-------
Finally, since the noise in transit cars increases with in-
creasing vehicle speed (as also discussed later in detail), the
sneeds listed in Table I are of some importance in assessing the
noise and the noise control problems.
Car Builders and the Procurement Process
In the past 15 years, ACF Industries and St. Louis Car Co.
have ceased all passenger car production and Budd has terminated
its production of self-propelled cars, leaving Pullman-Standard
as the only remaining old-line car builder.
However, new companies have entered the transit car building
field in the oast few years. Rohr Corp. supplied the cars for
the new BARTD system, the Boeing Vertol Co. has developed and
built a pair of state-of-the-art cars (SOAC) now undergoing test-
ing under the Urban Mass Transit Administration's Rapid Rail Sys-
tems and Vehicles Programs, LTV won a contract to supply vehicles
for the new Dallas airport system, and General Electric, who used
to supply only transit car components, has begun to bid as a prime
car supplier.
Transit systems wishing to purchase new cars generally pre-
pare detailed specifications, which are submitted to potential
suppliers for bidding.* Car builders generally do most of their
design work in the course of preparing bids. In effect, a bid
typically indicates little more than the proposed price for the
cars to be supplied; the successful bidder usually is the one
who can meet the prescribed specifications and schedules reli-
ably at- the lowest cost.
•Except for some of the most recent ones, these specifications
did not include any quantitative noise performance requirements;
some of the very newest ones, on the other hand, specify rather
stringent noise performance requirements, acceptance tests, and-
payment penalties for not meeting these requirements.
-------
Each oar proposed in response to a bid request is in essence
a new design aimed at meeting the specific requirements of the
procurement. Since the designer can take noise control techniques
and components into account during the early design stages, one
may expect that many of these noise control considerations can
be implemented at relatively low cost. However, except for some
very rare bold innovations, most new car designs draw heavily on
established technology, so that improved (and quieter) designs
tend more to evolve slowly (in a rather conservative industry)
than to appear overnight.
Rapid transit cars constitute a relatively complex assemblage
of systems and components. Builders typically build only the car
structure and body shell — they procure from other suppliers, in-
tegrate, and assemble all other parts, including such heavy items
as trucks, wheels, axles and propulsion motors, such major sub-
systems as controls, communication, and HVAC equipment, and such
smaller items as seats, doors, door operators, public address
systems, and lighting.
-------
NOISE IN TRANSIT CARS
Where Noise Originates
The primary sources of steady noise* in rapid transit cars
and the relation of these sources to passengers may be visualized
with the aid of Fig. 1, which shows a schematic section through
a transit car.
These sources, in typical order of importance, are:
1. Wheel/rail interaction
2. Propulsion (traction) system
3. Auxiliary (undercar) equipment
4. Air conditioning and distribution systems
The steady "roar" noise due to interaction between wheels
and rails typically constitutes the dominant noise component in
modern rapid transit cars running- on welded tangent brack. For
cars running on jointed track, an impact noise associated with
passage of the wheels over joints in the track is added to the
roar noise. Not much is known at present about the basic roar-
noise-producing mechanism, but it is thought to be associated
with wheel vibrations induced by small irregularities on the rail
interacting with the wheel tread, which also may contain small
surface irregularities. (It is well known that reduction of the
irregularities in the track — e.g., by grinding — reduces the
*By "steady" noise is meant a noise that is of long enough dura-
tion to make an appreciable contribution to the time-average
acoustic energy, computed for a trip or portion of a trip lasting
at least several minutes. Noise of short duration, such as the
screech produced by car wheels traversing tight curves, contri-
butes relatively little to the noise exposure of passengers, even
though this noise may be rather intense. Thus, short-duration
noise is excluded from consideration here.
-------
AIR DISTRIBUTION DUCTS
BODY SHELL
SUSPENSION
TRUCK
TRACTION SYSTEM
UNDERCAR EQUIPMENT
FIG. 1. SCHEMATIC SECTION OF TRANSIT CAR IN TUNNEL
10
-------
roar noise.) The wheel vibrations radiate "airborne" sound (much
like a loudspeaker membrane), but also are transmitted to the ve-
hicle shell via structural paths, leading to sound radiation from
the shell. The direct airborne radiation component generally is
by far the more significant.
The propulsion equipment typically includes one or more
traction motors per truck, reduction gearing, and fans or blowers
for cooling the motors. Each of these components tend to produce
both airborne noise and structural vibrations.
