VRLAmanual

A Guide To VRLA
Battery Formation
Techniques
www.firing-circuits.com
A Guide To VRLA Battery
Formation Techniques
By Mike Weighall and Bob Nelson
Mike Weighall is an independent Bob Nelson is an independent

consultant with 36 years’ experience consultant with over 23 years’
in the battery industry. He obtained experience in the VRLA battery
his Chemistry degree from the Industry. He obtained his Chemistry
University of Manchester Institute of degree at Northwestern University in
Science and Technology. He has 1963 and his PhD in Analytical
spent most of his working career Chemistry/Electrochemistry at the
associated with the battery industry, in a range of University of Kansas in 1966. After spending 11 years in
technical and managerial roles with major UK teaching and research at the university level, he joined
employers, including Lucas, Crompton, Cookson, and Gates Energy Products where he worked for 13 years in
ENTEK International. In recent years he has played an various positions dealing with the development and
important role in the ALABC (Advanced Lead Acid manufacture of both spiral-wound and flat-plate VRLA
Battery Consortium) as Chairman of the European products. He has also worked with other specialty VRLA
Technical Committee, member of the Research products during work tenures with Portable Energy
Management Team, and currently Chairman of the Products and Bolder Technologies. In between, he spent
Project Advisory Team on Separators. He has previously three years with ILZRO, where he was responsible for
written the “Battery Test Guide” for Digatron/ Firing organizing and managing the Advanced Lead Acid
Circuits. He has presented nine papers at International Battery Consortium. In addition to the publication of
Battery Industry Conferences, four of which were some 39 refereed papers, two book chapters and 22
published in the Journal of Power Sources. invited presentations at national and international
M.J. Weighall conferences during his academic career, he has
MJW Associates published 42 papers and given 41 presentations on
12 Low Stobhill VRLA battery technology over the past 23 years.
Morpeth
Northumberland Dr. Bob Nelson
NE61 2SG Recombination Technologies
Tel: +44 1670 512262 909 Santa Fe Drive
Fax: +44 870 056 0376 Denver
Mobile: +44 7977 459819 Colorado
Email: mjweighall@battery1.demon.co.uk CO 80204
Tel: +1 303 573 7402
Fax: +1 303 573 7403
Email: nelson909santafe@aol.com
© 2001 Firing Circuits, Inc.
www.firing-circuits.com Printed In U.S.A., All Rights Reserved

Table Of Contents
Paragraph Page Paragraph Page
1. Introduction ....................................................................... 1 8. Battery Design ................................................................ 26

8.1 Plate Height/ Plate Spacing Ratio ............................ 26
2. Plate Formation vs. Jar Formation .................................. 2
8.2 Battery Case Draft ................................................... 26
3. The VRLA Formation Process 8.3 Active Material Additives ......................................... 26
Jar Formation ........................................................................ 3
8.4 Electrolyte Additives ................................................ 27
3.1 The Filling Process ..................................................... 3
3.1.1 Acid Density for Filling ..................................... 4 9. Separator Optimization .................................................. 27
3.2 Fill-to-Form Processing .............................................. 4 9.1 Volume Porosity ....................................................... 29
3.3 Formation ................................................................... 4 9.2 Saturation Level ....................................................... 29
3.3.1 Battery Preparation for 9.3 Caliper ..................................................................... 29
Formation: Open Formation ..................................... 4
9.4 Compression ........................................................... 29
3.3.2 “Fill and Spill” Formation ................................. 5
9.5 Grammage ............................................................... 30
3.3.3 Saturation/ Electrolysis Formation ................... 5
9.6 Surface Area ............................................................ 30
3.4 Formation Time .......................................................... 5
3.5 Completion of Formation ............................................ 6 10. Separator Designs to Improve Wet Formation ............ 31
3.6 Formation Algorithms ................................................. 6
11. VRLA Gel Batteries ....................................................... 33
3.7 Initiation of Current Flow ............................................. 7
3.8 Constant Voltage Charging ........................................ 8 12. Formation Equipment and Layout ............................... 34
12.1 Battery Connections .............................................. 34
3.9 Constant Current Charging ........................................ 8
12.2 Formation Bay or Circuit Configurations ................ 35
3.10 Taper Current Charging ........................................... 9
12.3 Critical Maintenance of Formation Equipment ....... 36
3.11 Pulse Current Charging .......................................... 10
12.4 Power Quality and Equipment Costs ...................... 36
3.12 Rests and Discharges ............................................ 11
3.13 Sample Formation Algorithms & Profiles ................ 12 13. Battery Monitoring During Formation ......................... 37
3.13.1 A Simple Algorithm ...................................... 12 13.1 Electrical Monitoring .............................................. 37
3.13.2 More Typical Charge/ 13.2 Temperature Monitoring ......................................... 39
Rest/ Charge Algorithms ........................................ 12
13.3 Gas Monitoring ...................................................... 39
3.14 Development of a Suitable
Formation Algorithm ...................................................... 14 14. Post-Formation Handling and
In-Line Product Testing ...................................................... 40
4. Temperature Limits for VRLA Jar Formation ................. 15 14.1 Visual Standards .................................................... 40
5. VRLA Battery Manufacture using Plate Formation ....... 16 14.2 In-Line Product Testing .......................................... 41
14.2.1 Open-Circuit Voltage Measurement ............. 41
6. Technical and Theoretical Background .......................... 17 14.2.2 AC Impedance Measurements .................... 42
6.1 The Formation Process Explained ........................... 17 14.2.3 High-Rate Discharge Measurements .......... 43
6.2 Formation Processes and Ah Input .......................... 18
15. Troubleshooting: Problems and Solutions .................. 44
6.3 Key Differences Between
Flooded and VRLA Batteries .......................................... 19 16. References .................................................................... 48
7. Jar Formation – Additional Information ....................... 20
7.1 Battery Preparation for Formation –
Sealed Formation ........................................................... 20 Appendix 1:
7.1.1 Plate Curing and Carbonation ....................... 20 Glossary of terms and abbreviations ..................................... 49
7.2 Acid Filling ............................................................... 20
7.3 Control of Formation Temperature ............................ 23
7.4 Completion of Formation .......................................... 25
7.5 Alternative Jar Formation Options ............................ 25
Listing of Figures
Figure Page Figure Page
Figure 1. Examples Of Techniques For The Figure 13. 2.5 Ah And 25Ah Spiral-Wound
Initiation Of Formation Charging ............................ 7 Single-Cell Internal Temperatures
During Different Fill-To-Form Conditions. ............. 24
Figure 2. Examples Of Single-And Multi-Step
Current-Limited Constant-Voltage Figure 14. 6V/100 Ah Prismatic Battery
Formation Profiles .................................................. 8 Temperature Data During Fill-To-Form
Time With Different Conditions. ............................ 25
Figure 3. Examples Of Stepped Constant-Current
And Conventional CC Formation Profiles Figure 15. Solubility Of Lead Sulfate In
Compared To An Ideal Formation Curve. ............... 9 Sulfuric Acid At 25ºC. .......................................... 27
Figure 4. Examples Of The Progressive Influence Figure 16. Mean Pore Size Vs. Kr Bet
Of Temperature Monitoring On The Surface Area. ....................................................... 30
Efficiency Of The Formation Process. .................. 10
Figure 17. Impact Of Surface Area (m2/g) On
Figure 5. Taper Current Charging. ...................................... 11 Water Wicking Height While Under
20% Compression, After 24 Hours. ..................... 31
Figure 6. Examples Of Pulsed Charging
Algorithms. .......................................................... 11 Figure 18. Effect Of Fiber Mix And Segregation
On Vertical Wicking Speed. ................................. 31
Figure 7. Typical Constant-Current Formation
Profiles For A 12V/20Ah VRLA Battery. ................ 12 Figure 19. Upward And Downward Wicking
Height For Oriented And
Figure 8. Typical Constant-Voltage And Non-Oriented Fibers. ........................................... 32
Taper-Current Formation Profiles For
A 12V/20Ah Battery ............................................. 13 Figure 20. Battery Connections For Series
Strings, Series-Parallel Arrays And
Figure 9. Typical Constant-Current Formation Series-Parallel Matrixing. ..................................... 35
Profiles With Rests Or A Discharge
For A 12V/20Ah VRLA Battery. ............................. 14 Figure 21. AC Ripple Voltage And Current
Representation And Its Effect On
Figure 10. The Filling Process Within A Vacuum Cell Temperature And Cycle Lifetime. .................. 37
And Non-Vacuum Fill. .......................................... 21
Figure 22. Typical Self-Discharge Curves
Figure 11. Conceptual View Of the Filling For VRLA Batteries. ............................................. 41
Process For A VLRA Cell. .................................... 21
Figure 23. High-Rate Discharge Voltage/Time
Figure 12. Action On The Leading Edge Of The Curves For Acceptable And
Liquid In A VRLA Cell Filling Process. ................. 22 Unacceptable Battery Performance
On A 5-Second Test. ............................................ 43
Listing of Tables
Table Page
Table 1. Typical ampere-hour inputs in relation

to wet paste weight and dry cured
paste weight. ....................................................... 19
Table 2. Typical AC Impedance Values For A
Variety Of Thin-Plate VRLA Single Cells
And Batteries Fully Charged At 25ºC. .................. 48
Table 3. Sample OCV Chart Used In
Manufacturing To Sort Cells Or Batteries
After Formation Or Recharge. ............................. 49
1. Introduction
The purpose of this brochure is to guide the battery manufacturer in the formation of
VRLA (Valve Regulated Lead Acid) batteries. The information is nominally confined to
“small” VRLA batteries with capacities in the range 1.2 Ah to 100 Ah. Because “jar”
formation of VRLA batteries is far more difficult than plate formation, this aspect of VRLA
battery formation will comprise the largest section of the brochure.
The brochure is divided into several sections:

■ Sections 2 through 5 deal with practical issues related to VRLA battery
formation, and deal mainly with jar formation.
■ Section 6 deals with the technical and theoretical background.
■ Section 7 gives additional information about jar formation.
■ Sections 8 through 10 give battery and separator design guidance.
■ Section 11 is a brief overview of VRLA gel batteries.
■ Sections 12 through 14 deal with formation equipment, battery monitoring
and product testing.
■ Section 15 deals with troubleshooting formation problems.
www.firing-circuits.com 1
Practical issues related to VRLA battery and jar formation
2. Plate formation vs. jar The decision as to whether to plate form or jar form will
formation for VRLA batteries be based on a number of factors which the battery
manufacturer needs to take into account, and will be
The first decision the VRLA battery manufacturer has to
discussed in more detail later. Some general guidance is
take is whether to use a plate formation or jar formation
given below:
process, and this section highlights some of the issues
that need to be taken into account before making this Plate for mation – should be used in the following
formation
decision. circumstances:
circumstances:
■ Plates for tall batteries
Plate formation may result in fewer manufacturing and
technical problems in respect of battery design, process ■ Plates for large, high capacity batteries
control, quality, performance and life. The merits of plate ■ Plates for very long life batteries
formation are particularly apparent for larger, higher ■ Battery with high L/d ratio (>100)
capacity batteries where a long cycle life and/or (see section 10.1)
calendar life is required. However, particularly for the
smaller batteries being discussed in this brochure, Jar formation – consider in the following cir
formation cumstances:
circumstances:
many battery manufacturers are choosing jar formation. ■ Cylindrical battery design
This may be for reasons of cost and convenience, but ■ Thin plate prismatic battery design
may also be because the battery design does not lend ■ Battery with low L/d ratio (<100) (see section 10.1)
itself well to plate formation. This may apply for example
■ Large separator fringe area
with thin plate cylindrical or prismatic battery designs.
■ High separator grammage (>=2g/Ah)
The decision as to which process to use will be dictated
by the detailed battery design and manufacturing ■ High surface area separator
constraints as described in more detail later. However,

Other issues, which also need to be consider ed for jar
considered
there are other manufacturing and cost issues to be
for mation:
formation:
considered. The total cost of the plate formation/dry
■ Whether single cells or monoblocs, and how
charge process may be higher than jar formation when
one takes into account the following factors: many cells in the monoblocs (e.g. 3 or 6 cells).
■
This will have an impact on the efficiency of
The cost associated with the neutralizing and
cooling and temperature variations between cells.
cleaning or disposing of the plate wash water.
■ If plates have been cured to produce high levels
This water must be neutralized and cleaned of
heavy metals before it can be recycled or of tetrabasic lead sulfate (4BS) the plates may be
discharged into a public sewer system. more difficult to form, and require a higher charge
■
input during formation than for tribasic lead sulfate
The capital and operating cost of the dry charge
(3BS) cured plates.
operation (e.g. inert gas drying).
■ The inclusion of red lead in the positive paste mix
■ Post assembly charge and discharge cycles to
will assist jar formation and enable lower Ah input
recover the capacity loss that is inherent in the
and shorter formation times. It will also improve
dry-charge process.
the initial electrical performance.
■ Plate lug cleaning before final assembly.
In principle all VRLA batteries could use plates prepared
using plate formation/dry charge: but not all VRLA
batteries can be successfully jar-formed. The information
above is given for guidance only, and the suitability of a
particular battery for jar formation should be established
by careful experimentation.
Jar formation of VRLA batteries is actually quite a
complex process and will now be dealt with in detail.
2 www.firing-circuits.com
3. The VRLA Formation Process – reproducible in terms of fill weight and the final
Jar Formation saturation level is typically ~95% (i.e., the plate stack
doesn’t saturate).
3.1 The filling process
The formation process for valve-regulated lead-acid “Push” fill is a specialized technique for spiral-wound-
(VRLA) cells and batteries really begins with the filling type products where electrolyte is forced up through the
process. Several approaches can be taken to filling, wound element, either from the bottom or using a probe
including: in the wound-element mandrel space. This is faster than
■ Gravity top fill, single or multi-step the gravity-fill techniques (around 30-60 seconds) and,
■
thus, requires more care in thermal management.
Gravity bottom-up fill
■ “Push” fill where electrolyte is pumped into the Soft-vacuum filling involves drawing a moderate vacuum
cell or battery, usually from the bottom up (usable level and allowing the element to “suck in” electrolyte at
only with spiral-wound products) its own rate. As this approach doesn’t usually result in
■ Soft-vacuum fill (>~20mm Hg), single or multi- uniform electrolyte distribution there is often a “push-
step, possibly with a “push-pull” step to distribute pull” (pressure-vacuum) finishing step to physically
electrolyte more evenly move electrolyte around to help diffusion. The filling rate
■
is moderate (30-60 seconds) so thermal management is
Hard-vacuum fill (<~10mm Hg)
mandatory, along the lines of that given above for the
The first, a gravity top fill, is the simplest approach that gravity-fill approach.
can be used for any cell or battery (hereafter referred to, Hard-vacuum filling is a very rapid technique (on the
collectively, as “batteries”) and just involves pouring order of 1-10 seconds for sizes 1.2-25Ah) and is, thus,
electrolyte into the headspace at a rate that the battery attractive for high-volume manufacturing. However, it
can accommodate. It can be done slowly with a single also requires extreme care both during filling and for
addition or in several measured amounts. This is a processes prior to filling. In addition to speed, hard-
relatively slow process but it has advantages in that heat vacuum filling can result in uniform electrolyte
is generated slowly, there is not likely to be damage to distribution due to the almost total absence of air
the AGM separator and there is only a limited effect from displacement. However, the absence of air also means
carbon dioxide released from carbonated paste that the paste is very reactive and the rapid introduction
surfaces. There is the possibility of incomplete wetting of electrolyte results in very high heat generation over a
due to trapped gas pockets. Heat generation in larger short period of time. Thus, thermal management is
batteries can be counteracted by chilling the electrolyte critical with this type of filling and it is impractical over a
(typically to 0 to –10oC) and/or the unfilled elements and, size of ~50Ah due to the inability to dissipate the heat
if necessary, putting the filled battery into a chilled water rapidly generated, even with the chilling steps
bath. The measures used depend in large part upon the mentioned above. Poor thermal management can result
size of the battery. For small products (1.2-10Ah), simple in staining of the AGM separator by dislodged paste
bath cooling after fill is sufficient (and this may not even and/or plate deformation and case bulging due to heat
be necessary for very small products). For larger sizes and, possibly, steam generated from the filling reaction.
(10-100Ah), chilled electrolyte and bath cooling may be In this type of filling, the formation of hydration shorts
mandatory. Fill times are of the order of 10-40 minutes. (lead sulfate in the separator) is also possible due to the
Gravity bottom-up, or “dunk,” filling simply involves high temperatures and low acidity conditions that can
dipping a cell or battery into a bath of electrolyte (the be generated. Plate carbonation during processing is
case having a hole or holes in the bottom to allow also a problem because the rapid introduction of
ingress of acid) until wicking has resulted in complete electrolyte can result in a “burst” of liberated carbon
filling of the separator and plate pores. This is also a dioxide, which can help to defeat the vacuum created
slow process (several minutes), and has the advantages and result in low fill weights. Further liberation of CO2
and drawbacks listed above for gravity top fill. An added can cause regurgitation of electrolyte in extreme cases.
disadvantage is that the filling hole has to be sealed Separator damage can also result from the hydraulic
before the battery goes into formation. In fact, simply action of the electrolyte if it is added too quickly, thus
letting the battery take as much acid as it wants is very promoting plate-to-plate shorting due to the removal of
overlapping separator between adjacent plates.
3.1.1 Acid Density for Filling 3.3 Formation

This will depend on the battery application, the desired There are a number of factors to be considered in
final density, and the amount of sulphation achieved matching the correct formation algorithm to a given
during paste mixing. A typical filling acid density would product, among these being:
be 1.26 with a final density of 1.28. Finished acid ■ Product sealed or open?
densities are normally in the range 1.28 to 1.32, ■ Temperature control and the use of air or water
depending on the application. Standby batteries tend to ■ Formation time
have lower densities (1.28) while high-rate batteries
■ Desired level of completion of formation
(aircraft, engine start) may have a higher density
(~ 1.30). For deep cycle batteries the specified acid (i.e., % PbO2)
■ Formation algorithm used (CC, CV, taper, pulse,
density may be in the range 1.28 – 1.32. With open
formation the acid s.g. can be checked and adjustments rests, discharges?)
made if necessary. For sealed formation, calculations ■ Battery connection series-strings only, series-
need to be accurate as it is not possible to correct the parallel strings or series-parallel matrix?
acid s.g. after formation. Some VRLA battery ■ Monitoring parameters during formation
manufacturers monitor the fill/ formation weight loss and ■ Critical maintenance of formation equipment
then add an equivalent amount of water or dilute ■ Post-formation handling and product testing
electrolyte before sealing the battery.
3.3.1 Batter
Batteryy Preparation for For
Preparation mation:
Formation:
3.2 Fill-to-Form Processing Open For mation
Formation
The time gap between electrolyte filling and initiation of
For jar formation the easiest technique is “open”
the formation process is more important than may be
formation, which usually means a condition where the
realized. Even though VRLA batteries are electrolyte-
vent valve has not been put in place. (The alternative,
deficient and there is more than twice the amount of
less frequently used technique of sealed formation, is
pastes as there is acid (on an ampere-hour basis), a
discussed in section 7.1). Open formation may also
significant amount of acid remains unreacted if batteries
indicate a formation where the battery headspace is
are put into formation immediately after filling. The longer
open to the air. In either case, batteries are usually
the delay between fill and form the more lead sulfate is
flooded, or close to it, and have the capability of removal
formed. This facilitates the formation process, but it also
or addition of acid during processing. Open formations
increases the resistance of the unformed plates
are useful in that plate processing is not as critical (in
(particularly the positive), as lead sulfate is an insulator.
terms of carbonation), heat dissipation due to gassing is
This is usually overcome by using 10% or more of red
greater by about an order of magnitude than in sealed
lead, Pb3O4, in the positive paste. Longer stand times
formations (because the battery is formed in the flooded
after filling can also aggravate the conditions that can
state) and adjustments in saturation levels are possible
initiate hydration shorts by allowing lead sulfate to slowly
at any time. There are several approaches to open
dissolve and diffuse into the separator. With a good filling
formation for VRLA batteries, among them being:
process this is not a problem, as even a mildly acidic
■ Saturated or near-saturated condition with
condition will suppress lead sulfate solubility, particularly
provision for excess electrolyte handling;
if sodium sulfate is used as an additive in the fill
■ So-called “fill-and-spill” formation, where batteries
electrolyte. However, in cases where fill conditions
resulting in areas of the plate stack where hot electrolyte are formed saturated and then the electrolyte
depleted of acid can exist, batteries are put onto level is adjusted at the end of formation; a variant
formation as quickly as possible after fill. When this is of this is two-step formation, where the battery is
done, there is the danger of the battery overheating, as first formed with dilute electrolyte which is
the formation process generates heat, particularly early replaced after formation with a higher specific
when high plate resistances result in high I2 R heating, gravity acid closer to the desired operational
and additional heat is still being created by the ongoing level;
filling reaction (which is very exothermic). ■ Saturated or near-saturated formation open,
followed by saturation and electrolysis to achieve
In order to allow the filling reaction to go to completion
a target saturation level, usually ~95%.
and allow the battery to cool adequately, a fill-to-form
time of between 2 and 4 hours is recommended.
3.3.2 “Fill and Spill” For mation

