GB2337337A

GB2337337A - Method and means for determining the polarity of a capacitor

Info

Publication number: GB2337337A
Application number: GB9810138A
Authority: GB
Inventors: Richard J Payman
Original assignee: IFR Ltd
Current assignee: Aeroflex Ltd
Priority date: 1998-05-12
Filing date: 1998-05-12
Publication date: 1999-11-17
Anticipated expiration: 2018-05-12
Also published as: GB9810138D0; GB2337337B

Abstract

A method and means for testing a capacitors polarity comprises an alternating current supply 4 connected to one terminal of a capacitor under test 2 and a reference voltage, which may be ground, connected to the other terminal and a non-contact probe 26 located adjacent to the capacitor 2. The non-contact probe 26 senses the electric field generated by the capacitor 2 and provides an output signal relative to an alternating bias voltage which is applied to the said probe 26. A processed output signal from the non-contact probe 26 is arranged to indicate the polarity of the capacitor 2. The probe 26 may be a sensor plate arrangement connected via a screened cable to a differential amplifier 24 which is supplied with a common mode reference signal to the inventing and non-inverting inputs.

Description

1 APPARATUS FOR AND METHOD OF DETERMrqNG THE POLARITY OF A CAPACITOR

2337337 The present invention relates to an apparatus for and method of determining the polarity of a capacitor.

It is desirable to correctly determine the polarity of an electrolytic capacitor. If such a capacitor is installed incorrectly such that it is reversed biased, the performance of the capacitor may be degraded, its operating life may be shortened and in some cases the capacitor may even explode. Electrolytic capacitors tend to be constructed such that one of the plates encircles the other plate. As a consequence, one of the plates is screened, at least partially, by the other. This fact has been recognised and used in prior art capacitor polainity testers.

US 4745359 applies anti phase alternating current signals to the terminals of a capacitor under test. A probe is placed adjacent the body of the capacitor and may even be a moulded shape that fits around the body of the capacitor. The received signal is phase coherently detected in order to determine which terminal of the capacitor is connected to the outermost electrode thereof. However this does require that both terminals of the capacitor can be driven.

US 5 159526 appears to use the same test procedure as US 4745359. An alternating voltage from a test signal generator is buffered and applied to a first terminal of the capacitor whilst an anti phase version of the signal is generated by an inverter and supplied to a second terminal of the capacitor. A test signal at a frequency of approximately 100 Hz is used. It is expected that the phase of a voltage signal induced in a sensor plate will be approximately 90 out of phase with the reference signal generated by the test generator due to the reactive nature of the capacitor. Thus the signal received by the probe is buffered, amplified and then coherently detected with respect to a reference signal derived from the signal generator and then phase shifted by 90' so as to compensate for the phase shift induced by the capacitor. Again this technique relies on the ability to access both terminals of the capacitor, for example via a bed of nails tester, and then to drive both terminals with anti phase signals.

2 US 5502375 relates to a capacitor polarity tester that grounds a first terminal of the capacitor and applies an alternating signal to a second terminal of the capacitor. A sense plate is placed adjacent the capacitor and the signal is measured using an AC voltmeter. The second terminal is then grounded and a signal applied to the first terminal and a second voltage measurement is made. The ratio of these measurements is used to determine the polarity of the capacitor. As with the previous methods, a signal must be applied to both terminals of the capacitor.

These methods generally require the amplitude of the applied signals to be small, say 0.2 volts or so, in order to ensure that the capacitor is not significantly reversed biased during the test. These tests also assume that the capacitor can be driven symmetrically or that the same magnitude of signal can be applied to each terminal. These assumptions may be flawed when the capacitor is to be tested in-situ. The capkitor may be associated with other components within a circuit. One terminal of the capacitor may be coupled to resistors, inductors or sen-dconductor devices whilst the other terminal may be grounded. The presence of these other components may make it difficult to drive a signal onto one of the terminals of the capacitor. These components may also distort, attenuate or phase shift the signal.

EP 0722 091 A discloses a test method in which one terminal of a capacitor is supplied with a drive signal whilst the other terminal is connected to a reference voltage. A probe is placed near the capacitor and the voltage induced thereon is amplified and then supplied to a threshold detector in order to determine if the capacitor is the correct way round. This approach recognises that it may not always be possible to apply a signal to both terminals of a capacitor as one of them may be grounded. However, it still has limitations.