Auxiliary equipment, which generally is mounted under the
car, may include air conditioning compressors and condensers
(with associated fans, pumps, motors), air compressors and other
pneumatic system components, hydraulic systems, motor-alternator
sets, and electrical and electronic systems (some of which may
include cooling fans). Again, each of these items tends to pro-
duce both noise and vibrations.
Those portions of the air conditioning and distribution
systems which are not mounted under the car may also contribute
to the noise environment in the passenger space. For example,
noise is likely to be produced by air circulation fans, by air
flow in ducts, and by air emerging through grillages and per-
forations. For reasonably well designed equipment, air condi-
tioning noise tends not to be an important factor.
How Noise Reaches Passengers
Of all the aforementioned noise sources, only those asso-
ciated with the air distribution system communicate directly
with the passenger compartment. For all of the other sources
one may expect the noise to reach the passengers via a multitude
of paths. As indicated schematically in Fig. 2, these may in-
volve :
11
-------
REFLECTING SURFACES
SOURCEOF
NOISE AND
VIBRATIONS
SOUND TRANSMITTED
AIRBORNE SOUND (DIRECT)
VIBRATIONS OF CONNECTING STRUCTURES
OPENING
SOUND RADIATION
SOUND
IN CAR
INTERIOR
FROM BODY
VIBRATIONS
CAR
BODY
FIG. 2. SCHEMATIC DIAGRAM OF PATHS FOR SOUND TRANSMISSION INTO
CAR INTERIOR
-------
1. Transmission of airborne sound from the source to the
vehicle body, with sound entering the passenger com-
partment
(a) via openings (e.g., air intakes or exhaust
vents, gaps in door seals, open windows), or
(b) by setting the body shell into vibration,
causing it to radiate sound; and
2. Transmission of vibrations to the body shell via
structural paths (e.g., including bearings, mount-
ings, fastenings), resulting in airborne noise
radiation into the passenger compartment.
Transmission of (airborne) sound from sources outside the
car to the vehicle body may take place along relatively direct
"line of sight" paths, and along more circuitous paths involv-
ing reflections from the trackbed, the ground, and from tunnel
surfaces. For vehicles located in the open, one may expect
much of the airborne noise to reach the vehicle from its under-
side; for vehicles in tunnels, on the other hand, one may ex-
pect noise to reach it essentially from all directions. In
typical tunnels with little acoustic absorption, multiple re-
flections tend to make the sound field around vehicles rela-
tively uniform; since no sound can escape to the side, these
sound fields also tend to be relatively intense.
The Noise Environment in Cars
Since, as evident from the foregoing discussion, the noise
in a car depends to some extent on whether the car is in a tunnel
or in the open, it is reasonable to treat these two cases separ-
ately. In addition, the two most important ones of the previously
listed noise sources depend very significantly on the speed of the
vehicle, so that car speed may be expected to be an important para-
meter affecting the in-car noise.
13
-------
The available data* on the steady noise inside rapid transit
cars is summarized in Figs-. 3 and 4, in terms of (overall) A-
weighted noise levels, plotted as functions of speed. Correspond-
ing frequency spectra, as far as available, are collected in Ap-
pendix A. Presentation of the information here in terms of A-
weighted levels has been chosen because these levels have become
widely accepted as a basis both for judging noise annoyance and
for establishing hearing conservation criteria.
Figure 3 pertains to transit cars travelling on tangent
(straight) track, on the surface of the ground (not on elevated
structures), whereas Fig. 4 pertains to cars on similar track in
tunnels. The data in both figures corresponds to track that con-
tains no unusual roughness or irregularities.
The higher-speed data of Fig. 3 may be seen to fall into
three bands — two of which, if continued toward lower speeds, do
not encompass the lower speed data very well. This state of af-
fairs also is evident in Fig. 4 and has a reasonable explanation.
At zero speeds, the noise in a car is due only to air-handling
and auxiliary equipment; contributions from the propulsion system
and from dynamic wheel/rail interaction obviously are absent.
With increasing speed, these contributions increase until they
eventually predominate. Thus, the low-speed and higher-speed
regions of these two figures essentially correspond to dominance
*Data appearing in the literature without corresponding speed
information has not been included. Neither has such data from
which A-weighted overall levels cannot be deduced reliably.
