1. Test Results of a High Capacity Wayside Energy
Storage System Using Ni-MH Batteries for
DC Electric Railway at New York City Transit
Koki Ogura, Kazuya Nishimura, Takahiro Matsumura, Chiyoharu Tonda, Eiji Yoshiyama
GIGACELL Battery Center, Rolling Stock Company, Kawasaki Heavy Industries, Ltd., Kobe, Japan
Maurice Andriani, Willard Francis
Kawasaki Rail Car, Inc., Yonkers, New York, United States
Robert A. Schmitt, Anthony Visgotis, Nicholas Gianfrancesco
Power Substations, Subway – Electrical / Power, New York City Transit, New York, United States
Abstract— Over the past several years, mass transit systems in
the United States have been facing increasing demands on their
power systems. This is due to the several factors: increased
ridership, growing overall demand for power and limited
expansion of power generation facilities due to environmental
concerns. Finding the solutions for these problematic factors has
increased the demand for additional power sources at minimal
cost, an ability to recycle the regenerative braking energy created
by braking trains and line voltage stabilization. These demands
have been successfully addressed in a project that tested a
directly connected high capacity nickel-metal hydride (Ni-MH)
battery developed by Kawasaki. These controlled tests were
conducted at the New York City Transit (NYCT) early in 2010.
This paper presents the results of the tests on the Battery Power
System (BPS) using our GIGACELL batteries at the Far
Rockaway and Manhattan locations in the NYCT system.
Keywords— Wayside Energy Storage System, Nickel-metal
hydride (Ni-MH) battery, Direct Connection, Regenerative
Braking Energy, Voltage Stabilization, Emergency Power Source
I. INTRODUCTION
Energy storage application requirements for railways must
include compensation for voltage drops in railway power
supply lines, effective use of regenerative power and peak
power reduction. Various energy storage systems for railways
are being researched and developed, and their applications are
being tested and making progress all over the world[1][2]. It is
required that these energy storage systems must be capable of
quick charge and discharge, and possess a large power capacity.
Kawasaki recently developed a high-capacity, high-
performance battery based on nickel-metal hydride (Ni-MH)
technology named GIGACELL, which is shown in Figure 1.
By taking advantage of the low internal resistance
characteristics of GIGACELL batteries, the BPS can be
structured as a wayside energy storage system with the
GIGACELL batteries connected directly to the traction power
system without any power conversion system or controlling
device.
The BPS meets the requirements described above, and also
offers the benefits of compact size, high efficiency during
charging and discharging, and no adverse effects to signal
facilities such as electromagnetic interference (EMI). In
addition, when the BPS is installed between substations, it
stabilizes the railway power line voltage. Moreover, in the
unlikely event of an electric power failure at the substations,
trains can move to the nearest station only using the stored
energy in the BPS. This feature is commonly called the
emergency power supply for power outages.
Even though, Kawasaki has successfully demonstrated the
effectiveness of the BPS in Japan[3], this project in the NYCT
system represented a significant opportunity to confirm its
potential in the largest subway system in North America. The
verification tests were conducted at the Far Rockaway test
facility located adjacent to the NYCT test track used for the
new R160 trains and to the tracks used for the daily operation
of the A line trains.
II. BPS CONFIGURATION
Figure 2 shows an example of the BPS system diagram. As
can be seen in Figure 2, there is no control system such as a
DC-DC converter used with the BPS, and thus no EMI is
produced by this system. The BPS configuration consists of
battery units connected in parallel and then the entire battery
assembly
Figure 1. GIGACELL Battery

2. assembly is connected directly to the traction power system
with high speed circuit breakers (HSCBs) which provide
protection on both the positive and negative sides. In the event
of an abnormality, the BPS will be disconnected from the
traction power system by the HSCBs, thus allowing the traction
power system to continue to function. It should also be noted
that manual disconnect switches are also provided to isolate the
BPS from the traction power system for maintenance.
The condition of the BPS is monitored by the Battery
Monitoring System (BMS). As its name implies, the BMS
continuously monitors key performance characteristics of the
BPS, such as the internal temperature and the pressure of each
battery. In the unlikely event of a severe abnormality, the BMS
automatically disconnects the BPS from the traction power
system by opening the HSCBs. The intrinsically safe Ni-MH
technology and the BMS work together to create an extremely
safe, stable and reliable wayside energy storage system.