Formation 3.4 Formation Time
This formation approach involves formation in a flooded Very early work done by Ritchie at Eagle-Picher showed
state, followed by simple pouring off of excess that flooded lead-acid batteries could be formed in ~2
electrolyte. This results in a near-saturated condition weeks to a very high PbO2 level with minimal weight loss
following formation (trapped gas in the plate pores and consumption of just over the theoretical ampere-
ensures that some electrolyte is absorbed and, thus, hour input of 241 Ah/kg of PbO. This is not feasible in
there is a small amount of void space in the formed large-scale manufacturing, but it does set a baseline
battery), which may result in higher-than-usual against which more practical formation algorithms and
overcharge gassing and weight losses early in life and, times can be measured. Suggestions for typical ampere-
possibly, acid leakage during heavy overcharge. This hour inputs in relation to wet paste weights are given in
approach has been promoted by H&V for use with their section 6.2. In practice, small VRLA batteries are formed
Hovosorb II organic fiber/glass separator [1] (see also within 24-48 hours. The actual formation time will be
section 10) and is particularly well suited to dependent on a number of factors, but a general rule of
manufacturing processes with high manual labor inputs thumb is that cell/battery capacities of ~20Ah or less
and those where precise control over finished battery require a roughly 24-hour formation, while larger sizes
quality is not required. The saturation levels in the cells up to 100Ah can normally be formed in ~48 hours.
are not precisely known and significant cell-to-cell Smaller batteries are easier to form because they are
variations could exist. Heavy hydrogen gassing during more compact and voltage drops across the plates are
formation must be accounted for, but the high levels of less. They also tend to have better heat-dissipation
gas generation help with heat dissipation. properties and can be formed at higher currents.
In principle, a two-step formation could be used for Formation time is not the only criterion. Heat buildup is
VRLA products as is done for flooded lead-acid an issue that will tend toward longer formation times.
batteries. Here, the battery is filled with a relatively dilute High PbO2 levels move the time in the same direction,
electrolyte for the initial formation process, after which as does the requirement for lower formation weight
the forming acid is dumped and the battery is refilled losses. The positive plate (PbO2 level) is the key
with an electrolyte close to the desired final specific indicator of the completeness of formation (see also
gravity after a finishing charge. The major problem for section 7.4). In fact, the negative active material, or
VRLA batteries is that the AGM separator (unlike flooded NAM, forms relatively easily and it is rare that formation
lead-acid separators) holds most of its electrolyte and of the NAM is the limiting factor. If manufacturing
any manipulation of the formed battery is likely to result throughput were not an issue, all formations would be
in separator damage. done over several days, as most benefits flow from long
formation times. However, formation throughput is of
3.3.3 Saturation/Electrolysis For
Saturation/Electrolysis mation
Formation paramount importance and so the goal is almost always
In order to have an accurately known saturation level in to achieve complete formation in a minimum amount of
the region of 95% after formation, a method has been time. In order to accomplish this in large-scale
developed where a standard open formation is carried manufacturing, it is important to have a deeper
out, followed by over-saturation and pouring off of understanding of the chemistry of the formation process,
excess electrolyte (much like “fill and spill” above). The and more detail is given in section 6.1. The chemistry
fully-saturated, formed battery (still open) is then involved in formation is fairly complex, consisting not
subjected to a period of electrolysis at a known current only of the basic chemistry of conversion of lead sulfate
level to drive off an accurately-known amount of water, to sponge lead and lead sulfate, coupled with the
thus getting the battery to the desired saturation overcharge processes involving the decomposition of
percentage, after which completion of battery assembly water, but also more subtle issues such as electrolyte
can be carried out. diffusion and gas bubble formation.
3.5 Completion of formation 3.6 Formation Algorithms

One or more of the following parameters can be The best algorithms to use in forming VRLA batteries
measured to determine whether formation is complete: depend upon a number of factors, ranging from capital
■ Battery, cell and individual electrode potentials investment to desired product quality and the intended
become high and constant. Electrode potentials application. The battery manufacturer will need to carry
are determined using either cadmium wire or out tests to establish the best algorithm for his specific
mercury/mercurous sulfate reference electrodes manufacturing process and battery application, but the
on test batteries. following guidelines may be useful.
■ Top-of-charge voltages (TOCV) become constant, The optimum algorithm is likely to include a number
but will vary for individual VRLA batteries, of steps:
depending upon the amount of oxygen ■ Low initial current to minimize temperature rise at
recombination occurring at the end of formation; the start of formation. There may be a
the more oxygen reduction taking place the lower continuation of the heat production from the
the TOCV. This will be influenced by whether oxide/acid filling reaction. There may also be a
sealed or open formation is used. In fact, if a variable fill-to form hold time due to the time lag in
given formation system monitors TOCV values filling a formation circuit queue. The low initial
these can be used for matching small batteries current will compensate for possible high battery
into larger units, as this is a critical performance resistances. The low current charge should be
parameter. continued until the battery temperatures have
■ At the same time, electrolyte specific gravity fallen below 50°C.
becomes constant at some high level (relative to ■ One or more steps at a higher current during the
the starting density) due to conversion of sulfate main part of the formation process while battery
in the unformed plates to sulfuric acid. resistance is at its lowest and heat generation is
■ Both plates gas uniformly and strongly. at a minimum.
■ Temperature rises steeply toward the end of ■ One or more steps at a lower current as the
formation at a given applied current, reaching gassing phase is reached towards the end of
values as high as 65-75oC if the finishing current formation.
is not reduced. ■ The formation process may also include rest
■ Internal examination of a cell or battery would period(s) and/ or discharge step(s).
show that the plates are uniformly colored and are
without white spots (unformed lead sulfate); the The chosen charging approach will usually depend on
negatives are soft with a gray metallic sheen and the desired amount of capital investment and/ or
the positives are hard and are dark brown to experience from making flooded lead acid products.
black in color. The possible choices are:
■ Constant-voltage (CV)
In practice, most of these parameters cannot be ■ Constant-current (CC)
measured routinely, particularly in a large matrix of
■ Taper-current (TC)
batteries in a formation bay. With a given VRLA product,
■ Pulsed-current (PC)
the normal approach is to carry out test formation
algorithms on a few batteries, which are monitored and
When making the choice of appropriate charging
then autopsied in order to determine if they meet the
equipment, it should be noted that the charging
above criteria. Following the establishment of a
equipment from Digatron/ Firing Circuits Inc., offers
production-worthy formation algorithm, all batteries are
computer control with optional battery monitoring to
then manufactured using it, with some form of periodic
provide optimum control of charging parameters.
sampling and testing to ensure the desired level of
quality and uniformity. More information about each of these processes is
given on the next page.
Additional information is given in section 7.4
3.7 Initiation of Formation Current Flow Figure 1

Before formation current flow is initiated, most
manufacturing operating procedures include so-called Examples Of Techniques For The Initiation
“continuity checks,” where individual battery strings are Of Formation Charging
checked with an ohmmeter to ensure that the resistance
of the string, while very high, is not infinite (indicating a A. Low-current initiation
battery with an open connection (poor COS or squeeze
weld) or a defective formation connector lead-to-terminal Current,
amperes
contact). If an abnormally high or infinite resistance
reading is taken, the formation room personnel must
identify the source of the problem; otherwise, a complete
string of batteries will not be formed and, in a series-
parallel array, the other strings will receive too-high
current levels.
Initiating current flow can be difficult if the plates are Formation time, hours
heavily sulfated and/or the fill-to-form time has been

long, thus depleting most of the sulfuric acid in the B. Ramp-current initiation
electrolyte and raising the liquid resistivity. The use of
red lead in the positive paste, carbon in the negative Current,
amperes
paste, as well as sodium sulfate in the fill electrolyte will
help to minimize the high initial resistance.
If a high-inrush current level is applied when the initial
battery resistance is high, the voltage will be driven to
high values and most or all of the current will go into
heat generation and gassing. After a period of time
Formation time, hours
these processes will diminish and conversion of lead
sulfate to the active materials will take over and the
formation will proceed in a normal fashion. However, a C. Abrupt or high-current initiation
short initiation charge period of 0.5 to one hour at low
Current,
currents can be applied in order to generate some acid amperes
and improve the conductivity in the plates, or the current
can be ramped up slowly over an hour or so, as shown
in Figure1
Figure 1, before the main formation current is applied.
Initial resistance to proper current flow can be detected
either by an immediate rise in charge voltage to very
high levels (or to the voltage limit if constant-voltage
Formation time, hours
charging is used) or by a relatively sharp temperature
increase. Unless the battery plates are very heavily
sulfated, after a short period the voltage and
temperature will drop to normal levels, which are
temperatures below ~50oC and voltages of ~1.8-1.9
volts per cell. There will then be gradual rises in both
temperature and voltage, but because almost all of the
formation current is going into lead sulfate conversion
these increases will be very gradual.
3.8 Constant-Voltage Charging Figure 2

Current-limited constant-voltage (CV) charging is
commonly used for cyclic charging of VRLA batteries, Examples Of Single- And Multi-Step Current-Limited
but its utility in formation is more limited, largely due to Constant-Voltage Formation Profiles
cost and its effect on product uniformity. Precise voltage- A. Single-Step Current-Limited Constant-Voltage
control limits are expensive in terms of formation Formation Profile
electronics and in forming VRLA batteries they are not
necessary. The primary drawback is that, as in CV Voltage/
recharge, the current taper toward the end of formation Current
results in relatively long charge times. In order to
minimize this and speed up formation times, multi-step
CV algorithms can be used, by programming for current-
limit reductions when the voltage limit is reached. This
then becomes a stepped constant-current formation, but Formation time
with a voltage limit (usually ~16V for a 12V battery) to B. Multi-Step Current-Limited Constant-Voltage
minimize gassing and grid corrosion. Typical examples Formation Profile
of single-step CV and stepped-CV/CC algorithms are
shown in Figur
Figure e 2. The last step usually allows for a
current taper when the voltage limit is reached, the Voltage/
Current
duration depending upon the desired formation time.
There are a number of advantages and drawbacks to
CV charging. On the plus side, overcharge is minimized
due to the current taper during the finish of formation
and so the charge efficiency is relatively high and
Formation time
concerns about gas monitoring and ventilation are less
= Formation Charge Current
important. Balanced against this (in addition to cost) are = Formation Charge Voltage
a number of drawbacks:
■ With a significant time in the current-taper mode,
the total Ah input must be integrated electronically 3.9 Constant-Current Charging
(rather than simply timed as with CC formation).
In CC charging, voltage control is not required (although
■ In single-step CV, the long charge “tail” lengthens there is always a voltage limit such as 2.80 volts per cell)
formation time significantly. and this reduces the cost of the charging equipment.
■ Because voltage is applied to long strings or Using single-step or two-step formations at high currents
series-parallel arrays as a multiple of a given can also reduce formation times, but this results in lower
volts-per-cell, actual charge voltages for each cell charge efficiencies and large amounts of overcharge
can be highly variable. More seriously, paralleled and gassing. The most common approach is to use a
strings can draw different currents based upon single-step CC algorithm, possibly with one or more rest
their cumulative DC resistances; this can have the periods (see below) or discharges. This is not efficient,
effect of routing high currents through individual since at low currents overcharge is minimized but total
strings early in formation, which can result in large formation time is long whereas with high currents the
imbalances of total ampere-hours passed through forming time is shortened but the overall charge
different strings. In the extreme, this can result in efficiency is reduced. More innovative, multi-step
strings with low initial DC resistance going into algorithms are now in use where relatively high currents
thermal runaway, particularly for large batteries are used early in formation and lower finishing currents
with poor heat-dissipation capabilities. are then applied, either as a two-step or multi-step
■ If strict voltage control is desired, temperature- algorithm. Dramatic gains in charging efficiency can be
compensated charging must be used, which realized, as shown conceptually in Figur
Figuree 3 [2]. In some
further increases cost. Temperature issues can be cases this is done as a fixed, programmed algorithm
minimized by using lower charge voltages, but with defined current levels for pre-set time intervals.
this will increase formation times significantly. Other approaches involve monitoring of battery
parameters in order to apply optimal current levels for Figure 3

as long as possible. One example of this is shown in
Figure 4 [3]
Figure [3], where battery temperature is used as the Examples Of Stepped Constant-Current And
control variable. As can be seen, this allows for an initial Conventional CC Formation Profiles Compared To
high CC level, followed by step-downs to lower currents An Ideal Formation Curve.
Note The Disparity In Overcharge Amounts
based upon battery temperatures.
Stepped Constant-Current
The major advantages of CC charging are that it is
easily programmable, it is relatively rapid and the total I
ampere-hour input can be determined easily. In addition, Ideal current curve
the current level is controlled, so even in series-parallel Several-step current curve
arrays battery damage due to high charge currents as
noted above for CV formation is largely avoided.
However, several drawbacks also apply:
■ Single-step CC formation is either very lengthy
(low current) or very overcharge-intensive (high
current).
■ Heavy overcharge results in high heat production,
grid corrosion and gassing.
Current curve of several-step formation t
■ Voltage regulation on charge is not possible,
Conventional CC Formation
except for the high upper limits used (2.8-3.0 V/
Ideal current curve
cell or more) I
Several-step current curve
On balance, this is the simplest approach to formation
and is the most commonly used, particularly in multi-
step algorithms. As programmable controllers for
formation systems are now commonly available and
inexpensive these approaches, though seemingly
complex as shown in Figur
Figuree44, are very straightforward.
They are also a more tolerant approach when poorly
regulated input (i.e., “dirty”) power and/or inexpensive
formation electronics are used. Current curve of conventional formation t
3.10 Taper-Current Charging
Taper-current (TC) charging for formation combines formation current decreases because of the decreasing
some of the best aspects of CV and CC approaches voltage difference with the power-supply setting. Unlike
and is probably the least expensive of the three. As it is CC charging, when the battery array voltages climb into
not a common approach, Figur e 5 shows a typical circuit
Figure the gassing region the charge current is tapering off.
for TC charging, along with a typical charging curve. However, the current does not taper off as sharply as
This is really the simplest of circuits. A power supply is with CV charging because of the higher voltage setting.
wired in series with a load resistor and a battery string or Moreover, if the TOCV is low for the formation array due
strings to be formed. If desired, some form of sensing of to an unusually high oxygen-recombination current draw
battery parameters (voltage, temperature, etc.) can be the current at the end of formation will increase due to
included to provide feedback control. When formation is the widening gap between the power supply voltage,
initiated, current flows according to the rating of the load which is fixed, and the decreasing end-of-charge
resistor and the voltage difference between the power voltage. Initial and final currents are roughly set by
supply setting (typically a high value of 2.6-2.8 V/cell) choices of power-supply settings and resistor values.
and the battery array (which will initially have a very low These yield an approximate formation voltage/time
voltage). At the beginning of formation, the voltage profile, but the exact shape of the curve will vary
difference is great, on the order of ~1V/cell, and the considerably depending upon the charge-acceptance
inrush current is relatively high, as in CV charging. As properties of the battery array being formed. This can be
the cumulative battery array voltage climbs the viewed as both a strength and a weakness of this
Figure 4 approach. It is well suited to VRLA formations, as many

products do not require precise voltage control. In
Examples Of The Progressive Influence Of Temperature summary, TC charging has the following advantages:
Monitoring On The Efficiency Of The Formation Process. ■ Capital input for formation charging equipment is
low if inexpensive power supplies are used;
A. Normal formation. The batteries are placed in a
non-flowing water pool. however, poorly regulated supplies can not only
transmit, but also at times amplify, line fluctuations
to the charging system.
■ It allows for a relatively fast formation with only
Temperature degree C
moderate overcharge amounts by allowing high

Voltage V
Current A
inrush currents (relative to CC) as well as high

finishing currents (relative to CV).
■ To some extent, the formation profile is variable
according to the charge-acceptance
characteristics of the battery array being formed.
Time
There are also a number of drawbacks, as follows:
■ Because the current tapers, Ah input must be
B. Formation in flowing coolant, with maximum initial current.
Limit set on voltage but not on temperature. determined by electronic integration.
■ The amount of overcharge is high relative to CV or
multi-step CC charging.
■ Voltage and current are not controlled, so the
uniformity of formation profiles for different battery

Voltage V
Current A
lots is not high; total Ah input may vary

significantly with product uniformity going into
formation.
■ The use of unregulated power supplies can result
in shortened lifetimes in service.
Time
3.11 Pulsed-Current Charging
Pulsed-charge algorithms can be applied to the
C. Formation with initial maximum current limit,
followed by temperature and voltage limits. formation of VRLA batteries just as it is used in charge/
discharge service; typical algorithms are shown in
Figur
Figure e66. As can be seen, profiles analogous to CV or
TC charging as well as pulsed CC can be used. Note
that in the “off” periods, rests or partial or complete

discharges can be applied. The discharges are thought
Voltage V
Current A
to be beneficial in eliminating surface charges from the

plates, which can result in lower gassing levels; it has
not been unequivocally established if this is, indeed the
case. A good deal of work has been done on using
pulsed methods, but it remains unclear whether product
Time
quality gains can be realized with this approach. There
= Current are clear advantages in enhanced heat dissipation while
= Voltage allowing the use of relatively high currents (even late in
= Battery Temperature
= Coolant Temperature formation) and in reduced gassing due to the reductions
in coulombic input per pulse as the gassing region is
approached late in formation. While most battery
companies have investigated this for the above reasons,
it is not commonly used.
Figure 5 plate pores. The charge efficiency of the positive

electrode is relatively low even when completely formed,
Taper Current Charging. but in formation itself it is so poor that gassing of oxygen
Typical TC Circuit can begin after only a few hours, or even less. Later, the
Switch
negative plate will also begin gassing and in both cases
this hampers proper conversion of unreacted lead
oxides deep in the plates to lead sulfate and then
RTaper subsequent reaction of the sulfate to the active
+
Power + V materials. The first step requires acid generated at the
–
Supply – Sense plate surfaces early in formation to penetrate into the
plate interiors and the second reaction requires water to
produce sponge lead and PbO2. When either or both of
Power Supply Voltage - Cell Voltage (V) the plates goes into gassing, this will force liquids out of
Charge current ( I ) =
Load Resistor ( RTaper )
the plate pores and into the glass-mat separator;
eventually, with heavy gassing much of the electrolyte
Typical TC Current/Voltage/Ah Curves will be forced into the head space or even out of the
battery as regurgitated acid or acid spray.
I E
These conditions are easily avoided by inserting one or
Ampere-Hour Input
Charge Current
Charge Voltage
Ah more rest periods or discharges into the formation

algorithm. In both cases, when the charge voltage is
removed gassing ceases and time is allowed for water
and acid to diffuse into the plate interiors. This allows for
reaction of acid with any PbO remaining after filling and
fill-to-form. When formation is reinitiated more lead
Time
sulfate has been generated and water is present as a
part of the filling reaction. When formation is continuous
3.12 Rests and Discharges gassing seriously impedes these processes. Thus, use
of significant “off” times can actually result in faster, more
One of the major electrochemical problems in using any
complete formation processes. Rests or discharges can
of the above approaches in a continuous way (with the
be put in at fixed points in formation or they can be
exception of pulsed charging) is that gas generation can
initiated when a “trigger” voltage is reached. These
severely impede the efficiency of the formation process
considerations probably apply more to thicker-plate
by retarding the diffusion of acid and water within the
Figure 6
Examples Of Pulsed Charging Algorithms.
In all cases, the coulombic input decreases as the top-of-charge is approached.
+ +
Current
Current
– –
Time Time
Constant Period, Decreasing Amplitude Constant Amplitude, Decreasing Period
■ Pulse/Rest
■ Pulse/Discharge
■ Pulse/Discharge/Rest
products (2.0 mm or more) than those with thin plates Figure 7

(where diffusion paths are shorter and plate wetting is
more efficient due to the higher surface areas). Typical Constant-Current Formation Profiles
For A 12V/20Ah VRLA Battery.
Which approach is better? Discharges are clearly more
complex in terms of capital equipment and they will A. Voltage, Temperature and Gassing Curves for
lengthen formation time relative to rests due to the a One-Step CC Formation
Voltage, Temperature, Gassing Rate

requirement for replacing charge taken out during the
Formation Current, Amperes

discharge. Discharges are thought to be beneficial
because, in principle, they should increase the
porosities of the plates and further aid acid and water
penetration, as well as improve post-formation
discharge capacities. Little documentation is available
comparing the effects of rest periods and discharges, 2
so the technologist is left to weigh the possible benefits
given above against the significantly higher costs of
discharge equipment. Both are clearly beneficial in 0 4 8 12 16 20 24 28 32 36
reducing formation weight losses and in improving Formation Time, Hours
finished-product quality. For VRLA batteries requiring

high post-formation PbO2 levels (90% or greater) and B. Same Curves for a High-Inrush Two-Step
long shelf lives (low residual PbO levels in the formed CC Formation Algorithm
positive plates) the use of one or the other is almost