The presence of passengers in cars changes their acoustical
characteristics somewhat, and therefore also affects the noise
environment in cars to some extent. However, these effects are
relatively minor and generally well within the spread of the data
summarized here.
14
-------
120
T
T
T
110
100
CD
•o
hi
>
LU
UJ
CO
O
z
90
80
FIG
N NYCTA, R-44 (Ref.1)
C CTA, Budd (Ref. 2)
Ground Rail (Ref.4)
Underground Rail (Ref.4)
MBTA, South Shore (Ref.7)
MONTREAL, Line'2 (Ref.9)
PATCO (Ref. 10)
TTC, Gar 5414 (Ref. 3)
SOAC (Ref. 5}
Range of Data
u }BARTD, car 107 {
M
R
p
T
s
I
(See TABLE I for Abreviations)
U.
10
20
30 40 50
SPEED (mph)
60
70
80
. 3. STEADY NOISE LEVELS IN TRANSIT CARS ON TANGENT TRACK OF GOOD QUALITY
ABOVE GROUND
-------
120
I
no
100
CD
5 so
Ul
-J
Ul
CO
80
NYCTA, Various Lines and Cars (Ref. 9)
CTA, Various Lines and Cars (Ref. 2,6)
C3 SEPTA, Various Cars {Ref. 11)
M MBTA, South Shore (Ref.7)
P PATCO (Ref. 10)
T TTC, Several Cars (Ref. 3)
I Range of Data
(See TABLE I for Abreviatlons)
•4.
•KJ
•-S
LONDON
LISBON
STOCKHOLM
BERLIN
PARIS
(Rubber-Tired)
(Ref. 8) «
10
20
30
50
60
FIG. 4.
40
SPEED (mph)
STEADY NOISE LEVELS IN TRANSIT CARS IN SUBWAY TUNNELS
70
80
-------
of different noise sources. (The fact that one band of Fig. 3
also includes the lower-speed data probably is fortuitous.) Be-
cause of the lower noise levels at low speeds, and because tran-
sit systems tend to operate their vehicles at the greatest pos-
sible speeds consistant with safety and acceleration/deceleration
limitations, the lower-speed information is of limited interest.
Consequently, the later discussion of noise control costs focuses
on the higher-speed region.
The differences in the noise levels associated with the vari-
ous bands of Fig. 3 may be ascribed to differences in the car.
The data in the highest band (enclosed by solid lines, and in-
creasing on the average by about 4 dBA per 10 mph increase in
speed) corresponds to cars of somewhat older designs than the
data in the middle band (enclosed by long dashed lines, and in-
creasing on the average by about 2 dBA per 10 mph increase in
speed ). The lowest band (short dashed lines, also increasing
at 2 dBA per 10 mph) corresponds to a single very new demonstra-
tion vehicle.
Although the data pertaining to in-car noise in tunnels
does not suffice for the drawing of trend-indicating bands in
Fig. 4 like those of Fig. 3, bands are indicated in Fig. 4.
These have been established simply by shifting the upper two
bands of Fig. 3 upward (both by the same amount), so that they
enclose most of the significant higher-speed data. This 10 dBA
shift indicates that the noise level in a given vehicle at a
given speed is 10 dBA higher on the average when the vehicle is
in a tunnel than when it is on the surface.
From Fig. 3 one may determine that the noise level L in the
most quiet transit cars currently in service, when operating at
a speed V above ground, may be estimated from
17
-------
L(dBA) = 65 + 0.18 V(mph)
within ±5 dBA. In view of Fig. 4, one finds that one may estimate
the noise level in such cars in tunnels (for speeds above 20 mph)
by adding 10 dBA to the above-ground noise level obtained from
the foregoing relation.
One may also note that at any particular speed above 35 mph
the state-of-the-art car is about 7 dBA quieter on the average
than currently operating cars.
18
-------
NOISE REDUCTION AND ITS COSTS
Car Design Modifications for Noise Reduction
The most fruitful approach toward the reduction of noise
generally consists of modification of the noise sources so as to
reduce the noise generation. Application of this approach to
transit cars requires modification of the wheel/rail interaction
and possibly also of the propulsion and under-car equipment.
The only practical means presently available for reducing
wheel/rail roar noise at its source consists of replacing the
standard steel wheels in present use by "resilient" wheels. Sev-
eral such wheel designs are available and have been tested; all
incorporate rubber elements between the steel rim running sur-
faces and the central wheel discs, so as to achieve some vibra-
tion isolation between the rim and central disc.
