There have been many myths about lead-acid battery cycling.

This article will give the reader a better understanding of the more important aspects of battery cycling and battery choice. Different applications are discussed along with charger requirements and how the number of cycles varies with the depth of discharge. This article looks at applications where the recharge is not regular such as solar and wind charging. Intermittent diesel charging is also discussed.

This article does not include cycling applications such as those for motive power applications where the battery is completely isolated form the load for recharging.

In this article we are only interested in “true” cycling batteries. Batteries that are used regularly for applications such as switch closing & tripping or engine starting do not fall into the scope of this article.

A cycling battery can be defined as one that is subjected to frequent but not necessarily regular discharges and recharges. The discharge may be of relatively short duration but at a relatively high current or vice versa. The discharge may remove all of the capacity from the battery or only a little and the recharge may be a full recharge to 100% capacity or a partial charge.

In some cycling applications, the discharge may be considered as over a full year followed by a 100% recharge under controlled parameters. For this type of application the battery is usually removed from the site for recharging.

Solar applications are becoming more popular but present a new challenge. The night-time discharge will normally remove a reasonable capacity form the battery but there must be sufficient left in reserve to provide the next night-time demand even if the battery is not fully charged by the solar panel.

Typically, depending on the reliability of the weather, calculations can be made to minimize the battery and solar panel sizes. For example, in some applications, the capacity returned during daylight hours may be small or even negligible and the battery must have a large reserve. A calculation must be made to determine the minimum capacity required to bridge the gap between acceptable recharging periods. In applications where the solar energy is small, it may be necessary to install a secondary recharge system such as a diesel generator. It may be that this only needs to run for 24h once every month for some parts of the year. A system where the battery is only recharged fully by a diesel generator similar is desirable if a good cycle life is to be expected.

In other applications we may find just the opposite. Sun may be relied upon for 360 days a year but we must have something in reserve.

In all applications we need some reserve power not only to cover for times when the sun does not recharge the battery correctly but if an acceptable cycle life is to be achieved. We do not want to completely discharge the battery. An example of the number of cycles available form a battery may be 7000 cycles to a 10% depth of discharge to 1000 cycles when the depth is 80%. Cycling batteries should never be discharged below 80% discharged.

Clearly, the solar array will need to have current and voltage controlled.

Wind powered systems are very similar but can have the advantage of charging when there is no daylight. On the other hand, lack of acceptable wind force can severely disrupt the charging cycle.

In both solar and wind applications, it is not unusual for a diesel generator to be present on site. By carefully monitoring the battery voltage either on load or open circuit, the generator can be automatically started to recharge the battery.

These applications are to all intent and purpose similar to the solar or wind application.

Both solar panel and wind power charging characteristics will need to be integrated and meet a common characteristic for charging. From an electronics point of view this is simple.

It may be argued that if mains power is available then there will be no need for a battery to discharge. Some applications may have a limited a.c. power supply that is far less than the load demands during peak times. The addition of a battery to cater for these peak demand times is not unusual.

If the cycle is regular, the choice of battery capacity is relatively easy. We need to know the power requirements for the load and the power available from the a.c. mains and we can then choose a battery charging regime to suit.

Ideally, the charging cycle should put back at least what was removed but without overcharging the battery. Overcharging will reduce the battery life. The battery load will inevitable vary if only slightly and some sort of automatic charge termination will be required. To complicate issues, the load may have tight voltage limits and the charger will need to have an upper voltage limit if it is an “on line” battery system. It will be necessary to calculate the approximate power returned before the voltage limit is reached within the available time before the next discharge. Typically, there will be a conflict and it will be necessary to “over size” the battery. Most applications of this type involving regular discharges and reliable recharges will operate between about 80% charged and 80% discharged.

It would be good practice to 100% charge batteries in this type of application once a year. This technique will enhance the overall battery life and reduce plate sulphate which is often the life limiting factor in cycling applications.

Some simple but critical characteristics are required if the battery is to perform with an acceptable performance and life. Two parameters are required for charging the battery; current and voltage. Without an acceptable current the battery will not recharge correctly and without the correct voltage the required current will not flow in the battery circuit.

Manufacturers will advise design engineers and users of the minimum requirements but typically, a current of at least 10% amperes and 2.40Vpc are common recommendations.

A common error is for a large battery to be connected to a small solar array. For example, lets consider a load requirement of 15A for an 18-hour night. This will remove 270Ah and on the basis that we need to consider operating between 80% charged and 80% discharged, we need to add 40% to the calculated battery capacity resulting in a capacity requirement of about 376Ah. At this point we need a 38A charger and 2.40Vpc x 12 cells = 28.8V.

What we did not consider was that the sun may not provide enough power to recharge correctly every day. If we assume that we only have minimal power for 5 consecutive days then we need to revise the battery capacity to 15A x 18h x 5 plus 40% = 1890Ah. We now need a charger of 189A and 28.8V. This will work very well if the predictions on solar energy are correct. Again, some type of automatic control will be required to prevent the battery overcharging when the solar array is operating over several days at full power.

Now lets suppose the night-time load has increased to 30A and not 15A. The calculation becomes 30A x 18h x 5 plus 40% = 3780Ah. We now need a 378A solar array.

If this were left at 189A, i.e. 5%, the battery would not fully charge and would fail due to sulphation of the plates.

It is important to consider the operating temperature for cycling batteries. There are three main affects: a) the charging voltage that should be applied, b) the discharge performance which reduces as the operating temperature falls and c) the maximum temperature which can adversely affect the cell containers; most containers will start to soften when the temperature exceeds 50°C.

Manufacturers all give the charging voltage recommendations with respect to temperature and the graph below is typical.

All lead-acid batteries will cycle. However, the type of battery, depth of discharge, recharge characteristics and operating temperature will all affect the life. It has been said that VRLA AGM batteries do not cycle. This is not true. There are laboratory examples of VRLA AGM batteries achieving over 700 cycles. The depth of discharge was kept to a maximum of 80% and the charging programmeclosely controlled to ensure minimum overcharging with adequate recharging. Clearly, any laboratory experiment has to be put into context with the practicalities of a live working system and any published data from manufacturers must allow for some margin to cater for these practicalities in the system. Although VRLA AGM batteries will cycle, they do not represent the most cost effective type unless a short life is all that is required. Batteries that fail prematurely generally do so because they have been undercharged.

As explained briefly above, the depth of discharge affects the battery life and the graph below is a typical cycle life v depth of discharge for a tubular plate battery.

#BATTERY CAPACITY

Batteries for cycling applications are often rated at the 100h rate rather than the 10h rate. This can be confusing because the same product rated at the 10h rate may be 100Ah and at the 100h rate it may be 120Ah.Occasionally batteries may be rated at the 1000h rate. It is also worth remembering that when the discharge rate is longer than about 28 days, (672h) self discharges losses will be considerable. Calculations must take this into consideration. The self discharge losses of a lead acid battery are quite low and manufacturers will typically quote 2% per month but this will probably be over a 6 months period at 20°C. During the first month the loss can be 5% and for every 10°C above 20°C, the losses will be double. This means that at an ambient of 30°C the loss can easily be 10% over one month and this must be considered when sizing the battery.

The different rating does not affect the real performance but the user has to take care when comparing products.

Tubular batteries are arguably the best type for cycling applications and these come in two different types, Vented or Wet and VRLA GEL. The DIN Specification OPzS vented and DIN OPzV GEL tubular plate product has been designed “up to a specification” rather than down to a price and are the prefered choice by many specifiers and users. These DIN sizes typically start at 2V x 100Ah and go to 2V 3500Ah.