8
mandatory.
3.13 Sample Formation Algorithms & Profiles

3.13.1 A Simple Algorithm
Sample formation profiles will now be considered that
might be recommended for a typical 12V/20Ah VRLA
2
product. The simplest approach would be a single-
stage CC formation over, for example, 36 hours with a
total Ah input of four times the rated capacity, or 80Ah. 0 2 4 8 12 16 20 24 28 32 36
Over a 36-hour period this would be a CC level of Formation Time, Hours
~2.2A, as shown in FigurFiguree 7a. This approach results in
= Formation Current
relatively high temperatures toward the end of = Voltage
formation and large overcharge amounts and gassing = Temperature
= Gassing Rate
levels, but it will form the battery successfully. The pore
structure may not be optimal due to the low initial
current and so a modification of this would be to use a
two-step CC algorithm with, say, 2 hours at 8A (16Ah) 3.13.2 MoreT
More ypical Charge/Rest/Charge Algorithms
Typical
followed by 34 hours at 1.88A. (Figur e 7b). For a CV
(Figure An intermediate level of complexity can be applied
formation, somewhat more time may be required or a without going to the type of feedback approach used in
high inrush current may be needed, accepting a Figur
Figure e 3 (which is certainly acceptable). In this “typical”
somewhat lower charge input at 36 hours, as shown in case, two rest periods are introduced into the 36-hour
Figur
Figuree 8a. In order to increase the charge input toward formation with the focus on CC charging. The rest
the end of formation a taper-charge algorithm may be periods can be shifted toward the end of formation as
used, as shown in FigurFiguree 8b. This has a high inrush there is, initially, a great deal of lead sulfate from the
current as with CV but the current only tapers to ~30% filling process and it will take some time to consume this
of its initial value. This results in a higher Ah input, but material. (Given the current level imposed and the
also higher temperatures and more gassing (weight amount of lead sulfate expected to be present, a rough
loss) in the final 12 hours or so. time period can be calculated to where significant
Figure 8 during the rest of the formation. After this, a fixed CC

level can be used in combination with the two rest
Typical Constant-Voltage And Taper-Current Formation periods, as shown in Figur e 9a. The rest periods are
Figure
Profiles for a 12V/20Ah VRLA Battery beneficial not only in providing time for electrolyte
penetration but also for keeping the temperature down
A. Voltage, Temperature and Gassing Curves for
compared to a continuous one- or two-step CC
a One-Step CV Formation
algorithm. Because the time spent in overcharge and

resultant gassing is lower overall with rest periods
(even though the charging current is higher to

compensate for the 4-hour off time), weight losses are
also reduced somewhat.
If a discharge were to be used instead of the two rest
periods, it would be most beneficial to have it near the
end of the formation, as shown in Figur e 9b. As can be
Figure
seen, only a partial discharge is carried out; a
0 4 8 12 16 20 24 28 32..........48 complete discharge would obviously be more effective
Formation Time, Hours in promoting pore formation and electrolyte
penetration, but it would also require substantially more
B. Same Curves for a One-Step Taper-Current time for the full discharge and subsequent recharge. If
Formation Algorithm this were done within the 36-hour schedule time it
would require much higher charge current levels but it
could be done. However, as noted, there is no clear

evidence indicating that a discharge is more effective
than rest periods. One advantage for a discharge is
that it could be used as a matching tool for building
battery modules into high-voltage packages, using
discharge capacity and top-of-recharge data recorded
during formation. This would, of course, require that all
batteries be monitored and that the data be collected
0 2 4 8 12 16 20 24 28 32 36 and processed. The major cost, however, would be for
Formation Time, Hours equipment to carry out the discharges; in addition, if
the formation time were extended this would reduce
= Current
= Voltage
the battery throughput level somewhat and would
= Temperature require more formation stations to process the same
= Gassing Rate
number of batteries. However, finished battery quality
and uniformity would be improved significantly.
These are just a few examples of formation algorithms
gassing will begin; alternatively, voltage can be
that might be employed for the processing of VRLA
monitored, as shown here, and “trigger” levels used to
batteries. The great flexibility in choosing an
start the two rest periods). Four hours total have been
appropriate algorithm also introduces an equivalent
allocated for rest periods; this could be put into one or
amount of uncertainty. It is recommended that for a
two rests; but it makes more sense to use two. More rest
given VRLA product significant R&D work be put into
periods and longer total rest times may also be suitable
the definition of suitable formation conditions. While
for some VRLA thick-plate products.
each company has its own approaches, the following
In order to have a relatively fine pore structure in the is a recommended procedure that should work for the
positive active material, a short, high-inrush current majority of companies.
period has been used to provide smaller, more
numerous PbO2 seed crystals upon which to build
■ Weigh the batteries prior to formation, but at the

Figure 9
end of the fill-to-form period. Don’t assume that
Typical Constant-Current Formation Profiles With Rests the electrolyte fill weight can be added to the pre-
Or A Discharge For A 12V/20Ah VRLA Battery. fill battery weight in order to get the pre-formation
weight; all batteries, especially large ones that are
A. Current and Voltage Curves for a CC/Rest
processed open to the atmosphere (i.e., without
Formation Algorithm
8
the top and vent valve in place), will lose weight
between filling and formation due to evaporation
and, in some cases, acid spraying or

Formation Voltage, Volts
regurgitation. The primary loss is from

evaporation, which can be several percent of the
total fill weight.
■ To as great a degree as possible, batteries should
2
be configured as they will be in a formation bay in
manufacturing. Formation of single units or a few
0 2 4 8 12 16 20 24 28 32 36 in series when they will be in large series-parallel
Formation Time, Hours
arrays in production will not give an accurate idea
of the effectiveness of the formation process. In
B. Same Curves for a CC/Discharge/Recharge addition, thermal conditions should be close to
Formation Algorithm those that will be seen by the batteries in
8 manufacturing. Initial studies can be done with
Formation Voltage, Volts
forming of small numbers of batteries, but it

Charge
2 should not be assumed that product quality would

0 be the same as in full-scale manufacturing.
Discharge
■ Wire up the test batteries so that the following

parameters can be monitored: voltage, time,
-8 current (also with integration if CV or TC charging
is used, but also for CC to ensure that the correct
0 2 4 8 12 16 20 24 28 32 36
Formation Time, Hours Ah input is applied) and internal pressure (if
batteries are formed sealed). Reference-electrode
measurements should also be taken and at some
point several batteries should be formed with gas
3.14 Development of a Suitable Formation collection and analysis being done.
Algorithm ■ In order to get an idea of the capabilities of the
It is assumed that a 6V/100Ah VRLA battery is being battery design for formation, an initial run should
developed for Telecom use and it is necessary to find be done using a very simple one-step CC, CV
out how to form the product most effectively. The and/or TC charge, just to see how the battery
optimized formation algorithm will depend largely upon reacts to these “baseline” conditions. Then,
the desired formation time, the design of the battery and several preferred algorithms should be applied,
the user requirements. Without going into such details, covering a range of times and currents, using rest
the following steps can be used to define a suitable periods and, possibly, discharges (even if this is
algorithm. not to be done in manufacturing due to cost).
■ Take at least 12 filled modules and monitor
temperature between fill and formation; note the
battery temperatures at the initiation of formation.
■ After formation, batteries should be weighed and 4. Temperature Limits For VRLA
carefully inspected for cosmetic and product Jar Formation
defects (acid spray or leakage at lid/box seals or
For jar formation of conventional flooded batteries, a
terminal posts, label damage, etc.). Teardowns
maximum formation temperature of up to 65°C may
should be carried out to look at the plates in detail
be permitted with no apparent harmful effect on the
(visual inspection for white sulfate, color, hardness
battery performance: this is certainly the case for SLI
(PAM) or softness (NAM), distortion, massive
battery designs. Industrial battery designs may have
corrosion or growth). The separator should be
significantly longer formation times and lower
inspected for holes/tears, damage from filling,
recommended maximum formation temperatures
staining by expander or paste and the possible
(e.g. 50°C).
presence of lead sulfate (hydration shorts). The
latter can be determined using a sodium iodide The temperature during all stages of the filling and
solution sprayed on the separator; insoluble lead formation process is much more critical for VRLA jar
dioxide shows up as a bright yellow precipitate. formation. The control of temperature is necessary
Electrolyte should also be squeezed out of the from the initiation of formation until its completion.
separator at several points to determine specific Sometimes it involves active control and at other times
gravity levels. it dictates passive processing conditions. The latter is
■ Negative plates should be dried and prepared for true going into formation, where the battery has been
SEM, BET and porosimetry analysis; other tests filled with electrolyte and allowed to stand for some
may also be carried out. Positive active material time before being placed in the formation environment.
should be treated similarly. In addition, several With VRLA batteries, high formation temperatures may
positive plates should be stripped of active result in the formation of lead dendrites and/ or
material and the grids should be inspected and hydration shorts. Therefore the maximum formation
weighed for general or localized corrosion during temperature should be kept below 40°C: and normally
formation. Wet-chemical analysis of the NAM (free this will require water cooling or forced air-cooling. The
lead, sulfate) and PAM (PbO2, sulfate, unformed formation regime may also include brief rest periods.
PbO) should also be done. XRD should be Some VRLA battery manufacturers may specify a
applied, if available, to define the amounts of maximum temperature of 50°C or even 60°C, but there
alpha- and beta-PbO2 generated at different are risks associated with this approach. In comparing
locations on the positive plate surfaces. formation at 60°C with formation at 40°C, it has been
■ Taking all of the data above, several iterations of found that the PbO2 content is higher at 60°C, and the
formation algorithms should be applied to ensure a/b PbO2 ratio is lower. However, the higher
that the most effective algorithm has been temperature has an adverse effect on the negative
developed. plates, resulting in a decrease in battery capacity at
■ As a final step, a pre-production run should be high discharge rates. The surface area of the negative
carried out under actual manufacturing conditions plates is decreased if formation is carried out at high
to ensure that the development work done on a temperature, possibly because of deterioration of the
limited number of batteries (particularly the negative plate expander [4]. It is important to note that
thermal conditions and the series-parallel if the measured temperature at the top of the cell is
configurations) is relevant to full-scale production. 60°C, the maximum internal temperature inside the cell
■ In addition to the above analytical work, full may be significantly higher, 70°C or even as high as
electrochemical characterization of the formed 80°C. This has implications in respect of the stability of
batteries should be done to ensure that nominal the negative plate expander, and it has been found
quality levels and the desired uniformity have that the surface area of the negative plates is
been achieved using the selected formation significantly reduced. Localized overheating may also
algorithm. Self-discharge (shelf life) result in grid corrosion and/ or increased risk of lead
measurements should also be done to ensure that dendrite formation.
the degree of formation of the positive plate and
remaining unformed oxide amounts are
acceptable.
In practice, sufficient time must be allowed after filling A critical issue with plate formation is to ensure that the
and before the start of formation to allow the heat plates do not “bow” during formation. This is a particular
generated during the filling process to have passed its concern with positive plates if they have been over
peak. The thermal management during the filling pasted on one side. A high current density should also
process should not be too efficient or the exothermic be avoided. If the plates are bowed, the plate group
acid-oxide reaction may “shut down” if the battery is too pressure will be non-uniform when the plate group is
cold, and start up again – generating excessive heat – assembled which will cause other problems including a
when the formation process is started. The degree of reduction in battery life. Because VRLA batteries need to
cooling (or even heating) during formation needs to be be assembled with a controlled plate group pressure to
related to a number of factors including: ensure long battery life, bowed plates are a far more
■ Product size serious problem with VRLA batteries than with
■ Temperature at start of formation conventional flooded batteries.
■ Cooling technique Partly because of this issue concerning bowed plates,
■ Plant temperature the maximum temperature for plate formation should be
■ Sealed or open formation kept below 40°C. A higher maximum temperature is
unlikely to have an adverse effect on plate performance,
Additional information is given in section 7.3. but there is a greater risk that the plates will bow during
the formation process. A high current density should
5. VRLA battery manufacture also be avoided.
using PLATE FORMATION
In plate formation the pasted and cured positive and
negative plates are placed in bottomless slotted crates
with relatively wide spacing so that no separator is
needed. The positive and negative plates are placed
alternately so that the lugs of all the positive plates are
on one side of the crate, and all the negative plate lugs
are on the other side of the crate. “Tacked” or “tackless”
formation can be used. With “tacked” formation lead
bars are tacked onto the lugs joining all the positive
plates together, and all the negative plates together, to
form a 2v cell. In “tackless” or “burnless” formation, the
plate lugs make contact with lead bars wedged
between the walls of the crate and the plate lugs: a
special clamp ensures close contact between the plate
lugs and the lead bar. Or the plate lugs make contact
with lead bars at the bottom of the crate. The crates are
immersed in dilute sulfuric acid (e.g. 1.100 s.g.) and a
formation charge passed through the crates: a stepped
current may be used to maximize formation efficiency.
Because the sulfuric acid is present in excess, there is
rarely any problem with excessive formation
temperatures. After formation, the plates are washed
and dried to produce dry charged plates. (Oxygen
needs to be excluded during the drying of the negative
plates). This stage of the process is exactly the same for
VRLA batteries as it is for conventional flooded batteries.
The formation regime, total Ah input, and dry charge
process can be exactly the same as for conventional
flooded batteries.
Technical And Theoretical Background
6. Technical And Theoretical positive paste and/or 2-5% of graphitic carbon in the
Background negative paste provides some conductivity to aid with
current flow. Sodium sulfate in the electrolyte (typically
6.1 The formation process explained
1.5% by weight of electrolyte) is also useful in improving
The chemistry that takes place during formation has a conductivity. Some manufacturing processes also
lot to do with the performance and lifetime of the VRLA involve having a small amount of sodium sulfate in the
battery in service. It also has a large impact on how pastes themselves. All of these measures are more
batteries can be processed, particularly in the times important in thicker-plate VRLA designs (plate
required for proper formation. The chemistry taking thicknesses and plate spacings >~2.0 mm), as nominal
place during formation can be characterized on a resistance values are higher in such products. Even with
simple level as follows. [Theoretical treatments can be the above materials being present, there is a strong
found in textbooks written by Hans Bode (Lead-Acid initial resistance to current flow. Because of this,
Batteries) and Dietrich Berndt (Maintenance-Free formation algorithms often start with a short, low-current
Batteries, 2nd Edition) ]. step; immediate use of higher currents can result in
The purpose of the formation process is to convert the heavy gassing from water decomposition due to the
pasted plates to lead dioxide at the positive and lead at high voltage needed to overcome the high battery
the negative. After pasting, both positive and negative resistance.
plates have essentially the same composition, except When the electrical current is switched on, the
that the negative plates contain additional “non-leady” electrochemical reactions which take place convert the
additives (negative plate expanders and floc). The lead oxide and lead sulfate to lead dioxide (PbO2) at the
chemical composition of both the positive and negative anode (positive plate) and to lead (Pb) at the cathode
plates after pasting and curing is essentially lead (negative plate). In some products, formation is from the
monoxide (PbO), monobasic lead sulfate (PbO.PbSO4), grid strands out toward the plate surfaces and in others
and tribasic lead sulfate (3PbO.PbSO4) [TRB]. The it is the opposite. However, in both cases the lead
positive plate may also contain some tetrabasic lead sulfate-active material conversion also results in the
sulfate (4PbO.PbSO4) [TTB]. The relative proportions of production of sulfuric acid from the water present in the
TRB and TTB are important for formation because TRB plate pores and separator. This creates a strongly
pastes form much more easily than TTB pastes. TTB is conducting environment in the plate stack so that higher
formed during the plate curing process at temperatures currents can be applied at low voltages.
of ~ 70°C or above. TTB can also be unintentionally
created during the filling process if the internal battery 2PbSO4 + 2H2O Pb + PbO2 + 2H2SO4
temperature is at or above 70°C for an appreciable 2PbO Pb +PbO2
amount of time.
Sulfuric acid is regenerated during the electrochemical
When the unformed plate is immersed in lead sulfate, reactions, and because some sulfate was present in the
the acid reacts with the lead monoxide and basic lead basic lead sulphates of the cured plates, the final acid
sulphates as shown below: density at the end of formation is higher than that of the
PbO + H2SO4 PbSO4 + H2O filling acid.
PbO.PbSO4 + H2SO4 2PbSO4 + H2O For a significant period of time, the conversion of lead
sulfate to active materials proceeds with very high
3PbO.PbSO4 + 3H2SO4 4PbSO4 + 3H2O
efficiency, close to 100% in some cases where the rate
4PbO.PbSO4 + 4H2SO4 5PbSO4 + 4H2O of water diffusion into the plate pores is high enough to
A significant amount of heat is generated in these keep up with the current flow. This is a critical stage in
reactions, and the higher the initial acid density, the formation, as the basic pore structure of the plates is
greater the heat that is generated. Sulfuric acid is established as the chemistry proceeds. High-inrush
consumed in the reactions, so there is also a reduction currents (following a short, low-current step to promote
in acid density. The pastes are largely converted to current flow) are felt to be useful because they create
neutral lead sulfate on their surfaces. These conditions more seed crystals in the positive paste and a higher-
increase the resistance of the unformed battery to porosity lead dioxide structure is created; seed crystal
current flow. The use of 10% or more of red lead in the formation is also aided by the use of red lead as
described above (the red lead formula is Pb3O4 and 6.2 Formation processes and Ah input
each molecule contains one molecule of PbO2). (Additional information about lead acid battery formation
The conversion efficiency is greater at the negative plate of conventional “flooded” batteries is given in the
than at the positive plate. Negative plates form relatively Digatron/ Firing Circuits brochure “Lead Acid Battery
easily and it is almost always the positive electrode Formation Techniques” by Dr. Reiner Kiessling).
whose formation efficiency is poorer. Thus, at some point For conventional flooded batteries, the choice of
relatively early in formation the positive charge efficiency formation processes is as follows:
will decrease and gassing will begin. If individual plate 1. Plate formation. See Section 5 for details.
potentials are monitored during formation, it will be
2. “Two shot” jar formation (also known as box
observed that there are two clearly defined potentials for
formation). The battery containing dry unformed
the negative plate. At the higher potential (+0.1 to +0.2v
plates is assembled into the final container. The
with respect to a cadmium reference electrode), lead
battery is filled with dilute sulfuric acid (e.g. 1.120
oxide and lead sulfate are being reduced to lead. As the
s.g.) and subjected to a formation charge: this can
conversion process nears completion, there is a rapid
be a stepped charge to control the maximum
change of potential to about –0.2/-0.3v, at which
temperature (smooth out temperature peaks) and
potential the evolution of hydrogen gas by electrolysis of
maximize formation efficiency. After formation is
water becomes the predominant reaction. However, for
complete, the acid is tipped out, and replaced with
the positive plate, the difference between the potential at
a higher density acid. Because some of the low
which the lead dioxide is being formed and the potential
density acid is retained in the plates and
at which competing gassing reactions occur is less
separators, the density of the acid for refilling may
clearly defined. The plate potential is always very close
need to be as high as 1.350, to achieve a final
to that at which the electrolysis of water can occur.
density of 1.280. A short equalizing charge for 2
Therefore the formation process is inherently less
hours at a low current is desirable in order to mix the
efficient than at the negative plate. The positive plate
acid before the battery is finished and dispatched.
potential (with respect to a cadmium reference
Because the acid is more restricted in jar formation
electrode) at the end of formation is about 2.35-2.4v.
than in plate formation, additional care needs to be
Because the plate interiors are only partially wetted by taken to avoid excessive formation temperatures.
electrolyte during the filling and fill-to-form processes, 3. “One shot” jar formation. This is now generally
the complete conversion of pastes to active materials preferred over “two shot” jar formation. The battery
via lead sulfate formation depends upon continuous is filled with acid of density such that the density
penetration of the acid generated by electrolysis into the after formation is the correct density for battery
plate interiors. When heavy gassing in the late stages of dispatch without the need for further acid
formation physically expels acid from the plate pores adjustment. The required filling acid density will
this continued diffusion is limited. For this reason, rests depend on the battery design (e.g. interplate
or discharges are desirable in VRLA formations, spacing, active material to acid ratio etc.): for
particularly those carried out in an acid-starved example typically 1.230 to achieve a final acid
condition. density of 1.280. Extra care needs to be taken with
The overcharge processes result in the loss of oxygen one-shot formation compared with two-shot
and hydrogen gases in a 1:2 ratio equivalent to a given formation. The higher density filling acid results in a
amount of water. This concentrates the initial electrolyte high battery temperature during the first hour after
strength and reduces the liquid volume. A moderate filling, because of the vigor of the exothermic
water loss is unavoidable, but if this loss is substantial reaction between the acid and the active materials.
additions of water or electrolyte following formation must The formation regime needs to be designed
be made. carefully: possibly including one or more rest
periods so that the peak temperature does not
exceed 65°C. Also, a higher ampere hour input may
be required than for two-shot formation because the
efficiency of conversion of the positive active
material to lead dioxide is poorer the higher the
density of the formation acid.
Typical ampere-hour inputs in relation to wet paste ■ The valve-regulated battery is designed so that any
weight and dry cured paste weight are given in Table gases generated during charge are recombined
1 below. These figures should be used for guidance within the battery. Each cell contains a self-sealing
only as they will be influenced by a number of factors valve that vents gases to atmosphere if the pressure
in relation to the plate and battery design and within the cell rises above a preset limit. So the cell/
formation regime. battery is not hermetically sealed but is valve
regulated.
Table 1 ■
Ah/kg wet Ah/kg dry The separator (normally of microfine glass)
paste weight cured weight completely fills the space between the electrodes.
Theoretical 200 226 The sulfuric acid is contained within the pores of the
Plate formation (1.100 s.g.) 275 310 plates and the separators and there is no “free” acid.
2-shot jar formation (1.13 s.g.) 300 340 ■ The separator is not quite fully saturated with acid
1-shot jar formation (1.23 s.g.) 350 400
(e.g. 95% saturation), so that any oxygen gas
VRLA jar formation 350+ 400+
generated from the positive plate is able to pass
through unfilled pores in the separator and
For guidance, the Ah input in the above table for VRLA recombine with the active lead surface of the
jar formation is approximately equivalent to 4x the 5 hour negative plate.
rate capacity, dependent on the battery design and ■ As a result, the VRLA battery is unspillable,
active material density. It is possible that certain VRLA maintenance free throughout its design life, and can
battery designs may require higher Ah input than given be operated in any orientation.
in the table above, and extended formation times. ■ In the jar formation of VRLA batteries, the cell or
However, any increase in Ah input should be minimized
battery is not normally sealed until the formation
by experimentation to establish the optimum formation
process is complete and the separator saturation
regime (e.g. by including brief rest periods or even a
level is deemed to be correct. Therefore in “open”
brief discharge partway through formation). If choosing
formation (section 3.3) the recombination process is
jar formation, the battery manufacturer will normally
not an issue. However, because the separator
choose 1-shot rather than 2-shot formation. This is in
completely fills the space between the plates, the
spite of the fact that VRLA batteries tend to be specified
cell design and the properties of the separator have
with a higher final acid s.g. which therefore requires a
a critical influence on the acid filling and formation
higher filling acid s.g, resulting in greater heat
process. Since there is also less acid in a VRLA cell
generation during filling. However, on balance there is
than in a conventional flooded cell, thermal effects
no particular benefit in using a 2-shot jar formation
are also more important. (Sealed formation is dealt
process, because of the difficulty of controlling acid
with in section 7.1).
volumes and final acid density. (Section 3.13 has
■ Cell/ battery reproducibility & variability is also much
already given sample formation algorithms and profiles
more of an issue with VRLA batteries than
for VRLA jar formation).
conventional flooded batteries [5]. To make cells as
6.3 Key Differences Between Flooded and uniform as possible in the manufacturing process,
VRLA Batteries electrolyte amounts are accurately metered into the
The comments below relate to VRLA batteries cell/ battery elements in the filling operation.
containing special separators generically known as However, normal tolerances in upstream processes
“Recombinant Battery Separator Mats (RBSM)”. These & materials e.g. grid casting, pasting, separator
are typically glass separators also referred to as materials may result in plate groups with variable
“Absorptive Glass Mat” (AGM) or “Microfine Glass” amounts of void space within fixed case dimensions.
(MFG). However, other separator types may also be As a result, the filled and formed cells may have
used, for example containing a blend of glass and slightly different void volumes. In subsequent duty,
polymeric fibers. A discussion of gel VRLA batteries the cells may behave slightly differently during the
will follow later (section 11). charging process due to the small differences in the
available void volume, and cell-to-cell differences in
the compressed separator structure.
Additional information about Jar formation
7. Jar Formation – Additional AGM material and thus the headspace is easily flooded.
Information In order to handle the acid that is forced into the head
space during the gassing phase of formation, some
7.1 Battery Preparation for Formation –
manufacturers use devices fitted into the filling port to
Sealed Formation
take up the expelled electrolyte; when the formation
The first part of this brochure (section 3) dealt with current is reduced or terminated the acid can then flow
“open” formation, which is the more popular and widely back into the battery and be retained. This approach
used technique. It is also possible to have the battery also minimizes gassing water loss because when the
acid-starved and fully sealed during formation. This will overflow device has taken up acid the plate stack and
depend upon the battery design and size and the type separator have enough void space to allow for a
of thermal management contemplated. For efficiency of significant amount of oxygen recombination. Because
manufacture and to minimize handling it is preferable to an accurately measured amount of electrolyte can be
form a product in the starved state fully sealed. Thus, added at filling, the final saturation level can also be
following formation the battery can be checked for known precisely from the materials amounts and the
quality, be electrically tested and, if passed, be shipped formation weight loss. The external device to
to the customer with minimal handling. However, there accommodate free acid during formation returns the
are stringent requirements on the design and processing acid to the battery when gassing is completed; the re-
of a battery subjected to sealed formation. absorbed electrolyte amount then results in an
In order to be able to form a battery sealed, the accurately known saturation level. Following formation
processing must be such that little or no carbonation of the vent valve is put in place and the assembly of the
the plates has occurred (formation of lead carbonate by battery is completed.
reaction of the lead oxide in the paste with CO2 from the
7.1.1 Plate Curing and Carbonation
atmosphere); carbon dioxide liberated as the plates
form can cause expulsion of acid through the vent valve A detailed discussion of plate curing is outside the
in the form of acid spray. The battery must also be filled scope of this document. However, if sealed formation is
to a starved condition with no more than ~95% proposed, certain precautions need to be taken to
saturation. Forming sealed, weight losses are minimal minimize carbonation. This is also important if high
(~5% of the fill weight) so that there is only a small vacuum filling is used. The reaction of CO2 with the plate
increase in specific gravity relative to the nominal level. pastes is actually more likely to occur not during the
Because of the low gassing levels, heat dissipation from curing process itself but after drying, when the plates
gassing is low, so for large batteries more care must be are taken out of the curing ovens. To minimize
given to keeping the battery in the prescribed carbonation, plates should be cooled down in a dry
temperature range. If this is not done, acid spray may environment when removed from the ovens (~ 20°C and
again result due to elevated temperatures in the battery ~ 10% RH). Plates should also be used as soon as
during the gassing phase of formation. In practice, there possible after drying. The use of a desiccation system
are very few VRLA products that are formed sealed, due on the drying oven may also help.
to the above issues. While it is desirable and can be
7.2 Acid filling
done, the cost penalties must be weighed against the
ease of processing during and after formation. The glass mat separator has a critical role in electrolyte
filling. Any change in the physical properties of this
As noted above, it is relatively difficult to design and material can drastically change the quality of the filled
process VRLA products so that they can be formed and formed cell or battery. The separator structure,
sealed. Without adding electrolyte after formation, it is degree of compression and fiber composition have a
difficult to achieve a roughly 95% finished saturation significant influence on how well an unfilled element will
level without having acid leakage and spray during accept electrolyte. While high levels of compression are
formation. The starting saturation needs to be at 97- desirable for extended life, this may make the filling and
98%, so unless the battery has a large headspace there formation process more difficult. When the separator is
is likely to be physical loss of acid. This may also be the compressed, the pore size is reduced, and the space
case if a low surface area or a hybrid glass/organic fiber available for electrolyte between the plates is also
separator is used in the battery design. Such separators reduced. This will make the filling process more difficult.
do not “hold” electrolyte as well as a high-surface-area (Section 8 gives more detail concerning separator
optimization).
Figure 10 gets hotter, due to the exothermic reaction with the plate
pastes. In the extreme case this heat build up can
The Filling Process Within A Vacuum And generate steam, which will impede the further ingress of
Non-Vacuum Fill. acid/water, and if severe enough may also cause
Pre-Fill Stage Figur
buckling of the plate stack [6]. (Figure 12
Figure 12)
Vacuum Process Non-Vacuum Process
Another danger is the formation of hydration shorts/
dendrites. As the acid reacts with the plate pastes, the
sulfuric acid electrolyte becomes progressively more
dilute. Lead sulfate is relatively soluble in the hot
electrolyte with a pH close to that of water, and soluble
lead sulfate will diffuse into the separator. This will
hasten the formation of lead dendrites and/ or hydration
Plate pores, separator are evacuated Plate pores, separator are filled
of all air bubbles. with air. shorts. A short circuit may develop and be detected
during formation, or more subtly the battery will fail
Electrolyte Added prematurely in service due to the formation of lead
dendrites. Sodium sulfate is a useful additive to help to
prevent dendrite formation: however during the filling
and formation process the common ion effect may not
be strong enough to prevent dendrite formation if the
electrolyte turns to hot water during the latter stages of
the filling process.
Separator is open, all plate Ingress of acid is impeded