Reductions in the noise produced by the propulsion and aux-
iliary equipment sources usually may be obtained by choosing
quieter components (e.g., helical instead of spur gears, slow
centrifugal blowers instead of high-speed axial flow fans) and
by taking appropriate care in system design (to avoid turbulent
fluid flows, reduce mechanical vibrations, avoid impacts, rattles,
buzzing).
One may also reduce the noise reaching the passengers by
obstructing the dominant propagation paths. Thus, one may place
acoustical enclosures around noisy equipment components, and pos-
sibly even around the wheels (although wheel enclosures are like-
ly to be impractical). One may also increase the attenuation
provided by the body shell by sealing all openings as well as
possible, providing mufflers for all openings that cannot be
sealed, and using shell structures that permit less sound trans-
mission. Such structures, for example, might be of a double-wall
19
-------
or "shell within a shell" type. Similarly, one may impede the
propagation of vibrations (which lead to sound radiation in the
passenger space, as previously discussed), e.g., by use of vibra-
tion isolation in the form of rubber "shock mounts", elastomeric
bushings, or air springs.
Finally, one may reduce the intensity of the sound fields
generated in the passenger space by the various sources (and
paths) somewhat by increasing the acoustic absorption in the
passenger compartments, for example by installing acoustical
ceiling treatment, carpets and/or upholstery.
Costs and Benefits
Table II lists the various feasible car modifications that
may be expected to result in reductions of in-car noise, together
with the expected magnitudes of these reductions, and the associ-
ated estimated weight penalties and costs. For modifications
that affect noise in vehicles on grade differently from that in
vehicles in tunnels, two different values are indicated. The
initial costs of these noise control modifications listed in the
table represent the associated increase in cost of new cars; cor-
responding retrofitting of cars in current use is likely to be
prohibitively costly and is not considered here. The "Remarks"
column contains primarily notes concerning technical aspects of
the modifications.
Inspection of Table II leads one to the following conclu-
sions:
(1) Use of a floated interior shell is the one single modifica-
tion capable of providing the greatest noise reduction.
However, this modification involves considerable cost and
weight penalties.
20
-------
TABLE II. IN-CAR NOISE REDUCTIONS
DESIGN MODIFICATIONS
AND COSTS ASSOCIATED WITH TRANSIT CAR
MODIFICATION
gesilient Wheels
Quieter Components
Propulsion
Motor and cooling, fan
Gearing
Undercar Auxiliaries.
Electrical
Electronic
Motor-alternators
Hydraulic
Pneumatic
Air conditioning
Agoustlcal Enclosures for above
components
Vibration Isolation of above
eomoonents
Imoroved Vibration Isolation between
Trucks and. Body
liroroved Acoustical Performance of Bodj
Boubie->pane windows or acoustical glass
Secondary (.floated) floor
Tlgnter door seals
Air duct muffling
Floated (isolated) interior shell
Added Absorotlon Inside Cay
Quieted Air Distribution System
ecrease*** In
teady In-car
Noise, Above
30 mph (dBA)
5
2
3
N,E
N,E
S,E
N,E
S,E
S
N,E
2
2
fl on grade,
13 in tunnel
13 on grade.
12 in tunnel
* 1
2
(6 on grade,
10 In tunnel
2
5
Estimated Average
Incremental Costs
per car ($1000)*
Initial
3.2
(Approx,
$400/whael)
N
10
N
N
N
N
N
N
0.5
N
1.0
1 1.2
U$75/wlndow
2
1
1
25
1
0.5
Operating
-0.3/year
N
N
N
N
N
N
N
N
N
N
0.2/year(R)
0.2/year(R)
0.2/year(R)
If
0.2/year(R)
N
N
Weight
Penalty
per car**
(1000 Ib)
N
1
N
0.1
N
N
N
N
N
N
0.5
N
N
1.5
1.5
N
N
3.0
0.2
0.2
REMARKS
Operating cost reduction due to pos-
sibility of replacement of worn rims
Instead of entire wheels.
Modification of fan and cooling air
passages.
Higher quality gears, gear unit oil
cooling.
1 Primary noise due to air cooling, if
i ftnv
• aujr •
Noise due to pumps, valves, motors.
Use rotary Instead of reciprocating
equipment.