Flat plate pasted types of the correct design can also be good for cycling applications. However, the specific type must be suitable for this application. There are arguments in the market place that favour pasted plate to tubular plate. A suitably designed flat plate can be as good as a tubular plate type but in Euripe, the tubular plate excells.

Many manufacturers use special additives in the plates and electrolyte to aid cycle life. Some of these additives reduce the capacity slightly but more than make up for this loss by having a better cycle life and ultimately being more cost effective. There may be slight differences in the makeup of the battery but a significant change in cycle life can result.

Care has to be taken when choosing a cycling battery. Some manufacturers may quote 5000 cycles for a Tubular GEL product while a different manufacturer may quote 6000 cycles for what seems to be an almost identical product. The choice is difficult to make. Ask the question, does the manufacturer have a good track record for the product they are proposing and do they have a customer list? Can the manufacturer provide technical support if needed? What conditions have been used for recharging; was the recharge under laboratory conditions using constant current or at a more realistic constant voltage.

Research is essential to ensure the product is suitable for the application.

This article looks at the preferred designs for battery rooms and discusses how batteries should be laid out to give a safe environment. Alternative battery stand types are discussed to illustrate accessibility of the cells or monoblocs and safety considerations.

VRLA, Vented and Nickel Cadmium battery types are included.

Fully detailed information can be found in International Standards such as BS EN 50272-2:2001. This article gives an overview only to the more important subjects.

Battery rooms can be a hazardous place and all persons entering must be aware of the dangers. All too often, inexperienced people enter a room without receiving any safety information. Visitors who may have never been in a battery room previously are particularly vulnerable and must be give a short overview of the hazards. Topics discussed should include emergency evacuation procedure although this may be covered immediately after entering the room when emergency exit routs can be pointed out. High voltage and the need not to touch any part of the battery or stand should be explained. Battery acid and lead compounds and the risk of explosion due to the build up of explosive gasses should be discussed. The hazards with nickel cadmium batteries, which contain highly corrosive potassium hydroxide and give off hydrogen, should be discussed.

No persons should be allowed to enter a battery room without the correct clothing. Man made synthetic clothing may generate a static charge which can result in a spark which in turn may ignite the explosive gasses given off by all lead-acid and nickel cadmium battery types. This is less critical when considering VRLA batteries. For people not actually working on the battery or touching it, such as visitors, every day clothing is acceptable. The owner of the battery has a duty of care towards all persons entering a battery room.

A fully documented Method Statement and Risk Assessment must be covered for any person working on any part of a battery system. Even visitors may be required to sign the documents to confirm they have understood them and will comply with a simple Method Statement, even if they are not directly involved in the work. Lone working is possible when working in a battery room providing the activity is clearly understood and emergency procedures are in place.

As a final overview, all doors to the battery room must be anti-panic and open outwards.

Because battery rooms are a hazardous place, appropriate signage must be applied to the door. Doors should be locked to ensure only authorised persons can enter. A list of typical signs for lead acid batteries is given below. These signs are self-explanatory.

The examples given below are not exhaustive but do give the reader an appreciation of the hazards that may be encountered within a battery room. Where nickel cadmium batteries are installed, the appropriate warning for the potassium hydroxide, nickel and cadmium should be posted.

Batteries may be mounted on racks or in cabinets. When installed on racks, these may be of wood or steel and both may be insulated from earth. Generally speaking, battery stands are not earthed but isolated from earth. Some steel stands are earthed and consideration should be given to the method of earthling each component, such as rails and frames, to the next component as well as to “ground”. While steel stands are normally isolated from ground, steel enclosures are normally earthed. There is no Standard that covers earthling and the decision is left to the user.

Induced voltages may be generated into an insulated stand by ripple current in the battery circuit. Caution has to be applied before touching any metal stand rail or frame. Older UPS systems where notorious for inducing a high voltage and while modern systems generally have very low ripple current, caution should also be applied until it is established that no high induced voltage exists and there is no risk.

These should be designed and installed to provide good access. Multi-tier steel stands are very popular for UPS batteries. These may be up to 6 tiers high and measuring over 2m. Not very popular but still evident is some applications are very high stands with vented lead-acid cells or vented nickel-cadmium cells. Working at height on these batteries is not easy and special safety measures must be used. Very occasionally, even higher stands may be used. High stands present a high risk factor. Working at height has its own hazards and when working on live batteries extreme care must be taken. It is worth remembering that a battery cannot be isolated in the conventional way and will always be live, even when fully discharged. It only takes a few milliamperes to kill a person.

Stands that are two rows deep are generally easy to service but those of three or four or more rows deep may be difficult to service. Stretching over the front row(s) can result in electrocution or shorting of cells or monoblocs unless they are fully insulated. Insulating covers may be accidentally knocked off if care is not taken. A better design would be one having a total height where cells can be serviced without steps or ladders and where the depth of the stand is restricted so that the cells at the back of the stand may be reached easily. Tier pitching should also be considered at the design stage. Again, good spacing allows easy access and makes servicing easier.

Batteries housed in enclosures are notorious for having poor access. The writer has seen examples of enclosures, which are over 1m deep with less that 50mm between the top of cells and the underside of the shelf above. When the manufacturer was asked how the cells were installed, it was admitted that they were built up from the bottom and shelves added as required. This was a special design where this was the only way the required power could be provided in the volume available. Servicing the battery was impossible. Even long probes could not reach the terminals of the back monoblocs.

To cater for emergency exit situations the distance between stands or enclosures should be considered. It is reasonable to consider a “gang way” of not less than 600mm. However, if the “gang way” is partially closed such as during servicing, a larger gang way should be considered. It is not acceptable to use 600mm “gang-ways” and then obstruct them at any time.

Below are two examples of stand arrangement for 55 High Performance Planté cells. One offers good access whilst the other is much more compact making servicing difficult.

Typical 2 tier x 2-row stand for 55 x Planté cells – This is a good design giving good access.

Typical 4 tier x 3-row stand for 55 x Planté cells – This has a smaller footprint but is much higher, has less tier spacing and has restricted access for servicing the cells.

Ventilation recommendations are given in National and International Standards and this section is included to give the reader an appreciation of the risks involved when entering or working in a battery room.

Ventilation must be provided and be adequate to remove hydrogen from the room to a concentration of less than 4% which is the safe lower limit to prevent an explosion. It is preferable to use natural ventilation because forced ventilation systems can fail.

Many battery rooms do not have adequate ventilation and it is particularly important that when entering any battery room a Risk Assessment is carried out. It may be prudent to open battery room doors and allow any gasses to disperse before entering.

When batteries have been on boost charge such as constant current for vented cells, it is fundamentally important to consider a period of time for gasses to disperse.

Many users use hydrogen detection equipment to measure the concentration and ensure it is at a safe level before entering. The detection equipment may then be kept in the room whilst the technicians are working. The types that will give an audible alarm when the hydrogen level reaches a dangerous level are preferred. The alarm level should be set considerably higher than 4% and a normal suggested value is 10%.

Having discussed ventilation to remove explosive gasses, we need to consider heat generated by the battery both on charge and off charge. Looking specifically at cell or monobloc spacing to allow a flow of air between the units, it is reasonable to allow a minimum gap between VRLA cells and monoblocs of 5mm. It would also be prudent to increase this distance where the battery is housed in an enclosure. Batteries in enclosures are best mounted on rails rather than a solid shelf. Good designs use perforated shelves and increase the spacing to 15mm between cells or monoblocs. It is not unknown for batteries to go into thermal runaway for no other reason that the spacing between cells or monoblocs has been insufficient to cater for the natural heat generated whilst on float charge. Vented cells are less prone to these thermal problems but spacing for these should also be considered at the design stage.

