Testing process for the quality control of electromagnetically actuated switching devices

• US 5,629,864 A
• Filed: 11/22/1994
• Issued: 05/13/1997
• Est. Priority Date: 05/22/1992
• Status: Expired due to Fees

• Current Assignee: Siemens AG
• Original Assignee: Siemens AG
• Inventors: Noe, Ernst-Ludwig, Schwarz, Peter
• Primary Examiner(s): Voeltz, Emanuel T.
• Assistant Examiner(s): COLE, TONY MITCHELL
• Application Number:
  US08/335,877
• Time in Patent Office: 903 Days
• Field of Search: 364/506, 364/508, 73/584, 73/570, 73/587, 73/609, 73/632, 73/649, 73/658, 73/659, 324/415, 324/418, 324/420, 324/421, 324/422, 324/423, 324/424, 324/513, 340/683, 381/56
• US Class Current: 702/115
• CPC Class Codes: G01R 31/3278 of relays, solenoids or reed switches measuring contact resistance G01R27/205; high voltage magnetic switches G01R31/3271, G01R31/333; testing electric windings G01R31/72; monitoring of fail safe circuits H01H47/002

Abstract

Switching devices in which current contacts are made by actuating a magnetic armature, are tested by analyzing the vibrations or sound emission occurring under operating conditions. According to the present invention, the sound emitted by the switching device when it is actuated is detected by a sensor and is transmitted to an evaluation device for discriminating the time and/or frequency curve of individual sound components. With the process of the present invention, discriminating between the sound signals produced by the making of the contacts on the one hand, and the striking of the armature on the other hand, is possible. In addition, a sound analysis in a suitable time range permits any faults present in the switching device to be classified. Finally, especially in the application to flashing signal relays, whether the sound of the relay is proper, can be objectively determined. In all cases, particularly the high frequency sound components are evaluated.

Claims

1. A testing process for the quality control of electromagnetically actuated switching devices in which current contacts are made by actuating an armature, the process including steps of:

a) actuating the armature of a switching device to close the switching device;

b) detecting sound emitted by the switching device when its armature is actuated;

c) transmitting the sound detected in step (b) to an evaluation device;

d) discriminating sound signals caused by a making of the contact from sound signals caused by the armature striking a magnetic core by analyzing at least one of a time curve of individual components of the sound and a frequency curve of the individual components of the sound;

e) evaluating only the sound signals caused by the armature striking the magnetic core.

2. The method of claim 1 wherein the step of actuating the switching device closes a main current loop, further comprising a step of:

f) detecting an instant of the making of the contact based on a time when the main current loop is closed.

3. The process of claim 2 further comprising a steps of:

g) digitally sampling the sound signals after the instant of the making of the contact;

h) calculating a modified signal sample y(t) from the sound signals digitally sampled in step (g) according to the following relationship;

space="preserve" listing-type="equation">y(t)=Maximum(Abs(x(t)),(y(t-1)-d)),

wherein x(t) is a digitally sampled sound signal at time t, x(t-1) is a digitally sampled sound signal immediately preceding sample x(t), abs(x(t)) is an absolute value of x(t), and d is a decay constant;

i) determining an instant at which the armature contacts the magnetic core to be the instant of the maximum value of y(t); and

j) determining whether the instant at which the armature contacts the magnetic core determined in step (i) falls within a predetermined permissible time window for the switching device.

4. The process of claim 2 wherein the step of detecting begins when a coil voltage is applied to the magnetic armature.

5. The process of claim 1 further comprising steps of:

f) generating an envelope curve from the sound signal;

g) determining a time derivative of the envelope curve;

h) detecting an instant at which the armature strikes a magnetic core to be the time derivative determined in step (g);

i) determining whether the instant at which the armature strikes the magnetic core determined in step (h) falls within a predetermined permissible time window for the switching device.

6. The process of claim 1 wherein the step of detecting includes a sub-step of detecting at least one of airborne sounds and high frequency sounds by means of a sensor which is separated from the switching device.

7. The process of claim 1 wherein the step of detecting includes a sub-step of detecting at least one of structure-borne sounds and high frequency sounds by means of a sensor which physically contacts the switching device.

8. A testing process for quality control of an electromagnetically actuated switched device in which current contacts are made when an armature is actuated and in which, when properly functioning, the armature comes to rest during a time range, the process comprising steps of:

a) actuating the armature to close the switching device;

b) detecting a sound emitted by the switching device when the armature is actuated;

c) transmitting the sound detected in step (b) to an evaluation device;

d) discriminating individual components of the sound detected in step (b) based on at least one of a time curve of the individual components of the sound and frequency curve of the individual components of the sound;

e) analyzing the sound transmitted in step (c) in the time range; and

f) classifying the analyzed sound detected within the time range as at least one of a plurality of faults.

9. The process of claim 8 wherein the switching device is a relay operated with an alternating current.

10. The process of claim 8 wherein the step of analyzing in step (e) further includes sub-steps of:

i) calculating a cepstrum function from the sound; and

ii) determining whether a humming exists in the switching device based on the cepstrum function calculated in step (e)(i).

11. The process of claim 8 wherein the step of analyzing in step (e) further includes sub-steps of:

i) calculating an effective value of a signal amplitude from the sound; and

ii) determining whether a chattering exists in the switching device based on the effective value calculated in step (e)(i).

12. The process of claim 8 wherein the step of detecting includes a sub-step of detecting at least one of airborne sounds and high frequency sounds by means of a sensor which is separated from the switching device.

13. The process of claim 8 wherein the step of detecting begins when a coil voltage is applied to the magnetic armature.

14. The process of claim 8 wherein the step of classifying faults based on the sound analyzed in step (e) is performed by a self-learning classifier.

15. The process of claim 14 wherein the self-learning classifier includes a neural network.

16. The process of claim 8 wherein the step of detecting includes a sub-step of detecting at least one of structure-borne sounds and high frequency sounds by means of a sensor which physically contacts the switching device.

17. A testing process for the quality control of electromagnetically actuated switching devices in which current contacts are made by actuating an armature, the process comprising steps of:

a) actuating the armature to close the switching device;

b) detecting sound emitted by the switching device when the armature is actuated;

c) transmitting the sound detected in step (b) to an evaluation device;

d) discriminating individual components of the sound based on at least one of a time curve of the individual components of the sound and frequency curve of the individual components of the sound;

e) separating sound signals which are generated when the contact is made and generated when the armature strikes a magnetic core from sound signals caused by a post-switching oscillation of components; and

f) objectively determining whether the switching device has a proper sound based on the separated sound signals.

18. The process of claim 17 wherein the switching device is a motor vehicle flashing signal relay.

19. The process of claim 17 further comprising steps of:

digitizing the sound signal detected in step (b); and

defining a first time section in which a noise caused by the armature striking the magnetic coil lies;

defining a second time section in which the noise caused by the armature striking the magnetic coil has decayed; and

separately evaluating the first and second time sections.

20. The process of claim 17 wherein the step of detecting includes a sub-step of detecting at least one of airborne sounds and high frequency sounds by means of a sensor which is separated from the switching device.

21. The process of claim 17 wherein the step of detecting includes a sub-step of detecting at least one of structure-borne sounds and high frequency sounds by means of a sensor which physically contacts the switching device.

22. The process of claim 17 wherein the step of detecting begins when a coil voltage is applied to the magnetic armature.

23. The process of claim of claim 17 wherein the step of objectively determining whether the switching device has a proper sound uses a self-learning classifier.

24. The process of claim 23 wherein the self-learning classifier includes a neural network.