None of the above methods has addressed the fact that lead resistance within the test head of a test station, although small, can nevertheless give rise to unpredictable phase shifting of the signal across the capacitor if the capacitor is large, such that its impedance at the test frequency is of the same order as the lead impedance. Furthermore such lead impedance also reduces the voltage appearing across the capacitor which leads to unreliable measurements when using threshold based techniques, especially in the presence of noise.

3 According to a first aspect of the present invention there is provided a polarity tester for testing the polarity of a capacitor, comprising: signal supply means for supplying an alternating signal to a first terminal of a capacitor under test via a first conductive path, a second terminal of the capacitor under test being connected to a reference voltage via a second conductive path; a probe for sensing an electric field generated by the capacitor under test; probe bias means for biasing the probe with an alternating bias voltage; and processing means for processing an output of the non-contacting probe with respect to the bias voltage to obtain an output signal indicative of the polarity of the capacitor under test.

It is thus possible to provide a capacitor polarity tester in which only one terminal of the capacitor is driven. This provides considerable advandges as it facilitates testing of components in situ where one terminal of the capacitor may be connected to ground.

The provision of an AC bias signal to the non-contacting probe means that, from the point of view of the probe, the terminals of the capacitor are driven with anti-phase signals, even though one of the terminals is in fact held at a steady voltage.

Preferably at least one reference voltage probe is provided to make voltage measurements at or electrically adjacent the terminals of the capacitor. These reference voltages may then be supplied to the probe bias means.

Alternatively, a reference voltage probe may be selectively connectable to one end of the capacitor at a time and a measurement made, and then the measurements may be compared.

Alternatively one or more of the measurements may be replaced by an expected value of that measurement.

Reference voltage probes may be connected to the opposing terminals of the capacitor and the voltages measured thereby (at least one of which is an alternating voltage) processed to form the 4 bias voltage.

In a preferred embodiment of the invention the probe for sensing the electric field comprises a pair of spaced apart plates. The plates are connected to respective inputs of a differential amplifier. Buffer amplifiers, configured in an inverting or non-inverting mode, may also be included.

Advantageously, first and second measurements may be made with a reference probe connected to the first and second terminals of the capacitor, respectively, and then the amplitude, magnitude or power of the signals may be compared either ratiometrically or by forming a difference. At least one of the measurements may also be compared with a threshold in order to provide a measure of confidence that the capacitor is in lace and/or that the probe and/or signal generator is functioning correctly.

1n another embodiment, a phase sensitive measurement of the signal received by the probe may be made. The measurement may be made by a phase meter and this measurement compared with an acceptable range of values. Alternatively a phase shifter may be programmed with a compensating phase shift in order that a simple measurement of sign of the detected signal, or its magnitudes may be made. The compensating phase shift may be learnt from a known good board or calculated.

According to a second aspect of the present invention, there is provided a method of testing the polarity of a capacitor, comprising the steps of supplying an alternating signal to a first terminal of a capacitor under test via a first conductive path, a second terminal of the capacitor under test being connected to a reference voltage forming a signal return via a second conductive path; placing a non-contacting probe proximate to the capacitor under test in order to obtain a measurement of the electric field surrounding the capacitor under test; applying an alternating bias voltage to the non-contacting probe in order that the said electric field measurement is relative to the said bias voltage; measuring the noncontacting probe output relative to the said bias voltage in order to remove the bias voltage component from the measured output; and using a property of one or more measured output values to determine the polarity of the capacitor under test.

It is thus possible to provide an apparatus for and method of testing the polarity of a capacitor.

The present invention will further be described, by way of example, with reference to the accompanying drawings, in which:

Figure 1 illustrates an equivalent circuit of a test arrangement for testing the polarity of a capacitor; Figures 2 and 3 plot the magnitudes of voltaies measured at various nodes of the circuit for different values of capacitor at a fixed frequency of 100 Hz, and for a 1 millifarad capacitor at various frequencies; and Figure 4 schematically illustrates an apparatus constituting an embodiment of the present invention.

