The oldest and most common type of rechargeable battery is the lead-acid battery, found in applications from portable electrical and electronic equipment to heavy goods vehicles. Lead-acid batteries are used as either principal or reserve power sources and their reliability in the field is often critical. Because of their low cost they are likely to be in service for many years to come, despite recent progress in alternative technologies, but this type of battery does have some technical problems. Giving the user information regarding the health and the charge state of the battery is a critical issue, and presently there is no simple method of reliably monitoring either. Present methods are either labour or time intensive involving close, manual monitoring or automated testing during which time the battery is removed from active use. For a number of applications where reliability is important, batteries are disposed of rather than face a risk of failure during operation.

The usual parameters targeted to assess health and charge state are cell voltage and the specific gravity of the electrolyte. For example, DE 2 462 039 describes the use of an electrostatic capacitor to derive the specific gravity of a battery electrolyte relative to a standard electrolyte. The measurement of specific gravity gives an indication of the state of charge of a battery, but it is not suitable for all types of lead acid battery. For example, it is not conveniently measured for gelled or starved electrolyte cells. The conductivity of the electrolyte may also be measured, however this is unreliable as during charge and discharge the conductivity of the electrolyte changes such that it has a maximum value at an intermediate charge and a lower value at both high and low charge. This leads to a situation whereby a measurement of electrolyte conductivity cannot distinguish between states of charge.

Furthermore, it is not simple to measure the specific gravity of the electrolyte for sealed lead acid batteries. Other routines measure the battery impedance to an AC voltage. Yet further techniques use pulse-loading techniques to work the battery. The battery's health, i. e. the condition of the electrodes is determined from the response to this simulated pulse load. All battery health monitoring systems require some level of interpretation and user interaction. There are other drawbacks in existing systems, such as the need, in flooded electrolyte type cells, to manually dip a hygrometer in the electrolyte to measure its density and need for complex impedance and discharge

equipment to automatically perform an equivalence test. Following testing batteries can be serviced (e. g. by the addition of water) leading to prolonged operational lifetimes. However, even properly maintained batteries often give poor performance, failing unexpectedly.

In accordance with the present invention, an ionic concentration monitor comprises a pair of electrodes, a signal generator; processing means; and indication means; wherein in use the electrodes are inserted into a medium under test; and the capacitance between the electrodes is derived by the processing means from the frequency output by the signal generator, such that an indication of the ionic concentration of the medium under test is output to the indication means.

The present invention provides a reliable monitoring system which enables the charge state of the battery to be established more quickly, cheaply and more conveniently than in conventional systems. The electrodes form an electrochemical capacitor or super-capacitor. The mechanism for energy storage in such a capacitor is based upon the separation and accumulation of charged ions at the interface between an electrically conducting electrolyte and electrode. As the capacitor is charged, ions migrate through the electrolyte and accumulate at the electrodes forming an electrochemical double layer. The quantity of ions on the electrodes reflects the charge stored. The charge stored is thus proportional to the ionic concentration of the electrolyte. The energy stored in the capacitor, Ei, is given by; Ej =6rcoFV, where a represents the ions removed from the electrolyte, co is the initial ionic concentration, V is the applied potential and F is Faraday's constant. The capacitance is thus directly related to the ionic concentration. In contrast, prior art capacitative sensors, such as those described in DE 2 462 039, GB 2 222 683 A and WO 92/07251, employ electrostatic capacitors. These consist of plates separated by a dielectric, which does not conduct electricity. GB 2 222 683 A and WO 92/07251 use such capacitors in order to measure the complex dielectric constant of grain and soil respectively, the dielectric constant being related to the moisture content of the medium under test. An electrostatic capacitor cannot be used to measure ionic concentration. Furthermore, the charge density obtainable using a super-capacitor is far greater than that which can be obtained in an electrostatic capacitor. This allows the device of the present invention to be made physically small.

Any type of signal generator may be used, but preferably the signal generator is chosen from one with a continuous sinusoidal output; or an integrated circuit with a pulsed output.

Preferably, the indication means comprises one of a series of LED's; a liquid crystal display; or an output to a computer system.

Preferably, the monitor further comprises a temperature sensor.

Typically, the monitor comprises a battery charge indicator.

Preferably, the indicator further comprises a voltmeter, such that an indication of battery health may be derived.

The performance of lead acid batteries deteriorates over time due to the reaction between the lead plate electrodes and the sulphuric acid electrolyte. This forms an inactive skin of lead sulphate on the electrodes. By determining both the voltage of the battery, and the composition of the electrolyte in terms of its ionic concentration the present invention is able to provide a reliable indication of the overall health of the battery.

An example of an ionic concentration monitor in accordance with the present invention will now be described with reference to the accompanying drawings in which: Figure 1 is a block diagram of an example of a monitor according to the present invention; Figure 2 illustrates one example of a signal generator for use in the monitor of Fig. 1; Figure 3 is an illustration of the monitor of Fig. 1 in use in a battery; Figure 4 illustrates an alternative arrangement for a signal generator for use in the monitor of Fig. 1; and, Figure 5 is a graph illustrating the effect of change of sulphuric acid concentration on capacitance for a lead acid battery using the monitor of Fig. 1.

