METHOD AND DEVICE FOR MONITORING AN ENGAGEMENT PROCESS OF AN ENGAGING PINION OF A STARTER MOTOR

• US 20140236453A1
• Filed: 06/19/2012
• Published: 08/21/2014
• Est. Priority Date: 07/08/2011
• Status: Active Grant

• Current Assignee: Seg Automotive Germany GMBH
• Original Assignee: Robert Bosch GmbH
• Inventors: Casey, Simon, Rocco, Tony
• Granted Patent: US 9,500,173 B2
• Time in Patent Office: Days
• US Class Current: 701/101
• CPC Class Codes: F02D 2041/1417 Kalman filter
• CPC Class Codes: F02D 2041/288 for performing a transformation into the frequency domain, e.g. Fourier transformation
• CPC Class Codes: F02N 11/08 Circuits or control means specially adapted for starting of engines
• CPC Class Codes: F02N 11/0848 with means for detecting successful engine start, e.g. to stop starter actuation
• CPC Class Codes: F02N 11/0851 characterised by means for controlling the engagement or disengagement between engine and starter, e.g. meshing of pinion and engine gear
• CPC Class Codes: F02N 11/0855 during engine shutdown or after engine stop before start command, e.g. pre-engagement of pinion
• CPC Class Codes: F02N 2200/044 Starter current
• CPC Class Codes: F02N 2200/047 Information about pinion position
• CPC Class Codes: F02N 2200/065 Relay current

Abstract

A method for monitoring an engagement process of an engaging pinion of a starter motor for a vehicle drive, in which an armature of the starter motor is moved, includes detecting a magnetic coil signal while current is applied to the magnetic coil of the magnet, in order to obtain a signal profile, particularly a current profile; generating a frequency profile by transforming the signal profile using a time-frequency transform; comparing the frequency profile to at least one reference frequency profile; and monitoring the engagement process based on the comparison.

Claims

1-14. -14. (canceled)

15. A method for monitoring an engagement process of an engaging pinion of a starter motor for a vehicle drive, in which an armature of the starter motor is moved by a magnet, comprising:

detecting a magnetic coil signal when current is applied to a magnetic coil of the magnet, in order to obtain a signal profile;

generating a frequency profile by transforming the signal profile using a time-frequency transform;

comparing the frequency profile to at least one reference frequency profile; and

monitoring the engagement process based on the comparison.

16. The method as recited in claim 15, wherein it is ascertained during the monitoring of the engagement process whether the engagement process is able to be carried out successfully.

17. The method as recited in claim 15, wherein, in response to the generating of the frequency profile, at least one of the following transforms is used:

a discrete Fourier transform, a fast Fourier transform (FFT), a short-time Fourier transform (STFT), and a wavelet transform.

18. The method as recited in claim 15, wherein the comparing includes ascertaining an error value based on a mean square error between the frequency profile and the at least one reference frequency profile.

19. The method as recited in claim 15, wherein the comparing includes correlating the frequency profile with the at least one reference frequency profile.

20. The method as recited in claim 15, wherein the comparing takes place to a first number of reference frequency profiles, which are each associated with a successfully carried out engagement process, and a second number of reference frequency profiles, which are each associated with an unsuccessfully carried out engagement process.

21. The method as recited in claim 15, wherein the monitoring of the engagement process includes ascertaining a particular reference frequency profile from which the frequency profile has a smallest deviation.

22. The method as recited in claim 15, wherein the signal profile is submitted to a scanning rate conversion before the generating of the frequency profile.

23. The method as recited in claim 15, wherein the signal profile is brought to a specified number of scanning values before the generating of the frequency profile.

24. The method as recited in claim 15, wherein the generating of the frequency profile is replaced by observing of the signal profile, so as to obtain an observation profile;

and

wherein the frequency profile is replaced by the observation profile, and the comparing of the frequency profile to at least one reference frequency profile is replaced by a comparison of the observation profile to at least one reference frequency profile.

25. The method as recited in claim 24, wherein the observation profile is obtained by a Kalman filtering of the signal profile or by using a real-time model of mechanical kinematics and/or electrical parameters.

26. A device for monitoring an engagement process of an engaging pinion of a starter motor for a vehicle drive, wherein an armature of the starter motor is moved by a magnet, the device comprising:

a sensor configured to detect a magnetic coil signal when a current acts on a magnetic coil of the magnet, so as to obtain a signal profile, and

a processor configured;

to generate a frequency profile by transforming the signal profile using a time-frequency transform;

to compare the frequency profile to at least one reference frequency profile; and

to monitor the engagement process based on the comparison.

27. A device for monitoring an engagement process of an engaging pinion of a starter motor for a vehicle drive, wherein an armature of the starter motor is moved by a magnet, the device comprising:

a sensor configured to detect a magnetic coil signal when a current acts on a magnetic coil of the magnet, so as to obtain a signal profile, and

a processor configured;

to generate an observation profile by observation of the signal profile;

to compare the observation profile to at least one reference frequency profile; and

to monitor the engagement process based on the comparison.

28. The device as recited in claim 27, wherein the observation profile is obtained by a Kalman filtering of the signal profile or by using a real-time model of mechanical kinematics and/or electrical parameters.