Figure 1 illustrates the effective circuit of a capacitor test apparatus within a bed of nails tester. A signal generator is located within the bed of nails tester and can be attached to one or more of the nails via programmable switches. The tester will, in all probability also include an interface head. Although the wiring in such an arrangement is often supposed to be of a reasonable quality, there will nevertheless be some series resistance in the wiring between the signal generator and the device under test. This series resistance may become significant with respect to the impedance of the capacitor under test, especially if the capacitor has a high capacitance. Thus, in figure 1, the capacitor 2 under test is shown as being connected to a signal generator 4 via resistors RI and R2 which represent the resistance of the wiring within the tester. Commonly one side of the capacitor is connected to ground. This may have the effect that during the test only one terminal of the capacitor can be driven from the signal generator.

6 Figure 2 plots the voltage at various points within the circuit shown in figure 1 for a test frequency of 10011z and R I = R2 = 1.5 ohms. It has been assumed that the circuit is driven from a signal generator having no output impedance. This is of course impossible and will only result in further increases in the series resistance.

The series impedance of the circuit is complex and given by:

impedance = R1 + R2 + OwC) Thus, for capacitors having an capacitance of 1 microfarad, the impedance of the capacitor is large compared to the other components in the circuit, and consequently almost all of the applied signal appears across the terminals of the capacitor. However, as the capacitance increases past 100 microfarads, the impedance of the capacitor become comparable with that of the resistors R1 and R2. By the time the capacitor has a value of 1 millifarad, approximately 45% of the applied signal voltage appears across R2, as represented by the curve V(R2). Additionally a further 35% appears across RI, as represented by the curve V(R1). The voltage appearing across the capacitor is not equal to the applied voltage minus V(R1) minus V(R2), as the capacitor has a reactive impedance. The voltage appearing across the capacitor is therefor phase shifted with respect to the signal supplied by the signal generator, and is shown by the curve Vc. It can be observed that the value of Vc can remain significant, say 20% of the applied signal even when each of the resistors has little less than 50% of the applied signal voltage appearing across it.

Figure 3 is a similar plot, except this time the capacitor value remains fixed at 1 millifarad and the test frequency is allowed to vary. It can be seen that better accuracy could be obtained by using test frequencies significantly lower than 1 00Hz. Frequencies in the region of 20 Hz would give a good result, but would require a significantly longer period in which to conduct the test. Furthermore, as the frequency drops the response of the sensor probe degrades significantly as it is effectively capacitively coupled to the capacitor under test, and this coupling degrades with a reduction frequency. Furthermore, the sense amplifier associated with the probe often forms 7 a high pass filter (inadvertently) with the probe.

A significant improvement in test performance can be achieved by providing ftirther voltage probes which attach to the terminals of the capacitor under test.

In the circuit shown in Figure 4, additional voltage sensing probes 10 and 12 are provided which, in use, are connected to or adjacent the terminals of the capacitor 2 under test. The probes 10 and 12 are connected via resistors 14 and 16 to the non-inverting input of a buffer amplifier 18. In the arrangement shown in figure 4, the non-inverting input sees the average voltage applied across the capacitor when the resistors 14 and 16 have equal values.

The output of the buffer 18 is supplied via resiAor 20 to the noninverting input of a differential amplifier 24, and via resistor 22 to the inverting input of the differential amplifier 24. The resistors 20 and 22 have large values (several TYM) so as to allow the reference signal derived by averaging the voltages seen by the probes 10 and 12 to be applied as a common mode signal to the differential amplifier, whilst not swamping the signal received via a sensor plate 26 connected to the amplifier 24. It might be expected that in a simplified installation the action of the buffer 18 might be dispensed with if the value of the resistors 20 and 22 is significantly greater than the value of resistors 14 and 16.

The sensor is an electrically non-contacting probe and need only comprise a single plate connected to one input of the differential amplifier. However superior performance is achieved if the sensor comprises a pair of plates. one plate, in use, faces towards the capacitor under test whilst the other is more distant. This arrangement gives improved immunity to large stray fields but, more importantly, serves to balance the impedance seen by the inputs of the differential amplifier. The connections between the plates and the amplifier may be made using a screened cable with the screen connected to the output of the buffer 18.

An output of the differential amplifier is fed to a signal measurement block 30 which may include a phase detector, an AC voltmeter and a threshold circuit.