III. ATTRIBUTES OF THE GIGACELL BATTERY
The BPS attributes include the following.
・ Direct connection across the power line
・ No EMI
・ High efficiency during charging and discharging
・ High energy capacity
・ Rapid charge and discharge
・ Stable energy at times of peak power demand
・ Reduced energy costs
Specifications and a photograph of BPS installation are
provided in Table 1 and Figure 3 respectively.
TABLE I. BPS SPECIFICATIONS
Items Value
Total Voltage (Nominnal) 670 Vdc
Battery Capacity 600 Ah
Energy Capacity 400 kWh
Parallel Number of Battery Unit 4 Units
Series Number of Battery Module 17 Modules
Internal Resistance 25 mΩ
IV. TEST RESULTS
A. Verification of 3rd Rail Voltage Stabilization when R160
Train Accelerated at Full Throttle
・ Objective
Verify that the BPS provides supplemental power which
enables the 3rd rail voltage to be stabilized when the train is
accelerated at full throttle. R160 test train (10 cars train) is
shown in Figure 4.
・ Results
1) Without BPS, the 3rd rail voltage changed from 673 V to
555 V. The voltage drop is 118 V. (Figure 5)
2) With BPS, the 3rd rail voltage changed from 671 V to 608
V. The voltage drop is 63 V. The BPS provided peak current
of 2,900 A during the train acceleration. (Figure 6)
3) Therefore, the 3rd rail voltage was stabilized by 55 V (118 -
63 = 55) with the BPS connected and the train was
accelerated at full throttle. Moreover, the ripple voltage
waveform became smoother after the BPS was connected.
Figure 2. An example of BPS System Diagram
Figure 3. BPS Installation at Far Rockaway
Figure 4. NYCT R160 Test Train

3. B. Verification of the 3rd Rail Stabilization at the Service
Line (A-Line)
・ Objective
Verify that the BPS provides supplemental power that
enables the 3rd rail voltage to be stabilized during peak power
demands.
・ Results
1) Without BPS, the 3rd rail voltage fluctuated between 707 V
and 585 V. The voltage drop is 122 V. (Figure 7)
2) With BPS, the 3rd rail voltage fluctuated between 679 V and
615 V. The voltage drop is 64 V. (Figure 8)
3) Therefore, the 3rd rail voltage was stabilized by 58 V. (122 -
64 = 58)
Figure 5. 3rd Rail Voltage when Train Accelerated
at Full Throttle [Without BPS]
Figure 7. 3rd Rail Voltage [Without BPS]
Figure 8. 3rd Rail Voltage and Battery Current [With BPS]
Figure 6. 3rd Rail Voltage and Battery Current when Train
Accelerated at Full Throttle [With BPS]
300
400
500
600
700
800
0:00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 0:00
Time
3rdRailVoltage[V]
300
400
500
600
700
800
Min: 585 V
Max: 707 V3rd Rail Voltage
350
400
450
500
550
600
650
700
750
-30 -20 -10 0 10 20 30 40 50 60 70 80
Time [sec]
3rdRailVoltage[V]
-1000
0
1000
2000
3000
4000
5000
6000
7000
BatteryCurrent[A]
671 V
608 V
3rd Rail Voltage (= Battery Voltage)
Battery Current
63 V
Powering
Discharge
350
400
450
500
550
600
650
700
750
-30 -20 -10 0 10 20 30 40 50 60 70 80
Time [sec]
3rdRailVoltage[V]
-1000
0
1000
2000
3000
4000
5000
6000
7000
BatteryCurrent[A]
555 V
673 V
118 V
3rd Rail Voltage
Battery Current = 0 A
Powering
300
400
500
600
700
800
0:00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 0:00
Time
3rdRailVoltage[V]
-700
0
700
1400
2100
2800
BatteryCurrent[A]]
Max: 679 V
Min: 615 V 3rd Rail Voltage (= Battery Voltage)
Battery Current
Discharge
Charge

4. C. Verification of Regenerative Energy Enhancement and
Utilization by R160 Test Train
・ Objective
Verify that the BPS enhances the generation of regenerative
energy, captures the energy generated by the regenerative
braking during train operation at all speeds, stores it and
provides it for supplemental use as necessary.
