Texas
Virginia
Washington
Yard
Eugene
Allentown
Connellsville
Conway (East)
Conway (West)
Enola (East)
Enoia (West)
Pittsburgh
Pittsburgh
Rutherford (East)
Rutherford (West)
Chattanooga
Knoxville
Memphis
Nashville
Beaumont
Fort Worth
Houston
Alexandria (North)
Alexandria (South)
Bluefield
Lamperts Point
(empty)
Lamperts Point
(loaded)
Lamperts Point
Newport News
Roanoke
Pasco
Seattle
]
Railroad
S. P.
CNJ/LV
C&O/B&O
P. C.
P. C.
P. C.
P. C.
U. R. R.
Mon-Conn.
Reading
Reading
Sou
Sou
S. L. S. F.
L&N
S. P.
M. P./T. P.
S. P.
R. F. P.
R. F. P.
N&W
N&W
N&W
N&W
C&O/B&O
N&W
B. N.
B. N.
Number of
Tracks
32
19
15
54
56
33
36
23
22
33
18
50
46
50
56
12
44
48
49
39
13
36
36
30
15
56
47
16
C-4
-------
State
Wisconsin
Yard
Milwaukee
Railroad
C.M.S.P.&P.
Number of
Tracks
35
Abbreviations of Railroad Names Used in this Table*
L&N - Louisville and Nashville
Sou — Southern
M.P. - Missouri Pacific
St. L.S.W. - St. Louis Southwestern
S.P. - Southern Pacific
U.P. - Union Pacific
D&RGW - Denver and Rio Grande
Western
AT&SF - Atchison, Topeka and
Santa Fe
P.C. - Penn Central
S.C.L. - Seaboard Coast Line
C.M.S.P.&P. - Chicago, Milwaukee,
St. Paul and Pacific
I.H.B. - Indiana Harbor Belt Railway
B.R. Chgo - Belt Railway of Chicago
B.N. - Burlington Northern
I.C.G. - Illinois Central Gulf
A. & S. - Alton and Southern
T.R.R.A. - Terminal Railroad Assoc. of
St. Louis
C.N.W. - Chicago and North Western
C.R.I.P. - Chicago, Rock Island and Pacific
E.J. & E. — Elgin, Joliet, and Eastern
C&O/B&O - Chesapeake and Ohio
Baltimore and Ohio
B&M — Boston and Maine
D.T.&I. - Detroit, Toldeo, and Ironton
E.L. - Erie Lackawanna
N&W - Norfolk and Western
S.L.S.F. - St. Louis San Francisco
CNJ/LV - Central Railroad of New Jersey
Lehigh Valley
U.R.R. - Union Railroad
Mon-Conn. — Monongahela Connecting
Reading - Reading Company
M.P./T.P. - Missouri Pacific/Texas Pacific
R.F.P. — Richmond, Fredericksburg and
Potomac
These abbreviations reflect mergers; the abbreviations on the accompanying map frequently
do not reflect mergers.
C-5
-------
APPENDIX D
SUMMARY OF YARD
NOISE IMPACT STUDY
-------
SUMMARY OF YARD NOISE IMPACT STUDY
INTRODUCTION
The rail yard modeling study of noise impact on people used data collected at the Cicero
Yard of the Burlington Northern near Chicago Illinois. The study included the analysis of eight
railroad yards from a population density and yard layout standpoint which led to the selection of
the Cicero Yard for more detailed analysis. Characteristics of the noise emitted from the Cicero
Yard under a range of operating conditions were studied and a model of the yard was developed.
The model was then used to predict the impact on people (environmental noise levels) of various
noise abatement activities on different aspects of the Cicero Yard operation.
CASE STUDIES OF RAILROAD YARDS
Eight yards having a wide range of characteristics were selected in order to compare yard
traffic with population densities near them. Such a comparison provides a basis for determining
the number and frequency of exposure of people to noise from railroad yards. Figures D.I - D.8
are maps of the yards that were studied. Although no detailed studies of the zoning around the
yards were attempted, the maps provide some indication of land use. The configuration of the
yards and the traffic through the yards were determined by telephoning the yard superintendants
or the yard masters. Table D. 1 summarizes the population and traffic data for the yards.