4 8 The test circuit can be used in various ways. In a first sequence, the AC voltage source is connected across the capacitor 2 under test and the first and second voltage reference probes 10 and 12 are also connected across the capacitor under test. The buffer amplifier 18 forms a buffer signal which is equal to the average voltage across a capacitor, and this is supplied to the differential amplifier 24 as a common mode signal. As a result of this, the sensor plate 26 is supplied with an alternating bias voltage which is substantially the average of the voltage seen across the capacitor. Thus, from the point of view of the sensor plate 26, the capacitor is supplied with two equal anti-phase signals. One of these signals is effectively screened by the other. Thus the signal detected by the sensor plate and subsequently amplified by the differential amplifier effectively results from the signal supplied to the outermost plate of the capacitor. The output of the amplifier is supplied to a signal processor which performs a phase coherent detection of the received signal in order to determine the polarity of the capacitor. The inclusion of the resistive paths to and from the capacitor means that there is an arbitrary phase shift in the voltage seen across the capacitor. Thus the tester may be calibrated against a known good board and for each capacitor the phase shift may be measured. Advantageously the measured phase shift is used to set up an acceptable range of phase values to be used during the test. During subsequent testing a phase meter may be used to measure the phase shift. However, it is often easier to use the learnt phases to control a phase shifter to introduce a compensating phase shift between the measured signal and a phase reference signal such that, if the capacitor is correctly installed the signals are in phase, and if the capacitor polarity is reversed the signals are in antiphase, or vice versa. Thus the capacitor is deemed to be acceptable if the measured phase falls within the predetermined phase range. The magnitude of the output of the differential amplifier may also be thresholded in order to provide a level of confidence that the probe was adequately coupled to the capacitor under test. This test may also indicate the absence of a component.

In an alternative version of the test, the signal is applied to the capacitor under test and the first voltage reference probe is connected to one end of the capacitor. The second reference voltage probe is left open circuit or omitted. Thus the sensor plate then measures the received signal with respect to the voltage at one end of the capacitor. This measurement is then sent to the signal 9 processor 30 where its magnitude is measured, for example by a RMS voltmeter. The probe 16 is then moved to the second end of the capacitor under test and a similar measurement is taken. The ratio of the measured signals is then taken. This is indicative of which way round the capacitor has been inserted.

It is thus possible to provide an apparatus for and methods of testing the polarity of a capacitor in situ in a circuit board where only one terminal of the capacitor is driven with an AC signal.

The apparatus and method is capable of testing any electronic component in which the internal structure is such that one is effectively screened by the other.

Claims

1.

A polarity tester for testing the polarity of a capacitor, comprising: signal supply means for supplying an alternating signal to a first terminal of a capacitor under test via a first conductive path, a second terminal of the capacitor under test being connected to a reference voltage via a second conductive path; a probe for sensing an electric field generated by the capacitor under test; probe bias means for biasing the probe with an alternating bias voltage; and processing means for processing an output of the non-contacting probe with respect to the bias voltage to obtain an output signal indicative of the polarity of the capacitor under test.

2. A polarity tester as claimed in claim 1, in which the probe bias means comprises a first reference voltage probe for measuring a first reference voltage.

3. A polarity tester as claimed in claim 2, in which the probe bias means further comprises a second reference voltage probe and is arranged to derive the bias voltage as a function of the first and second reference voltages.

A polarity tester as claimed in claim 3, in which the probe bias means fonTis an average of the first and second reference voltages.

5. A polarity tester as claimed in claim 2, in which the first reference voltage probe is selectively connectable to, or adjacent, the terminals of the capacitor under test in turn.

6. A polarity tester as claimed in claim 3, in which the probe bias means is arranged to select between the first and second reference voltages.

7. A polarity tester as claimed in any one of the preceding claims in which the processing means comprises a differential amplifier arranged to form the difference between the 11 output of the probe and the bias voltage.

8. A polarity tester as claimed in any one of the claims 1 to 6, in which the processing means is a differential amplifier having a common mode reference signal derived from the bias voltage supplied to the inverting and non-inverting inputs thereof.

9. A polarity tester as claimed in claim 8, in which the non-contacting probe comprises a sense plate connected to one of the inputs of the differential amplifier.

10. A polarity tester as claimed in claim 8, in which the non-contacting probe comprises a pair of spaced apart plates, a first one of the plates being connected to the non-inverting input of the amplifier and a second one of the plates being connected to the inverting input of the amplifier.

A polarity tester as claimed in claim 10, in which connections between the noncontacting probe and the differential amplifier are made via a screened cable with the screen connected to the probe bias means.