If the filling process is poor, individual cells may also
pores are accessible, rapid by air bubbles, slower rate have “dry areas” after filling in which little or no liquid is
reaction ensues. of acid with pastes. present. These dry areas will slowly become wetted
during and after formation, but massive grid corrosion
Figur
Figuree 10 is a schematic view of what happens in may result due to the high temperatures and alkaline
vacuum and non-vacuum filling processes (section conditions prior to and during formation.
3.1). A high-vacuum fill allows faster acid ingress,
therefore shorter filling time and higher productivity. Figure 11
However, because it removes air from most of the Conceptual View Of the Filling Process For
plate pores, it greatly increases the reactivity and thus A VLRA Cell.
the rate of heat generation [6]. The battery design and Fill Electrolyte
manufacturing process need to be able to cope with
this rapid surge of heat.
When electrolyte is added to the cell, the ideal situation
is that all areas are wetted as much as possible by the positive negative
same amount of acid so that there is perfectly uniform plate plate
“sponge” “sponge”
distribution of electrolyte throughout the plate stack
plate height, I
when the filling process is completed. This ideal

situation is difficult or impossible to achieve in practice,
as there is a dynamic competition between the Dry
separator and the plate surfaces for the electrolyte [6] area?
(as shown in figur
figuree 11
11). As the electrolyte penetrates into
the plate stack, it is held up by the separator (the
Separator
capillary forces tend to hold the electrolyte fairly
strongly), and at the same time the electrolyte is
depleted by the exothermic reaction of the sulfuric acid
interplate spacing, d
with the plate pastes. As the liquid front penetrates
deeper into the stack it becomes more dilute and also When I/d is high, proper filling is difficult or impossible
Figure 12
Action On The Leading Edge Of The Liquid In A VRLA Cell Filling Process.
Fill Electrolyte
PbSO dissolves in liberated CO2 from

hot water, releases carbonated pastes
soluble PB(II) into exerts back-pressure
separator Pb(II) hydration shorts, on electrolyte
dendrites expander is leached
Pb(II) out by hot water
Hot Water
pressure, heat Steam paste particles are
from steam may dislodged by steam,
buckle plates, stain separator
soften and bulge
plastic case
Separator
Positive Pasted Plate Negative Pasted Plate
Heat dissipation can be aided by the following options: ■ Use multiple fill ports and channels in the battery
■ A cell design with a high surface area to volume case to guide acid and remove the “y” factor.
ratio, which allows a longer time for acid ingress, ■ Use push/pull massage after electrolyte
and thus a longer time for heat dissipation. introduction.
■ Chilling of the unformed element and/or electrolyte ■ Use a lower surface area AGM: but this may have
prior to filling. Chilling the electrolyte will be more an adverse effect on performance.
effective than chilling the unfilled elements. ■ Reduce compression to create a higher mean pore
■ Chilling of the filled element, using water rather size: but depending on the battery application this
than air cooling due to its greater heat capacity. may have an adverse effect on battery life.
■ Use more separator and more electrolyte so that
Dendrite growth and grid corrosion result from poor acid
the heat-sink properties are improved.
ingress and distribution. The following factors
are important in promoting faster, more effective Unfortunately, design or materials changes that improve
acid ingress: battery performance and/or life also tend to make proper
■ Minimize heat build-up and steam formation. filling more difficult. This includes e.g. high surface area
■ Avoid significant “carbonation” of unfilled elements glass fibers, high levels of compression, and thin plate
by taking special care in the drying and cooling of designs. Thin plate designs make possible the creation of
the unformed battery plates. small, powerful batteries. But they also mean higher
■ Use a “fluffier” more open separator with a lower surface areas, smaller interplate spacing and generally
grammage for a given caliper. However, this may greater l/d ratios (section 10.1). These impact negatively
have inferior compression characteristics, with on the filling process, therefore extra care needs to be
implications for battery life. taken in filling and formation, to avoid tipping over the
■
“knife-edge” into battery problems.
Use a high or rough vacuum fill.
7.3 Control of Formation Temperature poor that the degree of formation and initial discharge
Thermal management of the battery must be tailored capacities are reduced. It is an unfortunate fact that
to its size, the type of filling process and the formation negative plates are formed best at temperatures of
conditions used. Obviously, there is no set formula for 40°C or less; positives form better at higher
all VRLA products, so handling of batteries prior to temperatures. When positives are formed below 40°C
and early in formation must be developed by the there is an increase in the alpha-PbO2 content, a
manufacturer. In general, it can be said that very small decrease in total PbO2 and a fine pore structure
VRLA batteries (25Ah or less) are much more tolerant susceptible to clogging by lead sulfate during
of temperature and cells more so than 6 or 12V discharge can result. Because of all of these factors, a
batteries; for these products minimal thermal good formation system for small VRLA products
management is required (but the risk of over-cooling is should include some form of forced, circulated air that
high). For larger products the thermal requirements can be either heated or cooled, depending upon the
become more stringent and in the 60-100Ah range, ambient plant conditions, the product size and shape
particularly in multi-cell batteries, great care must be and the amount of heat production at each stage of
taken that the batteries are adequately cooled, but are formation.
not cooled too much. Larger batteries (25-100Ah), particularly those with
The VRLA battery manufacturer could use the unfavorable envelopes for heat dissipation (a cubic
following checklist in order to determine the necessary structure is the worst case), almost always require
temperature control to apply: cooling during formation. This is necessary to avoid
■ The size of the product (i.e., single cell or 6/12V high damaging temperatures and also to have
battery) uniformity of temperature among the cells in a 6- or
■ The battery envelope (i.e., the surface-area-to- 12V battery. In the best case (small size, favorable
volume ratio for heat dissipation). envelope, long low-current formation) passive cooling
■ How has the battery been handled prior to the (radiation, convection) can be employed, but there is
start of formation? likely to be a large swing in product quality between
■
summer and winter conditions unless the plant has an
Battery temperature at the start of formation
outstanding air-conditioning system. It is more likely
■ How will the batteries be cooled or heated that active cooling will have to be employed and this
during formation (ambient, forced air, water, can range from forced ambient air to forced chilled air
circulated water)? to stationary water bath to chilled, circulated water
■ What is the plant temperature (air conditioning, bath. The use of water for cooling compared to air is
summer or winter)? significantly more efficient due to the higher heat
■ How much heat will be generated in the capacity (four times that of air) and thermal
formation process and what is the duration? conductivity (roughly 15 times greater) of water.
■ Is the battery formed sealed or open, starved or Overall, heat transfer away from batteries during
flooded (i.e., what are the contributions of the formation is roughly nine times more efficient for water
oxygen cycle to heat generation and of gassing over air. Clearly, water is preferable to air for a number
to heat dissipation)? of reasons but it usually involves a higher capital input
and more maintenance. In some cases it may not be
The smaller the battery and the higher the surface- necessary, but perhaps the best argument for its use
area-to-volume ratio the more easily heat is dissipated. is that with the efficient cooling of circulated water
In the extreme, heat dissipation may be so good that shorter formation times can be used for some
small VRLA products actually require heating during products, thus maximizing the capital input for
formation! This is particularly true in a plant formation equipment. Different products will each
environment with poor temperature control i.e., hot in require a specific type of temperature control.
the summer and cold in the winter. Small VRLA
batteries require a minimum temperature for efficient
formation and a maximum above which damage can
occur. At the low end, formation of the PAM can be so
Figure 13
2.5 Ah And 25Ah Spiral-Wound Single-Cell Internal Temperatures During Different Fill-To-Form Conditions.
140
A D cell (2.5 Ah), 20
minutes in 10ºC cool
120 and wash, balance in
23º air.
B BC cell (25 Ah), 30
100 minutes in 10ºC cool
and wash, balance in
Temperature, ºC
23º air.
80 C BC cell, air cooling
(23ºC) only.
D BC cell, electrolyte at
60 10ºC, 30 minutes in
10ºC cool and wash,
balance in 23ºC air.
40 C
B
A
20 D
0
20 40 60 80 100 120 140 160 180 200
Time After Fill, Minutes
Several examples of the relative effects of air and evolution begin and this also contributes to a higher
water cooling are shown in Figur es 13 and 14 for
Figures temperature level. If the battery is formed in the starved
several different battery sizes. These figures compare state there is a further contribution to heat generation by
cell and battery temperatures with different the oxygen-recombination reaction involving oxygen
combinations of chilled and room-temperature reduction at the negative electrode. Finally, if the battery
electrolyte and the use of air and water during the fill- is formed sealed heat dissipation due to gassing is
to-form period. As can be seen, there are significant minimized, whereas maximal heat dissipation occurs if
differences in heat dissipation using air and water and the battery is formed in a flooded state and/or is open to
this also applies during formation [7]. the atmosphere.
Thermal management during the formation process To some extent the formation algorithm can be used to
itself is also very important. As noted earlier, at the control the temperature during the formation process,
beginning of formation the battery resistance is quite and smooth out temperature peaks. For example, longer
high and if high inrush currents are employed ohmic formation times at lower currents not only minimize
heat generation can be substantial. The formation ohmic heating but they also provide longer times for
reaction itself is highly exothermic, as is the continuing heat dissipation. This is usually in conflict with
oxide/acid neutralization reaction that takes place as manufacturing pressures, which favor the shortest
acid generated by electrolysis penetrates deep into possible formation times. The use of rests or discharges
the plates as formation proceeds. The heat of reaction will provide time for heat dissipation, but they also
for the formation process is ~394 kJ/mole and the heat lengthen formation. However, there are other technical
of neutralization is ~161 kJ/mole – both of which are advantages to these steps as discussed in section 3.13.
substantial. Later in formation the overcharge
processes of water decomposition and hydrogen
Figure 14
6V/100 Ah Prismatic Battery Temperature Data (Middle Of Center Cell) During Fill-To-Form Time With Different Conditions.
110
A Room temperature
electrolyte.
100 Cooling water at 8ºC.
B -20ºC electrolyte.
Cooling water at
90 A 10.5ºC.
C -20ºC electrolyte.
Temperature, ºC
Cooling in air at 21ºC.

80
70
47ºC at
B 360 Min.
60
C
50
40
20 40 60 80 100 120 140 160 180 200
Time After Fill, Minutes
7.4 Completion of formation 7.5 Alternative Jar Formation Options

In theory, formation is complete when there is conversion The battery manufacturer may wish to consider some
to 100% lead at the negative and 100% lead dioxide at rather more unusual assembly/formation options to
the positive, but this is not possible in practice. The overcome some of the disadvantages of jar formation
required degree of conversion will also depend to some already mentioned.
extent on the battery application. Close to 100% One option might be to place the assembled plate
conversion may be possible for the negative plates, but group in a container which is larger than the final
is more likely to be in the range 90-95% conversion to container so that the plate group is under little or no
lead dioxide for the positive plates. A lower PbO2 compression during formation. After formation, surplus
percentage may be acceptable for SLI batteries because acid is drained off, and the plate group is placed in the
it is assumed that the battery will be used fairly quickly final container to give the required degree of
and even though it is not completely formed the battery compression/plate group pressure. This approach would
will be able to start the vehicle. In addition, the shallow eliminate the problems already mentioned, which can
discharge service will slowly help to complete formation; occur when formation is attempted with a high plate
because the battery is usually charged by the alternator group pressure. In particular it would eliminate the
at a relatively low voltage and will typically only be at 60- problem of changes in plate group pressure during the
70% state of charge in use. On the other hand, many formation process (arising from changes in the volume
industrial batteries are required to deliver nearly 100% of of the battery plates during formation).
rated capacity when they are put into service by the end
user; in addition, they may be held on the shelf for long
periods of time before they are commissioned (thus
requiring very low levels of unformed oxide which
equates to a low self-discharge rate). Because of these
considerations, “completeness of formation” must be
relative to the product and its intended use.
Battery and separator design guidance
8. Battery Design 8.1 Plate height/plate spacing ratio (L/d)