The population information was taken from the 1970 Census of Housing, Block Statistics for
each city. The total populations for the cities studied were obtained from the 1970 Census of
Population, U.S. Summary. Population densities were derived for strips 250 or 500 ft wide for the
entire length of the yards and/or for a total of 2000 ft from the retarders. Often, separate popu-
lation density estimates were made for each side of a yard, since people are not evenly distributed
around yards. Figures D. 1 - D.8 contain graphs of the population distribution for each area.
The population of the cities in which the yards are located ranges from 67,058 (Cicero) to
1,800 (Roseville). Population cannot be considered an index of urbanization since all of the towns
are in urbanized areas generally outside a larger urban city. No yard located in a "rural" area was
studied as sufficiently detailed population statistics were not available for a yard located in other
than urbanized areas.
STATISTICAL ANALYSIS OF NOISE NEAR RAILROAD YARDS
Many methods of describing community noise have been proposed, studied, and evaluated, but
the most suitable method for describing environmental noise and its effect on people, in EPA's
D-l
-------
8
* o
r •§ §
-------
DISTANCE FROM
YARQJOUNDARY
(feet)
2000
1500
1000
500
Streets
Railroad Tracks
1000
2000 feet
Both Sides Averaged Together
FIG. D.2. MAP AND POPULATION DENSITY PROFILES FOR THE ELKHART,
INDIANA HUMP YARD.
D-3
-------
Q
o:
s:
o
LU
LU
>-
a:
o
Li.
u_
o
o:
a.
o
2=
o
I—I
F—
«=c
Q-
o
a.
o
a.
-------
o
QC
IE
±£
CXL
UJ
or
o
CO
UJ
u_
o
o;
a.
a.
O
o.
Q
z:
•a:
Q.
H
D-5
-------
a
ac
Q.
s:
ID
re
to
i—i
o
OJ
QC
o
LU
I
I—
O
u_
>-
I—
I—I
oo
LU
O
O
1—4
h-
-------
Q
cc
•a:
UJ
•X
a:
o
OL
O
O
a:
a.
UJ
o
o
t—I
I—
«t
a.
O
Q.
vo
•
Q
D-7
-------
01
o. —
CD
Q
CQ
_i
-------
o-
on
o
ot:
o
CO
U-
o
QE:
a.
co
z
LU
O
UO «
<0
— O O O O O
O O O O
O »O O "">
fSJ — ^~
O O O Q
0000
>O O ""> O
Q.
O
Q- Q
Q£
Q -
•a:
a.
a. -s.
CD
•
Q
H
CM
D-9
-------
o GE:
Q
o
o;
s
>•—;
«=t
n.
o
O
oe:
O
j —
o
Q-
r- -j
I
f
\
1
i
(
j
j
!
•*•-
t-
i™
i O f
'
f
, j
I
|
1
r -"—• *
i - I
o
C'J
o
CNJ
I
o
o
JZ
i ** o i
J* 0 1
m
« r-
r- (
i o
o
Of O
«3
»y> O
O
l» 0
01 --
0- 1
O O
r—> O
d. U°>
fflj
£i. -
c.
•••- o
0 «/>
c o
^ un
OJ
!
; O
I*
0
i*
^0
c -
.a
£
c
o
I ''' m
' i" ^
; tx
o
O-
h'
~"
t' t-
i's *"
i - j_
J V *'
*o "^
-- t?
r - ,
..<
•V
.,,.
\
'"'
«i*
.iL
O '"
-H "•-!
OJ "J
P) CO
x: r
t! n
,3 '.
\O CJ
rH I/N
CO ^3-
^ r-<
O '-O
O C*-
VO f\J
IX LfN
,'j ,_!