12. A polarity tester as claimed in claim 2, in which the processing means is arranged, in use, to make a measurement of the amplitude or power of a first output signal when the first reference voltage probe is electrically connected to one of the terminals of the capacitor and to store a result of the measurement, to make a measurement of the amplitude of power of a second output signal when the first reference voltage probe is electrically connected to the other terminal of the capacitor, and to use the relationship between the first and second signals to output a signal indicative of the polarity of the capacitor.

13. A polarity tester as claimed in claim 12, in which the ratio of the first and second signals is used to determine the polarity of the capacitor under test.

14. A polarity tester as claimed in claim 12, in which the difference between the first and 12 second signals is used to determine the polarity of the capacitor under test.

15. A polarity tester as claimed in claim 12, in which at least one of the first and second signals is compared with a predetermined threshold in order to provide a measure of confidence that the capacitor under test is not missing.

16. A polarity tester as claimed in claim 3, in which, in use, the first reference voltage probe is connected to a first terminal of the capacitor under test and the second reference voltage probe is connected to a second terminal of the capacitor under test.

17. A polarity tester as claimed in claim 16, in which the processing means is arranged to detect the phase of the output signal anid the phase reference signal, wherein the phase reference signal is derived from the signal supply means.

18. A polarity tester as claimed in claim 16, in which the output signal of the processing means is synchronously detected with reference to a phase reference signal and the synchronously detected signal is used to determine the polarity of the capacitor.

19. A polarity tester as claimed in claim 18, in which the sign of the synchronously detected signal is used to indicate the polarity of the capacitor.

20. A polarity tester as claimed in claim 18, in which the amplitude of the synchronously detected signal is used to indicate the polarity of the capacitor under test.

21. A polarity tester as claimed in any one of the preceding claims, in which the reference voltage is ground.

22. A polarity tester as claimed in claim 18, in which the phase reference signal is derived from the signal supply means.

13

23. A polarity tester as claimed in claim 18, in which a predetermined phase shift is introduced into the phase reference signal or the output signal of the signal processing means.

24. A polarity tester as claimed in any one of the preceding claims, in which the probe is electrically non-contacting with the capacitor under test.

25. A method of testing the polarity of a capacitor, comprising the steps of supplying an alternating signal to a first terminal of a capacitor under test via a fust conductive path, a second terminal of the capacitor under test being connected to a reference voltage forming a signal return via a second conductive path; placing a non-contacting probe proximate to the capacitor under test in order to obtain a measurement of the electric field surrounding the capacitor under test; applying an alternating bias voltage to the noncontacting probe in order that the said electric field measurement is relative to the said bias voltage; measuring the noncontacting probe output relative to the said bias voltage in order to remove the bias voltage component from the measured output; and using a property of one or more measured output values to determine the polarity of the capacitor under test.

26. A method as claimed in claim 25, in which a first bias voltage is derived from a first reference voltage supplied by a first reference voltage probe connected to one of the first and second terminals of the capacitor under test, to provide a first measured output value; a second bias voltage is derived from a second reference voltage supplied by the first or a second reference voltage probe connected to the other one of the fast and second terminals of the capacitor under test, to provide a second measured output value; and the relationship between the first and second measured output values is used to detern-line the polarity of the capacitor.

27. A method as claimed in claim 26, in which the relationship of the first and second measured output values is the ratio of the root mean square or amplitude of the first 14 measured value to the root mean square or amplitude of the second measured value.

28. A claim as claimed in claim 26, in which the relationship of the first and second measured output values is the difference between the root mean square of the first measured value and the root mean square of the second measured value.

29. A method as claimed in claim 25, in which a first reference voltage probe is connected to one of the first and second terminals of the capacitor under test; as second reference voltage probe is connected to the other of the first and second terminals of the capacitor under test; and the bias voltage is derived as a function of the voltages measured by the first and second reference voltage probes.

30. A method as claimed in claim 29, in which the property of the measured output value used to determine the polarity of the capacitor under test is the phase of the measured output value with respect to the phase of the applied alternating signal.

31. A method as claimed in claim 29, in which the function of the voltages measured by the first and second reference voltage probes is the instantaneous average of the voltages measured by the first and second reference voltage probes.

32. A method as claimed in claim 3 1, in which the said relative phase of the measured output value is compared with a predetermined acceptable phase range and the capacitor under test is judged to be correctly inserted if the said relative phase is within the phase range.

33. A method as claimed in claim 32, in which the predetermined acceptable phase range is derived from an expected phase learnt by investigation of a known good component and the phase measurements stored for use in subsequent tests.