The battery design has a critical influence on the filling The ratio of plate height to plate spacing (L/d) can be
and formation of VRLA batteries. Some of the potential used as a rough measure of the difficulties
problems with VRLA battery filling and formation can be encountered in filling. For a L/d ratio equal to or less
minimized or eliminated by careful battery design. than 50, easy filling results. If the ratio is between 50
Unfortunately, some of the design strategies to improve and 100, care should be taken to avoid potential
filling and formation may have an adverse effect on problems. Filling becomes more difficult when the
battery performance and life, so some compromise may ratio is between 100 and 200, and is almost
be necessary. impossible at ratios above 200 [8]. It can be deduced
from this that the worst case is a tall battery with a
The battery design parameters which may influence close plate spacing: the best case is a short, narrow
VRLA cell/battery filling and formation include: battery with a wide plate spacing.
■ Battery height: tall batteries are harder to fill than
short ones. 8.2 Battery case draft
■ Battery width: short, wide batteries are more Battery case draft can result in a 10% compression
difficult to fill if filled from a single filling port. change from the top to the bottom of the plate. For
■
example, with a target compression of 25%, the
Plate thickness and interplate spacing
compression may actually vary from 20% to 30%
■ Plate height and plate height/interplate spacing
between the top and bottom of the plate. The effect of
ratio
this on the performance and life of the battery may be
■ Filling port position highly significant. This effect should not be ignored in
■ Battery case draft the formation process as well. The separator at the
■ Active material additives (expander/reinforcing bottom of the cell will be subject to a higher
fibers) compression resulting in a smaller pore structure
■ Gravity: liquid will only wick so high before being which will influence the speed at which the acid fills
defeated by gravity: not a factor in gravity filling the separator during the acid filling process: there will
(top to bottom). be a slower acid drip speed as the acid approaches
the bottom of the plate. Because smaller pores
■ Separator properties:
have a greater force to pull liquid, this may also
- Volume porosity and pore structure (mean pore
increase stratification. [9], [10]. There is a slower
size). Finer pores wick more slowly but to greater
final heights acid drip speed as the acid approaches the bottom
- Saturation of the plate.
- Compression: results in a finer pore structure with
high tortuosity, therefore slower wicking. For 8.3 Active material additives
example, 15% compression will double wicking Additives in the active material can also affect the
time to a given height. filling operation. The expander or reinforcing fibers in
- Caliper (thickness at defined pressure) the paste may interact and result in excessive
- Grammage (g/m2) gassing during acid addition. This will result in a
- Surface area/ fiber diameter. Finer fibers (higher longer fill time or even in an unacceptable product.
surface area) result in finer pores, hence slower Care needs to be taken when any new material is
wicking.
used since the VRLA battery should be considered as
- Wettability
a system and all the ingredients interact.
- Fiber structure (coarse/fine fibers; inclusion of
synthetic fibers). Organic fibers inhibit wicking
(not wetted by sulfuric acid), and also promote
faster drainage.
- Fringe area of separator (area of separator not
covered by plates)
It can be seen that the separator properties are critical,
and these are discussed in more detail in Section 9. The
other critical battery design parameters are discussed
as follows.
Figure 15 9. Separator Optimisation

The separator properties have a critical impact on acid
Solubility Of Lead Sulfate In Sulfuric Acid At 25ºC.
filling and jar formation [6]. Any change in the physical
properties of this material can drastically change the
Solubility of Lead Sulfate (mg/liter)
7
quality of the filled and formed cell or battery. The type
6
of separator used is dictated more by the intended
5
battery application, but its properties can also partially
4
determine the filling and formation conditions used.
3
During the filling process the acid wicking rate is
2
important. The acid wicking rate is primarily a function of
1
the mean pore size of the glass-mat separator; this, in
0
1 1.05 1.1 1.15 1.2 1.25 1.3 1.35 1.4
turn, is largely a function of the fiber mix (represented by
Density of Sulfuric Acid (mg/liter) the fiber specific surface area as measured by BET), the
density of the glass mat and the compression level in
the unfilled plate stack. In practice, wicking is only
8.4 Electrolyte additives directly applicable for “top-down” gravity and “bottom-
Most VRLA battery manufacturers use sodium sulfate up” filling methods where wicking is the primary mode of
as an additive to the electrolyte. It is added to the fluid transport. For soft- and hard-vacuum filling
electrolyte in powder form, at about 1% by weight. techniques separator properties also have a role, but the
Sodium sulfate acts by the common ion effect to vacuum level and filling speeds are additional control
prevent the harmful depletion of sulfate ion which is a elements, in addition to the electrolyte temperature and
danger in the discharge of acid-starved batteries. its resultant viscosity. A further variable is the use of 10-
The addition of sodium sulfate provides an 20% organic fibers mixed with glass, as in the H&V IIP-
“inventory” of sulfate ions that are available without 15 material (see also section 10). The organic fraction
increasing grid corrosion [11]. The solubility of lead confers greater tensile strength and it also facilitates
sulfate increases significantly as the concentration of filling due to the hydrophobic nature of the organic
the sulfuric acid electrolyte decreases, as shown in fibers. Since the organic fibers are not wetted by sulfuric
figure 15
figure 15. The solubility increases more than acid, the electrolyte is not “held” as strongly as by glass
fourfold as the sulfuric acid density decreases from fibers. This clearly facilitates filling, but it can result in
1.300 to 1.100 kg/l. Under certain conditions of flooding of the negative-plate pores with acid and
over-discharge, the amount of dissolved lead electrolyte regurgitation and spray may be significant
sulfate may be such that on recharging the reduced during formation.
lead will form metallic bridges between the plates. The actual separator compression in the plate group will
The addition of sodium sulfate will reduce this risk. influence the ease of acid filling and jar formation as well
Alternative electrolyte additives may be used, which as impacting on the performance of the battery. High
have a different mode of operation. This class of separator compression has been shown to be beneficial
additives is known as dendrite prevention additives in extending the life of VRLA batteries by inhibiting
(DPA) [13]. These operate by actively seeking out positive plate expansion, but unfortunately the process
and deactivating the lead dendrite growths. They are of filling the battery with acid becomes more difficult.
polar organic compounds that are believed to When the separator is compressed it reduces the pore
deactivate a growing lead growth by coating its tip size significantly and also reduces the space available
with a layer of oriented molecules. Once the lead for electrolyte between the plates. This adversely affects
growth is deactivated, these molecules are available the wicking properties of the electrolyte. However,
to move onto the next site. smaller pores and higher compressions may mitigate
variations in saturation and acid strength in the vertical
plane (stratification). It is also important to optimize the
ratio of plate and separator pore volumes to ensure
sufficient electrolyte.
The easiest filling is achieved by using a combination of valve “sticking” (where the valve won’t open and release
glass and organic fibers with a low specific surface area gas at the design opening pressure).
(~0.8-1.4 m2/g) in a low-density material (i.e., high
If a battery is designed for sealed formation, the
percent porosity of ~95% or more) that has a relatively
separator chosen should have a relatively high surface
low compression level (25-30 kPa dry or less) in the
area (~1.6m2/g or more), be all glass and should have at
assembled, unfilled plate stacks. This gives an open
least moderate compression in the finished plate stack.
structure that is not completely wetted by the electrolyte
This is to ensure that the separator reservoir holds its
and, thus, has the best chance for uniform fluid
electrolyte as tightly as possible without having
distribution. This type of separator would be best suited
unacceptably poor filling characteristics. Moreover, the
to gravity filling. However, this type of separator and cell
amount of separator per ampere-hour of capacity should
construction also is most susceptible to electrolyte
be relatively great (1.4-2.0g/Ah) and the saturation level
drainage and stratification, particularly in deep-cycling
after filling should be ~95% or slightly less; it is virtually
applications. On the other hand, the best type of
impossible to form a VRLA battery sealed in the fully
separator to minimize drainage and stratification is a
saturated or flooded state. These parameters will allow
high surface-area glass (~2.0-2.6m2/g), high density
use of a sealed formation even with fairly aggressive
(90-92% porosity) all-glass with high compression. This
algorithms (high currents, short times). However, care
will give excellent deep-cycling results but it is extremely
should be taken to minimize heavy gassing by using
difficult to fill, particularly in large batteries. One might
multiple rest periods when gassing potentials are
think that a high-vacuum fill would be best for this type
reached. These will also allow more electrolyte to “soak”
of separator but, in fact, this would only be true in
into the plates and more void space will be created in
relatively small VRLA batteries (~25Ah or less) due to
the separator, enhancing electrolyte retention.
the large amounts of heat generated in short times in
high-vacuum fills. If the battery configuration cannot The rest of this section will give further background
dissipate the large burst of heat generated by the filling information concerning the influence of the separator
process there can be permanent damage in the form of choice and separator properties on the filling and
plate buckling, separator staining by paste and/or formation process. Typical design parameters are given
expander, bulging of the case and destruction of below, and are then discussed in more detail:
terminal seals; internal cell temperatures in excess of Volume
110oC can be achieved for relatively long periods of porosity:
porosity: 92%
time. Gravity fills with this type of separator system will Saturation: 95%
take much longer times (possibly up to 30-40 minutes),
Compression:
Compr ession: 30%
but thermal issues will be minimal.
Acid utilization: 8.8 – 9.5 ml/ Ah
In the formation process itself the separator can have an Separator
influence on gas management and electrolyte caliper: Related to interplate spacing and
distribution. The importance of this influence will depend required degree of compression
in part on whether “open” (section 3.3.1) or “sealed” Separator
(section 7.1) formation is used. For open formations, the grammage: > 2g/Ah preferred
type of separator used is not critical, as provisions are Separator
available for gas escape and fluid management, sur face ar
surface ea:
area: > 2m2/g preferred to minimize
whether by using a completely open top or by having stratification, but filling will be more
hollow tubes attached to the cell fill ports. For sealed difficult.
formations, however, the separator type and amount, the Jar for mation
formation
saturation level, the formation algorithm and the design Ah input: 4+ times rated capacity. It is
necessary to be careful about too
headspace are all important. Forming sealed under much overcharge during formation:
pressure can obviously lead to acid regurgitation and this may damage the positive plates
spray, but it also puts pressure on all of the seal areas and increase the overall acid
and can damage the functioning of some vent valve density.
designs by allowing electrolyte to leak into the seal area
between the valve and the fill stem, possibly leading to
9.1 Volume Porosity 9.3 Separator Caliper

This is a very important figure as it will determine how There is a standard BCI method for measurement of
much acid the separator will hold and is therefore a separator caliper (thickness). Because of the “fluffy”
critical parameter in the battery design and the nature of the RBSM, the caliper is measured at a
determination of the battery capacity. For 100% glass controlled pressure of 10 kPa. Although this may be
separators, the figure for the volume porosity in the referred to as the “uncompressed” thickness, it is
nominally uncompressed state is typically in the range important to note that this is referenced to a controlled
92-95% (as measured at 10 kPa). Compression of the pressure of 10 kPa. Specifications of e.g. “20%
separator will reduce this by a few percent, so that in the compression” or “30% compression” are referenced to
compressed state it will typically be 90-92%. this thickness as measured at 10 kPa. The required
separator caliper needs to be calculated in relation to
There has been some experimentation with “hybrid”
the interplate spacing and the specified % compression.
separators containing a percentage of polymeric fibers,
For example, a battery with an interplate spacing of
or non-glass separators such as the Daramic AJS
0.11cm and a design separator compression of 30% will
separator. These separators may have a lower volume
require separator material with a caliper of 0.16cm.
porosity and will therefore hold less acid. This needs to
When compressed by 30%, this separator material will
be taken into account in the battery design.
have a caliper of 0.11cm.
9.2 Saturation Level Recent research has highlighted the importance of plate
The cell/ battery is not normally sealed until the group pressure rather than % compression as the key to
formation process is complete and the separator extending battery life (see comments below).
saturation level is deemed to be correct (normally 95%)
(except for sealed formation, section 7.1). If in doubt, the 9.4 Separator Compression
battery manufacturer should err on the side of over- ALABC research work has shown that high compression
rather than under- saturation of the separator. If the battery designs can extend battery life by maintaining a
separator is over-saturated, and the cell is then sealed, high pressure against the positive plates and eliminating
the recombination process will be less efficient initially, or minimizing the phenomenon known as “premature
some water (as hydrogen and oxygen gases) will be lost capacity loss”. In fact, it may be more relevant to refer to
from the system, and the efficiency of the recombination “plate group pressure” rather than % compression.
process will increase, preventing further water loss. Some recent separator designs are less compressible
However, if the separator is under-saturated in relation to but may be able to maintain a higher pressure against
the design value, the cell may contain insufficient acid to the positive plates than conventional glass separators
meet its design capacity. [13]. Unfortunately, a high plate group pressure/ high
compression design may also be more difficult to fill. A
An accurate calculation of the amount of acid absorbed
higher compression will generally result in lower fill rates.
by the plates and separator is needed when precision
When the separator is compressed it reduces the pore
acid filling is used [12]. The separator needs to be 90-
size significantly and also reduces the space available
95% saturated, which corresponds to the separator in
for electrolyte between the plates. This adversely affects
the area between the plates holding about 6g of acid for
the wicking properties of the electrolyte.
every 1g of separator. The fringe area of the separator
also needs to be considered. Larger fringe areas allow Another issue that may need to be considered is that
for additional acid and thermal capacity in the battery: of changes in plate group pressure during formation.
this will help in the thermal management of the cell. There are changes in the volume of both positive and
There may also be some contribution to the low rate negative active materials during the formation process
capacity of the cell. The acid gravity is normally as the lead oxides are converted to lead at the
between 1.290 and 1.320 and for guidance a figure of negative plates and lead dioxide at the positive plates.
from 9 to 11 ml of acid per Ah as measured at the 20- This may have some effect on the separator
hour discharge rate should be used in the battery compression and applied plate group pressure. This
design calculation. Fully discharged, the acid gravity needs to be taken into account in the design of the
can be around 1.08. battery and the specification of the separator and the
initial compression level.
However, recent ALABC research has shown that Figure 16

under some circumstances the separator
compression may drop significantly after formation Mean Pore Size Vs. Kr BET Surface Area.
[14]. The exact causes of this are not yet certain, but
20
may be related to a relatively low initial plate group
18
pressure. There may be a critical compression that
Mean Pore Size (microns)

16
holds the fibers in place (for a particular separator)
14
and if there is significant gassing at the end of the
charging process there may be a loss of integrity in 12
the fiber mat. 10

8
While this problem is not yet fully understood, the
6
following design issues need to be considered to
4
minimize the risk of loss of plate group pressure
2
during jar formation:
■ Assemble cells with the maximum practicable 0.0 0.5 1.0 1.5 2.0 2.5 3.0
plate group pressure (> 40 kPa) BET Surface Area, m2/g

■ Maximize available acid volume and increase
separator grammage to >= 2g/Ah. The separator surface area for a glass separator is
■ Increase the fine fiber content of the separator. related to the ratio of coarse/ fine fibers. A lower surface
■ Use a formation algorithm that minimizes the area (higher proportion of coarse fibers) separator has
gassing at the end of charge. advantages in the filling process, but may have other
disadvantages depending on the battery application. A
9.5 Separator Grammage higher surface area correlates to a smaller pore structure
The separator grammage is the weight of the and results in a lower wicking rate, but a greater ultimate
separator per unit area. The amount of acid held by wicking height [17]. (Figur e 17)
(Figure 17). The smaller pore
the separator is very important, therefore separator structure will also help to decrease stratification within
grammage as well as thickness needs to be the cell. The pore structure of the separator provides for
considered carefully in the battery design. The aim a highly tortuous path which helps to prevent dendrite
should be to maximize available acid volume, which growth and minimizes the size of any dendrite if formed.
will improve the heat capacity of the battery and However, this also creates a tortuous path for acid and
enable thermal management to be improved. The air movement. This increases the filling time for each
amount (g/ Ah) of separator used will also have an cell. A battery designed for deep cycling should use a
impact on battery processing and performance. A high surface area separator, but extra care will need to
higher amount of separator (around 2g/ Ah or greater) be taken during the filling process. The higher surface
will have a beneficial impact on practical levels of area separator will require additional time to add the
compression, gas recombination and acid acid, since the acid wicks more slowly through finer
stratification. This also implies a greater plate spacing pores. Also, the way the acid is added to the battery is
and larger electrolyte reservoir. This will make it more critical. If the acid is added too rapidly from the top, the
practical to carry out jar formation on the completed air within the plates and the separator may not have
cell/ battery [15]. enough time to escape, and dry spots may result. If the
filling process allows the acid to wick up the separator,
9.6 Separator Surface Area entrapped air can escape since it does not have to
The surface area of the glass mat is very important diffuse through the electrolyte. It is also necessary to
because it has a great deal to do with wicking during allow sufficient time to allow for complete filling of the
fill and fluid movement in fill/formation. There is a pores of the separator. With a high surface area
reasonably well-defined relationship between surface separator, an advantage of the longer time for acid
area and pore size, as shown in figur e 16. This curve
figure ingress could be a longer time for heat dissipation. The
was constructed from data on various separator filling procedure is critical to providing a quality VRLA
samples from a wide range of manufacturers [16]. battery.
Figure 17 10. Separator designs to improve

wet formation
Impact Of Surface Area (m2/g) On Water Wicking Height
For many years H & V have marketed a hybrid glass/
While Under 20% Compression, After 24 Hours.
organic separator Hovosorb II, covered by US patent
8.00 4,908,828 [18], [19]. This separator contains a synthetic
Wet/Dry Weight of Separator
7.00 fiber with reinforcing glass strands, the balance being

6.00 microglass. The synthetic fibers are hydrophobic, and
5.00 these hydrophobic sites within the separator matrix offer
4.00 controlled wetting properties of the separator and
3.00 modification of the recombination process. It is claimed
2.00 that this assists the filling and formation process, and
1.00
allows a “fill and spill” formation technique to be used.
0 20 40 60 80 100 120 140 The unformed cells can be flooded with electrolyte prior
Height (cm) to formation, and drained of excess free electrolyte after
formation. It is claimed that the use of Hovosorb II
= 0.8
= 1.3
together with a “fill and spill” formation system may
= 2.2 result in a more uniform cell-to-cell electrolyte
= Hybrid @ 1.0
concentration than is obtained with volumetrically filled,
in-container formed cells.
With Hovosorb II, because of the hydrophobic sites
Figure 18 within the separator, recombination can occur even
when the separator is fully saturated. This is because
Effect Of Fiber Mix And Segregation some pores within the separator remain unfilled, even in
On Vertical Wicking Speed. the presence of excess free electrolyte, so that oxygen
1000
is provided with a path for transfer to the negative
Slower Wicking electrode surface. Oxygen recombination may even be
900
800 enhanced at those areas on the negative electrode
700
600
surface that are in direct contact with the non-wettable
Time / s
500 material contained in the separator. The separator also

400 has improved compression properties compared with
300
200
100% glass separators. Hovosorb II-P-15 is a refinement
100 of the original Hovosorb II, and has improved puncture
Faster Wicking
0 resistance.
0 25 50 75 100 125 150
(Height)2 / cm 2 An alternative approach is to use separators that consist
= 100% Fines of two or more layers of different fibers: this may be
= 20% Fines helpful in the filling process since layers of coarse fibers
Note: Strips of AGM not compressed. = 0% Fines
All AGM samples are single = AMER-GLASS are soaked more quickly. Battery filling is made easier
layered except AMER-GLASS, = AGM(B) with an “oriented” separator that has separate layers of
a multi-layered AGM. = AGM(A)
coarse and fine fibers: the fine fibers against the positive
plate, and the coarse fibers against the negative plate.
This has a very fast wicking characteristic both upward
and downward [20] [8]. In figur e 18 the influence of fiber
figure
mix and segregation on the vertical wicking speed is
shown. Figur e 19 shows the upward and downward
Figure
wicking height for oriented and non-oriented fibers.
During the filling process, the fine fiber component
absorbs acid quickly, but when the battery is filled from
the top, the looser, coarser fiber structure permits an
Figure 19
Upward And Downward Wicking Height For Oriented And Non-Oriented Fibers.
80 = Non Oriented AGM