IA O
O O
rH rH
CO Cg
vo" 0
rH P,
on co
CO P~i
r: °.
r'
o
o
or)
(.j
C^"
"•O
i -:
...
i
I ; ,
I
* ;^
i , - <
' '
«
\ J >
1
""•i
•' ' £<
' i> v.
-i e,
•i '7' t '
r" r t-
-NJ v," .H
CO
,_r
on
VD
O
f-
LT
OO
f\.
"
O
o
CO
I
T
C\j
jv"
-7
!
1
iri i*i
i (-
j >
if, •
„_
[! t'.
^ /,
'
•-. F
, , 5
r- , '
T^ 14 '
-^ ^1
£,
U, 'I-
C G
O O
C. 0
^ c
ti ;.
0 0
0^3 -^
_
1
yp rH
-3- -3-
cr- iA
rH U",
m fn
CO O
o> co
O fNi
on ao
o^ o
•^T
oj -=r
IT* r-'
o
o
i-\,
-4
-3
"i "• '
i '*- '"
f. (!
0' ' '
i. M
(1; f
ii> «-
j- l
1 ,'
';•
''(" 17)
~ 111 HI
4> u. V.
,H n 1
" " , 'l
>" '" V
.': o i'
,
• " I,
!0 J> -j> •»
V. ' ' --. J .
v* ,i r1 "T1
E.' iT. i-t S-»
0 J> *'
L'N ^ (D fl)
-J- +* t. t.
d c
o o
4J 4^
0 0
rt n
cd a>
u ^
...
vo <^o
o t—
a- co
ON nn
rH t~-
rH o^
h- rH
-^- rH
-
rn
=r >o
E — -A
'"* "^
•n
a
o
^r
T
^
!*-!
r--
-i
r,
tn J>
(, n ""
1' f« d
v. r is
1 ' ' ( r I
? , ^'
f«'l
tJ. 1 f '
l' "
.*' *"
iT* 1
,3 d» f) f,
'1 ill C) . 1
'M t I 0[t
aj (.-> ^
^ d FH e
c c
o o
o o
o flf
>.
i.1 wi
i i i > (,
.^ ij j>
r (H C
0 rnrH
C C
O 0
•H -H
4-> +J
CO CO
3 SH
W 2
rH
on
e--
ry
m
rg
o
rH
ry on
rH CO
O IA
o co
rH
ON CO
CO CO
rn* IA
rH CO
t*~ rj
o> o
rH ^T
O
O
CO
r—
rvi
vo*
>I
Z*
. QJ
f-H -r <
>
* (
c c:
rtl O
i. *'
0< O
;-. en
in*
"!1
i',
m
^
ti.
^ £!
'i ^'
^i
a) t, ^->
ifl a
*H .-> I,
•i' tli
R* t, (1>
w a> c:
V 1 > A'
»*fl E^J C* t^
t. r. t: o
iJ "H ti
at (^
oo c: o -H
VO 0 C <*
C C
0 0
o o
0) 4)
«? ca
c c
V 9)
CD (O
-
_ry VO
ry
o
o
LA
CO
O
^
0
O -H
*H X
x: o
rj
fli C>
T) O
-^ t:
t, »-i
A) *'
(fl (P
r< m
tn
j, di
ri f.
V« i?
»' 4-
fct
t]
S C
WI -H
i: t< T
,« t, Hi
*-i tO tl
*4» 1
cn to Li faO
t > i ; o. ^H
L c o a
4-> t. W
0 60
ON ?! ON
^r ^r-^r
n n
l« (U
c c c*o c -o
o o o t- o t-
«H -H *H C1) «H <8
O t< t> T3 O 0) O 0)
co nJ w d W co
>> >> 4> 0)
na» nii> n^H M«H
alb L. 0) 3 O t, O
CO Z W 3B
rH lA O> VO
ON CM CO ^
O rH 1*^ lf>
CM f\* OJ
ry t»- (M m -
rH t— ^T tn
CO ON ^O O
H rH -*
co o no f—
t*— vo vo ^
rH ON Oa ON
on cvi r-t vo
ON ON rn jy
MA "1
rH rH CM
CT\ 0 O VO
IA ON f— C—
ry vo rH CNJ
rH rH
O «
O
IA
IA
cr*
CO
fj1
<-H
vl t)
rH t-t
(11 ^H
O O
m
* a.
rH I,
., 1 0)
r* jf.
