Electrical transmission or interconnection systems – Plural load circuit systems – Control of current or power
Reexamination Certificate
2001-06-15
2004-05-18
Toatley, Jr., Gregory J. (Department: 2836)
Electrical transmission or interconnection systems
Plural load circuit systems
Control of current or power
C323S283000
Reexamination Certificate
active
06737760
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention is directed to a method for negative feedback controlling electrical power delivered to an electrical load, which method comprises generically, and as known to the skilled artisan, monitoring the electrical power delivered to the load, thereby generating a monitoring signal, forming in dependency of the monitoring signal and of a rated value signal a control deviation signal and adjusting—via a controller as e.g. a proportional or proportional-integral controller—the electrical power delivered and monitored in function of the control deviation signal.
The present invention thereby departs especially from such method and power supply for delivering high power of at least 100 VA to a load.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide such a negative feedback control method and, accordingly, such a negative feedback controlled power supply for superior accuracy of controlled power delivered to the load, with respect to a rated power value to be delivered.
Under a first aspect, the present invention departs from a method for negative feedback controlling electrical power delivered to an electrical load as mentioned above, whereat monitoring the electrical power delivered to the load results first in an analog monitoring signal, which is then analog to digital converted so as to result in a digital monitoring signal.
Under a second aspect, the present invention departs from a method for negative feedback controlling electrical power delivered to an electrical load as mentioned above, whereat adjusting the electrical power delivered to the load and monitoring is performed as a function of the control deviation signal—again via a respective controller—by means of Pulse-Width Modulation (PWM).
In a method for negative feedback controlling electrical power, whereat the controlled value, namely the power delivered to the load, is digitalized as under the first aspect of the present invention, the overall accuracy of the negative feedback control significantly depends on the accuracy of the analog to digital conversion of the monitored signal. As is well known in the art of analog to digital conversion, noise of the analog input signal leads to digital output signal jitters by at least one least significant bit (LSB). This problem is customarily resolved by oversampling, i.e. by establishing a sampling rate which is considerably higher than necessitated by the spectrum of the analog signal to be converted. In fact, by oversampling, a multitude of digital samples are generated in a predetermined time frame, and the respective digitalized output value is formed by averaging the digital samples. Nevertheless, oversampling is always limited by the layout of a specifically considered converter not suited to handle sampling rates above a specific maximum rate. Rising the sampling rate over such limit makes it necessary to apply a differently designed A/D-converter which is often much more expensive and more critical to operate.
Under the generic object of the present invention as outlined above, and under the first aspect, it is therefore an object of the present invention to provide for an increased conversion accuracy of A/D-conversion of the monitoring signal, thereby avoiding the above mentioned drawbacks.
This object is inventively resolved by the method mentioned above and according to the preamble of claim 1, at which the analog monitoring signal is analog to digitally converted by performing the analog to digital conversion at least twice in parallel.
Thereby, at least two analog to digital converters, each construed for a predetermined sampling rate, are operated in parallel. This leads to an overall conversion with at least double that sampling rate, wherefore each of the converters is construed for, if both converters—as preferred—are operated at equal sampling rates.
Under the second aspect of the present invention the following has to be considered: Whenever a method for negative feedback controlling electrical power delivered to an electrical load comprises adjusting the electrical power delivered to the load and monitored by means of PWM, accuracy of adjustment is significantly limited by the predetermined minimum pulse-width adjustment increment by which pulse length may be at all adjusted.
It is therefore a further object of the present invention, at a method as mentioned above, whereat, within the negative feedback loop, adjusting of the electrical power delivered to the load and monitored in function of the control deviation signal is performed by pulse-width modulation with a predetermined pulse repetition period, to improve adjustment accuracy beyond the limit set by the predetermined minimum pulse-width adjustment increment.
This object is resolved inventively and according to the characterizing part of claim 2 by calculating from the control deviation signal a desired pulse-width adjustment increment. Such a desired pulse-width adjustment increment will normally not accord with an integer multiple of the predetermined minimum pulse-width adjustment increment of the PWM considered. Therefore, and according to this second aspect of the present invention, a pulse of predetermined, preferably fixed length is not applied in every pulse repetition period, but only so often in time that, averaged over time, the adjustment by the integer multiple on one hand and/or the pulse accords with a pulse-width modulation adjustment with said desired pulse-width adjustment increment.
Thereby, one can exactly deal with any desired adjustment according to any fractions of the predetermined minimum pulse-width adjustment increment.
If e.g. the minimum predetermined pulse-width adjustment increment is of 30 nsec. and an actual control deviation would necessitate a pulse-width adjustment by an increment only of 10 nsec., then the calculation will reveal in preferred mode that the pulse of predetermined length (as in a most preferred embodiment a pulse which is the predetermined minimum pulse-width adjustment increment of 30 nsec.) is to be applied in every third pulse repetition period. This leads, time-averaged as by filtering, to the same result as applying a 10 nsec. pulse in every period of PWM.
If e.g. the control deviation signal reveals a desired pulse-width adjustment increment of 40 nsec., then there will be applied in each pulse repetition of the PWM one predetermined minimum pulse-width adjustment increment and, additionally, in every third period an additional predetermined minimum pulse-width adjustment increment, if, as in the most preferred mode, the pulse is selected to be just this increment.
In other words, this inventive technique combines customary pulse-width adjustment technique of PWM with a pulse frequency modulation (PFM) technique, at which pulses of a predetermined fixed length, namely as preferred according to the predetermined minimum pulse-width, are applied at a varying frequency.
According to the wording of claim 3, in a preferred embodiment of the inventive method and power supply both aspects of the invention as outlined above, namely of inventive A/D-conversion and of inventive adjustment by PWM and superimposed PFM, are combined.
Although the present invention under all its aspects may be used or realized for lower power supply, in a far preferred mode of operation, it is realized for controlling the electrical power of at least 100 VA delivered to a load. Monitoring the electrical power delivered to the load may be performed by monitoring the current or voltage delivered to the load.
In spite of the fact that the above mentioned A/D-conversion, at least twice in parallel, might be applied in cases, where both conversions are performed at minimum required sampling rates, additionally to the inventively performed parallel conversion, each of the A/D-conversions is performed with oversampling.
In spite of the fact that A/D-conversion in parallel could be performed in some cases at mutually different sampling rates, e.g. at sampling rates of integer ratio, A/D-conversions are per
Eisenberg Jacob
Institut Paul Scherrer
Paul Scherrer Institut
Toatley , Jr. Gregory J.
LandOfFree
Method for negative feedback controlling electrical power... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for negative feedback controlling electrical power..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for negative feedback controlling electrical power... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3236809