Several factors need to be considered when designing a battery room floor. For VRLA batteries the simplest of protection is normally acceptable but rooms housing vented battery types need to be impermeable for battery acid or alkaline for nickel cadmium types. An alternative to having a complete floor treated is a bunded floor lip, which prevents any spillage from spreading beyond the immediate vicinity of the battery. As a further alternative, drip trays may be used.

For any battery type, the floor must be capable of withstanding the point loading of the stands. Good battery stand manufacturers are capable of providing the point loading details and advising on designs suitable for spreading the load. Whilst point-loading issues can normally be overcome by load spreading plates, in some instances the floor may not be capable of withstanding the battery weight as a whole. If this is suspected, a structural survey will need to be carried out.

For vented batteries, room lighting should be intrinsically safe to avoid explosions caused by sparks igniting hydrogen gas. For VRLA batteries normal lighting may be used. This does assume that the correct battery charging characteristics are used and correct ventilation has been provided.

Lighting should be sufficient for maintenance technicians to be able to see the complete battery without difficulty. With some enclosure design with restricted access almost any type of fixed lighting will not provide the level of illumination required for the service technician. Hand held torches are fine providing they are fully insulated and for vented battery types the unit must be intrinsically

It is easy to assume that battery rooms need not be anything special. However, this article, which is only intended to give the reader “food for thought”, illustrates the need to be cautious when designing a battery room, stand or enclosure. Reference to National and International standards must be made for more detailed information on this subject.

Battery internal resistance and short circuit current values are available from battery manufacturers. The method used to arrive at the published values varies but when a method recognised by International Standards is used a comparison between products can be considered.

Searching the internet will reveal many papers discussing actual case studies where a battery has been shorted and interesting results have been obtained. With no fuse or battery circuit breaker in the system short circuits may result in fires and catastrophic failure. Alternatively, the protection may work and isolate the battery from the point of failure and the load resulting in loss of power to the equipment which the battery was intended to protect. It therefore follows that the subject of battery short circuit current can have at least two points of view when looked at in practical terms.

The internal resistance may be used to calculate the theoretical short circuit current but the method used is open to debate. Never the less, values of the internal resistance may be used to estimate the actual short circuit current in a battery system.

This article discusses how the battery manufacturer arrives at the published internal resistance and short circuit currents. It also looks at how the short circuit current may be estimated in a practical system.

Some manufacturers carry out actual short circuit tests to determine the characteristics. The test method normally involves fully charging the product which has been “characterised” to establish the real performance compared with the actual performance. It is not unusual for the real performance to be considerably better than that published. For this reason, the manufacturer will state that the values obtained are from actual testes on a typical product taken from production. However, there is more to this than simply carrying out the test. The value of the “short” in terms of resistance must be considered. Clearly it is not acceptable to simply connect two connectors coming from the cell pillars and running the test. In one extreme the connectors may melt and act like a fuse, and in the other extreme, the connectors may be large solid copper bars which cannot be easily attached to the cell terminals. Typically, the manufacturer will look at an accepted Standard which will give some guidance to the method that should be used.

The results of actual short circuit testing are very interesting. Within the first few microseconds no current flow is seen. After a few milliseconds the current starts to flow and will rapidly increase to a peak before falling back and eventually reducing to a value somewhere close to the nominal published value. The current will continue to fall and in some cases, may flow for several hours before reaching zero.

Actual short circuit testing is interesting but they do not necessarily give repeatable value that can be used for comparison purposes. Circuit resistance, temperature and actual state of charge all contribute to variable results. The point where the current should be measure has the largest variation. Measurements taken after 1 second will show a large variation when compared with those taken after 1 minute. With this in mind, tests have shown that a “projected” result formulated from actual tests give the most repeatable results. Because the tests are based on the same datum points, comparisons to other products can be made with good accuracy.

The methods which is used by most manufacturers involves discharging at two different rates and plotting the U/I characteristics. Typically, one discharge will be at a moderately high current and after a pre-determined discharge time the actual voltage and actual current will be recorded. The pre-determined discharge time will be quite short to minimize the Ah capacity removed but will be sufficiently long for reasonably stable results to be obtained. The discharge will be terminated and the product will be stood on open circuit and allowed to recover for a few minutes before a second discharge is carried out. This second discharge will be at a higher current of typically three times the value of the first current. Again, after a pre-determined time the actual voltage and actual current will be recorded and the test completed.

The internal resistance is then calculated as follows: -

The short circuit current is then calculated by ohms law where 2.00V may be used in the calculation. Some manufacturers may use the typical open circuit voltage of the cell which will largely be dependent upon the specific gravity of the cell. Values between 2.05V and 2.15V may be used.

The value of internal resistance may be used to estimate the short circuit current in a practical circuit.

Consider a 250 Ah cell having a published internal resistance of 0.33mΩ.

The battery has a total of 24 cells connected in series giving a total cell resistance of 0.33 x 24 = 7.92mΩ. The external circuit has an estimated resistance of 0.5 mΩ.

The estimated short circuit current is: -

I = (24 x 2.00V) / ((24 x 0.33mΩ) + (0.5mΩ) = 48V / 8.42mΩ = 5,701A

In comparison, the published short circuit current for a single cell is 6,150A.

Consider a 2500 Ah cell having a published internal resistance of 0.049mΩ. This battery has 240 cells and the external circuit has a resistance if 21mΩ. The short circuit current is estimated to be:-

(240 x 2.00V) / ((240 x 0.049mΩ) + 21mΩ)) = 480V / (11.76mΩ + 21mΩ) = 480V / 32.76mΩ = 14,652A.

The internal resistance and short circuit values depend on the type of battery under consideration but as a guide, the following are typical values and are given as comparisons only. Manufacturers publish these values and the user should consult these for the correct value.

The values below are for guidance only and consider a 100Ah single cell. Values for multi-cell monoblocs will be different.

From the above table it can be seen that the VRLA AGM product has a much higher short circuit current and lower internal resistance. This characteristic is the result of thinner plates with smaller plate pitch and very low resistance separators coupled with a higher specific gravity electrolyte which enhances the open circuit voltage and initial discharge current. It is noted that the internal resistance of cells does not substantially vary with state of charge until the cell is more than about 80% discharged.

The short circuit current of industrial standby batteries may be extremely high, even when the nominal characteristics do not suggest this.

In a real live situation, even with small batteries, it is not unusual for currents to be several thousand amperes. For large batteries such as those used in Power Stations, short circuit currents may exceed 40k amperes.

Even when the battery is not fully charged, the short circuit current is very similar to the published value because the internal resistance does not vary substantially until the cell approaches fully discharged.

Typical Method of Determination of Internal Resistance

This article discusses the manufacturing principals and processes in the manufacture of VRLA batteries. The article discusses the manufacturing processes and how deviations from the ideal can seriously affect the finished product.

The article assumes the reader has a reasonable knowledge of the construction of both AGM and GEL types. The process described here is very much simplified but will give the reader an understanding of the processes involved in manufacturing VRLA batteries.

It has been said that it is very easy to manufacture a bad VRLA battery and quite difficult to manufacture a good performing product having a durable long life. It is the writer’s belief that this is true.

The following notes will help those not familiar with the manufacturing process to understand this article.

a) Lead (Pb): the main ingredient of the negative plate. b) Lead dioxide (PbO2): the main ingredient of the positive plate. c) Dilute sulphuric acid (H2SO4): the third component to make the battery work. d) Lead grids: the essential component which the positive or negative active materials are applied. The grids are the main current carrying component responsible for delivering power from the active materials to the terminals. Lead as Pb active material is a reasonable conductor of power but lead dioxide as PbO2 is a poor conductor and good grid design is essential. e) Plate lug: the upper most part of the grid which is “welded” to the group bar. In battery language, the welding process is called “burning”. f) Plate wires: the vertical and horizontal wires of the plate grid. g) Group burning process: this is the term used in the process where the plates are welded to the group bar or where the group bar is welded to the pillars. h) Flag: the section of the group bar which protrudes above the plate group which is used for the inter cell weld.