Primary noise due to compressors,
valves. Use rotary Instead of recip-
rocating equipment.
Primary noise due to compressors,
condenser cooling air fans.
Enclosures Include provision for
cooling, Including muffling of air
passages for air cooling.
Reduces transmission of vibrations
originating- from wheel/rail interac-
tion and propulsion components.
Double-pane windows Imply need for
added sash and structural complexity.
Require development to be practical;
necessitate more frequency replace-
ment to maintain seal.
Cleanablllty requirements usually
limit design.
Includes appropriate windows and
door seals.
Space limitations, cleaning require-
ments and vandal-proofing limit de-
sign and usable materials.
Space limitations limit design.
N • leellelble 3 • source contributes significantly only to noise In stationary
' * or slowly moving cars
E « modifications affect exterior noise primarily ' ,, , «.„„„ »
R • repairs, replacement, and maintenance
•Typical car cost $250,000 to 300,000
••Typical oar weighs 60,000 to 100,000 16. Cost of weight penalty IS $1.50 to *2.00 per pound,
•••Amounts of decrease Indicated correspond to implementation of only one modification at a time. Decreases due to multiple
modifications are not additive in general.
-------
(2) Many other modifications, which effect limited noise reduc-
tions, may be implemented at little cost.
(3) Many modifications affect only the low-speed in-car noise,
and not the high-speed noise, which is of primary interest
here.
Incremental Costs of Quieter Cars
The decreases in the high-speed in-car noise expected to be
obtained by use of virtually all technically sensible combina-
tions of noise control modifications are indicated in Table III,
together with the associated incremental costs.
For purposes of preparing this table, it was assumed that
anyone desiring quieter cars at minimum cost would install qui-
eter motor and cooling fans, at the same time improving the vi-
bration isolation of the noisy propulsion and undercar compo-
nents, since these two modifications are estimated to reduce
the noise by 3 dBA, at essentially zero incremental cost. In
addition, it was assumed that of the four approaches involving
minor design improvements and/or development — namely: (1) im-
proved vibration isolation between trucks and body, (2) improved
air duct muffling, (3) increased acoustical absorption inside
car, and (4) tighter door seals — one would always implement
either all or none.
Figure 5 shows the noise reductions obtained with the vari-
ous combinations of noise control modifications, as a function
of the initial incremental costs they add to a car. This figure
permits one to select that combination which gives the greatest
amount of noise reduction for a given incremental initial cost,
or to determine the minimum cost associated with a given amount
of noise reduction. In addition, Fig. 5 also permits one to
22
-------
S3
Q.
E
O
*-. ^
< LJ
CD >
3 o
CO
2 <
2u 10
I- w
= o
Q Z
o
i
220
"*
-•-
170 O18 2°
6 2
23
19 14
21
Numbers Correspond to Noise Reduction Combination
Code of Table IT. Points are Plotted to Represent
Average Noise Reductions for Above -Ground and
in-Tunnel Operation
10 20 30
INCREMENTAL INITIAL COST OF CAR ($ 1000)
40
FIG. 5. INCREMENTAL COSTS OF NOISE REDUCTION
-------
eliminate from consideration some combinations that are clearly
less cost-effective than others; for example, since combinations
17 and 18 produce the same amount of noise reduction, but 17 is
less costly, one would be inclined not to consider 18 further.
However, in order to consider the total costs of noise-
control design modifications more meaningfully, one must consider
the operating costs in addition to the initial costs on which Fig.
5 is based. One may reduce initial and operating costs to a sin-
gle index by discounting the future incremental costs to the pre-
sent day at an appropriate interest rate and adding this discounted
cost to the increase in initial cost. The result is the net dis-
counted cost increase. Corresponding values are shown in Table III,
based on a car service life of 25 years and on an assumed annual
interest rate of Q% (which is a representative value for public
projects).
Figure 6 is analogous to Fig. 5, but is based on the afore-
mentioned net discounted cost increase, instead of on the incre-
mental initial cost. The same remarks made above in relation to
Fig. 5 apply also to Fig. 6.
Table IV summarizes the minimum costs associated with achiev-
ing various levels of in-car noise reduction by car design modifi-
cations. It is a coincidental effect of the various combinations
of initial and operating costs listed in Table III (as well as of
the car life times and interest rates used in the discounted value
computations) that the noise control modifications which are
most desirable on the initial cost basis are also most desirable
on the net discounted value basis.
