Many, if not most storage batteries receive the bare minimum of attention until one day they are called into service and are expected to provide a dependable source of power for a pre-planned period. But how can users be sure that their batteries will be equal to the demands placed upon them? The answer is that only regular testing can provide this level of assurance.
The next question is the type of testing to use. Let’s start by making it clear that the only way to reveal the full information about the condition of a battery is by load testing. Essentially, this involves discharging the battery into a dummy load and recording its performance.
It is usually recommended that the condition of important battery installations is thoroughly evaluated by performing a discharge test at least once a year. Specially designed test sets, such as those in the Programma Torkel range from Megger, are available to make this process as easy and convenient as possible.
For complete discharge testing, the battery must be taken out of service for the duration of the test. This can easily be as long as two days to allow a full discharge/charge cycle to be completed. Clearly this test method, while offering the most accurate results possible, is costly, time consuming and often inconvenient. How many users, for example, are happy to be without standby power for two full days, or to go the cost and inconvenience of having an alternative battery available to provide cover while the testing is carried out?
One way of addressing this problem is to carry out limited discharge testing. With Torkel test sets, for example, testing can be carried out by discharging the batteries by up to 80% without taking them out of service.
This is a useful solution, but it is not perfect. The test is still time consuming and, although the batteries remain in service, should they be called upon to supply power at the point of deepest discharge during the test, they will only have 20% of their full capacity available.
For these reasons, many alternative methods of attempting to assess battery conditions have been proposed.
One of the simplest, or so it seems, is to measure the battery’s float voltage – the voltage at its terminals when it’s on charge. This is undeniably an important parameter, and if it is abnormal it indicates that further investigation is needed. If, however, the float voltage is normal, this gives no indication whatsoever about the condition of the battery.
Another approach is to measure the specific gravity of the electrolyte, particularly in lead acid batteries where sulphate is part of the electrochemical reaction. When the battery is discharged, some of the sulphate migrates to the plates and the specific gravity of the electrolyte falls. Conversely, when the battery is fully charged, all of the sulphate is in the electrolyte and the specific gravity rises.
Measuring specific gravity therefore gives an indication of the battery’s state of charge, but it has never been possible to show that there is a link between specific gravity and battery capacity.
In fact, the only approach which has been demonstrated as providing information about battery capacity reliably and without taking the battery out of service is impedance testing. The impedance of a cell is, in fact, a measure of its ability to deliver current and it has been shown that impedance correlates with battery capacity.
The correlation is not 100% – as stated earlier the only 100% certain way of determining battery performance is by load testing. Nevertheless, impedance testing is an excellent way of locating weak batteries in a bank.
This method of testing is used in Megger’s BITE 3 range of battery test sets. It works by applying an AC voltage to the battery and measuring the resulting current flow. Since the applied voltage is known, the impedance of the battery can be calculated using Ohm’s law. Extensive research has confirmed that this impedance is, to a good approximation, inversely proportional to the capacity of the battery.
The benefits of impedance testing are easy to see. It is completely non-invasive, and it is fast. It takes only around 30 minutes, for example, to test a typical substation battery bank. In addition, the charge held by the battery is not affected at any stage of the test and, as already mentioned, there is no need to take the battery off line.
All of this having been said, it’s still important to remember that impedance testing is an indirect measure of battery capacity. While valuable, it can, therefore, never offer the same degree of accuracy and certainty as a full load cycle test.
What then, is the best way of monitoring the condition of important battery banks? Megger suggests using both load cycling and impedance testing.
Full load cycle testing carried out once per year will yield accurate information and provide an excellent baseline for impedance testing. Impedance testing carried out on a much more regular basis will provide reassurance throughout the year that the condition of the battery has not deteriorated significantly.
Of course, loss of capacity is not the only ill that can befall a battery bank. It has been stated, for example, that loose inter-cell connectors cause around 50% of the failures in battery banks.
The connectors loosen because of the heating and cooling that takes place during charging and discharging – the cell terminal posts expand and contract and, because the lead from which they are made is very malleable, they cold flow with every cycle. Fortunately, problems of this kind can be detected relatively easily by testing with a low-resistance ohmmeter, such as Megger’s DLRO.
Another common problem is earth faults, which are often caused by water ingress on large distributed systems. In large batteries, these faults can be notoriously difficult to find using conventional methods based on resistance measurements.
A better solution, and one that can be used whether the battery is on-line or off-line, is to inject an AC signal into the battery system. The amplitude of the resulting signal in the battery feeder cables can then be easily measured with a clamp-type probe. Since the signal amplitude is inversely proportional to the fault impedance on the feeder, the location of the fault can readily determined.
The Megger BGFT battery ground fault test set uses this approach, and provides a fast convenient and reliable method of detecting and locating faults of up to 100 kohms.
Other factors relating to battery operation deserving of careful attention include float current, ripple voltage and operating temperature. Float current is the current delivered by the charger when the battery is in its fully charged state. Normally this current is small, but if it starts to increase for any reason, the temperature of the battery will rise.
The increase in temperature allows more current to flow, which further increases the battery temperature, allowing even more current to flow. The result is thermal runaway and, in extreme cases, this has been known to lead to battery meltdown, particularly in VRLA batteries, which have no free electrolyte that can evaporate to help keep the batteries cool.Fortunately, the onset of thermal runaway is usually relatively slow – of the order of months – so regular monitoring of float current can prevent problems of this kind.
A high ripple current is often indicative of a fault in the battery’s charging circuitry – possibly the failure of a diode in a bridge rectifier.
Excess ripple current again increases the heating of the battery, thereby shortening its life. Regular checking is recommended and the better battery testers, such as Megger’s BITE products, incorporate facilities for this.
Finally, regular measurement of the battery’s operating temperature is invaluable, since high temperatures invariably lead to premature failure. As a rule of thumb, battery life is halved for every 10ºC increase in temperature. This means that a battery with a rated life of 20 years, which is operated at 30ºC rather than 20ºC, will only have a life of 10 years.
Throughout this article, particular emphasis has been given to the desirability of regular testing. The reason is simple. One test in isolation may well provide useful information, but a series of tests carried out over a period of time will not only provide much more information, but will also make the information vastly easier to interpret.
A sudden change in a test result that has previously remained almost constant, for example, immediately suggests that further investigation is desirable, even if both the old and new values in isolation would be considered as falling within the acceptable range.
Specialised software is available from suppliers like Megger, which makes the storage, trending and analysis of battery test data straightforward tasks.
At first sight, monitoring the condition of storage batteries may appear to be a complicated undertaking.
As this article has shown, however, if the testing regime is properly planned and split into simple tasks that can readily be carried out with modern instruments, the overall effort and inconvenience is small.
This is particularly true when a comparison is made with the inconvenience and costs – direct and consequential – of a battery failure. Megger’s advice on batteries is, therefore, straightforward – keep an eye on your batteries with regular testing…
Damon Mount
Sales Manager, Megger UK
If you want to share thoughts or feedback then please leave a comment below.