All lead acid batteries have three active ingredients; lead (Pb), lead dioxide (PbO2) and dilute sulphuric acid (H2SO4). The principals have been discussed in a previous article titled “HOW DOES A LEAD-ACID BATTERY WORK”.

In addition to the active materials, pure lead or lead alloy grids are required to “hold” the lead and lead dioxide. Separators are required to keep the plates apart and AGM batteries have an absorbent separator to hold acid whilst GEL batteries have special jellified electrolyte.

It is easy to say the active materials are lead (Pb), lead dioxide (PbO2) and dilute sulphuric acid (H2SO4), but, the purity of these materials is critical to performance and life of the finished product. Impurities in the dilute sulphuric acid can affect the initial performance or the life of the product, in some cases resulting in unexpected catastrophic failure. At first sight the purity of the lead used in battery manufacture seems to be fantastic with values of 99.97% pure being quoted. However, 99.99% pure lead is not only superior but desirable... A purity of 99.97% and even lower may be used but the impurities such as iron, manganese, copper and other harmful elements must be closely controlled. Interestingly, it has been proven that trace elements of gold in the lead will enhance both performance and life. Unfortunately, most of the gold today is extracted from lead before the battery manufacturer takes receipt.

In principal the manufacture of the lead oxide which then becomes lead for the negative plates and lead dioxide for the positive plates should be a simple process. However, the process has to be controlled finitely. Several methods are used to manufacture lead oxide but the most popular are the Barton Pot and Ball Mill. In both, lead oxide is produced in a vessel where fine particles carried away by “extraction” fans are then used to produce the active material. The particle size is very small but the superficial volume is high making the particles act more like dust than lead.

Ball Mill Active Material Manufacturing Machine

The particles are critically controlled for size, density and “free lead”. Variations in the oxide characteristics will affect the battery performance and life. For example, high “free lead” content will reduce the overall efficiency but can improve the overall life of the product. The finished product characteristics can be “manipulated” by varying the lead oxide characteristics within the Barton Pot or Ball Mill manufacturing process.

The next stage of manufacture involves mixing the lead oxide powder with acid, water, and proprietary additives to produce a paste have a consistency similar to thick toothpaste. This is then pasted onto the lead grids. For VRLA GEL cells using Tubular Plates, the process is quite different and the active material is “inserted” into the tubes as slurry or a dry powder.

Tubular plate cells and increasingly pasted plate types have a rather special lead compound added; triplumbic tetroxide (Pb3O4). This material enhances the plate formation acting as a catalyst to produce a more even and consistent active material in the finished product. It is normally produced by a specialist manufacturing plant rather than the battery manufacturer. Relatively speaking it is several times more expensive to produce when compared with lead oxide (PbO2).

It is normal to add a colour element to differentiate between positive and negative paste. Carbon black is often added which arguably improved conductivity. This is a good identifier.

Areas open to error are:- a) Poor quality control of impurities in the lead, lead alloys and dilute sulphuric acid. b) Poor control of particle size and free lead content. c) Poor control of paste density.

Corrosion of the grids is at the front of the battery manufacturers mind and many different alloys and manufacturing methods have been tried to minimise this corrosion. It does not always follow that thick grids will last longer than thin grids. Thick grids will undoubtedly last longer than thin grids, all other aspects remaining the same. Development has taken place and today thin grids with exceptional corrosion resistance are now common place.

The alloys used are normally pure lead, tin, calcium or cadmium. Some exotic alloys are also used but these are very special and not commonly available and in most instances are regarded as “experimental”.

As with the active material, the purity of the lead and alloying element play an important part in achieving the design life. The design of the grids also affects the performance and life of the product. Radial grids are popular with automotive batteries and are increasing in popularity for VRLA industrial products. Inboard lugs, tapered group bars, large inter cell connectors etc all play an important part in the overall design which ultimately will affect the internal resistance and terminal voltage on discharge.

Thin grids have, for many years, been frowned upon but with the availability of very pure lead and by using the correct additives such as tin, superior products have been developed. It is interesting to note that the very first VRLA AGM cells used pure lead grids. The level of tin in the plates varies but typically this is between 2% and 3%. The “hardening agent” such as calcium and cadmium are introduced at much lower levels and it is true to say that the lower the better. Typically 0.05% to 0.10% calcium is added. Some batteries do not contain any “hardening agent” making them delicate to handle in manufacture but resulting in an exceptional product with a very long life.

The lead grids are usually manufactured using a gravity casting process. Grid weight is controlled by the operator attending to release agents and mould “maintenance” which is ongoing throughout the casting process. Grids are weighed at regular intervals to ensure compliance with the specification. An under-weight grid will result is a lower performance and a reduced life while a grid heavier than the design specification will cost more in material and profitability will suffer.

Casting Machines

Grids are also manufactured using a rolled and punching method. This latter method results in a very stable grain structure similar to forged steel when compared with case iron. With a more stable grain structure corrosion resistance is enhanced. When a lead tin alloy is used for rolled and punched grids they can be very thin with a finished grid thickness as low as 0.6mm. By using robotic machinery to assemble finished plates into the containers, thinner plates can be accommodated which results in better active material utilisation and superior high rate performance. Unfortunately there is a cost to this process because thin plates are more difficult to handle in a volume production cycle. For many years finished plates were assembled into the containers by hand.

Before pasting the grids with the active material, they are allowed to age harden for several days and sometimes even weeks. The grids become more robust which makes the next stage of pasting easier.

Areas open to error are:-

a) Poor quality control of impurities in the lead. b) Poor control of additive concentrations. c) Poor mixing of the lead and alloying agents. d) Poor control of the casting temperature e) Poor control of the grid thickness.

Battery Grid Plates

The paste has to be added to the grids and automatic machinery is used for this process. Grids are fed, horizontally, under the head of the pasting machine which normally consists of counter rotating “screws” which force the paste into the grids. The underside of the grid is said to be “flush” while the upper side has a small amount of “over paste”, typically about 0.1mm. At this stage we can call the product “plates”. These plates are fed into a “flash drying oven”. After passing through the flash dryer, plates are sufficiently dry to allow stacking. The flash dryer is a critical area where a good or bad product may be manufactured. If the flash drying temperature is too high a “skin” can form which inhibits plate formation. Both performance and life will be adversely affected. If flash drying is inadequate, plates cannot be stacked separately because they will stick together. With plates not separated correctly the curing process will be affected.

Grids Being Fed In To A Pasting Machine

Plates After Pasting

Before flash drying, some special battery types are returned to the start of the pasting process and turned over before being pasted again. This results in the grid wires being embedded within the plate active material which enhances life. The down side is the additional cost with questionable benefit.

Flash Dryer

For Tubular plates the term pasting is inaccurate because the tubes are filled with positive active material in a power of slurry form. However, this only applies to the positive plate. Negative plates follow the same process for all lead acid battery types.

Areas open to error are:-

a) Inaccurate blending of the dry lead oxide with acid and water. b) Incorrect quantity of additives and / or ineffective mixing. c) Insufficient quantity of active material inserted into the tubes for Tubular plate cells. d) Under pasting resulting in insufficient weight and the consequential loss of performance. e) Missing pellets, i.e. parts of the grid not pasted or sections missing. f) Bent plates as a result of machine settings not correct.

