United States Nuclear Regulatory Commission - Protecting People and the Environment

ACCESSION NO #: 9705120062

Bell Labs                          Lucent Technologies

     Innovations for Lucent        Bell Labs Innovations


                              Fred C. Laman  Lucent Technologies Inc.

                    Micro Electronics Group  Room 1Q-105

                              Power Systems  3000 Skyline Drive

                                             Mesquite, Texas 75149-1802

                    Telephone 972-284-2871   EMAIL flaman@lucent.com

                    Facsimile 972-284-3579

March 3, 1997

                           REPORT OF DISCOVERY


                          IN ROUND CELL BATTERY

                               FRED LAMAN


                           REPORT OF DISCOVERY


On February 12, 1997 a four cell string of high gravity Round Cells

failed a capacity discharge test.  The test was carried out at the C&D

Technical Center in Conshohocken PA., as part of the Nuclear Round Cell

User's Council test program.  The discharge test failed because one cell

went into reversal immediately following the start of the discharge. 

This cell was manufactured in 1991 (serial # R9105-97461) and used in a 

spare battery at McGuire's Nuclear Generating Station.  The remaining

three cells in the string were new cells recently manufactured.  The

discharge was done at a constant current of 508 amps, corresponding to a

2 hour discharge rate, to an end voltage of 1.75/cell.  The reversed cell

reached a potential of -2.63 volt within 1 second and -3.32 volt after 47

seconds, at which time the discharge was terminated.  During this time

the reversed cell was gassing heavily and produced a smell of hydrogen

sulfide.  Subsequent autopsy of the failed cell revealed that the harness

of the positive electrode was completely disconnected from the four lugs

connected to the top positive plate.

Cell history

The failed cell from the Mcguire spare battery had been subjected to the

same discharge testing as the batteries in service at McGuire.  During a

total number of five tests consisting of a combination of capacity

discharge testing and service load profile testing, no additional voltage

loss associated with poor harness/lug connections had been observed. 

After the cell was transferred to Conshohocken the first capacity

discharge test using a two hour rate, showed an additional voltage loss

of approximately 100 mV and a delivered capacity of approximately 80% of

rated capacity.  In the subsequent discharge done under the same

conditions the cell was driven into reversal.

Cause of failure

Further examination of the disconnected harness and positive plate

towards incomplete welds as the most likely contributor to the

disconnection.  The weld burns were narrow and shallow.  The fact that no

discharge problems were experienced during testing done at the McGuire

Nuclear Generating Station indicates that extensive handling with

associated mechanical and thermal stresses possibly contributed to the

final failure.

Consequences of such failure

Consequences of such a cell reversal, if it occurs in a battery servicing

a load, are loss in capacity and excessive gassing.  The major factor in

excessive gassing is the generation of hydrogen gas which can form

explosive mixtures in air.  Smaller amounts of hydrogen gas also be

formed during cell reversal, especially when the cell temperature is

raised.  Although this gas is highly toxic its smell can be detected at

very low concentrations.  The extent of capacity loss and gas generation

depends on the specific application, such as battery configuration,

battery cut-off voltage and most importantly initial discharge current in

the load profile.

Capacity loss 

The Coup de Fouet, which is a rapid voltage loss occurring at the very

discharge of a lead-acid battery, is critical when the initial discharge

current is very high.  The service load profile for the Nuclear

Generating Plant is characterized by a high initial current, which

generally lasts approximately on minute.  If the initial discharge

current is enough to lower the Coup de Fouet discharge voltage below the

battery cut-off voltage, then the battery will shut down almost

immediately and the deliverable capacity is very small.  In this case the

amount of hydrogen gas generated during reversal is also limited.

Hydrogen gas generation

Assuming that water decomposition is the main electrochemical reaction

following cell reversal, then the total amount of hydrogen gas generated

in the reversed cell can estimated based on the discharge current and

battery string capacity delivered before the cut-off voltage is reached,

according to:

          Volume of hydrogen in liters = 0.42 x delivered capacity in



Based on the available data, values for delivered capacity and the volume

of hydrogen gas, generated is estimated for two different initial

discharge currents.  The lower current represents the case where the Coup

de Fouet voltage is above the battery voltage, while the higher current

represents the case where the Coup de Fouet voltage is below the battery

cut-off voltage.  This example is based on one reversed cell in a 60 cell

string battery with 100% capacity and a cut off voltage of 105 volt.

          initial             available      amount of

          current             battery        hydrogen gas

                              capacity       generated

          500 A               60%            250 liters

          1000 A              < 1%           < 4 liters


     Test records from the manufacturer show that the cell serial #

97461, did pass the bond resistance test as per Round Cell specification

KS-20724 drawing L-905503.  This test is required to evaluate the harness

to positive plate lug weld.  The cell also passed a subsequent discharge

test specified by McGuire Nuclear Generating Station.  Most importantly

in the entire Round Cell manufacturing history of 800,000 cells and 25

years this is the first time that a complete disconnection of harness and

positive plate lugs have been reported as the cause of failure.  From

this Round Cell manufacturing history, combined with the weld test

results as the manufacturer, it is concluded that the cell failure

described in this ROD is an anomaly, occurring extremely rarely.


Page Last Reviewed/Updated Monday, March 30, 2020