A monitor 1 according to the present invention comprises a pair of conducting electrodes 2,3 which form part of a signal generator 4, in this case an oscillator with a sinusoidal output. The signal output from the sinusoidal oscillator is processed by a processor 5 and output to a display 6 to give an indication of the state of charge of the battery. The monitor 1 has its own power supply (not shown) which can be a direct DC supply or an AC supply, which has been converted to DC. Although the example is described with reference to a battery, the same monitor can be used to monitor the ionic concentration of other liquids such as for process control in chemical engineering plants or quality control for desalination plants.

The electrolyte in a lead acid battery is sulphuric acid. During discharging the concentration of sulphuric acid changes according to the overall reaction: Pb + Pb02 + 2H2SO4--+ 2PbSO4 + 2H20 Lead sulphate plates out onto the electrodes and on recharge, Pb and Pb02 is formed. The density of the electrolyte also decreases during discharging, and is related to the state of charge. The decrease in the concentration of sulphuric acid will have an effect on the interface between the electrolyte and any electrode. This interface consists of a charged region, or double layer, and has a specific capacitance. The capacitance is also related to the surface area and type of the electrodes. During charge and discharge the number of ions present at the surface of the electrodes changes and so the capacitance between the electrodes changes.

The monitor could make use of the battery electrodes, but this is more complex and would not be possible whilst the battery was being charged. Furthermore, when the technique is used for monitoring the ionic concentration of other species, for example in a chemical or water treatment plant, there may be no electrodes available to use.

In this example the signal generator 4 is an electronic oscillator, where the output is a sine wave. The frequency of oscillation is given by: where fosc is the frequency of oscillation, L is the circuit inductance and C is the circuit capacitance (i. e. that between the electrodes 2,3, depending on the battery electrolyte). As the capacitance, or inductance, of the circuit is varied then the oscillation frequency changes. The oscillator converts a steady input into an oscillating input, the only direct input being the DC supply and the output being a changing signal. In Fig. 2 a typical circuit diagram for such a device is shown. The frequency of the sinusoidal output is dependent on the inductor and variable capacitor shown in parallel.

In use, as shown in Fig 3, the electrodes 2,3 of the monitor 1 are inserted into an electrolyte 7 of a battery 8 which is being monitored and take no part in the in the charge or discharge reaction of the battery. The electrodes 2,3 are electrically insulated from battery electrodes 9,10 and the battery electrodes 9,10 are connected to equipment (not shown) for which they are supplying power. As the

battery discharges during use, the frequency of the oscillator changes and the processor 4 compares the frequency that is produced by the oscillator with data in a look up table in order to derive the capacitance for that battery electrolyte at that frequency. The look up table is specific to a particular design of battery and type of electrolyte. The display 6 then displays an indication of the charge of the battery. This can be a simple illumination of a green or red LED according to whether the result is satisfactory or not, or a text indication of the same. With a suitably programmed processor, the actual numerical concentration could be displayed and even instructions on how to correct any deficiencies.

Another method of timing involves generating a pulsed signal, which has a frequency which is also dependent on the value of a capacitor within the circuit. Non- sinusoidal oscillators rely on the charging and discharging of a capacitor through a resistance and may be known as relation oscillators, their waveforms often being a rectangular output. In Fig. 4 a common set-up for such an oscillator is shown, the main part of the circuit consisting of two resistors and a capacitor. When the capacitor is variable, the frequency of the rectangular pulsed output may be altered. This timing method may be achieved by using a conventional integrated circuit (IC), such as one from the 555 family of IC chips. The frequency, f, of the output is related to the capacitance, C, and the sum of two resistors used in the circuit, R1 and R2, and is: <BR> <BR> f = I<BR> 0.7C(RI+R2) It can be seen that as the capacitance increases then the output frequency decreases and so the concentration of the ionic species can be derived using a look up table as before. The invention is not limited to these types of signal generator, and others may equally be used.

The graph in Fig. 5 shows how the capacitance in Farads between the two monitor electrodes 2,3 varies as the sulphuric acid concentration (expressed in terms of moles/dm3) changes. Although the effect was based on platinum electrodes, the effect will be similar for other electrode materials. For use in a battery monitor the electrode material should be one that is not attacked by sulphuric acid. Suitable alternatives to platinum include silver, gold, titanium, tungsten, molybdenum and glassy or vitreous carbon. The choice of electrode material is clearly less constrained if the monitor is to be used in a less aggressive medium, for example stainless steel electrodes may be used in a desalination plant.

The monitor uses the electrochemical properties of the electrolyte to influence the oscillation of a electrical circuit providing an essentially non intrusive and reliable method of charge determination, either as a self contained unit or, in more complex systems, as one part of a test procedure. For example, measuring the state of health (or physical deterioration with time) of the battery by measuring the voltage and the temperature as well as the ionic concentration of the electrolyte as described above, then using the processor 4 to compare the results to an idealised table and indicate to the user whether the cell is healthy or unhealthy.

Other forms which the monitor could take are a dip cell which is inserted into the electrolyte to take a measurement and then removed, or the monitor could be built into new batteries and provided with an interface to a computer system for continuous or intermittent data logging, in addition to the basic in-situ display.

Alternatively, only the electrodes may be built into the battery and provided with external connections. This would allow a user to plug the rest of the monitor onto the battery, assess its state of health and move on to the next battery to be tested.

This would lead to time and cost savings.

In large or tall lead acid batteries, stratification of the electrolyte may occur.

This may cause the concentration of the electrolyte to be non-uniform throughout the battery. Clearly, in this situation a number of monitor electrode pairs could be placed at various points. This would ensure that the outputs of each were representative of the local environment, and combined would provide an accurate assessment of the overall state of health of the battery.