Selection Of Standby Batteries
For Outdoor Systems

Philip Shumard

Hawker Energy Products
617 N. Ridgeview Drive
Warrensburg, MO 64093
Phone: 660-429-7582, Fax: 660-429-1758
Email: philip.shumard@hawker.invensys.com 

Presented At WesconŽ / IC EXPO '97, Nov. 5, 1997

Introduction

Batteries have been used to provide standby power for many decades, especially in the telecommunications industry. Initially, the standby power was provided by rooms of flooded lead acid batteries. The fact that the batteries were in a centralized, stationary location meant that monitoring, maintenance, and replacement were relatively simple, and the operating environment could be controlled.

More recently, system demands combined with improvements in battery technology have resulted in increased use of decentralized systems. Decentralization, which typically means outdoor installations, makes monitoring, maintenance, and replacement more difficult and expensive. In addition, the environmental conditions are typically poorly controlled, if controlled at all. Considerations for selecting a battery in these decentralized installations are much different from considerations in climate controlled, centralized settings due to the fact that battery life and performance are very dependent on temperature and other related factors.

It is not uncommon for the battery to have the shortest life of any component in an electronic system. This means that the proper selection of a standby battery can have a direct impact on the maintenance interval and therefore the operating cost of the system. It also means that it is critical to have a well designed power system; including the battery, the charger, and any monitoring and thermal management. In order for outdoor electronic systems to have long, reliable life, with a minimum requirement for field service, it is necessary to select a battery with long, reliable life.

Pure Lead Tin Technology

Pure Lead Tin Technology (Pure Pb/Sn Technology) is uniquely suited to outdoor applications. The use of high purity lead, greater than 99.99% pure, results in low self discharge rates and low positive grid corrosion rates.

Positive grid corrosion is the typical failure mode of lead acid batteries in standby applications. As the grid, which is the primary current conductor through the electrode, corrodes, the internal resistance of the cell increases until the cell ultimately fails. Using traditional lead calcium (Pb/Ca) technologies, long battery life can only be achieved by utilizing very thick electrodes with thick grids. Using thick grids, the life is extended because more grid must corrode before the battery fails.

The rate of corrosion is related to the alloy of lead which is used. Calcium, antimony, and many other materials which are commonly used in lead acid cells greatly accelerate the rate of grid corrosion. Manufacturers use alloys such as these primarily because they add rigidity to the grids. Rigid grids are easy to handle through the manufacturing process.

Pure lead corrodes at a very slow rate compared to alloys of calcium and antimony. High purity lead also minimizes the amount of hydrogen and oxygen gas which are generated from the electrolysis of water in the cell, preventing dry out. Using pure lead, it is possible to manufacture batteries with very thin electrodes which still have a long life.

Thin electrodes result in improvements of active material utilization, especially when batteries are discharged at high rates. Using traditional alloys, it is possible to develop batteries with high power densities but short life, or batteries with long life but low power density. It is not possible to develop batteries which can deliver both. With pure lead, it is possible to have very high power densities combined with long life. The low rate of gas generation allows the battery to operate at very high temperatures without drying out.

A drawback to the use of pure lead alone in the grid is the fact that it has a limited and unpredictable cycle life. This is due to the fact that there is a tendency to develop a resistive layer at the grid to active material interface. This resistive layer can also cause the battery to have difficulty in recovering from deep discharges. The addition of a minimal amount of tin resolves these problems. Further, the addition of tin does not result in significant increases in gas generation or self discharge. It also has a minimal effect on the rate of grid corrosion so that thin electrodes can still be utilized.

Aside from long life combined with high power density, Pure Lead Tin Technology also results in benefits of reduced self discharge rates, a wider operating temperature range, superior low temperature performance, and the ability to fast recharge. These features match up well to the requirements of outdoor applications.

Constraints In Outdoor Applications

There are several constraints involved in the selection of batteries for outdoor or remote applications which do not exist for indoor systems. The engineer responsible for battery selection should understand these factors.

Temperature. In general, the most important external parameter which will effect the life and performance of a standby battery is temperature. High temperature will reduce battery life and low temperature will reduce battery capacity. Temperature will also effect how the battery will be charged.

graph1.gif (6560 bytes)
Graph 1: Low Temperature Capacity

Graph 1 shows the superior performance of Pure Pb/Sn Technology at low temperature. The slope of the curves represents the effect of temperature on capacity. At low temperatures, the advantage in energy density which Pure Pb/Sn Technology has over traditional Pb/Ca standby batteries is even further enhanced.