24
-------
TABLE III. NOISE REDUCTIONS DUE TO COMBINATIONS OF MODI
FICATIONS, AND ASSOCIATED COSTS
•s
0
o
g
^1
43
§
MO >
O O
>>Q C<
•d o.
O + B
(Q M
X
X
X
X
X X
X
X
X
X
X
X X
X
X
X X
fc
o
o
fo
•0
0)
43
a
0
H
fo
X
X
X
X
X
X
X
X
r-t
r-i
(U
co
•o
4>
Float
X
X
X
X
X
n
H
4>
4)
4J
C
0)
•H
i-H
i-l
03
£
X
X
X
X
X
X
X
X
X
*
0)
£
o
flj
"
V
09
3
5
1-3
3-2
6-10
5
7
4-5
5-4
8-12
7.5
7.5-9
8
12-15
'11
6.5-7
8.5-9.5
9
11.5-15.5
10
16-8.5
11-12
14
•€»•
+>
M
O
O
iH
at
•H
43
•H
C
M
C
•H
4)
a
a
«
o
a
_
4
1.2
2
25
3.2
4
1.2
2
25
3.2
5.2
6
29
7.2
3-2
4.4
5.2
28.2
7.2
32.2
6.4
10.4
crating Cost (K$/Year)
8
•H
0)
3
£
o
c
M
_
0.2
0.2
0.2
0.2
-0.3
0.2
0.2
0.2
0.2
-0.3
0.4
0.4
0.4
-0.1
0.4
-0.1
-0.1
-0.1
0.6
0.1
0.1
0.3
L Cost Increase* (K$)
•o
0>
43
O
u
a
43
_
14.7
3-3
4.1
27.1
*0
6.1
3.3
4.1
27.1
-0
9.5
10.3
33.3
7.1
7.5
3.3
4.3
27.1
13.6
33.3
7.5
13.6
•For Q% annual Interest rate, 25 year life time
**Where two numbers are given, the first pertains to above-ground
and the second to in-tunnel operation.
25
-------
TABLE IV
MINIMUM COSTS ASSOCIATED WITH NOISE REDUCTION MODIFICATIONS
In-Car Noise Reduction
Above 30 mph
Incremental Costs ($1000)
Initial* Net Discounted*
5 dBA
10 dBA
15 dBA
3.2 [11]
7.2 [15]
32.2 [21]
*0 [11]
7.1 [15]
33.3 [21]
•Numbers in brackets refer to best combination of noise control
design modifications listed in Table III.
26
-------
fo
•vj
a.
E
< UJ
§§
GO
z <
2 LJ
|- C/>
o ^
s*
«•«
o
I
10
1
1
1
0
• 15
170Q18
12®
?" 7A
7* *16
8OO9
>6
• 4
n *3
• 23
20
• ,
13
* t22
• 2
• 19
10
_ „„.,
• 5
• 21
• 14
Numbers Correspond to Noise Reduction Combination
Code of Table EL Points are, Plotted to Represent
Average Noise Reductions for Above-Ground and
in-Tunnel Operation
10 20 30
NET DISCOUNTED COST INCREASE ($1000)
40
FIG. 6. NET DISCOUNTED COST INCREASE (FOR 8% ANNUAL
LIFE) ASSOCIATED WITH NOISE REDUCTION
INTEREST RATE, 25 YEAR
-------
REFERENCES
1. A. Paolillo, Memorandum to D.T. Scannell re "Noise Measure-
ments, R-44 Contract Cars," 21 August 1972.
2. W. Patterson, "Results of Field Tests on CTA Train, Conducted
6 September 1972".
3. Toronto Transit Commission, "Interim Report on Penn Resilient
Wheels," October 31, 1969. (Test run between 27 August and
4 Septmeber 1969.)
4. G.P. Wilson, "BARDT Prototype Car 107 Noise Tests with Stand-
ard, Damped and Resilient Wheels: Ballast and Tie Tangent
Track and Short Radius Subway Curves," Final Report, June
21, 1972.
5. F.N. Hiuser, "State-of-the-Art Car: Ready for Revenue
Testing," Railway Age, January 8, 1973.
6. C. Hanson, E.E. Ungar, Measurements taken on 11 June 1973
on CTA Douglas Service.