W 3
O £1
ii, t-j ,.
-------
judgment, is the day/night sound level (re: Levels Document). Ljn may be obtained from an
analysis of statittical records of noise (Schultz, 1972). Details of this procedure are in enclosure A
of section 8 of this document. "Time records" usually means magnetic tape recordings made at
the measurement site with rugged, portable, high-quality tape recorders. Permanent recordings
permit processing a given noise record in several different ways, freeing the investigator from the
restrictions imposed by the particular analysis that might be suitable in the field.
Figure D.9 shows portions of a time history of noise measured around 5:00 a.m. near resi-
dences about 400 ft from the boundary of a railroad yard. The record from which Figure D.9 was
constructed was produced by playing a magnetic tape recording of the noise through an A-weighting
network into a graphic level recorder. The figures show some significant noise events that are not
associated with railroad operations. Those events must be iliminated from statistical analysis of
the information on the tapes if the results are to be descriptive of railroad noise only.
An edited tape, from which all non-railroad noises were removed, was prepared by selectively
interrupting a re-recording of the original tape. Both the unedited and the edited tapes of railroad
noise were processed using an electronic statistical analyzer and a digital computer, to produce
statistical analyses like the one shown in Figure D. 1 Oa. The tape which was generated is shown in
Figure D.9. Figure D. 1 Ob shows the result of a statistical analysis of the edited version of the tape
that generated Figure D. lOa. The solid lines in Figure D.I Ob represent the data from Figure D.lOa.
Figure D.I Ob shows that editing out extraneous events did not cause large changes in the
statistical properties of the recorded noise, and the effect is typical of cases for which editing was
possible. For times when the community was active, it was impossible to discriminate between
noises due to railroad operations and other noises.
Figure D.I 1 shows the results of a statistical analysis of an edited tape recording of noises at
the boundary of a busy yard. Even though a few diesel trucks traveled along a street adjacent to
the boundary, editing the recorded sounds produced negligible changes in their statistical properties.
Figures D. 12a and D. 12b demonstrate a contrasting situation. Figure D. 12a shows the
results of statistical analysis of an unedited tape recording of noises at the boundary of the yard
described above during a period of relative inactivity. Since much of the noise in the vicinity was
extraneous (mostly diesel trucks), editing changed the statistical properties of the recorded noise.
Figure D. 1 2b shows the effect of editing this tape. Even though there were few readily noticeable
railroad noises during the period covered by Figure D.I2, the continuous background noise is
higher at the boundary of the yard than in the community, illustrating the contributions of
continuously idling locomotives and other noises associated with the activities of men and machines
assigned to the yard.
"Energy Mean Level" is one of the parameters shown in the computer listing - in Figures D. 10
through D. 12. That parameter, usually called "LEQ" is the level of the continuous sound that
D-ll
-------
J
o H
o
o
-
CC
o
I—
OO
cc
<
oc
t
CD
OC.
o o
Qi
> o
LU CO
o
LU
—I O
LU a:
oo u_
O
(O
2
3000*0
o
10
(V)9P] 13A31 QNOOS 11VW3AO
0
H
En
D-12
-------
GRAPHICAL OUTPUT OF STATISTICAL KOZSC DATA
CICERO IAPD, KAY 17,1973, 5120 A,II,, VEST 3PTH SI,
1
99.9
99
P
B 90
R
C
E
N
T
I 50
L
E
I.