At high temperatures, batteries tend to be effected by both dry out and grid corrosion. Both of these factors affect capacity and the internal resistance of the battery. The fact that internal resistance is affected is a major factor because it means that a battery which might be able to deliver adequate capacity at a low discharge rate may not be able to deliver higher discharge rates.

Graph 2 shows a comparison of the float life at high temperature between Pure Pb/Sn Technology and traditional Pb/Ca technology. In this graph the discharge is conducted at a 5 hour rate which is relatively low. It can be seen that the degradation in performance of the Pure Pb/Sn Technology product is much slower than for traditional Pb/Ca standby product.

graph2.gif (6609 bytes)
Graph 2: Float Life At High Temperature

While Graph 2 shows the performance difference for low rate discharges, Graph 3 shows the change in internal resistance during the float life test. While the Pure Pb/Sn Technology product shows a slow rise in internal resistance over the battery life, the Pb/Ca product's internal resistance increases dramatically. This will prevent the Pb/Ca product from delivering high rate discharges or from delivering high rate pulses.

graph3.gif (6793 bytes)
Graph 3: Increase In Internal Resistance During High Temperature Float

Duty Requirements. The frequency which the battery will be called on to deliver a discharge, and the duration of the discharge, are critical factors in the selection of a standby battery. Traditionally, standby applications have used batteries which were designed to give maximum life under float charge conditions. Very little attention has been paid to the battery's ability to cycle, or deliver repeated discharges. In many industrialized locations this is acceptable because the battery would typically have several months between discharges. The battery may see less than a dozen discharges during its entire life. In this scenario, a battery which is designed specifically for float charge can give good life.

Increasingly, however, systems are being installed in developing countries and in remote areas. In many of these areas, there is a lack of reliable infrastructure to deliver line power. In such areas, a standby battery can be discharged quite frequently. In many cases, batteries designed to maximize float life are unable to deliver the cycle life needed. The result is that battery failures due to cycling occur.

For this reason, design engineers should be very concerned with the ability of the battery to survive significant cycling. With traditional standby batteries, this will result in a tradeoff. Traditional Pb/Ca batteries which are capable of delivering good cycle life tend to deliver unacceptably short life in standby applications.

Pure Pb/Sn Technology is uniquely suited to deliver both float life and cycle life. Pure lead technology, without tin, delivers outstanding standby life, but does not perform well if cycled frequently. Lead calcium batteries are also unable to deliver good cycle life without compromising standby life.

Graph 4 shows a comparison of cycling performance between pure lead and Pure Pb/Sn Technology. The batteries are on a float charge at the manufacturer's recommended float voltage. Every 21 hours, the batteries are discharged. This is an extreme cycling condition and since the batteries are being charged at a float voltage, even the Pure Pb/Sn batteries are losing capacity after 20 days, but the pure lead product failed very early. While a standby battery would not be subjected to this extreme level of cycling, it would not be uncommon for batteries to have relatively deep discharges several times per week in some installations.

graph4.gif (16042 bytes)
Graph 4: Pure lead versus lead-tin product undergoing daily cycle with float recharge

Total Battery Related Costs

The design engineer who is selecting a battery is ultimately looking for the lowest cost solution. Pure Pb/Sn Technology will seldom be the lowest priced, but it is important to look beyond purchase price when selecting a battery. Instead, the designer should look at the lifetime battery cost. Factors such as battery life, replacement costs, and the cost of battery failure must also be taken into account. The following equation gives the cost per year resulting from the battery system:

C = (P + R)/ Lbattery

where; C = battery related cost per year

    P = purchase price of the battery

    R = replacement costs (not including battery price)

    Lbattery = service life of battery

This equation looks at the relationship between purchase price, replacement cost, and battery life. The value of R, the replacement cost of the battery, can be very high for outdoor systems, frequently much higher than the purchase price of the battery. The systems may be separated by great distances, they may be located on poles, or they may be far from roads. These factors all add to the replacement cost.

Since the value of P, the purchase price of the battery, will typically be small compared to the replacement cost, R, using a battery with a lower purchase price will not reduce the total system cost. The best approach is to increase the battery life, Lbattery . This can best be accomplished by purchasing a durable battery. For most outdoor systems, the longer life of Pure Pb/Sn batteries is justified.

Graph 5 shows the relationship between battery price, replacement costs, and battery life. As the graph shows, as the cost of replacing a battery increases, it becomes very easy to justify the price premium of Pure Pb/Sn Technology.

graph5.gif (4699 bytes)
Graph 5: Relationship Between Battery Price, Replacement Cost, And Battery Life

Testing/Qualification Methods

While considerable information can be gleaned from the literature published by battery manufacturers, the ultimate decision concerning the selection of a battery should be based on product testing. Virtually every standby battery application is different in some respect. It is critical that the product testing should reflect the requirements and conditions of the intended application.