7. E.J. Rickley, R.W. Quinn, "MBTA Rapid Transit System (Red
Line) Wayside and In-Car Noise and Vibration Level Measure-
ments," Report DOT-TSC-OST-72-31, August 1972.
8. E.W. Davis, "Comparison of Noise and Vibration Levels in
Rapid Transit Vehicle Systems," National Capital Transporta-
tion Agency. Operations Research Inc. Technical Report 216.
April 1964.
9. C.M. Harris, B.H. Aitken, "Noise in Subway Cars," Sound and
Vibration, 5., No. 2, Febrary 1971, pp. 12-14.
10. E.E. Ungar, Measurements taken on 12 July 1973.
11. J.W. Vigrass, Memorandum to R.B. Johnston, on "Resilient
Wheels," 2 July 1973.
-------
APPENDIX A
IN-CAR NOISE SPECTRA
A-l
-------
IUU
0
z
00
UJ 90
\-
0
0
0
SOUND PRESSURE LEVEL IN ONE-THI
(dB re 2x 10~5 N/m2
£> » 01 -4 C
o»o o o o c
^*
1 1
^VN
\
\
,x\
\
A,
^•s*—
40 mph (
1 1
•
**—T£
81 dBA)
1 1
/"^^
\
X
s
30mph (
1
"T
50 mph
^
v>
76 dBA)
1 1
(85 dBA)
t
^ \
*\
1 1
\
X
63 125 250 500 1000 2000 4000 8000 16,000
ONE-THIRD OCTAVE BAND CENTER FREQUENCY (Hz)
FIG. A.I. NOISE LEVELS IN TffC CAR 5414 ON TANGENT TRACK
-------
100
o
CO
UJ
u
o
o
(E
90
80
UJ Z
§r
z2 70
_j co
y ~ eo
-------
I
tfl
100
Q
ffi
UJ 90
3
0
0
0
— T 80
$1
UJ Z
§?
SOUND PRESSURE LEVEL If
(dB re 2x 1
£ w
-------
O
•z.
<
-.a
LJ
<
h
O
o
\-
i
UJ
X
-> CM
UJ
>
O
CO
31.5 63 125 250 500 1000 2000 4000 8000
ONE-THIRD OCTAVE BAND CENTER FREQUENCY (Hz)
16,000
FIG. A.4. NOISE LEVELS IN CTA CARS 2251, 2252 (BUDD CO.) ON WELDED AND GROUND
RAIL, ON GRADE, TIES ON BALLAST
-------
3--
I
9O
m
UJ 80
h-
6
Q
*- 70
UJ
E^ 60
LJ
rr
3
CO
(./i
UJ
E
Q.
O
cn
50
40
30
31.5
PIG. A.5.
O
O
A
a
a
O
MADRID TALGO 76.5 dBA
LISBON 86dBA
PARIS (RUBBER TIRED) 75 dBA
BERLIN 77.5dBA
STOCKHOLM 81dBA
LONDON TUBE 83 dBA
I I
I !
I i
I i
I I
I
63 125 250 500 1000 2000 4000
OCTAVE BAND CENTER FREQUENCY (Hz)
NOISE LEVELS IN SEVERAL EUROPEAN SUBWAY CARS
8000 16,000
-------
CO
STATE ST
CHICAGO
O DEARBORN ST. 75 dBA
O LONDON TUBE
A PARIS (RUBBER TIRED)
NEW YORK
STOCKHOLM
TORONTO
31.5
FIG. A.6.
125 250 500 1000 2000 4000
OCTAVE BAND CENTER FREQUENCY (Hz)
NOISE LEVELS IN SEVERAL SUBWAY CARS
8000 16,000
-------
100
-o
A
O I DEARBORN ST. 91 dBA
© PHILADELPHIA 92dBA
Q TORONTO SOdBA
D NEW YORK 86dBA
STATE ST.
'
31.5
125 250 500 1000 2000 4000
OCTAVE BAND CENTER FREQUENCY (Hz)
FIG. A.7. NOISE LEVELS IN SEVERAL NORTH AMERICAN SUBWAY CARS AT 30 mph
8000 16,000
-------
->
CD
UJ
u
o
V.
Z
I
LJ
o?
SOUND PRESSURE LEVEL
dB re 2x
uu
90
80
70
60
50
40
31
1 1
^
\
1 1
1 1
\
V-
1 1
v^
^
1
1 1
^
x
BERLIN
40 mph,
1 1
SEATTl
45 mph
x^
^S
SUBWAY
( 75 dBA)
1 1
.E MONC
, (91.5 c
/
\
1
)RAIL
BA)
— .