After pasting and flash drying the plates are cured. There are several methods for curing plates but the most popular is to use large curing ovens where the oxygen content, temperature and humidity are controlled. It is possible to cure plates without going into a controlled environment but a radically different manufacturing process has to be used later in the manufacturing cycle. The curing process takes time and even when using curing ovens the process takes many hours.

Plates coming out of the ovens should be completely dry.

Areas open to error are:-

a) Poor control of the oven temperature, humidity and oxygen content. b) Curing time too short. c) Stacking in the oven too tight leading to uneven or inadequate drying.

Plates Being Loaded In To Curing Oven

Curing Ovens

Before the plates can be assembled into the cells, a finishing process is required to remove paste from the plate lugs, and edges. Loose material must also be removed and this is particularly important for AGM product. A plate having a piece of dried active material on the surface will eventually puncture the separator and a short will be inevitable. Many manufacturers over paste the grids and then plane them down to the correct thickness.

Areas open to error are:-

a) Inadequate cleaning leading to dry paste debris not being removed. b) When plates are planed to the correct thickness, this may not be to the desired design leading to under or over compression in AGM cells.

Plate stacking is where the positive plates, negative plates and separators come together to form a “group” or “stack”. This is more critical in AGM cells compared with GEL types.

Stacking can be either by hand or machine. For AGM cells the separator is more often than not wrapped round one plate or both plates. This may be the positive or negative plate. Occasionally, wrapping is not carried out and a single sheet is inserted between the plates. A very popular process is to wrap both positive and negative plates because two thin layers of separator are better than one thick layer.

For GEL cells a conventional separator is used and the plate stacking process is very much simplified. Double separation is often used which generally is in the form of two different types of separator “bonded” together to form one piece which makes plate stacking very simple. Generally, this process is dome automatically be machine.

Areas open to error are:-

a) Plates can be damaged which may ultimately find their way into finished cells. b) Separator damage may occur.

#GROUP BURNING

The term “burning” is used to describe the “welding” process when the plates are connected to the group bar and pillars.

This process involves loading the plate group (see plate stacking above) into a jig ready for group “burning”. The burning process may be by hand or machine. In the case of monoblocs, several cells are normally assembled into a multi cell jig.

Finished Battery Plate Ready To Be Assembled In To The Burning Jig

For large cells typically over 500Ah, group burning is generally done by hand involving skilled operators burning (welding) the plate lugs to group bars. The group bar may already be an integral part of the cell pillar or may be formed with molten lead by the operator. The principal is simple to describe but great skill is required to obtain a grain structure which is corrosion resistant. Too much local heat will result is a coarse grain structure at the plate lug / group bar interface which will be liable to rapid corrosion. If too little heat is used, a poor electrical connection will result.

Typical Hand Burning Process   Smaller capacity cells and monoblocs tend to be assembled using a “cast on process”. This involves loading the plate groups into a jig with the plate lugs uppermost; the jig is then inverted with the plate lugs facing vertically down. The lugs of the complete group or groups are then cleaned and flux is applied... The plate lugs are then dipped into molten lead which effectively welds them together. The molten lead is in a cavity which then forms the group bar and pillar or inter cell “flag”.

Areas open to error are:-

a) Plates not burned (welded) to the group bar adequately. This may lead to local corrosion and premature failure. This may be caused by inadequate lug cleaning, too much or too little flux being applied, plate lugs being misaligned. b) Too much heat applied during the burning process resulting in a coarse grain structure susceptible to corrosion. c) Poor choice of lead alloy leading to early corrosion of the group bar / pillar.

This is the mechanical process of inserting the plate group into the container. It is purely mechanical but can lead to early failure of the product if not controlled.

Exploded View Of A Cell Group

VRLA AGM cells have a microfiber separator which relies on being compressed in the finished product. As a plate group it will the larger than the cell container and must be compressed and “shoe horned” into the container. The procedure may involve a machine to ram the group through a tapered “shoe horn” or for small cells a hand insertion method may be used.

VRLA Battery Assembly Line

VRLA GEL cells are easier to assemble because the group is not a compression fit into the container. Simple machines or a hand insertion method may be used.

Areas open to error are:-

a) Over compression at the “shoe horn” stage leading to fractures in the micro fibre. This may ultimately lead to loss of compression and failure of the product. b) Incorrect alignment of the “shoe horn” to the container leading to snagged plates which can lead to internal shorts.

For monoblocs, an inert cell connection must be made. Very occasionally this is done externally in a similar way to connecting single cells in series. However, most monoblocs have an inter cell weld. This is normally achieved by an “extrusion fusion” method which, in mechanical terms is a spot weld. The process has to be controlled very closely because the weld is “through” the inter cell wall and it has to be acid tight as well as being electrically correct.

The process involves applying pressure to the group flags which will force lead through the cell wall. With pressure still applied, a high current is applied which melts the lead at the electrodes. With pressure still applied, the current flow is stopped to allow the lead to solidify and finally, the pressure applied by the electrodes is removed.

Areas open to error are:-

a) Poor alignment of the electrodes. b) Inadequate electrode pressure c) Bad current control. d) Worn electrodes

Typical Cast On Strap Machine

Cast On Strap Machine Being Loaded

Having inserted the plate group into the cell, and completed the inter cell welds for monobloc types, the lid must be fitted. The lids are normally sealed to the container by either; a) heat seal method or b) adhesive bonding. Some small cell lids are ultrasonic sealed but this method is the quite rare for industrial batteries above 25Ah.

As the term implies, heat sealing involves applying heat and bonding the lid to the container. It is normal for a hot plate to heat the mating surfaces of both lid and container and while the mating surfaces are still in a semi-molten state the two components are brought together. The method is well proven and very reliable.

Adhesive bonding is very popular because a “hot plate” does not have to be manufactured for every combination of lid and container. Adhesives normally come in the form of the two-part mixed at the point of application. Large cells and those with low volume production runs are more than often adhesive bonded. Modern adhesives are extremely strong and reliable but some polymers cannot be easily bonded and heat sealing is the only solution.

It is normal to pressure test cells after lid sealing. Often very high critically important product, every cell may be tested using helium which is very searching.

Areas open to error are:-

a) Poor heat bonding caused by inaccurate machine setup. Too hot, to cold, wrong cycle time and misalignment of tooling contribute or cause poor heat bonding. b) Adhesive bonding problems can occur due to inadequate adhesive being applied, or a poor mixing of the two parts.

Some products have plates formed prior to cell assembly in a process known as “tank formation”. The alternative formation process is known as “container formation” where cells are filled with acid and the formation process is carried out.

Arguably, “tank formation” is more reliable because the temperature can be more closely controlled and the acid circulation is achieved. The down side is that the process is considerably more expensive than “container formation”.

Cells going through the “container formation” process must first be filled with acid which normally ends up as the finished cell electrolyte. However, some large GEL cells are formed with a disposable liquid acid which is then dumped before the cell is filled with the finished acid mix. Inn GEL cells the acid mix is formed by adding fine silica, and other proprietary ingredients to dilute sulphuric acid. This produces a milky acid mix which is thixotropic. The mixture is poured into the cells and soon sets into a GEL forming the electrolyte.

The process for VRLA AGM cells following the container formation process, normally involves filling with cold acid typically of about +5C. As soon as the acid reaches the plates which at this stage are in the lead oxide (PbO) form, considerable heat is generated and this must be controlled. Cells or monoblocs are normally formed in water filled baths to control the temperature. Filling the cells is quite complex because the micro-fibre separator does not allow instant filling. Typically, a vacuum will be pulled before acid is allowed to flow into the cell. Most processes involve several pulses of pulling a vacuum and acid flow.

The actual formation process involves putting typically, 7 to 10 times the nominal Ah capacity of the cell into the cells. For example, a nominal 100Ah cell will have 700Ah to 1000Ah input before it may be considered fully formed.