E
V 10
E
L
S
1
.1
.01
I i i I i I I i ' i 1 1 I 1 I i
• • M • i 4
1
1
I
1
J
1
I
1
1
1
• ,
1
• ,
1
',
1
1
1
1
1
I ,
•
• ,
I §
1 ,
1
1
1
1
» _
*l
>3
2
1
8
X
06
M
A
.1
,2
.3
h
050 (!55 C60 065 070 075 080 085 169$ 095 100 105 110 115 12(9
CUMULATIVE DISTRIBUTION
MAXIMUM KOJSC LEVEL • 82,5
MIUIMUn KOXSE LEVEL « 31,3
KOISE J-OLI.UTIPK LEVEL » 71,1
STAHTARD DtVIATlOK » U,9
ENERGY KtAK LEVEJ, • 58,6
FIG. D.lOa.STATISTICS OF NOISE IN A COMMUNITY 400 FT FROM THE
BOUNDARY OF A RAILROAD YARD (TOTAL NOISE; UNEDITED
TAPE).
D-13
-------
OUiP'-'t OF SUXlSIXCAi NCISE PftlA
CICERO TARD, HA* 17,1973, 5120 A.K,, WEST 3BTH SI,
(EDITED)
* I i I t 1 I I i i « I « I I
«*,»
99
P
E 90
R
C
E
N
T
Z 50
L
C
I
I
V 10
F
I
S
1
.1
,01
0
_____ <
t
> 1 1 i
* • i
J
1
I
I
!
J
1 *
\
I
J
1
1
1
J
1
I .
|\
| \
t \
' • ^v
1 , >w
'', \ A TOTAL NOISE
i, \ / (UNEDITED TAPE-
RAIL YARD '',,X FROM FIG. o.lOa)
NOISE ONLY / ' . V
(EDITED TAPE I—7 ".X
1 1^
•
•
40 0U5 050 055 060 065 C70 075 0B0 085 090 095 100 105 110 115 1'
CUMUI.ATZVZ PZSTRIBVTION
KAXIMUW NOISE J.EVSL « 80,0
MINIMUM NOISK J,EVEL » 31,3
KO:i.E POiLUTiOH tEVEi " 6H,3
S1ARDARD DEVmiOH " 3,3
EKbEG* MfcAU tEVE^ e 55,6
•J
2
1
s
z
0G
M
A
•• 1
*2
.3
|r
20
FIG. D.lOb.
STATISTICS OF NOISE IN A COMMUNITY 400 FT FROM THE
BOUNDARY OF A RAILROAD YARD (COMPARISON OF EDITED
AND UNEDITED TAPES).
D-14
-------
GflApHICAL OUTPUT OF STATISTICAL iiOlSE DAIA
CICCPO YAH>, MA- 17,1973, 5M6 A.M.,
(EDITED)
AVE,
99.9
99
P
£ 90
K
C
E
N
T
I 50
L
E
• I i i t i (
l I I I i l i I i I I I I I | I I 4 I I I
I
I
I
I
i i
I I I I I
. 1
I
I
I
'•
I
I
-b 070 i
CUI1ULATIVE
100 105 110
I
+ 3
t
4-
:2
i
i s
: i
: M
: A
-3
woist LEVEL « 97,5
MINIMUK NOJSt IEVEL » 65,0
HOISj; PPJ.IUXJ.OK LEVEL » 82.6
STANDARD DEVOTION « 3,3
MtAN LlVft B 71,?
FIG. D.ll.
STATISTICS OF NOISE AT THE BOUNDARY OF A RAILROAD YARD
WHILE THE YARD WAS BUSY (RAILROAD NOISE ONLY; EDITED
TAPE).
D-15
-------
CAi. OUTPUT v STf>* If.TICAL 1'OlSE DATA
CICERO VAi'.D, «AJ 17,1973, 3:3i1 A.M., OGDEW AVfi,
99.9
99
P
E 90
a
C
E
N
T
I 50
L
E
1
E
V i0
fc
L
S
4
• 1
.01
1 ' » 1 t • 1 I i « 1 1 1 1 1
• « 1 ' ' ' « « ' ' « t • 1 M ! 1 i 1 1 i 1 » « 1 I 1 I 1 i <
I
J
I
' « 1
1
1
1
1
', <
J
1
1
«
1
1 4
1 i
J ' !