\
1 1
— -*.
\
1 1
.5 63 125 250 500 1000 2000 4000 8000 16,0
OCTAVE BAND CENTER FREQUENCY (Hz)
FIG. A.8. NOISE LEVELS IN TWO TRANSIT VEHICLES AT 40 AND 45 mph
-------
APPENDIX B
LIST OF TRANSIT SYSTEMS PERSONNEL
B-l
-------
LIST OF TRANSIT SYSTEMS PERSONNEL CONTACTED
System
Office Address and Telephone
Individuals Contacted
Chicago Transit Authority (CTA)
co
CJ
Cleveland Transit System (CTS)
Massachusetts Bay Transit Authority
(META)
New York City Transit Authority
(NYCTA)
Port Authority Transit Corporation
(PATCO)
Port Authority Trans Hudson (PATIO
Southeastern Pennsylvania Transit
Authority (SEPTA)
Bay Area Rapid Transit District
(BARTD)
Merchandise Mart Plaza. Rm. 7-144
Chicago, Illinois 6065A
(312) 664-7200
1401 East Ninth Street
Cleveland, Ohio 44114
(216) 781-5100
500 Arborway
Jamaica Plain, Boston, Mass. 02130
(617) 722-6162
370 Jay Street
Brooklyn, New York 11201
(212)
Lindenwold Yard,
Lindenwold. Nevf Jersey
(609) 963-8300
Rm. 65E, 1 World Trade Center
New York, N. Y. 1004?
(212) 466-3524
200 West Wyoming Avenue
Philadelphia, Penna. 19140
(215) 329-4000
800 Madison Street
Oakland. Calif.
(415) 788-2278
Prank J. Cihak, Chief Equipment Engineer
Equipment Research/Development Department
(Ext. 516)
Glenn M. Anderson, Senior Equipment
Engineer, Rapid Transit Section
Equipment Research/Development Department
Michael (Tim) Browne, Research Specialist
Research and Planning
(Ext. 385)
John J. Williams
Planning and Development
Anthony Paolillo
Environmental Staff Division
J.W. (Bill) Vigrass
Maintenance Superintendent
(Ext. 35)
Nat Streitman, staff of Edward Parrelly,
Assistant Chief, Rail Planning Division
B.J. Krant, Manager
Administration
Public Relations Department
-------
BIBLIOGRAPHIC DATA
SHEET
K"P
P5W9-74-012
3. Recipient's Accession No.
4. Title and Subtitle
Noise in Rail Transit Cars
Incremental Costs of Quieter Cars
5. Report Date
June 1O7A
6.
Author(s)
E.E. Ungar
&• Performing Organization Kept.
No.
Performing Organization Name: and Address
Bolt Beranek and Newman
10. Project/Task/Work Unit No.
It. Contract/Grant No.
EPA No. 68-01-153*
2. Sponsoring Organization Name and Address
Environmental Protection Agency
Office of .Noise Abatement & Control
13. Type of Report & Period
Covered
Final
14.
IS. Supplemenrary Notes
16. Abstracts tj.s. rail rapid transit systems, car operations, and the car
building industry are described in relation to the procurement of quieter
cars. The noise environment of passengers in rapid transit cars is
discussed and the major noise sources and paths of noise transmission into
cars are delineated.
For essentially all combinations of car noise-control modificatior
deemed technically and economically feasible for implementation in new
vehicles, estimates are presented of the associated noise reductions,
initial costs, and operating costs. It is concluded that significant
reductions in in-car noise under typical operating condicitions can be
achieved at incremental costs that are small percentages of the total
car costs.
17. Key Words and Document Analysis. 17o. .Vrcriptors
Transit systems
Car Operations
Car Builders
Car design andi.onodifications
Noise in transit r.«»rs
Costs and benefit*, of noise reduction
17b. Identifiers/Open-Ended Terms
17c. COSATI Field/Group
18. Availability Statement
Available at NTIS
19.. Security Class (This
Report)
UNCLASSIFIED X
20. Security Class (This
Page „
UNCLASSIFIED X
21. No. of Pages
22. Price
FORM NTIS-35 (IO-70)
USCOMM-DC 40379-1-
-------