In order to make testing procedures match the application, the design engineer must tailor the testing to some degree. In this section, some of the basic test methods will be outlined. Using this information, combined with knowledge of the application, the test engineer can design an appropriate test plan to evaluate batteries for specific applications.

Capacity Testing. The most basic testing to be performed for an application is determination of runtime which the battery can deliver. This testing should be conducted at the actual rates required by the application. It is especially important to verify that the battery is able to perform at the maximum discharge rates required. The testing should also be conducted across the full temperature range of the application. The most critical test point would be the maximum discharge rate combined with the lowest temperature.

A common error during the process of selecting batteries to test is to screen batteries based on the labeled capacity. This capacity applies only to a single discharge rate and is not indicative of the capacity at other rates. Screening should be based on the battery's capacity only at the applicable discharge rates.

Float Life Testing. The classic test to determine life in standby applications is high temperature float testing. In this type of test the battery is charged, at a float voltage, at an elevated temperature. The battery is discharged periodically to determine when the battery fails.

This type of high temperature float testing is a valid test and should typically be part of the battery selection process. At the same time, the designer should understand the limitations of this type of test. High temperature is used in this type of test for 2 purposes. First, high temperature is recognized as a method of stressing the battery and accelerating failure. For this reason, life in test is used to predict life in the application. Second, it proves the ability of the battery to perform at the high temperature. The latter of these reasons is straight forward, but the former requires some discussion.

High temperature accelerates certain failure modes in the battery. The most important of these failure modes is positive grid corrosion.

For a given battery design, at the manufacturers recommended float voltage, it is possible to construct a graph of battery life as a function of battery temperature. The shape of this curve will vary somewhat from battery design to battery design and most battery manufacturers can provide specific data showing this relationship. In the absence of this information, a rough approximation is to assume that the life of the battery is reduced by half for each 8° to 10°C rise in temperature.

Using this information, it is possible to conduct a float test at high temperature, in the range of 50°C to 80°C, and based on the life achieved in the test predict the life at reduced temperatures. This method of predicting life can be very reliable as long as the failure mode in the test and the failure mode in the application are both positive grid corrosion.

The failure mode of the battery in the test is relatively easy to determine once the test is completed, either by the design engineer or by the battery manufacturer. It is more difficult to prove that the predominant failure mode in the application will be positive grid corrosion, and in many cases it will not be the failure mode.

Essentially, the high temperature float testing can provide an outside limit to the potential life of the battery. If no other factor causes the battery to fail, it will fail at the time predicted by this test. The most likely factor to cause the battery to fail earlier than predicted in the high temperature float test is the fact that the battery is frequently discharged, especially if the discharges are to a fairly deep depth of discharge.

Further testing beyond the high temperature float test are required for virtually all standby applications. Only in applications which are certain to have periods of several months between discharges would it be appropriate to predict life based solely on this test.

Cycle Life Testing. If the application will require frequent discharges, cycle testing should be conducted. The test should include the same charge/float voltages which will be used in the application. The depth and frequency of the discharges should be representative of the application. Most batteries will perform poorly when fully discharged daily and charged at float conditions. Batteries which cycle well under these conditions are very likely to have very short float life. A more reasonable test is to conduct discharges either less frequently or to a shallow depth of discharge. If there is a real possibility that daily full discharges will be required, the charger should accommodate this by providing a multistep charger which charges at a cyclic voltage and then switches to a float voltage. The testing should include the multistep charger.

Conclusion

The objective of the engineer responsible for selecting batteries for outdoor systems should be to minimize the cost of batteries over the life of the system. Due to the costs associated with battery replacement in decentralized outdoor sites, this objective is best achieved by selecting a battery which can deliver a long life. This is true even if there is a price premium for the longer life battery.

There are several factors which impact the life of batteries in these applications. Extreme temperature, both high and low, is an extremely important factor which effects battery life and performance. The depth and frequency of discharge are also important factors. It is critical that the battery which is selected for an application must be able to survive in its working environment.

Design engineers should ultimately make battery selection decisions based on testing. Testing should reflect the real life conditions which will exist in the application.

An ideal choice in many outdoor applications is Pure Pb/Sn Technology. This technology provides long life and high performance in extreme temperatures. It is able to deliver long cycle life when required. Using this technology, engineers can greatly reduce maintenance costs and maximize system reliability.

 

[whitepapers/_private/footer.htm]