Electrical pulse counters – pulse dividers – or shift registers: c – Applications – Measuring or testing
Reexamination Certificate
2000-03-07
2001-09-11
Lam, Tuan T. (Department: 2816)
Electrical pulse counters, pulse dividers, or shift registers: c
Applications
Measuring or testing
C377S023000
Reexamination Certificate
active
06289072
ABSTRACT:
TECHNICAL FIELD
This invention relates to a method of computing speed with a sensor that develops one or more electrical pulses per unit movement of a device such as a rotary shaft, and more particularly to a method for estimating very low speeds of the device.
BACKGROUND OF THE INVENTION
One very common and relatively inexpensive technique for measuring the speed of a moving part is to produce electrical pulses in synchronism with movement of the part and to count the number of pulses produced per unit interval of time. For example, a variable reluctance sensor may be mounted relative to a rotary shaft for producing a pulse each time a gear tooth or notch formed on the shaft passes the sensor. The pulses may be applied to a control circuit (which may include a microprocessor) that counts the pulses and maintains a record of the time at which each pulse is received. In such case, the speed of the shaft (in revolutions per second, for example) can be computed as a simple function of the counter value, the time elapsed between the first and last pulses of the stored count, and number of pulses produced per revolution of the shaft. For the sake of convenience, the speed calculation is usually repeated at a known update rate, as may be defined, for example, by the loop time of a microprocessor-based controller. In other words, the controller periodically executes a software routine that checks the counter value and computes a new speed value. A limitation of this technique occurs when the speed of the shaft is so low that the pulse frequency is lower than the update frequency; that is, when no pulses have been received since the previous speed calculation. At such point, there is insufficient information to update the calculated speed, although it can be deduced that the previously calculated value is no longer correct.
When the calculated speed is used for control purposes, the above-described limitation also limits the ability to perform any control function based on the calculated speed. For example, when the speed of a vehicle engine or transmission is calculated and used in a control strategy to engage a transmission clutch, it becomes difficult or impossible for the control strategy to achieve high shift quality at low vehicle speeds.
Known techniques for handling the above-described low speed limitation include simply holding the value of the previous speed calculation until the next pulse is received, or assuming that the speed is zero and ramping the indicated speed to zero using a first order filtering technique. However, these approaches are inadequate for control purposes where accurate low speed information is important. In the transmission shift control application, for example, certain shifts occur at very low speeds, such as when shifting between forward and reverse speed ranges, and high shift quality is difficult to achieve when the above-mentioned techniques are used to estimate speed.
Accordingly, what is needed is a speed estimation method that develops a more accurate low speed indication from an inexpensive pulse-type sensor.
SUMMARY OF THE INVENTION
This invention is directed to an improved speed indication method for a sensor that develops one or more electrical pulses per unit movement of a movable part, wherein the indicated speed is updated based on the number of speed pulses generated per unit time, but is constrained by a limit function based on the time elapsed without the generation of a speed pulse. The indicated speed is updated in accordance with a first methodology when at least one pulse has been generated since the last update, and in accordance with a second methodology when no pulses have been produced since the last update. In the first methodology, a speed is computed based on a count of the produced pulses and a time interval spanning the production of the counted pulses, and the indicated speed is updated in accordance with the computed speed. In the second methodology, a speed bound is computed based on a no-pulse interval during which no pulses have been produced, assuming that a pulse is produced when the speed bound is computed, and the indicated speed is updated in accordance with the lower of the computed speed bound and the indicated speed as previously updated. The number of successive updates of the indicated speed with no pulses being produced by the sensor is counted, and the indicated speed is set to zero when the count reaches a threshold. When one or more pulses are produced after a period of no pulses, the first methodology updates the indicated speed in accordance with the lower of the computed speed and the computed speed bound, providing a transition from the second methodology to the first methodology. As a result, the indicated speed provides a smoothly varying and more accurate representation of the speed of the part at very low speeds, enabling more accurate low speed control.
REFERENCES:
patent: 4890231 (1989-12-01), Frantz
patent: 5117698 (1992-06-01), Baumoel
patent: 5157461 (1992-10-01), Page
Hubbard Gregory A
Runde Jeffrey Kurt
General Motors Corporation
Hargitt Laura C.
Lam Tuan T.
LandOfFree
Near-zero speed estimation method for a pulse counting speed... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Near-zero speed estimation method for a pulse counting speed..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Near-zero speed estimation method for a pulse counting speed... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2542132