・ Results
1) Without BPS, the regenerative energy from the train is 1.34
kWh (Figure 9).
2) With BPS, the regenerative energy from the train is 2.94
kWh (Figure 10), which is 2.19 times more energy (Figure
11).
3) With the BPS connected the regenerative energy stored by
the BPS is 2.10 kWh, therefore 71.4 % of the regenerative
energy was stored by the BPS(Figure 12).
D. Verification of the Use of BPS as an Emergency Power
Source
・ Objective
Verify that the BPS alone can provide sufficient power to
operate a single train whenever there is no power being
supplied by the substation and to calculate the total number of
trains that could be moved by BPS alone with all auxiliary
equipment such as lighting and air conditioning on, at a
maximum train speed of approximately 10 mph.
・ Results
Figure 13 (a) shows the speed and distance traveled by the
train. Figure 13 (b) shows the power consumption of the train
during the emergency run test. As can be seen, during
powering the maximum power consumption is approximately
800 kW. Figure 14 shows the discharge of the BPS during the
emergency run test. The starting point of approximately 300 A
is caused by the constant load of auxiliary power equipment.
During powering, battery current discharge exceeds 1,600 A in
order to maintain 10 mph speed, and the battery voltage
momentarily drops to approximately 600 V.
(a) Speed
Figure 9. Regenerative Energy from Test Train [Without BPS]
(b) Power
(b) Power
(a) Speed
Figure 10. Regenerative Energy from Test Train [With BPS]
Figure 11. Comparison of Regenerative Energy
Figure 12. Regenerative Energy From Test Train and
Absorbed Regenerative Energy by BPS
0
10
20
30
40
50
0 10 20 30 40 50 60 70 80 90 100 110
Time [sec]
Speed[mph]
-2000
-1000
0
1000
2000
3000
4000
0 10 20 30 40 50 60 70 80 90 100 110
Time [sec]
Power[kW]
Regenerative Energy
1.34 kWh
0
10
20
30
40
50
0 10 20 30 40 50 60 70 80 90 100 110
Time [sec]
Speed[mph]
-2000
-1000
0
1000
2000
3000
4000
0 10 20 30 40 50 60 70 80 90 100 110
Time [sec]
Power[kW]
Regenerative Energy
2.94 kWh
-1500
-1000
-500
0
500
1000
1500
-10 -5 0 5 10 15 20
Time [sec]
Power[kW]
Regenerative Energy
2.94 kWh
Absorbed Regenerative Energy
2.10 kWh
Test Train
BPS
Braking
Test TrainTest Train

5. The summary of this result is as follows.
1) The train ran on the test track for a round trip distance of
8,200 feet and used 5.5 % of the BPS Charge.
2) The train ran on the service line for a round trip distance of
3,000 feet and used 2.8 % of the BPS Charge.
3) Therefore, supposing the next station is 4,000 feet away, a
fully charged BPS alone can move up to 17 trains (10 cars /
train) to the next station.
E. EMI Test Results
・ Objective
Determine if there is any EMI impact due to the BPS.
・ Results
NYCT measured the EMI levels at the site and did not
observe any impact when the BPS was operating.
F. Emergency Power Supply at Central Substation
The BPS was relocated to Central Substation (CSS), which
is in the center of Manhattan, during the month of June and was
recommissioned in the 1st week of July 2010. CSS is
surrounded by 5 other substations in the immediate area in
Manhattan. They provide the large amount of power required
to support the power demands of the many NYCT trains
running in Manhattan around CSS.
On August 5th, while undergoing a minimum 6-month test
in this extremely busy transit location, both rectifiers in Central
Substation unexpectedly shutdown causing an unanticipated
test of the BPS as an emergency power supply. During this
event, which lasted for 30 minutes, the BPS demonstrated its
ability to assist the surrounding substations to support the
NYCT system power requirements. As shown in Figure 15, the
BPS was able to maintain an average voltage of 600 V while
supplying peak current of over 2,500 A to accelerating trains.