1 1
'. :
I :
i <
!
:
!
:
1 4
f i
1 1
1 :
1 :
1 i
1
I :
i
t-3
!
>
i
I
I
k
12
t
1
*
I
k
11
I
I
: s
X
>PG
H
K
-1
i.
-2
k
^3
k
070 (575 060 085 tf9J H9b 1B0 105 110
CUKULATIVE DISTRIBUTION
Kcusi; IEVEI. « 96, e
t-OlSL tKVEi « 77, S
SE VOLLUTiOtl LEVEL » 90,7
' DEVIATION * 2,7
MtAN LEVEL «» 83,9
FIG. D.12a. STATISTICS OF NOISE AT THE BOUNDARY OF A RAILROAD
YARD WHILE THE YARD WAS QUIET (TOTAL NOISE; UN-
EDITED TAPE).
D-16
-------
GRAPHICAL OUTPUT OF STATISTICAL NOISE DATA
CICEPO USD, HAy 17,1973, 3J30 A.M., OGOEN AVK,
(EDITED)
1 ' ' « 1 • • 1 1 1 1 1 , , , ,
* * « y
99
P
E 921
R
£
N
T
I 50
E
L
E
V :n
E
L
S
1
. 1
.1-1
0«
w
*3
J
«
f
t
*
J2
t
t •
. r TOTAL NOISE :
/ (UNEDITED TAPE- ti
PIG. D.12a) i
*
9
1 f s
1 : I
*0G
J H
I A
1
; ,
• i !
i L *
• i •
I k \
1 S
1 1 *
RAIL YARD j' \ «'2
NOISE ONLY /. \ *
(EDITED TAPE)-/ I \
•' ^ '
• 1 4,_3
' 1 J
" \ :
'"•• 1 :
M'05b'o.s0'a55*f'.LVt'.;,5*K7rr74*;^<1iir^:,';srn:-:ji5': •••.•••?*
CUMULATIVE DISTRIBUTION
M^.XIKUtt KOlSi;
MIKIKUn MDlSfc tEVEL
NOIS}; POLLUTiON IEVU,
STANDARD DtVlATIOW
EKKEGY Ki,AN 1.EVE1,
PF.HCEMILE
133.3
150
L90
t99
FIG. D.12b.
62,5
6«,0
1.6
67^14
66.5
63,5
63,3
«2.7
62,5
U7,9
STATISTICS OF NOISE AT THE BOUNDARY OF A RAILROAD
YARD WHILE THE YARD WAS QUIET (COMPARISON OF
EDITED AND UNEDITED TAPES).
D-17
-------
•fr
CICERO YAKD
MAY 17,1973
70
f™ — -I
o
CNJ
ffi
-o
O
Lu
55
50
/\ :
OGDEN AVE
(AT THE YARD
BOUNDARY)
W. 30 TH (ABOUT
400 FT. FROM THE -J
YARD BOUNDARY) -|
L^^^L^— I
TIME OF DAY (HOURS PAST MIDNIGHT)
6
FIG. D.13. MEASURED LEQ VS TIME OF DAY FOR POINTS IN AND NEAR A
RAILROAD YARD (20-MIN RECORDINGS, SAMPLED 10 TIMES/
SEC),
D-18
-------
would be associated with an amount of eneigy equal to the sum of the energies of a collection of
discontinuous sounds. The discontinuous sounds are analyzed for a specified period of time, and
LgQ is calculated for that same period. Figure D-13 shows plots of the computer-calculated Lgq's
for the observations described above.
MODELING YARD NOISE IMPACT ON PEOPLE
The two types of raihoad switching yards are flat yards and hump yards. In a flat railroad
yard there are two major sources of noise - locomotives and car impact. In hump yards the squeal
caused by cars passing through rctarders is significant.
The development of a yard noise model lor this Background Document involves the computa-
tion of L£>N* for yards which (1) describes the activities of locomotives, (2) determines the
probabilities of occurrence of various levels of retarder squeal and car impact noise, and (3) inte-
grates the cumulative acoustic energy that is developed at a given point in the space surrounding
the yard.