UpWard Wicking Height (cm)
(mixed fibers)
70 = Oriented AGM
(segregated fibers)
60
50
40
30
Compressed @ 10 kPa; band 5cm
20
10
0
50 100 150 200 250 300 350 400 450 500
Wicking Time (Min.)
40
UpWard Wicking Height (cm)
35
30
25
20
15
10
5 Compressed @ 10 kPa; band 5cm
0
5 10 15 20 25 30 35 40
Wicking Time (Min.)
easier access to the electrolyte which then Another possible option is to include a thin microporous
permeates instantaneously to the fine fiber side. sheet as part of the separator system: this may help to
When the process is in reverse and acid is spilled out eliminate the problem of lead dendrite formation. This
of the battery, the forces binding the electrolyte to the might also help to control the diffusion of oxygen from
coarser fiber structure are weaker, so that electrolyte the positive to the negative plate. An example of such a
will be preferentially lost from this part of the AGM. microporous separator is the DuragardTM separator
The desired partial saturation of the separator is thus recently announced by ENTEK International at 7ELBC in
quickly reached. This multi-layered AGM such as that Dublin. Amer-Sil has also developed a composite
manufactured by Amer-Sil, has faster wicking separator which includes a microporous sheet between
properties which may be of great value in the “gray two layers of glass. Results with this separator system
zone” of filling where the ratio of plate height to plate have been reported in the ALABC Research
spacing is between 50 and 200. Programme.
VRLA gel batteries overview
11. VRLA Gel Batteries the gel as the gel cracks and shrinks. The free acid
can be removed from the top of the cell before sealing
Gel is an older technology than RBSM technology for
the cell and this will ensure that the cell is in a
valve regulated lead acid batteries: gel batteries have
recombinant state prior to going into service. The
been around for many years but have not been widely
electrolyte removed from the cell can be measured to
used except for special applications. The gel system
provide an accurate volume of the electrolyte present
has an inferior performance at high discharge rates,
in the cell. The total amount of electrolyte in the gel
therefore it is not suitable for applications such as SLI
cell may be slightly less (~ 80-90%) than that added to
requiring a high rate discharge capability. The gel
a comparable AGM cell.
technology may be more appropriate for tall cells
because the gel system does not suffer from The electrolyte is immobilized in a 3-dimensional
stratification problems. The gel is produced by adding structure set up by the very fine particles of silica [20].
finely divided (fumed) silica to sulfuric acid: the In an aqueous medium, these particles are fixed by
concentration of silica in the gel is about 6%. Fumed the presence in their midst of negatively charged
silica is a high purity silica manufactured by high sulfate ions. The gel formation happens at a molecular
temperature hydrolysis of chlorosilanes in a hydrogen/ level and allows the electrolyte to be present in the cell
oxygen flame. It is typified by a small particle size and in an altered state that is neither liquid nor solid. It
a very high surface area, in the range 200 – 400 m2/g. does not have the required mechanical strength
When the fumed silica is mixed with sulfuric acid, a properties to separate the plates, therefore a
viscous solution is formed, which develops into a mechanical separator is needed. The separator in the
thixotropic gel on standing. It has been found that the gel cell is purely a spacer, similar to the separator in a
inclusion of phosphoric acid is also beneficial, this conventional flooded cell. Gas transfer occurs via void
gives the battery a much longer cycle life, greater than spaces which develop within the gel structure as it
the addition of phosphoric acid to a flooded battery dries out slightly and opens up under the pressure of
design. The optimum H3PO4 concentration is the oxygen gas bubbles. The density of the gel will
17-30g/L. A typical process for preparation of the gel depend on the silica content of the gel. The higher the
is given below: gel density (higher silica content) the harder it will be
for crevice creation, and the greater the need for a
6% by weight of fumed silica is added to 1.280 S.G.
very open separator.
sulfuric acid with continuous stirring. 20g/ L of 85%
phosphoric acid is added and stirring continued. There may be different consistencies of the gel
Stirring is continued while pouring the gel into the dependent on the silica content of the gel: a “soft” gel
battery: the gel will set as soon as the stirring ceases. has a lower silica content than a “hard” gel. The “soft”
The battery should be filled under vacuum and stirring gel will have a relatively weak structure and only very
should not be stopped until the gel is in the battery. low shear forces are required to break it. The
consistency of the gel will influence how many cycles
Because the gel itself is unable to prevent the
of discharge/ charge are needed to achieve
penetration of lead dendrites that can cause short
recombination. Initially when the silica gel is formed
circuits between the plates, a conventional separator
there is total water saturation. The gel structure is
is required, which has ribs on both sides of the
completely filled with the electrolyte, which is also
separator. Stratification is negligible, because the gel
present in the active material and in the gel/ electrode
is more strongly fixed in the plates and separators. It is
surface interfaces. For the first few cycles, the gelled
not possible to carry out jar formation of gel batteries,
VRLA cell functions similarly to a flooded lead/ acid
the battery needs to be assembled with dry charged
cell and water loss occurs, particularly at the end of
plates before adding the gel. Sufficient time needs to
the charging periods. As water is lost in the initial
be allowed for the gel to “set up”, and the cells should
charging cycles, there is a slight dry-out of the gel
then be given an equalizing charge. During this
structure, which creates micro-channels in the gel
equalizing charge free electrolyte may percolate from
VRLA gel batteries overview / Formation equipment, battery monitoring and product testing
through which gas can pass. The “softer” the initial 12. Formation Equipment And
gel, the sooner the recombination process starts, Layout
because the process of gel cracking can start more
Apart from the choice of electronic equipment (rectifiers,
easily.
power resistors, power supplies, etc.), there are some
The formation of the gel is not influenced by the critical practical issues in choosing battery connections
type of separator used or its design. However, the and configuring formation bays. These can have as
separator does have an important influence on the much to do with resulting product quality and uniformity
filling process and the ease of oxygen transport as the choice of an effective formation algorithm.
during the recombination process. The separator for
gel batteries is a conventional separator as may be 12.1 Battery Connections
used in flooded lead acid batteries but with some This topic is often neglected, but in practice can have a
important design differences. There are deep ribs lot to do with product uniformity and scrap levels. There
on the negative side of the separator as well as on are several common methods of hooking units together
the positive side of the separator. The ribs are in a formation bay, depending primarily upon the
normally vertical and relatively widely spaced to hardware used, the degree of automation, the numbers
permit easy filling with the gelled electrolyte. The of units handled and the types of battery terminals.
interplate spacing is also very important: with a thin A few common connection methods are as follows:
plate battery design having a close plate spacing, ■ Wires and alligator clips (usually for larger
filling with the gelled electrolyte may be difficult if batteries)
not impossible.
■ Manual plug-in of flat tabs to forming strips
For gelled VRLA cells, the separator porosity needs (usually for small single cells/batteries)
to be as high as possible. This is because the ■ Loading of trays that may contain 20-50 small
separator needs to retain as much of the electrolyte batteries and insertion into formation bays using
as possible in its structure and to minimize the ‘snap-on’ contacts or pressure springs.
barrier to oxygen transport through the separator.
The pore size of the separator is also important in Whatever the connection method, a few common areas
the context of oxygen transport through the of concern apply:
separator. The optimum mean pore size needs to be ■ Are the battery terminals and formation contacts
in the range 1 - 10µ. Separators with a mean pore clean?
size < 1µ (µ=micron) severely inhibit oxygen ■ If a pressure contact is used, are any components
transport through the separator so that the only fatigued so that spring pressure is not adequate?
route for oxygen transport is around the edges or ■ Are the contact surface areas substantial so that
over the top of the cell group. The maximum rate of
localized resistive heating and/or oxidation
the internal oxygen cycle is lower in gel cells than in
doesn’t take place?
cells with AGM separators. The maximum rate is
■ Are the battery terminals and formation contacts
typically 10 A/100 Ah in AGM batteries, and 1.5 A/
made from low-resistance materials that will not
100 Ah in gel batteries [21].
produce significant voltage drops? Are these
materials easily corroded or oxidized?
These issues may seem trivial, but many formation

connectors have minimal contact areas with battery
terminals and the contacts and terminals are made from
materials that are easily corroded by battery acid (steel,
tin, copper) and/or are air oxidized at the high
temperatures generated at the contact points (copper,
copper-bronze). The use of more rugged materials such
as nickel plating will minimize the above problems, but
such materials may have significantly higher resistivities.
High scrap levels or high formation recharge category
can be created by the improper choice of connector
Formation equipment, battery monitoring and product testing
designs and/or materials and, more commonly, by Figure 20

infrequent inspection and cleaning of formation
equipment contacts. Battery Connections For Series Strings, Series-Parallel
Arrays And Series-Parallel Matrixing.
The cleaning of battery connections is also a key health
A. Simple Series-String Battery Connections
and safety issue. Hydrogen and oxygen gases are
generated during the formation process: a potentially
explosive mixture of gases is formed, which only needs
a spark from a corroded connector to cause an B. Series-Parallel Array Connections
explosion or even a fire. This has the potential to cause
injury or death to personnel in the area: quite apart from
the almost certain damage to batteries and/ or
equipment.
Additional information about gas monitoring is given in
section 13.3.
12.2 Formation Bay or Circuit C. Series-Parallel Matrix Connections

Configurations
There are three approaches to how batteries are
configured in formation circuits:
■ Series strings
■ Series-parallel arrays
■ Series-parallel matrices
By far the most common approach is to have a series

string of batteries charged by a single rectifier or power
supply. This is an effective approach and can result in resistances. Thus, current would flow through the
good finished battery uniformity if voltage drops due to remaining strings, but at significantly higher levels than
high contact resistances are not an issue. The major planned. If undetected by personnel or monitoring
drawback is that one open contact (due to one open equipment, this would result in over-formation, and
internal battery connection or to poor or broken possible destruction, of the remaining batteries.
formation bay mechanical contacts) can result in the
The most effective approach, but also the most
loss of a whole string if not detected.
complicated from a mechanical standpoint, is to use
Series-parallel arrays are useful in that more batteries series-parallel matrix connections where two or more
can be formed from a single rectifier or power supply by series strings are put in parallel, but there are also cross-
dividing the available current (with all strings having the connections at some interval so that a matrix is created.
same total voltage), but there is a danger of over- This approach, along with the two previous
formation or even thermal runaway if one string, due to constructions, is shown in Figur e 20. The major
Figure
even a slightly lower overall resistance, draws a advantages of matrixing are that current is distributed
disproportionate amount of the total formation current more uniformly and in the event of an “open” battery or
available at the expense of the other strings. If this is contact current continues to flow around the defective
even a subtle effect there may be a significant variation position. Batteries immediately adjacent to the defective
in product quality string-to-string. This can be avoided in position are affected to some extent (i.e., they
the setup of the formation circuits by using “steering experience somewhat higher formation current levels),
diodes” or other electronic measures to ensure that but those further removed are not noticeably influenced
roughly equal currents flow through each of the when cross-connections are made between all units.
paralleled strings. If an “open” condition occurs in one The important point is that current continues to flow
of the strings, however, it is lost as would be the case in through all of the strings in a roughly equal fashion.
a simple series-string arrangement; in addition, the total Complete strings are not lost as in series-string and
current available will now be distributed among the series-parallel designs and whole strings are not heavily
remaining strings according to their individual series overcharged in the event of an open battery. Matrixing is
usually done for single cells or small batteries where Cooling and heating equipment should also be
large numbers of units are formed in individual bays. maintained effectively. If forced-air heating or cooling
With these products, trays or carousels are used in is used, CFM and temperature measurements should
which connections are made automatically so that be taken frequently in each of the formation bays to
cross-connections can be done between each unit. ensure that airflows are correct. If water baths are
This approach gives the most uniform current used for cooling, water should be checked frequently
distribution and ensures that the effects of defective for temperature, flow rate (if applicable) and pH to
units are minimal, but there is an added cost. ensure that acid buildup is not severe.
12.3 Critical Maintenance of Formation All removable trays, racks, tables, etc. should be
Equipment washed and cleaned of acid frequently in order to
extend life and to minimize the occurrence of
To obtain optimal results from formation equipment
cosmetic rejects in batteries formed at later times
frequent maintenance is required. Some of this
(primarily terminal and label damage).
involves electrical equipment such as rectifiers and
DC power supplies, which must be calibrated on a 12.4 Power Quality and Equipment Costs
regular schedule (this is essential for ISO 9000 or QS
An important practical issue to consider is one of
9000 certification). The quality of the incoming AC line
power quality and equipment cost and how these will
power should also be analyzed periodically. Computer
impact upon a specific VRLA product that is being
control and monitoring equipment must similarly be
formed. Power quality varies greatly throughout the
maintained and calibrated, but perhaps the most
world and in some areas it is very poor. High levels of
critical aspect of formation-room equipment is the
AC ripple and harmonics can and do feed directly
actual hardware that accommodates the batteries and
into formation charging equipment; if filtering
which may be exposed to the high temperatures and
electronics is not included in the formation rectifier or
acid fumes and spray that go along with such close
power supply, large amounts of AC ripple will then be
contact.
superimposed upon the nominally DC charge profile.
One of the most important parts of maintaining a In formed batteries out in field service it has been
formation room is to carry out regularly scheduled shown that AC ripple can result in severely-shortened
abrasive or chemical cleaning of hardware used to lifetimes by generating heating that results in
connect batteries in formation bays. Oxidized or accelerated PAM softening and grid corrosion, as
corroded contacts can contribute large voltage drops shown in Figur
Figuree 21 [22]. Due to their low impedances,
to battery strings and this can cause undercharging or VRLA batteries are affected to a greater extent by AC
no charging at all in extreme cases. This can be ripple than comparable flooded lead-acid products
detected if current/voltage monitoring is done on all because low ripple currents translate into higher
battery strings, but if this is not done some batteries ripple voltages with the low impedances attributable
will be taken off formation with low voltages or in a to VRLA products.
“dead” state (i.e., zero voltages) and it may be
AC ripple is a form of “mini-cycling” that can wear out
attributed to the batteries and not the connectors.
a battery prematurely in service, but it can also have
Because of this, connector hardware must be cleaned
negative effects in formation, depending upon its
regularly, often enough that oxide and corrosion
amplitude, symmetry and frequency [23]. In addition
buildup cannot accumulate to levels that affect
to temperature effects and enhanced grid corrosion,
contact resistances. If an automated monitoring
ripple can reduce oxygen over voltage values thereby
system is not in place, poor contacts can be detected
increasing gassing during formation. Ripple is also
manually by taking voltage and current readings on
individual batteries or strings.
Figure 21 13. Battery Monitoring During

Formation
AC Ripple Voltage And Current Representation
Formation system monitoring is essential for both quality
(Upper Figure) And Its Effect On Cell Temperature
And Cycle Lifetime. and safety reasons, as well as to keep scrap levels
minimal. Formation scrap is the most expensive
because units that are scrapped here have the
Volts
7.0 maximum input of materials and labor of any stage of
VCM manufacturing other than finishing. Therefore, it is
6.5 imperative that some form of monitoring equipment be
Test Battery is 8 Ah 12V SLA Type
used, even if this is only a formation-room worker with a
multimeter.
Amps
12
ICM 13.1 Electrical Monitoring
6
The outputs of the charging system must be monitored
0
for compliance with values for voltage and current as set
-6
out in the manufacturing documentation. In addition,
-12 measurements may be taken to ensure good power
0 10 20 30 40 50 60 70
quality in terms of low levels of AC ripple and harmonics
Time (mSEC)
reaching the batteries being formed. Continuous
Correlation of Battery Voltage and Charger Ripple Current recording of these data by Quality or Manufacturing
Ripple current (effective val.) personnel, with SPC charting posted in the formation
Item
Net Current
0.1C 0.2C 0.5C 0.8C 1.0C
area, is highly recommended.
Value
Temperature For the batteries themselves, the following parameters
Rise (ºC) 0.1 0.2 0.5 1.2 4.0 6.0
should be monitored and analyzed (monitoring alone is
Proportion not enough: the data must be interpreted and
of life (%) 100 97 93 77 61 50
appropriate action taken if necessary):
■ Initial currents (CV or TC charge) or voltages (CC
charge)
not always symmetrically-imposed on the DC signal ■ Top-of-charge (TOC) voltages (peak voltage
and so there can be a net increase or decrease in the toward the end of a CC charge step, usually the
total charge applied (i.e., greater or lesser Ah last one at the end of formation)
formation inputs, respectively) depending upon ■ Finishing currents (CV or TC charge)
whether it is skewed to the charge or discharge side. ■ Integrated ampere-hour input
Given all of the above, it is highly recommended that
■ Presence of any timing kickouts and whether or
formation power supplies be provided with the proper
not they were properly re-initiated
filtering equipment (capacitors and chokes) so that
■ Overall voltage-time or current-time formation
minimal AC ripple is fed to the forming VRLA
batteries. It should be noted that the charging profiles (sampling basis only)
equipment available from Digatron/ Firing Circuits ■ Discharge capacities if the formation algorithm
provides inductive filtering to minimize ripple current. includes a discharge
Initial currents or voltages are important because they For CV and TC charging, integration of ampere-hour
indicate whether or not the initial current flow is at the inputs is necessary to ensure that the proper amount of
prescribed level and that most or all of it is going into the formation current is being applied; for CC charging,
formation process (conversion of lead sulfate to the monitoring of the applied current level and the step
active materials) and not into gassing because of high times is sufficient to determine accurately the Ah inputs.
resistances. Low currents on CV or TC charge or
A serious practical problem in many formation systems
abnormally high voltages on CC charge indicate that
is proper re-initiation of the formation profile following an
some or all of the unformed units are heavily sulfated
unscheduled power interruption. Ideally, the electrical
and/or there may be connector/terminal problems (high
system should pick up exactly where it left off but even
resistances). Even though current is flowing the
when this is done the forming batteries will have
distribution between strings may not be uniform and if
experienced an unscheduled “rest” of some duration (or
timed formation algorithms are used without current
even two or more). If interruptions occur early in
integration this will result in low Ah inputs to the
formation the effect is probably minimal but during
batteries. Electrical energy that goes into resistive
overcharge the liquid diffusion and gassing processes
heating or gassing will, obviously, not count toward
will be significantly affected. If it occurs within the first
forming the active materials but it will be included in the
hour or so, the formation schedule can simply be re-
total formation Ah input.
initiated. If it occurs in the last hour or so, early
Careful voltage monitoring during CC formation termination will probably not affect product quality
algorithms is useful as a feedback tool for triggering rest severely. The bigger problems are if and when they
or discharge steps and for defining product quality; as occur in the middle of the formation, if they are extensive
with the other parameters discussed, TOC “windows” and if there are multiple outages. For manually
are defined for manufactured products and these values controlled formation this can create confusion and
are used for accept/reject purposes. When voltages are errors. For computer-controlled systems proper
low (when they should be high), this indicates that the programming can account for outages, but in any event
batteries are being under formed; if voltages are in the the overall manufacturing schedule will be disrupted.
gassing region (above ~2.35 V/cell) for appreciable However, if there is no compensation for outages the
periods of time weight losses and, possibly, grid uniformity of product quality will be greatly affected, as
corrosion will be high, battery internal temperatures will each group of formed batteries may have widely
be elevated and dangerous over-formation of the different Ah inputs.
positive plate is possible. If TOC voltages are unusually
If one or more discharges are part of the formation
high (as they would be for a flooded lead-acid product),
algorithm their duration and Ah output can be used as
it indicates complete saturation of the plate stacks,
an indicator of the quality of the formed batteries;
which may not be desired for some products formed
beyond this, the values can be used in matching VRLA
sealed with ~95% saturation levels. (This depends on
batteries into larger arrays. In matching of modules to
whether the batteries are formed “open” or “sealed”, see
make higher-voltage (series) and ampere-hour (parallel)
sections 3.3.1 and 7.1). While it is often not possible to
batteries it is a reliable rule-of-thumb that matching of
monitor every battery, compliance of the overall voltage-
like-capacity modules is of paramount importance –
time or current-time profiles with manufacturing standard
more so than the nominal capacity. Thus, high- and low-
curves should be ensured. Non-compliance can
capacity modules must be matched with comparable
indicate problems with either the formation equipment or
partners rather than having them mixed. A battery
the filled batteries going into formation or both.
comprised of all low-capacity modules will give better
It is also important to monitor finishing currents during initial performance than one comprised of a mixture of
CV or TC charging, as they can indicate rectifier or high- and low-capacity units. Obviously, the best
power supply charging voltages being set too high or performance is obtained when all high-capacity
one or more strings in a series-parallel array being open. modules are matched.
High finishing currents can also indicate that batteries
are too hot or they have low saturation levels and, thus,
high levels of oxygen recombination. In the extreme,
high finishing currents can be a precursor to a part of
the formation bay going into thermal runaway.
13.2 Temperature Monitoring If monitoring of air or water is done as described above