The process normally involves a controlled discharge part way through to check the capacity and “exercise” the active materials.

The formation process is critical the performance and life of the finished product.  

Areas open to error are:-

a) Inadequate acid filling. This is more likely to occur on AGM cells and large tubular plate GEL cells. b) Poor mixing of the silica and acid for GEL cells. c) Poor temperature control during the formation process. d) Inadequate capacity input.

Most products are checked for voltage and some type of conductance, impedance or performance testing is carried out prior to despatch.

This is the “final” QA check that the product normally must pass before packing and shipment.

Areas open to error are:-

a) Inadequate checks allowing defective product through and into the supply chain.

Machine For Final Testing Of Voltage & Impedance

This document covers the more obvious and more critical processes used to manufacture a VRLA battery.

Because batteries are not regarded as “exciting” and are a “low tech” product, the complexity of manufacture is not interesting to the lay man. This article is written to illustrate that the processes are far from simple and extremely complex.

The article only “skims” over the subject of metals used but these are critical if a high quality; long life product is to be manufactured.

This article discusses the effects of storing VRLA AGM & GEL types and also vented Planté, Pasted Plate and Tubular types.

The article covers the effects of time, temperature, humidity and light and also discusses sulphation; lead dendrite growth and corrosion of the internal lead parts as well as corrosion to the pillars and UV aging of the plastic containers. The relationship between open circuit voltage and state of charge is also discussed.

The article is intended to give the reader a better understanding to the reasons why battery manufacturers make specific recommendations for storing batteries which should not be ignored. It is not intended to be a definitive article and the battery manufacturer’s instructions should be followed at all times.

Battery manufacturers quote very different maximum storage times for their batteries and this can be confusing for the installer and user. One manufacturer may quote 3 months whilst another may quote 2 years. In the first instance we need to look at the storage temperature because the discrepancy in time may be explained by different temperatures. It is also important to confirm if the storage time is from the date of manufacture or first charge which is normally printed on the product, or from the date of dispatch from the manufacturing plant. Never the less, the storage time for “like to like” circumstances may be significantly different depending on many factors including design, materials and manufacturing differences.

Product quality in terms of material purity affects the storage time as does the minimum state of charge that the manufacturer is happy that the product may be successfully commission charged and what the commissioning charge recommendations are.

It is true to say that the longer the product has been stored; the more attention to commission charging is required.

For product supplied filled and nominally charged, the product will “self discharge” and lead sulphate will form on both positive and negative plates; and the specific gravity of the electrolyte will fall. The longer the storage time and the more difficult it is to convert this lead sulphate back to active materials. Where the product has been stored for a long period of time, some lead sulphate may never be re-converted and a permanent loss of capacity will result.   Product that is supplied “dry charged” is more easily dealt with but this condition is reserved for vented Planté, vented Pasted Plate and vented Tubular types. Although the “dry charge” will be lost over the storage time, the condition can be recovered because the electrolyte is added after the storage time and only a limited amount of sulphate can be formed during storage. However, most manufacturers stipulate a storage time based on an 80% state of charge being achieved after filling with dilute sulphuric acid. A long storage time may result in no real capacity being seen after filling which can normally be recovered providing the correct charging regime is followed. This charging regime may be different than that required for a storage time within the manufacturer’s limits.

Fig 1 below shows the typical loss of capacity with time. This is an example only and should not be used to calculate the state of charge for any specific battery type. It can be seen that the capacity loss is more pronounced in the first months of storage before a more gradual fall off occurs.

The correct storage temperature is arguably the most important aspect to consider. The higher the temperature and the quicker the capacity will fall and the shorter the recommended storage time.

Typically, manufacturers consider a storage temperature of 20ºC when quoting storage times. It is unwise to store a battery at above 40ºC for two main reasons: a) the storage time will be very short and b) at this temperature aging of the plastic components is accelerated. Whilst low temperatures will reduce the open circuit losses and give an extended storage time, it is normally recommended not to go below +5ºC. However, storage temperatures as low as -40ºC, and even below this for some types, are possible without problems providing the product is allowed to “thaw out” over several days until a more reasonable temperature above +5ºC is achieved before charging the battery. Very low temperatures will make plastic containers more brittle and special conditions may apply if the product is transported at these extreme temperatures.

In Fig 2 below the storage time for different storage temperatures can be found. It can be seen that if the storage temperature is 20ºC, the storage time is shown as 100%. If the storage temperature is 30ºC, the graph shows 50% and if the manufacturer quotes a storage time of 18 months at 20°C the storage time at 30ºC becomes 9 months.. Similarly, if the storage temperature is 10ºC, the storage time will be 200%, i.e. 36 months. This graph is in the most part correct but it would be unwise to consider a storage time of 400% of the nominal for the reasons illustrated in the above section, “EFFECTS OF TIME”. Similarly, the maximum storage temperature should be considered as 40 ºC for the reasons given in this section above.

Storage humidity can have detrimental effects on the product. Terminal pillars are typically manufactured from lead alloys or brass and both will oxidize over time but this is accelerated in high humid environments. This corrosion should be removed prior to making the electrical connections. Failure to do this may lead to a high resistance joint and could, in extreme circumstances, lead to dramatic failures. The author of this document has firsthand experience of a battery fire developing as the direct result of severely oxidized terminals of a Planté battery where the terminals were not cleaned prior to installation. The storage time was less than 4 months but the humidity was extremely high.

Internal corrosion is completely unseen in VRLA product but can be considerable. In vented product assembled into transparent containers, “flaking” of the positive group bars and pillars inside the cells can become extensive. This may be severe enough to cause early battery failure. The “flakes” are not normally seen in storage but develop rapidly during the commissioning charge process. Typically, this is not noticed until the battery is several years old when failure due to internal shorts develops. To the experienced battery technician this type of failure can be clearly identified as a storage issue and not one of manufacture, over charging or undercharging. The structure of this corrosion is quite unique depending on how it was formed.

Corrosion to the plates is more subtle. Corrosion to the “weld” area between plates and group bars and group bars to pillars are the first to be affected. In some grid designs, the nodes between horizontal and vertical wires are affected. It is stressed that this corrosion is extremely rare and is insignificant if the product is stored correctly.

When batteries are discharged the lead oxide (PbO2) of the positive plates and lead (Pb) of the negative plates are converted to lead sulphate (PbSO4). The extent that this occurs depends on the depth of discharge. The more rapid the discharge and the faster sulphate will be formed. When batteries are stored the self discharge is low but may be significant over time and this will be higher at elevated temperatures. Ultimately, lead sulphate will be formed and if the state of charge is very low, the specific gravity of the electrolyte will also be low and may be as low as 1.050 s.g. Under these circumstances, the lead sulphate (PbSO4) will react with the electrolyte (H2SO4) and some lead may go into solution. When the battery is eventually charged following storage, lead dendrites may be formed resulting in shorts between positive and negative plates.

Lead dendrites are rare because manufacturers often include chemicals to reduce their formation. Also, these lead dendrites only occur in VRLA AGM cells. Lead dendrites are not seen in vented products and VRLA GEL cells.

Battery containers can be affected by UV light. The effect is commonly known as UV degradation and sunlight contains a significant amount of UV light. The result of UV light can often be seen as cracking of the containers. Occasionally crazing of the container corners can be seen and a loss of colour definition can occur. Significantly many manufacturers include additive in the manufacture of containers to make them UV stable. However, this is very difficult where transparent containers are used such as with Planté and other vented products where clear containers are used.

It should be remembered that UV will affect the product once installed and direct sunlight on the battery should be eliminated.