Figure D. 14a shows calculated Lj^ profiles for group retarders in a typical yard — the
Cicero Yard in Chicago. Figure D. 14b shows Lr)'>j profiles for car-car impacts. Figure D. 14c shows
^DN Pr°files f°r locomotive operations in the yard.
The calculated Lpj^j profiles in Figure D. 14 are based on observed levels and frequencies of
occurrence of various noises. In addition to the usual geometric attenuation, atmospheric
absorption and ground attenuation effects (Beranek, 1971) were included in the construction of
the figure. The levels for the individual noise events at the measurement points shown in
Figure D. 14 were consistent with the points of origin of the events also shown in Figure D. 14.
The noise levels for retarders and rail car impacts arc considerably lower than those for loco-
motives, so that the total noise levels from all sources is approximately that of locomotives alone,
as shown in Figure D.14. The noise levels determined from magnetic tape recordings of noise
emissions at the West 30th measurement point are also in good agreement with the total noise
emission levels (approximated by locomotive noise), as noted in Figure D.14c.
Retarder noise levels and impact noise levels in Figure D.14 generally would be dominant at
community observation points if the locomotive noise levels were lowered by 10 dB(A). Thus,
retarder and car impact noise will replace locomotive noise as the most obtrusive noise in the
community near the Cicero Yard, if locomotive exhausts can be muffled sufficiently to lower their
noise by 10 dB(A) (assuming that no other sources of locomotive noise produce levels comparable
to exhaust noise levels).
*Enclosure A of section 8.
D-19
-------
RETARDSR MEASUREMENTS
IMPACT MEASUREMENTS
CONTINUOUS SOUND LEVEL
RECORDINGS
LOUDSPEAKERS
(a) Retarder Squeals
FIG. D.14a. LDN PROFILES FOR BURLINGTON NORTHERN'S CICERO YARD
D-20
-------
'ill
i^L
9 RETARDER MEASUREMENTS
ASUREMENTS
N • CONTINUOUS SOUND LEVEL
(b) Impacts
FIG. D.14b. (CONT.)
D-21
-------
c
TOTAL LOCOMOTIVE L ON «66, 61 TO ««s MEASURED
( J:OO TO 7-00 A.M.)
1303 NF LOCOMOTIVE IN THROTTLE
4, S0% OP THE TIME
O 1500 HP LOCOMOTIVE,
CONTINUOUS IDLE
(c) Locomotives
FIG. U,L4c. (CONT.)
-------
ORIGINAL EQUIPMENT
20 dB RETARDER
QUIETING
10 dB LOCOMOTIVE
QUIETING
10 dB LOCOMOTIVE
QUIETING AND
20 dB RETARDER QUIETING
70 80 90
LDN [dB(A)]
100
FIG. D.15
NUMBER OF PEOPLE EXPOSED TO VARIOUS I BY CICERO
YARD OPERATIONS.
D-23
-------
Figure D.I 5 shows the number of people exposed to various L(jn around the Cicero Yard.*
Figure D, 15 indicates that a muffler which quiets locomotive exhaust noise by 10 dB(A) will
•Jecrease by 400 the number of people exposed to Ljn of 65 or more from the Cicero Yard opera-
tions (assuming that no other sources of locomotive noise produce levels comparable to exhaust
noi!>e levels). The figure also shows that barriers providing a 20 dB(A) reduction of retarder noise
would decrease by 200 the number of people exposed to Lcjn of 65 or more.
Analysis in more detail of Figure D. 15 shows that at the time of the study, at the Cicero Yard
approximately 4,800 people or more were exposed to noise levels higher than the L»->•.. .'•>» . ,. „ . OS. , . ,. „, ,
which is a 42% improvement. From a hearing conservation point of view, the number of exposed
people would shrink to 0, which is a 100% improvement.
*Population densities for use in construction of Figure D.I5 were obtained from the U.S.
Department of Commerce, Bureau of the Census.
D-24
------- |