Temperature monitoring of the formation room then monitoring of battery surface temperatures is not
environment, some of the formation equipment and the necessary, as the temperature differences will be quite
batteries themselves is important. VRLA batteries are small. However, in cases where no active cooling or
very sensitive to the formation temperature, so if a heating is applied (e.g., bench formation with no
formation room is being used without active heating or treatment) it is a good idea to monitor battery skin
cooling (forced air, water) it is imperative that temperatures (selected batteries on a sampling basis,
temperature be monitored at a number of independent particularly center batteries in large groupings) in order
positions throughout the room to ensure that a nominal to correlate this information with the formation algorithm
formation temperature exists and that it is relatively timing and to ensure that batteries do not overheat or
uniform within a range of, at most, several degrees form too cold.
Centigrade. If the formation-room temperature is too
13.3 Gas Monitoring
high the batteries will be over-formed or, in the extreme,
can go into thermal runaway. This will result in high In the later stages of formation, the forming batteries
gassing levels and weight losses, possibly high levels of give off various gases, particularly when they are formed
grid corrosion and poor positive-plate quality. If the open. Early in formation, the positive plate goes into
ambient temperature is too low under-formation can overcharge, generating significant amounts of oxygen
result, with low initial discharge capacities (particularly gas. Vented oxygen poses no particular problems in
the positive plate) and high levels of unformed oxides formation rooms. Later, when negative plates go into
(resulting in poor shelf life). In some VRLA products, if overcharge hydrogen gas is given off. This is also not a
the initial formation is done at too-low temperatures it is problem unless it reaches a level of at least 4% by
difficult or impossible to effectively recharge the partial pressure; at this point and above it forms an
batteries following formation; the reason for this is not explosive mixture with air in the presence of a spark
known. Both the nominal temperature for the formation source (which are usually abundantly present in
room and consistency throughout the year are formation environments). Other gases that are routinely
important, particularly if the manufacturing plant is not generated during formation are carbon monoxide and
heated and air-conditioned. The worst condition is to carbon dioxide. In cases where battery internal
simply form batteries on tables in a plant without temperatures reach levels of 160-190oC (this normally
temperature regulation of any type; this will result in only occurs in a thermal-runaway situation), hydrogen
variable product quality throughout the year and it is sulfide can be generated on negative plates during
likely that batteries produced during either the summer heavy overcharge. Hydrogen sulfide is extremely toxic to
or winter (or both) will be of inferior quality. humans, even at very low concentrations. Moreover, it
attacks any copper or copper-coated electrical
It should be noted that critical formation equipment such components and forms an insulating surface coating of
as rectifiers, power supplies, power resistors, monitoring copper sulfide, rendering the devices useless.
equipment and control computers must be located in a
completely separate room from the formation room itself. While it is unusual in practice to have significant
The high temperatures and acidic atmosphere in the amounts of hydrogen sulfide generated, it is
formation room can damage delicate electronic recommended that several types of gas monitors be
equipment and shorten their lifetimes considerably. installed in manufacturing formation rooms. At the very
least, hydrogen monitoring should be employed for
As noted earlier (section 7.3) as a general rule small safety purposes. Monitors should be placed at locations
VRLA batteries (~25Ah or less) may form better if heated in the formation room where hydrogen gas can be
and larger batteries require cooling for optimal generated in large amounts or where it may accumulate.
formation. In either case, appropriate regulation of the Clearly, this also calls for adequate air movement and
temperature immediately around the batteries (as ventilation to ensure that hydrogen gas buildup does not
opposed to the formation room ambient temperature) will occur in pockets that may explode. Carbon dioxide and
result in superior and consistent product quality. Thus, carbon monoxide are not likely to be generated in large
temperature monitoring of forced air or still or circulated quantities but CO monitors are common and
water used to heat or cool batteries, respectively, is also inexpensive and it takes little investment to install
recommended. several of these in a formation room.
Perhaps the most serious concern from a safety and recharged). Multicell batteries which clearly have
equipment standpoint is the potential buildup of one or more dead cells (e.g., a 12V battery with an
hydrogen sulfide, which is generated when VRLA OCV of ~10V or 8V) should also be identified and
batteries are seriously overheated during overcharge. scrapped. OCV monitoring directly after formation
Incidents are documented where telecom and UPS may not be feasible for single cells or small
installations have been destroyed by VRLA batteries that batteries, but for larger monoblocks it is useful in
generated large quantities of hydrogen sulfide on float removing scrap batteries and thus minimizing
when the systems went into thermal runaway [24]. future wasteful handling and storage.
Human toxicity is likewise severe, so it is highly ■ For batteries formed open, completion of
recommended that hydrogen sulfide monitoring be a assembly is done at this stage. This may involve
part of any VRLA manufacturing formation room. simply mounting one or more vent valves and
In most manufacturing operations, batteries in formation caps and/or it may require heat sealing, gluing or
will occasionally explode, or “pop”, due to hydrogen ultrasonic welding of an outer cover. In any case,
being present internally during overcharge and a spark areas of the battery involving these operations
being created from a poor COS or squeeze weld must be dry and completely free of acid.
contact. Such a defective connection will conduct ■ Batteries are then put into a stable environment,
current but at some point an arcing condition may occur. possibly temperature-controlled, for several days
If hydrogen gas within the battery ignites there may be until their electrical characteristics have stabilized
enough force to rupture the plastic case, creating plastic and they can be sorted for future disposition.
“shrapnel” near the battery. Clearly, this is a hazard for
14.1 Visual Standards
employees and by itself makes the wearing of safety
glasses or, better, face shields mandatory. The Many VRLA companies have a formal set of visual
frequency of this “popping” can be used as a quality standards for manufacturing personnel to use in
indicator for the upstream cast-on-strap and/or through- evaluating batteries for shipment following formation.
the-wall squeeze welding operations; if they occur Sample criteria that may be used for sorting are as
frequently the welding equipment and procedures follows:
should be carefully inspected. ■ Category 1. Batteries with only minor defects that
Category
can be cleaned up and are acceptable after such
14. Post-Formation Handling and treatment.
In-Line Product Testing ■ Categor
Category y 2. Batteries with cosmetic defects that
When formation is completed, batteries are not will not affect performance and are acceptable
necessarily ready for shipment immediately, although electrically can be used in closed-case
some manufacturers with very strong confidence in applications.
product quality (no doubt backed up by extensive ■ Category 3. Batteries with major cosmetic damage
Category
sampling of formed batteries over a long period of time) but acceptable electrical performance can be
do ship directly after formation. For most manufacturers, reworked, if feasible.
however, batteries from formation are subjected to the ■ Categor
Categoryy 4. Batteries with major cosmetic damage
following general processes: that may impact performance and/or lifetime must
■ Batteries are cleaned of any excess acid and be scrapped.
inspected for physical case damage and possible
corrosive attack on terminal posts or labels. Any Each product must be evaluated and a series of visual
batteries that clearly have leaks in seal areas (lid/ standards covering all of the problems seen in post-
box, terminal posts) must be scrapped. formation batteries must be developed to put batteries in
■
one of the above four categories. Examples of common
Batteries immediately after formation will have very
problems included in the above categories are: case/
high open-circuit voltages (OCVs) due to trapped
label acid damage, poor seals/acid leakage, bulged or
gas and excess surface charge on the plates.
cracked cases, corroded terminals, improperly applied
However, OCV readings directly after formation will
labels and deep scratches or dents.
identify dead batteries (to be scrapped) and those
with low, but significant, voltages (to be
14.2 In-Line Product Testing Figure 22

Following cleaning and inspection after formation and
storage for several days, batteries are then put through Typical Self-Discharge Curves For VRLA Batteries.
a series of tests in order to determine product quality
A. Self-Discharge Curves for One- and Two-year Shelf Lives
and segregate them into several categories for further
handling. The primary tool used is OCV measurement, 2.2
but in some processes AC impedance and high-rate 2.15
Open-Circuit Voltage, V
discharge performance are also measured. It must be 2.1
stressed that all of these parameters change with time 2.05
off formation, so in assessing product quality charts with
2
listings of daily parameter levels should be employed
1.95
(see below).
1.9
14.2.1 Open-Circuit V
Open-Circuit oltage Measur
Voltage ement
Measurement 1.85
OCV values tend to drop on a daily basis by a few 1.8
millivolts per cell depending upon the product type, the 0 5 10 15 20 25
Time on Stand, Months
finished electrolyte strength and the effectiveness of
formation. In addition, batteries with higher voltages
B. Self-Discharge Curves At Different Storage Temperatures
after formation show a slower decrease in OCV values
than those with lower (but acceptable) post-formation
100
voltages. All batteries have a relatively steep drop in 90
% Nominal Capacity Available
OCV over the first couple of weeks and this then 80

“flattens out” to a lower rate. In addition, the rate of 70
decrease in OCV is a function of temperature, so it is 60
imperative that storage and measurement be done at a 50
controlled temperature level, within a window of 3-4oC at 40 0ºC
most. Typical self-discharge curves for VRLA single cells 30 10ºC
are shown in Figur
Figuree 22; multiples apply for multi-cell 20 20ºC
batteries. 10 40ºC
0
OCV measurements are usually not simply a “pass/fail” 10 20 30 40 50 60 70 80
Months of Storage from Full Charge
affair, as batteries are often sorted into a number of
categories, among these being (in order of increasing
OCV):
■ Dead, low voltage (e.g., <2.000V) or reverse- short circuit, or low fill weight; in any event, it is
polarity (all are reject category) considered too dangerous and wasteful to attempt to
■ Recharge category recover these batteries and they are also scrapped.
■ Low acceptable or high acceptable (to be “Recharge category” batteries have low OCV values,
subjected to further test and, if acceptable, but they are considered to be possibly recoverable as
shipping to customers) acceptable product. They are incompletely formed for
■ High-voltage (reject category) some reason and often have relatively high internal
impedances due to sulphation. Because of this they are
Dead batteries have either an internal open connection often recharged in groups using high-voltage (3.5-4.0 V/
(COS, intercell weld) or they were not properly cell or more) constant-current charging for some period
connected to the formation system during charge. of time, after which they are subjected to the same in-
Reverse-polarity batteries were formed in reverse and, line test criteria as for the original formation. Some
thus, positives were formed as negatives and vice- manufacturers will use their formation equipment for
versa; both are scrap. Low-voltage batteries also may recharging, using a short, modified algorithm. In some
have had a poor connection in formation or they may manufacturing processes a second recharge treatment
have a near-fatal internal problem such as an incipient is allowed, but it should be recognized that with each
follow-up charge treatment it becomes much more likely 14.2.2 AC Impedance Measurements
Measurements
that these batteries contain some type of defect that In addition to OCV, the measurement of AC impedances
may show up early in field service. Recharging, of batteries can be a valuable sorting tool. Here, it is
particularly multiple times, has the distinct drawback usually the case that a “pass/fail” value is used for each
that, if effective, it can mask these subtle defects (such product, as batteries of different designs, voltages and
as small leaks, hydration shorts, etc) so that defective Ah capacities will have different nominal impedance
product is shipped. Manufacturers must be very careful values. In batteries, single-cell impedances are additive
in setting recharge procedures and standards for depending upon the number of series-connected cells.
acceptance, as too-generous guidelines can and will Larger batteries (in terms of Ah capacity) and those with
result in higher levels of customer dissatisfaction and thinner plates have relatively lower impedances, as this
returns. parameter is a function of plate surface area and plate
The vast majority of batteries should normally be in the spacing. Typical impedance readings for a number of
“acceptable” categories. If a company’s batteries go into different VRLA single-cell and 12V products are given in
general use there may be only a single voltage window Table 2. These are all thin-plate products (thicknesses of
for product to be shipped immediately. However, if some ~1.2mm or less) and so the AC impedance values will
of a company’s markets involve the application of high- be lower than for comparable (in terms of voltage and
voltage systems or have high-performance requirements Ah ratings) thicker-plate batteries.
there may be two (or more) “acceptable” categories, AC impedance is also a function of the state-of-health
one (A-group) being superior in terms of OCV to the for a given VRLA product type and is an indicator of the
other (B-group). quality of the materials and processes in use. The higher
In general, higher post-formation OCVs (A category as the impedance reading the poorer the battery quality
opposed to B) indicate a better-formed product and, (relative to the nominal value for that specific product).
thus, one with longer shelf life and superior electrical Thus, the nominal acceptable impedance reading is an
performance. However, some batteries can be and are allowable maximum value, with batteries above that
over-formed (high-voltage, or HV, rejects), resulting in value being sorted out for recharging (moderately high)
too-high weight losses and possible internal damage or as scrap (very high). It should be noted that it is
(enhanced grid corrosion, low saturation levels, over- possible for VRLA batteries to have acceptable OCV
formed PAM, too-high electrolyte specific gravity). values but unacceptably high impedance readings, as
Clearly, if the formation process involves adjustment of these two parameters have little in common electrically.
electrolyte volumes after formation this type of defect is There is no acceptable minimum value for impedance
not as likely to be detected. For batteries formed in a because it can never be “too low.” In practice, the lower
sealed state this is particularly pertinent. In addition to the impedance the better the battery quality. Thus,
possible internal damage, voltages may be so high for average impedance readings for production batches of
these batteries that they will tend to be undercharged in batteries are an indicator of the effectiveness of the
applications where they are mixed with “normal OCV” manufacturing process and should be used by Quality
batteries in high-voltage strings. This will be particularly and Manufacturing personnel as such. AC impedance
true in some float applications; in others, their high values are usually taken at 1 kHz with a Hewlett-Packard
voltages may cause undercharging of the “normal” Model 4328A meter or an equivalent instrument. As this
batteries and, in extreme cases, partial discharge. involves a four-point measurement using delicate probes
An example of a chart used in sorting batteries after it is necessary to use great care and reproducibility in
formation is given in Table 3; this same Table can be taking measurements on batteries, as simply varying the
used for sorting batteries after recharge or a short boost location of the probes on the battery terminals or applied
charge of any type. Obviously, any charging process will measuring probe pressure can result in significant
be the “zero point” for the chart, with “days off form” variances in measured values. It should be noted that
really reading as “days off charge”, whatever the type AC impedance is a “no-load” test, so it says nothing
and duration of charge. about the capability of a battery to sustain certain
discharge currents and/or the integrity of the internal
battery connections. For this, an additional test is
needed.
14.2.3 High-Rate Discharge Measurements

Measurements Figure 23
In addition to OCV and (possibly) AC impedance, some
manufacturers also carry out a short-duration high-rate High-Rate Discharge Voltage/Time Curves For
discharge test, usually at about the 10C discharge rate Acceptable And Unacceptable Battery Performance
at room temperature. As this is not a complete
On A 5-Second Test.
discharge, variations in ambient conditions are not as
critical as they would be for determining full capacities. A. Discharge/Rebound Curve Typical for a "Strong" Battery
The discharge time is for ~5-10 seconds, so a recharge Discharge
is not necessary. This is a useful test, as it not only Initiated
defines finished-product electrical quality but it also acts
as a check on the integrity of the internal battery
Test Voltage
mechanical connections. Discharge
Kickout
It is often set up as only a “pass/fail” test, using a
minimum voltage threshold at 5 or 10 seconds; any
batteries whose voltages are at or above this voltage are Note the strong voltage during discharge and
accepted; those slightly below it may be put into an the sharp voltage rebound after kickout.
“acceptable but inferior” category for use in certain non-
Discharge/Rebound time, Seconds
critical low-rate discharge applications; alternately, they
may be sent back for recharge. In practice, it can also
B. Discharge/Rebound Curve Typical for a "Weak" Battery
be used as a test to sort nominally acceptable batteries
Discharge
(i.e., those whose voltages are above the threshold Initiated
value) into “normal” and “superior” categories,
particularly for use in high-rate discharge applications
Test Voltage
such as engine start or UPS.

This type of testing is somewhat subjective, as it
depends upon the shapes of the discharge and voltage Discharge
Kickout
rebound curves, as shown in Figur
Figuree 23. With
Note the weak voltage during discharge and
programmable testers, collection of data and analysis, the sluggish voltage rebound after kickout.
even on such a detailed level, is not only possible but
also commonplace. Discharge/Rebound time, Seconds
Figur
Figuree 23a shows a typical discharge and rebound
curve for a strong battery, one with outstanding pore
structures in the plates and good diffusion kinetics. On While post-formation testing and evaluation are not
discharge, the coup de fouet (initial, instantaneous technically a part of the formation process, it is hoped
voltage drop) is followed by strong voltage recovery and that the foregoing will illustrate the necessity and power
an increasing voltage at the 5-second point, indicative of of this stage of manufacturing in validating the
good electrolyte diffusion into the plate pores to support effectiveness of formation or catching and correcting its
the high discharge current. Following termination of the shortcomings. In some cases, it can also provide
discharge the voltage recovery is sharp, again valuable information on processes earlier in the
suggesting effective diffusion kinetics. Figur e 23b is an
Figure manufacturing stream (plate quality, COS integrity, heat-
example of the discharge and voltage recovery curve for seal effectiveness, etc.) This is an area that is often
an inferior battery. During the discharge the voltage is overlooked or given short shrift, but it is one that will
not strong and may even begin to drop off toward the have a major impact on the quality and uniformity of
end of the 5-second duration. The recovery is gradual, product reaching the end user.
indicating restricted diffusion of acid from the separator
reservoir into the plate pores. Such a battery will not be
a strong performer in high-rate applications.
Troubleshooting formation problems
15. Jar Formation Toubleshooting: 4. Electrical continuity check shows an “open” rreading
eading
Problems And Solutions on a string or circuit
circuit
a. Carefully inspect connectors for corrosion or oxidation
15.1 Filling and Fill-to-Form b. Inspect connectors for broken wires
1. Batteries overheat following acid filling c. Inspect wires coming from rectifiers/power supplies
a. Chill unfilled batteries to –30oC (Caution: this process d. Measure individual batteries for “open” resistance
is energy intensive and the low temperature may readings (with and without connectors attached)
cause damage to seal areas, and the battery case
may be susceptible to cracking during handling) 15.3 Formation Process
b. Chill electrolyte to –10oC
c. Immerse filled batteries in chilled water bath 1. Batteries are too hot during for
are mation
formation
d. Slow down filling speed a. Increase airflow or water circulation rate, decrease
cooling water temperature
2. Separator above plates is damaged during b. Reduce ambient temperature conditions
acid addition c. Reduce current levels, extend formation time
a. Slow down fill speed d. Put in rest periods to dissipate heat
b. Use acid-diffuser nozzle design e. Provide more space between batteries
f. Lengthen fill-to-form time, reduce initial battery
3. Lead sulfate hydration shor ts for
shorts formm in the temperature
glass-mat separator
glass-mat
g. Shorten time for high initial current charge, if used
a. Use sodium sulfate in electrolyte
h. Change plate materials, use red lead in positive paste,
b. Use slower filling to yield more uniform acid carbon in negative paste
distribution
i. Use sodium sulfate in electrolyte
c. Keep temperature down during the fill-to-form time.
j. Shorten fill-to-form time to reduce sulfation level, lower
d. As a last resort, put batteries onto formation battery resistance
immediately after fill
k. Some parallel strings receive too much current
4. Conversion of PbSO4 to PbO2 is poor l. Use pulsed-current algorithm
a. Increase fill-to-form time
2. Batteries are too cold during for
are mation
formation
5. Batteries rregurgitate
egurgitate acid after filling a. Increase airflow rate, air temperature
a. Use cooling bath after filling b. Increase cooling water temperature, slow circulation rate
b. Check dried plates for excessive carbonation c. Increase ambient temperature
d. Increase charge current, shorten formation time
e. Some parallel strings receive too little current
15.2 Pre-Formation Conditions
1. Batteries are too hot
are 3. For mation amper
Formation e-hour input is too low
ampere-hour
a. Allow longer fill-to-form time a. Check for power outages during the formation period
b. Use chilled water cooling between fill and formation (did the formation programmer compensate?)
c. Use a low-current initial charge for 1-2 hours b. Check rectifier/power supply setting - too low?
d. Reduce cooling efficiency during fill, fill-to-form; c. Ambient temperature too low for CV or TC formation
extend filling and fill-to-form times algorithms
d. Cooling air or water too cold
2. Batteries are rregurgitating
are egurgitating acid e. Check formation programmer for the correct formation
a Form open instead of sealed algorithm
b. Use external “acid tower” fitted to the fill port to take f. Batteries heavily sulfated, current-time profile is low
up expelled acid (CV or TC)
c. Modify cure/dry conditions to minimize plate g. Batteries under filled, resistances are high, current-
carbonation time profile is low
d. Process batteries rapidly after cure and dry, minimize h. Connector hardware is corroded, oxidized
time on floor before fill and formation i. Battery terminals are corroded, oxidized
3. Loading of for mation bays takes too much time,
formation
fill-to-form time is high and variable
ill-to-form
a. Use multiple acid fillers to speed up loading of bays
b. Use smaller numbers of batteries on the formation
circuits
4. For mation amper

Formation e-hour input is too high
ampere-hour 9. Toxic, explosive gases ar
Toxic, e generated in large
are
a. Check controller for correct algorithm current/voltage amounts, alar ms sound
alarms
settings, timing a. Too much overcharge ampere-hours on the negative,
b. Too-high rectifier/power supply setting high hydrogen levels
c. Ambient temperature too high for CV, TC formation b. Ventilation equipment inoperative, malfunctioning or
algorithms under-designed
d. Cooling/heating air/water too warm c. Some formation bays have batteries in thermal
e. Interrupted formation reinitiates from the beginning runaway, hydrogen sulfide is being generated –
evacuate and shut down affected bays!
5. End-of-charge currents ar
currents e too high in CV or
are
TC for mation
formation 10. Batteries explode during for mation, may cause
formation,
plastic “shrapnel”
plastic
a. Batteries are too hot due to poor temperature regulation
a. Poor internal battery connections cause sparks during
b. Rectifier/power supply voltage too high overcharge, H2 ignites, and cases are distorted or
c. Low fill weights in one or more cells and/or high ruptured
formation weight losses results in high O2-
recombination level 11. Batteries explode during formation, catch fir
formation, e
fire
d. Impurities lower oxygen over potential on positive, a. Use flame-retardant case/lid materials
increase recombination current b. Equip formation room with sprinklers, smoke detectors
e. Non-uniform distribution of current in a series-parallel
string formation bay
15.4 Post-Formation Handling and Visual
6. End-of-charge voltages are too high in CC for
are mation
formation Inspection
a. Batteries are flooded, no oxygen recombination 1. Acid and/or water damage to batteries
taking place
a. Apply measures as in Section 16.3, Section 5 to avoid
b. Too much electrolyte in negative-plate pores, acts like future acid damage
flooded battery
b. Clean and dry battery thoroughly, examine terminals,
c. End-of-charge current higher than the programmed level labels for damage and scrap if severe
7. End-of-charge voltages (EOCV) are too low in CC
are 2. Acid damage to for mation har
formation dwar
hardware
dware
for mation
formation
a. Clean thoroughly before re-use
a. Low fill weights in some cells and/or high formation
weight losses create too much void space, therefore b. Replace steel, copper connectors if damage is severe
too much oxygen recombination, lowers EOCV
3. Oxidation damage to for mation har
formation dwar
hardware
dware
b. Too little electrolyte in negative-plate pores, too much
a. Clean thoroughly before re-use
recombination
b. Replace if damage is severe
c. High impurity levels reduce hydrogen over potential at
the negative and/or the oxygen over potential at the 4. Exposed steel parts ar
parts e rusted
are
positive
a. Clean with abrasive if minor
d. Batteries are too hot, O2-reduction current increases,
lowers EOCV 5. Ter
Ter minal plating is corr
erminal oded or bur
corroded ned away by acid
burned
e. End-of-charge CC current is lower than the or shor ting, rrespectively
shorting, espectively
programmed level a. Scrap battery if severe; if minor use in closed-case
applications
8. Batteries rregurgitating
egurgitating acid, acid spray occurs
a. Batteries too hot going into formation 6. Ter
Ter minal posts ar
erminal e bur
are nished or par
burnished tially melted due
partially
b. Batteries are too hot for the reasons given in Section 1 to dead shor
shortt
above a. Scrap battery
c. Heavy, deep carbonation continues to generate CO2
from plates 7. Minor scratches and/or dents to batter
batteryy case
d. Voltages and/or currents are too high, gassing is excessive a. Rework and use as normal
e. Attach acid tower to fill port to catch regurgitated acid b. Train formation room personnel in better handling
f. Use all-glass separator, higher surface area separator, techniques
higher stack compression to hold electrolyte more
8. Major dent or cracked/chipped case that may
tightly
indicate inter nal damage
internal
g. Battery headspace is inadequate
a. Scrap battery
h. Batteries are overfilled and/or void space is too low
(high paste/grid weights)
i. Equip formation room with sulfuric acid mist monitoring,
alarms
9. Acid leakage evident fr om lid/box seal, vent- valve