Within reasonable limits, there is a direct relationship between the open circuit voltage of a lead acid cell and the specific gravity of the electrolyte. For this relationship to be accurate, the cell must have been on open circuit for at least 12 hours and preferably 24 hours. It therefore follows that it will be reasonably accurate for product that has been in store for several days or more. The temperature limits for this technique are reasonably wide but for best accuracy this should be taken as +10°C to +30°C.

If we know the fully charged and fully discharged specific gravity of the cell then we can calculate the state of charge. This is illustrated below in Fig 3 for a typical VRLA AGM product.

The mathematical relationship is: - OCV – 0.84 = the s.g. of the electrolyte.

Example 1: OCV = 2.15V S.G. = 2.15 – 0.84 = 1.31 s.g..

Example 2: OCV = 1.95V S.G. = 1.95 – 0.84 = 1.11 s.g.

Note: Even at an open circuit of 2.00V, the cell will only be about 25% charged or, more to the point, 75% discharged.

A typical average rate of losses is 2% per month which equates to 3 years for a 75% discharged cell. Recovery from this very low state of charge is unlikely without a considerable capacity loss. A reasonable minimum open circuit voltage is typically about 2.10V for VRLA AGM cells. The product should be recharged if this voltage is reached because anything lower than this is likely to result in a permanent capacity loss. The 2.10Vpc represents about 75% charged.

The mathematical formula is relevant to all lead acid cells, not only VRLA types. However, the fully charged and fully discharged s.g. will be different for different battery types; particularly vented types compared with VRLA.

The fully charged specific gravity of Planté cells is typically in the order of 1.215, for Vented Pasted Plate types a value of about 1.250 is typical and for Vented Tubular Plate about 1.280. For vented cells it is more accurate to use a hydrometer or density meter to measure the specific gravity and from this the state of charge can be determined.

To make best use of this technique it is important that the fully charged and fully discharged specific gravity is known and at what reference temperature. For VRLA AGM cells where the voltage is measured and cross referenced to the state of charge the technique is reasonably accurate. This is because there is a large difference between the fully charged and fully discharged s.g.; typically 1.310 when fully charged to 1.110 when fully discharged.

The storage conditions can interact and cause serious problems. In particular, whilst the store time may be extended at lower temperatures, this time frame must not be extended beyond the recommended limit as advised by the manufacturer.

Where high storage temperatures such as above 30°C are experienced, consideration to the humidity is very important. High temperatures with high humidity will cause rapid corrosion to terminals and if this is not addressed, high resistance electrical connections may result.

In conclusion, always follow the manufactures instructions.

This article discusses charging of lead-acid industrial standby batteries on float charge and “off mains” locations. It does not include charging of batteries on full cycling applications such as daily traction duties or similar.

Charging of parallel battery strings, inrush currents, ripple voltage and ripple current are also discussed in outline.

The majority of industrial standby batteries are charged using a Full Float Charging system but Pulse Charging is becoming popular particularly for UPS systems. In the main, the chargers are of the thyristor or switch mode type. Both types have some advantages over the other and for a detailed evaluation of each type it is advisable to contact the charger manufacturer direct and discuss the options.

In Full Float Charging it is intended to continuously charge the battery 24 hours a day, 365 days a year and the only exception will be when a mains power failure occurs or during testing and servicing.

Generally, battery manufacturers prefer this method of charging because it is fully automatic in providing the current that the battery requires, irrespective of the state of charge. The battery and charger are in parallel. This system is a full “no break” system. See the Diagram 1 below.

Diagram 1.

When voltage is applied to the battery terminals, current will flow. It has to be remembered that voltage does not charge the battery; it is the resultant current that will maintain the battery in a constant state of charge or recharge following a discharge. The value of the current going into the battery will depend on its state of charge, battery temperature and type of battery. Typically, but depending on the battery type; the applied voltage will be between 2.23Vpc and 2.28Vpc for a battery temperature of 20oC. Different applied voltages are used for different battery temperatures. The maximum resultant current will depend on the charger current output rating. When the battery is fully charged, the current will be very low and may be as small as 10mA per 100Ah of battery capacity. However, International Standards give guide lines of 1mA per Ah to 8mA per Ah depending on battery type. These values are published as being the typical maximum and have a large “safety margin” in them. Never the less, the current will be very small and in most cases, it cannot be measure without the correct current measuring instruments. A simple “clamp on” ammeter is unlikely to measure the current or the displayed value may be widely inaccurate. It is impossible to accurately measure the current in a fully charged battery with a 200A or even a 20A “clamp on” ammeter when the actual current in a 100Ah battery will normally be in the order of 25mA.

Chargers are normally current rated to supply the full load plus the minimum required for the battery which is typically between 5% and 10% C10 amperes depending on the battery type and manufacturers recommendations. If the system is for a short standby time an interesting scenario may arise. For example, if the load is 250A but is only required for 5 minutes, then the battery capacity will be low and in the order of 75Ah depending on the operating temperature and minimum voltage. If we consider a recharge current of 10% C10A i.e. 7.5A for the battery then the charger will need to have a current rating of about 250A to cater for the standing load of 250A plus 7.5A to recharge the battery, i.e. 257.5A. If a power failure occurs the battery will discharge into the load at the 250A required until the battery is exhausted, or until the mains power is restored, or until the load is disconnected. Either way, the battery will not be fully charged when the charging source is restored. If the load is disconnected from the battery for the recharge, the full charger current capability will be available to recharge the battery.

Under these circumstances the initial recharge current flowing in the battery will be considerable but because the voltage is limited this is not normally an issue. The back EMF of the battery being charged will rise very fast and will meet the float voltage quickly. In some instances, the time to the float voltage may be only a few seconds. However, there is the possibility for this current to flow for many seconds or even longer and there is a danger that any fuses or OCB protection in the battery circuit may be affected. For this reason, battery chargers are often designed in two sections, one for the load and one for battery recharging. Batteries themselves are rarely affected by this high initial recharge current and the effects can largely be ignored for Full Float Charge systems. To avoid any problems with fuses or OCB protection, this is often omitted between the battery and charger.

For float recharge systems, the time to fully charged does not change significantly for recharge currents between 3% and 10% and even at higher currents the time to fully charged still remains more or less the same at typically 3 days. However, the time to a state of charge of about 80% or less does change. The value of 80% is typically the state of charge when the float voltage is reached and this is more or less irrespective of the current available to the battery. This is illustrated in Fig 1 below for recharge currents between 3% and 10% C10A. Up to an 80% state of charge the Ah efficiency is close to 100%. Beyond this point, the efficiency gradually falls away until the battery is fully charged when the efficiency can be said to be 0%. Once the battery is fully charged the energy is lost in heat, battery corrosion electrolyte losses etc.

Fig 1.

The relationship between recharge current and voltage is illustrated in Fig 2 below.

It can be seen that the charging current falls rapidly once the float voltage is reached and in this case this is after approximately 11 hours of charging. Referring to Fig 1, it can be seen that a charge condition of 80% is achieved after approximately 11 hours when using a current limit of 7% C10A.

Fig 2.

Pulse Charging can be in several forms but as the term suggests, the charge is delivered in a series of pulses.

Pulses of charge may be delivered to the battery at time intervals dependant on the charger settings. The interval between charging may be as short as 1 second or many minutes. Some systems charge for 1 second, then rest for 1 second and this is continuous. Where the pulse intervals are short, the control voltage will be close to the normal Full Float charge value. Some systems charge at the Full Float voltage until the current has fallen to a pre-determined level or for a set time at which point the battery is put onto open circuit. Over several hours the voltage will gradually fall and at a pre-determined value charging recommences. Some systems charge a higher voltage such as 2.35Vpc before returning to the normal 2.23Vpc to 2.28Vpc value. For these systems, the battery is not connected directly to the load but only to the charger. A fast acting static switch is employed to reconnect the battery to the load in the event of a mains power failure. This system is illustrated in Diagram 2 below in outline for a UPS system.