Overall, the BPS supplied 95 kWh to the system during the 30-
minute period of the event. At the conclusion of the event,
when rectifier power was restored, the BPS was fully charged
within 5 minutes. And returned to normal operating conditions.
V. CONCLUSIONS
The tests successfully demonstrated the following.
1) The BPS stabilized the dramatic fluctuations in the third rail
line voltage.
2) The BPS efficiently captured the regenerated braking energy
and used this energy as needed, thus reducing the contract
energy requirements and also CO2 emissions.
3) The BPS demonstrated that the peak demand can be reduced.
4) The BPS started a 10-car train from a complete standstill and
operated it for a full round trip on the test track while all its
lights and auxiliary equipment were “On.” It proved that up to
17 10-car trains could be moved to the next station during an
emergency power outage.
(b) Power
(a) Speed and Distance
Figure 14. BPS Voltage and Current during Emergency Run Test
Figure 15. BPS Voltage and Current during Emergency
Power Supply at Central Substation
Figure 13. Test Train Measurements during Emergency Run Test
0
5
10
15
20
0 50 100 150 200 250 300
Time [sec]
Speed[mph]
0
1200
2400
3600
4800
Distance[ft.]
-300
0
300
600
900
1200
0 50 100 150 200 250 300
Time [sec]
Power[kW]
250
300
350
400
450
500
550
600
650
700
9:20 9:30 9:40 9:50 10:00 10:10 10:20
Time
3rdRailVoltage[V]
-3000
-2000
-1000
0
1000
2000
3000
4000
5000
6000
BatteryCurrent[A]
Battery Current
9:38 AM 10:08 AM
3rd Rail Voltage (= Battery Voltage)
Emergency Power Supply
Discharge
Charge
300
350
400
450
500
550
600
650
700
0 50 100 150 200 250 300
Time [sec]
3rdRailVoltage[V]
-500
0
500
1000
1500
2000
2500
3000
3500
BatteryCurrent[A]
Battery Current
3rd Rail Voltage(= Battery Voltage)
Current for Auxiliary Power Equipment (Lightning, Air-conditioning, etc)

6. 5) The BPS was easily installed by direct connection to the
third rail line voltage without any electronic controls and had
no measured EMI impact.
Based on the above conclusions, it has been confirmed that
the application of BPS with GIGACELL batteries connected
directly to the railway traction power line makes it possible to
reduce energy consumption through an effective use of braking
regenerative power and that BPS responds quickly to
discharging during powering as well as to charging by
regenerative power during braking. GIGACELL batteries are
applicable to other regenerative devices such as in cranes,
elevators, and robots and to renewable energy sources such as
wind power systems and photovoltaic systems.
GIGACELL batteries have opened up the possibility to
utilize the new types of energy, such as power systems
equipped with a BPS as a hub for utilizing renewable energy. It
also serves as means for providing safe and secure
transportation in the event of a power interruption, a means of
reducing power demand, as protection against voltage drops,
and as a tool to avoid canceled regeneration thereby helping to
build a system to enhance the reliability of rail transportation as
well as contribute to the energy conservation. By combining
the technology that we have accumulated as a rolling stock
manufacturer with the battery technology we have developed,
we will continue proposing environmentally friendly
equipment that improves the safety and efficiency of railways.
ACKNOWLEDGEMENT
This test program was sponsored by the New York State
Energy Research and Development Authority (NYSERDA)
and was performed with the cooperation and support of New
York City Transit (NYCT).
REFERENCES
[1] A. Okui, S. Hase, H. Shigeeda, T. Konishi, T. Yoshi, “Application of
Energy Storage System for Railway Transportation in Japan,”
International Power Electronics Conference (IPEC2010), pp. 3117-3123,
2010.
[2] H. Lee, E. Joung, G. Kim, C. An, “A Study on the Effects of Energy
Storage System,” International Conference on Information and
Multimedia Technology (ICIMT2009), pp. 28-32, 2009.
[3] K. Tsutsumi, T. Matsumura, “Revolution in Storage Battery Technology
and Adoption by Electric Railways,” Science & Technology in Japan,
vol. 26, no. 102, pp. 21-24, 2009.