from 3. Recharge-categor
Recharge-categoryy batteries (voltage is >LV but
>LV
area or ter
area minal post seal ar
terminal ea.
area. below “acceptable”)
a. Lid/box seal area – scrap battery or rework for a. Slightly low acid fill weight
closed-case use b. Slightly inferior connector contact during formation
b. Terminal post seal area – see below c. Formed adjacent to dead battery in series-parallel
c. Vent-valve area – clean and dry matrix formation (less-than-normal Ah input)
d. One or more cells may have an incipient hard short or
10. Top and/or sides of batter
Top y ar
battery e bulged visually
are hydration shorts in the separator
or to touch
e. Missing or damaged/leaking vent valve
a. If minor, hold for evaluation; if major, scrap battery
f. Large air leak due to cracked case or seriously
11. Dir t, oil and/or gr
Dirt, ease contamination
grease defective seal
a. Clean thoroughly and use g. Incomplete formation (high PbO level in PAM), poor
acid penetration/no rests
12. Acid around ter
around minal post(s) (may be simple acid
terminal h. If all batteries in a bay are recharge category; check
contamination during for mation or a defective post seal)
formation rectifier/power supply setting (may be low)
a. Clean away acid and put battery on a 2- or 3-hour i. If all batteries in one string of a series-parallel array are
overcharge recharge category-low string current draw, other
b. If no leakage, accept as good product; if leakage string(s) may be HV
occurs during overcharge, scrap battery j. Ambient temperature low for CV or TC formation
algorithm (all batteries affected)
k. Formation kicked out prematurely, not reinitiated (all
15.5 Post-Formation Electrical Evaluation - batteries affected)
OCV, Impedance, High-Rate Discharge. l. Battery terminals corroded or oxidized
1. Batter
Batteryy is dead – zero voltage
zero
a. Internal “open” connection in battery 4. High-voltage (HV) batteries, >~2.2 V/cell
b. No fill acid a. String draws too-high current in series- parallel array;
all batteries in string are HV
c. No connector contact on formation
b. All batteries in a bay are HV; check for high rectifier/
d. Battery was short-circuited during handling (terminal power supply setting
damage will identify this)
c. Formation kicked out; reinitiates from the beginning
e. Battery is part of a whole string that received no (too-high Ah input, high weight losses for all batteries
formation current in a bay)
f. For a-d, batteries must be scrapped; for e batteries d. Ambient temperature high for CV or TC formation
may be put back through formation although they may algorithm, high Ah input (all batteries affected)
not form properly and it is safer to scrap them also.
5. High impedance rreadings
eadings
2. Low-voltage (LV) batter
(LV) batteryy (<2.000 V/cell)
a. Fill weights too low
a. One or more cells shorted out
b. Batteries over-formed, high weight losses
b. Low acid fill weight (probably HV category also)
c. Poor connector contact on formation c. Incomplete formation (dead, LV or recharge
d. Battery was partially short-circuited during handling category also)
(terminal damage) d. Poor plate processing, usually at cure/dry
e. Battery is part of a string that received very low e. Meter probes too high on terminals
formation current
f. Meter probes corroded or oxidized
f. Battery is poorly formed, went onto formation heavily
sulfated g. Impedance meter out of calibration
g. Early formation kickout malfunction not reinitiated (all h. Battery terminals corroded or oxidized
batteries affected)
h. The usual procedure is to scrap any LV battery
6. High-rate discharge (HRD) voltage rreading

eading too low 3. Excessive grid corrosion after for
corrosion mation
formation
a. Battery over-formed or incompletely formed, high a. Poor filling process created watery or dry areas,
impedance creates large voltage drop on HRD rampant alkaline corrosion occurred in these areas
(impedance reject) (the grid or areas of it may be completely corroded
b. Poor battery internal connections (affects HRD but not after formation)
impedance measurements- impedance is a “no-load” b. Dry areas in separator occur during filling due to tight
test) fiber structure, trapped air (very localized), again
c. HRD tester probes and/or battery terminals corroded alkaline corrosion occurs
or oxidized c. Low negative-plate potential pushes positive potential
d. Poor connection of tester probes to battery terminals into high-corrosion region in CV formation (occurs over
e. Discharge current set too high or HRD tester out of entire plates)
calibration d. Formation Ah input and/or temperature are too high
f. Poor electrolyte distribution between separator and e. Impurities present (chloride, organic compounds that
plates (too little in plates) break down to form acetic acid) catalyze extensive
g. Low active-material porosities, restricted electrolyte grid corrosion
diffusion
4. High levels of unformed oxide, par
unformed ticularly in the
particularly
h. Plate stack compression too low, results in poor positive plate
separator/plate contact
a. Low Ah input, batteries under formed
i. Grid-PAM passivation layer present
b. High percentage of formation charge goes into heat
production, gassing, grid corrosion
15.6. Battery Tear-Down and Analysis c. Heavy gassing doesn’t allow acid, water penetration
into plate interiors during formation (use rest periods)
1. Negative plate shows areas of white lead sulfate
areas
d. Dense lead sulfate layer on plate surfaces (too-fine
a. Insufficient formation Ah input oxide, heavy calendering during pasting, too long fill-
b. Fill-to-form time too long, too much lead sulfate formed to-form time) inhibits fluid penetration
(high resistance) e. High orthorhombic PbO level in the positive paste
c. Long fill-to-form time, no carbon black in expander creates an impervious structure for liquid penetration
d. High impurity level, lowers hydrogen over potential,
negative can’t be formed 5. High alpha:beta-PbO2 ratio in PAM
PAM
e. Too much oxygen recombination due to low fill weight a. Increases with >temperature
and/or high formation weight loss (too much void b. Decreases with >acid density
space in the separator) c. Decreases with >current density
f. p:n value too low (<~0.8), negative- plate sulfate can’t d. Decreases with increasing amount of positive-paste
all form out sulfate content
g. Solutions: use high-purity materials, increase carbon e. Generally, lack of acid penetration increases alpha-
black amount, use p:n ratio >/=1.0 PbO2 percentage
h. CV formation voltage and/or temperature too low f. High alpha-PbO2 levels form in areas of the plate stack
where the fill electrolyte is dilute; most prevalent in
2. Positive plate shows areas of white lead sulfate
areas thin-plate products, pastes with little or no sulfation
a. Insufficient formation Ah input (high density)
b. Fill-to-form time too long, as in 1b
c. High impurity level, lowers oxygen over potential,
positive material can’t be formed
d. High Ah input into grid corrosion, doesn’t go into lead
sulfate conversion
e. High negative-plate voltage (flooded?) doesn’t allow
sufficient positive-plate polarization to complete CV
formation
f. CV formation voltage and/or temperature too low
References
16. References Table 2.

[1] G. Zguris, Batteries International, April, 1991, pp. 80-81
Typical AC Impedance Values For A Variety Of Thin-Plate
VRLA Single Cells And Batteries Fully Charged At 25ºC.
[2] H. Chen, et al. Journal of Power Sources 59 (1996), pp.
59-62. 2V/2.5Ah Cell 5.0 milliohms
[3] C. Ressel, Batteries International, April, 1990, pp. 24-25 6V/2.5Ah Battery 15.0 milliohms
[4] Exide Technologies, First Semi-Annual Report, March 2V/5.0Ah Cell 3.5 milliohms
2001. 6V/5.0Ah Battery 10.0 milliohms
[5] R.F.Nelson, The Battery Man, May 1998. 2V/25Ah Cell 1.5 milliohms
[6] R.F.Nelson, Batteries International, 36, July 1998, 95-103. 12V/25Ah Battery
(Hawker SBS) 7.0 milliohms
[7] K. Matthes, B. Papp and R.F. Nelson, Power Sources 12,
12V/25Ah Battery
T. Keily and B. Baxter, eds., 1988, p.1. (Hawker Genesis) 5.0 milliohms
[8] A.Ferreira, J.Power Sources 78 (1999), 41-45 12V/35Ah Battery
[9] The Battery Man, January 1995. (Hawker SBS) 5.5 milliohms
12V/36Ah Battery
[10] J.Power Sources 67 (1997). (Hawker Genesis) 4.5 milliohms
[11] A.Ferreira, Batteries International, 46, Jan 2001, 43-50 6V/100Ah Battery 1.8 milliohms
[12] J.Power Sources 73 (1998) 60. By comparison, thicker-plate VRLA products
[13] M.J.Weighall, ALABC Project No. R/S-001, Final Report, have the following published impedance values:
October 2000. 12V/2Ah Yuasa Battery 80 milliohms
2V/2.6Ah Portable
[14] CSIRO, ALABC Project S1.1, Progress Report 1, July-
Energy Products Cell 30 milliohms
December 2000. 2V/4.3Ah Portable
[15] R.F.Nelson, Batteries International, 43, April 2000, 51-60 Energy Products Cell 24 milliohms
[16] R.F.Nelson, Batteries International, 34, Jan 1998, 87-93
[17] G.Zguris, 16th Annual Battery Conference on
Applications and Advances (2001), 163-168.
[18] Hovosorb® Technical Manual, April 1993.
[19] M.J.Weighall, ALABC Project No. R/S-001, Final Report,
October 2000, 44.
[20] A.Ferreira, Batteries International, 36, July 1998, 83-90.
[21] Dr D.Berndt, Oxygen Cycle Meeting, 7ELBC, Dublin,
Ireland, 19th September 2000
[22] P.R. Stevenson and O. Enoki, Proceedings of the 5th ERA
Seminar, London, U.K., April, 1988 Paper 3.2
[23] R.F. Nelson and M. A. Kepros, Proceedings of the 14th
Long Beach Battery Conference, IEEE, 1999, pp. 281-
287
[24] R.S. Robinson and J.M. Tarascon, J. Power Sources 48
(1994), 277-84
References
Table 3.
Sample OCV Chart Used in Manufacturing to Sort Cells or Batteries After Formation or Recharge
Days Off Form HV A Categories B R LV-Dead

4 >2.199 2.199 - 2.170 2.169 - 2.140 <2.140 <1.999
5 >2.196 2.196 - 2.167 2.166 - 2.137 <2.137 <1.999
6 >2.194 2.194 - 2.165 2.164 - 2.135 <2.135 <1.999
7 >2.192 2.192 - 2.163 2.162 - 2.133 <2.133 <1.999
8 >2.190 2.190 - 2.161 2.160 - 2.131 <2.131 <1.999
9 >2.188 2.188 - 2.159 2.158 - 2.129 <2.129 <1.999
10 >2.186 2.186 - 2.157 2.156 - 2.127 <2.127 <1.999
11 >2.185 2.185 - 2.156 2.155 - 2.126 <2.126 <1.999
12 >2.184 2.184 - 2.155 2.154 - 2.125 <2.125 <1.999
13 >2.183 2.183 - 2.154 2.153 - 2.124 <2.124 <1.999
14 >2.182 2.182 - 2.153 2.152 - 2.123 <2.123 <1.999
15 >2.181 2.181 - 2.152 2.151 - 2.123 <2.123 <1.999
16 >2.180 2.180 - 2.151 2.150 - 2.123 <2.123 <1.999
17 >2.179 2.179 - 2.150 2.149 - 2.123 <2.123 <1.999
18 >2.178 2.178 - 2.149 2.148 - 2.123 <2.123 <1.999
19 >2.177 2.177 - 2.148 2.147 - 2.123 <2.123 <1.999
20 >2.176 2.176 - 2.147 2.146 - 2.123 <2.123 <1.999
21 >2.175 2.175 - 2.146 2.145 - 2.123 <2.123 <1.999
22 >2.174 2.174 - 2.145 2.144 - 2.123 <2.123 <1.999
23 >2.173 2.173 - 2.144 2.143 - 2.123 <2.123 <1.999
24 >2.172 2.172 - 2.143 2.142 - 2.123 <2.123 <1.999
25 >2.172 2.172 - 2.143 2.142 - 2.123 <2.123 <1.999
26 >2.171 2.171 - 2.142 2.141 - 2.123 <2.123 <1.999
27 >2.170 2.170 - 2.141 2.140 - 2.123 <2.123 <1.999
28 >2.169 2.169 - 2.140 2.139 - 2.123 <2.123 <1.999
29 >2.168 2.168 - 2.139 2.138 - 2.123 <2.123 <1.999
30 >2.168 2.168 - 2.139 2.138 - 2.123 <2.123 <1.999
31 >2.167 2.167 - 2.138 2.137 - 2.123 <2.123 <1.999
32 >2.166 2.166 - 2.137 2.136 - 2.123 <2.123 <1.999
33 >2.165 2.165 - 2.136 2.135 - 2.123 <2.123 <1.999
34 >2.164 2.164 - 2.135 2.134 - 2.123 <2.123 <1.999
35 >2.164 2.164 - 2.135 2.134 - 2.123 <2.123 <1.999
36 >2.163 2.163 - 2.134 2.133 - 2.123 <2.123 <1.999
37 >2.163 2.163 - 2.134 2.133 - 2.123 <2.123 <1.999
38 >2.162 2.162 - 2.133 2.132 - 2.123 <2.123 <1.999
39 >2.162 2.162 - 2.133 2.132 - 2.123 <2.123 <1.999
40 >2.161 2.161 - 2.132 2.131 - 2.123 <2.123 <1.999
41 >2.160 2.160 - 2.131 2.130 - 2.123 <2.123 <1.999
42 >2.160 2.160 - 2.131 2.130 - 2.123 <2.123 <1.999
43 >2.159 2.159 - 2.130 2.129 - 2.123 <2.123 <1.999
44 >2.159 2.159 - 2.130 2.129 - 2.123 <2.123 <1.999
45 >2.158 2.158 - 2.129 2.128 - 2.123 <2.123 <1.999
46 >2.157 2.157 - 2.128 2.127 - 2.123 <2.123 <1.999
47 >2.157 2.157 - 2.128 2.127 - 2.123 <2.123 <1.999
48 >2.156 2.156 - 2.127 2.126 - 2.123 <2.123 <1.999
49 >2.156 2.156 - 2.127 2.126 - 2.123 <2.123 <1.999
50 >2.155 2.155 - 2.126 2.125 - 2.123 <2.123 <1.999
51 >2.155 2.155 - 2.126 2.125 - 2.123 <2.123 <1.999
52 >2.154 2.154 - 2.125 2.124 - 2.123 <2.123 <1.999
53 >2.154 2.154 - 2.125 2.124 - 2.123 <2.123 <1.999
54 >2.154 2.154 - 2.125 2.124 - 2.123 <2.123 <1.999
55 >2.153 2.153 - 2.124 2.123 - 2.123 <2.123 <1.999
56 >2.153 2.153 - 2.124 2.123 - 2.123 <2.123 <1.999
57 >2.153 2.153 - 2.124 2.123 - 2.123 <2.123 <1.999
58 >2.153 2.153 - 2.124 2.123 - 2.123 <2.123 <1.999
59 >2.152 2.152 - 2.123 2.123 - 2.123 <2.123 <1.999
60 >2.152 2.152 - 2.123 2.123 - 2.123 <2.123 <1.999
Appendix
APPENDIX 1 – Glossary Of Terms And Abbreviations

VRLA Valve Regulated Lead Acid
SLI Starting, Lighting, Ignition
UPS Uninterruptible Power Supply
V Voltage
I Current
R Resistance
Ah Ampere Hour
OC Open Circuit
OCV Open Circuit Voltage
TOCV Top Of Charge Voltage
EOCV End Of Charge Voltage
LV Low Voltage
HV High Voltage
CC Constant Current
CV Constant Voltage
TC Taper Current
PC Pulse Current
DC Direct Current
AC Alternating Current
HRD High Rate Discharge
Pb Lead
PbO Lead Monoxide
PbO2 Lead Dioxide
Pb3O4 Red Lead
H3PO4 Phosphoric Acid
CO Carbon Monoxide
CO2 Carbon Dioxide
3BS/ TRB Tribasic Lead Sulfate
4BS/ TTB Tetrabasic Lead Sulfate
NAM Negative Active Material
PAM Positive Active Material
SEM Scanning Electron Microscope
BET Technique for surface area determination
XRD X-Ray Diffraction
S.G./ s.g. Specific Gravity
H&V Hollingsworth and Vose
AGM Absorptive Glass Mat
RBSM Recombinant Battery Separator Mat
MFG Microfine Glass
L/d ratio Plate height/ Plate spacing ratio
Kpa Kilo Pascals
CFM Cubic Feet per Minute
COS Cast on Strap
= Factory Locations
= Representatives
Providing World-Wide Support
for our Customers
Digatron/Firing Circuits, Inc. Digatron Hong Kong Co. Digatron (Qingdao) Digatron
230 Long Hill Cross Road Flat D, 3/F Cheerrick Bldg. Electronic Co., Ltd. Industrie-Elektronik GmbH
Shelton, CT 06484, U.S.A 65 Sau Fu Street 8 Keyuanjing 3 Road, Tempelhofer Strasse 12-14
Tel: 1 203 446 8000 Yuen Long, Hong Kong Hi-tech Zone 52068 Aachen, Germany
Fax: 1 203 446 8015 Tel: ++852 2478 1011 266101 Qingdao, China Tel: ++49 241 168090
E-Mail: info@firing-circuits.com Fax: ++852 2474 5060 Tel: ++86 532 8870 5292 Fax: ++49 241 166465
www.firing-circuits.com E-Mail: digatron@hongkong.com Fax: ++86 532 8870 5259 E-Mail: info@digatron.de
An ISO 9001 company E-Mail: www.digatron.de
info@digatron.com.cn

VRLAmanual

Încărcat de

Informații document

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

VRLAmanual

Încărcat de

Drepturi de autor:

Formate disponibile

A Guide To VRLA

Mike Weighall is an independent Bob Nelson is an independent

© 2001 Firing Circuits, Inc.

www.firing-circuits.com Printed In U.S.A., All Rights Reserved

Paragraph Page Paragraph Page

1. Introduction ....................................................................... 1 8. Battery Design ................................................................ 26

Figure Page Figure Page

Table 1. Typical ampere-hour inputs in relation

The brochure is divided into several sections:

constraints as described in more detail later. However,

3.1.1 Acid Density for Filling 3.3 Formation

3.3.2 “Fill and Spill” For mation

3.5 Completion of formation 3.6 Formation Algorithms

3.7 Initiation of Formation Current Flow Figure 1

heavily sulfated and/or the fill-to-form time has been

3.8 Constant-Voltage Charging Figure 2

parameters in order to apply optimal current levels for Figure 3

Figure 4 approach. It is well suited to VRLA formations, as many

moderate overcharge amounts by allowing high

inrush currents (relative to CC) as well as high

uniformity of formation profiles for different battery

lots is not high; total Ah input may vary

that in the “off” periods, rests or partial or complete

to be beneficial in eliminating surface charges from the

Figure 5 plate pores. The charge efficiency of the positive

Ah more rest periods or discharges into the formation

products (2.0 mm or more) than those with thin plates Figure 7

Voltage, Temperature, Gassing Rate

Formation Current, Amperes

finished-product quality. For VRLA batteries requiring

Voltage, Temperature, Gassing Rate

3.13 Sample Formation Algorithms & Profiles

Figure 8 during the rest of the formation. After this, a fixed CC

Voltage, Temperature, Gassing Rate

(even though the charging current is higher to

could be done. However, as noted, there is no clear

■ Weigh the batteries prior to formation, but at the

and, in some cases, acid spraying or

regurgitation. The primary loss is from

forming of small numbers of batteries, but it

2 should not be assumed that product quality would

■ Wire up the test batteries so that the following

Separator is open, all plate Ingress of acid is impeded

when the filling process is completed. This ideal

PbSO dissolves in liberated CO2 from

Positive Pasted Plate Negative Pasted Plate

Cooling in air at 21ºC.

7.4 Completion of formation 7.5 Alternative Jar Formation Options

8. Battery Design 8.1 Plate height/plate spacing ratio (L/d)

Figure 15 9. Separator Optimisation

9.1 Volume Porosity 9.3 Separator Caliper

However, recent ALABC research has shown that Figure 16

Mean Pore Size (microns)

the fiber mat. 10

plate group pressure (> 40 kPa) BET Surface Area, m2/g

Figure 17 10. Separator designs to improve

7.00 fiber with reinforcing glass strands, the balance being

500 material contained in the separator. The separator also

80 = Non Oriented AGM

These issues may seem trivial, but many formation

designs and/or materials and, more commonly, by Figure 20

12.2 Formation Bay or Circuit C. Series-Parallel Matrix Connections

By far the most common approach is to have a series

Figure 21 13. Battery Monitoring During