Diagram 2.

This method of charging is normally reserved for commission charging vented batteries such as Planté, Tubular and Pasted plate types. Constant Current Charging should not be used with VRLA cells without specialist information from the battery manufacturer.

The best way to charge by constant current is to have a charger where both the voltage and current are adjustable. Initially, both voltage and current control should be set to a minimum. The voltage and current should be adjusted slowly until current starts to flow. Typically, once the current has started to flow, the voltage should be adjusted to the maximum available for the charger and the current control adjusted to the correct value. During this process, the actual voltage seen across the battery will not change because the current controller limits the current flowing and hence the voltage across the battery. At this point the current can be adjusted to the required value. Charging will then continue at a constant current and as the battery reaches about 80% charged, the voltage will be seen to increase. However, if the maximum voltage of the charger cannot be adjusted to about 3.00Vpc, then the current may not be constant for the complete charging process. For example, if the maximum voltage of the charger is 2.50Vpc, the current will remain constant until this voltage is reached and then the current will taper off and the voltage will remain steady at 2.50Vpc: this is not constant current charging.

Figure 3 below shows the typical voltage characteristics when charging a vented cell from fully discharged at a constant current of 10%

More information on charging with a voltage limit of 2.50Vpc can be seen below in the section titled Limited Voltage Charging.

Fig 3.

Limited voltage charging is similar to Full Float Charging but instead of using the float voltage which is typically in the order of 2.23Vpc to 2.28Vpc, a higher voltage is used. This type of charging system is often found where the voltage of the charger has an upper limit which is below the natural cell voltage maximum which is in the order of 2.80Vpc to 3.00Vpc. Voltage limits above 2.40Vpc should only be used under guidance from the battery manufacturer.

Limits to the maximum charger voltage are often imposed to protect the load from a voltage that may cause damage. A voltage of 2.40Vpc is often referred to as an equalizing charge. When fully charged, the current flowing in a VRLA cell at 2.40Vpc is typically 8 times that flowing at the float voltage of 2.23Vpc to 2.28Vpc, i.e., about 8mA per Ah. For Vented products such as Planté, Tubular or Pasted Plate, the current will be typically 20mA per Ah.

The higher the voltage limit, the shorter the recharge time will be. However, once the cell voltage reaches the voltage limit of the charger, the current diminishes at a rapid rate. This is illustrated below in Fig 4.

Fig 4

This charging regime is often misinterpreted as constant current because, in some cases the current is seen to be constant for some time before the voltage limit is reached and because it is not a float charge mode. Frequently, chargers have two settings “float” and “equalising”.

The time to fully charged from fully discharged is shorter than using the Full Float method but is considerably longer than using Constant Current. Some manufacturers will not accept this method for commission charging vented products such as Planté, Tubular and Pasted Plate types.

Table 1 below gives an indication of the time taken to reach the voltage limit and the time to fully charged. The current available is taken as 7% C10A.

Table 1

Off Mains Charging is often used in rural locations where no a.c. mains supply is available. Examples are country houses and telecommunication systems in remote locations. In the vast majority of applications a generator is provided but to avoid having to run this continuously a battery is used. The battery may be charged by a wind generator or solar system, or both, in addition to a generator. In these applications, the battery will at some time need to be brought back to a near fully charged condition if a satisfactory life is to be obtained. Continual undercharging will result is sulphation of the battery plates and premature failure. For systems with wind or solar power, a full recharge may only be required after many months and in some cases this may extend to a full year. In all cases, a full recharge once per year is normally recommended by the battery manufacturer. Typically a constant current recharge is used but a recharge at a value of about 2.50Vpc may prove sufficient.

For county houses the system usually consists of a d.c. generator and battery only. The d.c. power is supplied through an inverter to the normal 240V a.c. house load. To avoid unnecessary noise, the generator may be run through the day and turned off at night. Some systems may run on the battery for several days before recharging is necessary. Again, a full recharge is advisable at least once a year.

To reduce the battery size and if no other charging source is available other than by generator, an automatic system may be used. Country houses often use this system. To obtain a satisfactory life the battery may be charged daily up to a voltage of typically 2.35Vpc to 2.45Vpc and then reduced to a more value typical of a full float system. In this way, the battery will receive a good charge without being overcharged.

It may be argued that a full float system may be more appropriate to avoid overcharging but there is a real risk of undercharging if a value in the order of 2.23Vpc to 2.28Vpc is used. The Full Float Charging is intended for batteries which will only be discharged rarely and typically no more than 4 times a year. The Full Float Charger will not adequately recharge a battery on an Off Mains installation which is subject to regular discharges.

Providing all the strings in the battery are in good condition, no issues should be experienced. However, charging of parallel strings can present some interesting characteristics. For example, as on discharge, the recharge current in each string may be marginally different and providing they are within 10% of each other it is reasonable to consider as normal. However, if the currents are more than 10% difference it would be prudent to investigate. The cause may be they are a different state of charge and then the question arises – why? If the battery is new, the differences will stabilize after a good charge.

Occasionally, battery systems are made up on multiple strings but test discharged as individual strings. In these circumstances, it is important to recharge each string independently following a discharge test and before reconnecting to the other string or strings. If this is not carried out and the discharged battery is connected to the fully charged strings, two unwanted characteristics will occur. First, there will be a very large inrush current into the discharged string from the charged string(s). Protection may be affected. Second, considering a two string system, the second string will not be fully charged because of the energy transferred to the first string. Testing of the second string will not be correct because it will not be fully charged unless it is allowed to remain on float charge for several days. The system is further complicated when more than two strings are involved.

The inrush current into a fully discharged battery can be very high and in the order of 3 or 4 times the Ah capacity of the battery. This is not normally a problem because the back EMF of the battery will rise very quickly, within seconds in some cases, to meet the applied charging voltage and the current will rapidly fall away.

Under normal conditions, the charging source will have a current limit and the maximum inrush current will be the maximum of the charging source.

#RIPPLE CURRENT AND RIPPLE VOLTAGE

International standards discuss this but in all cases; the manufacturer’s instructions should be followed.

Excessive ripple voltage will cause a higher than desirable ripple current and batteries have been known to fail after a very short time when the charging source has excessive ripple.

Corrosion due to cycling by the ripple or heat will reduce the battery life.

In Planté batteries, cases have been recorded where catastrophic failure has occurred within a few hours. Rapid corrosion of the positive plates takes place which is normally seen as a white deposit in the base of the cells. This is easily identified because the containers are transparent. In VRLA cells with opaque containers, it is not possible to identify the problem until it is too late. VRLA cells are affected very differently to Planté or other vented products. Because the state of charge of the negative plates in a VRLA cell are inevitable in a less than 100% state of charge, it is these where the problem is seen. Corrosion of the negative grid wires takes place and the state of charge is very low. Ripple will cause heating and the life may be reduced without abnormal corrosion characteristics. However, the occurrence of “a.c” corrosion is rare in VRLA cells and the effects are not fully proven.

At first, the charging of industrial standby batteries seems simple but to obtain a satisfactory life and performance it should not be taken lightly.

The manufacturer’s instructions should be followed without deviation and any suggestion of using an alternative method of charging must be approved by the manufacturer.

Manufacturers are increasingly concerned about the effects of pulse charging which is being promoted as an energy saving feature. However, there is no evidence that pulse charging does save energy and it has been argued that pulsing can cause additional heat within the cells and reduce the life. Long term “field testing” continues.